ULTRASOUND PROBE EXAMINATION

Risk studies

ULTRASOUND PROBE EXAMINATION

RISK STUDIES

There are numerous studies concerning the risk of being infected

during an endocavitary examination despite protection of the

probe by a single-use sheath and low-level disinfection of the

probe by a disinfectant wipe or a spray.

Early studies have shown that single-use condoms or sheaths have

macroscopic and microscopic leakages in high proportions (1-9%).

Thus, they do not provide satisfactory protection of the probe

and of the patient.

More recent studies have shown that the protocol [single use

sheath + low-level disinfection] does not prevent contamination

of the probes. 2-3% of endovaginal probes used in the next

patient are infected with High Risk HPV (Human Papilloma Virus).

HPV 16 is a highly resistant virus to glutaraldehyde (GTA) and

ortho-phthalaldehyde (OPA).

Risk Studies

1. Condom perforation during transrectal ultrasound guided (TRUS)

prostate biopsies: a potential infection risk - Junaid Masood Æ

Stelios Voulgaris Æ Olisa Awogu Æ Choudhary Younis Æ Andrew J. Ball Æ Tom

W. Carr - Springer Science+Business Media B.V. 2007

2. Transvaginal ultrasound probe contamination by the human

papillomavirus in the emergency department - Shuk Ting Christine Ma,1

A C Yeung,2 Paul Kay Sheung Chan,2 - Colin A Graham1 - 2012

3. High Risk HPV Contamination of Endocavity Vaginal Ultrasound

Probes: An Underestimated Route of Nosocomial Infection? Jean

S.Casalegno1*, Karine Le Bail Carval2, Daniel Eibach1,3, MarieLaure

Valdeyron4, G.Lamblin2, Georges Mellier2, Bruno Lina1, P.Gaucherand2,

P.Mathevet2, Yahia Mekki1* 1 Laboratory of Virology, Hospices Civils de

Lyon, Lyon, France, 2 Gynecology Obstetrics Dpt, Hospices Civils de Lyon,

Lyon, France, 3 European Public Health Microbiology Training Programme,

European Centre for Disease Prevention and Control, Stockholm, Sweden, 4

Preventive Medecine Dpt, Hospices Civils de Lyon, Lyon, France, – Plos One

2012

4. Medical Device Alert – MHRA - 2012 5. Survey of microbial contamination of ultrasound probe handles

following high level dinsfection with ortho-phtaladehyde – G.R.

McNally, A.Ngu – Medical Imaging, Royal Hospital for Women, Sydney, NSW,

Australia - 2013

6. Impact of Vaginal-Rectal Ultrasound Examinations with Covered and

Low-Level Disinfected Transducers on Infectious Transmissions in

France - Sandrine Leroy, Fatima M’Zali, Michael Kann, David J. Weber and

David D. Smith - Infection Control and Hospital Epidemiology - The

University of Chicago Press on behalf of The Society for Healthcare

Epidemiology of America – 2014

7. Persistence of Microbial Contamination on Transvaginal Ultrasound Probes despite Low-Level Disinfection Procedure - Fatima M’Zali1*,

Carole Bounizra1, Sandrine Leroy2, Yahia Mekki3, Claudine Quentin-

Noury1, Michael Kann1 - Université Bordeaux Segalen – Plos One 2014

8. Susceptibility of high-risk human papillomavirus type 16 to

clinical disinfectants - Craig Meyers Journal of Antimicrobial

Chemotherapy, 2014

1. Condom perforation during

transrectal ultrasound guided

(TRUS) prostate biopsies: a

potential infection risk - Junaid Masood Æ Stelios Voulgaris Æ Olisa Awogu Æ

Choudhary Younis Æ Andrew J. Ball Æ Tom W.

Carr - Springer Science+Business Media B.V.

2007

ORIGINAL ARTICLE

Condom perforation during transrectal ultrasound guided(TRUS) prostate biopsies: a potential infection risk

Junaid Masood Æ Stelios Voulgaris Æ Olisa Awogu ÆChoudhary Younis Æ Andrew J. Ball ÆTom W. Carr

Received: 29 January 2007 / Accepted: 15 March 2007 / Published online: 21 July 2007

Springer Science+Business Media B.V. 2007

Abstract

Introduction Transrectal ultrasound (TRUS) guided

prostate biopsies are amongst the most common

outpatient diagnostic procedures performed in uro-

logy practice. Of concern appear to be recent reports

of infectious complications following this procedure

in which contamination of the biopsy equipment was

the likely source. This study looks at the rate of

condom perforation during prostate biopsy and we

look to highlight the potential problems, which may

arise as a result of inadequate cleansing of the

equipment between cases during a busy prostate

biopsy clinic

Material and methods All patients attending for

prostate biopsies over a three-month period in our

institution were included in the study. All condoms

(latex) used were made by the same manufacturer and

were checked prior to the procedure and found to

have no leaks. The biopsy gun was inserted through

an externally placed needle guide, as is standard

practice in many departments in the UK. After the

end of each procedure the condom was removed from

the rectal probe and filled once again with water to

assess for perforations. Two experienced surgeons

carried out all the procedures.

Results 10 out of 107 patients were found to have at

least one perforation in the condom. In some of the

condoms there were multiple perforations.

Discussion We have demonstrated a significant

condom perforation rate (9%) amongst patients under-

going prostate biopsies. This raises the serious issue of

hygiene and cross infection, particularly with blood

borne communicable diseases such as hepatitis and

HIV unless strict disinfection and sterilization proto-

cols are followed between patients. Perforation of the

condoms used during TRUS guided prostate biopsy

and hence faecal and blood contamination of the

biopsy equipment could potentially have far-reaching

implications for urologists and the infection control

community. Although the risk of cross infection is

probably small this serious issue needs addressing.

Introduction

Transrectal ultrasound (TRUS) guided prostate biop-

sies are amongst the most common outpatient diag-

nostic procedures performed in urology practice. The

procedures are generally performed in follow-up to

elevated levels of prostate-specific antigen (PSA) or

abnormal digital rectal examinations. Septicaemia is

a known but rare complication of this procedure [1].

J. Masood S. Voulgaris O. Awogu C. Younis A. J. Ball T. W. Carr

Department of Urology,

Southend University Hospital, Essex, UK

J. Masood (&)

47 Priory Mews, 44 Station Avenue, Prittlewell Southend,

Essex SS2 5EP, UK

e-mail: Junaido@aol.com

123

Int Urol Nephrol (2007) 39:1121–1124

DOI 10.1007/s11255-007-9213-y

In one recent study infectious complications occurred

in 10% of patients undergoing this procedure without

antibiotic prophylaxis and in 3.7% of patients who

had antibiotic prophylaxis [2].

Of concern appear to be recent reports of infec-

tious complications following this procedure in which

contamination of the biopsy equipment was the likely

source [3]. There have also been other reports

suggesting that the ultrasound couplant gel is a

potential source of infection with contamination

occurring at the time of manufacture with organisms

that degrade parabens, which are commonly used as

stabilizing agents [4].

Given the above facts, perforation of the condoms

used during TRUS-guided prostate biopsy and hence

faecal and blood contamination of the biopsy equip-

ment could potentially have far-reaching implications

for urologists and the infection control community.

To our knowledge most urology departments per-

forming TRUS guided prostate biopsies simply

disinfect the ultrasound probes for a short and often

inadequate time period between patients.

This study looks at the rate of condom perforation

during prostate biopsy and we aim to highlight the

potential problems that may arise as a result of

inadequate cleansing of the equipment between cases

during a busy prostate biopsy clinic.

Materials and methods

All patients attending for prostate biopsies over a

three-month period in our institution were included in

the study. All study patients had eight cores of tissue

(22 mm) taken from the same locations around the

prostate. All patients were given a peri-prostatic

block using 5 ml of 1% lignocaine through the needle

guide, which was fitted outside of the condom as is

standard. The biopsy gun was also inserted through

this externally placed needle guide, as is standard

practice in many departments in the UK (Fig. 1). The

rectal probe was disinfected by placing it in a

commonly used disinfectant called Mikrozid AF

liquid (Schulke and Mayr, UK) between cases for a

short period. All condoms (latex) used were made by

the same manufacturer and were checked prior to the

procedure and found to have no leaks. All of the

condoms had 5 ml of standard low-viscosity ultra-

sound couplant gel inserted prior to insertion over the

rectal probe. After the end of each procedure the

condom was removed from the rectal probe and filled

once again with water to assess for perforations. Two

experienced surgeons carried out all the procedures.

Results

Ten out of 107 patients were found to have at least

one perforation in the condom (Fig. 2). In some of the

condoms there were multiple perforations. All of the

patients, including those in whom condom perfora-

tions had been discovered, were contacted at the end

of the study and no one appeared to have suffered any

unexpected ill effects. One patient was admitted with

coliform sepsis and treated with intravenous antibi-

otics.

Discussion

Recent reports have highlighted contaminated med-

ical equipment (needle guides) as the cause of

infectious complications from prostate biopsy with

cases of infections that were likely related to

contamination of TRUS prostate equipment that had

not been adequately cleaned (i.e., by brushing) or

properly sterilized and had been rinsed improperly

with tap water after reprocessing [3]. Probe swab

Fig. 1 A standard trans-rectal ultrasound probe with externally

attached biopsy apparatus

1122 Int Urol Nephrol (2007) 39:1121–1124

123

samples after condom perforation have also been

shown to be positive for bacterial growth in patients

undergoing transvaginal sonography [5].

Commonly used disinfectant fluids such as the one

we use claim to be effective against bacteria, fungi,

and most viruses but by their very nature cannot

eradicate all pathogens and hence are not ideal

products.

The level of disinfection or sterilization is depen-

dent on the intended use of the object: critical items

(such as surgical instruments, which contact sterile

tissue), semi-critical items (such as endoscopes,

which contact mucous membranes), and noncritical

items (such as stethoscopes, which contact only skin)

require sterilization, high-level disinfection, and low-

level disinfection, respectively. Cleaning must always

precede high-level disinfection and sterilization [6].

According to the Spaulding system [7] for repro-

cessing medical devices prostate biopsy needle

guides are critical devices because the needles that

pass through them penetrate sterile tissue. After

adequate manual cleaning, critical devices must be

sterilized before reuse. Steam sterilization is the

preferred method for reprocessing heat-stable medi-

cal devices. Manual cleaning to remove biologic

material is a necessary first step in reprocessing any

medical device; disinfection and sterilization proto-

cols do not work effectively on visibly soiled

surfaces. Because the lumens of needle guides and

needle guide support channels and assemblies are

long and narrow, manual cleaning is difficult without

the use of special equipment designed to clean the

device [3].

The practice of rinsing the needle guide in tap

water after reprocessing might contribute to its

contamination as Pseudomonas aeruginosa is well

known to colonize tap water and has the ability to

form biofilms on medical devices, which are difficult

to remove [3].

Burkholderia cepacia infections have been re-

ported as a result of intrinsic contamination of the

ultrasound couplant gel by paraben-degrading micro-

organisms [4]. This is another potential hazard if the

condom is perforated during prostate biopsy.

As well as infectious complications from the

aforementioned organisms leakage of viral particles

such as human immunodeficiency virus (HIV) has

been demonstrated through perforated latex condoms

in vitro [8]. Such transfer of virus from a patient into

the biopsy equipment is a potential hazard during

prostate biopsy if the condom surrounding the probe is

perforated. This could represent a serious infection

hazard to subsequent patients if strict sterilization

protocols are not followed for the reusable equipment

used during TRUS-guided prostate biopsy. Transrec-

tal probes are, to the best of our knowledge, only

disinfected for a short period between patients in most

UK urology departments. Adequate cleaning, disin-

fection and sterilization is more difficult in systems

employing ultrasound transducers that have internal

channels through which the needle guides and biopsy

needles. These channels could easily harbor patho-

gens and simply flushing the channel between proce-

dures in not sufficient. Systems employing ultrasound

transducers where the needle guides are externally

fitted are easier to disinfect, as there are no internal

channels where pathogens could hide.

Conclusion

We have demonstrated a significant condom perfora-

tion rate (9%) amongst patients undergoing prostate

biopsies. This raises the serious issue of hygiene and

Fig. 2 A water-filled condom demonstrating a perforation

caused during the procedure

Int Urol Nephrol (2007) 39:1121–1124 1123

123

cross-infection, particularly with blood-borne com-

municable diseases such as hepatitis and HIV unless

strict disinfection and sterilization protocols are

followed between patients. Although the risk of

cross-infection is probably small this serious issue

needs to be addressed.

References

1. Crundwell MC, Cooke PW, Wallace DM (1999) Patients’

tolerance of transrectal ultrasound-guided prostatic biopsy:

an audit of 104 cases. BJU Int 83:792–795

2. Puig J, Darnell A, Bermudez P, Malet A, Serrate G, Bare M,

Prats J (2006) Transrectal ultrasound-guided prostate biop-

sy: is antibiotic prophylaxis necessary? Eur Radiol

16(4):939–943

3. Gillespie J, Arnold KE, Kainer MA, Jensen B, Arduino M,

Hageman J, Srinivasan A (2006) Pseudomonas aeruginosa

infections associated with transrectal ultrasound guided

prostate biopsies— Georgia, 2005. CDC-MMRW

55(28):776–777

4. Hutchinson J, Runge W, Mulvey M, Norris G, Yetman M,

(2004) Valkova N et al Burkholderia cepacia infections

associated with intrinsically contaminated ultrasound gel:

the role of microbial degradation of parabens. Infect Control

Hosp Epidemiol 25:291–296

5. Amis S, Ruddy M, Kibbler CC, Economides DL, Maclean

AB (2000) Assessment of condoms as probe covers for

transvaginal sonography. J Clin Ultrasound 28(6):295–298

6. Rutula WA, Weber DJ (2004) Disinfection and sterilization

in health care facilities: what clinicians need to know. Clin

Infect Dis 39:702–709

7. Spaulding EH (1968) Chemical disinfection of medical and

surgical materials chapter 32). In: Lawrence CA, Block SS

(eds) Disinfection, sterilization and preservation. Lea &

Febiger, Philadelphia, PA, pp 517–531

8. Carey RF, Herman WA, Retta SM, Rinaldi JE, Herman BA,

Athey TW (1992) Effectiveness of latex condoms as a

barrier to human immunodeficiency virus sized particles

under conditions of simulated use. Sex Transm Dis

19(4):239–244

1124 Int Urol Nephrol (2007) 39:1121–1124

123

2. Transvaginal ultrasound probe

contamination by the human

papillomavirus in the

emergency department - Shuk Ting

Christine Ma,1 A C Yeung,2 Paul Kay Sheung

Chan,2 - Colin A Graham1 - 2012

Transvaginal ultrasound probe contamination by thehuman papillomavirus in the emergency department

Shuk Ting Christine Ma,1 A C Yeung,2 Paul Kay Sheung Chan,2

Colin A Graham1

ABSTRACTObjective To determine if human papillomavirus (HPV)DNA can be detected on the transvaginal sonography(TVS) probe in the emergency department (ED) andwhether the current barrier method plus disinfection canprevent HPV contamination of the TVS probe.Methods This was a two-part cross-sectional study. Inthe first part, surveillance samples were taken from theTVS probe for HPV DNA detection daily for 2 months. Inthe second part, patients presenting with earlypregnancy complications were identified in the ED andhigh vaginal swabs were taken for HPV DNA testing.Several probe swabs were taken to identify ifcontamination was possible in cases where theprocedure was done on an HPV carrier.Results A total of 120 surveillance samples wereobtained, nine of which (7.5%) tested positive for HPVDNA. In the second part, 76 women were recruited, ofwhom 14 (18.4%) were HPV carriers. After theprocedure and disinfection of the probe, three out of the14 probe samples (21%) were HPV DNA positive.Conclusions HPV is commonly encountered in the EDand contamination of the TVS probe with HPV ispossible. Although it is difficult to prove the viability andinfectivity of the virus, vigilant infection control measuresshould be maintained.

INTRODUCTIONHuman papillomavirus (HPV) is the most commonsexually transmitted disease worldwide with 10%e20% of both men and women having molecularevidence of HPV infection.1 The cumulative risk ofacquiring HPV infection is reported to be 45% at3 years after the first sexual relationship, and theoverall prevalence is 25% in sexually active youngwomen.2 Most infections are subclinical and tran-sient. Risk factors for infection include earlier age ofcoitarche, more sexual partners, smoking and reducedimmunity.3 4 Nevertheless, infection is common evenin those without identifiable risk factors.HPV is a circular double-stranded DNA virus

consisting of more than 150 genotypes. There isa well-established relationship between cervical neo-plasia and certainHPV subtypes, namely types 16, 18,31, 33 and 45. Persistent infectionwith these high-riskHPVs may lead to abnormal cervical cell changes,increasing the risk of cervical cancer.5 Among thosehigh-risk HPV types, types 16 and 18 together causeabout 70% of all cases of cervical cancer.6 Other typesof sexually transmitted HPV (type 6 and 11) areresponsible for genital condylomata.Bedside ultrasound examination is gaining

importance in the everyday practice of the emer-

gency department (ED). The use of ultrasonog-raphy is mainly focused on biliary disease,intrauterine pregnancy and abdominal aorticaneurysms,7 as well as looking for peritoneal fluidand pericardial temponade in trauma patients.8

Many studies have demonstrated that emergencyphysician performed ultrasonography can be veryuseful in the management of early pregnancybleeding,9e12 with potential reductions in thelength of inpatient stay under the care of thegynaecology team. Patients identified as having anintrauterine pregnancy can be safely dischargedfrom ED with proper advice and an early follow-upappointment at an early pregnancy assessmentclinic. This decreases treatment time in the ED by55%, and saves total costs of 63% per patientwithout major adverse outcomes.13 14 In ourdepartment, there was a dramatic reduction in thenumber of gynaecological admissions from 75% to26% when we introduced this in 2009.15

The use of transvaginal sonography (TVS) hasconsequently become more popular in EDs inrecent years. The TVS probe is routinely protectedby a condom, acting as a physical barrier tocontamination. Studies have shown that theperforation rate of these condoms ranged from0.9% to 5%.16e18 One large scale study showed thatthe condom perforation rate was 2%, with 65% ofthe leakage points being <10 cm from the tip.18

With these expected perforation rates, staff areadvised to follow the proper steps for disinfectionof the TVS probe after each scan to prevent cross-infection.There are few studies on this issue of TVS probe

contamination with HPV in the ED. It is unclearwhether the current disinfection method is sufficientto clear up the virus in cases of contamination.The aims of the study were (1) to determine if

any HPV DNA could be detected on the TVS probeand its contamination rate and (2) to evaluate ifHPV DNA was detectable on a TVS probe whichwas used on patients with confirmed vaginal orcervical HPV infection despite following therecommended barrier method and disinfectionprocedure.

MATERIALS AND METHODSStudy designTwo independent cross-sectional studies wereconducted.

SettingThe studies were conducted in the ED of a teachinghospital in Hong Kong which has an annual EDattendance of around 150 000 persons. The study

1Accident and EmergencyMedicine Academic Unit, TheChinese University of HongKong, Hong Kong2Department of Microbiology,Faculty of Medicine, TheChinese University of HongKong, Hong Kong

Correspondence toDr Shuk Ting Christine Ma,Resident, Accident & EmergencyMedicine Academic Unit, TheChinese University of HongKong, 2/F, Main Clinical Blockand Trauma Centre, Prince ofWales Hospital, Shatin, N.T,Hong Kong;christinema524@gmail.com

Accepted 3 June 2012

Ma STC, Yeung AC, Chan PKS, et al. Emerg Med J (2012). doi:10.1136/emermed-2012-201407 1 of 4

Original article EMJ Online First, published on July 3, 2012 as 10.1136/emermed-2012-201407

Copyright Article author (or their employer) 2012. Produced by BMJ Publishing Group Ltd under licence.

took place between December 2011 and February 2012. It wasapproved by the Local Institutional Clinical Research EthicsCommittee.

MethodsOur departmental protocol calls for the use of a condom (53 mmplain, Pleasure Latex Products, Selangor Darul Ensan, Malaysia)as barrier protection for the TVS probe for every examination.Dry tissue paper was used to wipe away the excess gel on theprobe after condom removal, followed by the use of T-spray(Pharmaceutical Innovations, Inc., Newark, New Jersey, USA).T-spray is a bactericidal, fungicidal and virucidal disinfectingdetergent, specifically designed for ultrasound probes andmammography compressor plates. The TVS probe was then leftto dry in air for at least 5 min.

Specimen collection took place in the ultrasound room of theED which is equipped with one transvaginal transducer. Theinstrument was available for all ED medical staff to use whenclinically required. Cotton wool swabs were used to takesamples from the TVS transducer head and to take the highvaginal swabs.

Part 1: surveillance study on contamination of the TVS probeTwo sets of samples were taken daily at 08:00 and 20:00 h fora 2-month period at times when the instrument was not in use.Although there was no formal blinding, the specimen collectorswere unaware of how or by whom the instrument had beenused or whether the probe had been cleaned. The transducerhead up to 10 cm from the tip was swept in a 3608 fashion. Thecotton wool part was then stored in normal saline solution.

The samples were stored frozen until DNA extraction usingQIAamp DNA Mini Kit (Qiagen, Hilden, Germany). The pres-ence of HPV DNA in the samples was detected by PCR usingconsensus primers, PGMY09e11, which target a 450 bp regionof the L1 gene. This consensus PCR covers more than 40 types ofmucosal HPV. Briefly, the PCR was conducted in a 50 ml reactionmix containing 5 ml of extracted DNA, 200 mM deoxynucleotidetriphosphates, 0.06 mM of PGMY09 and PGMY11 primers, and1.25 m of HotStarTaq Plus polymerase (Qiagen). The cyclingconditions were as follows: activation of polymerase at 958C for5 min, 40 cycles of denaturation at 948C for 1 min, annealing at558C for 1 min and extension at 728C for 1 min, followed bya final extension at 728C for 8 min. Amplification was visualisedby agarose gel electrophoresis.

Part 2: serial samples from patients and the TVS probeThe investigator identified patients who presented with symp-toms of possible early pregnancy complications and requireda TVS as a component of routine clinical care. A total of fourswabs were taken as follows. First, a high vaginal swab from thepatient (swab 1) for HPV DNA detection was taken during thespeculum examination. A probe swab (swab 2) was taken beforepatient contact to rule out pre-existing probe contamination.After evaluation with TVS, the condom was removed andanother probe swab (swab 3) was taken. The probe disinfectionprocedure was then completed and the last probe swab (swab 4)was collected. Swabs 1e4 were sent for HPV DNA detectionusing PCR as described above.

PatientsPatients who were $18 years of age and confirmed to be preg-nant with a positive urine pregnancy test, with a gestational ageof #12 weeks, whose chief complaint was vaginal bleeding orabdominal pain were eligible for inclusion in the study. Exclu-

sion criteria included women with diseases causing immuno-suppression or currently on immunosuppressive agents andpatients with a confirmed spontaneous abortion (defined asproducts of gestation found in the genital tract on speculumexamination).

Statistical analysisc2 Test and Fisher ’s exact test were used to compare HPVpositivity and risk factors for categorical data. Continuousvariables were compared using the t test. All statistical testswere two-tailed and a p value of 0.05 or less indicated statisticalsignificance. Data were analysed with SPSS software V.20 forWindows.

RESULTSA total of 120 surveillance samples were collected over 60 days,of which nine (7.5%) were HPV DNA positive. Of the nine thatwere positive for HPV DNA, eight were collected in the first20 days of the study period and two were consecutive samples.The results are summarised in figure 1. Despite an averagecontamination rate of 7.5%, the positive results were unevenlydistributed. When we divided the study period into three parts(40 samples each), the contamination rate in the first period wasup to 20%.In the second part of the study, 78 patients were initially

recruited but two of them were subsequently excluded asproducts of gestation were identified on speculum examination.Of the remaining 76 patients, most of them were from HongKong, three from Mainland China and two patients were fromother countries. The mean age of the patients was 31.3 years(range 19e46 years) and their pregnancies had a median gesta-tional age of 7.8 weeks (range 4e12 weeks). Most of the womenwere non-smokers (79%, n¼60) and 4% (n¼3) were ex-smokers.In all, 55% (n¼42) of the women had had a normal cervicalsmear while 43% (n¼33) had never had a cervical smear done.One woman (2%) had a history of a low-grade squamous cellintraepithelial lesion. Only one woman reported a history ofsexually transmitted disease. The TVS revealed a viable fetus in41% (n¼31) of patients and 12% (n¼9) of fetuses were non-viable. The TVS showed an empty uterus and uncertain viabilityin 18% (n¼14) and 29% (n¼22) of patients, respectively.Out of 76 women, 14 (18.4%) had HPV DNA detected from

the high vaginal swab (swab 1). Their demographic data and theswab results are summarised in table 1. About 78.5% showed noHPV contamination on the TVS probe after the procedure

Figure 1 Result of surveillance transvaginal sonography probe samples(n¼120). HPV, human papillomavirus.

2 of 4 Ma STC, Yeung AC, Chan PKS, et al. Emerg Med J (2012). doi:10.1136/emermed-2012-201407

Original article

following standard disinfection. Three out of 14 last samples(swab 4) taken after disinfection were HPV DNA positive. Acomparison of the characteristics between HPV positive andHPV negative women is shown in table 2.

DISCUSSIONWe found that 7.5% of surveillance samples were HPV DNApositive. It showed that either the ED staff did not fully complywith disinfection procedures after every ultrasound examinationor the disinfection procedures did not remove all the HPV DNA.This finding confirms previous results that HPV DNA can stillbe found on medical instruments after sterilisation.6

HPV is a physically stable and resistant virus with longdurability.19 A recent in vitro study demonstrated that it cansurvive on a wet surface for at least 7 days, and carries aninfection ratio of 30%. Desiccation reduced its infectivity witha lower infection ratio of 10%.19

HPV is a very common infection. Our results showed an 18%carrier rate which is within the range of 10%e20% carrier ratereported worldwide.1 It is spread predominantly through sexualintercourse, although other routes of transmission have beenpostulated.6 20 A number of studies have demonstrated thatHPV DNA can also be found in the environment. A studyconducted in a genitourinary medicine clinic found mucosalHPV DNA present on the examination beds, the instrumenthandles and in the patients’ washrooms.21 Even after propersterilisation procedures, HPV DNA was still detected on the

medical instrument surfaces used in patients with confirmedHPV genital tract infection.6

The fact that most HPV DNA positivity concentrated in thefirst 20 days of the study period may be explained by theHawthorne effect. Medical staff became more alert and possiblymore thorough in cleaning the instrument (TVS probe) inresponse to learning that a study was going on. It seems likelythat increasing alertness among staff may have led to meticu-lous disinfection which reduced the contamination rate in latterpart of the study.In the second part of the study, three probe swabs were

positive for HPV DNA after examining HPV carrier patients.Condom breakage may be a possible explanation and the rate of3.9% as observed in this study is comparable with previousreports.18 On removal of the condom, the mucus or contami-nated gel may contact the probe surface if the condom is turnedinside out. Using a dry towel to wipe out the gel with anammonia based disinfectant (T-spray) may not be enough toremove the HPV DNA. After cleaning, the probe could also becontaminated by the environment or gloves which had previ-ously been in contact with the external genitalia.The detection of HPV DNA does not necessarily indicate the

presence of viable virus, and thus its infectivity is unknown. Todetermine the infectivity of HPV is technically difficult as thereis no feasible in vitro or in vivo culture system for HPV.Nevertheless, a good cleansing procedure should aim atremoving any viral DNA.

Table 1 Demographic data and swab results for HPV DNA positive patients

Age (years) Gestational age (weeks) Smoking status Previous pap smear TVS result* Swab 1y Swab 2 Swab 3 Swab 4

30 10 Non-smoker Normal VF ++ 41 10 Non-smoker Never NVF ++ 31 10 Non-smoker Never VF ++ 24 10 Non-smoker Never VF +++ 36 4 Smoker Normal UV +++ ++ 24 6 Non-smoker Normal VF +++ 29 7 Non-smoker Never NVF ++ 29 6 Non-smoker Normal EU +++ 46 11 Non-smoker Normal NVF ++ +

26 9 Non-smoker Never UV + 34 12 Non-smoker Normal VF + +

20 6 Smoker Normal EU + 23 10 Smoker Never UV + + +

27 8 Non-smoker Normal VF +

The HPV DNA PCR results were expressed semiquantitatively according to the intensity of the PCR amplicons.*VF, viable fetus with fetal heart pulsation seen; NVF, non-viable fetus, fetal heart pulsation absent for fetal pole >1 cm; UV, uncertain viability, intrauterine sac seen without fetal pole or absentfetal heart pulsation in a fetal pole <1 cm; EU, empty uterus, no intrauterine sac identified.ySwab 1, high vaginal swab of the patient; Swab 2, TVS probe swab before patient contact; Swab 3, TVS probe swab after procedure and before disinfection; Swab 4, TVS probe swab afterstandard disinfection.HPV, human papillomavirus; TVS, transvaginal sonography.

Table 2 Demographic data of study subjects according to the human papillomavirus (HPV) results of high vaginal samples

All (n[76) HPV positive (n[14) HPV negative (n[62)

Age (years) (mean 6SD) 31.365.48 3067.22 31.665.04 p¼0.34Age range (years) 19e46 20e46 19e44

Gestational age (weeks) (mean 6SD) 7.862.18 8.562.35 7.662.13 p¼0.16Normal previous pap smear 42 (55%) 8 (57%) 34 (55%) OR¼1.1 (95% CI 0.34 to 3.54), p¼1.0Smoker 13 (17%) 2 (14%) 11 (18%) OR¼0.77 (95% CI 0.15 to 3.9), p¼1.0Viable fetus 31 (41%) 6 (43%) 25 (40%) OR¼1.11 (95% CI 0.34 to 3.59), p¼1.0Non-viable fetus 9 (12%) 3 (21%) 6 (10%) OR¼2.5 (95% CI 0.55 to 11.7), p¼0.35Empty uterus 14 (18%) 2 (15%) 12 (19%) OR¼0.69 (95% CI 0.14 to 3.5), p¼1.0Uncertain viability 22 (29%) 3 (21%) 19 (31%) OR¼0.62 (95% CI 0.15 to 2.47), p¼0.75

c2 Test and Fisher exact test were adopted for categorical data. t Test was adopted for continuous numerical data (age and maternity).

Ma STC, Yeung AC, Chan PKS, et al. Emerg Med J (2012). doi:10.1136/emermed-2012-201407 3 of 4

Original article

A TVS probe is considered to be a semicritical item whichcontacts mucous membranes or non-intact skin, according tothe Centers for Disease Control and Prevention (CDC) guideline.This category of devices should be free from all microorganismswhile small numbers of bacterial spores are permissible. Probecovers are encouraged to reduce microbial contamination buthigh-level disinfection procedures should be followed. High leveldisinfectant formulations can contain glutaraldehyde, glutaral-dehyde with phenol/phenate, ortho-phthalaldehyde, hydrogenperoxide, and both hydrogen peroxide and peracetic acid.22 Theexposure times vary with different compounds. The T-spraywhich our ED is currently using is a quaternary ammonia baseddisinfectant which is recommended for non-critical equipmentused on intact skin according to the CDC guideline. This may bean explanation of the failure to clear all HPV DNA on the probesurface. We have subsequently changed our TVS disinfectantmethod to the Tristel TRIO wipes system (Tristel Solutions,Cambridge, UK) which is a chlorine dioxide based disinfectant.

We also observed that there was a significantly higher rate ofnon-viable fetuses found in the group of HPV carriers (OR 2.5). Aprevious study raised the possibility that HPV may be an aetio-logical agent of spontaneous abortions in early pregnancy.23

Although the pathophysiology of this viral infection withrespect to fetal pathology is not clear, our finding may addevidence to this observation. Future studies should aim at inves-tigating the association between HPV carriage and spontaneousabortion.

Given these findings, proper disinfection with high leveldisinfectant should be done after each TVS procedure. It is notpossible, on the basis of these results, to confirm or exclude thepossibility of acquiring HPV infection through contaminatedinstruments. Further studies may well be targeted at cutaneoussubtypes of HPV that could be transmitted through trans-abdominal ultrasound transducers; the prevalence of these maybe even higher.

Acknowledgements We would like to thank the medical and nursing staff of theEmergency Department at Prince of Wales Hospital for their help with this study.

Contributors MSTC, CSKP and CAG designed the study and reviewed and edited thefinal manuscript. MSTC did the data collection, literature review and wrote the firstdraft of the manuscript. CSKP supervised the laboratory work. YCMA performed thelaboratory work. CAG supervised the study overall. All authors have seen andapproved the final manuscript.

Funding The study was supported by the Research Fund from the Department ofMicrobiology, The Chinese University of Hong Kong.

Competing interests None.

Ethics approval Ethics approval was provided by Joint CUHK - New Territories EastCluster Clinical Research Ethics Committee.

Provenance and peer review Not commissioned; externally peer reviewed.

Data sharing statement There are no additional unpublished data from the study.

REFERENCES1. Koutsky LA, Galloway DA, Holmes KK. Epidemiology of genital human

papillomavirus infection. Epidemiol Rev 1988;10:122e63.2. Carter JR, Ding Z, Rose BR. HPV infection and cervical disease: a review. Aust N Z J

Obstet Gynaecol 2011;51:103e8.3. Winer RL, Lee SK, Hughes JP, et al. Genital human papillomavirus infection:

incidence and risk factors in a cohort of female university students. Am J Epidemiol2003;157:218e26.

4. Koutsky L. Epidemiology of genital human papillomavirus infection. Am J Med1997;102:3e8.

5. Schiffman M. Integration of human papillomavirus vaccination, cytology, and humanpapillomavirus testing. Cancer 2007;111:145e53.

6. Ferenczy A, Bergeron C, Richart RM. Human papillomavirus DNA in fomites onobjects used for management of patients with genital human papillomavirusinfections. Obstet Gynecol 1989;74:950e4.

7. Schlager D, Lazzareschi G, Whitten D, et al. A prospective study of ultrasonographyin the ED by emergency physicians. Am J Emerg Med 1994;12:185e9.

8. Ma OJ, Mateer JR, Ogata M, et al. Prospective analysis of a rapid trauma ultrasoundexamination performed by emergency physicians. J Trauma 1995;38:879e85.

9. Wong TW, Lau CC, Yeung A, et al. Efficacy of transabdominal ultrasoundexamination in the diagnosis of early pregnancy complications in an emergencydepartment. J Accid Emerg Med 1998;15:155e8.

10. Stein JC, Wang R, Adler N, et al. Emergency physician ultrasonography forevaluating patients at risk for ectopic pregnancy: a meta-analysis. Ann Emerg Med2010;56:674e83.

11. Adhikari S, Blaivas M, Lyon M. Diagnosis and management of ectopic pregnancyusing bedside transvaginal ultrasonography in the ED: a 2-year experience. Am JEmerg Med 2007;25:591e6.

12. McRae A, Murray H, Edmonds M. Diagnostic accuracy and clinical utility ofemergency department targeted ultrasonography in the evaluation of first-trimesterpelvic pain and bleeding: a systematic review. CJEM 2009;11:355e64.

13. O’Rourke D, Wood S. The early pregnancy assessment project: the effect ofcooperative care in the emergency department for management of early pregnancycomplications. Aust N Z J Obstet Gynaecol 2009;49:110e14.

14. Durston WE, Carl ML, Guerra W, et al. Ultrasound availability in the evaluation ofectopic pregnancy in the ED: comparison of quality and cost-effectiveness withdifferent approaches. Am J Emerg Med 2000;18:408e17.

15. Rotheray KR, Woo WW, Graham CA. An early pregnancy assessment cliniccombined with emergency physician ultrasound reduces admissions. Emerg MedAustralas 2010;22:194.

16. Storment JM, Monga M, Blanco JD. Ineffectiveness of latex condoms in preventingcontamination of the transvaginal ultrasound transducer head. South Med J1997;90:206e8.

17. Rooks VJ, Yancey MK, Elg SA, et al. Comparison of probe sheaths for endovaginalsonography. Obstet Gynecol 1996;87:27e9.

18. Milki AA, Fisch JD. Vaginal ultrasound probe cover leakage: implications for patientcare. Fertil Steril 1998;69:409e11.

19. Ding DC, Chang YC, Liu HW, et al. Long-term persistence of human papillomavirus inenvironments. Gynecol Oncol 2011;121:148e51.

20. Pao CC, Tsai PL, Chang YL, et al. Possible non-sexual transmission of genital humanpapillomavirus infections in young women. Eur J Clin Microbiol Infect Dis1993;12:221e2.

21. Strauss S, Sastry P, Sonnex C, et al. Contamination of environmental surfaces bygenital human papillomaviruses. Sex Transm Infect 2002;78:135e8.

22. Guideline for Disinfection and Sterilization in Healthcare Facilities. 2008. http://www.cdc.gov/hicpac/Disinfection_Sterilization/17_00Recommendations.html (accessed 20Mar 2012).

23. Hermonat PL, Han L, Wendel PJ, et al. Human papillomavirus is more prevalent infirst trimester spontaneously aborted products of conception compared to electivespecimens. Virus Genes 1997;14:13e17.

PAGE fraction trail=4

4 of 4 Ma STC, Yeung AC, Chan PKS, et al. Emerg Med J (2012). doi:10.1136/emermed-2012-201407

Original article

3. High Risk HPV Contamination of

Endocavity Vaginal Ultrasound

Probes: An Underestimated

Route of Nosocomial Infection? Jean S.Casalegno1*, Karine Le Bail

Carval2, Daniel Eibach1,3, MarieLaure

Valdeyron4, G.Lamblin2, Georges Mellier2,

Bruno Lina1, P.Gaucherand2, P.Mathevet2,

Yahia Mekki1* 1 Laboratory of Virology,

Hospices Civils de Lyon, Lyon, France, 2

Gynecology Obstetrics Dpt, Hospices

Civils de Lyon, Lyon, France, 3 European

Public Health Microbiology Training

Programme, European Centre for Disease

Prevention and Control, Stockholm,

Sweden, 4 Preventive Medecine Dpt,

Hospices Civils de Lyon, Lyon, France, –

Plos One 2012

10/25/12 PLOS ONE: High Risk HPV Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimat…

1/8www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0048137

High Risk HPV Contamination of Endocavity VaginalUltrasound Probes: An Underestimated Route ofNosocomial Infection?Jean­sebastien Casalegno1*, Karine Le Bail Carval2, Daniel

Eibach1,3, Marie­Laure Valdeyron4, Gery Lamblin2, Hervé

Jacquemoud5, Georges Mellier2, Bruno Lina1, Pascal Gaucherand2, Patrice Mathevet2, Yahia

Mekki1*

1 Laboratory of Virology, Centre de Biologie et Pathologie Est, Hospices Civils de Lyon, Lyon, France, 2Gynecology Obstetrics Departement, Hospices Civils de Lyon, Lyon, France, 3 European Public HealthMicrobiology Training Programme, European Centre for Disease Prevention and Control, Stockholm,Sweden, 4 Preventive Medecine Departement, Hospices Civils de Lyon, Lyon, France, 5 TechnicalSupport Departement, Hospices Civils de Lyon, Lyon, France

Abstract

Background

Endocavity ultrasound is seen as a harmless procedure and has become a common gynaecological

procedure. However without correct disinfection, it may result in nosocomial transmission of

genito­urinary pathogens, such as high­risk Human Papillomavirus (HR­HPV). We aimed to

evaluate the currently recommended disinfection procedure for covered endocavity ultrasound

probes, which consists of “Low Level Disinfection” (LLD) with “quaternary ammonium compounds”

containing wipes.

Methods

From May to October 2011 swabs were taken from endovaginal ultrasound probes at the

Gynecology Department of the Lyon University Hospital. During the first phase (May–June 2011)

samples were taken after the ultrasound examination and after the LLD procedure. In a second

phase (July–October 2011) swab samples were collected just before the probe was used. All

samples were tested for the presence of human DNA (as a marker for a possible transmission of

infectious pathogens from the genital tract) and HPV DNA with the Genomica DNA microarray (35

different HPV genotypes).

Results

We collected 217 samples before and 200 samples after the ultrasound examination. The PCR was

inhibited in two cases. Human DNA was detected in 36 (18%) post­examination samples and 61

(28%) pre­examination samples. After the ultrasound LLD procedure, 6 (3.0%) samples contained

HR­HPV types (16, 31, 2×53 and 58). Similarly, HPV was detected in 6 pre­examination samples

(2.7%). Amongst these 4 (1.9%) contained HR­HPV (types 53 and 70).

RESEARCH ARTICLE

10/25/12 PLOS ONE: High Risk HPV Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimat…

2/8www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0048137

Conclusion

Our study reveals that a considerable number of ultrasound probes are contaminated with human

and HR­HPV DNA, despite LLD disinfection and probe cover. In all hospitals, where LLD is

performed, the endovaginal ultrasound procedure must therefore be considered a source for

nosocomial HR­HPV infections. We recommend the stringent use of high­level disinfectants, such

as glutaraldehyde or hydrogen peroxide solutions.

Citation: Casalegno J­s, Le Bail Carval K, Eibach D, Valdeyron M­L, Lamblin G, et al. (2012) High Risk HPV

Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimated Route of Nosocomial Infection? PLoS

ONE 7(10): e48137. doi:10.1371/journal.pone.0048137

Editor: Rui Medeiros, IPO, Inst Port Oncology, Portugal

Received: August 2, 2012; Accepted: September 20, 2012; Published: October 24, 2012

Copyright: © 2012 Casalegno et al. This is an open­access article distributed under the terms of the Creative

Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided

the original author and source are credited.

Funding: This study has been funded by Germitec. Study sponsors had no involvement in the study design, the

sample collection, the analysis, the interpretation of data, the writing of the manuscript and in the decision to submit

the manuscript for publication. None of the authors have any participation or involvement in the Germitec society.

Competing interests: The authors have declared that no competing interests exist.

* E­mail: jean­sebastien.casalegno@chu­lyon.fr (JSC); yahia.mekki@chu­lyon.fr (YM)

INTRODUCTION

Human papillomavirus (HPV) is now recognized as the major etiological cause of invasive cervical

cancer and cervical intraepithelial neoplasia worldwide [1]–[3]. More than 100 human HPV genotypes,

classified into high risk (e.g. HPV 16, 18) and low risk types (e.g. HPV 6, 11) [2], are known to infect

the anogenital tract [4]. The high risk genotypes have been found to be closely associated with

cervical cancer [3], [4] and high grade cervical intraepithelial neoplasia. Worldwide it has been

estimated that 70% of cervical cancers are due to HPV types 16 and 18 [1]. Genital HPV infection is

mainly a sexual transmitted disease [3], however perinatal transmission from mother to child has

been shown as well [5]. So far little is known on a possible nosocomial source of HPV infection.

HPV has the potential for nosocomial transmission due to high resistance and persistence in the

environment. It has been shown that HPV retains 30% of its infectivity, even after dehydration for 7

days [6]. More recently, a study has reported that HPV 16 remains infectious for at least 7 days on a

10/25/12 PLOS ONE: High Risk HPV Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimat…

3/8www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0048137

wet surface [7]. These results suggest that fomites represent a possible nosocomial source of HPV

infection [8]. Indeed the detection of HPV on medical equipment, such as forceps and cryoprobe tips

has been reported, despite proper disinfection procedure [9].

Endovaginal cavity ultrasound, seen as a rapid and completely harmless procedure, has become a

common medical diagnostic tool. The close contact between the probe and the cervix uteri or vaginal

wall may represent a potential vector for sexual transmitted infections, such as HPV. In 2010 Kac

noticed HPV contamination of endovaginal and endorectal ultrasound probes, despite usage of specific

probe covers [10]. French national disinfection guidelines classify covered endovaginal ultrasound

probes as non­critical devices and therefore undergo low level disinfection with products such as

quaternary ammonium compounds [11]. The use of high level disinfection has been questioned,

because of the potential of shortening the life of the transducer, the increased toxicity and a more

time intensive disinfection procedure. The objective of our study was to investigate the proportion of

contaminated endovaginal ultrasound probes with HPV in order to evaluate the antimicrobial efficacy of

the current standard disinfection procedure.

MATERIALS AND METHODS

Study settings

We performed a prospective study, conducted during two periods in one ward of the gynaecology

department of the Lyon University Hospital “Femme Mère Enfant”. Permission to enter the wards and

collect the samples has been granted by the head­chief of the gyneco­obstetric department. No patient

information of any kind has been gathered and no human samples were tested in this study, therefore

no patient consent was required by the local ethical committee. During the first period, from the 2nd of

May to the 1stof July 2011, probe samples were obtained immediately after the endovaginal ultrasound

was performed and the probe was disinfected. In the second sampling period, from the 2nd of July to

the 10th of October 2011, we assessed a potential risk of HPV transmission to the next patient, with

samples being collected just before the probe was used on a new patient. The mean time interval to

the previous disinfection procedure was 71 minutes (range: 16 minutes to 8 hours).

Standard disinfection procedure

The disinfection of the probes was performed by nurses under supervision of the technician in charge

of the sampling. Before the study all participating nurses were specifically trained on how to handle

the probe in order to avoid contamination via hands or the cover itself. Probes were used with a

disposable probe cover (93/42/EEC CE mark) in compliance with national health recommendations

[11]. After the examination, the probe cover was carefully removed without contaminating the probe

and the probe was disinfected with low level disinfection wipes (Sani­Cloth Active), containing

“quaternary ammonium compounds”.

Sampling

Samples were taken from two endocavity ultrasound probes used on two ultrasound machines

10/25/12 PLOS ONE: High Risk HPV Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimat…

4/8www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0048137

(General Electric, VOLUSON­Vaginal Probe RIC5­9D). Probe samples were collected by a specifically

trained technician, no more than 15 minutes after the end of the disinfection process. The sample was

taken with a dry swab, applied lengthwise across the surface of the probe. The swab was then

immediately suspended in a viral transport medium (EMEM) and send to the microbiological lab. Delay

between sampling and lab processing did not exceed four hours.

DNA extraction, amplification and genotyping

The samples were sent to the virology laboratory within less than four hours and stored at −20°C. All

samples were extracted using a NucliSens easyMAG instrument (Biomerieux, Marcy L'étoile, France)

according to the manufacturer's recommended procedures. HPV amplification and genotyping were

conducted using the Genomica SAU (Genomica, Spain) microarray test [12]. This assay allows

hybridization of amplified and biotinylated 450 bp fragments from the L1 region of HPV virus. This

region is highly conserved and specific for each HPV type [4]. The assay detects 35 HPV types: 20 high

risk (HPV 16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 70, 73, 82, 85) and 15 low risk

(HPV 6, 11, 40, 42, 43, 44, 54, 61, 62, 71, 72, 81, 83, 84, 89) [11]. Amplification and detection of a

human house­keeping gene (cystic fibrosis transmembrane conductance regulator gene) is conducted

in each assay in order to ensure sample quality.

Extractions and PCRs for all samples were run in specific series to avoid any risk of contamination

from clinical samples. Each series included a negative control in order to test for contamination during

the extraction procedure. One swab of each batch, as well as the EMEM medium batch was tested for

the absence of HPV contamination.

Analysis

Statistical analysis was performed with EpiInfo software (V 3.5.1 CDC). Data were expressed as the

percentages of positive sample. Inhibited samples were excluded from the calculation.

RESULTS AND DISCUSSION

In our study, no break of the probe covers was detected in any case, and visual inspection did not

reveal any presence of blood or body fluids on the probe.

During the first study period from May, 2nd to July 1st 200 probes samples were obtained after

endovaginal use and standard disinfection. The PCR was inhibited in two cases. Of the remaining 198

samples, 7 samples (3.5%) were HPV positive. 1 sample showed low risk HPV and 6 samples (3.0%)

were positive for at least one high risk HPV type. The detected HPV types include HPV 53, HPV 16, HPV

58 and HPV 31. Two samples revealed more than one HPV type, with a maximum of 4 different HPV

types per sample being detected (Table 1). When investigating the sample order, it appeared that

three out of four subsequent samples were positive for HPV 58. This observation suggests a persistent

probe contamination with the same HR­HPV type, despite the application of three disinfection

procedures.

10/25/12 PLOS ONE: High Risk HPV Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimat…

5/8www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0048137

Table 1. Results of the 13 Sets of Samples forwhich HPV was Isolated from UltrasoundProbes after Removal of the Probe Cover underRoutine Conditions.

doi:10.1371/journal.pone.0048137.t001

Download: PowerPoint slide | larger image (47KB PNG) | original image (252KB TIFF)

In the second study period, from July, 2nd to October, 10th 217 samples were collected before the

ultrasound examination. One PCR was inhibited. We detected HPV in six out of 216 (2.8%) samples.

Four samples were positive with the HR­HPV types HPV 53 and HPV 70 and two samples revealed LR­

HPV. We did not detect more than one HPV per sample (Table 1).

The test sample quality control (use for in vitro diagnosis), based on a human housekeeping gene,

detected the trace of human DNA in 63/216 (29.2%) and 39/198 (19.7%) samples, before and after

the probe use.

This study shows a probe contamination with high risk oncogenic HPV DNA after low level disinfection,

despite the use of probe covers. We observed a 2.2% rate of HR­HPV contamination of probes (1.8%

before use and 2.5% after use). This result is in agreement with the results observed by Kac G et al

[10], who reported 8.2% (95% CI, 4.0%–15.5%) of vaginal­rectal ultrasound probe covers

contaminated with HPV, and 0.9% (3/336) contaminated after retrieval of the probe cover, but before

low level disinfection (LLD). In a recent publication, Ma ST detects a proportion of 7.5% probes

positive for HPV DNA [13]. However this study did not assess the HPV genotype. Our study therefore

adds to the evidence, that HR­HPV DNA on endovaginal probes can be routinely detected after low­

level disinfection and despite covering of the probe.

The high rate of human DNA on the probes most likely represents human DNA from cells of the genital

tract. We tried to avoid any hand contact with the probe during the sampling and the ultrasound

examination, but it cannot be ruled out completely. The presence of human DNA from the genital tract

on the ultrasound probe can be seen as a marker for a contamination with potential pathogens,

attached to cells of the genital tract.

This study may have several limitations. The HPV detection method is based on nucleic acid detection

and does not detect the presence of infectious virus particles. The HPV diagnostic faces the same

problem, as no HPV cell culture is available up to date. Therefore it is not possible to assess the

infectious potential of HPV. Although we are not able to conclude on the infectivity of the detected

HPV, considering the potential for HPV resistance on fomites [7] as well as the close and prolonged

contact of the probe with the cervix, the detection of HR­HPV DNA on the probe raises concerns.

This risk of swab or probe contamination, due to hand contact was a main concern during the study.

Necessary precautions (e.g. training of the sampling staff, use of gloves) were taken to avoid this

identified risk, therefore direct HPV contamination of the swab is very unlikely. However, despite the

10/25/12 PLOS ONE: High Risk HPV Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimat…

6/8www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0048137

use of gloves, we cannot definitely exclude manual contamination of the probe during probe handling.

The main HPV type detected in our study is HPV 53. This high detection rate of HPV 53 is a main

feature of the genital HPV genotype distribution of our population in Lyon, as reported in a previous

study (12). On the other hand HPV 16, the most prevalent HR­HPV genotype, is detected less than

expected. Given the limited number of HPV detected in the study, the HPV genotype distribution should

be interpreted cautiously.

The use of dry swabs for sampling could have resulted in a loss of sensitivity. We tried to avoid the

problem, by suspending the swab into a viral transport medium (EMEM) and transport the samples into

the lab within four hours.

This study was carried out in an almost standard disinfection situation. The staff was aware of the

study and this may have strengthened the compliance to the disinfection procedures. Under different

circumstances, such as emergency units or wards with a high frequency ultrasound use, compliance

may be lower and HPV detection therefore likely to be higher.

The study focused on the detection of HPV on the ultrasound probe. Patients were not sampled in the

study, as the main objective was to evaluate the disinfection procedure for covered endocavity

ultrasound probes and not routes of transmission.

Presently the French disinfection guidelines [11] allow to classify endovaginal ultrasound probes as

non­critical devices, as a probe cover is used and changed for every patient. Non­critical medical

equipment is low­level disinfected with substances, such as quaternary ammonium compounds or

phenolics. They are not effective against non­enveloped viruses, such as HPV, fungi and bacterial

spores [14], [15], [16]. Taken into account the high rate of perforations for commercially available

probe covers and to a lesser extent for condoms the “CDC Guideline for Disinfection and Sterilization

in Healthcare Facilities” [17] clearly recommends to categorize endovaginal ultrasound probes as

semicritical devices. This change in the category implies the use of high­level disinfectants, which also

eliminate non­enveloped viruses. Chemical high­level disinfectants, such as hydrogen peroxide,

glutaraldehyde or peracetic acid or non­chemical alternatives, such as UVC light can be used [17],

[18]. This study was conducted in cooperation with the Hygiene department of the hospital. The results

were immediately reported to the responsible authorities. Upon this study it was decided to change the

disinfection procedure for endovaginal ultrasound probes from a low­level disinfection to a high­level

disinfection procedure. Momentarily a high­level disinfection with UVC light is under evaluation.

In conclusion the study shows a considerable number of endovaginal ultrasound probes being

contaminated with HPV after the use of low­level disinfection. Low­level disinfection seems to be

inefficient for preventing contamination of endocavity ultrasound probes with HR­HPV and pose a risk

of nosocomial transmission of HR­HPV during an ultrasound procedure. We therefore recommend the

use of high­level disinfection procedures after each endovaginal ultrasound examinations.

ACKNOWLEDGMENTS

We are grateful to the nurses, medics and technicians for their support.

10/25/12 PLOS ONE: High Risk HPV Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimat…

7/8www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0048137

AUTHOR CONTRIBUTIONS

Conceived and designed the experiments: YM HJ MLV KLBC. Performed the experiments: KLBC GL GM

PG PM. Analyzed the data: JSC DE YM. Contributed reagents/materials/analysis tools: HJ YM. Wrote

the paper: JSC DE YM BL.

REFERENCES

1. Bosch FX, Manos MM, Munoz N, Sherman M, Jansen AM, et al. (1995) Prevalence of human papillomavirus in

cervical cancer: a worldwide perspective. J Natl Cancer Inst 87: 796–802.

2. Munoz N, Bosch FX, de Sanjose S, Shah KV (1994) The role of HPV in the etiology of cervical cancer. Mutat Res

305: 293–301.

3. Munoz N, Castellsague X, de Gonzalez AB, Gissmann L (2006) Chapter 1: HPV in the etiology of human cancer.

Vaccine 24 Suppl 3S3/1–10.

4. zur Hausen H (2002) Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer 2:

342–50.

5. Martinelli M, Zappa A, Bianchi S, Frati E, Colzani D, et al. (2012) Human papillomavirus (HPV) infection and

genotype frequency in the oral mucosa of newborns in Milan, Italy. Clin Microbiol Infect 18: E197–9.

6. Roden RB, Lowy DR, Schiller JT (1997) Papillomavirus is resistant to desiccation. J Infect Dis 176: 1076–9.

7. Ding DC, Chang YC, Liu HW, Chu TY (2011) Long­term persistence of human papillomavirus in environments.

Gynecol Oncol 121: 148–51.

8. Ferenczy A, Bergeron C, Richart RM (1989) Human papillomavirus DNA in fomites on objects used for the

management of patients with genital human papillomavirus infections. Obstet Gynecol 74: 950–4.

9. Gage JC, Schiffman M, Solomon D, Wheeler CM, Castle PE (2010) Comparison of measurements of human

papillomavirus persistence for postcolposcopic surveillance for cervical precancerous lesions. Cancer Epidemiol

Biomarkers Prev 19: 1668–74.

10. Kac G, Podglajen I, Si­Mohamed A, Rodi A, Grataloup C, et al. (2010) Evaluation of ultraviolet C for disinfection

of endocavitary ultrasound transducers persistently contaminated despite probe covers. Infect Control Hosp

Epidemiol 31: 165–70.

11. Haut Conseil de la Santé Publique (2007) Gaines de Protection à usage unique pour dispositifs médicaux

réutilisables: recommandations d'utilisation. French. Haut Conseil de la Santé Publique website. Available:

http://www.hcsp.fr/docspdf/avisrapports/hcspr20071214_GainesProtec.pdf. Accessed 2012 September 28.

12. Casalegno JS, Benchaib M, Le Bail Carval K, Piaton E, Mathevet P, et al. (2011) Human papillomavirus genotype

distribution among French women with and without cervical abnormalities. Int J Gynaecol Obstet 114: 116–9.

13. Ma ST, Yeung AC, Chan PK, Graham CA (2012) Transvaginal ultrasound probe contamination by the human

papillomavirus in the emergency department. Emerg Med J.

14. Milki AA, Fisch JD (1998) Vaginal ultrasound probe cover leakage: implications for patient care. Fertil Steril 69:

10/25/12 PLOS ONE: High Risk HPV Contamination of Endocavity Vaginal Ultrasound Probes: An Underestimat…

8/8www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0048137

All site content, except where otherwise noted, is licensed under a Creative Commons Attribution License.

409–11.

15. Storment JM, Monga M, Blanco JD (1997) Ineffectiveness of latex condoms in preventing contamination of the

transvaginal ultrasound transducer head. South. Med. J. 90: 206–8.

16. Hignett M, Claman P (1995) High rates of perforation are found in endovaginal ultrasound probe covers before

and after oocyte retrieval for in vitro fertilization­embryo transfer. J. Assist. Reprod. Genet. 12: 606–9.

17. Center for Disease Control and Prevention (2008) Guideline for Disinfection and Sterilization in Healthcare

Facilities. Center for Disease Control and Prevention website. Available:

www.cdc.gov/hicpac/pdf/guidelines/Disinfection_Nov_2008.pdf. Accessed 2012 July 30.

18. Bloc S, Mercadal L, Garnier T, Komly B, Leclerc P, et al. (2011) Evaluation of a new disinfection method for

ultrasound probes used for regional anesthesia: ultraviolet C light. J Ultrasound Med 30: 785–8.

4. Medical Device Alert – MHRA -

2012

Medical Device Alert

Ref: MDA/2012/037 Issued: 28 June 2012 at 14:00

Device

Reusable transoesophageal echocardiography, transvaginal and transrectal ultrasound probes (transducers). All models. All manufacturers.

Problem Action

The MHRA is aware of an incident where the death of a patient from hepatitis B infection may have been associated with a failure to appropriately decontaminate a transoesophageal echocardiography probe between each patient use. The MHRA is issuing this alert to advise users to appropriately decontaminate all types of reusable ultrasound probes.

Review, and if necessary update, local procedures for all ultrasound probes that are used within body cavities to ensure that they are decontaminated appropriately between each patient use, in accordance with the manufacturer’s instructions. Ensure that staff who decontaminate medical devices are appropriately trained and fully aware of their responsibilities.

Action by

Trust decontamination leads. Healthcare professionals using these devices and staff responsible for reprocessing medical devices.

Be aware of the MHRA’s guidance document ‘Managing Medical Devices’ (available from our website www.mhra.gov.uk). Be aware of the Department of Health’s publications (England only): Choice Framework for local Policy and Procedures 01-06 – Decontamination of flexible endoscopes: Operational management manual 13536:1.0. Available from Space for Health, sign-in required:

CAS deadlines

http://www.spaceforhealth.nhs.uk/England/topics/choice-framework-local-policy-and-protocols-01-06-%E2%80%93-decontamination-flexible-endoscopes

Action underway: 11 July 2012 Also be aware of similar advice as/when published by the devolved administrations.

Action complete: 19 July 2012

Note: These deadlines are for systems to be in place to ensure the actions are undertaken.

Medicines and Healthcare products Regulatory Agency Page 1 of 4

Issued: 28 June 2012 at 14:00 Ref: MDA/2012/037

Distribution This MDA has been sent to: • NHS trusts in England (Chief Executives) • Care Quality Commission (CQC) (Headquarters) for information • HSC trusts in Northern Ireland (Chief Executives) • NHS boards in Scotland (Equipment Co-ordinators) • Local authorities in Scotland (Equipment Co-ordinators) • NHS boards and trusts in Wales (Chief Executives) • Primary care trusts in England (Chief Executives)

Onward distribution Please bring this notice to the attention of relevant employees in your establishment. Below is a suggested list of recipients. Trusts CAS and SABS (NI) liaison officers for onward distribution to all relevant staff including: • Adult intensive care units • All wards • Anaesthetists • Coronary care departments

• A&E departments

• A&E departments

• Cardiologists

• Clinical governance leads

• Colposcopy departments

• Day surgery units

• Decontamination leads

• Directors of infection prevention and control

• Endoscope reprocessing units

• Endoscopy units

• Gastroenterology departments

• General surgery

• Gynaecology departments

• Gynaecology nurses

• Health and safety managers

• Infection control departments

• Infection control nurses

• Infection prevention and control directors

• Intensive care nursing staff (adult)

• Intensive care units

• Microbiologists

• Outpatients

• Radiographers

• Radiologists

• Radiology departments

• Risk managers

• Sonographers

• Sterile services departments

• Theatres

• Ultrasound departments

• Urologists

Primary care trusts CAS liaison officers for onward distribution to all relevant staff including:

• Community hospitals

• Infection control teams

Independent distribution Establishments registered with the Care Quality Commission (CQC) (England only) This alert should be read by:

• Hospitals in the independent sector

• Independent treatment centres

Medicines and Healthcare products Regulatory Agency Page 2 of 4

Issued: 28 June 2012 at 14:00 Ref: MDA/2012/037

Please note: CQC and OFSTED do not distribute these alerts. Independent healthcare providers and social care providers can sign up to receive MDAs directly from the Department of Health’s Central Alerting System (CAS) by sending an email to: safetyalerts@dh.gsi.gov.uk and requesting this facility.

England If you are in England, please send enquiries about this notice to the MHRA, quoting reference number MDA/2012/037 or 2011/007/026/081/015

Technical aspects John McManus or Sharon Knight Medicines & Healthcare products Regulatory Agency Floor 4 151 Buckingham Palace Road London SW1W 9SZ

Tel: 020 3080 7226 or 020 3080 7202 Fax: 020 8754 3965

Email: john.mcmanus@mhra.gsi.gov.uk sharon.knight@mhra.gsi.gov.uk

Clinical aspects Nicola Lennard Medicines & Healthcare products Regulatory Agency Floor 4 151 Buckingham Palace Road London SW1W 9SZ

Tel: 020 3080 7126 Fax: 020 8754 3965

Email: nicola.lennard@mhra.gsi.gov.uk

How to report adverse incidents Please report via our website http://www.mhra.gov.uk

Further information about CAS can be found at https://www.cas.dh.gov.uk/Home.aspx

Northern Ireland Alerts in Northern Ireland will continue to be distributed via the NI SABS system.

E

nquiries and adverse incident reports in Northern Ireland should be addressed to:

Northern Ireland Adverse Incident Centre Health Estates Investment Group Room 17 Annex 6 Castle Buildings Stormont Estate Dundonald BT4 3SQ

Tel: 02890 523 704 Fax: 02890 523 900

Email: NIAIC@dhsspsni.gov.uk

http://www.dhsspsni.gov.uk/index/hea/niaic.htm

How to report adverse incidents in Northern Ireland Please report directly to NIAIC, further information can be found on our website http://www.dhsspsni.gov.uk/niaic

Further information about SABS can be found at http://sabs.dhsspsni.gov.uk/

Medicines and Healthcare products Regulatory Agency Page 3 of 4

Issued: 28 June 2012 at 14:00 Ref: MDA/2012/037

Scotland Important note: For decontamination advice in Scotland, contact: HFS Decontamination Team Email: nss.hfsdeconteam@nhs.net Tel: 0141 207 1857

Enquiries and adverse incident reports in Scotland should be addressed to:

Incident Reporting and Investigation Centre Health Facilities Scotland NHS National Services Scotland Gyle Square 1 South Gyle Crescent Edinburgh EH12 9EB

Tel: 0131 275 7575

Fax: 0131 314 0722

Email: nss.iric@nhs.net

http://www.hfs.scot.nhs.uk/online-services/incident-reporting-and-investigation-centre-iric/

Wales Enquiries in Wales should be addressed to: Improving Patient Safety Team Medical Directorate Welsh Government Cathays Park Cardiff CF10 3NQ

Tel: 029 2082 3922

Email: Haz-Aic@wales.gsi.gov.uk

MHRA is an executive agency of the Department of Health © Crown Copyright 2012

Addressees may take copies for distribution within their own organisations

Medicines and Healthcare products Regulatory Agency Page 4 of 4

5. Survey of microbial

contamination of ultrasound

probe handles following high

level dinsfection with ortho-

phtaladehyde – G.R. McNally, A.Ngu – Medical Imaging, Royal Hospital for

Women, Sydney, NSW, Australia - 2013

6–9 October 2013, Sydney, Australia Oral communication abstracts

Results: Type of previa, lacunes, vascularity, history of Cesareansection or placenta previa correlated significantly with peripartumcomplications. However, there was no independent predictivefactor for peripartum complications in multivariable analysis. Whenwe formulated scoring system including type of previa, lacunes,vascularity, history of Cesarean section and placenta previa, allpatients with total score ≥ 6 needed Cesarean hysterectomy. Whentotal score were ≥ 4, 37.5% of patients needed uterine arteryembolization or Cesarean hysterectomy.Conclusions: Scoring system based on ultrasonographic andclinical factors would be useful for the prediction of peripartumcomplications in pregnancies complicated by placenta previa.

OC20.04: Table 1. Peripartum complication according to score ofrisk factors

score No transfusion embolization hysterectomy

0-3 56 8 (14.3%) 0 04-5 12 4 (33.3%) 2 (16.7%) 06-7 4 4 (100%) 0 4 (100%)sum 72 16 2 4

OC20.05Survey of microbial contamination of ultrasound probehandles following high-level disinfection withortho-phthalaldehyde

G.R. McNally, A. Ngu

Medical Imaging, Royal Hospital for Women, Sydney, NSW,Australia

Objectives: Intracavity ultrasound probes require high-leveldisinfection between patients to prevent transmission of infection.Ultrasound probe manufacturers advise against soaking probe han-dles in liquid high-level disinfectants such as ortho-phthalaldehyde(OPA) as this may result in damage to the probe. Probe handles aretherefore not routinely disinfected and this may provide a route fortransmission of infectious agents between patients and healthcareworkers. This study surveyed the bacterial contamination ofultrasound probe handles in two Australian ultrasound clinics.Methods: Transvaginal ultrasound probes were used in routineclinical procedures before being disinfected with OPA according tothe manufacturer’s instructions. Following disinfection, ultrasoundprobe handles were sampled for bacterial contamination to establishcolony counts and species identification.Results: Of the 100 probe handles tested, 93 showed contaminationwith bacteria. The majority of organisms recovered were Staphylo-coccus spp and smaller numbers of Bacillus spp, Streptococcus sppand Pseudomonas spp were also identified. The clinically importantpathogen Staphylococcus aureus was recovered from 11 samples.Conclusions: Ultrasound probe handles can become contaminatedduring routine clinical use following high-level disinfection withOPA. Contaminating organisms typically include skin flora andenvironmental bacteria but some are potentially serious pathogens.Technologies that enable the disinfection of both the ultrasoundprobe and handle may help to reduce the risk of transmissionbetween patients and healthcare workers.

OC20.06Usefulness of uterine artery Doppler velocimetry indiagnosing placenta accreta in placenta previa patients

H. Cho1, H. Hwang2, J. Kwon1, Y. Park1, Y. Kim1

1Department of Obstetrics and Gynecology, YonseiUniversity Health System, Seoul, Republic of Korea;2Department of Obstetrics and Gynecology, KonkukUniversity School of Medicine, Seoul, Republic of Korea

Objectives: To evaluate the efficacy of uterine artery Dopplervelocimetry in diagnosing placenta previa accreta.Methods: Clinical records of all deliveries between April 1991through March 2013 were retrospectively analyzed. Cases withIUGR, pre-eclampsia, multiple pregnancies, fetal anomalies, chro-mosomal abnormalities, and maternal medical illnesses wereexcluded. Total of 18,337 cases were included in the study, ofwhich 421 had placenta previa without accreta and 52 placenta pre-via with accreta, with histologic confirmation. All cases underwentuterine artery Doppler velocimetry to measure the mean resistanceindex (RI) at third trimester. Age, parity, previous abortion history,previous Cesarean section, gestational age at delivery, gender andbirth weight were included for comparison.Results: Significant differences were found for age (P= <0.0001),parity (P= <0.0001), previous abortion (P= <0.0001) , previousCesarean section (P= <0.0001) , gestational age at delivery (P=<0.0001), and birth weight (P= <0.0001) with control (n= 18,337)and placenta previa group ( n= 473). Uterine artery Doppler MeanRI was similar between two groups.

Multiple logistic regression was used to identify independentrisk factors for placenta previa with accreta. No difference wereobserved in all variable except Uterine artery Doppler Mean RIbetween placenta previa without accreta (n =421) and placentaprevia with accreta ( n =52). Uterine artery Doppler Mean RI lesserthan 0.40 (OR 3.39 ; 95% Cl 1.41, 8.16) was associated withincreased risk of having placenta previa with accreta.Conclusions: This study suggests that uterine artery Doppler RI isreduced in placenta previa with accreta patients when comparedwith placenta previa without accreta or normal pregnancy. Whenuterine artery Doppler mean RI is lesser than 0.40 there is 3.39-foldincreased risk of having placenta previa with accreta. Consequently,uterine artery Doppler velocimetry can be a useful tool for predictingplacenta accreta.

OC20.07Lower uterine segment in the first trimester of pregnancy: aquality based approach

L.J. Salomon1,2, O. Castaing2, M. Kuleva1, J. Bernard1,N. Fries2, M. Fontanges2, G. Haddad2, R. Mangione2,P. Boukobza2, D. Combourieu2, P. Bussiere2, M. Le Gac2,P. Coquel2, P. Billen2

1Maternite, Hopital Necker Enfants Malades, AP-HP,Universite Paris Descartes, Paris, France; 2College Francaisd’Echographie Foetale, Paris, France

Objectives: Transvaginal ultrasound assessment of the lower uterinesegment (LUS) performed as early as first trimester may providerelevant information regarding obstetrical outcome. We aimed tostandardize the evaluation of the lower uterine segment and todevelop a score-based quality control LUS images in the firsttrimester of pregnancy.Methods: This was a ‘‘flash’’ study carried out by the FrenchUltrasound College for a 15 day-period in 2012. Obstetrician-sonographers from routine ultrasound units all over the countrywere recruited by e-mail after a short internet-based teachingprogram. LUS was evaluated based on a predefined standardized USsagittal plane obtained by transvaginal ultrasound examination. All

Ultrasound in Obstetrics & Gynecology 2013; 42 (Suppl. 1): 1–47. 41

6. Impact of Vaginal-Rectal

Ultrasound Examinations with

Covered and Low-Level Disinfected

Transducers on Infectious

Transmissions in France - Sandrine

Leroy, Fatima M’Zali, Michael Kann, David J.

Weber and David D. Smith - Infection Control

and Hospital Epidemiology - The University

of Chicago Press on behalf of The Society for

Healthcare Epidemiology of America – 2014

Impact of Vaginal-Rectal Ultrasound Examinations with Covered and Low-Level DisinfectedTransducers on Infectious Transmissions in FranceAuthor(s): Sandrine Leroy, Fatima M’Zali, Michael Kann, David J. Weber and David D. SmithSource: Infection Control and Hospital Epidemiology, (-Not available-), p. 000Published by: The University of Chicago Press on behalf of The Society for Healthcare Epidemiologyof AmericaStable URL: http://www.jstor.org/stable/10.1086/678604 .

Accessed: 07/11/2014 01:45

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

The University of Chicago Press and The Society for Healthcare Epidemiology of America are collaboratingwith JSTOR to digitize, preserve and extend access to Infection Control and Hospital Epidemiology.

http://www.jstor.org

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

infection control and hospital epidemiology december 2014, vol. 35, no. 12

o r i g i n a l a r t i c l e

Impact of Vaginal-Rectal Ultrasound Examinations with Coveredand Low-Level Disinfected Transducers on Infectious

Transmissions in France

Sandrine Leroy;1 Fatima M’Zali;2 Michael Kann;2,3 David J. Weber;4 David D. Smith5

background. The risk of cross-infection from shared ultrasound probes in endorectal and vaginal ultrasonography due to low-leveldisinfection (LLD) is difficult to estimate because potential infections are also sexually transmitted diseases, and route of contamination isoften difficult to establish. In France, the widely used standard for prevention of infections is through the use of probe covers and LLDof the ultrasound transducer by disinfectant wipes. We performed an in silico simulation based on a systematic review to estimate thenumber of patients infected after endorectal or vaginal ultrasonography examination using LLD for probes.

study design. We performed a stochastic Monte Carlo computer simulation to produce hypothetical cohorts for a population of 4million annual ultrasound examinations performed in France, and we estimated the number of infected patients for human immunode-ficiency virus (HIV), herpes simplex virus, hepatitis B virus, hepatitis C virus, human papilloma virus, cytomegalovirus, and Chlamydiatrachomatis. Modeling parameters were estimated by meta-analysis when possible.

results. The probability of infection from a contaminated probe ranged from 1% to 6%, depending on the pathogen. For cases ofHIV infection, this would result in approximately 60 infected patients per year. For other common viral infections, the number of newcases ranged from 1,600 to 15,000 per year that could be attributable directly to ultrasound and LLD procedures.

conclusions. Our simulation results showed that, despite cumulative use of probe cover and LLD, there were still some cases of denovo infection that may be attributable to ultrasound procedures. These cases are preventable by reviewing the currently used LLD and/or upgrading LLD to high-level disinfection, as recommended by the US Centers for Disease Control and Prevention.

Infect Control Hosp Epidemiol 2014;35(12):000-000

Affiliations: 1. Laboratoire de Biostatistique, Epidemiologie, Sante Publique et Informatique Medicale, Centre Hospitalier Universitaire (CHU) de Nımes,Nımes, France; and EA 2415 Unit, Montpellier 1 University, Montpellier, France; 2. Universite de Bordeaux, Microbiologie Fondamentale et Pathogenicite,Unite de Mixte de Recherche (UMR) 5234, Bordeaux, France; and Centre National de la Recherche Scientifique, Microbiologie Fondamentale et Patho-genicite, UMR 5234, Bordeaux, France; 3. CHU de Bordeaux, Bordeaux, France; 4. Division of Infectious Diseases, University of North Carolina Schoolof Medicine, Chapel Hill, North Carolina; 5. Department of Biostatistics, City of Hope, Duarte, California.

Received January 3, 2014; accepted August 5, 2014; electronically published November 5, 2014. 2014 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2014/3512-00XX$15.00. DOI: 10.1086/678604

In ultrasonography, a vaginal probe and all endocavitaryprobes without a probe cover are considered semicritical de-vices, because they have direct contact with mucous mem-branes (eg, vagina, rectum, and pharynx). Endorectal andvaginal ultrasonography are widely used as important diag-nostic tools in gynecology, obstetrics, and urology. Such en-docavitary ultrasonography is seen as a harmless procedurebecause of the absence of ionizing radiation. However, thecost of transducers precludes a single-use-only strategy. En-dovaginal and transrectal ultrasonography are considered asat least medium-risk procedures, in which “semicritical” in-struments come into contact with mucous membranes andrequire high-level disinfection (HLD) rather than steriliza-tion.1-3 Although endocavitary ultrasonography probes mightbe considered even less critical instruments, because they are

routinely protected by single-use disposable probe covers,leakage rates of 0.9%–2% for condoms and 8%–81% forcommercial probe covers have been reported in the literature.4

For maximum safety, the key infection control issue con-cerns the risk of contamination and the need for specificcleaning and disinfection procedures to ensure a high degreeof protection against infectious disease transmission evenwhen a disposable cover is used, as recommended in theUnited States, Canada, and Australia.1-3,5 Cleaning with a de-tergent and water solution is important as the first step inproper disinfection, because chemical disinfectants act moreeffectively on clean surfaces. Because of the potential disrup-tion of the barrier sheath, additional HLD with chemicalagents (eg, glutaraldehyde, aldehydes, and hydrogen perox-ide) is necessary. Least desirable, but routinely used, are wipe

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

000 infection control and hospital epidemiology december 2014, vol. 35, no. 12

figure 1. Model steps. LLD, low-level disinfection; Prob, proba-bility; Pt, patient.

disinfection methods, often containing quaternary ammo-nium compounds, which are classified as a low-level disin-fection (LLD) method.

The main pathogens of concern for both transrectal andendovaginal ultrasound examinations are human immuno-deficiency virus (HIV), cytomegalovirus (CMV), human pap-illoma virus (HPV), enteric gram-negative pathogens (eg,Escherichia coli and Klebsiella species). Clostridium difficile isa pathogen of specific concern for transrectal ultrasound, andNeisseria gonorrhoeae and Treponema pallidum (syphilis) arespecific concerns for endovaginal ultrasound. A recent sys-tematic review and meta-analysis estimated a pooled preva-lence of 12.9% (95% confidence interval [CI], 1.7%–24.3%)for pathogenic bacteria remaining on the probe after cleaningand LLD, even when a disposable cover is used, and 1.0%(95% CI, 0.0%–10.0%) for frequently occurring viruses(HPV, herpes simplex virus [HSV], and CMV) on endovagi-nal and rectal probes.6 The pooled prevalence of infectedpatients after transrectal ultrasound and guided biopsies wasestimated to be 3.1% (95% CI, 1.6%–4.3%). However, thesystematic review confirmed that very few cases of contam-inated patients with an established route of contaminationfrom endocavitary ultrasonography had been reported. Thisfinding does not mean that infectious risk of bacterial andviral transmission attributable to shared probes between pa-tients and failure of LLD procedures does not exist. The keyquestion is to estimate it in the presence of daily LLD practice,despite the use of a disposable cover.

An attempt to estimate the number of patients infected byendovaginal ultrasonography probes has been made by theFrench Sanitary Institute (INVS).7 However, the modelingtechniques used were relatively crude, applying a multipli-cative model, assuming that an infected patient would onlyinfect the following one and not taking into account howprobe covers and their manipulation before cleaning theprobe would affect the amount of virus or bacteria left on

the probe head. Our objective was to perform a moresophisticated modeling combined with the previous meta-analysis to specify this infectious risk for broader infectiousagents. Our models focused on viral and bacterial quantitativevariation across the patient flow and were dependent on (i)the order of infected (or uninfected) patients and (ii) themicroorganism’s ability to stay on the probe after removalof the cover. After quantifying these parameters, our modelcould then be turned into an infection risk estimate for pa-tients.

methods

Study Design

A flexible, detailed simulation model was built to estimatethe number of patients annually who are exposed an infec-tious disease by means of a contaminated ultrasonographyprobe.

Mathematical Model Building

A stochastic Monte Carlo computer simulation produced hy-pothetical cohort data for a population of 4 million exami-nations performed annually in France.8 Our model steps areshown in Figure 1, and the modeling was restricted to vaginaland/or rectal ultrasonography examination without any in-vasive procedures (eg, needle biopsies). Our simulations re-quired the following probabilities to be specified into themodel to compute the risk of having a vaginal-rectal ultra-sonography-transmitted infection with an LLD and coveredprobe:

• Probability 1 was a combination of (i) the probability thata random patient underwent the examination with an activeinfection and presence of the pathogen in the vaginal orrectal mucosa and corresponding fluids (Pr1a) and (ii) theprobability that the infected patient contaminated the cov-ered probe or the handles of the probe during an exami-nation (Pr1b).

• Probability 2 resulted from the standard operating proce-dure of preparing the probe for the next patient (sheathremoval, Pr2a; LLD, Pr2b; new cover and gel, Pr2c). Prob-ability 2 is the probability that the probe remained con-taminated just before the next examination despite adher-ence to standard disinfection procedures.

• Probability 3 was the probability that the probe contami-nated the new patient, as a combination of probability ofpathogen transmission from the covered probe to the newpatient (Pr3a), with the probability that the next patientwas potentially receptive (Pr3b).

We built a model for each of the following pathogens thatwe considered to be most relevant in an ultrasonographysetting: HIV, hepatitis B virus (HBV), hepatitis C virus(HCV), HPV, HSV, CMV, and Chlamydia trachomatis. Ofnote, the pathogens that we considered persist on inert surfacesfor longer than several hours, and we did not include the time

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

impact of vaginal-rectal ultrasound examinations on infectious transmission 000

table 1. Parameters Used in Simulations

Probability 1 Probability 2 Probability 3

Pathogen

Pr1a2(disease

prevalence)

Pr1b(transmissionfrom patient

to probe)

Pr2b(probability

that a pathogenremained aftercleaning/LLD)

Pr2c (pathogentransmission tocover exterior

despite gel andsheath; our data)

Pr3a(probability of probecontamination froman infected patient)

Pr3b(potentially

receptivepatients)a

HIV 0.002124 0.2 0.75613,15b 0.023–0.147 0.001525 1.00HBV 0.006526c 0.8 0.75613,15b 0.023–0.147 0.5517,18d 0.3513

HCV 0.005326e 0.3 0.75613,15b 0.023–0.147 0.00527,28 1.00HSV2 0.1829,30 0.5 0.75613,15b 0.023–0.147 0.03620 0.8229,30

HPVBaseline model 0.0819f 0.5 0.75613,15b 0.023–0.147 0.5631,32 0.9219

Empirical model … … 0.04921b 0.023–0.147 0.5631,32 0.9219

CMV 0.533,34 0.5 0.75613,15b 0.023–0.147 0.0420g 0.5018,20

Chlamydia trachomatis 0.01520 0.5 0.75613,15b 0.023–0.147 0.535-37 0.98520

note. Our estimate for Pr2a across all pathogens was 8.9%, from the Kac et al13 study. CMV, cytomegalovirus; HBV, hepatitis Bvirus; HCV, hepatitis C virus; HIV, human immunodeficiency virus; HPV, human papilloma virus; HSV2, herpes simplex virus type2; LLD, low-level disinfection.a Evaluated as (1 disease prevalence) or (1 prevalence of vaccinated people).b Results from pooling data using a meta-analysis technique; see “Methods.”c Prevalence of hepatitis B surface antigen–positive individuals.d This was estimated on the basis of the median of the range of probabilities of sexual transmission.26,30

e Prevalence of individuals with an active HCV infection (ie, HCV RNA positive).30

f The probability of sexual transmission ranged from 5% to 100% according to Burchell et al,19 with a median of 40% that wasconsidered for estimating the probability of probe transmission from an infected patient.g The sexual transmission probability remained unknown; we approximated with the only available data in the literature, which wasfor CMV transmission related to breast-feeding.20

period between 2 examinations in our model. The assumedvalues for each pathogen’s parameters used as probability es-timates in the simulations are summarized in Table 1.

The Pr1a estimate was based on the pathogen prevalenceand its presence in human fluids, such as vaginal secretion.For each random patient who entered into the ultrasonog-raphy clinic, the Pr1a estimate was the probability in a Ber-noulli trial (0 p no infection, 1 p infection) to determinewhether the patient carried an active infection. Consequently,we simulated a group of patients with active infections froma binomial distribution with probability Pr1a. In the case ofHBV and HCV, we estimated the probability of Pr1a2 fromthe population-based prevalence of patients who were hep-atitis B surface antigen positive and the population-basedprevalence of patients with an active HCV infection (ie, HCVRNA positive), respectively.9

Pr1b estimated the pathogen transmission from an infectedpatient to the external surface of the probe cover. Very fewdata were available in the literature: only patients with AIDSwere found to be able to contaminate semicritical dental de-vices;10 risk of contamination for dental devices was foundto be 0%–60% for HCV11 and 15%–75% for HBV.12 We es-timated Pr1b using the percentage proposed by a consensusof experts.9 The covered probe was considered to be exposedto the pathogen from a Bernoulli trial with probability Pr1b.

Probability Pr2a that a pathogen is found on the probe

after sheath removal was obtained from the Kac et al13 study.In this study, the viral genomes of HPC, CMV, and EBV weresearched on the endovaginal or endorectal probe covers afterexamination just before removing the cover and then on theprobes after sheath removal.13 The authors estimated that8.9% (95% CI, 3.5%–19.7%) of the probes are contaminatedwith the pathogen when the pathogen was found on the probecover. Probability Pr2b that a pathogen remained on theprobe after cleaning and LLD was obtained by pooling datafrom 2 studies using a meta-analysis calculation approach.14,15

For Pr2b estimation, we assumed that LLD was as effectivefor viruses as for bacteria, although this assumption isarguable.

Our group empirically estimated the probability (Pr2c) thatthe contamination remaining on the probe would contami-nate the next patient despite use of new gel and new cover.We estimated this probability by an experiment in a controlledradiologic clinic setting. These results are described in detailin M’Zali et al,16 but we summarize the experiment here. Thisexperiment was to simulate a routine examination. The ma-terials included a blue phantom mannequin used for vaginalultrasound medical training, Conformite Europeenne probecovers (recommended in France), and an ultrasound system.We used a strain of Pseudomonas aeruginosa as the source ofcontamination into which the probe was soaked. We mea-sured the number of times that the exterior surface of the

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

000 infection control and hospital epidemiology december 2014, vol. 35, no. 12

table 2. Results of Simulations across All Pathogens

No. of patients, mean SD

PathogenPatients arriving

with active infection

Probe becomescontaminated

for next ultrasonographypatient despite LLD

and use of probe covers

Uninfected patientsleaving with infectionafter ultrasonography

examination

HIV 8,392 197 1,274 75 40 20HBV 26,033 319 15,738 248 1,383 164HCV 21,177 315 4,789 151 151 63HSV2 719,914 1,459 86,574 583 9,707 2,900HPV

Baseline model 319,961 1,025 15,709 274 8,085 193Empirical model … … 14,848 255

CMV 2.0E06 1,895 755,959 1,532 22,549 7,690Chlamydia trachomatis 60,017 511 22,680 277 4,025 183

note. CMV, cytomegalovirus; HBV, hepatitis B virus; HCV, hepatitis C virus; HIV, human immunodeficiencyvirus; HPV, human papilloma virus; HSV2, herpes simplex virus type 2; LLD, low-level disinfection; SD, standarddeviation.

probe cover and the mannequin were contaminated with P.aeruginosa under routine examination conditions, and theseestimates were our basis for Pr2c.

Pr3a estimates were assumed to be similar to the proba-bility of sexual transmission of the pathogen, when available(eg, the Pr3a estimate for HBV).17,18 If these estimates werenot available, it was assumed to be equivalent to accidentalblood exposure risk. The probability of sexual transmissionof HPV ranged from 5% to 100% according to Burchell etal,19 with a median of 40% that was considered for estimatingthe probability of probe transmission from an infected pa-tient. Data for the sexual transmission probability of CMVwas not available; we approximated the sexual transmissionprobability of CMV by the probability of transmission frombreast-feeding.20 Pr3b represented the potentially receptivepatients percentage (eg, 100%) for HIV or HCV but tookinto account prevalence of disease when the disease was morefrequent (CMV, HPS, and HSV2) or the prevalence of vac-cinated people (HBV).

Casalegno et al21 estimated in a cohort study the numberof ultrasonography probes contaminated by HPV just beforethe next patient in a cohort study of patients who underwentendovaginal ultrasonography examination. Based on thisstudy, and for HPV specifically, we constructed 2 models: an“empirical model” and the baseline (BL) HPV model. TheBL HPV model is comparable to the simulation model thatwe considered for the other pathogens. In the “empiricalmodel,” HPV probabilities of Pr1b to Pr2b, inclusive, wereestimated empirically by pooling results from Casalegno etal21 and our data using a meta-analysis calculation approachdescribed elsewhere.

The mathematical modeling is further described in theAppendix. The following are the key assumptions for ourmodel:

• Every step in the probe contamination/decontaminationprocess was considered as an independent trial, with theexception that we allowed for the possibility of a contam-inated probe infecting every subsequent subject until theprobe was decontaminated or 20 examinations.

• Patients were drawn from a random Bernoulli trial withthe probability of infection equal to the prevalence ofpathogen.

• Transmission probabilities were the ones previously definedand appear in Table 1.

• Modeling did not take into account that patients may havecontracted more than 1 infection; we assumed that trans-mission of more than 1 pathogen resulting in multiple in-fections per patient or per probe is unlikely.

• We did not account for differences between vaginal or rectalmucosa, either in probability of transmission or probabilityof infection.

After discussion with experts in the field, we used a geometricdistribution with a probability that the pathogen remainedon the probe to simulate who was at risk for contaminationwhen examined after an infected patient. However, the geo-metric distribution was truncated at a limit of at most 20patients subsequent to the index case patient. The simulationprogramming was performed in the R statistical computinglanguage.22

results

Estimated Parameters Used for the Modeling

All parameters used for the modeling are presented in Table1. Pr1a2, Pr1b, Pr2c, Pr3a, and Pr3b came from literaturewith reference indexed. Pr2b was estimated at 0.756 afterpooling data from relevant studies with a meta-analysis tech-nique, and Pr1 to Pr2b inclusive was estimated at 0.049 for

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

impact of vaginal-rectal ultrasound examinations on infectious transmission 000

figure 2. Example Monte Carlo simulation results from the assumptions in Table 1. Histograms and kernel density estimates are ofhuman immunodeficiency virus (HIV; A) and human papilloma virus (HPV; B). The 2 competing models for HPV are shown in black(baseline model) and dark gray (model based on empirical data). CI, confidence interval; inf, infections.

the HPV empirical model. For all pathogens, Pr2a was 8.9%(95% CI, 3.5%–19.7%) based on the study by Kac et al.13

Modeling Results

The results of 4 million simulations per pathogen appear inTable 2. First, the number of infected patients who arrivedfor ultrasonography matched reasonably with the assumeddisease prevalence in the Pr1a2 column of Table 1. The sim-ulated rates of an infected patient contaminating a probedespite LLD procedures ranged from 5% in the case of HPVto over 30% for CMV and C. trachomatis. The rate of infectiontransmission from an infected patient to an uninfected patientranged from 0.7% for HIV (63 of 8,392) to over 6% for bothHCV and C. trachomatis.

We had 2 methods of estimating the number of new HPVinfections due to ultrasonography. The first method, the BLmodel, used the same simulation that we applied to the otherpathogens. However, Casalegno et al21 and our empirical dataaddressed the same question in HPV with empirical data.This reduced our simulation complexity; in the empiricalHPV model, we only applied the probabilities to the trans-mission of HPV to an uninfected patient from the step whenHPV is transferred from the contaminated probe to the ex-terior of the probe cover. The results of the HPV BL andempirical HPV models had similar estimates of de novo in-fection cases. The BL model, despite its complexity, gave moreconservative estimates of cases than the more empiricallybased HPV model. This may be an indication that our sim-

ulated number of cases may be similarly conservative for theother pathogens.

Figure 2 shows 2 examples of the simulation results for 2of the pathogen results in Table 2. The plot in Figure 2Ashows the estimated number of patients with cases of HIVinfection who arrive for an ultrasonography examination peryear in France and our estimate for the resulting simulatednumber of uninfected patients who are exposed to and areinfected by a probe contaminated by HIV. The plot in Figure2B has a similar interpretation for the 2 HPV models. Oursimulated HPV results were consistent and suggest that theestimates mirror the actual number of HPV infections re-sulting from ultrasonography due to our use of the publishedempirical data for probe contamination.

discussion

It is understood that there are risks of endocavity infectionfor any diagnostic procedure in which the instrument is notintended for single-patient use. There are differing opinionsas to how to quantify the risk in the endorectal and vaginalultrasonography setting. Although all practitioners under-stand that the risk of a contaminated ultrasonography probeis nonzero, there has not been an attempt thus far to syn-thesize and summarize the existing data across a broad spec-trum of pathogens.

Building on our recent results,6,23 we considered a system-atic review of the available published data from ultrasonog-

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

000 infection control and hospital epidemiology december 2014, vol. 35, no. 12

table a1. Simulated Patients

PatientActive

infection? Simulation result

1 No Patient does not carry an active pathogen2 No No active pathogen3 No No active pathogen4 No No active pathogen5 Yes Patient is carrying an active pathogen6 No No active pathogen7 Yes Patient is carrying an active pathogen8 No No active pathogen9 No No active pathogen10 No No active pathogen

raphy clinics in France to estimate the probabilities at eachstep in the process of pathogen transmission during ultra-sonography. Our approach extends an earlier in silico studyby the French Sanitary Institute.7 The rates of de novo infectedcases from our simulations range between 1% and 6% riskfor the procedures that we considered. These cases are likelyto go undetected by health authorities, and patients often donot self-report infections that were attributable to ultraso-nography. This may explain why there are no available datapublished in the literature. Our simulation study is, to ourknowledge, the first one in this area of research and providesthe first estimation of the infectious risk related to endovag-inal and endorectal ultrasonography covered probes after LLDprocedures.

There are limitations to our findings. Our risk estimatesare population based and not age corrected. The probabilityassumptions on which our simulations were built tended tobe point estimates from relatively small studies. Some prob-ability estimates, such as the probability of pathogen trans-mission to external cover after application of gel and a sheath,were used across multiple pathogens. However, it may be thecase that this estimate is suitable only for a specific pathogen;we were unable to find comparable probabilities for eachpathogen of interest. Other limitations include the probabi-listic assumptions of our model. We assumed independenceof the efficacy of disinfection methods and probe covers. Itis more likely that infection control standards vary widelybetween ultrasonography clinics. It is not possible to modelthese types of clinic-to-clinic variations without supportivedata, but these data are difficult and expensive to collect fromboth a sampling and a laboratory perspective. Finally, wepossibly overestimated the number of HIV-infected patientsafter ultrasonography examination, because Pr2b used formodeling estimated LLD efficacy for bacteria. However, HIVis a fragile virus for which LLD efficacy may be higher thanit is for bacteria.

Despite these limitations, our simulation results may drawattention to the cost in terms of patients and illness due tocurrent infection control standards across ultrasonographyclinics in France. We believe that it is possible to reducesignificantly the number of new cases of infection estimatedin our study arising from contaminated ultrasound probeswith enhanced disinfection procedures and standards. Ourestimates indicated that 5%–30% of infections can be elim-inated by introducing improved standards for those clinicsthat rely on LLD. Indeed, in contrast to LLD, HLD has beenshown to be up to 100% effective in different settings. Kacet al13 reported that 3.4% of endovaginal and transrectal trans-ducers were contaminated by pathogenic bacteria and that1.5% had viral contamination, all of which contaminationdisappeared after ultraviolet-C HLD.23 When our estimatednumber of cases is multiplied by the lifetime cost to treatpatients with these types of infections, increased infectioncontrol procedures may be worth the investment.

In conclusion, our data synthesis of the infectious disease

literature and corresponding simulations showed that theremay be a case to be made for improved infection controlprocedures in ultrasonography clinics where endocavitary ul-trasonography probes are used routinely. There is a need forthis topic to receive additional study, particularly in areas thatdo not put ultrasonography patients at risk. Simulation stud-ies and studies with blue phantom mannequins may still serveto improve upon the state of the data until additional dataon humans are available.

acknowledgments

Financial support. This study was funded by Germitec.Potential conflicts of interest. D.J.W. is a paid consultant for Germitec.

All other authors report no conflicts of interest relevant to this article. Allauthors submitted the ICMJE Form for Disclosure of Potential Conflicts ofInterest, and the conflicts that the editors consider relevant to this articleare disclosed here.

Address correspondence to Sandrine Leroy, MD, PhD, EA2415 Unit,Montpellier 1 University, 34295 Montpellier, Cedex 5, France (sandrin.leroy@gmail.com).

appendix

This is an example of our simulation of patient flow. First,we simulate a series of patients. In this example, we willassume 10 patients who arrive in an ultrasonography clinicin France (Table A1).

Which of the Patients Has an Active Infection?

Each of the 10 women has a probability of walking into theclinic with an active viral infection (probability Pr1a). Forthis example, we will use an arbitrary probability of Pr1a p10% of a viral pathogen in the general population.

Patient 5 and Patient 7 have an infection. This puts thesubsequent patients at risk for infection.

Next, we will simulate whether the probe is contaminatedby the pathogen. This probability is Pr1b. For this example

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

impact of vaginal-rectal ultrasound examinations on infectious transmission 000

with 10 patients, we will assume that the Pr1b probability isa constant probability of 0.20.

Did Probe Become Contaminated?

First, we will simulate whether patient 5 infects the probe.For our simulation’s Bernoulli trial, we perform a biased cointoss with the probability of infection. For example, if therandom number generator gives 0.2 or less, then patient 5has contaminated the probe. If the random number generatoris greater than 0.2, then there is no infection.

Who May Be at Risk from Contamination?

Patient 6 and patients 8–10 may be at risk of an infectedprobe. However, our simulation assumes that patient 7 is notat risk from patient 5, because patient 7 already has the in-fection in question. Additionally, the risk of infection is notcumulative for patients 8 through 10; we assume that theymay only be infected by the most recent patient with an activeinfection (patient 7 in this example).

In the case in which an infected patient has an ultrasoundand contaminates the probe, the simulation chooses howmany subsequent patients are at risk. The number of sub-sequent patients who may be infected was drawn from atruncated geometric distribution.

Did Probe Become Contaminated Despite LLD?

The simulation checks whether the probe is cleaned withprobability Pr2a, once per patient, for each patient.

Did the Probe Cover Prevent Infection?

Similarly, we can simulate whether the infection is transferredto the exterior of the probe cover. This is probability Pr2b.

If the probe has been disinfected or if the infection wasnot transferred to the exterior of the probe cover, then thereis no risk of contamination to the next patient. If both theLLD and the probe cover failed to prevent the infection, thenthe probe has become contaminated for the next uninfectedpatient. The probability of transmission from probe to patientis Pr3a. For most of our pathogens, we assumed that theprobability of the infection being transmitted from the probeto the next uninfected patient was approximately the sameas a sexual transmission.

Finally, the subject may not be receptive to the infection.That probability is modeled by Pr3b.

references

1. American Institute of Ultrasound in Medicine (AIUM). Guide-lines for Cleaning and Preparing Endocavitary US Transducersbetween Patients. Laurel, MD: AIUM, 2003.

2. Commission Special Securite Sanitaire Comite Technique desInfections Nosocomiales et des Infections Liees Aux Soins.

Gaines De Protection a Usage Unique Pour Dispositifs MedicauxReutilisables: Recommandations d’Utilisation. Paris: Haut Conseilde Sante Publique, 2007.

3. US Food and Drug Administration Center for Devices and Ra-diological Health (FDA/CDRH). Guidance for Industry: Guid-ance for Manufacturers Seeking Marketing Clearance of Ear, Nose,and Throat Endoscope Sheaths Used as Protective Barriers. Wash-ington DC: FDA/CDRH, 2000.

4. Amis S, Ruddy M, Kibbler CC, Economides DL, MacLean AB.Assessment of condoms as probe covers for transvaginal sonog-raphy. J Clin Ultrasound 2000;28(6):295–298.

5. Australian Society for Ultrasound in Medicine (ASUM). Guide-lines for Disinfection of Intracavitary Transducers: Policies andStatements. Crow’s Nest, New South Wales: ASUM, 2005.

6. Leroy S. Infectious risk of endovaginal and transrectal ultra-sonography: systematic review and meta-analysis. J Hosp Infect2013;83(2):99–106.

7. Antona D, Bernillon P, Coignard B, Gallay A, Larsen C, Lot F.Analyse du Risque Infectieux Lie aux Echographies Endocavitaires,en l’Absence de Protection ou de de Sinfection des Sondes EntrePatients. Saint-Maurice, France: Institut de Veille Sanitaire Fran-caise, 2008.

8. Assemblee Nationale XIIIe Legislature. Deuxieme Seance duLundi 9 Mars 2009.

9. Institut de Veille Sanitaire. Infectious risk related to non-ster-ilization between patients of semi-critical dental devices. 2009.http://www.invs.sante.fr/publications/2009/risques_chirurgie_dentaire/risques_chirurgir_dentaire.pdf.

10. Lewis DL, Arens M, Appleton SS, et al. Cross-contaminationpotential with dental equipment. Lancet 1992;340(8830):1252–1254.

11. Artini M, Scoarughi GL, Papa R, et al. Specific anti cross-infection measures may help to prevent viral contamination ofdental unit waterlines: a pilot study. Infection 2008;36(5):467–471.

12. Hu T, Li G, Zuo Y, Zhou X. Risk of hepatitis B virus transmissionvia dental handpieces and evaluation of an anti-suction devicefor prevention of transmission. Infect Control Hosp Epidemiol2007;28(1):80–82.

13. Kac G, Podglajen I, Si-Mohamed A, Rodi A, Grataloup C, MeyerG. Evaluation of ultraviolet C for disinfection of endocavitaryultrasound transducers persistently contaminated despite probecovers. Infect Control Hosp Epidemiol 2010;31(2):165–170.

14. Buffet-Bataillon S, Vallee A, Lebrun B, Cormier M, Poulian P,Jolivet-Gougeon A. Controle microbiologique de la desinfectionde sondes endovaginales et d’echographie transoesophagienneau CHU de Rennes. In: Program and abstracts of the 20thCongress of the Societe Francaise d’Hygiene Hospitaliere(SFHH) and Societe des Infirmiers et Infirmieres en HygieneHospitaliere de France (SIIHH). Nice, France: SFHH andSIIHH, 2009. Abstract 312009.

15. Kac G, Gueneret M, Rodi A, et al. Evaluation of a new disin-fection procedure for ultrasound probes using ultraviolet light.J Hosp Infect 2007;65(2):163–168.

16. M’Zali F, Leroy S, Kann M, Quentin-Noury C. A novel approachfor in vitro evaluation of the potential risk of patient contam-ination during endovaginal ultrasound examinations. In: Pro-gram and abstracts of the 24th European Congress of ClinicalMicrobiology and Infectious Diseases. eP275. Barcelona, Spain:

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

000 infection control and hospital epidemiology december 2014, vol. 35, no. 12

European Society of Clinical Microbiology and Infectious Dis-eases, 2014.

17. Denis F, Trepo C, Alain S, Chastel C. Virus des Hepatites B etDelta. Paris: Elsevier; 2004.

18. Lee WM. Hepatitis B virus infection. New Engl J Med 1997;337(24):1733–1745.

19. Burchell A, Winer R, de Sanjose S, Franco EL. Chapter 6: ep-idemiology and transmission dynamics of genital HPV infection.Vaccine 2006;24(suppl 3):S3/52–61.

20. Hayashi S, Kimura H, Oshiro M, et al. Transmission of cyto-megalovirus via breast milk in extremely premature infants. JPerinatol 2011;31(6):440–445.

21. Casalegno JS, Le Bail Carval K, Eibach D, et al. High risk HPVcontamination of endocavity vaginal ultrasound probes: an un-derestimated route of nosocomial infection? PLOS ONE 2012;7(10):e48137.

22. R Development Core Team. R: a language and environment forstatistical computing. Vienna, Austria: R Foundation for Statis-tical Computing, 2013.

23. M’Zali F, Bounizra C, Leroy S, Mekki Y, Quentin-Noury C, KannM. Persistence of microbial contamination on transvaginal ul-trasound probes despite low-level disinfection procedure. PLOSONE 2014;9(4):e93368.

24. Pillonel J, Cazein F. The fight against HIV/AIDS and sexuallytransmitted infections in France: 10 years of surveillance, 1996–2005. Saint-Maurice, France: Institut de Veille Sanitaire, 2007:32–35. http://www.invs.sante.fr/publications/2007/10ans_vih/index.html.

25. Downs AM, De Vincenzi I. Probability of heterosexual trans-mission of HIV: relationship to the number of unprotected sex-ual contacts. European Study Group in Heterosexual Trans-mission of HIV. J Acquir Immune Defic Syndr Hum Retrovirol1996;11(4):388–395.

26. Institut de Veille Sanitaire. Prevalence des hepatites B et C enFrance en 2004. Saint-Maurice, France: Institut de Veille Sani-taire, 2007.

27. Lot F, Desenclos J. Epidemiologie de la transmission soignant/soigne: risque lie au VIH, VHC et VHB. Hygienes 2003;11:96–100.

28. Jagger J, Puro V, De Carli G. Occupational transmission of hep-atitis C virus. JAMA 2002;288(12):1469; author reply 1469–1471.

29. Herpes simplex. In: Heymann D, ed. Control of CommunicableDiseases Manual. 18th ed. Washington DC: American PublicHealth Association, 2004:268–272.

30. LeGoff J, Saussereau E, Boulanger MC, et al. Unexpected highprevalence of herpes simplex virus (HSV) type 2 seropositivityand HSV genital shedding in pregnant women living in an EastParis suburban area. Int J STD AIDS 2007;18(9):593–595.

31. Corey L, Wald A, Patel R, et al. Once-daily valacyclovir to reducethe risk of transmission of genital herpes. New Engl J Med 2004;350(1):11–20.

32. de Lima Rocha MG, Faria FL, Goncalves L, Souza Mdo C,Fernandes PA, Fernandes AP. Prevalence of DNA-HPV in malesexual partners of HPV-infected women and concordance ofviral types in infected couples. PLOS ONE 2012;7(7):e40988.

33. Bate SL, Dollard SC, Cannon MJ. Cytomegalovirus seropreva-lence in the United States: the national health and nutritionexamination surveys, 1988–2004. Clin Infect Dis 2010;50(11):1439–1447.

34. Boeckh M, Geballe AP. Cytomegalovirus: pathogen, paradigm,and puzzle. J Clin Invest 2011;121(5):1673–1680.

35. Katz BP. Estimating transmission probabilities for chlamydialinfection. Stat Med 1992;11(5):565–577.

36. Clad A, Prillwitz J, Hintz KC, et al. Discordant prevalence ofChlamydia trachomatis in asymptomatic couples screened usingurine ligase chain reaction. Eur J Clin Microbiol Infect Dis 2001;20(5):324–328.

37. Quinn TC, Gaydos C, Shepherd M, et al. Epidemiologic andmicrobiologic correlates of Chlamydia trachomatis infection insexual partnerships. JAMA 1996;276(21):1737–1742.

This content downloaded from 31.34.124.196 on Fri, 7 Nov 2014 01:45:55 AMAll use subject to JSTOR Terms and Conditions

7. Persistence of Microbial

Contamination on Transvaginal

Ultrasound Probes despite Low-

Level Disinfection Procedure - Fatima M’Zali1*, Carole Bounizra1,

Sandrine Leroy2, Yahia Mekki3, Claudine

Quentin-Noury1, Michael Kann1 -

Université Bordeaux Segalen – Plos One

2014

Persistence of Microbial Contamination on TransvaginalUltrasound Probes despite Low-Level DisinfectionProcedureFatima M’Zali1*, Carole Bounizra1, Sandrine Leroy2, Yahia Mekki3, Claudine Quentin-Noury1,

Michael Kann1

1 Universite Bordeaux Segalen, Microbiologie Fondamentale et Pathogenicite Unite Mixte de Recherche 5234, Bordeaux, France, 2 Centre Hospitalier Universitaire de

Nımes, Service de Biostatistique, Epidemiologie Clinique, Sante Publique, Informatique Medicale, Nımes, France, 3 Laboratoire de Virologie, Centre de Biologie et

Pathologie Est, Hospices Civils de Lyon, Lyon, France

Abstract

Aim of the Study: In many countries, Low Level Disinfection (LLD) of covered transvaginal ultrasound probes isrecommended between patients’ examinations. The aim of this study was to evaluate the antimicrobial efficacy of LLDunder routine conditions on a range of microorganisms.

Materials and Methods: Samples were taken over a six month period in a private French Radiology Center. 300 specimensderived from endovaginal ultrasound probes were analyzed after disinfection of the probe with wipes impregnated with aquaternary ammonium compound and chlorhexidine. Human papillomavirus (HPV) was sought in the first set of s100samples, Chlamydia trachomatis and mycoplasmas were searched in the second set of 100 samples, bacteria and fungi inthe third 100 set samples. HPV, C. trachomatis and mycoplasmas were detected by PCR amplification. PCR positive sampleswere subjected to a nuclease treatment before an additional PCR assay to assess the likely viable microorganisms. Bacteriaand fungi were investigated by conventional methods.

Results: A substantial persistence of microorganisms was observed on the disinfected probes: HPV DNA was found on 13%of the samples and 7% in nuclease-resistant form. C. trachomatis DNA was detected on 20% of the probes by primary PCRbut only 2% after nuclease treatment, while mycoplasma DNA was amplified in 8% and 4%, respectively. Commensal and/orenvironmental bacterial flora was present on 86% of the probes, occasionally in mixed culture, and at various levels (10-.3000 CFU/probe); Staphylococcus aureus was cultured from 4% of the probes (10-560 CFU/probe). No fungi were isolated.

Conclusion: Our findings raise concerns about the efficacy of impregnated towels as a sole mean for disinfection ofultrasound probes. Although the ultrasound probes are used with disposable covers, our results highlight the potential riskof cross contamination between patients during ultrasound examination and emphasize the need for reviewing thedisinfection procedure.

Citation: M’Zali F, Bounizra C, Leroy S, Mekki Y, Quentin-Noury C, et al. (2014) Persistence of Microbial Contamination on Transvaginal Ultrasound Probes despiteLow-Level Disinfection Procedure. PLoS ONE 9(4): e93368. doi:10.1371/journal.pone.0093368

Editor: Rui Medeiros, IPO, Inst Port Oncology, Portugal

Received June 18, 2013; Accepted March 4, 2014; Published April 2, 2014

Copyright: 2014 M’Zali et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The laboratory received a grant from Germitec (www.germitec.com). The fact that the authors received this external funding does not alter theiradherence to PLOS ONE policy. The funders had no role in the study design, data collection and analysis or preparation of the manuscript.

Competing Interests: The authors have declared that no competing interests exist.

* E-mail: fatima.mzali@u-bordeaux2.fr

Introduction

Endovaginal ultrasonography is commonly used in gynecology

and obstetrics for investigation of suspected disease and pregnancy

complications and for medically assisted procreation. Transvaginal

as other endocavitary probes are considered semi-critical devices

since they are not intended to penetrate skin or mucous

membranes but only to come into contact with them. Being at

lower risk of infection, sterilization of these equipments is neither

required nor feasible. To minimize even further the risk, the

endocavitary ultrasound probes are covered with a single use

sheath, after coating the probe with a gel enabling sound

transmission. Nevertheless, probe covers can fail, and probes can

be contaminated by pathogens present in human secretions

resulting in their nosocomial transmission. Thus, disinfection of

the probes between patients is needed. However, there are no

consensual guidelines for transvaginal probe disinfection. Health

authorities such as the Centers for Disease Control and Prevention

[1] or the American Institute of Ultrasound in Medicine [2]

recommend a High-Level Disinfection (HLD) even for covered

probes. HLD technologies consist of immersion in glutaraldehyde,

hydrogen peroxide, or peracetic acid, and then rinsing and drying.

They present many drawbacks such as possible deterioration of the

transducer, chemical damage to the mucosa of patients and

practitioners, toxic effects on the gametes and embryos, impaired

imaging, and in all cases the time devoted to the procedure. As a

consequence, current practice compliance with this standard is

PLOS ONE | www.plosone.org 1 April 2014 | Volume 9 | Issue 4 | e93368

poorly followed [3,4]. For this reason, other countries recommend

a Low-Level Disinfection (LLD) procedure based on probe wiping

with a single use towel (pre)impregnated with products such as

quaternary ammonium compounds or phenolics [5]. Although

very few cases or outbreaks of hospital acquired infections linked to

endovaginal ultrasound procedures have been documented [6–9],

the risk of cross infection must not be dismissed. Indeed, some

reports have evidenced bacterial and/or viral contamination of

LLD disinfected endovaginal probes [10–14]. Nevertheless, none,

to our knowledge, has investigated contamination of vaginal

ultrasound probes by both viruses and bacteria, including

Chlamydia trachomatis and mycoplasmas, together with fungi.

While new HLD technologies for ultrasound probes such as gas

plasma or ultraviolet C light systems are in evaluation to comply

with current workflow [12,15,16], the question on whether there is

a need to perform HLD between patients remains. The aim of this

study was to assess the antimicrobial efficacy of the LLD

procedure for transvaginal ultrasound probes on a range of

potentially pathogenic microorganisms under routine conditions.

Materials and Methods

Study settingsOver a 6-month period (between April and September 2012) a

prospective study was conducted in a large private French

Radiology center. No patient information of any kind has been

gathered, no human samples were tested in this study and the

observed procedure complied with the national recommendations.

Therefore no patient consent was required by the local ethical

committee. A total of 300 consecutive samples were taken from

vaginal ultrasound probes just after LLD disinfection of the probe.

Standard disinfection procedureThe disinfection of the probes was performed by the clinician.

Probes were covered with a medical CE mark disposable sheath.

After examination, the probe cover was carefully removed to avoid

probe contamination, and a visual inspection was performed to

detect any break of the probe cover and any blood or body fluids

on the probe. Then, the probe was cleaned with a non sterile dry

tissue paper to eliminate the gel, and disinfected using wipes

(Prodene, France) that are preimpregnated with a solution of

ethanol/water, propylene glycol, myristalkonium chloride, men-

thol, and chlorhexidine digluconate.

SamplingSamples were taken from three endovaginal ultrasound probes

using three ultrasound machines (Voluson E8, GE healthcare,

USA) each one placed in a separate scanning room. All specimens

were collected by a trained Microbiologist (F.M.), less than five

minutes after the disinfection of the probes by the clinicians. The

entire surface of the ultrasound probe was thoroughly sampled

using flocked swabs (Copan Diagnostic, France). Swabs were

immediately placed in transport media and brought to the

laboratory. Delay between sampling and laboratory processing

never exceeded three hours. The first 100 samples were analyzed

for HPV detection, the next set of 100 samples for C. trachomatis

and mycoplasmas, and the last set of 100 samples for bacterial and

fungal screening. Specific flocked swabs and Universal Transport

Medium (UTM-RT, Copan Diagnostic, France) were used for

HPV, C. trachomatis and mycoplasmas, and eSwabs in Amies liquid

(Copan Diagnostic, France) were used for the other microorgan-

isms. For quality control purpose, samples were collected once a

week from the ultrasound rooms using Count-Tact Agar plates

(BioMerieux, France), and from the ultrasound gel bottles.

DNA extraction and amplificationDNA was extracted from the samples using the semi-automated

magnetic system NucliSENS, easyMag (BioMerieux, France)

according to the manufacturer’s instructions. PCR amplifications

were carried out to screen for the presence of HPV, C. trachomatis

and mycoplasmas using previously described primers and condi-

tions [17–19]. All samples giving a PCR positive product were

further subjected to a nuclease treatment to remove any free DNA

and potentially leave likely viable microorganisms before another

PCR assay was performed. The nuclease treatment consisted of

mixing 900 ml of the sample with 2 mg Nuclease S7 (Sigma,

France) in presence of 10 mM CaCl2 during 2 h at 37uC. The

enzyme’s activity was stopped by the addition of 30 mM EDTA.

Each series included a negative control in order to test for

contamination during the extraction procedure, and a positive

control. One swab from each batch, as well as the transport

medium batch, was tested for the absence of microbial contam-

ination.

Culture and identification of bacteria and fungiAliquots of 100 ml of the transport medium were spread on a

series of agar plates (BioMerieux, France), either selective

(Gardenella specific agar, Sabouraud agar) or not (Mueller Hinton

agar, chocolate-polyvitex and horse blood agar). Plates were then

incubated at 37uC overnight (Mueller Hinton agar plates), at 30uCfor 2–5 days (Sabouraud plates), or at 37uC in a 5% CO2 enriched

atmosphere for up to 48 h (chocolate-polyvitex, blood and

Gardenella agar plates). After incubation, colony forming units

(CFU) were enumerated. The ultrasound gel analysis was

performed as above. Microorganisms grown on these plates and

on the Count-Tact Agar plates were identified to species level and

their antibiotic susceptibility was determined when relevant by

conventional methods.

AnalysisStatistical analysis of the number of pathogens was performed

using the Stata 12/SE software (Statacorp LP, Texas). Data were

expressed as number and 95% confidence interval (CI).

Results and Discussion

The results of this study revealed that despite LLD, the

ultrasound probes remained substantially contaminated by clini-

cally significant microorganisms, including HPV, C. trachomatis,

mycoplasmas, Gram-positive and Gram-negative bacteria. These

results are in accordance with those of the few studies on this topic

[10–14].

In the first subset of 100 samples screened by PCR for the

presence of HPV, 13% (95% CI: 6–20) were positive. Such

contamination rate is higher than previously described. For

instance, Ma et al. [13] found that 7.5% of surveillance samples

taken daily from vaginal transducers when the instrument was not

in use, were positive for HPV DNA; interestingly, three of the 14

probe samples collected from HPV colonized patients were

contaminated by HPV DNA. Casalegno et al. [11] reported

3.5% of HPV contaminated endovaginal probes, including 3% for

at least one high risk (HR) type; furthermore, they detected HPV

in 2.7% in pre-examination samples, including 1.9% of HR-HPV,

and apparently the same HR-HPV persisted on an endovaginal

probe despite three disinfection procedures. In both studies,

endovaginal probes were used with similar probe covers and LLD

procedure (wipes impregnated with quaternary ammonium

compounds) as ours. HPV DNA was detected either by PCR

covering more than 40 types of mucosal HPV [13] or by a

Contaminated Transvaginal Probes despite LLD

PLOS ONE | www.plosone.org 2 April 2014 | Volume 9 | Issue 4 | e93368

microarray kit detecting 35 HPV genotypes [11], while the

methodology used in the present study focused on 22 mucosal

HPV genotypes. Thus, the lower rates of HPV contamination on

disinfected endovaginal probes found in the literature might be

due to differences in examined population, hygiene practice or, as

suggested by some authors, the use of dry swabs resulting in a loss

of sensitivity [11]. In an earlier study, Amis et al. [10] indicated

that none of the condoms covered probes used for transvaginal

sonography, wiped with a dry tissue and then with a 70%

isopropyl alcohol towel were positive for herpesvirus but only few

probes were examined (n = 26) and alcohol is known to shorten the

working life of the probe. Recently, Kac et al [12] reported 1.5% of

viral (Epstein-Barr virus, human cytomegalovirus and HPV)

contamination on endovaginal/transrectal transducers after re-

moval of the probe covers. After HLD using both disinfection with

disinfectant impregnated towel and a 5-min cycle in an ultraviolet

C chamber, no viral genome was detected.

In the second subset of 100 samples screened by PCR for the

presence of C. trachomatis and mycoplasmas, 20% (95% CI: 12–28)

and 8% (95% CI: 3–13) were positive, respectively. To our

knowledge, no previous study has investigated the presence of

these organisms on ultrasound transducers. Primers used for C.

trachomatis detection amplify the cryptic 7.5-kb plasmid present in

all serotypes; false negative reactions thus should only be

encountered with the exceptional strains harboring a partly

deleted plasmid or no plasmid at all [18]. PCR amplification

used for Mycoplasma detection target 16S rRNA sequences that are

genus-specific and react with all members of the genera Mycoplasma

(including Mycoplasma genitalium and Mycoplasma hominis), Ureaplasma

(in particular Ureaplasma urealyticum), Spiroplasma and Acholeplasma

[19].

HPV cannot be propagated in tissue culture, and C. trachomatis

and mycoplasmas are difficult to cultivate. Therefore, the accurate

detection of these microorganisms in patients’ samples relies on

molecular biology techniques such as PCR amplifications, which

are the most sensitive and specific tests [19–21]. However, DNA

detection does not necessarily indicate the presence of viable and

infective microorganisms. In an effort to select for infectious viral

particles or bacteria, positive samples have been subjected to a

second PCR amplification after DNase treatment. The percent-

ages of positive samples fell by twofold for HPV (7%; 95% CI: 2–

12) and mycoplasmas (4%; 95% CI: 0–8). A tenfold decrease was

observed for C. trachomatis (2%, 95% CIX-X), may be reflecting

both its high prevalence in female genital tract [20] and its limited

survival in the environment and/or low resistance to disinfectants.

HPV is the most common sexually transmitted virus and is now

recognized as the major etiological cause of invasive cervical

cancer. C. trachomatis is the first agent of sexually transmissible

diseases, and causes in women, cervicitis, pelvic inflammatory

disease and its sequelae, i.e. infertility, ectopic pregnancy, and

chronic pelvic pain [20]. ‘‘Genital mycoplasmas’’ are frequently

isolated from the genital tract. M. genitalium is increasingly

identified as the causative agent of pelvic inflammatory disease

[22]; M. hominis, and U. urealyticum may induce a variety of

genitourinary infections [23]. Perinatal transmission from mother

to child has been demonstrated for HPV and C. trachomatis, the

latter being responsible for neonatal conjunctivitis and pneumonia

[20]. Genital mycoplasmas are involved in a number of adverse

outcomes of pregnancy [23]. Considering the clinical impact of

these pathogens their absence would have been desirable.

In the third subset of 100 samples taken for screening for

bacterial/fungal contamination, no agent specifically responsible

for sexually transmitted diseases such as Neisseria gonorrhoeae, or for

vaginosis/vaginitis such as Gardnerella vaginalis and Candida albicans

were found. In addition, organisms potentially deleterious for

neonates, such as group B streptococci or Escherichia coli, were

absent. Amis et al. [10] and Sykes et al. [14] did not evaluate the

presence of C. albicans. Kac et al. [12] searched fungi and did not

encounter any on endovaginal/transrectal probes even just after

removal of the cover probe. None of them looked for gonococcus

or gardnerella. Although all pathogenic organisms that can be

transmitted through endovaginal ultrasonography [24] have not

been investigated, the study has encompassed the most represen-

tative ones.

In contrast, 86% (95% CI: 79–93) of our samples, were

contaminated by commensal and/or environmental bacterial

flora, occasionally in mixed culture (Table 1). Skin flora, including

coagulase-negative staphylococci (CNS, 73%), Micrococcus sp.

(20%), methicillin susceptible Staphylococcus aureus (4%), viridans

streptococci (2%), and Corynebacterium sp. (1%) was found

predominantly and often in high numbers (10-.3000 CFU/

probe). Environmental flora (Pseudomonas stutzeri, Shewanella putrefa-

ciens and Aeromonas sp. 2%; Pseudomonas aeruginosa, Acinetobacter

baumannii, Flavobacterium oryzihabitans, and Comamonas acidovorans,

1%) was less represented and in lower amounts (10-90 CFU/

probe). It is difficult to say which ones are the ‘‘pathogenic

bacteria’’. Indeed, as suggested by Koibuchi et al., [25] even

coagulase-negative staphylococci and Corynebacterium spp. as some

Bacillus spp. can cause critical infectious diseases in immunosup-

pressed patients. S. aureus, a part of the skin microbiota is one of the

main causes of hospital-and community-acquired infections which

can have serious consequences, and methicillino-resistant strains

pose therapeutic issues [25]. Enterobacteria, which are the

dominant aerobic flora of the digestive tract, may also be

encountered [12]. Thus, outbreaks caused by S. aureus and SHV-

5 producing Klebsiella pneumoniae after endovaginal ultrasonography

have been reported [6,9]. Environmental flora, mainly composed

of non-fermentative Gram-negative bacilli is responsible for

nosocomial infections in debilitated patients. Outbreaks due to

P. aeruginosa [26,27], Burkholderia cepacia [28,29], Achromobacter

xylosoxidans [30] and recently multidrug resistant bacteria [31]

have been increasingly associated with transrectal ultrasonogra-

phy. These data indicate that full consideration of bacterial

contamination of endocavity ultrasound probes is essential. Sykes

et al., [14] under similar conditions as in this study, observed that

83.3% of the samples from the transvaginal ultrasound equipment

grew skin/environmental organisms, and 6.7% grew ‘‘potential

pathogens’’, one of which being S. aureus. Amis et al. [10] found

only one of 46 transvaginal probes positive for bacteria

(Acinetobacter spp.) using isopropyl alcohol wipes. Kac et al. [12]

reported 3.4% of contamination by pathogenic bacteria on

endovaginal/transrectal transducers, all of which disappeared

after HLD.

The mechanism by which probe contamination occurs is

unclear. With regard to specifically genital pathogens (HPV, C.

trachomatis, mycoplasmas), an incidental perforation of the probe

cover before/during the examination, or leakage of blood or

secretions at the open rim of the sheaths might be involved. In our

study, neither the damage of the cover nor the presence of blood

or other body fluids on the probe after cover removal was detected

by visual inspection. However, the possibility of contamination due

to microscopic damage of the sheaths cannot be excluded. The

Centers for Disease Control and Prevention and the American

Institute of Ultrasound in Medicine recommend the use of

condoms rather than cover probes because they are less prone

to perforations (1–9% and up to 81% in one study) [1,10].

However CE marked probe covers are preferred on the basis that

condoms are not adapted to all types of transducers and may have

Contaminated Transvaginal Probes despite LLD

PLOS ONE | www.plosone.org 3 April 2014 | Volume 9 | Issue 4 | e93368

a lower coverage of the heads. In the study of Kac et al. [12] both

types of covers performed similarly. In this study, considering the

high frequency and level of skin bacteria, manual contamination of

the probes should not be excluded, although necessary precautions

(e.g. trained personnel, use of gloves…) were taken. Alternatively,

in this large urban Radiology center with a high frequency

ultrasound usage, the probe may have been either inconsistently

cleaned and disinfected, or sporadically contaminated after LLD

procedure by the environment or by gloves previously in contact

with the external genitalia [11,13]. Samples taken from the

ultrasound room and the ultrasound gel did not show significant

microbial contamination.

In conclusion, this study demonstrates that a high proportion of

endovaginal ultrasound probes remain contaminated despite the

use of medically adapted probe covers and conventional LLD

procedure. Therefore, these equipments actually could represent a

potential vehicle for cross-transmission. To our knowledge there

are no data on how many of these pathogens have to be inoculated

in order to cause infection. Nevertheless, in order to prevent risks

of cross contamination; it is advisable that the endovaginal

ultrasound disinfection procedure is reviewed. More studies using

other brands of probe covers, and disinfecting towels or novel

decontaminating approaches are warranted.

Author Contributions

Conceived and designed the experiments: FM MK. Performed the

experiments: FM CB. Analyzed the data: FM SL YM CQ MK.

Contributed reagents/materials/analysis tools: FM SL. Wrote the paper:

FM CQ MK.

References

1. Rutala WA, Weber DJ, and the Healthcare Infection Control Practices Advisory

Committee (HICPAC) (2008) Guideline for disinfection and sterilization in

healthcare facilities. Available: http://www.cdc.gov/hicpac/pdf/guidelines/

disinfection_nov_2008.pdf. Accessed 2013 May 7.

2. American Institute of Ultrasound in Medicine (2003) Guidelines for Cleaning

and Preparing Endocavitary Ultrasound Transducers Between Patients.

Ultrasound Obstet Gynecol 7: 94.

3. Gray RA, Williams PL, Dubbins PA, Jenks PJ (2012) Decontamination of

transvaginal ultrasound probes: review of national practice and need for national

guidelines. Clin Radiol 67: 1069–77.

4. Backhouse S (2003) Establishing a protocol for the cleaning and sterilisation/

disinfection of ultrasound transducers. BMUS Bulletin 11: 37–39.

5. Haut Conseil de la Sante Publique (2007) Gaines de protection a usage unique

pour dispositifs medicaux reutilisables: recommandations d’utilisation. Rapport,

decembre 2007. Available: http://www.hcsp.fr/explore.cgi/hcspr20071214_

GainesProtec.pdf. Accessed 2013 May 7.

6. Gaillot O, Maruejouls C, Abachin E, Lecuru F, Arlet G, et al. (1998)

Nosocomial outbreak of Klebsiella pneumoniae producing SHV-5 extended-

Spectrum b-lactamase, originating from a contaminated ultrasonography

coupling gel. J Clin Microbiol 36: 1357–60.

7. Leroy S (2013) Infectious risk of endovaginal and transrectal ultrasonography:

systematic review and meta-analysis. J Hosp Infect 83: 99–106.

8. Lesourd F, Izopet J, Mervan C, Payen JL, Sandres K, et al. (2000) Transmissions

of hepatitis C virus during the ancillary procedures for assisted conception. Hum

Reprod 15: 1083–5.

9. Ohara T, Itoh Y, Itoh K (1998) Ultrasound instruments as possible vectors of

staphylococcal infection. J Hosp Infect 40: 73–7.

10. Amis S, Ruddy M, Kibbler CC, Economides DL, MacLean AB (2000)

Assessment of condoms as probe covers for transvaginal sonography. J Clin

Ultrasound 28: 295–8.

11. Casalegno JS, LeBail Carval K, Eibach D, Valdeyron ML, Lambin G, et al.

(2012) High risk HPV contamination of endocavity vaginal ultrasound probes:

an underestimated route of nosocomial infection? 7(10): e48137. doi:10.1371/

journal.pone.0048137

12. Kac G, Podglajen I, Si-Mohamed A, Rodi A, Grataloup C, Meyer G (2010)

Evaluation of ultraviolet C for disinfection of endocavitary ultrasound

transducers persistently contaminated despite probe covers. Infect Control Hosp

Epidemiol 31: 165–70.

13. Ma ST, Yeung AC, Chan PK, Graham CA (2012) Transvaginal ultrasound

probe contamination by the human papillomavirus in the emergency

department. Emerg Med J. Available: http://emj.bmj.com/content/early/

2012/07/02/emermed-2012-201407.long. Accessed 2013 May 7.

14. Sykes A, Apleby M, Perry J, Gould K (2006) An investigation of the

microbiological contamination of ultrasound equipment. Br J Infect Cont 7:

16–20.

15. Bloc S, Mercadal L, Garnier T, Komly B, Leclerc P, et al.(2011) Evaluation of a

new disinfection method for ultrasound probes used for regional anesthesia:

ultraviolet C light. J Ultrasound Med. 30: 785–8.

16. Kac G, Gueneret M, Rodi A, Abergel E, Grataloup C, et al. (2007) Evaluation

of a new disinfection procedure for ultrasound probes using ultraviolet light.

J Hosp Infect 65: 163–8.

17. De Roda Husman AM, Walboomers JMM, van den Brule AJC, Meijer CJLM,

Snijders PJF (1995) The use of general primers GP5 and GP6 elongated at their

3’ ends with adjacent highly conserved sequences improves human papilloma-

virus detection by PCR. J Gen Virol 76: 1057–62.

Table 1. Quantification and identification of the bacterial flora present on the transducers.

Bacterial species Number of positive samples Estimated number of CFU on the probe per sample

Gram positive cocci and bacilli

Coagulase-Negative Staphylococci 73 10-.3000

Micrococcus sp. 20 10-.3000

Staphylococcus aureus 4 10-560

Streptococcus sp. 2 10

Corynebacterium sp. 1 20

Gram negative bacilli

Pseudomonas stutzeri 2 10–20

Shewanella putrefaciens 2 20–90

Aeromonas sp 2 30–40

Flavobacterium oryzihabitans 1 10

Pseudomonas aeruginosa 1 30

Acinetobacter baumannii 1 10

Comamonas acidovorans 1 20

doi:10.1371/journal.pone.0093368.t001

Contaminated Transvaginal Probes despite LLD

PLOS ONE | www.plosone.org 4 April 2014 | Volume 9 | Issue 4 | e93368

18. Santos C, Teixeira F, Vicente A, Astolfi-Filho S (2003) Detection of Chlamydia

trachomatis in endocervical smears of sexually active women in Manaus-AM,Brazil, by PCR. Braz J Infect Dis 7: 91–5.

19. Van Kuppeveld FJM, van der Logt JTM, Angulo AF, van Zoest MJ, Quint

WGV, et al. (1992) Genus- and species-specific identification of Mycoplasmas by16S rRNA amplification. Appl Environ Microbiol 58:2606-15. Author’s

correction (1993) in Appl Environ Microbiol 59: 655.20. Bebear C, de Barbeyrac B (2009) Genital Chlamydia trachomatis infections.

Clin Microbiol Infect 15: 4–10.

21. Abreu AL, Souza RP, Gimenes F, Consolaro ME (2012) A review of methods fordetect human papillomavirus infection Virol J 9: 262 Abreu, et al. Virology

Journal 2012, 9: 262. Available: http://www.virologyj.com/content/9/1/262.Accessed 2013 May 7.

22. Mc Gowin CL, Anderson-Smits C (2011) Mycoplasma genitalium: an emergingcause of sexually transmitted disease in women. PLoS Pathog 7(5): e1001324.

doi: 10.1371/journal.ppat.1001324.

23. Judlin F (2003) Genital mycoplasmas. Gynecol Obstet Fertil 31: 054–9.24. Institut de veille sanitaire (2008) Analyse du risque infectieux lie aux

echographies endocavitaires en l’absence de protection ou de desinfection dessondes entre patients. Rapport fevrier 2008. Available: http://www.invs.sante.

fr/publications/2008/risque_echographie/080205_analyse_risque_infectieux_

echographies_vf.pdf. Accessed 2013 May 7.

25. Koibuchi H, Kotani K, Taniguchi N (2013) Ultrasound probes as a possible

vector of bacterial transmission. Med Ultrason 15: 41–44.

26. Gillespie JL, Arnold KE, Noble-Wang J, Jensen B, Arduino M, et al. (2007)

Outbreak of Pseudomonas aeruginosa infections after transrectal ultrasound-guided

prostate biopsy. Urology 69: 912–4.

27. Paz A, Bauer H, Potasman I (2001) Multiresistant Pseudomonas aeruginosa

associated with contaminated transrectal ultrasound. J Hosp Infect 49: 148–9.

28. Hutchinson J, Runge W, Mulvey M, Norris G, Yetman M, et al. (2004)

Burkholderia cepacia infections associated with intrinsically contaminated ultra-

sound gel: the role of microbial degradation of parabens. Infect Control Hosp

Epidemiol 25: 291–6.

29. Organ M, Grantmyre J, Hutchinson J (2010) Burkholderia cepacia infection of the

prostate caused by inoculation of contaminated ultrasound gel during tranrectal

biopsy of the prostate. Can Urol Assoc J 4: E58–E60.

30. Olshtain-Pops K, Block C, Temper V, Hidalgo-Grass C, et al. (2011) An

outbreak of Achromobacter xylosoxidans associated with ultrasound gel during

transrectal ultrasound guided prostate biopsy. J Urol 185: 144–7.

31. Williamson DA, Barrett LK, Rogers BA, Freeman JT, Hadway P, et al. (2013)

Infectious complications following transrectal ultrasound-guided prostate biopsy:

new challenges in the era of multidrug-resistant Escherichia coli. Clin Infect Dis.

In press.

Contaminated Transvaginal Probes despite LLD

PLOS ONE | www.plosone.org 5 April 2014 | Volume 9 | Issue 4 | e93368

8. Susceptibility of high-risk

human papillomavirus type 16 to

clinical disinfectants - Craig

Meyers Journal of Antimicrobial

Chemotherapy, 2014

Susceptibility of high-risk human papillomavirus type 16to clinical disinfectants

Jordan Meyers1†‡, Eric Ryndock2†, Michael J. Conway2§, Craig Meyers2* and Richard Robison1

1Department of Microbiology and Molecular Biology, Brigham Young University, Provo, UT 84602, USA; 2Department of Microbiology andImmunology, Pennsylvania State College of Medicine, Hershey, PA 17033, USA

*Corresponding author. Tel: +1-717-531-6240; Fax: +1-717-531-4600; E-mail: cmm10@psu.edu†Authors contributed equally.

‡Present address: Department of Medicine, Brigham and Woman’s Hospital, Boston, MA 02115, USA.§Present address: Department of Foundational Sciences, Central Michigan University, Mount Pleasant, MI 48859, USA.

Received 7 October 2013; returned 18 November 2013; revised 31 December 2013; accepted 6 January 2014

Objectives: Little to nothing is known about human papillomavirus (HPV) susceptibility to disinfection. HPV is esti-mated to be among the most common sexually transmitted diseases in humans. HPV is also the causative agentof cervical cancers and other anogenital cancers and is responsible for a significant portion of oropharyngeal can-cers. While sexual transmission is well documented, vertical and non-sexual transmission may also be important.

Methods: Using recombinant HPV16 particles (quasivirions) and authentic HPV16 grown in three-dimensionalorganotypic human epithelial culture, we tested the susceptibility of high-risk HPV to clinical disinfectants.Infectious viral particles were incubated with 11 common clinical disinfectants, appropriate neutralizers wereadded to inactivate the disinfectant and solutions were filter centrifuged. Changes in the infectivity titres ofthe disinfectant-treated virus were measured compared with untreated virus.

Results: HPV16 is a highly resistant virus; more so than other non-enveloped viruses previously tested. The HPV16quasivirions showed similar resistance to native virions, except for being susceptible to isopropanol, the triplephenolic and the lower concentration peracetic acid-silver (PAA-silver)-based disinfectant. Authentic virus andquasivirus were resistant to glutaraldehyde and ortho-phthalaldehyde and susceptible to hypochlorite and thehigher concentration PAA-silver-based disinfectant.

Conclusions: We present the first disinfectant susceptibility data on HPV16 native virions, which show that com-monly used clinical disinfectants, including those used as sterilants in medical and dental healthcare facilities,have no effect on HPV16 infectivity. Policy changes concerning disinfectant use are needed. The unusually highresistance of HPV16 to disinfection supports other data suggesting the possibility of fomite or non-sexual trans-mission of HPV16.

Keywords: hospital sterilants, papillomavirus, cancer, glutaraldehydes, ortho-phthalaldehydes

IntroductionDue to the specific life cycle requirements of human papilloma-virus (HPV), infectious virus has been difficult to produce in labora-tories and an assay for infectious virus has only recently becomeavailable. The ability to produce infectious virus outside of hostanimals is a great benefit to basic research; it often requires lesstime and is more cost-effective. HPV has a life cycle stringentlytied to differentiated epithelial tissue. This has required the devel-opment of special systems to make in vitro propagation possible.Because of the historical difficulty in producing high enough titresof infectious HPV particles and the lack of a suitable assay to testfor infectivity, little to nothing is known about HPV susceptibility to

disinfection. Disinfectants have been tested against many import-ant viruses and these studies are important to public health asthey provide information that can be used to reduce the preva-lence of infection, transmission and reinfection. Presently, hospi-tals’ and other healthcare institutes’ use of disinfectants toinactivate HPV is based on what is used for other viruses or simplyon what someone thinks should be effective. Two systems(recombinant based and organotypic) have been developed toproduce high amounts of infectious HPV particles in the labora-tory. Infectivity can now be measured by using reverse transcrip-tion quantitative PCR (RT-qPCR) that detects the viral E1^E4transcript. Detection of this transcript signals infectious particlesthat were able to achieve cell entry and start their early viral

# The Author 2014. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.For Permissions, please e-mail: journals.permissions@oup.com

J Antimicrob Chemother 2014; 69: 1546–1550doi:10.1093/jac/dku006 Advance Access publication 4 February 2014

1546

at University of N

orth Carolina at C

hapel Hill on A

pril 21, 2016http://jac.oxfordjournals.org/

Dow

nloaded from

programmes. HPV16 was used in these initial experimentsbecause it is responsible for up to 60% of all HPV-associatedcancers.

The adaptations of virus-like particle (VLP) technologies, quasi-virus and organotypic raft culturing systems have made it possibleto obtain high concentrations of HPV particles.1,2 Here, we presentthe first disinfectant susceptibility data on HPV16 native virions,which show that commonly used clinical disinfectants, includingthose used as sterilants in medical and dental healthcare facil-ities, have no effect on HPV16 infectivity. HPV16 is a highly resist-ant virus, more so than other non-enveloped viruses previouslytested. The HPV16 quasivirions showed similar resistance toHPV16 native virions, except that they were susceptible to isopro-panol, the triple phenolic and the lower concentration peraceticacid-silver (PAA-silver)-based disinfectant. Both authentic virusand quasivirus were resistant to glutaraldehyde (GTA) andortho-phthalaldehyde (OPA) and susceptible to hypochlorite andthe higher concentration PAA-silver-based disinfectant. Policychanges concerning disinfectant use are needed. Given theunusually high resistance of HPV16 to disinfection, our findingssupport other data suggesting the possibility of fomite or non-sexual transmission of HPV16.

Methods

Organotypic culturesHuman foreskin keratinocytes (HFKs) were isolated and maintained fromnewborn circumcision tissue specimens as previously described.3 The useof HFK tissues to develop cell lines for these studies was approved by theInstitutional Review Board at the Pennsylvania State University College ofMedicine and by the Institutional Review Board at Pinnacle HealthHospitals. Discarded, de-identified tissues were exempt from needinginformed patient consent. Informed consent was waived by bothInstitutional Review Boards. Full-length HPV16 was electroporated intolow-passage HFK monolayer cultures and keratinocyte lines shown to sta-bly maintain HPV16 genomes were selected for organotypic cultures.Organotypic ‘raft’ epithelial cultures were grown as previously described.4

Briefly, HPV16-infected HFK cells were seeded onto rat-tail type-1 collagenmatrices containing J2 3T3 feeder cells. After epithelial cells were grown toconfluence, these matrices were lifted onto stainless steel grids to estab-lish a liquid–air interface. Raft cultures were then fed by diffusion using

E medium. Raft culture epithelia were allowed to differentiate and thenharvested. Tissue was collected and stored for particle isolation.

Quasivirus productionQuasivirus was produced as described by Buck et al.1 with minor modifica-tions. Briefly, 293TT cell lines, known to express high levels of SV40 largeT antigen, were co-transfected with codon-modified HPV16 capsid genesL1 and L2 and with full-length HPV16. The HPV16 capsid gene plasmidcontained the SV40 origin of replication for expression. After 48 h, cellswere harvested for particle isolation.

Particle isolationRaft culture tissue and 293TTcells were homogenized in 0.6 mL of ice-cold1 M NaCl/0.05 M Na-phosphate buffer with a 7.5 mL homogenizer. Thehomogenizer was washed twice with 200 mL of 1 M NaCl/0.05 MNa-phosphate buffer, pH 8. The solution was centrifuged at 10500 rpmfor 10 min at 48C and the supernatant was transferred to a 1.8 mLNalgene cryovial. The crude viral stocks were stored at 2208C.

Disinfection testingThe disinfectants used in the study were ethanol (Fisher Scientific), isopro-panol (Sigma–Aldrich), GTA (Cidex and Cidexplus; Advanced SterilizationProducts), OPA (Cidex OPA; Advanced Sterilization Products), phenol(CiDecon; Decon Labs), PAA-silver (STERIPLEX SD and STERIPLEX SD Plus;sBioMed) and hypochlorite (Activate; Deardorff Fitzsimmons Corp.).

Each disinfectant was prepared according to manufacturer’s recom-mendation and stated dilution (see Table 1) immediately before testing.Virus and disinfectant were mixed thoroughly and incubated for 45 minat room temperature. After incubation, appropriate neutralizers wereadded to the virus/disinfectant solutions. The neutralizer used for thethree aldehyde disinfectants was 7% (w/v) glycine. The neutralizer usedfor the PAA-silver-based disinfectants contained 0.1% Tween 80, 1% pep-tone, 1% cysteine and 0.5 M Tris buffer, pH 7.5. This second general neu-tralizer was also used for all the other disinfectants tested. The solutionwas then centrifuged in Amicon Ultra centrifugal filters 100000 molecularweight cut-off (MWCO) (Millipore) at 4000 rpm for 10 min. More neutralizerwas added to the columns and they were recentrifuged at the same speedand duration. Two more washes were performed with HaCaT culturemedium prepared as previously described.5 The resulting liquid containingthe treated virus was collected and assayed for infectivity.

Table 1. Effectiveness of clinical disinfectants on HPV virions

Disinfectant Native virion (log10 reduction) +SD Quasivirion (log10 reduction) +SD

70% Ethanol 20.789 0.106 0.197 0.53095% Ethanol 20.076 0.481 0.307 0.12370% Isopropanol 20.770 0.186 4.675 0.41595% Isopropanol 20.272 0.023 4.435 0.1962.4% GTA 20.856 0.179 20.041 0.0143.4% GTA 20.306 0.232 20.145 0.2320.55% OPA 0.017 0.200 0.109 0.180Phenol 20.319 0.380 4.218 0.1440.25% PAA-silver 20.857 0.195 1.359 0.4081.2% PAA-silver 5.150 0.971 4.946 0.5480.525% Hypochlorite 4.862 0.623 5.087 0.413

All tests were performed at least five times and the averages are shown.

HPV16 disinfectant susceptibility

1547

JAC

at University of N

orth Carolina at C

hapel Hill on A

pril 21, 2016http://jac.oxfordjournals.org/

Dow

nloaded from

Infection and RT-qPCR infectivity testingHaCaT cells, an immortalized keratinocyte line kindly provided by NorbertFusenig (German Cancer Research Center), were seeded onto 24-wellplates at a density of 50000 cells/well. Cells were grown and maintainedin the HaCaT culture medium mentioned previously and incubated at 378Cin 5% CO2. After 48 h, the disinfectant-treated virus was added to the sub-confluent cells. Virus and cells were incubated for 48 h at 378C. The abilityto infect HaCaT cells after 48 h of incubation was determined by the pres-ence of the spliced HPV16 E1^E4 mRNA species. Total mRNA was har-vested with the SurePrep TrueTotal RNA Purification Kit (Fisher Scientific)and diluted to a final concentration of 100 ng/mL with 500 ng of totalRNA in each PCR. Amplification of both the viral target and endogenouscellular control target was performed as described previously, except allPCRs were performed in a 7900HT SDS thermal cycler and 96-well opticalreaction plates (Applied Biosystems).5 Relative expression was calculatedusing the 2–DDCt method, as efficiency of both the housekeeping gene andthe target gene was found to be similar.

ResultsThe disinfectants tested in this study were 70% and 95% ethanoland isopropanol, 2.4% and 3.4% GTA, 0.55% OPA, a triple phen-olic, a 0.25% and 1.2% PAA-silver-based disinfectant and0.525% hypochlorite (Table 1). The alcohols were chosen due totheir widespread use as surface disinfectants and hand sanitizers,both in public and clinical settings. Considering previous studiesthat show high levels of HPV DNA on fingers, especially in patientswith current genital infections, we sought to determine whetheralcohols were effective against HPV.6,7 GTA and OPA were testeddue to their widespread use as disinfectants, and in the case ofGTA also as a sterilant, in medical and dental healthcare facilities.The remaining disinfectants were chosen as representatives ofother common disinfectant types.

In HPV16 authentic viruses, only hypochlorite (4.86 log10 reduc-tion) and the 1.2% PAA-silver-based disinfectant (5.15 log10

reduction) were able to produce .99.99% reduction in infectivity.All other disinfectants showed slight or no reduction in infectivity(Table 1).

Comparing the two systems we used to produce virus particles,the organotypic culture system is more time-consuming andmore expensive. This has made the recombinant particle technol-ogy very attractive to researchers. However, in organotypic cul-tures, the tissue differentiation along with the accompanyingphysiological conditions that occur in stratified tissue are moresimilar to what would be found in mucosal tracts normallyinfected by HPV16. We sought to investigate whether differencesexist, with respect to disinfectant susceptibility, between HPV16quasivirus and HPV16 organotypic-produced virus. HPV16 quasi-virions were subjected to the same disinfectants and conditionsas the raft virions to compare their susceptibility profiles (Table 1and Figure 1). The results suggest that quasivirions are generallymore susceptible to disinfection than organotypic tissue-derivedvirus. Specifically, quasivirions showed a susceptibility to both70% (4.68 log10 reduction) and 95% (4.44 log10 reduction) isopro-panol and the triple phenolic (4.22 log10 reduction) disinfectantthat was not seen in the raft-produced virions. Also, the lower per-centage PAA-silver-based disinfectant (0.25%) produced a mod-erate reduction in infectivity (1.36 log10 reduction) of quasivirusthat was not seen with organotypic tissue-derived virus.

It is interesting to note that both GTA and OPA were ineffectivein producing any significant reduction in infectivity for both par-ticle types. GTA was further tested at longer contact times (upto 48 h) and still failed to inactivate HPV to any measurabledegree (data not shown). One proposed disinfection mechanismfor aldehydes is the cross-linking of important lysine residues inviral capsid proteins. Downstream effects of these capsid modifi-cations may include inhibition of viral entry, uncoating and trans-port of viral genomes to the correct cellular compartmentsnecessary to initiate viral programmes.8 Aldehydes are alsoeffective at denaturing nucleic acids, suggesting that if viral capsid

6

NativeQuasi

***

**

***

***

***

***5

4

3

Log

10 r

ed

uc

tio

n o

f vir

al

titr

e

2

1

0

–1

–2

70

% E

tha

no

l9

5%

Eth

an

ol

70

% I

sopro

pa

no

l9

5%

Iso

pro

pa

no

l

2.4

% G

TA

3.4

% G

TA

0.5

5%

OPA

Phen

ol

0.2

5%

PA

A-

silv

er

1.2

% P

AA

-si

lver

0.5

25

%H

ypo

chlo

rite

Figure 1. Susceptibility of HPV virions to clinical disinfectants. Both authentic virus and quasivirus were incubated with the indicated disinfectants for acontact time of 45 min. Disinfectants were neutralized and the virus was added to HaCaT cells for infection. Data shown are the averages of at least fiveindependent experiments, with errors shown as the standard deviations of all experiments. **P,0.01. ***P,0.001.

Meyers et al.

1548

at University of N

orth Carolina at C

hapel Hill on A

pril 21, 2016http://jac.oxfordjournals.org/

Dow

nloaded from

integrity was compromised, the viral DNA or RNA might also beinactivated. Knappe et al.9 showed that there are lysine residuesin the L1 capsid protein that are important for efficient infection.Structural data on bovine papillomavirus have also shown lysineresidues located on the invading arm of the L1 protein that maybe important for pentamer integrity.10 Although quasivirions andorganotypic-derived virions differed with respect to their suscep-tibility to certain disinfectants, their common resistance to thealdehydes leads to the assumption that these residues are eithernot accessible or do not exhibit the proper spatial separationnecessary for cross-linking. This commonality suggests thatthese two types of particles have some structural similarities.The two disinfectants that were effective against both particletypes were the 1.2% PAA-silver-based formulation and the hypo-chlorite. Both agents are strong oxidizers, suggesting thatoxidization-based disinfectants may be effective at inactivat-ing HPV.

GTA- and OPA-based disinfectants are often used in protocolswith longer contact times. Therefore, we tested both GTA (3.4%)and OPA (0.55%) at a contact time of 24 h on native HPV16 vir-ions. Testing was completed as described above. The 24 h GTA(20.039 ± 0.129) and OPA (20.270 ± 0.620) treatments did notresult in a decrease in viral infectivity (Table 2).

The addition of neutralizers alone, coupled with centrifugationthrough Amicon Ultra centrifugal filters, had no effect on infectiv-ity. All studies shown in Figure 1 and Table 1 were repeated at leastfive times. All studies in Table 2 were repeated six times. Due to theimportance of the results concerning nosocomially and iatrogeni-cally acquired infections, all studies were performed in two loca-tions, by different researchers, with multiple batches of virusand cells.

DiscussionTo date, this is the first report of a virus shown to be resistant toinactivation by GTA. Because of its broad range of effectivenessagainst other microbial pathogens, GTA has been the disinfectantof choice in many clinical situations. It has been proven effectiveagainst a large number of non-enveloped viruses, including ade-noviruses, parvoviruses, caliciviruses and many entero-viruses.8,11,12 Other than the resistance to GTA and OPA, HPV16shows many similarities to other highly resistant non-envelopedviruses such as caliciviruses and parvoviruses. These viruses areall susceptible to high concentrations of hypochlorite and a parvo-virus in one study was also susceptible to a PAA-silver-based dis-infectant as well as PAA alone.11 Non-enveloped viruses aregenerally mildly or completely resistant to alcohol-based

disinfection and that resistance was also seen with organotypic-derived HPV16 in our system.

Numerous studies have suggested, albeit based on limiteddata, that non-sexual transmission routes exist for HPV. Smithet al.13 evaluated concordance of HPV detection in mothers andinfants. They found that among HPV DNA-positive mothers, asmall percentage of their infants also had detectable HPV DNA.Interestingly, this study also found that two infants were HPVDNA-positive, whereas their mothers had no detectable HPVDNA. Samples were taken at birth and the PCR assays were per-formed in a separate facility to lower the risk of maternal inocula-tion and contamination. These findings suggest that in clinicalsettings there exists a low risk of nosocomial infection with HPV.Taken together with our disinfectant data, it is conceivable thatmany of the disinfectants currently used may not be effective ineliminating hospital sources of infectious HPV. Studies looking atprevalence among women with no reported sexual history haveyielded mixed results.14,15 Kjaer et al.14 followed a group of virginsfor 2 years and at the start of the study all were HPV16DNA-negative and seronegative to VLP HPV16. Only those whoinitiated sexual intercourse during the study were found to havedetectable HPV16 DNA and/or seroconvert. Tay et al.15 deter-mined the occurrence of HPV infection by colposcopy and hist-ology and found that .50% of the virgins in their study hadcolposcopic evidence of HPV infection with .80% of those con-firmed by histology. The variations seen in these studies may bedue to differences in detection methods, sample size, geographyor even processes of patient referral. These variations in methodsand design are common and often lead to the inability to drawclear conclusions about some aspects of the natural historyof HPV infection. Our data support the possibility of HPVfomite-related transmission, autoinoculation and nosocomialtransmission, even via instruments that were considered ‘sterile’.Because native HPV16 has been shown to be very resistant tochemical disinfectants, especially those routinely used in handsanitizers and instrument processing, new infection control proce-dures are warranted. As routine disinfection and hygiene may notbe sufficient to remove possible sources of HPV, the risk of reinfec-tion may be compounded in populations with active infections.Additionally, our data show that quasivirions and similar particlesmay not accurately measure the effectiveness of disinfectantson HPV.

AcknowledgementsExcellent technical help was provided by Ms Janice Melici and Terri Bills.

FundingThis study was funded in part by the National Institute of Allergy andInfectious Diseases (R01AI57988) and the BYU Mentoring EnvironmentGrant (MEG) Program.

Transparency declarationsC. M. has received speaker honoraria from Merck, Quest Diagnostics, GSK,Wyeth and Bristol-Myers Squibb, and has performed research funded byMerck, The Phillip Morris External Research Program, NexMed, GSK,

Table 2. Efficacy of disinfectants on HVP16 virions at 24 h contact time

DisinfectantNative virion

(log10 reduction) +SDQuasivirion

(log10 reduction) +SD

3.4% GTA 20.039 0.129 20.128 0.1140.55% OPA 20.270 0.620 0.058 0.052

Data shown are the averages of six independent experiments.

HPV16 disinfectant susceptibility

1549

JAC

at University of N

orth Carolina at C

hapel Hill on A

pril 21, 2016http://jac.oxfordjournals.org/

Dow

nloaded from

OriGenix and Interferon Sciences Inc. R. R. has performed research fundedby sBioMed, Inc. All other authors: none to declare.

References1 Buck CB, Pastrana DV, Lowy DR et al. Generation of HPV pseudovirionsusing transfection and their use in neutralization assays. Methods MolMed 2005; 119: 445–62.

2 Meyers C, Frattini MG, Hudson JB et al. Biosynthesis of humanpapillomavirus from a continuous cell line upon epithelial differentiation.Science 1992; 257: 971–3.

3 Visalli RJ, Courtney RJ, Meyers C. Infection and replication of herpessimplex virus type 1 in an organotypic epithelial culture system. Virology1997; 230: 236–43.

4 Meyers C, Mayer TJ, Ozbun MA. Synthesis of infectious humanpapillomavirus type 18 in differentiating epithelium transfected withviral DNA. J Virol 1997; 71: 7381–6.

5 Conway MJ, Alam S, Christensen ND et al. Overlapping and independentstructural roles for human papillomavirus type 16 L2 conserved cysteines.Virology 2009; 393: 295–303.

6 Sonnex C, Strauss S, Gray JJ. Detection of human papillomavirus DNA onthe fingers of patients with genital warts. Sex Transm Infect 1999; 75: 317–9.

7 Winer RL, Hughes JP, Feng Q et al. Detection of genital HPV types infingertip samples from newly sexually active female university students.Cancer Epidemiol Biomarkers Prev 2010; 19: 1682–5.

8 Chambon M, Archimbaud C, Bailly JL et al. Virucidal efficacy ofglutaraldehyde against enteroviruses is related to the location of lysineresidues in exposed structures of the VP1 capsid protein. Appl EnvironMicrobiol 2004; 70: 1717–22.

9 Knappe M, Bodevin S, Selinka HC et al. Surface-exposed amino acidresidues of HPV16 L1 protein mediating interaction with cell surfaceheparan sulfate. J Biol Chem 2007; 282: 27913–22.

10 Wolf M, Garcea RL, Grigorieff N et al. Subunit interactions in bovinepapillomavirus. Proc Natl Acad Sci USA 2010; 107: 6298–303.

11 Eterpi M, McDonnell G, Thomas V. Disinfection efficacy againstparvoviruses compared with reference viruses. J Hosp Infect 2009;73: 64–70.

12 D’Souza DH, Su X. Efficacy of chemical treatments against murinenorovirus, feline calicivirus, and MS2 bacteriophage. Foodborne PathogDis 2010; 7: 319–26.

13 Smith EM, Parker MA, Rubenstein L et al. Evidence for verticaltransmission of HPV from mothers to infants. Infect Dis Obstet Gynecol2010; 2010: 326369.

14 Kjaer SK, Chackerian B, van den Brule AJC et al. High-risk humanpapillomavirus is sexually transmitted: evidence from a follow-up studyof virgins starting sexual activity (intercourse). Cancer Epidem Biomar2001; 10: 101–6.

15 Tay SK, Ho TH, Sookim LT. Is genital human papillomavirus infectionalways sexually-transmitted? Aust NZ J Obstet Gyn 1990; 30: 240–2.

Meyers et al.

1550

at University of N

orth Carolina at C

hapel Hill on A

pril 21, 2016http://jac.oxfordjournals.org/

Dow

nloaded from