ORIGINAL RESEARCH

Phlebitis risk varies by peripheral venous catheter site and increases

after 96 hours: a large multi-centre prospective study

Giancarlo Cicolini*, Lamberto Manzoli*, Valentina Simonetti, Maria Elena Flacco, Dania

Comparcini, Lorenzo Capasso, Angela Di Baldassarre & Ghaleb Eltaji Elfarouki

Accepted for publication 1 March 2014

Correspondence to G. Cicolini:

e-mail: g.cicolini@unich.it

*These authors equally contributed to this

study.

Giancarlo Cicolini MSN PhD

Nurse Director

Department of Medicine and Aging

Sciences, University of Chieti-Pescara, Italy

and ASL 02 Abruzzo, Italy

Lamberto Manzoli MD MPH

Associate Professor of Epidemiology and

Public Health

Department of Medicine and Aging

Sciences, University of Chieti-Pescara, Italy

Valentina Simonetti MSN RN

PhD Student

Department of Medicine and Aging

Sciences, University of Chieti-Pescara, Italy

Maria Elena Flacco MD

Graduate School of Public Health Intern

Department of Medicine and Aging

Sciences, University of Chieti-Pescara, Italy

Dania Comparcini MSN RN

PhD Student

Department of Medicine and Aging

Sciences, University of Chieti-Pescara, Italy

Lorenzo Capasso MD

PhD Student

Department of Medicine and Aging

Sciences, University of Chieti-Pescara, Italy

C ICOL IN I G . , MANZOL I L . , S IMONETT I V . , F LACCO M.E . , COMPARC IN I

D . , CAPAS SO L . , D I BALDASSARRE A . & ELTA J I ELFAROUK I G . ( 2 0 1 4 ) Phle-

bitis risk varies by peripheral venous catheter site and increases after 96 hours: a

large multi-center prospective study. Journal of Advanced Nursing 00(0), 000–

000. doi: 10.1111/jan.12403

AbstractAims. This multi-centre prospective field study evaluated whether peripheral

venous catheter site of insertion influences the risk of catheter-related phlebitis.

Potential predictors of phlebitis were also investigated.

Background. Millions of patients worldwide use peripheral venous catheters,

which frequently cause local complications including phlebitis, infection and

obstruction. Although phlebitis predictors have been broadly investigated,

uncertainties remain on the potential effect of cannulation anatomical site,

duration and the appropriate time for catheter removal.

Design. A prospective cohort design was carried out from January–June 2012.

Methods. The clinical course of each patient who received a new peripheral

venous catheter for any cause in five Italian hospitals was followed by trained

nurses until catheter removal. The presence of phlebitis was assessed every

24 hours using the Visual Infusion Phlebitis score. Analyses were based upon

multilevel mixed-effects regression.

Results. The final sample consisted of 1498 patients. The average time for

catheters in situ was 65�6 hours and 23�6% of the catheters were in place beyond

96 hours. Overall phlebitis incidence was 15�4%, 94�4% of which were grade 1.

The likelihood of phlebitis independently increased with increasing catheter

duration, being highest after 96 hours. Compared with patients with catheter

placed in the dorsum of the hand (22�8% of the sample), those with the catheter

located in the antecubital fossa (34�1%) or forearm were less likely to have a

phlebitis of any grade.

Conclusions. Antecubital fossa and forearm veins may be preferential sites for

peripheral venous cannulation. Our results support Centers for Disease Control

and Prevention recommendations to replace catheters in adults no later than

96 hours. A relevant proportion of healthcare personnel did not adhere to such

guidelines – more attention to this issue is required.

© 2014 John Wiley & Sons Ltd 1

continued on page 2

Angela Di Baldassarre MD PhD

Professor of Motor Sciences

Department of Medicine and Aging

Sciences, University of Chieti-Pescara, Italy

Ghaleb Eltaji Elfarouki MPH PhD

Professor

Center for Research on Health and Social

Care Management, SDA Bocconi, Milan,

Italy

Keywords: catheter site, catheter time in situ, catheterization, iatrogenic events,

infusion-related phlebitis, nurses, nursing, peripheral venous catheter, phlebitis,

thrombo-phlebitis

Introduction

Peripheral venous catheters (PVCs) are used in the care of

millions of patients worldwide, usually for intravenous fluid

infusion and other vital, often life-saving, clinical interven-

tions (McCallum & Higgins 2012). It has been estimated

that as many as one in three hospital patients have a PVC

in place at any given time (Reilly et al. 2007).

The use of PVCs puts patients at risk for local and systemic

infective complications, including bloodstream infections

(Cicolini et al. 2009). While PVCs rarely cause systemic

complications (Maki et al. 2006), they are frequently associ-

ated with phlebitis, catheter-related infection and obstruction

of the catheter (Bregenzer et al. 1998); complications which

lead to longer admissions, additional healthcare costs and

discomfort and morbidity for patients (McCallum & Higgins

2012). This has evoked the need for thorough patient assess-

ment and careful catheter management and observation and

several studies sought to investigate the determinants of phle-

bitis to devise strategies and guidelines for risk reduction

(Maki & Ringer 1991, Lundgren et al. 1993, Bregenzer et al.

1998, Curran et al. 2000, Karadag & Gorgulu 2000, Cornely

et al. 2002, Karadeniz et al. 2003, Uslusoy & Mete 2008,

Rego Furtado 2011).

In addition to patient-related factors, which are often

resistant to modification, a significant relationship was doc-

umented, independently and in combination, between phle-

bitis risk and catheter material and size (needle gauge)

(Maki & Ringer 1991, Lundgren et al. 1993, Bregenzer

et al. 1998, Curran et al. 2000, Cornely et al. 2002, Abbas

et al. 2007, Singh et al. 2008, Uslusoy & Mete 2008, Cico-

lini et al. 2009, Forni et al. 2010). Cicolini et al. (2009),

for example, found that phlebitis incidence was significantly

lower in patients with 22 gauge vs. 16–18 gauge cannulas.

The frequency of catheter insertion and indwelling time

have also been found to be related to phlebitis risk (Maki

& Ringer 1991, Bregenzer et al. 1998, Webster et al. 2007,

2008, 2010, Uslusoy & Mete 2008, Lee et al. 2009, Rego

Furtado 2011, Mestre Roca et al. 2012, Rickard et al.

2012). Finally, the type of infused fluid was also found to

affect phlebitis risk and applying aseptic, non-touch, princi-

ples at insertion reduced the incidence of infective compli-

cations (Maki & Martin 1975, Melly et al. 1975, Sheth

et al. 1983, Maki & Ringer 1991, O’Grady et al. 2011).

Some major uncertainties remain on the relationship

between catheter anatomical site and risk of phlebitis: the

latest Centers for Disease Control and Prevention (CDC)

guidelines recommend the use of the upper extremities for

peripheral catheterization with no more detailed or specific

Why is this research needed?

• The latest Centers for Disease Control and Prevention

guidelines recommend the use of the upper extremities for

peripheral catheterization with no specification of preferred

anatomical site.

• There is a debate whether dwell times of intravenous cann-

ulae could be safely extended after 96 hours if clinically

indicated.

What are the key findings?

• Placing peripheral venous cannulae in the antecubital fossa

and forearm veins rather than in the dorsum of the hand

may reduce the risk of phlebitis by 30–50%.

• The incidence of phlebitis was highest for catheters left in

place more than 96 hours.

• Healthcare personnel often do not follow current recom-

mendations of catheter indwelling time.

How should the findings be used to influence policy/practice/research/education?

• Antecubital fossa and forearm veins may be the preferred

anatomical sites for peripheral intravenous cannulation.

• Peripheral venous catheters should remain in situ not more

than 96 hours after insertion.

• Operators should be sensitized to the clinical importance to

follow current recommendation on peripheral venous cann-

ulation management.

2 © 2014 John Wiley & Sons Ltd

G. Cicolini et al.

recommendation about the preferred site of insertion to

minimize risk of infection (O’Grady et al. 2011).

Background

Many authors investigated the influence of anatomical site

on phlebitis rate with contrasting results; a major source of

difficulty is the wide heterogeneity in defining and classify-

ing anatomical sites among studies (Maki & Ringer 1991,

Bregenzer et al. 1998, White 2001, Cornely et al. 2002,

Karadeniz et al. 2003, Kagel & Rayan 2004, Webster et al.

2007, Uslusoy & Mete 2008, Cicolini et al. 2009, Lee et al.

2009, Forni et al. 2010, Rego Furtado 2011, Mestre Roca

et al. 2012). Further, the definition of catheter-related infec-

tions remains unclear in the literature and clinical presenta-

tions vary. Thus, what constitutes the preferred catheter

site in terms of infection risk rather than convenience

remains an unresolved issue.

To our knowledge, only one study investigated and com-

pared, as its main objective, the relationship between differ-

ent anatomical sites and incidence of phlebitis with a view to

identifying the safest location (Cicolini et al. 2009). This

observational study was conducted on a sample of 427

patients from surgical and medical wards in one Italian hos-

pital; the incidence of phlebitis was found to be highest in

patients with cannulae inserted in the dorsal aspect of the

hand. The authors therefore advocated the use of antecubital

fossa veins rather than hand and forearm veins. Phlebitis was

assessed using the checklist from (Lundgren et al. 1993); all

insertions were carried out by nurses. A major limitation was

that the proportion of patients receiving catheters in the dor-

sal aspect of the hand was significantly lower than that of

other sites; furthermore, many patients were admitted to gen-

eral surgery wards where the lengths of stay were typically

shorter than for internal medicine units.

Another unsolved issue pertains to the effect of catheter

duration and the appropriate time for removal: prominent

national guidelines state that elective removal of peripheral

intravenous catheters should be considered if the PVC has

been in situ for 72–96 hours to reduce risk of complications

(Department of Health 2011). Other guidelines have been

more ambitious, suggesting replacement every 48–72 hours

(Tagalakis et al. 2002). In fact, some studies have been con-

ducted to investigate the possible merit of elective replace-

ment, i.e. in the absence of any clinical complications, of

peripheral intravenous catheters to prevent phlebitis occur-

ring. A systematic review conducted to search for relevant

randomized controlled trials identified three small trials,

with samples in two of the trials including patients requir-

ing parenteral nutrition (Idvall & Gunningberg 2006).

Although time intervals for elective replacement in the stud-

ies did vary, the overall conclusion was of limited scientific

evidence, suggesting but not confirming that elective

replacement of peripheral intravenous catheters reduces the

incidence (and severity) of phlebitis. In these studies, how-

ever, data collectors were not blinded as to whether the

patients were in intervention or control groups; and control

groups were those, where catheters were replaced on signs

and symptoms of complication and not merely on an elec-

tive scheduled basis.

The study

Aim

The primary objective was to evaluate whether PVC site of

insertion influences the risk of catheter-related phlebitis. We

also investigated other potential predictors of phlebitis with

a special emphasis on indwelling time (time in situ, i.e. dura-

tion between insertion and removal) for venous catheters.

Design

A prospective cohort design was carried out on a mixed

population of patients from five Italian hospitals (Ancona,

Chieti, Gallipoli, San Benedetto del Tronto, Vasto) based

on three different regions along the Adriatic coast (Marche,

Abruzzo and Puglia).

Participants

The criteria for inclusion were: patients admitted to medical

or surgical wards in an ordinary setting (no day-hospital or

day-surgery), age 18 years or above, written informed con-

sent. Patients were excluded if they had recognized cogni-

tive impairments, were on renal dialysis, were admitted or

had their catheter placed in an emergency department, were

known to be immune-compromised or had known blood

infections.

Sample size estimation

Sample size estimation was based on the following parame-

ters: 10% and 20% of phlebitis at the end of the follow-up

in the subjects with cannulae in the antecubital fossa and in

the dorsum of the hand, respectively (Cicolini et al. 2009);

alpha 0�05; 5% loss to follow-up. Using such parameters in

a two-tailed Chi-square test, 234 subjects were required in

each group to achieve 80% statistical power. Considering

multiple adjustments in multivariate modelling, we conser-

vatively decided to enrol at least 334 patients in each group.

© 2014 John Wiley & Sons Ltd 3

JAN: ORIGINAL RESEARCH Predictors of infusion-related phlebitis

Data collection

Study protocol and outcomes

Five trained nurses (one for each hospital), in coordination

with local personnel, took part in recruiting eligible

patients, collecting and recording their data up to the

96 hours after admission; it was presumed that all newly

inserted catheters would be removed by then. The 96-hour

rule has been violated in several cases and this is discussed

in detail in the results section. The recruiting nurses acted

as observers and could not make decisions to insert cathe-

ters or remove them; they did however remind local staff of

current CDC guidelines (O’Grady et al. 2011). This divi-

sion of labour was adopted both to simulate best routine

clinical practice conditions and to reduce the potential influ-

ence of local health personnel on data recording and collec-

tion. The primary outcome of interest was the occurrence

of phlebitis; the presence of phlebitis (and degree or grade)

was assessed at 24, 48, 72 and 96 hours following catheter

insertion. Thrombo-phlebitis is defined as the presence of at

least two symptoms/signs among erythema, swelling or pal-

pable venous cord (Maki & Ringer 1991).

The research nurses were trained to recognize and classify

infusion-related phlebitis implementing the Visual Infusion

Phlebitis score proposed by Gallant and Schultz (Gallant &

Schultz 2006) with grades of severity from 1–5. In addition

to the number of hours for catheters in situ and the ana-

tomical placement site, the following variables were

recorded at the time of catheter insertion for all partici-

pants: age, gender, body mass index (BMI), smoking status,

place of admission, category of admission, type of person-

nel inserting catheter, whether patient was on antibiotic

therapy, type of infusion, cannula needle gauge, catheter

material and type of dressing used.

In the previous single-site observational study, three

upper extremity venous access locations were defined: ant-

ecubital fossa, forearm veins and the dorsal aspect of the

hand (Cicolini et al. 2009). Both studies have similar

exclusion criteria and it was decided to use the same clas-

sifications for cannula position. More specifically, when

we refer to antecubital fossa veins, they include the med-

ian cubital vein, median cephalic vein, the cephalic vein

(along the lateral aspect of the antecubital fossa) and the

basilic vein (on the postero-medial aspect of the wrist);

forearm veins include the superficial median vein of the

forearm, the cephalic vein (along the radial aspect of the

forearm) and the part of the basilic vein posterior to the

ulnar border of the forearm; hand veins include the dorsal

metacarpal veins. The different factors mentioned thus far

were each considered as potential predictors of phlebitis or

confounders of the association between catheter site and

phlebitis.

Ethical considerations

The study protocol received full approval from the institu-

tional review board and the local ethics committee of the

coordinating centre at the University of Chieti.

Data analysis

Initially the difference in the incidence of phlebitis of any

grade during follow-up according to the anatomical site of

the catheter was investigated using stratification per each

defined time-point: 24, 25–48, 49–72, 73–96 and 97–

120 hours. Kaplan–Meier survival estimates and log-rank

test by site were also used without stratification, including

the whole follow-up period. The validity of constant inci-

dence ratios over the follow-up period was checked using

Nelson-Aalen cumulative hazard estimates (Hosmer &

Lemeshow 1999).

The potential independent predictors of phlebitis were

evaluated using Cox proportional hazards analysis and then

multilevel mixed-effects regression, the latter to estimate the

potential impact of time (Rabe-Hesketh & Skrondal 2008).

In both analyses, all recorded covariates were included into

the final models a priori, although those that are continuous

or ordinal variables were tested for transformation into sev-

eral alternative dummy variables. Multicollinearity, interac-

tions and higher-power terms were tested for all covariates

in Cox analysis; Schoenfeld test was performed to check

the validity of proportional hazards assumption.

Although it is not specifically recommended in most guide-

lines to remove the catheter on the detection of grade 1 phle-

bitis (slight pain or redness close to the insertion site), this

was requested by the approving ethics committee. We there-

fore advised personnel in participating centres to remove

catheters on observing phlebitis of any grade. During the

study, catheters remained in situ despite a grade 1 phlebitis

in 32 cases, but only until one subsequent time-point. In the

analysis, all patients including the above 32 were censured at

the time-point where phlebitis was observed.

The mixed model was set as repeated regression analysis

(with one observation every 24 hours of follow-up), using a

patient’s identification number and hospital/centre as the

cluster levels, both cases assuming an independent correla-

tion structure. However, all models were repeated using an

exchangeable correlation structure, with marginal increases

in standard errors and no qualitative change. Time was

included in the model either as an ordinal variable or using

4 © 2014 John Wiley & Sons Ltd

G. Cicolini et al.

dummy variables for each 24-hour unit (0–24 hours being

the reference category). Given that all dummy odds ratios

(ORs) of time were significantly associated with phlebitis

and were quite linearly decreasing, time was treated as

ordinal in the final model: the OR of time is thus to be

interpreted as the variation in the odds of phlebitis for

every 24-hour increment (Parruti et al. 2010).

Given that the results of the survival and multilevel analy-

ses were substantially similar, we only reported the estimates

from the mixed model to avoid redundancy and include the

estimate of the effect of time. A two-tailed P value of 0�05was deemed significant for all analyses, which were per-

formed using Stata 10.1 software package (Stata Corp.,

College Station, TX, USA, 2007). Results with averages are

given along with 95% confidence intervals.

Validity and reliability

We followed participants’ clinical course using the Visual

Infusion Phlebitis (VIP) scale, that has shown content valid-

ity, inter-rater reliability higher than 85% and clinically

feasibility (Gallant & Schultz 2006) and it is recommended

in the Infusion Nursing Standards of Practice to determine

when an intravenous catheter should be removed (Infusion

Nurses Society 2006).

Results

Recruitment was carried out from January–June 2012,

inviting a total of 1539 (eligible) inpatients. Only 41 eligi-

ble patients did not consent to participate and thus the final

sample was 1498 patients (mean age 60�8, SD 20�7; 41�2%males – Table 1). Overall, 22�8% of the participants had

their catheter inserted on the dorsum of the hand, 34�1% in

the antecubital fossa and 41�1% in veins in the forearm.

The average time for catheters in situ was 65�6 hours (SD

34�5), a relatively lengthy period considering that many

patients had their cannulae removed due to phlebitis or no

longer requiring them. Although current guidelines recom-

mend that all peripheral catheters be removed and replaced

if necessary before reaching 96 hours, in 23�6% of cases the

same catheter was in place beyond the 96 hour of follow-

up (Table 1). Often this was not an oversight but due to

conscious decisions from personnel or resistance from

patients. In such cases, given that the study was not inter-

ventional in nature, the researchers’ role was confined to

strongly recommending the immediate removal of over-stay-

ing catheters to local health professionals. We subsequently

planned an additional visit at 120 hours postinsertion,

where was evaluated whether catheters were removed and

Table 1 Detailed characteristics of the study sample (n = 1498).

Variables Overall Variables Overall

Mean age in

years, SD

60�8 (20�7) Catheter time

in situ, hours

Age class in

years, % (n)

Mean time (SD) 65�6 (34�5)

18–39 21�2 (318) ≤24 13�7 (205)

40–59 35�8 (536) 25–48 21�8 (327)

60–100 43�0 (644) 49–72 21�9 (328)

73–96 19�0 (284)

Male gender,% (n) 41�2 (617) 97–120 23�6 (354)

Mean Body Mass

Index (weight/

height2) (SD)

25�5 (4�1) Catheter

site, % (n)

BMI class, % (n) Dorsum of

the hand

22�8 (341)

<20 5�2 (78) Antecubital

fossa

34�1 (511)

20–24�99 44�5 (666) Forearm

veins

43�1 (646)

25–29�99 39�3 (589)

≥30 11�0 (165) Type of

infusate, % (n)

Isotonic

solution

95�9 (1430)

Smoking status, % (n) Hypertonic

solution

4�1 (61)

Current smoker 12�8 (189)

Former smoker 22�0 (324) Needle

gauge, % (n)

Never smoked 65�2 (962) 14G 0�3 (4)

16G 6�8 (102)

Location (ward) in

hospital, % (n)

18G 40�5 (607)

Internal Medicine,

Geriatrics and

Cardiology

34�2 (513) 20G 45�6 (683)

General Surgery 13�9 (208) 22G 2�4 (36)

Orthopaedics 26�1 (391) 24G 4�3 (65)

Obstetrics &

Gynaecology

17�2 (258)

Other Surgical

Units

8�5 (127) IV catheter

material, % (n)

Teflon 6�3 (94)

Type of patient

admission, % (n)

Polyurethane 90�5 (1356)

Planned 33�1 (496) Polyvinyl 0�7 (10)

Urgent 66�9 (1002) Polyethylene 2�5 (38)

Personnel inserting

catheter, % (n)

Type of cannula

dressing, % (n)

Nurse 94�7 (1419) Gauze and tape 43�1 (646)

Physician 5�3 (79) Transparent

polyurethane

patch

56�9 (852)

Those receiving

antibiotic

therapy, % (n)

48�7 (730)

© 2014 John Wiley & Sons Ltd 5

JAN: ORIGINAL RESEARCH Predictors of infusion-related phlebitis

the timing; if a cannula was still in situ at this point, which

was the case in 5�3% of the sample, local health profession-

als were requested to have it removed immediately.

Overall, 231 patients (15�4%) were recorded as having an

episode of phlebitis, of which 94�4% were grade 1 phlebitis;

the most serious degree observed was grade 3. When univari-

ate analysis was conducted, the frequency of phlebitis sub-

stantially and significantly differed according to several

variables: age, smoking status, inpatient hospital ward, type

of admission, whether on antibiotic therapy or not, catheter

time in situ, catheter anatomical site, needle gauge and cath-

eter material (Table 2). However, at multivariate (multilev-

el) analysis, only catheter insertion site, length of time in situ

Table 2 Incidence of phlebitis (%) and selected variables

(n = 231 in the overall sample of 1498 patients).

Variables

Phlebitis,

% Variables

Phlebitis,

%

In overall sample 15�4 Catheter timein situ, hours≤24 2�6

Age category

in years

* 25–48 4�9

18–39 11�3 49–72 6�040–59 14�5 73–96 3�860–100 18�2 97–120 13�0

Gender Catheter

insertion site

*

Females 15�8 Dorsum of

the hand

19�0

Males 14�9 Antecubital

fossa

12�9

Forearm 15�5Body mass index

<20 16�7 Type of infusate

20–24�99 15�9 Isotonic

solution

15�2

25–29�99 14�1 Hypertonic

solution

19�7

≥30 17�6Needle gaugeused

*

Cigarette smoking

status

* 14G/16G 14�0

Current 9�5 18G 11�7Former 24�1 20G 16�2Never smoked 12�9 22G/24G 33�7

Admission hospitallocation (ward)

* Cathetermaterial

*

Internal Medicine,

Geriatrics and

Cardiology

24�6 Other

materials

26�1

General Surgery 16�3 Polyurethane 14�3Orthopaedics 5�6Obstetrics &

Gynaecology

13�9 Type of cannula

dressingOther Surgical

Units

10�2 Gauze and tape 14�0

Transparent

polyurethanepatch

16�5

Type of

admission

*

Planned 11�0Urgent 17�6

Personnel inserting

catheterNurse 15�6Physician 11�4

Whether on

antibiotic therapy

*

No 11�7Yes 19�5

*P < 0�05 – Chi-squared test.

Table 3 Results of the multivariate analysis predicting phlebitis

during follow-up period (mixed model with two cluster levels: hos-

pital and identification).

Variables OR (95% CI) P*

Catheter insertion site

Dorsum of the hand (Ref.) 1 –

Antecubital fossa 0�66 (0�46–0�95) 0�026Forearm veins 0�52 (0�32–0�84) 0�008

Catheter time in situ,

additional 24 hours

1�05 (1�03–1�07) <0�001

In-patient location (ward)

Internal Medicine,

Geriatrics, Cardiology (Ref.)

1 –

General Surgery 0�61 (0�38–0�98) 0�041Orthopaedics 0�53 (0�29–0�95) 0�034Obstetrics & Gynaecology 1�50 (0�80–2�78) 0�2Other Surgical Units 0�66 (0�35–1�25) 0�2

Age, 1 additional year 1�00 (0�99–1�01) 0�5Male vs. female gender 0�91 (0�66–1�24) 0�5Body Mass Index,

1 additional (kg/m2)

1�01 (0�98–1�04) 0�7

Cigarette smoking status

Current (Ref.) 1 –

Former 1�35 (0�98–1�85) 0�064Never smoked 0�88 (0�51–1�52) 0�7

Urgent vs. planned admission 1�21 (0�85–1�72) 0�3Physician vs. nurse

inserting catheter

0�73 (0�35–1�52) 0�4

Whether on antibiotic

therapy (yes vs. no)

1�09 (0�81–1�47) 0�6

Type of infusion:

Hypertonic vs. isotonic solution

1�16 (0�63–2�13) 0�6

Needle gauge, 1 level greater 1�11 (0�93–1�33) 0�2Catheter material:

Polyurethane vs. others

0�67 (0�42–1�07) 0�095

Type of cannula dressing:

Transparent polyurethane

patch vs. gauze and tape

0�97 (0�66–1�44) 0�9

6 © 2014 John Wiley & Sons Ltd

G. Cicolini et al.

and inpatient hospital ward were found to be significantly

associated with incidence of phlebitis (Table 3). In

particular, when compared to patients with a catheter placed

in the dorsum of the hand, those with the catheter located in

the antecubital fossa or forearm were less likely to have an

episode of phlebitis of any grade recorded (Adjusted ORs

0�66 and 0�52, respectively; both P < 0�05). Also, the proba-bility of having phlebitis was lower for patients admitted to

general surgery and orthopaedics wards compared with

patients admitted to the medical (internal medicine) units

(ORs 0�61 and 0�51, respectively; both P < 0�05).Finally, the odds of having a recorded phlebitis episode

increased on average by 5% with every 24 hour increment

of catheter time in situ. All time-points were associated with

an independent and significant increase in the likelihood of

having phlebitis when compared with the first (0–24 hours –

Table 3). Notably, the frequency of phlebitis was highest for

those having the same catheter for 96–120 hours (Figure 1).

The association between phlebitis and catheter anatomi-

cal site was further evaluated (Table 4; Figures 1 and 2) to

assess whether differences by site were more apparent at

any time-point and varied (crossed) by time; also whether

episodes of phlebitis of higher severity (grade 2 or 3) were

more frequent in any of the sites (Table 4 only). In fact, the

incidence of phlebitis was found to be consistently higher at

any time-point among patients with catheters placed on the

dorsum of the hand; the incidence of medium grade and

severe phlebitis was too low to enable any meaningful com-

parison by anatomical site.

Discussion

The use of the upper extremities for peripheral catheter inser-

tion is recommended by current guidelines, yet with no speci-

fication of preferred anatomical site (O’Grady et al. 2011).

3.2

7.4 7.0

4.1

20.3

2.0

4.7 5.2

9.3

2.83.9

6.1

3.5

12.4

3.9

0.0

3.0

6.0

9.0

12.0

15.0

18.0

21.0

24.0

0-24 h 25-48 h 49-72 h 73-96 h 97-120 h

Back of the hand

Cubital fossa

Forearm

Figure 1 Incidence of phlebitis by cathe-

ter insertion site at each time-point.

Table 4 Incidence of phlebitis at different time-points; overall and by catheter insertion site.

0–24 hours 25–48 hours 49–72 hours 73–96 hours 97–120 hours Follow-up

% (*) n % (*) n % (*) n % (*) n % (*) n % (*) n

Catheter site

Dorsum of the hand 3�2 (0�0) 342 7�4 (0�4) 282 7�0 (0�0) 199 4�1 (0�0) 123 20�3 (0�0) 69 19�0 (0�3) 342

Antecubital fossa 2�0 (0�2) 510 4�7 (0�2) 449 5�2 (0�0) 324 3�9 (0�0) 203 9�3 (0�0) 108 12�9 (0�4) 510

Forearm 2�8 (0�5) 646 3�9 (0�5) 562 6�1 (0�4) 445 3�5 (0�3) 314 12�4 (1�1) 178 15�5 (1�7) 646

All sites 2�6 (0�3) 1498 4�9 (0�4) 1293 6�0 (0�2) 968 3�8 (0�2) 640 13�0 (0�6) 355 15�4 (0�9) 1498

*Percentage of patients with grade 2 or 3 phlebitis. None of the patients were found to have grade 4 or 5 phlebitis during follow-up.

0·00

0 25 50

site = Back of the handsite = Forearm

site = Cubital fossa

STaTaTM

75Hours of catheter permanence

100 125

0·25

0·50

0·75

1·00

Figure 2 Kaplan–Meier estimates of time to develop phlebitis by

catheter insertion site.

© 2014 John Wiley & Sons Ltd 7

JAN: ORIGINAL RESEARCH Predictors of infusion-related phlebitis

Therefore, the choice of anatomical site for catheterization is

determined by the operator on the basis of personal expertise,

habit, preferences and other factors such as the presence of

good quality veins. Several authors have suggested that the

anatomical site of insertion is a potential predictor of phlebi-

tis rate, yet no univocal consensus exists on anatomical site

selection (Maki & Ringer 1991, Bregenzer et al. 1998, White

2001, Cornely et al. 2002, Karadeniz et al. 2003, Kagel &

Rayan 2004, Webster et al. 2007, Uslusoy & Mete 2008,

Cicolini et al. 2009, Lee et al. 2009, Forni et al. 2010, Rego

Furtado 2011, Mestre Roca et al. 2012). To our knowledge,

this is the first multi-centre study specifically tailored to eval-

uate the relationship between site of insertion of PVCs and

risk of catheter-related phlebitis. Our results demonstrate

that, using the antecubital fossa and forearm veins, the risk

of phlebitis decreases by about 40%. This finding could be

expected, given that larger vessels may be less likely to

become inflamed than those with a narrow lumen and con-

firms previously published results that suggested that the ant-

ecubital fossa is a favoured site to minimize the risk of

phlebitis (Cornely et al. 2002). Furthermore, these results are

in accordance with others studies that showed an increased

risk of phlebitis when using veins in the (dorsum of the) hand

(Maki & Ringer 1991, Curran et al. 2000, Kagel & Rayan

2004, Lee et al. 2009, Forni et al. 2010). In an early study,

an increase in phlebitis incidence was observed when using

the hand and wrist as opposed to the forearm (RR: 0�71 and

0�60, respectively) (Maki & Ringer 1991). Another group

demonstrated a 1�5 risk ratio when the hand was used instead

of the forearms or lower extremities (Lee et al. 2009).

Nonetheless, the results of our study are in contrast to

those of other authors who reported higher phlebitis rates

for catheters inserted in the antecubital fossa (Uslusoy &

Mete 2008, Forni et al. 2010), or in the forearm (Bregenzer

et al. 1998, Karadeniz et al. 2003, Singh et al. 2008). A pos-

sible explanation given for the higher rate of phlebitis in the

antecubital fossa was excessive movement at the elbow joint

causing vessel wall traumas (Maki & Ringer 1991, Bregenzer

et al. 1998, White 2001, Cornely et al. 2002, Karadeniz

et al. 2003, Kagel & Rayan 2004, Webster et al. 2007, Uslu-

soy & Mete 2008, Cicolini et al. 2009, Lee et al. 2009,

Forni et al. 2010, Rego Furtado 2011, Mestre Roca et al.

2012). Furthermore, differences in study design (i.e. RCT vs.

observational), clinical setting (i.e. strictly general surgical or

also orthopaedic department) and sample size and character-

istics may help explain contrasting results across studies.

Another important consideration is the variation in classi-

fication and definition of different anatomical sites used.

For example, Bregenzer et al. (Uslusoy & Mete 2008) clas-

sified anatomical sites as ‘hand, forearm and other (includes

upper arm, leg)’; in contrast Karadeniz et al. (Bregenzer

et al. 1998) and Uslusoy and Mete (Karadeniz et al. 2003)

used ‘back of hand, wrist, antecubital area and forearm’.

Forni et al. (Uslusoy & Mete 2008) used ‘arm, antecubital

area, forearm, wrist, hand and leg’; Webster et al. (Forni

et al. 2010) used ‘hand, forearm, combination of sites and

others’; Lee et al. (Webster et al. 2007) used ‘hand, lower

extremity and forearm’ and finally Cicolini et al. (Lee et al.

2009) used the classification ‘cubital fossa, forearm and

hand’. This aspect should be taken into account when com-

paring results; it will certainly be advisable to use standard-

ized classification and terminology.

Clinical location

In addition to the particular anatomical site where can-

nulas are inserted, another factor proposed as being an inde-

pendent predictor of phlebitis incidence is the inpatient loca-

tion or hospital ward (Cicolini et al. 2009). We found a

reduced likelihood of developing phlebitis secondary to

peripheral intravenous cannulation for surgery or in the

orthopaedic wards when compared with the internal medi-

cine, geriatric and cardiology wards or services. Other stud-

ies found an increased risk of developing phlebitis associated

with medical (as opposed to surgical) wards (Lee et al.

2009, Mestre Roca et al. 2012). This has been attributed to

several factors such as the presence of underlying medical

conditions (e.g. cancer, immunodeficiency) or patient-spe-

cific risk factors related to hypercoagulability, including

inherited thrombophilias, which appear to increase the risk

of peripheral vein phlebitis. Other possible factors may be

operator-related, including the level of clinical staff experi-

ence (Maki & Ringer 1991, Mermel et al. 2001). One study

suggested that insertion in the emergency room, where

establishing venous access quickly is often necessary, could

be a risk factor for phlebitis (Tomford et al. 1984, Maki &

Ringer 1991, Soifer et al. 1998). In our study, we did not

consider the emergency room as a location.

Time in situ

Several research groups have concluded that a prolonged in

situ time for peripheral venous catheters is an independent

predictor of phlebitis incidence (Maki & Ringer 1991, Bre-

genzer et al. 1998, Lai et al. 1998). However, other studies

have challenged the need for routine re-sites, suggesting that

dwell times of intravenous cannulae may be safely extended

if clinically indicated (Tager et al. 1983, Maki & Ringer

1991, Bregenzer et al. 1998, Lai et al. 1998). Our results

support the recommendation of current CDC guidelines to

8 © 2014 John Wiley & Sons Ltd

G. Cicolini et al.

replace catheters in adults, at the latest, within 96 hours of

insertion (Bregenzer et al. 1998, Homer & Holmes 1998,

White 2001, Cornely et al. 2002, Webster et al. 2007,

2008, Van Donk et al. 2009, Hasselberg et al. 2010,

Rickard et al. 2010, 2012); most hospitals follow this rec-

ommendation.

It is important to note that many patients (13%) kept the

same catheter in situ beyond the maximum recommended

limit of 96 hours despite our pressure on local personnel

towards removal. Our researchers were more vigorous

towards removal when time exceeded 120 hours and after

phlebitis was observed. It appears that there is a need for

additional education of both healthcare personnel and

patients to raise awareness of the risks. On the other hand, a

strict epidemiological view may not be most apt at deciding

the cut-off; patient’s desires and comfort should also be con-

sidered, seeking a balance between danger and tolerability.

Our findings steer us towards recommending the use,

whenever possible, of antecubital fossa veins for peripheral

intravenous cannulation; future research efforts could seek

to confirm the reduced risk of phlebitis. Notably, the rate

of phlebitis (15�4%) was consistent with previous studies

that showed a range of phlebitis rate between 2�3% and

35% (O’Grady et al. 2011).

Limitations

In this study, as in most previous ones, the diagnosis of

phlebitis was made according to clinical signs at catheter

site, thus subject to operator-dependent variation (and

consequent variable investigators’ diagnosis). It is to note,

however, that blinding study subjects and operators to cath-

eter insertion site is practically very difficult and objective

criteria were strictly followed for phlebitis recognition and

classification.

Conclusions

Millions of catheter-related phlebitis worldwide may be

prevented by placing peripheral intravenous catheters into

the antecubital fossa or forearm veins rather than in the

dorsum of the hand and removing the cannulae within

96 hours. Educational campaigns on such issues are there-

fore urgently needed for healthcare professionals. A stan-

dardized classification of anatomical sites is warranted.

Acknowledgements

The authors acknowledge the collaboration of the following

nurses and nursing students, whose support was essential

for data collection: Dayana Di Vincenzo, Antonio Monteo-

dorisio, Andrea Cataldo, Valeria Rita Viva, Dunia Benl-

loch.

Funding

This research received no specific grant from any funding

agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

No conflict of interest has been declared by the author(s).

Authors contributions

All authors participated in all phases of the study (design,

data collection and interpretation of the results). LM and

MEF made the statistical analysis; VS, DC, ADB and LC

contributed to data collection; GC, GEE and LM wrote

the manuscript. All authors had full access to data and

are responsible for the integrity and the accuracy of the

data.

The authors have confirmed that all authors meet the IC-

MJE criteria for authorship credit (www.icmje.org/ethi-

cal_1author.html), as follows:

� substantial contributions to conception and design of,

or acquisition of data or analysis and interpretation of

data,

� drafting the article or revising it critically for impor-

tant intellectual content, and

� final approval of the version to be published.

References

Abbas S.Z., de Vries T.K., Shaw S. & Abbas S.Q. (2007) Use and

complications of peripheral vascular catheters: a prospective

study. British Journal of Nursing 16(11), 648, 650, 652.

Bregenzer T., Conen D., Sakmann P. & Widmer A.F. (1998) Is

routine replacement of peripheral intravenous catheters

necessary? Archives of Internal Medicine 158(2), 151–156.

Cicolini G., Bonghi A.P., Di Labio L. & Di Mascio R. (2009)

Position of peripheral venous cannulae and the incidence of

thrombophlebitis: an observational study. Journal of Advanced

Nursing 65(6), 1268–1273.

Cornely O.A., Bethe U., Pauls R. & Waldschmidt D. (2002)

Peripheral Teflon catheters: factors determining incidence of

phlebitis and duration of cannulation. Infection Control and

Hospital Epidemiology 23(5), 249–253.

Curran E.T., Coia J.E., Gilmour H., McNamee S. & Hood J.

(2000) Multi-centre research surveillance project to reduce

infections/phlebitis associated with peripheral vascular catheters.

Journal of Hospital Infection 46(3), 194–202.

© 2014 John Wiley & Sons Ltd 9

JAN: ORIGINAL RESEARCH Predictors of infusion-related phlebitis

Department of Health (2011) Peripheral Intravenous Cannula Care

Bundle. System, U. N. H., London.

Forni C., Loro L., Tremosini M., Trofa C., D’Alessandro F.,

Sabbatini T., Kapron M., Genco R., Schiavone M., Borri C.,

Bombino C., Notarnicola T., Amodeo A., Boschi R., Capezzali

D., Mosci D. & Mini S. (2010) Cohort study of peripheral

catheter related complications and identification of predictive

factors in a population of orthopedic patients. Assistenza

Infermieristica e Ricerca 29(4), 166–173.

Gallant P. & Schultz A.A. (2006) Evaluation of a visual infusion

phlebitis scale for determining appropriate discontinuation of

peripheral intravenous catheters. Journal of Infusion Nursing 29

(6), 338–345.

Hasselberg D., Ivarsson B., Andersson R. & Tingstedt B. (2010)

The handling of peripheral venous catheters–from non-

compliance to evidence-based needs. Journal of Clinical Nursing

19(23–24), 3358–3363.

Homer L.D. & Holmes K.R. (1998) Risks associated with 72- and

96-hour peripheral intravenous catheter dwell times. Journal of

Intravenous Nursing 21(5), 301–305.

Hosmer D. & Lemeshow S. (eds) (1999) Applied Survival Analysis.

John Wiley & Sons, New York.

Idvall E. & Gunningberg L. (2006) Evidence for elective

replacement of peripheral intravenous catheter to prevent

thrombophlebitis: a systematic review. Journal of Advanced

Nursing 55(6), 715–722.

Infusion Nurses Society (2006) Infusion nursing standards of

practice. Journal of Infusion Nursing 29(Suppl 1), S1–S92.

Kagel E.M. & Rayan G.M. (2004) Intravenous catheter

complications in the hand and forearm. Journal of Trauma 56

(1), 123–127.

Karadag A. & Gorgulu S. (2000) Effect of two different short

peripheral catheter materials on phlebitis development. Journal

of Intravenous Nursing 23(3), 158–166.

Karadeniz G., Kutlu N., Tatlisumak E. & Ozbakkaloglu B. (2003)

Nurses’ knowledge regarding patients with intravenous catheters

and phlebitis interventions. Journal of Vascular Nursing 21(2),

44–47; quiz 48–49.

Lai L.P., Lin J.L., Chen T.F., Ko W.C. & Lien W.P. (1998)

Clinical, electrophysiological characteristics and radiofrequency

catheter ablation of atrial tachycardia near the apex of Koch’s

triangle. Pacing and Clinical Electrophysiology 21(2), 367–374.

Lee W.L., Chen H.L., Tsai T.Y., Lai I.C., Chang W.C., Huang

C.H. & Fang C.T. (2009) Risk factors for peripheral intravenous

catheter infection in hospitalized patients: a prospective study of

3165 patients. American Journal of Infection Control 37(8),

683–686.

Lundgren A., Jorfeldt L. & Ek A.C. (1993) The care and handling

of peripheral intravenous cannulae on 60 surgery and internal

medicine patients: an observation study. Journal of Advanced

Nursing 18(6), 963–971.

Maki D.G. & Martin W.T. (1975) Nationwide epidemic of

septicemia caused by contaminated infusion products. IV.

Growth of microbial pathogens in fluids for intravenous

infusions. Journal of Infectious Diseases 131(3), 267–272.

Maki D.G. & Ringer M. (1991) Risk factors for infusion-related

phlebitis with small peripheral venous catheters. A randomized

controlled trial. Annals of Internal Medicine 114(10), 845–854.

Maki D.G., Kluger D.M. & Crnich C.J. (2006) The risk of

bloodstream infection in adults with different intravascular

devices: a systematic review of 200 published prospective studies.

Mayo Clinic Proceedings 81(9), 1159–1171.

McCallum L. & Higgins D. (2012) Care of peripheral venous

cannula sites. Nursing Times 108(34–35), 12, 14–15.

Melly M.A., Meng H.C. & Schaffner W. (1975) Microbiol growth

in lipid emulsions used in parenteral nutrition. Archives of

Surgery 110(12), 1479–1481.

Mermel L.A., Farr B.M., Sherertz R.J., Raad I.I., O’Grady N.,

Harris J.S. & Craven D.E. (2001) Guidelines for the management

of intravascular catheter-related infections. Infection Control and

Hospital Epidemiology 22(4), 222–242.

Mestre Roca G., Berbel Bertolo C., Tortajada Lopez P., Gallemi

Samaranch G., Aguilar Ramirez M.C., Cayla Buqueras J.,

Rodriguez-Bano J. & Martinez J.A. (2012) Assessing the

influence of risk factors on rates and dynamics of peripheral vein

phlebitis: an observational cohort study. Medicina Cl�ınica 139

(5), 185–191.

O’Grady N.P., Alexander M., Burns L.A., Dellinger E.P., Garland

J., Heard S.O., Lipsett P.A., Masur H., Mermel L.A., Pearson

M.L., Raad I.I., Randolph A.G., Rupp M.E. & Saint S. (2011)

Summary of recommendations: guidelines for the Prevention of

Intravascular Catheter-related Infections. Clinical Infectious

Diseases 52(9), 1087–1099.

Parruti G., Tontodonati M., Rebuzzi C., Polilli E., Sozio F.,

Consorte A., Agostinone A., Di Masi F., Congedo G., D’Antonio

D., Granchelli C., D’Amario C., Carunchio C., Pippa L.,

Manzoli L. & Volpi A. (2010) Predictors of pain intensity and

persistence in a prospective Italian cohort of patients with herpes

zoster: relevance of smoking, trauma and antiviral therapy. BMC

Medicine 8, 58.

Rabe-Hesketh S. & Skrondal A. (ed.) (2008) Multilevel and

Longitudinal Modelling using Stata. Stata Press, College Station,

TX.

Rego Furtado L.C. (2011) Incidence and predisposing factors of

phlebitis in a surgery department. British Journal of Nursing 20

(14), S16–S18, S20, S22 passim.

Reilly J., Stewart S., Allardice G., Noone A., Robertson C., Walker

A. & Coubrough S. (2007) NHS Scotland national HAI

prevalence survey. Final Report 2007, Health Protection

Scotland.

Rickard C.M., McCann D., Munnings J. & McGrail M.R. (2010)

Routine resite of peripheral intravenous devices every 3 days did

not reduce complications compared with clinically indicated

resite: a randomised controlled trial. BMC Medicine 8, 53.

Rickard C.M., Webster J., Wallis M.C., Marsh N., McGrail M.R.,

French V., Foster L., Gallagher P., Gowardman J.R., Zhang L.,

McClymont A. & Whitby M. (2012) Routine versus clinically

indicated replacement of peripheral intravenous catheters: a

randomised controlled equivalence trial. Lancet 380(9847),

1066–1074.

Sheth N.K., Franson T.R., Rose H.D., Buckmire F.L., Cooper J.A.

& Sohnle P.G. (1983) Colonization of bacteria on polyvinyl

chloride and Teflon intravascular catheters in hospitalized

patients. Journal of Clinical Microbiology 18(5), 1061–1063.

Singh R., Bhandary S. & Pun K.D. (2008) Peripheral intravenous

catheter related phlebitis and its contributing factors among

10 © 2014 John Wiley & Sons Ltd

G. Cicolini et al.

adult population at KU Teaching Hospital. Kathmandu

University Medical Journal (KUMJ) 6(24), 443–447.

Soifer N.E., Borzak S., Edlin B.R. & Weinstein R.A. (1998)

Prevention of peripheral venous catheter complications with an

intravenous therapy team: a randomized controlled trial.

Archives of Internal Medicine 158(5), 473–477.

Tagalakis V., Kahn S.R., Libman M. & Blostein M. (2002) The

epidemiology of peripheral vein infusion thrombophlebitis: a

critical review. American Journal of Medicine 113(2), 146–151.

Tager I.B., Ginsberg M.B., Ellis S.E., Walsh N.E., Dupont I.,

Simchen E. & Faich G.A. (1983) An epidemiologic study of the

risks associated with peripheral intravenous catheters. American

Journal of Epidemiology 118(6), 839–851.

Tomford J.W., Hershey C.O., McLaren C.E., Porter D.K. & Cohen

D.I. (1984) Intravenous therapy team and peripheral venous

catheter-associated complications. A prospective controlled study.

Archives of Internal Medicine 144(6), 1191–1194.

Uslusoy E. & Mete S. (2008) Predisposing factors to phlebitis in

patients with peripheral intravenous catheters: a descriptive

study. Journal of the American Academy of Nurse Practitioners

20(4), 172–180.

Van Donk P., Rickard C.M., McGrail M.R. & Doolan G. (2009)

Routine replacement versus clinical monitoring of peripheral

intravenous catheters in a regional hospital in the home

program: a randomized controlled trial. Infection Control and

Hospital Epidemiology 30(9), 915–917.

Webster J., Lloyd S., Hopkins T., Osborne S. & Yaxley M. (2007)

Developing a Research base for Intravenous Peripheral cannula

re-sites (DRIP trial). A randomised controlled trial of hospital in-

patients. International Journal of Nursing Studies 44(5), 664–

671.

Webster J., Clarke S., Paterson D., Hutton A., van Dyk S., Gale C.

& Hopkins T. (2008) Routine care of peripheral intravenous

catheters versus clinically indicated replacement: randomised

controlled trial. British Medical Journal 337, a339.

Webster J., Osborne S., Rickard C. & Hall J. (2010) Clinically-

indicated replacement versus routine replacement of peripheral

venous catheters. Cochrane Database Systematic Review (3),

CD007798.

White S.A. (2001) Peripheral intravenous therapy-related phlebitis

rates in an adult population. Journal of Intravenous Nursing 24

(1), 19–24.

The Journal of Advanced Nursing (JAN) is an international, peer-reviewed, scientific journal. JAN contributes to the advancement of

evidence-based nursing, midwifery and health care by disseminating high quality research and scholarship of contemporary relevance

and with potential to advance knowledge for practice, education, management or policy. JAN publishes research reviews, original

research reports and methodological and theoretical papers.

For further information, please visit JAN on the Wiley Online Library website: www.wileyonlinelibrary.com/journal/jan

Reasons to publish your work in JAN:

• High-impact forum: the world’s most cited nursing journal, with an Impact Factor of 1·527 – ranked 14/101 in the 2012 ISI Jour-

nal Citation Reports © (Nursing (Social Science)).

• Most read nursing journal in the world: over 3 million articles downloaded online per year and accessible in over 10,000 libraries

worldwide (including over 3,500 in developing countries with free or low cost access).

• Fast and easy online submission: online submission at http://mc.manuscriptcentral.com/jan.

• Positive publishing experience: rapid double-blind peer review with constructive feedback.

• Rapid online publication in five weeks: average time from final manuscript arriving in production to online publication.

• Online Open: the option to pay to make your article freely and openly accessible to non-subscribers upon publication on Wiley

Online Library, as well as the option to deposit the article in your own or your funding agency’s preferred archive (e.g. PubMed).

© 2014 John Wiley & Sons Ltd 11

JAN: ORIGINAL RESEARCH Predictors of infusion-related phlebitis