1

Oral gentamicin gut decontamination for prevention of KPC-producing Klebsiella 1 pneumoniae infections: the relevance of concomitant systemic antibiotic therapy 2 3 Running title: Gut decontamination from KPC-Kp. 4 5 Carlo Tascini 1, Francesco Sbrana 2, Sarah Flammini 1, Enrico Tagliaferri 1, Fabio Arena 3, 6 Alessandro Leonildi 1, Ilaria Ciullo 1, Francesco Amadori 1, Antonello Di Paolo 4, Andrea Ripoli 7 2, Russell Lewis 5, Gian Maria Rossolini 3, 6, 7, Francesco Menichetti 1#, and the GENGUT 8 study group 9 10 (1) Infectious Diseases Unit, Cisanello Hospital, Pisa, Italy 11 (2) Fondazione Toscana Gabriele Monasterio, Pisa, Italy 12 (3) Dipartimento di Biotecnologie Mediche, Università di Siena, Siena, Italy 13 (4) Division of Pharmacology, Dep. of Clinical and Experimental Medicine, Un. of Pisa, Italy 14 (5) Infectious Diseases Unit, S’Orsola-Malpighi Hospital. Department of Medical Sciences 15 and Surgery, Un. of Bologna, Bologna, Italy 16 (6) Dipartimento di Medicina Sperimentale e Clinica, Sezione di Medicina Critica e Medicine 17 Specialistiche, Università di Firenze, Florence, Italy 18 (7) S.O.D. Microbiologia e Virologia, Azienda Ospedaliera Universitaria Careggi, Florence, 19 Italy 20 21 22 Corresponding author: Francesco Menichetti, MD 23

Infectious Diseases Unit - Cisanello Hospital 24 Via Paradisa 2 25

56100 Pisa, Italy 26 Phone: +39 050 996739 27 Fax: +39 050 571021 28 E-mail: menichettifrancesco@gmail.com 29

30

AAC Accepts, published online ahead of print on 13 January 2014Antimicrob. Agents Chemother. doi:10.1128/AAC.02283-13Copyright © 2014, American Society for Microbiology. All Rights Reserved.

2

Abstract 31

Gut colonization represents the main source for KPC-producing Klebsiella pneumoniae 32

(KPC-Kp) epidemic dissemination. Oral gentamicin, 80 mg four-times daily, was administered 33

to 50 consecutive patients with gut colonization by gentamicin susceptible KPC-Kp in case of 34

planned surgery, major medical intervention, or need for patient transferral. The overall 35

decontamination rate was 68% (34/50). The median duration of gentamicin treatment was 9 36

days (interquartile range 7-15) in decontaminated patients as compared to 24 days 37

(interquartile range 20-30) in those with persistent colonization (p<0.001). In the six-months 38

period of follow-up, KPC-Kp infections were documented in 5/34 (15%) successfully 39

decontaminated patients as compared to 12/16 (73%) persistent carriers (p<0.001). The 40

decontamination rate was 96% (22/23) in patients receiving oral gentamicin only as 41

compared to 44% (12/27) of those treated with oral gentamicin and concomitant systemic 42

antibiotic therapy (CSAT) (p<0.001). The multivariate analysis confirmed CSAT and a KPC-43

Kp infection as the variables associated with gut decontamination. In the follow-up period, 44

KPC-Kp infections were documented in 2/23 (9%) of patients treated with oral gentamicin 45

only and in 15/27 (56%) of those receiving also CSAT (p=0.003). No difference in overall 46

death rate was documented between different groups. Gentamicin-resistant KPC-Kp strains 47

were isolated from stools of 4/16 persistent carriers. Peak gentamicin blood levels were 48

below 1 mg/L in 12/14 tested patients. Oral gentamicin showed to be potentially useful for gut 49

decontamination and prevention of infection due to KPC-Kp, especially in patients not 50

receiving CSAT. The risk of emergence of gentamicin-resistant KPC-Kp should be 51

considered. 52

53

Keywords: KPC-producing K. pneumoniae; Infection prevention; Oral gentamicin gut 54 decontamination. 55 56

3

Introduction 57

Carbapenem-resistant Klebsiella pneumoniae producing KPC-type beta-lactamase (KPC-Kp) 58

are multi-drug resistant Gram-negative bacilli with limited therapeutic options and high 59

associated mortality rates (> 40%) [1-4]. Gut colonization represents the main human 60

reservoir for epidemic dissemination in hospitals and requires active surveillance for prompt 61

carrier identification and infection control measures such as isolation or cohorting with 62

dedicated staff [5, 6]. Furthermore, gut colonization seems to be associated with a 63

substantial risk (around 10%) of developing subsequent KPC-Kp infection [7, 8] and may be 64

a contraindication for some surgical procedures, organ transplantation and other major 65

medical interventions. Gut colonization by KPC-Kp may last for months, even in the absence 66

of hospital readmission [9]. Therefore, gut decontamination for KPC-Kp colonization could be 67

of interest as a complementary approach for removing patient from isolation, reducing 68

transmission, and preventing subsequent infectious episodes in already colonized patients. 69

In fact, various regimens for KPC-Kp gut decontamination have been investigated. Using an 70

oral gentamicin regimen of 80 mg qid, Zuckerman et al. reported decontamination of KPC-Kp 71

from 10/15 (66%) haematological patients without selecting gentamicin-resistant strains [10]. 72

Oren et al., using oral gentamicin 80 mg qid, reported decontamination of carbapenem-73

resistant K. pneumoniae from 11/26 (42%) patients [11]. In a randomized double-blind 74

placebo-controlled trial in 40 patients with KPC-Kp positive rectal swab, Saidel-Odes et al., 75

reported that administration of an oral gentamicin/polymyxin E combination cleared KPC-Kp 76

colonization in 12/20 patients (60%) without selecting gentamicin or colistin-resistant KPC-Kp 77

strains [12]. However, Brink et al. found that use of colistin for selective digestive tract 78

decontamination was associated with the emergence of a colistin-resistant OXA -181-79

producing K. pneumoniae strain [13]. 80

Oral gentamicin may be an ideal agent for gut decontamination from KPC-Kp for several 81

reasons, including: i) gentamicin is rapidly bactericidal "in vitro" against gentamicin-82

susceptible KPC-Kp strains; ii) the oral formulation is virtually non-absorbable, without 83

systemic activity and toxicity; iii) the drug has a limited spectrum and no activity against 84

4

anaerobes, being less disruptive to the gastrointestinal flora; and iv) gentamicin is not 85

typically administered as part of combination regimens for serious KPC-Kp infections [1]. 86

The aim of our pilot study was to describe the microbiological and clinical response of gut 87

decontamination with oral gentamicin in 50 consecutive patients colonized by KPC-Kp, 88

receiving or not concomitant systemic antibiotic therapy (CSAT). 89

90

Materials and Methods 91

We performed a pilot, non-blinded prospective study in three Italian hospitals to assess the 92

feasibility of administering oral gentamicin for KPC-Kp gut decontamination. Because no 93

published data were available regarding the possible treatment efficacy of oral 94

decontamination therapy for KPC-Kp, we did not attempt to estimate a priori the required 95

sample size for this pilot study. In a eight-months period consecutive adult patients with a 96

rectal swab culture positive for KPC-Kp within the previous 7 days were considered eligible 97

for the study if they had: i) planned surgery; ii) need for major medical intervention that could 98

predispose to serious infection (e. g., cancer chemotherapy or immunosuppression); or iii) 99

need for transferral outside the hospital. All patients provided written informed consent to 100

participate in the study. Patients who met the inclusion criteria were administered oral 101

gentamicin (80 mg four times daily) for at least 8 days. The need for continued oral 102

decontamination therapy beyond 8 days was determined by the physicians caring for the 103

patient, who were not blinded to follow-up rectal swab culture results. Infection control 104

procedures for patients with KPC-Kp gut colonization were similar at all participating 105

institutions, consisting of patient isolation or cohorting, with strict contact precautions, and 106

dedicated caregivers and equipment as described in CDC guidelines [14]. 107

We used a validated direct screening method on rectal swab culture to identify patients with 108

KPC-Kp gut colonization [15]. Rectal swab cultures were performed at the time of hospital 109

admission and every 4 days thereafter. K. pneumoniae stool isolates were identified with the 110

mini-API system (bioMérieux, Mercy L’Etoile, France) or MALDI-TOF mass spectrometry 111

(Vitek MS, bioMérieux). Detection of blaKPC genes was confirmed by PCR as previously 112

5

described [16]. Antimicrobial susceptibility testing was performed using a disk diffusion 113

method according to EUCAST version 3.1 guidelines [17], and MIC of gentamicin, colistin 114

and fosfomycin were determined with E-test, according to the manufacturer’s instruction (AB 115

Biodisk, Solno, Sweden). Interpretation of susceptibility data was according to EUCAST 116

breakpoints (version 3.1) [17]. Pulsed-field gel electrophoresis (PFGE) of XbaI-digested 117

genomic DNA of KPC-Kp to compare isolates longitudinal isolates from patients who failed 118

decontamination was performed and interpreted as described previously [18]. 119

Gentamicin peak plasma concentration was determined with a fluorescent immunoassay 120

method (Abbott, Abbott Park, IL) on day three of treatment on samples obtained 2 hours after 121

drug administration. 122

Study endpoints: The primary study outcome was decontamination of KPC-Kp gut 123

colonization defined by at least two consecutive negative rectal swab cultures. Patients with 124

two or more consecutive positive rectal swabs were considered to have failed oral 125

decontamination and were considered persistent carriers. We also evaluated clinical 126

outcomes of patients administered oral gentamicin for KPC-Kp decontamination, including 127

infectious episodes caused by KPC-Kp and death for any cause during a six-months period 128

after oral gentamicin therapy, in decontaminated and persistent carrier groups. A separate 129

analysis was performed with the 23 patients receiving oral gentamicin alone and with the 27 130

patients who received CSAT. 131

Statistical analysis: Data were expressed as mean +/- standard deviation, median and 132

interquartile range or proportions, as appropriate. Comparisons were performed with 133

unpaired t-test, Mann-Whitney test or chi-square test with continuity correction, as 134

appropriate. Gut decontamination events during oral gentamicin therapy were analysed by 135

the Kaplan–Meier method, with log-rank test for curve comparison in patients treated or not 136

with CSAT. To evaluate the influence of CSAT on gut decontamination we performed a 137

univariate logistic regression analysis adjusting for several potential confounders (KPC-Kp 138

infection, ICU stay, days of gentamicin treatment, gender and age). Predictors found to be 139

statistically significant (p<0.05) at the univariate logistic regression were then included in a 140

6

multivariate logistic model. All computations were done with R statistical software (R, version 141

2.11.1, 2010; R Development Core Team 2006, R Foundation for Statistical Computing, 142

Vienna, Austria). A value of p < 0.05 was considered significant. 143

144

Results 145

Fifty consecutive patients with KPC-Kp positive rectal swab cultures were enrolled in the 146

study over 8 months. All the KPC-Kp isolates from rectal swabs showed susceptibility to 147

gentamicin (MIC50 1.5 mg/L; MIC90 2 mg/L; range 0.125 - 2 mg/L). All strains were resistant to 148

carbapenems; 27/50 (54%) were resistant to colistin, 24/50 (48%) to tigecycline and 7/50 149

(14%) to fosfomycin. 150

CSAT was administered in 27/50 of the study patients due to clinically documented 151

pneumonia (n=1) and 26 microbiologically and clinically documented infections due to: 18 152

KPC-Kp (bloodstream infection-BSI, n=14; ventilator-associated pneumonia, n= 2; peritonitis 153

n=1, urinary tract infection-UTI n=1), 2 P. aeruginosa (BSI n=1, cellulitis n=1), 3 E. coli (BSI 154

n=2, UTI n=1), 2 S. aureus (BSI n=1, cellulitis n=1), and 1 S. epidermidis BSI. The majority of 155

patients (22/27) received combinations of two or more antibiotics, whereas five patients 156

received monotherapy regimens. The prescribed antibiotics were: tigecycline (n=15 patients), 157

meropenem (n=12), colistin (n=12), gentamicin (n=8), fosfomycin (n=4), imipenem (n=4), 158

cotrimoxazole (n=3), piperacillin/tazobactam (n=3), rifampin (n=2), teicoplanin (n=1). 159

The enrolled patients were treated with oral gentamicin for a median of 16 days (interquartile 160

range 10-27), and the overall decontamination rate in the entire study population 161

administered oral gentamicin was 68% (34/50). The median duration of gentamicin therapy 162

was 9 days (interquartile range 7-15) in decontaminated patients, versus 24 days 163

(interquartile range 20-30) in those with persistent colonization (p <0.001). At 6 months 164

follow-up, KPC-Kp infections were documented in 5/34 (15%) patients successfully 165

decontaminated, compared to 12/16 (73%) of patients with persistent colonization (p <0.001). 166

7

No difference in overall mortality was observed between decontaminated and persistently 167

colonized patients. Patient characteristics and clinical outcomes in successfully 168

decontaminated patients and persistent carriers are summarized in Table 1. 169

The probability of a persistent carriage status during oral gentamicin therapy was statistically 170

lower in patients treated with gentamicin only as compared to those receiving CSAT (p 171

=0.007) (Figure 1). The decontamination rate was 96% (22/23) in patients receiving oral 172

gentamicin only versus 44% (12/27) in those also treated with CSAT (p <0.001). In the follow-173

up period, KPC-Kp infection were documented in 2/23 (9%) of patients treated with oral 174

gentamicin only and in 15/27 (56%) of those receiving also CSAT (p = 0.003). No difference 175

in overall mortality rate was documented between the two treatment groups. Patient 176

characteristics, microbiological and clinical outcomes in patients receiving gentamicin only 177

and those also treated with CSAT are summarized in Table 2. 178

The univariate analysis identified CSAT, KPC-Kp infection and ICU stay as significant 179

variables (Table 3). The multivariate analysis confirmed CSAT and KPC-Kp infection as the 180

variables associated with gut decontamination (Table 4). 181

Genotype analysis performed with PFGE on strains of 5/17 patients who failed gut 182

decontamination with gentamicin showed that KPC-Kp strains were genetically related 183

(Correlation index > 90%). 184

Gentamicin-resistant KPC-Kp strains were isolated during oral gentamicin therapy in 4/16 185

failed patients (25%). 186

Mean of gentamicin peak blood concentrations, measured in 14/16 patients, were 0.7 ± 1.1 187

mg/L (range: 0 – 3.1 mg/L). Only two patients showed a gentamicin blood level above 1 188

mg/L. No drug treatment was interrupted for any reason and no documented side effects 189

attributable to gentamicin oral therapy were observed. 190

191

Discussion 192

Gut colonization with KPC-Kp represents the main source for the epidemic dissemination of 193

carbapenem-resistant Enterobacteriaceae, and is associated with a substantial risk of 194

8

developing subsequent infection [7, 8]. Selective gut decontamination of KPC-Kp, therefore 195

could be an important complementary approach for reducing a colonized patient’s risk for 196

spreading or developing severe infection with Multi-drug resistant (MDR) Enterobacteriaceae 197

[10, 11, 12]. 198

In our population of 50 KPC-Kp colonized patients treated with oral gentamicin, we observed 199

a favourable microbiological outcome (overall gut decontamination rate: 34/50; 68%). 200

Our results are similar to those reported in a small group of haematological patients treated 201

with the same regimen (10/15; 66%) [10], and in a cohort of medical and surgical population 202

receiving oral gentamicin in combination with polymyxin E (12/20; 60%) [12]. Recently, 203

Feldman et al. reported that spontaneous resolution of carriage may occur in the majority of 204

patients with remote (> 4 months) versus recent acquisition [19]. However, Oren et al. 205

reported a spontaneous eradication of carbapenem-resistant enterobacteriaceae from stools 206

in only 7/102 patients (7%) after a median follow-up period of 140 days (range 20-737) [11]; 207

furthermore, some patients (such as those represented in our study) may benefit from more 208

rapid clearance of KPC-Kp colonization due to planned major therapeutic interventions. 209

It is noteworthy that infectious episodes due to KPC-Kp documented in the six-months follow-210

up period were significantly lower in the group of successfully decontaminated patients (5/34; 211

15%) with respect to the group of persistently colonized patients (12/16; 73%). Interestingly, 212

Zuckerman et al. reported that 5/15 patients who failed gentamicin decontamination 213

subsequently died, with three of these deaths due to KPC-Kp BSI [10]. Our experience 214

seems to support the relevant role of gut decontamination in preventing subsequent 215

infectious episodes due to KPC-Kp. 216

In our study we observed a significant difference in decontamination rate in the group of 217

patients receiving oral gentamicin only (22/23; 96%) compared to patients also receiving oral 218

gentamicin during CSAT (12/27; 44%). Selective pressure of CSAT on gut microbiota may 219

favour the persistent carriage of KPC-Kp, thereby contributing to the higher failure rates 220

observed versus monotherapy. This hypothesis is supported by the reports of Schwaber [20] 221

suggesting the role of antibiotics use as predictors for KPC-Kp acquisition and Ben-David 222

9

[21] identifying antibiotic exposure during the prior three months as one of the risk factor for 223

persistent carriage. In our experience 16/27 (59%) patients treated with CSAT received a 224

carbapenem antibiotic. The specific role of carbapenems in favouring selective pressure and 225

the acquisition of KPC-Kp has also been reported [22, 23]. However, we cannot exclude the 226

possibility that CSAT was a marker for more severely-ill, debilitated patients who are more 227

likely to have a natural history of persistent KPC-Kp colonization. 228

The genotype analysis, although performed on few strains isolated from 5/17 patients who 229

failed gut decontamination with gentamicin, show a high correlation rate and no acquisition of 230

new clones. 231

It is noteworthy that patients treated with gentamicin only suffered for less KPC-Kp infectious 232

episodes (2/23, 9%) as compared to the group receiving also CSAT (15/27, 56%). This 233

significant difference suggests that gut decontamination for asymptomatic KPC-Kp carriers 234

may be an effective strategy worthy of evaluation in prospective randomized trials, especially 235

among patients who do not require immediate additional systemic antimicrobial therapy. 236

Considering the small sample size, it was also not surprising we were unable to detect an 237

overall survival difference between successfully and non-successfully decontaminated 238

patients who did not receive CSAT (22% vs. 37%, p = 0.406 ). Detection of such a difference 239

would likely require a prospective randomized trial with several hundred patients per 240

treatment arm. 241

According to the multivariate analysis results CSAT and KPC-Kp infections are the variables 242

associated with gut decontamination; these variables might identify a sub-group of more 243

critically ill patients less responsive to gentamicin oral therapy. 244

Oral gut decontamination is associated with some risks. Gentamicin oral therapy may favour 245

the emergence of resistant KPC-Kp, especially in patients who failed to respond to gut 246

decontamination regimen. Gentamicin-resistant KPC-Kp strains were in fact isolated in our 247

study from 4/16 (25%) persistent carriers. Although previous experience with oral gentamicin 248

did not mention the emergence of resistant strains [10, 12], this potential risk should be 249

considered and gut decontamination should be interrupted as soon as patients fail. 250

10

The virtual lack of any systemic activity of oral gentamicin (only 2 patients had blood levels > 251

1 mg/L) underscored why the oral regimen was associated with a good safety profile in our 252

study, with no evidence of nephrotoxicity, ototoxicity, or drug interruption due to adverse 253

effects. 254

Finally, a limitation of this pilot study was the lack of a control group and blinding by 255

physicians caring for patients. 256

An additional limitation was the potential overestimation of gut decontamination due to the 257

lower sensitivity of the culture method as compared to PCR [15]. Low-level carriage may not 258

be detectable with culture-based methods, but two consecutive negative results are certainly 259

suggestive either of true decontamination or of significant reduction of the gut colonization 260

burden, and have been used as the basis for relaxation of strict contact precautions in 261

infection control programs [14]. 262

However, given our experience that suggests selective oral decontamination therapy of KPC-263

Kp with oral gentamicin is safe and possibly effective, we believe larger scale, prospective 264

randomized trials should be considered to evaluate the promising, complementary approach 265

for reducing risk of severe infection with these notoriously difficult to treat MDR 266

Enterobacteriaceae. 267

268

269

11

Acknowledgements: 270

Partial data related to this study have been presented in ECCMID meeting in April 2013 271

(ECCMID R2812). 272

Funding: This work was supported in part by a grant from FP7 project EvoTAR 273

(no.HEALTH-F3-2011-2011-282004) to G.M.R. 274

Disclosures: No conflict of interest for each authors. 275

In the past 2 years, C.T. has been paid for lectures on behalf of Pfizer, Novartis, Merck, and 276

Astellas. 277

278

The GENGUT study group: Paola Lambelet - Viareggio, Francesco Forfori - Pisa, Bruno 279

Viaggi - Pisa, Francesca Papini - Pisa, Lucio Urbani - Pisa, Paolo Malacarne – Pisa. 280

281

12

References 282

1) Qureshi ZA, Paterson DL, Potoski BA, Kilayko MC, Sandovsky G, Sordillo E, Polsky 283

B, Adams-Haduch JM, Doi Y. 2012. Treatment outcome of bacteremia due to KPC-284

producing Klebsiella pneumoniae: superiority of combination antimicrobial regimens. 285

Antimicrob Agents Chemother. 56:2108-13. 286

2) Tumbarello M, Viale P, Viscoli C, Trecarichi EM, Tumietto F, Marchese A, Spanu T, 287

Ambretti S, Ginocchio F, Cristini F, Losito AR, Tedeschi S, Cauda R, Bassetti M. 2012. 288

Predictors of mortality in bloodstream infections caused by KPC-producing Klebsiella 289

pneumoniae: importance of combination therapy. Clin Infect Dis. 55:943-50. 290

3) Sbrana F, Malacarne P, Viaggi B, Costanzo S, Leonetti P, Leonildi A, Casini B, Tascini 291

C, Menichetti F. 2013. Carbapenem-sparing antibiotic regimens for infections caused by 292

Klebsiella pneumoniae carbapenemase-producing K. pneumoniae in intensive care unit. Clin 293

Infect Dis. 56:697-700. 294

4) Bulik CC, Nicolau DP. 2011. Double-carbapenem therapy for carbapenemase-producing 295

Klebsiella pneumoniae. Antimicrob Agents Chemother. 55:3002-4. 296

5) Munoz-Price LS, Quinn JP. 2013. Deconstructing the infection control bundles for the 297

containment of carbapenem-resistant Enterobacteriaceae. Curr Opin Infect Dis. 26:378-87. 298

6) Buehlmann M, Bruderer T, Frei R, Widmer AF. 2011. Effectiveness of a new 299

decolonisation regimen for eradication of extended-spectrum β-lactamase-producing 300

Enterobacteriaceae. J Hosp Infect. 77:113-7. 301

7) Borer A, Saidel-Odes L, Eskira S, Nativ R, Riesenberg K, Livshiz-Riven I, Schlaeffer 302

F, Sherf M, Peled N. 2012. Risk factors for developing clinical infection with carbapenem-303

resistant Klebsiella pneumoniae in hospital patients initially only colonized with carbapenem-304

resistant K pneumoniae. Am J Infect Control. 40:421-5. 305

8) Schechner V, Kotlovsky T, Kazma M, Mishali H, Schwartz D, Navon-Venezia S, 306

Schwaber MJ, Carmeli Y. 2013. Asymptomatic rectal carriage of blaKPC producing 307

carbapenem-resistant Enterobacteriaceae: who is prone to become clinically infected? Clin 308

Microbiol Infect. 19:451-6. 309

13

9) Zimmerman FS, Assous MV, Bdolah-Abram T, Lachish T, Yinnon AM, Wiener-Well Y. 310

2013. Duration of carriage of carbapenem-resistant Enterobacteriaceae following hospital 311

discharge. Am J Infect Control. 4:190-4. 312

10) Zuckerman T, Benyamini N, Sprecher H, Fineman R, Finkelstein R, Rowe JM, Oren 313

I. 2011. SCT in patients with carbapenem resistant Klebsiella pneumoniae: a sigle center 314

experience with oral gentamicin for the eradication of carrier state. Bone Marrow Transpl. 315

46:1226-30. 316

11) Oren I, Sprecher H, Finkelstein R, Hadad S, Neuberger A, Hussein K, Raz-Pasteur 317

A, Lavi N, Saad E, Henig I, Horowitz N, Avivi I, Benyamini N, Fineman R, Ofran Y, 318

Haddad N, Rowe JM, Zuckerman T. 2013. Eradication of carbapenem-resistant 319

Enterobacteriaceae gastrointestinal colonization with nonabsorbable oral antibiotic treatment: 320

A prospective controlled trial. Am J Infect Control. 41:1167-72. 321

12) Saidel-Odes L, Polachek H, Peled N, Riesenberg K, Schlaeffer F, Trabelsi Y, Eskira 322

S, Yousef B, Smolykov R, Codish S, Borer A. 2012. A Randomized, Double-Blind, 323

Placebo-Controlled Trial of Selective Digestive Decontamination Using Oral Gentamicin and 324

Oral Polymyxin E for Eradication of Carbapenem-Resistant Klebsiella pneumoniae Carriage. 325

Infect Control Hosp Epidemiol 33:14-19. 326

13) Brink AJ, Coetzee J, Corcoran C, Clay CG, Hari-Makkan D, Jacobson RK, Richards 327

GA, Feldman C, Nutt L, van Greune J, Deetlefs JD, Swart K, Devenish L, Poirel L, 328

Nordmann P. 2013. Emergence of OXA-48 and OXA-181 carbapenemases among 329

Enterobacteriaceae in South Africa and evidence of in vivo selection of colistin resistance as 330

a consequence of selective decontamination of the gastrointestinal tract. J Clin Microbiol. 331

51:369-72. 332

14) Guidance for control of infections with carbapenem-resistant or carbapenemase-333

producing Enterobacteriaceae. Available from: http://www.cdc.gov/hai/organisms/cre/cre-334

toolkit/index.html 335

14

15) Giani T, Tascini C, Arena F, Ciullo I, Conte V, Leonildi A, Menichetti F, Rossolini 336

GM. 2012. Rapid detection of intestinal carriage of Klebsiella pneumoniae producing KPC 337

carbapenemase during an outbreak. J Hosp Infect. 81:119-22. 338

16) Giani T, D'Andrea MM, Pecile P, Borgianni L, Nicoletti P, Tonelli F, Bartoloni A, 339

Rossolini GM. 2009. Emergence in Italy of Klebsiella pneumoniae sequence type 258 340

producing KPC-3 carbapenemase. J Clin Microbiol. 47:3793-4. 341

17) European Committee on Antimicrobial Susceptibility Testing (EUCAST). Clinical 342

breakpoint. EUCAST. Available from: http://www.eucast.org/clinical_brekpoints 343

18) Tenover FC, Arbeit RD, Goering RV, Mickelsen PA, Murray BE, Persing DH, 344

Swaminathan B. 1995. Interpreting chromosomal DNA restriction patterns produced by 345

pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol. 33: 2233-346

2239. 347

19) Feldman N, Adler A, Molshatzki N, Navon-Venezia S, Khabra E, Cohen D, Carmeli Y. 348

2013. Gastrointestinal colonization by KPC-producing Klebsiella pneumoniae following 349

hospital discharge: duration of carriage and risk factors for persistent carriage. Clin Microbiol 350

Infect. 19:E190-6. 351

20) Schwaber MJ, Klarfeld-Lidji S, Navon-Venezia S, Schwartz D, Leavitt A, Carmeli Y. 352

2008. Predictors of carbapenem-resistant Klebsiella pneumoniae acquisition among 353

hospitalized adults and effect of acquisition on mortality. Antimicrob Agents Chemother. 354

52:1028-33. 355

21) Ben-David D, Masarwa S, Navon-Venezia S, Mishali H, Fridental I, Rubinovitch B, 356

Smollan G, Carmeli Y, Schwaber MJ; Israel PACF CRKP (Post-Acute-Care Facility 357

Carbapenem-Resistant Klebsiella pneumoniae) Working Group. 2011. Carbapenem-358

resistant Klebsiella pneumoniae in post-acute-care facilities in Israel. Infect Control Hosp 359

Epidemiol. 32:845-53. 360

22) Gupta N, Limbago BM, Patel JB, Kallen AJ. 2011. Carbapenem-resistant 361

Enterobacteriaceae: epidemiology and prevention. Clin Infect Dis. 53:60-7. 362

15

23) Patel G, Huprikar S, Factor SH, Jenkins SG, Calfee DP. 2008. Outcomes of 363

carbapenem-resistant Klebsiella pneumoniae infection and the impact of antimicrobial and 364

adjunctive therapies. Infect Control Hosp Epidemiol. 29:1099-106. 365

366

367

368

16

TABLE 1. Patient characteristics and clinical outcomes in decontaminated patients vs. 369

persistent carriers. 370

371

Decontaminated Patients (n= 34)

Persistent Carriers (n= 16)

p

Male 23 (68%) 12 (75%) 0.843

Mean age, years ± SD

66 ± 11 59 ± 15 0.141

ICU’s patients 2 (6%) 8 (50%) 0.001

Median duration of treatment, days (range)

9 [7 – 15] 21 [15 – 30] < 0.001

KPC-Kp infection a 5/34 (15%) 12/16 (73%) < 0.001

Overall death a 10/34 (29%) 5/16 (31%) 0.843

372 a in the 6-month follow-up period 373

374

17

TABLE 2. Patient characteristics, microbiological and clinical outcomes in patients treated 376

with gentamicin only vs. those receiving concomitant systemic antibiotic therapy (CSAT). 377

378

Oral gentamicin only

(n= 23)

Oral gentamicin plus CSAT

(n= 27) p

Male 15 (65%) 20 (74%) 0.710

Mean age, years ± SD 67 ± 12 61 ± 15 0.108

ICU’s patients 1 (4%) 9 (33%) < 0.001

Median duration of treatment, days (range)

14 [11 – 19] 18 [12 – 29] 0.311

Gut decontamination 22/23 (96%) 12/27 (44%) < 0.001

KPC-Kp infection a 2/23 (9%) 15/27 (56%) 0.003

Overall death a 5/23 (22%) 10/27 (37%) 0.406

379 a in the 6-month follow-up period 380

18

TABLE 3. Univariate logistic regression for gut decontamination event. Odds Ratio, 95% 381 Confidence Interval for Odds Ratio and p-value of each predictor. 382

383

OR 95% CI OR p

CSAT 0.036 0.002 – 0.213 0.002

KPC-Kp infection 0.057 0.011 – 0.231 <0.001

ICU’s patients 0.062 0.008 – 0.305 0.002

Male gender 0.697 0.165 – 2.548 0.598

Age (years) 1.028 0.985 – 1.076 0.200

Duration of Gentamicin treatment (days) 0.966 0.913 – 1.015 0.198

384

19

TABLE 4. Multivariate logistic regression for gut decontamination event. Odds Ratio, 95% 385

Confidence Interval for Odds Ratio and p-value of each predictor. 386

387

OR 95% CI OR p

CSAT 0.105 0.005 – 0.801 0.048

KPC-Kp infection 0.139 0.021 – 0.754 < 0.001

ICU’s patients 0.132 0.011 – 0.992 0.066

388

20

FIGURE 1. Probability of persistent carriage in patients treated with oral gentamicin only 389

[SOLID LINE] vs. those receiving also concomitant systemic antibiotic therapy [DOT LINE] (p 390

= 0.007). 391