Antimicrobial Monotherapy versus Combination Therapy for the Treatment of Complicated Intra-Abdominal Infections

Transcription

1 Antimicrobial Mono versus Combination Therapy for the Treatment of Complicated Intra-Abdominal Infections Sarah E. Petite, 1 * Seth R. Bauer, 2 Jessica E. Bollinger, 2 Christine L. Ahrens, 2 and Lisa M. Harinstein 2 1 Department of Pharmacy Practice, University of Toledo College of Pharmacy and Pharmaceutical Sciences, Toledo, Ohio; 2 Department of Pharmacy, Cleveland Clinic, Cleveland, Ohio STUDY OBJECTIVE It is unknown if b-lactam mono is sufficient for complicated intra-abdominal infections or if broader coverage is required, such as with vancomycin. This study sought to determine the clinical outcomes of piperacillin/tazobactam () mono compared to combination with vancomycin and for complicated intra-abdominal infections among patients within a surgical intensive care unit (ICU). DESIGN Retrospective cohort study. SETTING Three surgical ICUs at a tertiary academic medical center. PATIENTS Four hundred seventeen patients with a secondary peritonitis identified by International Classification of Diseases, Ninth Revision codes who received either mono (228 patients) or and vancomycin combination (189 patients). MEASUREMENTS AND MAIN RESULTS The primary outcome was day 28 clinical cure; secondary outcomes included day 7 clinical cure, length of stay (LOS), and mortality. There were no statistically significant differences between the mono and combination groups with respect to day 28 clinical cure (33.9% vs 25.5%, p=0.064), day 7 clinical cure (23.6% vs 17.6%, p=0.14), or 28-day mortality (7% vs 7.9%, p=0.72). LOS in the ICU was significantly shorter in the mono group (6 days) compared with the combination group (7 days; p=0.04); however, hospital LOS was not significantly different. CONCLUSIONS No difference was observed in clinical cure rates at day 28 or day 7 between those who received mono compared to and vancomycin combination. KEY WORDS infectious disease, critical care, vancomycin. (Pharmaco 2016;36(11): ) doi: /phar.1847 Complicated intra-abdominal infections (ciais) are associated with significant mortality in intensive care unit (ICU) patients and a leading cause of infectious death. 1, 2 Appropriate treatment for intra-abdominal infections consists of a combination of source control and Conflicts of Interest and Source of Funding: None were declared. *Address for correspondence: Sarah E. Petite, Department of Pharmacy Practice, University of Toledo College of Pharmacy, 3000 Arlington Avenue, MS #1013, Toledo, OH 43614; sarah.petite@utoledo.edu. Ó 2016 Pharmaco Publications, Inc. antimicrobial targeted at the most likely causative pathogens, including facultative and aerobic gram-negative bacilli and anaerobic organisms. 3 Depending on patient-specific risk factors, other bacterial pathogens such as Enterococcus species and methicillin-resistant Staphylococcus aureus (MRSA) may require empiric coverage, resulting in combination antimicrobial. 3, 4 Prior literature has demonstrated conflicting results when Enterococcus has not been sufficiently covered by empiric antibiotics and isolated from ciais; some studies have shown this to be a predictor of treatment failure. 5, 6

2 INTRA-ABDOMINAL INFECTION TREATMENT Petite et al 1139 However, other studies have demonstrated no difference in clinical outcomes. 7, 8 Data are limited for ICU patients regarding the need for empiric enterococci coverage. Initial antimicrobial providing enterococcal coverage includes ampicillin, piperacillin/tazobactam (), and vancomycin, with the recommendation that an agent should be chosen based on the local antibiogram. 3 This recommendation leaves room for interpretation and offers the ability to use combination, typically the addition of vancomycin to a b-lactam, to enhance the likelihood of empiric coverage of enterococci and MRSA. However, combination with vancomycin added to a b-lactam/b-lactamase inhibitor raises safety and antimicrobial resistance concerns. While the incidence of nephrotoxicity with vancomycin is reported to be 5 35%, the rate may be higher with concomitant administration of a b-lactam Prolonged durations of vancomycin, trough concentrations of 15 mg/l, and total daily doses > 4 g are other factors associated with vancomycin nephrotoxicity. 12 In addition, use of combination may result in exposure to unnecessary antibiotics. With an emphasis on antimicrobial stewardship, recent evidence has shown that decreasing the duration of antimicrobial exposure in patients with ciais did not affect treatment outcome but decreased the number of days patients received antimicrobials. 13 Investigating mono versus combination may aid in further minimizing antimicrobial exposure for treatment of ciais. The goal of this study was to determine if there is clinical benefit gained from the addition of vancomycin to compared to mono for the treatment of ciais. Methods This was a noninterventional, retrospective, single-center cohort study that included all patients admitted to the surgical intensive care unit (SICU) at Cleveland Clinic between September 2010 and July This study was approved by the institutional review board with waiver for written informed consent. Complicated intra-abdominal infection was defined based on the site of infection. Possible sites of infection included intra-abdominal abscess, gastric or duodenal ulcer perforation, peritonitis with fecal contamination, large or small intestinal perforation with or without abscess or fecal contamination, and appendicitis, cholecystitis, or diverticulitis complicated by perforation with or without abscess formation. Patients were identified via International Classification of Diseases, Ninth Revision (ICD-9) codes as having a ciai treated with at least 72 hours of. In addition, patients had to have either a white blood cell (WBC) count > 12,000 or < 4000 cells/mm 3 or a temperature 36 C or 38 C to be included. 14 Patients were excluded if they were less than 18 years of age, had not undergone a surgical procedure for source control within 48 hours after starting antimicrobials, had pancreatitis or primary peritonitis, were immunocompromised, had a concomitant infection at a site other than the abdomen, or received 24 hours of active antimicrobial other than within the 72 hours before the initiation of. Active antimicrobial was defined according to the Infectious Diseases Society of America guideline recommendations. 3 Patients were further divided into two groups: those who received only mono or those who received combination with vancomycin initiated within 48 hours after starting. The primary objective was to determine if there was a difference between the two treatment groups in the clinical cure rate at day 28 or at hospital discharge if prior to 28 days. Secondary objectives included differences in the rate of clinical cure at day 7, microbiology results, incidence of nephrotoxicity, as well as hospital and ICU length of stay and mortality. Data collected from each patient s medical record included demographic data (age, gender), height, weight, components of the Acute Physiology and Chronic Health Evaluation (APACHE) III score, daily antimicrobial treatment (PIP/ TAZ, vancomycin, or other antimicrobials), microbiological data, length of stay after PIP/ TAZ initiation, timing of antibiotic, daily temperature and WBC count, and mortality. Renal function was assessed at baseline (day 1 of ) and day 7 by recording serum creatinine and the need for renal replacement. Clinical cure was defined as temperature C for 48 hours and WBC count ,000 cells/mm 3. The definition of clinical failure was any of the following: (i) addition of second line agents for 72 hours, (ii) unplanned reoperation or percutaneous drainage, (iii) death, (iv) temperature 36 C or 38 C or (v)

3 1140 PHARMACOTHERAPY Volume 36, Number 11, 2016 WBC count 12,000 or 4000 cells/mm rescue was defined as reinitiation of 72 hours after completing the initial course of. Vancomycin rescue was defined as initiation of vancomycin 48 hours after initiation in the mono group and 72 hours after completing the initial course of vancomycin in the combination group. Other antimicrobials were defined as rescue if patients received 72 hours of. APACHE III scores were collected on day 1 of SICU admission to characterize baseline severity of illness. Day 1 vasopressor requirements were defined as any requirement of vasopressor medication on the first day of. Vancomycin troughs were assessed in the combination group for appropriateness (drawn no earlier than prior to the fourth dose) and within goal range (10 20 mcg/ml). Nephrotoxicity was defined using the Kidney Disease Improving Global Outcomes (KDIGO) criteria. 15 All three stages of acute kidney injury per the KDIGO criteria were included in the definition of nephrotoxicity. Baseline renal function was defined as day 1 of. Due to potential discrepancies in recorded urine output, only the serum creatinine criteria of the KDIGO criteria was used to assess nephrotoxicity. Statistical Analysis In previous ciai studies, a clinical cure rate of 78.8% was considered acceptable. 14 Assuming this clinical cure rate at day 28, a difference of 15% between the two treatment groups with 80% power and 5% a would require at least 142 patients in each group. Nominal data were analyzed using either Fisher s exact test or v 2 test and continuous data was reported using the Mann Whitney U test. The Mann Whitney U test was chosen for continuous data because assessment of normality with the Shapiro Wilk test demonstrated that the data were not normally distributed. All tests of significance were two-tailed, and p<0.05 was considered statistically significant. A propensity score analysis was performed to determine the probability that patients would receive combination, with variables being included if deemed by the study investigators to influence prescribing of combination. 16 The variables considered for the propensity score analysis included receipt of vasopressors at day 1, APACHE III score, infection type, MRSA or enterococcal infection within the past year, and MRSA nares colonization culture positive. A multivariable logistic regression was performed to assess for factors associated with clinical cure, including the propensity score for combination prescription. Variables were chosen based on biological plausibility for affecting the outcome and statistical significance (p<0.05) on univariable analysis. To avoid duplication of included variables, those variables included in the propensity score analysis were not included in the multivariable analysis. All statistical analyses were performed using JMP Pro version 10 (SAS Institute Inc., Cary, NC). Results are presented as n (%) or median (interquartile range [IQR]) when in tabular format. Results During the study period, 664 patients were identified (Figure 1). The primary reason for study exclusion was immunosuppression or receipt of an insufficient duration of study medication. There were 228 and 189 patients included in the mono group and combination group, respectively. Baseline characteristics were similar between the groups with a few notable exceptions (Table 1). The requirement of vasopressors on day 1 of was significantly more frequent in the combination group (p=0.005). In addition, significantly more patients in the combination group were colonized with MRSA, based on positive nares cultures (p=0.018). Infection types were similar between groups; however, appendicitis occurred more frequently in the mono group (p=0.014). The most common type of surgical 247 pa ents excluded 116 immunosuppressed* 95 <72 hours 19 concomitant infec on** 17 prior ac ve an microbial N= pa ents iden fied + Vanco N=189 Figure 1. Study enrollment flowchart. *89 transplant, 22 rheumatologic disorders on immunosuppressants, 5 human immunodeficiency virus. **Concomitant infections: 7 Clostridium difficile, 8 urinary tract infections, 4 respiratory infections.

4 INTRA-ABDOMINAL INFECTION TREATMENT Petite et al 1141 Table 1. Baseline Characteristics Characteristic n=228 + Vanco n=189 p Gender (male) 116 (50.9%) 99 (52.4%) 0.76 Age (years) 63 (51 73) 62 ( ) 0.75 Weight (kg) 80.1 ( ) 83 ( ) 0.32 Height (in) 67 (64 70) 67 (64 70) 0.49 APACHE III 67 (54 82) 68 (52 87) 0.46 n=214 n=176 Day 1 vasopressors 27 (11.8%) 42 (22.2%) Hospital 115 (50.4%) 87 (46%) 0.38 admission within 1 year MRSA nasal 44 (19.3%) 57 (30.2%) colonizer MRSA infection 9 (4%) 4 (2.1%) 0.28 within 1 year Enterococcus 26 (11.4%) 17 (9%) 0.42 infection within 1 year Cephalosporin 29 (12.7%) 23 (12.2%) 0.54 use within 1 year Infection type a Peritonitis 129 (56.6%) 119 (62.9%) 0.19 with fecal contamination Intraabdominal 67 (29.4%) 54 (28.6%) 0.86 abscess Diverticulitis 51 (22.4%) 33 (17.5%) 0.21 Intestinal 30 (13.2%) 36 (19.1%) 0.1 perforation Cholecystitis 34 (14.9%) 23 (12.2%) 0.42 Appendicitis 10 (4.4%) 1 (0.5%) Procedure type b Open 162 (71.1%) 125 (66.1%) 0.28 Percutaneous 123 (54%) 112 (59.3%) 0.28 drainage Laparoscopic 24 (10.5%) 16 (8.5%) 0.48 APACHE = Acute Physiology and Chronic Health Evaluation; MRSA = methicillin-resistant Staphylococcus aureus; = piperacillin/tazobactam; Vanco = vancomycin. Data presented as n (%) or median (interquartile range). a More than one infection type was possible for included patients. b More than one procedure type was possible for included patients. procedure experienced at study inclusion was an open procedure, such as an open laparotomy (p=0.28). The primary outcome of day-28 clinical cure occurred more frequently in the mono group than the combination group (33.9 vs 25.5%; p=0.064), but this did not reach statistical significance (Table 2). Clinical failure was driven by two outcomes: continued abnormality in WBC count or elevation in temperature (39.5% vs 48.7%; p=0.16) and escalation in antimicrobial (53.5% vs 56.6%; p=0.53). Day 7 clinical cure also occurred more frequently in the mono group (23.6% vs 17.6%; p=0.14). Similar to the day-28 outcome, a continued abnormality in WBC count or temperature elevation (59.7% vs 70.4%; p=0.02), and escalation in antimicrobial (33.6% vs 41.3%; p=0.095) were the primary reasons for clinical failure. Table 2. Day 28 and Day 7 Clinical Cure n=228 + Vanco N=189 p Day 28 outcome Day 28 clinical 77 (33.9%) 48 (25.5%) cure a Elevated WBC 90 (39.5%) 92 (48.7%) 0.16 count or temperature Unplanned 38 (16.7%) 38 (20.1%) 0.37 procedure Escalation in 122 (53.5%) 107 (56.6%) day mortality 16 (7%) 15 (7.9%) 0.72 n=226 n=188 Day 7 outcome Day 7 clinical 53 (23.6%) 33 (17.6%) 0.14 cure Elevated WBC 135 (59.7%) 133 (70.4%) 0.02 count or temperature Unplanned 11 (4.8%) 13 (6.9%) 0.37 procedure Escalation in 76 (33.6%) 78 (41.3%) day mortality 2 (0.1%) 3 (1.6%) 0.5 = piperacillin/tazobactam; Vanco = vancomycin. a Patients may have fulfilled multiple criteria for clinical failure. The duration of the initial was 8.5 ( ) days and 8.8 ( ) days in the mono and combination groups, respectively (p=0.68). Patients in the combination group were initiated on vancomycin 0.25 (0 12) hours within the start of. Initial vancomycin duration was 5.2 ( ) days in the combination group. Vancomycin troughs were drawn appropriately in 56.4% of patients and 71% were within goal range. Length of stay in the SICU was significantly longer in the combination group (7 [3 21] days vs 6 [2 14] days; p=0.04). However, hospital length of stay was not significantly different between the groups (22 [14 33] days vs 20 [13 31] days; p=0.072). Rescue antimicrobial is presented in Table 3. The most common rescue antimicrobial agents used were vancomycin (25.4% vs 28%; p=0.55) and meropenem (23.7% vs 29.1%; p=0.21). No differences in microbiologic outcomes were observed in regard to gram-positive organisms (16.7% vs 20.1%; p=0.37). The most commonly isolated gram-positive pathogens were Enterococcus faecalis (4.4% vs 5.3%; p=0.67), Enterococcus faecium (9.2% vs 10%; p=0.77), and MRSA (2.2% vs 3.7%; p=0.36). There were significantly more gram-negative organisms isolated in the combination group (32.5% vs 43.3%; p=0.02); however, there

5 1142 PHARMACOTHERAPY Volume 36, Number 11, 2016 were no significant differences in the species isolated. The most commonly isolated gram-negative organisms were Escherichia coli (16.2% vs 12.7%; p=0.31), Pseudomonas aeruginosa (10.5% vs 14.8%; p=0.19), and Klebsiella pneumoniae (7% vs 5.8%; p=0.62). There was a trend toward a significantly higher number of fungal organisms isolated in the combination group (19.3% vs 27%; p=0.06) and the most frequently isolated species were Candida albicans (13.2% vs 15.9%; p=0.43) and Candida glabrata (3.5% vs 9%; p=0.04). Further evaluation of the individuals who had a culture positive for enterococci or MRSA was performed. The enterococcal susceptibilities to the study medications are reported in Table 4. The only risk factor significantly associated with a culture positive for enterococci was a history of an enterococcal infection in the past year (p=0.008). Severity of illness (APACHE III or day 1 requirement for vasopressors), hospital admission within the past year, and cephalosporin use within the past year were not significantly associated with development of an enterococcal infection. In the 12 individuals with a positive MRSA culture, MRSA infection in the previous year (p=0.049) and MRSA nares culture positive (p<0.001) were significantly associated with the isolation of MRSA. In regard Table 3. Rescue Antimicrobial Therapy Table 4. Enterococcus Sensitivities Outcome n=228 n=31 + Vanco n=189 p Outcome Received rescue 122 (53.5%) 107 (56.6%) 0.53 Aminoglycosides 5 (2.2%) 7 (3.7%) 0.23 Ciprofloxacin 3 (1.3%) 3 (1.6%) 1 Colistimethate 2 (0.9%) 1 (0.5%) 1 Daptomycin 23 (10.1%) 23 (12.2%) 0.49 Imipenem/cilastatin 1 (0.4%) 3 (1.6%) 0.33 Linezolid 7 (3.1%) 1 (0.5%) Meropenem 54 (23.7%) 55 (29.1%) (14%) 20 (10.6%) 0.29 Tigecycline 12 (5.3%) 8 (4.2%) 0.62 Vanco 58 (25.4%) 53 (28%) 0.55 = piperacillin/tazobactam; Vanco = vancomycin. + Vanco n=29 p Amp S + Vanco S 12 (38.7%) 12 (41.4%) 0.83 Amp S + Vanco R 2 (6.5%) Amp R + Vanco S 1 (3.2%) 2 (6.9%) 0.51 Amp R + Vanco R 16 (51.6%) 16 (51.7%) 0.99 = piperacillin/tazobactam; Vanco = vancomycin; Amp = ampicillin; S = sensitive; R = resistant. to the gram-negative organisms isolated in this patient population, 67.6% and 65.9% were sensitive to in the mono and combination groups, respectively (p=0.91). The incidence of nephrotoxicity was not significantly different between groups; 21.8% of patients on mono and 27.4% of patients on combination developed some degree of renal impairment during the study period (p=0.18). The need for renal replacement was similar between the mono and combination groups (10.7% vs 13.3%; p=0.42). No significant differences were observed with regard to the stages of nephrotoxicity in the treatment groups. Due to the retrospective nature of the study, we were unable to capture the reasons for empiric initiation of combination, such as perceived severity of illness. To account for channeling bias with prescription of combination, a propensity score for receipt of combination was calculated. A total of 384 patients were included in this analysis due to missing values that did not allow for APACHE III scores to be calculated in all 417 patients. A multivariable logistic regression, including the propensity score for combination prescription, was subsequently performed to assess for factors significantly associated with day 28 clinical cure (Table 5). No factors included in the model were found to be significantly associated with clinical cure at day 28. Discussion To the best of our knowledge, this is the first study to evaluate the use of combination PIP/ TAZ and vancomycin to treat ciais in a patient population with a high severity of illness. We found no differences in the rate of clinical cure at day 28 and day 7 between patients that received empiric mono compared to and vancomycin combination. Table 5. Multivariable Logistic Regression Day 28 Clinical Cure Outcome OR 95% CI p treatment group Positive bacterial culture Positive C. glabrata culture Propensity score for vancomycin receipt CI = confidence interval; OR = odds ratio; = piperacillin/ tazobactam.

6 INTRA-ABDOMINAL INFECTION TREATMENT Petite et al 1143 The low rate of clinical cure found in this study compared to previous ciai trials is likely due to the severity of illness in this patient population, demonstrated by the APACHE III scores While the clinical cure study definition utilized in this study was similar to other ciai trials, the high severity of illness in this patient population may have resulted in lower rates of clinical cure using this definition. Previous studies have evaluated a less critically ill patient population that is more likely to achieve a clinical cure at day 28. There are limited data regarding appropriate clinical cure rates in an ICU patient population and the study was powered from available cure rates. The predominant reason for clinical failure was escalation of antimicrobial. Due to the retrospective nature of this study, it is difficult to determine if the escalation was indicated based on the clinical picture. Microbial resistance to was observed in approximately one-third of patients in each group and one-half of isolated Enterococcus was resistant to ampicillin and vancomycin; however, this does not account for all antimicrobial escalation that occurred. In addition to the use of combination broadspectrum antimicrobials, prolonged durations of antimicrobial are often used for ciai treatment. 13 The guideline-recommended duration of antimicrobials for treatment of ciais is 4 7 days depending on patient response; however, the optimal duration of is unknown. 3,20 One study evaluated a fixed antimicrobial course of 4 days compared to a maximum of 10 days, both in conjunction with adequate source control of the infection. 13 Similar incidences of recurrent intra-abdominal infection, surgical-site infection or death occurred in the two groups but no assessment of differences between antimicrobial regimens was performed. Although APACHE III scores were not significantly different between groups on day 1 of SICU stay, more patients received vasopressors on day 1 of in the combination group. This discrepancy is likely due to APACHE III scores being calculated within 24 hours of ICU admission, while day 1 of PIP/ TAZ could have occurred at any time during the ICU stay. This illustrates the potential difference in study groups on day 1 of antimicrobial. The differences in severity of illness between groups may, at least partially, explain the numeric differences found with day 28 and day 7 clinical cure rates. Despite no statistically significant differences found with day 28 clinical cure, there was a trend toward higher rates of clinical cure in the mono group. However, patient group was not significantly associated with 28-day clinical cure on multivariable analysis. This analysis accounted for imbalances between the two groups by including the propensity score for combination prescription in addition to other factors that may have influenced clinical cure rates. Patients in the combination group had a significantly longer SICU length of stay. This may be due to a more complicated clinical picture, but because of the study design, we were unable to determine the reason for longer length of stay. The combination group also had a numerically higher incidence of nephrotoxicity, which may be due in part to the higher percentage of patients requiring vasopressors at baseline. This finding also may have contributed to a longer SICU length of stay. Our institution s incidence of ampicillin and vancomycin-resistant Enterococcus is > 50% of isolates. In addition, the next most common Enterococcus species isolated in this patient population was sensitive to both ampicillin and vancomycin. There is minimal additional clinical benefit gained from vancomycin for empiric Enterococcus coverage in this patient population because more than half of isolates will not respond to this antimicrobial agent. As we were unable to identify risk factors in this study for Enterococcus, other than a history of an enterococcal infection, our data suggest that PIP/ TAZ mono is sufficient for ciai treatment. If a patient has risk factors for Enterococcus, should be tailored to the institutional antibiogram. In patients with risk factors at our institution, empiric coverage for vancomycin-resistant Enterococcus with daptomycin or linezolid is considered. Strengths of this study include the large sample size and methods used to account for confounding variables. The use of the propensity score for vancomycin receipt allowed for the control of variables that may cause channeling bias for combination, such as severity of illness on day 1. The propensity score and multivariable analysis were strengths that controlled for identified factors imbalanced between groups. Limitations to this study were the clinical cure and failure definitions utilized. There is no set definition in the literature for clinical failure, specifically the duration of rescue antimicrobial. Due to the low rates of clinical cure

7 1144 PHARMACOTHERAPY Volume 36, Number 11, 2016 observed in this study, there is a potential for unaccounted variables that led to these outcomes. However, there were high rates of antimicrobial resistance in this patient population that also contributed to clinical failure. Patients were allocated to the combination group if vancomycin was initiated within 48 hours of. This time period was chosen to provide adequate time for clinical evaluation of the patient, but we were unable to assess the intention of empiric combination or escalation of with the addition of vancomycin. Source control was not directly assessed and reoperation was used as a surrogate marker for this outcome. In addition, the power analysis was based from studies in a lower severity of illness and may not be an appropriate clinical cure rate in this patient population. Conclusions This study did not detect a difference in clinical cure rate when utilizing and vancomycin combination compared to PIP/ TAZ mono for the treatment of ciais. Treatment should be tailored to each institution s local antibiogram. At our institution, the high rates of vancomycin-resistant Enterococcus demonstrate that the addition of vancomycin to empiric ciai may not be warranted. In this study, the addition of empiric vancomycin to b-lactam/b-lactamase inhibitor for ciais did not improve clinical outcomes. References 1. Brun-Buisson C, Doyon F, Carlet J, et al. Incidence, risk factors, and outcome of severe sepsis and septic shock in adults. A multicenter prospective study in intensive care units. French ICU Group for Severe Sepsis. JAMA 1995;274: Zahar JR, Timsit JF, Garrouste-Orgeas M, et al. Outcomes in severe sepsis and patients with septic shock: pathogen species and infection sites are not associated with mortality. Crit Care Med 2011;39: Solomkin JS, Mazuski JE, Bradley JS, et al. Diagnosis and management of complicated intra-abdominal infection in adults and children: guidelines by the Surgical Infection Society and the Infectious Diseases Society of America. Clin Infect Dis 2010;50: Fierobe L, Decre D, Muller C, et al. Methicillin-resistant Staphylococcus aureus as a causative agent of postoperative intra-abdominal infection: relation to nasal colonization. Clin Infect Dis 1999;29: Burnett RJ, Haverstock DC, Dellinger EP, et al. Definition of the role of enterococcus in intraabdominal infection: analysis of a prospective randomized trial. Surgery 1995;118:716 21; discussion Sitges-Serra A, Lopez MJ, Girvent M, et al. Postoperative enterococcal infection after treatment of complicated intraabdominal sepsis. Br J Surg 2002;89: Ohlin B, Cederberg A, Forssell H, et al. Piperacillin/tazobactam compared with cefuroxime/metronidazole in the treatment of intra-abdominal infections. Eur J Surg 1999;165: Teppler H, McCarroll K, Gesser RM, et al. Surgical infections with enterococcus: outcome in patients treated with ertapenem versus piperacillin-tazobactam. Surg Infect (Larchmt) 2002;3: Burgess LD, Drew RH. Comparison of the incidence of vancomycin-induced nephrotoxicity in hospitalized patients with and without concomitant piperacillin-tazobactam. Pharmaco 2014;34: Gomes DM, Smotherman C, Birch A, et al. Comparison of acute kidney injury during treatment with vancomycin in combination with piperacillin-tazobactam or cefepime. Pharmaco 2014;34: Meaney CJ, Hynicka LM, Tsoukleris MG. Vancomycin-associated nephrotoxicity in adult medicine patients: incidence, outcomes, and risk factors. Pharmaco 2014;34: Elyasi S, Khalili H, Dashti-Khavidaki S, et al. Vancomycininduced nephrotoxicity: mechanism, incidence, risk factors and special populations. A literature review. Eur J Clin Pharmacol 2012;68: Sawyer RG, Claridge JA, Nathens AB, et al. Trial of shortcourse antimicrobial for intraabdominal infection. N Engl J Med 2015;372: Food and Drug Administration, U.S. Department of Health and Human Services. Draft Guidance for Industry on Complicated Intra-Abdominal Infections: Developing Drugs for Treatment; Availability. Docket No. FDA-2012-D-0973; Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Section 2: AKI Definition. Kidney Int Suppl 2012;2: D Agostino RB Jr. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17: Solomkin JS, Reinhart HH, Dellinger EP, et al. Results of a randomized trial comparing sequential intravenous/oral treatment with ciprofloxacin plus metronidazole to imipenem/cilastatin for intra-abdominal infections. The Intra-Abdominal Infection Study Group. Ann Surg 1996;223: Solomkin JS, Wilson SE, Christou NV, et al. Results of a clinical trial of clinafloxacin versus imipenem/cilastatin for intraabdominal infections. Ann Surg 2001;233: Solomkin JS, Yellin AE, Rotstein OD, et al. Ertapenem versus piperacillin/tazobactam in the treatment of complicated intraabdominal infections: results of a double-blind, randomized comparative phase III trial. Ann Surg 2003;237: Schein M, Assalia A, Bachus H. Minimal antibiotic after emergency abdominal surgery: a prospective study. Br J Surg 1994;81: