Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 
  • Users Online: 276
  • Home
  • Print this page
  • Email this page

 Table of Contents  
Year : 2017  |  Volume : 3  |  Issue : 2  |  Page : 256-262

Prophylactic antibiotics for extracorporeal membrane oxygenation in critically-Ill adults

1 Department of Pharmacy, University of Maryland Medical Center, Baltimore, MD, USA
2 Department of Epidemiology and Public Health, University of Maryland School of Medicine, Baltimore, MD, USA
3 Department of Medicine, University of Maryland School of Medicine, Baltimore, MD, USA
4 Department of Surgery, University of Maryland School of Medicine, Baltimore, MD, USA

Date of Web Publication9-Jan-2018

Correspondence Address:
Dr. Omayma A Kishk
University of Maryland Medical Center, 22 S. Greene Street, Room S8A06, Baltimore, MD 21201
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/IJAM.IJAM_90_16

Rights and Permissions

Context and Aims: Primary objective of this study was to estimate time to infection after extracorporeal membrane oxygenation (ECMO) initiation among patients who received prophylactic antibiotics compared to those who did not.
Subjects and Methods: This retrospective cohort study included adult patients initiated on ECMO between February 1, 2009, and June 30, 2012, in Intensive Care Units at an 816-bed tertiary care hospital.
Results: Prophylaxis defined as an antibiotic initiated at start of ECMO without other bacterial infectious indications was categorized into three levels: (1) No prophylaxis, (2) single antibiotic, and (3) combination two or more antibiotics with activity against methicillin-resistant Staphylococcus aureus and Gram-negative bacteria. Of the 96 patients included, 37 (39%) patients acquired a sterile site infection; 17 of which were caused by of multidrug-resistant organisms. There was no significant difference in the proportion of new infections by antibiotic exposure group. By 500 h on ECMO, all patients had a 60% probability of developing infection. There was a trend that patients on combination antibiotics were less likely to develop a sterile site infection than patients who were not exposed to antibiotics after adjusting for age and comorbid illnesses (adjusted hazard ratio 0.44, 95% confidence interval 0.17–1.13) (P = 0.09).
Conclusions: Patients on ECMO on single and combination antibiotic regimens demonstrated no significant difference in the hazard for the development of a new sterile site infection compared to no antibiotic group.
The following core competencies are addressed in this article: Medical knowledge, Patient care, Practice-based learning and improvement, Systems-based practice.

Keywords: Antibiotic prophylaxis, critically ill, extracorporeal membrane oxygenation, multidrug-resistant organism

How to cite this article:
Kishk OA, Stafford KA, Pajoumand M, Williams CP, Thom KA, Kon ZN, Herr DL, Amoroso A. Prophylactic antibiotics for extracorporeal membrane oxygenation in critically-Ill adults. Int J Acad Med 2017;3:256-62

How to cite this URL:
Kishk OA, Stafford KA, Pajoumand M, Williams CP, Thom KA, Kon ZN, Herr DL, Amoroso A. Prophylactic antibiotics for extracorporeal membrane oxygenation in critically-Ill adults. Int J Acad Med [serial online] 2017 [cited 2022 Sep 28];3:256-62. Available from: https://www.ijam-web.org/text.asp?2017/3/2/256/222484

  Introduction Top

The use of extracorporeal membrane oxygenation (ECMO) for life support is associated with high rates of hospital-acquired infections, ranging from 7.5% to 46%,[1],[2] with the highest incidence in the adult population.[3] The majority of these are bloodstream infections.[4] During severe respiratory and cardiac failure, the empiric use of antibiotics is common. The continued use of antibiotics throughout the duration of ECMO support is also a frequently used intervention to attempt to minimize the acquisition of serious nosocomial infections. The literature contains sparse information related to the indication, type, or duration of prophylactic antibiotic therapy in patients treated with ECMO and their relationship to the subsequent acquisition of hospital acquired infections.[3],[4],[5],[6],[7],[8] As a result, there is no clear consensus for the use of prophylactic antibiotics for the duration of ECMO support and current practice varies widely. In a survey conducted in 132 Extracorporeal Life Support Organization centers,[9] 74% of the centers routinely administered antibiotic prophylaxis during ECMO, however, only 49% of these centers made use of standardized protocols for antibiotic prophylaxis.

At our institution, until 2013, there was no standard protocol for the use of prophylactic antibiotics when a patient was placed on ECMO support, allowing for an opportunity to study the utility of this intervention to reduce infections. We performed a retrospective cohort study of patients on ECMO support to investigate the association between prophylactic antibiotics and the development of sterile-site infections related to ECMO support.

  Subjects and Methods Top

The study was performed at the University of Maryland Medical Center (UMMC), an 816-bed tertiary care hospital located in Baltimore, Maryland, which houses a level-1 trauma center and an extracorporeal life-support referral center. All adult patients between the ages of 18 and 89 years of age that were initiated on ECMO at UMMC between February 1, 2009, and June 30, 2012, were eligible for inclusion in the study. Patients who were on ECMO for <48 h or who developed an infection within 48 h of cannulation were excluded from the study.

We conducted a retrospective cohort study to estimate the association between three antibiotic prophylaxis strategies and the rate of acquisition of a sterile site infection occurring after at least 48 h of ECMO support. We further assessed the proportion of multidrug-resistant organisms (MDROs) as a cause of these infections and the occurrence of Clostridium difficile as secondary outcome. This study was approved by the University of Maryland Institutional Review Board.

Prophylaxis was defined as an antibiotic started at the time of ECMO cannulation and was continued without any other documented infection or infectious indication in the chart. Prophylactic antibiotic exposure was categorized into three clinically relevant levels: (1) no prophylactic antibiotics, (2) a single prophylactic antibiotic, and (3) a combination of more than one prophylactic antibiotic, including at least one drug with broad-spectrum against Gram-negative (GN) bacteria and one drug with activity against methicillin-resistant Staphylococcus aureus (MRSA). Combination antibiotic prophylaxis constituted being on one of the following antibiotic regimens at the time of cannulation (1) a broad-spectrum beta-lactam antibiotic and an anti-MRSA antibiotic, (2) a broad-spectrum beta-lactam antibiotic, an aminoglycoside, and an anti-MRSA antibiotic, or (3) a polymyxin antibiotic (polymyxin B or colistin) and an anti-MRSA antibiotic. Anti-MRSA antibiotics included vancomycin, linezolid, daptomycin, or rifampin. Broad-spectrum beta-lactam antibiotics with activity against GN bacteria included cefepime, ceftriaxone, aztreonam, piperacillin/tazobactam, meropenem, imipenem, or doripenem.

Acquisition of a new infection was defined as recovery of a bacterial organism identified from sterile sites (e.g., blood, cerebrospinal fluid, pleural fluid) at least 48 h after initiating ECMO and up to 7 days after decannulation. An MDRO was defined as the acquisition of MRSA, vancomycin-resistant enterococci, and multidrug-resistant GN (MDR-GN) bacteria. MDR-GN rods bacteria are defined at our institution as those GN bacteria resistant to two of the following three drugs: piperacillin/tazobactam, cefepime, or meropenem (or imipenem if tested). C. difficile infections were identified from clinical samples sent by the primary team using polymerase chain reaction. Additional data collected included demographic information, presence of comorbid diseases, calculated Charlson score, indication for ECMO, veno-venous (VV) or veno-arterial (VA) ECMO, duration of ECMO course, sterile site infections, antibiotics given before and during ECMO support, and hospital and Intensive Care Unit length of stay. All data were collected from direct chart review, except the presence of comorbid diseases used to calculate the Charlson score, which was obtained from the UMMC Central Data Repository.

Data were first investigated using univariate analysis to describe the frequency and distribution of the outcome, primary exposure, and covariates. The crude association between antibiotic exposure and acquisition of an infection was initially investigated in bivariate analysis using a Chi-square test. The association between baseline covariates and both the exposure group and sterile site infection were investigated in bivariate analysis using Chi-square tests or Fisher's exact test for categorical variables and t-tests for continuous variables. Time to event analysis was performed using the Kaplan–Meier method. Differences in time to event were evaluated using the log-rank test. Patients that did not experience an infection were censored at the time of discharge or at the end of the study period. Cox proportional hazards models were used to estimate the hazard of sterile site infection by exposure group. Rates of MDRO per 1000 ECMO hours for each exposure group were calculated. All statistical analyses were carried out using SAS 9.3 (SAS Institute Inc., Cary, NC, USA).

  Results Top

A total of 139 patients underwent ECMO for respiratory or cardiac support during the study period. A total of 96 patients were included into the study; 43 patients were excluded from the study (41 were on ECMO for <48 h, and two patients were cannulated at an outside hospital). Fifty-four percent of patients were male, and 60% were white with a mean age of 48 years [Table 1]. Sixty-two percent of the cohort received VV-only ECMO support. Five patients were switched from VA ECMO support to VV ECMO support. Seven patients had infectious indications for the use of ECMO. Of these, four had pneumonia caused by influenza A. Eleven patients were not placed on any antibiotics at the initiation of ECMO support. Twenty-four patients (25%) were placed on a single antibiotic at the initiation of ECMO. Of these 24 patients, 13 (54%) were treated with vancomycin, four with piperacillin/tazobactam, four with carbapenems, four with moxifloxacin, two with clindamycin, and one each on cefepime, cefazolin, metronidazole, azithromycin, ceftriaxone, and sulfamethoxazole/trimethoprim. Sixty-one patients (64%) were placed on a combination antibiotic regimen at the initiation of ECMO. Of these patients, 57 patients received a broad-spectrum beta-lactam antibiotic and an anti-MRSA-antibiotic combination, four patients received an aminoglycoside in addition to a broad-spectrum beta-lactam antibiotic and an anti-MRSA antibiotic. There was no difference between prophylaxis groups and the type of ECMO (P = 0.89) or indication for ECMO (P = 0.95). Patients who received a combination antibiotic regimen were on average older and had a higher body mass index, but this was not statistically significant. The average duration of ECMO support lasted 335 h (95% confidence interval (247–423). A total of 44/96 (46%) patients died while on ECMO; 7 (63%) in the no antibiotic group, 5 (21%) in the single antibiotic group, and 32 (52%) in the combination antibiotic regimen group (P = 0.01) as seen in [Table 1].
Table 1: Baseline characteristics of patients receiving extracorporeal membrane oxygenation by antibiotic prophylaxis regimen type

Click here to view

Thirty-seven patients (39%) acquired a sterile site bacterial infection, of which 17 (18%) acquired an MDRO. Although there was no statistical difference in the proportion of acquired infections or acquired MDRO infections between prophylaxis groups, patients exposed to combination antibiotic prophylaxis acquired the lowest proportion of new sterile site infections (34%) and the highest proportion of MDRO infections (20%) across exposure groups. While there was no statistical difference between groups, the majority of C. difficile infections occurred among patients on combination antibiotic regimens [Table 2]. Patients who were exposed to a single antibiotic had 0.48 times the risk of developing a sterile site infection compared to patients with no antibiotic exposure after controlling for age and Charlson score. Similarly, patients who were exposed to combination antibiotic regimens had 0.44 times the risk of developing a sterile site infection compared to patients who were not exposed to any antibiotics after controlling for age and Charlson score [10] [Table 3]. While patients placed on a single antibiotic or a combination antibiotic regimen were at decreased risk of developing a new sterile site infection compared to patients not exposed to antibiotics, the difference was not statistically significant [Table 4]. By 500 h on ECMO support, all exposure groups had approximately 60% probability of acquiring a new sterile site infection [Figure 1].
Figure 1: Kaplan–Meier plot of time to first sterile site bacterial infection by regimen category (1) no antibiotics (2), single antibiotic (3), combination antibiotic regimen

Click here to view
Table 2: Extracorporeal membrane oxygenation characteristics of patients by antibiotic prophylaxis regimen type

Click here to view
Table 3: Acquisition of sterile site bacterial infection by antibiotic regimen type

Click here to view
Table 4: Cox proportional hazard ratios for rate of sterile site bacterial infection while on extracorporeal membrane oxygenation

Click here to view

  Discussion Top

To better define the effect of antibiotic prophylaxis use during ECMO support, we investigated three levels of antibiotic prophylaxis with the associated acquisition of a new sterile site bacterial infection. We evaluated the proportion of the new infection acquisition, time to onset of infection acquisition, and acquisition of MDROs and C. difficile infection. Because of the complexities of clinically confirming ventilator-associated pneumonia in patients supported by ECMO, we focused only on acquisition a sterile site infection proven by culture. Even with this more stringent definition, 39% of our patients acquired a sterile site infection. Our findings compare to the high reported rate of ECMO-related nosocomial infections ranging from 7.5% to 46% in the adult population.[1],[2],[3]

We found that exposure to antibiotic prophylaxis regardless of the type of regimen (e.g., single versus combination) did not significantly alter the proportion of patients acquiring an infection. There was also no significant association between antibiotic prophylaxis strategy and acquisition of an MDRO infection; however, although not statistically significant, the combination group demonstrated a trend toward more C. difficile infections and a higher proportion of infections which were caused by MDRO.

A potential benefit of prophylaxis would be to delay the onset of new infections, and we did find that time to infection was slower in patients who received single or combination agent prophylaxis compared to those who did not receive antibiotics, after controlling for patient age. Another possibility is that the delay in time to acquisition could actually represent a delay in time to diagnosis, as the prophylactic antibiotics could clinically mask the onset of infection, which could potentially be of more harm than benefit. Despite the initial delay observed, the longer a patient was supported on ECMO, the risk of infection increased. By 500 h of support, we observed a 60% probability of infection regardless of the type of prophylaxis regimen used.

Our study represents one of the largest single-center adult ECMO studies completed. It addresses the important clinical question of the effect of prophylaxis antibiotics on acquired infections during ECMO support. The paucity of research on this subject remains a major limitation to form guidance for the field.[8] We recognize the limitations of the observational nature of the study, particularly the potential for confounding by indication. These are recognized complicated patients and as our results show the majority of patients receive initial empiric antibiotics. It is possible that the higher mortality rate seen in the combination antibiotic prophylaxis group is related to the fact that no one protocol existed, and that patients with greater acuity of illness likely received broader initial empiric antibiotic treatment and this greater acuity also attributed to a higher rate of acquisition of new infections. The severity of illness measures (e.g., acute physiology and chronic health evaluation II scores) was not collected and calculated for our patients, and we were unable to factor this into our analysis. It would be difficult even for a large prospective multicenter study to fully control for confounding by indication, as the initial use of empiric antibiotic treatment would be difficult to randomize in this population. As an observational cohort, we were unable to control the sample size. The small numbers of patients in the no antibiotic group and single-antibiotic group limited our power to detect a difference as large as what we observed to 70%. This study does not support the use of routine prophylaxis antibiotics during ECMO support but does indicate the need for a larger, multicenter study to determine whether antimicrobial prophylaxis may have a beneficial role in the prevention of infection for patients receiving ECMO and if so, which antibiotics are indicated, at what time point they should be initiated, and for how long. Without such a study, there is no evidence to support the routine use of prophylaxis antibiotics throughout ECMO support.

  Conclusions Top

We found no significant difference in the proportion of infections that occurred in patients on ECMO support who were treated with continuous antibiotic prophylaxis, (either single or combination regimens) compared to those who received no continuous antibiotic prophylaxis. Patients who received prophylactic antibiotics progressed to a new sterile site infection slower than patients who did not receive prophylactic antibiotics, but the rate was not significant. Those patients on combination prophylaxis antibiotic regimens had more C. difficile infections, higher mortality, and a higher proportion of infections with MDR organisms. Based on our finding, we endorse the use of preemptive culture surveillance, careful monitoring of infections, and use of appropriate antibiotics for infections that do occur and do not recommend the use of prophylactic antibiotics during ECMO support.

Financial support and sponsorship

KAS was partially supported by the Building Interdisciplinary Research Careers in Women's Health (BIRCWH) program award (K12 HD43489-15).

Conflicts of interest

There are no conflicts of interest.

  References Top

Steiner CK, Stewart DL, Bond SJ, Hornung CA, McKay VJ. Predictors of acquiring a nosocomial bloodstream infection on extracorporeal membrane oxygenation. J Pediatr Surg 2001;36:487-92.  Back to cited text no. 1
O'Neill JM, Schutze GE, Heulitt MJ, Simpson PM, Taylor BJ. Nosocomial infections during extracorporeal membrane oxygenation. Intensive Care Med 2001;27:1247-53.  Back to cited text no. 2
Bizzarro MJ, Conrad SA, Kaufman DA, Rycus P. Extracorporeal life support organization task force on infections, extracorporeal membrane oxygenation. Infections acquired during extracorporeal membrane oxygenation in neonates, children, and adults. Pediatr Crit Care Med 2011;12:277-81.  Back to cited text no. 3
Sun HY, Ko WJ, Tsai PR, Sun CC, Chang YY, Lee CW, et al. Infections occurring during extracorporeal membrane oxygenation use in adult patients. J Thorac Cardiovasc Surg 2010;140:1125-32.e2.  Back to cited text no. 4
Vogel AM, Lew DF, Kao LS, Lally KP. Defining risk for infectious complications on extracorporeal life support. J Pediatr Surg 2011;46:2260-4.  Back to cited text no. 5
Hsu MS, Chiu KM, Huang YT, Kao KL, Chu SH, Liao CH. Risk factors for nosocomial infection during extracorporeal membrane oxygenation. J Hosp Infect 2009;73:210-6.  Back to cited text no. 6
Nagappala K, Gonzales R, Lelli J, Langenburg S. Pediartric extracorporeal membrane oxygenation: Is antibiotic prophylaxis necessary? Crit Care Med 2010;38:U195.  Back to cited text no. 7
O'Horo JC, Cawcutt KA, De Moraes AG, Sampathkumar P, Schears GJ. The evidence base for prophylactic antibiotics in patients receiving extracorporeal membrane oxygenation. ASAIO J 2016;62:6-10.  Back to cited text no. 8
Kao LS, Fleming GM, Escamilla RJ, Lew DF, Lally KP. Antimicrobial prophylaxis and infection surveillance in extracorporeal membrane oxygenation patients: A multi-institutional survey of practice patterns. ASAIO J 2011;57:231-8.  Back to cited text no. 9
Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 1987;40:373-83.  Back to cited text no. 10


  [Figure 1]

  [Table 1], [Table 2], [Table 3], [Table 4]

This article has been cited by
1 Antimicrobial Prophylaxis in Extracorporeal Membrane Oxygenation: Is the Debate Still Open?
Giorgia Montrucchio, Marcello Zagagnoni, Luca Brazzi
Annals of the American Thoracic Society. 2022; 19(5): 870
[Pubmed] | [DOI]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Subjects and Methods
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded99    
    Comments [Add]    
    Cited by others 1    

Recommend this journal