Objectives: Urinary tract infections are among the most common infections after liver transplant, especially soon after surgery. This study analyzed urinary tract infections or bacteriuria, their causative agents, and related risk factors in the early period after liver transplant in hospitalized adult transplant recipients in the main liver transplant referral center in Iran.
Materials and Methods: In this prospective study, 389 consecutive adult patients who underwent liver transplant at the Nemazee Teaching Hospital were enrolled between October 2014 and October 2015. Risk factors were compared for patients who developed urinary tract infections or bacteriuria (“infection group”; n = 63 [16.2%]) and patients without evidence of infection (“control group”; n = 211 [54.2%]). Patients with sites of infection other than the urinary tract were excluded. Antimicrobial susceptibility testing was performed using the Kirby-Bauer disk-diffusion method. Univariate and multivariate analyses compared variables between the 2 groups.
Results: Seventy-nine episodes of urinary tract infections or bacteriuria occurred in the infection group. Multiple logistic regression analysis showed that female sex, hospitalization 2 to 7 days before transplant, and frequency of abdominal exploration were 11.0, 5.9, and 3.0 times more common in the infection group than in the control group. The chance of infection rises 1.1 times with each one unit increase of body mass index. The most common infection causes were gram-negative bacteria (n = 50; 63.3%), predominantly Escherichia coli (n = 24; 30.4%); followed by gram-positive bacteria (n = 20; 25.3%), predominantly Enterococcus species (n = 14; 17.8%) that had a high incidence of vancomycin resistance (n = 10; 71.4%); and non–Candida albicans species isolates (n = 9; 11.4%).
Conclusions: Urinary tract infections are a common infection in hospitalized adult patients soon after liver transplant. Female sex, hospitalization shortly before transplant, more frequent abdominal exploration, and higher body mass index substantially increased the risk of developing such infections in this period.
Key words : Bacteriuria, Iran, Nosocomial infections, Solid-organ transplant, Vancomycin-resistant enterococci
Postoperative infection is one of the most common complications in liver transplant (LT) recipients.1 Infections, especially in the first month after transplant, remain grave complications after LT despite all measures to reduce their risk.2,3 Urinary tract infections (UTIs) and bacteriuria are among the most common infections experienced by solid-organ transplant recipients.4 However, a limited number of published studies have investigated the prevalence, risk factors, and causes of UTIs in LT recipients worldwide and especially in Iran. The aim of this study was to determine the risk factors and prevalence of UTIs or bacteriuria among hospitalized adult patients after LT at the main referral center for liver transplant in Iran, the cause of these episodes, and the patients’ susceptibilities to antibacterial agents during a 1-year period.
Materials and Methods
The study population comprised 389 consecutive adult patients who underwent LT at Nemazee Teaching Hospital, affiliated with the Shiraz University of Medical Sciences, Shiraz, Iran, between October 2014 and October 2015. Of these, 145 (37.3%) were women, and patients’ ages varied from 18 to 69 years (mean, 42 y).
For all patients, the immunosuppressive regimen consisted of tacrolimus, mycophenolate mofetil, and prednisolone. Ampicillin/sulbactam and ceftizoxime were given as postsurgical prophylaxis after LT. Antibiotics were discontinued after 72 hours in all patients, except in those with bilioenteric anastomosis, who continued this regimen for 5 days.5 All patients received fluconazole for fungal prophylaxis and trimethoprim/sulfamethoxazole for Pneumocystis jirovecii. Ganciclovir was added to this regimen if one or more CMV pp65 antigen-infected cell per 50 000 leukocytes were detected. During the hospital stay, attending physicians were allowed to change the antibiotic regimen or add other antibiotics based on clinical and paraclinical findings suggestive of bacterial infections.
Urine culture samples were routinely collected for microbiological examination every other day and when clinically indicated. Significant culture growth was defined as the isolation of a single bacterium in counts of 100 000 colony-forming units/mL or more in a single specimen obtained through a clean-catch voided urine sample or ≥ 2 urine cultures with repeated isolation of the same uropathogenic (gram-negative bacteria or Staphylococcus saprophyticus) with ≥ 100 colony-forming units /mL in nonvoided urine specimens.6 Antimicrobial susceptibility testing was performed using the Kirby-Bauer disk-diffusion method. The range of follow-up times spanned the immediate period of hospitalization after transplant until either the time of discharge from the hospital or the time of death at the hospital.
The patients were divided into 2 groups: patients with UTIs or bacteriuria (infection group) and patients without any evidence of infection (control group). Patients with sites of infection other than the urinary tract were excluded. We compared the risk factors associated with the development of UTIs in the 2 groups in the immediate period of hospitalization after transplant. Factors related to UTI during the pretransplant period included age, sex, body mass index (BMI), comorbidities, hospitalization 2 to 7 days before LT, and hospitalization in the 3 months to 1 week before LT. Laboratory data, including serum creatinine and serum urea nitrogen levels, were recorded. Factors related to UTI during the perioperative period included duration of transplant surgery, vancomycin-resistant enterococci colonization at the time of admission, and antimicrobial prophylaxis. Posttransplant factors related to UTI included length of stay in days in the intensive care unit (ICU), between transplant and discharge, frequency of abdominal exploration, the need for renal replacement therapy, the time of Foley catheter removal, and whether the outcome was death. The data recorded for each episode of infection were the date, isolated organism, and susceptibility of isolates to antibiotics. For patients who died (8 in the infection group and 5 in the control group), the main causes contributing to death were recorded.
Data were analyzed using the chi-squared test or the Fisher exact test for discontinuous variables, the t test for continuous variables, and SPSS software (SPSS: An IBM Company, version 16, IBM Corporation, Armonk, NY, USA). Multiple logistic regression analyses were performed to compute the independent risk factors. A P value < .05 was considered significant. This study was approved by the ethics committee of Shiraz University of Medical Sciences. All of the protocols conformed to the ethical guidelines of the 1975 Helsinki Declaration. Written informed consent was obtained from all patients.
Among 389 patients who underwent LT, 79 episodes of UTI or bacteriuria were detected in 63 patients (16.2%) during hospitalization in the early posttransplant period (infection group). Most patients (211 [54.2%]) showed no evidence of any infection (control group), and the remaining 115 patients (29.6%) had sites of infection other than the urinary tract and were excluded. Bacteriuria or UTI developed within 1 to 21 days after LT (mean ± standard deviation [SD], 3.87 ± 4.85 d). In 13 patients (20.6%) with UTI or bacteriuria, evidence of infection at other sites (eg, surgical site, intra-abdominal, bloodstream, and respiratory tract) was also seen, and in 50 patients (79.4%) in the infection group, only bacteriuria or UTI was observed.
The results of univariate analysis showed that some variables were significantly detected in the infection group, whereas other variables did not significantly differ between the 2 groups, as shown in Table 1. The misclassification rate in the present study was 13%. Using multiple logistic regression analysis, it was found that female sex, hospitalization 2 to 7 days before LT, and frequency of abdominal exploration increased the risk of infection about 11, 5.9, and 3 times, and rising each one unit BMI increases the risk 1.1 times.
Among the 79 microbial strains reported in the patients’ urine culture, the most common were gram-negative (n = 50; 63.3%), with a predominance of Escherichia coli (n = 24; 30.4%); Pseudomonas species (n = 10; 12.6%); Klebsiella species (n = 7; 8.9%), Enterobacter species (n = 4; 5.1%); Acinetobacter species (n = 4; 5.1%); and Citrobacter species (n = 1; 1.3%) (Figure 1). Gram-positive bacteria comprised the most common strains (n = 20; 25.3%), of which Enterococcus species were the most abundant (n = 14; 17.8%) and in which there was a high incidence of vancomycin-resistant enterococci (n = 10; 71.4%); these were followed by Staphylococcus species (n = 5; 6.4%) and Streptococcus species (n = 1; 1.3%). All fungal strains (n = 9; 11.4%) were non-Candida albicans species isolates (Figure 1). The reported sensitivity rates of gram-negative isolates were 37.3% to gentamicin, 11.8% to ciprofloxacin, 2.0% to ceftriaxone, and 2.0% to trimethoprim/sulfamethoxazole. Susceptibility rates of gram-positive isolates were 36.8% to vancomycin, 15.8% to amikacin, and 10.5% to ampicillin.
Urinary tract infection is a common complication after LT, and two-thirds of UTI episodes in LT recipients occur during the first month after transplant.7 In this study, 63 hospitalized patients (16.2%) had positive urine cultures in the immediate period of hospitalization after transplant. We found that UTI developed within 1 to 21 days after LT. In other studies, the incidence of UTI during the first 4 weeks after LT is reportedly between 7.1% and 16.7%.2,8 In nonkidney transplant recipients, bacterial UTI is usually an early complication during 4 weeks after transplant and associated mainly with malnutrition and factors related to postoperative patient care, such as urinary catheterization and stays in the ICU.9 In the present study, multivariate analysis showed that female sex, hospitalization 2 to 7 days before LT, frequency of abdominal exploration, and high body mass index increased the risk of infection. Similar factors were reported in other studies, although at different rates.
Other possible factors associated with increased risk of UTI in LT recipients include age, mycophenolate mofetil or antithymocyte globulin use, need for immediate posttransplant dialysis, diabetes mellitus, number of episodes of acute graft rejection, length of hospitalization, duration of Foley catheterization, number of episodes of acute rejection, and increase in immunosuppression.3 Based on this univariate analysis, factors such as age, Model for End-Stage Liver Disease score, ICU and hospital stay, duration of Foley catheterization, admission in the previous 3 months before transplant, continuous renal replacement therapy, and death during the immediate period of hospitalization after transplant were found to be significant in the infection group. However, in multivariate analysis, these variables did not show any prognostic power (Table 1). The leading causes of death in the infection group (n = 8) were noninfectious ones, such as nonfunction of the newly transplanted liver or infections other than in the urinary tract.
In agreement with other studies, we found that gram-negative bacteria were the most common causative agents of post-LT infection (63.3%; E. coli 30.4%), followed by gram-positive bacteria (comprising 25.3%) and fungi (11.4%).2,10,11 Higher rates of yeast like fungi (21%) have been reported in other studies of LT patients.10 Using fluconazole as prophylaxis in all patients may have led to selection pressure and an increased rate of isolation of non-Candida albicans species, which can be fluconazole resistant.12 In contrast to most studies, Hjortrup and associates reported gram-positive bacteria as the predominant causative agents (67%) in post-LT infection.13 In our study, the most common gram-positive bacteria were Enterococcus species, of which vancomycin-resistant enterococci were the majority (74%), consistent with other reports from Iran.14 Also, the same pattern of antimicrobial susceptibility (with a high rate of resistance among gram-negative bacteria) was reported in our previous study and others from Iran among patients who developed UTIs after kidney transplant and other hospitalized patients with UTIs, which were mainly nosocomial.4,15
Urinary tract infection is a common infection in the early period after LT. Furthermore, the high antimicrobial resistance rate among isolated bacteria in these patients can be alarming. Female sex, higher frequency of abdominal exploration, history of hospitalization 2 to 7 days before LT, and high BMI can substantially increase the risk for developing UTI (11.0, 5.9,3.0, and 1.1 times, respectively) within the immediate period of hospitalization after transplant.
Volume : 15
Issue : 1
Pages : 190 - 193
DOI : 10.6002/ect.mesot2016.P68
From the 1Alborzi Clinical Microbiology Research Center, Shiraz, Iran; and the
2Shiraz Organ Transplant Center, Nemazee Teaching Hospital, Shiraz University of
Medical Sciences, Shiraz, Iran
Acknowledgements: Our thanks go to H Khajehei for copyediting of the manuscript. The authors have no conflicts of interest to declare.
Corresponding author: Zahra Jafarpour, Alborzi Clinical Microbiology Research Center, Shiraz, Iran
Phone: +98 713 647 4304
Table 1. Risk Factors Associated with Urinary Tract Infections and Bacteriuria in the Early Period After Liver Transplant
Figure 1. Bacteria Isolated From Urine in Hospitalized Patients in the Early Post Liver Transplantation