To Treat or Not to Treat: Bacteriuria in Kidney Transplant Recipients

#NephJC Chat

Tuesday February 9 9 pm Eastern

Wednesday January 10 9 pm IST

Wednesday January 10 9 pm GMT

Antibiotics versus no therapy in kidney transplant recipients with asymptomatic bacteriuria (BiRT): a pragmatic, multicentre, randomized, controlled trial

Julien Coussement, Nassim Kamar, Marie Matignon, Laurent Weekers, Anne Scemla, Magali Giral, Judith Racapé, Éric Alamartine, Laurent Mesnard, Mireille Kianda, Lidia Ghisdal, Concetta Catalano, Emine N Broeders, Olivier Denis, Karl M Wissing, Marc Hazzan, Daniel Abramowicz, Bacteriuria in Renal Transplantation (BiRT) study group

PMID: 32919076

Introduction

Urinary tract infections (UTI) are the most frequent infections (Vidal et al, Transpl Inf Dis 2012) after kidney transplantation (KT). The incidence of UTI varies from 10% to 98% (Castaneda et al Transpl Proc 2013). Fortunately, over the past few decades, the incidence of post-KT UTI has reduced due to improvements in surgical techniques, early removal of urethral catheters, and antibiotic prophylaxis therapy. However, it continues to be a matter of concern. A simple cystitis can quickly become a graft pyelonephritis and result in graft dysfunction or worse.  

Asymptomatic bacteriuria (ASB) is estimated to occur in up to 51% of kidney transplant recipients (KTR), although the estimates depend on the definition of ASB, follow-up period and the screening strategies used (eg Fiorante Kidney Int 2010; Green Eur J Clin Micr Inf 2013). It is noteworthy that most of these studies are retrospective. The impact of screening and treatment of ASB has been explicitly studied in urological procedures and for pregnant women (recommendation strength grade A-I in the IDSA guidelines, Nicolle et al 2019), sadly the same cannot be said for KTR (grade of evidence CIII).

In the paucity of firm evidence, the management of ASB in KTR largely varies between transplant centers. A recent European survey (Coussement et al 2018) highlighted that more than 70% of transplant physicians always screen for ASB during post-kidney transplant surveillance, and also have a tendency to treat it, often with quinolones or beta-lactam antibiotics. One can make a plausible case for this.  The denervation of the renal allograft and usage of immunosuppressive agents might mask the clinical features associated with UTI and confer higher susceptibility of the KTR to develop UTI. It is also quite plausible  that there are higher chances of ascending bacterial infection resulting in pyelonephritis in the realm of immunosuppression especially in the setting of prior manipulation of the urinary tract. This fear of developing severe infections, graft dysfunction, and higher mortality has fueled the practice of administering antibiotics in ASB.

Nevertheless, on account of the threat of emergence of multidrug-resistant (MDR) bacteria, increased financial stress, and higher incidence of Clostridium difficile infection, the utility of this strategy has been questioned in recent trials.  A Cochrane Review (Coussement et al 2018), reported that there was insufficient evidence to support the routine use of antibiotics for ASB treatment in  KTR.

Some small trials have already been conducted. In a noninferiority, multicenter RCT Sabé et al 2019,  found no significant differences in the incidence of acute graft pyelonephritis in the first year post-KT between the antibiotic treatment receiving group and no treatment receiving group. There was no apparent benefit of systematic screening beyond the second-month post-transplantation. Besides, concerns were raised about the increased antibiotic resistance rates. Similarly, in a quasi-randomized prospective study by Moradi et al 2005 comparing  43 KTR with treated ASB and 45 that did not receive therapy, no differences were found in the incidence of symptomatic UTI during 1-yr of follow-up. Also, ASB did not affect kidney function in the short term. El Amari et al 2011, demonstrated that restricting antibiotic treatments for ASB in the absence of pyuria, occurring later than 1 month posttransplantation, might be safe in KTR. Origen et al 2016, compared antibiotic treatment with no treatment for ASB in 112 KTR beyond the second month after transplantation, no differences were seen in the incidence of acute graft pyelonephritis or lower urinary tract infections.

Despite the negative results in the above-mentioned trials, there has been no consensus on whether ASB in KTR should receive antibiotic therapy. And even if one decided to treat, what is the ideal time posttransplant period for the same? 
This brings us to Antibiotics versus no therapy in kidney transplant recipients with asymptomatic bacteriuria (BiRT): a pragmatic, multicentre, randomized, controlled trial to determine whether antibiotics are beneficial in KTRs who have ASB beyond the first 2 months post-transplant. 

The Study

Methods

A prospective, randomized, parallel-group, multicentre, open-label, superiority trial done in France (seven sites) and Belgium (six sites) (NCT01871753)

Study Population

Inclusion criteria

  • KTR  (≥18 years of age) were eligible if they had ASB, and were ≥2 months post-transplantation. Patients developing ASB while on cotrimoxazole prophylaxis were considered eligible.

ASB was defined as the isolation of a single bacterial species at ≥105 CFU/mL in a urine specimen from a patient without symptoms of UTI. A second positive culture was not considered necessary before enrolment however, specimens were sent whenever possible.

Exclusion criteria

  • Non-functioning native bladder (e.g., bladder dysfunction requiring intermittent self-catheterization, orthotopic ileal neobladder). 

  • Recurrent acute graft pyelonephritis (i.e., ≥ 2 episodes in the last year)

  • Use of antibiotics at the time of the asymptomatic bacteriuria episode (except for prevention of Pneumocystis jirovecii pneumonia)

  • Neutropaenia (≤ 500 neutrophils/mm3)

  • Major increase in immunosuppression (Solumedrol bolus and/or use of thymoglobulin) or any other treatment of an acute graft rejection in the two months prior to recruitment

  • Presence of an indwelling urinary device (e.g., urethral catheter, ureteral catheter,nephrostomy) 

  • Combined transplantation (i.e., liver-kidney, lung-kidney, heart-kidney)

  • End-Stage Renal Disease (ESRD) requiring dialysis

  • Urinary tract surgery in the two months prior to recruitment, and/or planned in the subsequent 2 weeks

  • Kidney transplant recipients who could not return for regular follow-up 

  • Pregnant women or women wishing to become pregnant during the course of the study

Randomization and masking

Patients were centrally assigned (1:1) to either antibiotics or no therapy using an internet-based randomization service. Randomization was stratified by sex and age (<50 versus ≥50 years). Investigators were masked to the randomization sequence whereas participants and clinicians were not blinded to allocation.

Procedure 

Antibiotics in the antibiotic group were selected by the treating physician and were administered for 10 days. The Control group did not receive any antibiotic treatment for ASB. Participants were followed for 12 months post-randomization, with study visits scheduled at 1, 2, 4, 6, 8, 10, and 12 months. For each follow-up visit, a urine culture and a pre-established questionnaire, also included history taking, temperature measurement, and blood analysis was performed. If ASB occurred again at a study follow-up visit, antibiotics were re-administered in the antibiotic group but not in the control group.

Primary Outcome: 

  • Incidence of symptomatic UTI during the 1-year follow-up, defined as the association of 

(a) one or more symptoms/signs from a prespecified list (of cystitis, pyelonephritis, prostatitis, or bloodstream infection due to UTI 

(b) a positive urine culture (i.e. isolation of a bacterial organism at ≥104 CFU/mL). 

Secondary outcomes:

  • The proportions of participants with pyelonephritis (i.e., association of (1) symptom(s) of pyelonephritis [fever and/or chills and/or kidney pain], and (2) a positive urine culture [i.e., isolation of a bacterial organism at ≥104 CFU/mL])

  • UTI requiring hospital admission

  • Bloodstream infection due to UTI

  • Biopsy-proven graft rejection, or graft loss

  • All-cause mortality during the one-year study period. 

  • The total number of symptomatic UTI episodes/participant over the one-year study period

  • Change in serum creatinine from baseline to end-of-study. 

Statistics simplified

The 1-year cumulative incidence of symptomatic UTI was estimated to be 20% among untreated patients and a reduction in the incidence of symptomatic UTI from 20% to 6%  represented the minimal clinically important difference. A sample size of 198 participants was needed to have an 80% chance of detecting a reduction in the incidence of symptomatic UTI from 20% to 6% as significant at the 5% level, considering a potential 10% loss to follow-up. The primary analysis was by intention-to-treat. For the primary outcome, Kaplan–Meier survival curves were used  to estimate the cumulative incidence of symptomatic UTI. 

To investigate the consistency of the study conclusions among different subpopulations, an analysis was undertaken in three pre-specified subgroups: (a) time between transplantation and study inclusion <6 versus ≥6 months; (b) baseline estimated glomerular filtration rate <40 versus ≥40 mL/min/1.73 m2; and (c) resistant organism at baseline versus other organisms. Change in serum creatinine throughout the follow-up period was compared using a two-way repeated-measures ANOVA.

Funding

This work was supported by three research grants: Fonds Erasme pour la Recherche Medicale, Fonds David et Alive Van Buuren, and Fonds Carine Vyghen (all to JC). The funders of the study had no role in the study design, in the collection, analysis and interpretation of data, or in the report writing. 

Results

199  kidney transplant recipients with ASB were randomly assigned to receive antibiotics (100 participants) or no therapy (99 participants). As can be seen, there were some protocol violations and some patients in the antibiotic group did not receive them, and some in the control group did receive antibiotics. 


Table 1 depicts the baseline demographic characteristics of patients. 

 At study inclusion, 27.1% of the patients (54/199) were in the first post-transplant year and Escherichia coli was the most common organism responsible for the inclusion episode of ASB (see below). Fluoroquinolones were the most frequently prescribed agents at baseline. 

Outcomes 

On an intention-to-treat basis, the risk of symptomatic UTI did not differ significantly between participants in the antibiotic group and those in the no- therapy group (27/100, 27% versus 31/99, 31%, univariate Cox model: HR 0.83, 95%CI: 0.50-1.40, p =0.49). (See Figure 1).

Figure 2 from Coussement et al

The characteristics of these 58 symptomatic UTI episodes are summarized in table 2.

Table 2 from Coussement et al

Antibiotics did not significantly reduce the cumulative incidence of symptomatic UTI in any of the pre-specified subgroups (Table 3 and figure below)

Table 3 from Coussement et al

Supplementary figure from Coussement et al

Secondary clinical outcome

Antibiotics had no significant impact on any secondary clinical outcome ( See Table 3). Of interest,  there was no significant difference in pyelonephritis between study groups (17%) in the antibiotic group versus (16%) in the no-therapy group, RR 1.05, 95%CI 0.56-1.96, p 0.87. Similarly, ASB treatment did not significantly result in improvement of kidney function.

One month after randomization, a urine specimen was sent for culture.  Among them, the prevalence of ASB was significantly lower in the antibiotic group than in the no-therapy group (29%, versus 66%, p < 0.001). The participants in the antibiotic group also had a significantly lower total number of ASB episodes during the complete follow-up period. 

Importantly, the median number of days receiving antibiotics for any cause was five times higher in the antibiotic group (p < 0.001).

Impact of ASB treatment on antibiotic resistance

The study determined the impact of antibiotics on antibiotic resistance, focusing on 155/199 participants (77.9%) who had at least one further episode of bacteriuria during the follow-up. Compared with the patient's baseline episode of ASB, the second episode of bacteriuria was more frequently caused by bacteria resistant to clinically relevant antibiotics (i.e., ciprofloxacin, cotrimoxazole, or third-generation cephalosporin) in the antibiotic group than in the no-therapy group (18% versus 4%, p 0.003) 

Safety  

Overall, 93 serious adverse events (SAEs) were reported 28%  in the antibiotic group versus 23% in the no-therapy group (p 0.44). See table 3 above for more.


Discussion 

Even though the antibiotics for treatment of ASB resulted in fewer subsequent cases of bacteriuria, the study failed to show any clinical benefit including a reduction in symptomatic UTI over the 1-year study period. It is worth noting that there was a  five time higher consumption of antibiotics among participants in the antibiotic group. Moreover, this was significantly associated with the emergence of resistant organisms in the urine.

Another striking finding in the study was that treating ASB did not reduce the incidence of pyelonephritis or improve any of the other graft-related outcomes such as kidney function, graft rejection, and graft loss. Often in the kidney transplant setting, there is a lower threshold to treat ASB mainly because of this fear of pyelonephritis and graft dysfunction, which seems to be unfounded. 

Strengths of the study 

  • Randomized design

  • Large population size

  • High level of adherence with the study protocol 

  • Demonstration of clear microbiological effects of antibiotics 

  • Rigorous examination of benefits and harms of antibiotics 

  • To assess the external validity of the trial findings, the study group also performed an observational co-study in some of the trial sites that showed that the characteristics of the BiRT study participants resembled those of kidney transplant recipients who have ASB in usual care in terms of sex, age, kidney function, and time post-transplant.

Limitations of the study 

  • Participants and physicians were not blinded to treatment allocation (may increase in biasing in determining the symptoms of UTI which is partly subjective). However, the primary outcomes were adjudicated with the help of assessors blinded to treatment allocation.

  • Mild to moderate effects of antibiotics on the risk of symptomatic UTI was not ruled out as this study was powered to detect a large effect.

  • Only 13% of the participants were included in the first 6 months after transplantation hence findings cannot be extrapolated to the initial months of the posttransplant period.

  • The regimen of  10-day antibiotic administration to treat ASB was a bit long. While this duration was selected to be of sufficient length to be potentially effective, this choice may also have impacted the estimates for the outcomes of antimicrobial resistance and antibiotic consumption. 

This trial adds to the previous trials discussed and mentioned above. In a subsequent paper (Coussement et al 2020) the authors also perform a meta analysis, with consistent results of no difference between treating ASB compared to watchful expectancy. 

Figure from Coussement et al NDT 2020

Conclusion 

The treatment of ASB with antibiotics was associated with fewer subsequent cases of bacteriuria, however, this microbiological effect did not translate into significant clinical benefit, including symptomatic UTI, pyelonephritis, or improvement in kidney function. Furthermore, this strategy increases antibiotic use and promotes the emergence of resistant organisms. As the White Rabbit says, ‘Don’t just do something, stand there!’ After all, it is difficult to make the asymptomatic patient feel better.

Summary prepared by Priti Meena
Nephrologist,

New Delhi, India