Prophylactic Ureteric Stents in Renal Transplant Recipients

A Multicenter Randomized Controlled Trial of Early Versus Late Removal

P. Patel; I. Rebollo-Mesa; E. Ryan; M. D. Sinha; S. D. Marks; N. Banga; I. C. Macdougall; M. C. Webb; G. Koffman; J. Olsburgh


American Journal of Transplantation. 2017;17(8):2129-2138. 

In This Article


The results of our trial performed in a routine care setting showed that early ureteric stent removal at 5 days is feasible and reduces the incidence of stent complications compared with cystoscopic stent removal at 6 weeks. This finding was strongly statistically significant when analyzed PP and remained significant when crossovers were removed from the analysis as potential modifiers. Analysis by ITT showed a 12% (p = 0.055; odds ratio [OR] 2.11, 95% CI 1.04–4.29) reduction in overall stent complications. The power of the study was sufficient to show a clinically useful benefit of early stent removal, with data on primary and secondary outcomes collected from 97% of the cohort. The benefit for the primary outcome with early stent removal was mirrored in the pediatric subgroup, although the study was not powered to show significance in this group.

Our cohort population was diverse but excluded patients with urinary diversion because of different requirements for transplant stent removal. Postrandomization exclusion allowed for problems encountered during organ retrieval, benching, and transplantation. Of the 459 patients screened, 138 (30%) were excluded for ineligibility. The main reason for ineligibility was abnormality of either recipient or donor urinary tract anatomy (35.8%). The study was not blinded because we were unable to conceal the intervention from surgeons or patients; the procedure for stent removal is quite different between groups. A 3-mo end point for stent complications was chosen because previous studies suggest UTIs are most frequent during this period, when immunosuppressant use is maximal. The end point for monitoring MUCs was 6 mo because stenosis can present later in the posttransplantation period.

UTIs were the predominant complication reported and the only stent-related complication reported in the early removal group. The timing of UTI episodes was spread across the 3-mo follow-up period. However, the majority of UTIs in the late group occurred coincidentally with the median duration of stent placement in this group. A prior history of recurrent UTIs was not a predictive variable for posttransplant UTIs or stent complications.

Incidence of stent symptoms was low. Symptoms consisted of discomfort, storage symptoms, or mild hematuria and affected only the late group (8.7%). This finding is consistent with other transplant studies and is much lower than studies in native ureters (80%).[3,12,13] Despite studies reporting lower incidence of urinary symptoms or bother with stents in transplant patients, no previous studies used a validated questionnaire to assess QoL in this population. The Functional Assessment of Chronic Illness Therapy symptom-targeted questionnaire FAIT-U was selected because it focuses on urinary symptoms, and the EQ-5D was used as general health status assessment. At week 1, there was no difference in these measures between groups, perhaps because of the significant morbidity associated with transplantation in both groups at this point. Interestingly, although both groups showed equal general improvement in health status at week 6 (evaluated by EQ-5D), symptom assessment evaluated by FAIT-U was significantly worse (p = 0.01) in the late group, which had stents in situ until this time. There appeared to be significant patient morbidity related to the length of time that the stent was indwelling.

Prophylactic stenting to prevent MUCs in renal transplantation is recommended by current evidence.[3] The incidence of MUCs in our study (4 of 206, 1.94%) is in keeping with the literature. The low rate of MUCs in our late-removal group (1 of 26, 0.8%) emulates the benefit shown in the Cochrane review.[3] The incidence of MUCs in our early group (3 of 80, 3.75%) remains better than that in an unstented population (median MUC rate of 7%) and suggests that, despite a short stent period, a lower MUC rate is still achievable.

This study did not assess MUCs as the primary outcome because, based on our current incidence of MUCs, it was estimated that 952 patients would be required for 80% power and a 5% two-sided type I error rate, which we could not achieve. Three patients in the early group had transient obstruction due to edema or clot, which we classified as an intermediate urological complication (IUC). These IUCs were seen within 4 days of early stent removal, diagnosed on ultrasound with hydronephrosis, and managed with a temporary stent or nephrostomy. Patients with IUCs may have benefited from a longer initial period of stenting, highlighting a limitation of this study, which lacks comparison at a range of time intervals between 5 days and 6 weeks.

Although not statistically significant, there appeared to be a relationship between surgeon experience performing the ureteric anastomosis and the incidence of urological complications. Of the seven urological complications (four MUCs and three IUCs), six occurred where a junior surgeon performed the ureteric anastomosis (OR 0.506, 95% CI 0.094–2.430); three of the four MUCs occurred where the ureteric anastomosis was performed by a junior surgeon (OR 0.231, 95% CI 0.011–1.866). The recent Cochrane review by Wilson et al also commented that surgeon experience influenced surgical outcome, with the incidence of MUCs lower in studies in which a single experienced surgeon was involved (median 6.3% vs. 10.3%).[3]

The technique for stent placement in the early group had a learning curve. This is represented by the crossovers from the early to the late group (n = 21), which were mainly due to technical difficulties in attaching the stent to the catheter (n = 15); other reasons were intraoperative nonurological complications during surgery (n = 5) and trial intervention missed by the surgical team (n = 1). The technical difficulties encountered were mainly in locating the catheter through the cystotomy to enable attachment to the stent (n = 11). Adjustments with catheter placement and the volume placed in the catheter balloon helped overcome these difficulties. We did not assess whether the length or size of the cystotomy was increased to overcome difficulties in technique, which could potentially increase risk of MUCs. The strings on the stent need to be placed through the distal drainage eye of the catheter to prevent occlusion of the catheter balloon, rather than around the distal catheter, which occurred on one occasion. Two of the urethral catheter balloons in the pediatric population were fractured during the process of locating it through the cystotomy. This occurred in small 8F urethral catheters and perhaps suggests a limitation for the technique of catheters of ≥10F size.

There were five complications with stent removal in the early group. The urethral catheter and stent fell out early (n = 2), which may have been a result of inadequate volume placement in the catheter balloon, inadvertent damage to the catheter balloon, or a fault with the catheter. The string snapped during removal, requiring cystoscopic removal (n = 1); occlusion of the catheter balloon with the string (n = 1) required percutaneous deflation; and difficulty with catheter removal required cystoscopic removal (n = 1).

A decision to delay stent removal in the early group was made in anuric patients with delayed graft function (DGF) and where a transplant biopsy was needed, to avoid potential complications of bleeding and clot obstruction (n = 2). In recipients of kidneys from standard criteria donors after brainstem death, where DGF is not common (20%), it seems reasonable to use the early stent-removal technique with the current exclusion criteria. However, because it can still be hard to predict DGF, it seems reasonable to delay early stent removal in this group until after a transplant biopsy may be required, at around 1 week after transplant. Recipients of deceased donor kidneys that have a high risk of DGF, such as kidneys from extended criteria donors after brainstem death or from donor after cardiac death, may be advised not to have the stent attached to the catheter but would still benefit from earlier cystoscopic stent removal, perhaps within 4 weeks rather than 6 weeks after transplant.[8]