Kidney transplantation is the treatment of choice for end-stage kidney disease (ESKD). Transplantation improves both quantity and quality of life for recipients. Currently, and for the foreseeable future, kidney graft demand greatly exceeds supply. A large body of research is developing into strategies to expand the donor pool and increase the supply of organs. Concurrently other groups are looking at ways to extend the life of kidney grafts by both reducing the burden of chronic allograft nephropathy and reducing the number of recipients dying with functioning grafts. There remain a small percentage of grafts lost in the early months post-transplantation to acute rejection and technical complications including vascular thrombosis. Major urological complications (MUCs) mostly originate from the vesicoureteric anastomosis, present early after transplantation (within three months) (Kumar 2004) and contribute to patient morbidity, graft loss and mortality (Rigg 1994; Thomalla 1990). The first kidney transplants were anastomosed using a transvesical approach (Merrill 1956). This technique, now more commonly known by the eponym Leadbetter-Politano (L-P)(Politano 1958), has been generally superceded by the extravesical ureteroneocystostomy (Lich-Gregoir) (L-G)(Konnak 1972) which is less technically demanding. Despite this there remains a significant morbidity with a recent review of case control studies reporting mean incidences of between 3% and 5% (Mangus 2004).
vesicoureteric complications present either as urine leaks or collecting system obstruction. In the absence of technical complications, ureteric ischaemia (often related to the retrieval procedure) is thought to be chiefly responsible for the early ureteric complications post-transplantation (Karam 2004). Both ureteric leak and obstruction have been successfully treated with "double-J" stent insertion, prompting surgeons to contemplate the use of prophylactic stents (Insall 1995).
Stents have been successfully used in general urological practice for a number of years. The double-J stent consists of a straight tube with anchoring "j" loops at either end. In conventional urological practice the stent is straightened and inserted over a guidewire either via a cystoscope and/or under radiological guidance. Removal of the guidewire causes both ends to curl into their natural anchoring conformation. Placement of a stent during transplantation is a relatively simple procedure and requires no additional instrumentation or imaging. However, a further procedure to remove the stent is always required, may be inadvertently delayed and incurs further healthcare costs.
Conventional native ureteric repairs over stents are widely accepted to have a better outcome (Turner 1982). In addition they have been successfully used in pyeloplasty, ureterovesical reconstruction and in the management of stone disease (Baum 1982; Finney 1978). The therapeutic benefits of stents in transplantation are still disputed but may include simplifying the creation of a watertight ureteric to bladder anastomosis and reduction of anatomical kinking (French 2001; Kumar 2004). The most significant theoretical complication is an increase in the number and severity of urinary tract infections (UTI). Other possible complications include persistent haematuria and bladder discomfort, stent migration, breakage, encrustation and complications during removal.
A number of centres have adopted a policy of prophylactic stent insertion with endoscopic removal at a designated time post-transplantation in an effort to reduce the MUCs rate (Lin 1993). However, it has been suggested that a tension-free anastomosis with an intact blood supply is the only effective strategy to avoid both early and late ureteric complications, and the benefits of prophylactic stenting are outweighed by their possible complications. As a consequence many units continue to selectively stent only difficult anastomoses or in circumstances where the vesicoureteric viability may be additionally compromised (Thomalla 1990).
This review aimed to examine the benefits and harms of routine ureteric stenting to prevent urological complications in kidney transplants recipients.
- Reduction in incidence of MUCs
- Increase in the incidence of UTI and haematuria
- Idiosyncratic complications (migration, malposition, irritation, encrustation)
- Complications from stent removal
Criteria for considering studies for this review
Types of studies
All randomised controlled trials (RCTs) and quasi-RCTs looking at the use of double-J stents to prevent urological complications.
Types of participants
We included all studies designed to examine the impact of the use of stents in recipients of kidney transplants. Specifically we aimed to include studies regardless of the type of graft (live donor, cadaveric), technique of ureteric implantation (extravesical, transvesical) and patient group (e.g. paediatric, elderly, multiple organ and previous transplant recipients).
Studies were assessed against a number of criteria including the demographics of organ donor, recipient, surgical techniques and medical management to allow comparison and stratification for known MUCs risk factors.
Specific exclusion criteria stipulated by our protocol included studies detailing the analysis of horseshoe kidneys, patients with abnormal bladder function and urinary diversion.
Types of interventions
For a trial to be included one randomised group of graft recipients must have had the vesicoureteric anastomosis stented at the time of implantation with a suitable prosthesis. Our protocol stipulated that the stent must have remained in for at least 14 days to be considered an adequate "treatment period".
Types of outcome measures
- Urological complications related to the transplant ureter (urine leak/ obstruction) within three months of transplant.
- Patient survival.
- Graft survival.
- Stent complications including pain, haematuria, UTI, migration, irritative symptoms, stone formation, secondary obstruction from crusting.
- Complications from stent removal.
Search methods for identification of studies
We searched the Cochrane Renal Group's Specialised Register (up to 8 January 2013) through contact with the Trials' Search Co-ordinator using search terms relevant to this review.
The Cochrane Renal Group’s Specialised Register contains studies identified from:
- Monthly searches of the Cochrane Central Register of Controlled Trials CENTRAL;
- Weekly searches of MEDLINE OVID SP;
- Handsearching of renal-related journals & the proceedings of major renal conferences;
- Searching of the current year of EMBASE OVID SP;
- Weekly current awareness alerts for selected renal-journals;
- Searches of the International Clinical Trials Register (ICTRP) Search Portal & ClinicalTrials.gov
Studies contained in the Specialised register are identified through search strategies for CENTRAL, MEDLINE, EMBASE based on the scope of the Cochrane Renal Group. Details of these strategies as well as a list of handsearched journals, conference proceedings and current awareness alerts are available in the 'Specialised Register' section of information about the Cochrane Renal Group.
See Appendix 1 for search terms used in strategies for this review.
Searching other resources
- Reference lists of nephrology textbooks, review articles and relevant studies.
- Letters seeking information about unpublished or incomplete trials to investigators known to be involved in previous studies.
Data collection and analysis
Included and excluded studies
The review was carried out initially by three authors (CHW, AAB, DMM, DAR) and a fourth (DAR) was consulted for a specific urological transplant opinion. The search strategy described was used to obtain titles and abstracts of studies that may have been relevant to the review. The titles and abstracts were screened independently by CHW and AAB who discarded studies that were not applicable. Both primary authors independently assessed the retrieved abstracts and determined which studies satisfied the inclusion criteria. Data extraction was carried out by the same authors independently using standard data extraction forms. Further information required from the original author was requested by written correspondence and relevant information obtained in this manner has been included in the review. The inclusion of studies was then discussed and agreed with the third and fourth authors (DMM and DAR).
The quality of studies to be included was assessed independently by CHW and AAB without blinding to authorship or journal using the checklist developed for the Cochrane Renal Group. Discrepancies were resolved by discussion with DMM. The quality items assessed were allocation concealment, blinding, intention-to-treat analysis and completeness to follow-up.
- Adequate (A): Randomisation method described that would not allow investigator / participant to know or influence intervention group before eligible participant entered in the study.
- Unclear (B): Randomisation stated but no information on method used is available.
- Inadequate (C): Method of randomisation used such as alternate medical record numbers or unsealed envelopes; any information in the study that indicated that investigators or participants could influence intervention group
- Blinding of participants: Yes/no/not stated.
- Blinding of outcome assessor: Yes/no/not stated.
- Blinding of data analysis: Yes/no/not stated.
- Yes: Specifically reported by authors that intention-to-treat analysis was undertaken and this was confirmed on study assessment.
- No: Not reported and lack of intention-to-treat analysis confirmed on study assessment. (Patients who were randomised were not included in the analysis because they did not receive the study intervention, they withdrew from the study or were not included because of protocol violation).
- Not stated.
Completeness of follow-up
Number of participants loss to follow-up.
Outcomes were expressed as risk ratio (RR) with 95% confidence intervals (CI) where appropriate. Absolute risk reductions are given as numbers needed to treat (NNT), a figure that quantifies the number of patients that must be stented to prevent one MUCs. Data was pooled using the random effects model but the fixed effects model was also analysed to ensure robustness of the model chosen, susceptibility to outliers and during sub group analysis. Heterogeneity was analysed using a Chi squared test on N-1 degrees of freedom, with a P value of 0.05 used for statistical significance and with the I² test (Higgins 2003). I² values of 25%, 50% and 75% correspond to low, medium and high levels of heterogeneity. When statistical heterogeneity was present subgroup analysis was to be used to explore possible sources. Adverse effects were collated and reported in tabular form.
Description of studies
Seven RCTs and quasi-RCTs (1154 patients) were identified (Bassiri 1995; Benoit 1996; Dominguez 2000; Guleria 1998; Kumar 1998; Osman 2004; Pleass 1995). All were published as full articles in English language journals. No dual publications or non-English language studies were identified. Three of the seven authors (Dominguez 2000; Osman 2004; Pleass 1995) answered enquiries about study design and results; whilst their assistance was invaluable often original data was unavailable and the basis for their response was personal memory. We did not pursue statistical attempts to identify publication bias.
Risk of bias in included studies
As expected none of the studies attempted to blind patient, investigator or assessor to treatment allocation. Overall the studies were of low or medium quality, however four of the RCTs (Guleria 1998; Kumar 1998; Osman 2004; Pleass 1995) detailed a robust method of randomisation which would be unaffected by physician pre-conceptions. One study (Dominguez 2000) was designed with a treatment group randomised to receive stents universally and one arm receiving stents if the senior surgeon present deemed them necessary, a policy commonly termed "selective" stenting. For two studies the randomisation method was unclear from the paper and the authors did not respond to attempts at clarification (Bassiri 1995; Benoit 1996). Bassiri 1995 and Osman 2004 presented their results censored for either graft loss or patient death unrelated to urological complications or stents, during our meta-analysis these results were adjusted to include the patients as originally randomised. Pleass 1995 randomised patients to four groups: L-G or L-P with or without stent. These results were summated into "with" and "without stent" groups. Only Guleria 1998 detailed a loss to follow-up before three months.
The inclusion and exclusion criteria of each study were difficult to assess. Only one study investigator (Dominguez 2000) reported patients assessed, enrolled or withdrawn prior to randomisation in keeping with modern CONSORT guidelines (Ioannidis 2004). Three studies (Bassiri 1995; Dominguez 2000; Osman 2004) specifically excluded kidneys and recipients with abnormal urinary tracts. One study included multi-visceral recipients (Dominguez 2000) and in three studies (Guleria 1998; Kumar 1998; Osman 2004) the kidneys were only live donor derived. Paediatric recipients were specifically excluded by Osman 2004 and Pleass 1995 whilst in other studies the protocol was unclear. No study reported results on our specific exclusion criteria recipients. The intervention period in Dominguez 2000 was only seven to 10 days, out with our protocol stated minimum period of 14 days.
Benoit 1996, Guleria 1998 and Kumar 1998 stated that the operations reported on were consecutive and detailed no exclusions. None of these authors responded to attempts to contact and the inclusion/ exclusion criteria from their protocols could not be confirmed.
In summary, the study designs were heterogeneous with disparate donors, intervention periods, outcome assessments and statistical analysis, however common themes of randomisation and control were present in all studies and the primary research question of each study followed the principle of this review. After consultation between all four reviewers, we decided to include all the studies in this meta-analysis.
Effects of interventions
Urine leak and obstruction
The MUCs (leak and/or stenosis) incidence ranged between 0 and 4 % in stented patients (median 1.0%) and between 0 and 17.3% (median 7.0 %) in the non-stented patients (additional Table 1 - Incidence of major urological complications (MUCs)) There were significantly fewer events in the stent group ( Analysis 1.1.1: RR 0.24, 95% CI 0.07 to 0.77, P = 0.02; NNT 13). This treatment effect appeared uniform for both complications, with urine leak ( Analysis 1.1.2: RR 0.29, 0.12 to 0.74, P = 0.009) and ureteric stenosis ( Analysis 1.1.3: RR 0.27, 0.09 to 0.81, P = 0.02) similarly reduced.
As was to be expected with considerable differences in the clinical interventions, diagnostic techniques and donors there was a level of statistical heterogeneity in these outcomes; although this did not reach significance (χ² = 11.48, P = 0.07, I² = 48%).
The most substantial finding during subgroup analysis was the influence of surgeon experience on outcome. In studies where only a single experienced surgeon performed or supervised all the operations, the incidence of MUCs in the control group was lower (median 6.3 % versus 10.3% ( Analysis 1.2.1 same surgeon: RR 0.39, 95% CI 0.08 to 1.86; Analysis 1.2.2 many surgeons: RR 0.13, 95% CI 0.01 to 1.16; Table 2 - Influence of surgeon number on incidence of MUCs) and hence the benefit of stent placement was lower (NNT 30 versus 10).
Of the two studies in which both LP and LG techniques were utilised, only Pleass 1995 reported results on LP procedures separately and showed a significantly greater incidence of MUCs in the unstented LP group, and a concomitantly larger risk reduction with the use of stents. No significant influence of the donor organ type (cadaveric, live, multiorgan) or stent type could be identified in this small number of studies. Similarly we could not identify a correlation with the length of stent placement and the incidence of MUCs.
In Dominguez 2000 (stenting versus selective stenting), six (4.4%) patients in the control group received a prophylactic stent. Despite this, there were more MUCs in the control group than the stent group. Three studies specifically reported better graft function or less MUCs whilst the stent was in situ. Bassiri 1995 reported no MUCs in the first two weeks post-transplant in the stented group versus two in the non-stented group. Benoit 1996 showed a significant difference in kidney function early post-transplant (mean serum creatinine at one week 302 µmol/L versus 388 µmol/L ) and Dominguez 2000 reported no MUCs within 20 days in the group that had been stented, whilst all the patients with MUCs in the non-stented group presented within 13 days.
Patient and graft survival
( Table 3 - Graft loss and patient mortality)
No studies directly attributed patient mortality to the use of ureteric stents. Benoit 1996 reported graft loss secondary to use of ureteric stents: two kidneys had to be removed after stent encrustation with corynebacterium could not be eradicated with antibiotics.
There was considerable heterogeneity in definitions, reported morbidity and relative incidence of UTI. Meta-analysis identified a significant increase in UTI associated with stents ( Analysis 1.3: RR 1.49 95% CI 1.04 to 2.15, P = 0.03) with significant statistical heterogeneity (Chi² = 15.32, P = 0.02, I² = 60.8%). This statistic was independent of the diagnostic criteria ( Analysis 1.4), with an equivalent RR reported between studies diagnosing infection on the basis of culture alone ( Analysis 1.4.1: RR 1.44, 95% CI 0.85 to 2.45) and studies requiring clinical symptoms for confirmation of infection ( Analysis 1.4.2: RR 1.23, 95% CI 0.80 to 1.90). Subgroup analysis identified the risk to be associated with the antibiotic regime ( Analysis 1.5). Studies detailing prophylaxis (Bassiri 1995 excluded) with long-term cotrimoxazole 480 mg/d (or 960 mg alternate days) had an equivalent RR for infection between stented and non-stented patients ( Analysis 1.5.1: RR 0.97, 95% CI 0.71 to 1.33) compared with other regimes Analysis 1.5.2: RR 1.78, 95% CI 1.45 to 2.21). The cotrimoxazole regime detailed by Dominguez 2000 included only three doses of cotrimoxazole 480 mg/wk.
Graft implantation requires bladder incision and there will always be a level of haematuria after kidney transplantation. As a symptom or clinical finding, haematuria was sporadically reported by investigators and definitions varied widely (additional Table 4 - Incidence of haematuria). Two studies reported cases of ureteric clot retention (Benoit 1996; Dominguez 2000) and only three studies gave specific definitions for haematuria (Bassiri 1995; Osman 2004; Pleass 1995). Of the three studies that gave figures for haematuria, two studies reported more morbidity in the stented group and one the non-stented group. There was no evidence in the six studies that the presence of a stent per se pre-disposed to recurrent or severe haematuria Analysis 1.6: RR 0.74, 95% CI 0.37 to 1.48).
( Table 5 - Other stent-related complications)
The most significant complications reported were the two graft losses by Benoit 1996. Only two stent breakages were reported, both in patients receiving simultaneous pancreas-kidney transplants with bladder exocrine drainage. Encrustation appeared to be a significant problem for stents left in for longer periods (Bassiri 1995; Pleass 1995), although this statistic was not reported in the original paper from Pleass 1995. The highest rate of stent migration and expulsion was reported by Guleria 1998 who used the longest (24 cm) stent. Cases of "forgotten" stents were reported in Kumar 1998, none of these cases was associated with excess morbidity. Stents appeared well tolerated whilst in situ with only two studies reporting an excess of irritative symptoms and pain not associated with infection (Guleria 1998; Kumar 1998). No studies reported complications from routine removal or secondary obstruction from encrustation.
In summary the maximum reported non-infectious complications were irritative symptoms 5.6% (Guleria 1998), breakage 2.0% (Benoit 1996), migration/malposition/expulsion 7.4% (Guleria 1998), encrustation/urolithiasis 5.7% (Bassiri 1995) and "forgotten" stents 7% (Kumar 1998).
MUCs require interventions varying in severity from a radiological nephrostomy through to surgical urinary diversion. Consequently any intervention which can safely and cost effectively lower their incidence are to be welcomed. Our analysis demonstrates that universal prophylactic ureteric stenting reduces the incidence of MUCs significantly (RR 0.24, 95% CI 0.07 to 0.77).
A recent review and meta-analysis included only five of the RCTs (796 patients) and reported a similar treatment effect (Mangus 2004). However, there is a significant "caveat" to these conclusions. RCTs with unblinded treatment allocation have been acknowledged to overestimate the beneficial effects of medical interventions by up to 40% (Schulz 1995). Such an overestimate would comprehensively negate the treatment effect described here. Of the studies included, Dominguez 2000, which had a shorter intervention period than our protocol originally allowed and compared universal stenting with selective stenting, would be expected to show the least evidence of effect. This was not born out by his results. It was slightly surprising therefore that despite six of the seven studies showing a reduction in the MUCs incidence, only three of the seven studies recommended universal prophylactic stenting.
One explanation for this may be an "experience" effect. The four studies with a single senior author performing or supervising all the operations had a lower complication rate in the control group when compared with studies where more than one surgical team performed the operations. This has been suggested previously and may explain the conclusions of senior study investigators, cognisant of their considerable personal experience, extolling a policy of selective stenting (French 2001).
We were unable to identify a particular stent design associated with a better outcome as a wide variety of stent designs, calibres and lengths were used. Non-infectious adverse events appeared more common in studies using longer stents (Guleria 1998; Osman 2004) and or longer treatment periods (Bassiri 1995; Pleass 1995), although there was no evidence that haematuria was more common. These idiosyncratic stent-related complications were mostly managed by endoscopic removal of the stent - a minimally invasive procedure - which was unavoidable after initial stent placement and has been successfully combined with other necessary interventions for transplant recipients (e.g. peritoneal and haemodialysis catheter removal). Irritative symptoms and stent-related pain were much less frequent than would be expected from studies examining the use of stents in native ureters after general urological procedures where the incidence may be as high as 80% (Joshi 2003). Kumar 2004 has suggested that the combination of a denervated kidney and the high anterior placement of the stent reduce the combination of vesicoureteric reflux pain and trigonal irritation. However, none of the studies specifically used a validated questionnaire to quantify the urinary symptoms or bother associating with either stent usage or removal and this may well be an underestimate of the morbidity burden.
Two grafts were lost as a result of stent related infectious complications and the pooled results indicated a general increased risk of UTI with their use (RR 1.49, 05% CI 1.04 to 2.15). Cotrimoxazole, at a dose of 480 mg/d or 960 mg every other day, is standard therapy in most units and recommended in published guidelines for all kidney transplant recipients as prophylaxis against Pneumocystis carinii pneumonia (EBPG 2002). Our subgroup analysis data also suggests this regime is effective at reducing the incidence of UTI to levels found in unstented transplant urinary tracts (RR 0.97, 95% CI 0.71 to 1.33). It should be emphasised that this statistic, in common with all the analyses, is based on a small number of studies and open to reporting error.
Only one study reported a cost-benefit analysis and although the morbidity associated with stents appears minimal and the therapeutic benefit sizeable, there remain unanswered questions with regards to economic and quality of life issues. In this respect a study of selective versus universal stenting, utilising a stent specific quality of life instrument , would not only provide the opportunity for a more realistic cost benefit analysis of universal prophylactic stenting but also demonstrate whether surgeons can identify intra-operatively urinary tracts that need to be stented. Such a study would also need to stratify surgeons by experience to have significant worldwide implications for practice.
Implications for practice
Routine prophylactic stenting reduces the incidence of MUCs and is to be recommended on the basis of the currently available RCTs. Stents are associated with specific complications and experienced surgeons may adopt a policy of selective stenting, as opposed to universal stenting, to minimise the potential morbidity of stents. Cotrimoxazole, at a dose of 480 mg/d, effectively reduces the risk of UTIs associated with stent placement. Transplant units currently using this antibiotic regime as prophylaxis for pneumocystis pneumonia should not notice an excess of stent related infections.
Implications for research
The optimal stent calibre, length, design and duration remain to be determined and further studies addressing these issues could be instituted. Well designed studies of stenting versus selective stenting, with appropriate patient selection protocols, randomisation, quality of life questionnaires, cost-benefit analyses and surgeon stratification would answer the more clinically important question of whether it is possible and desirable to identify the appropriate patients to stent at the time of graft implantation.
This review has been co-published with Transplantation Oct 2005 (Wilson 2005)
We would also like to thank Dr Nicholas Brooks, Dr David Cranston, Dr Francis Keeley and Dr Petra Macaskill for their editorial advice during the preparation of this review.
Data and analyses
- Top of page
- Authors' conclusions
- Data and analyses
- What's new
- Contributions of authors
- Declarations of interest
- Index terms
Appendix 1. Electronic search strategies
Last assessed as up-to-date: 27 March 2013.
Protocol first published: Issue 3, 2004
Review first published: Issue 4, 2005
Contributions of authors
Writing of review - CHW, AAB, DMM
Screening of titles and abstracts - CHW, AAB
Quality assessment - CHW, AAB, DAR
Data extraction - CHW, AAB
Data analysis - CHW, AAB, DMM
Resolution of discrepancies/disagreements - DMM, DAR
Declarations of interest
Medical Subject Headings (MeSH)
Anastomosis, Surgical; Hematuria [etiology]; Kidney Transplantation [*adverse effects]; Postoperative Complications [*prevention & control]; Randomized Controlled Trials as Topic; Stents [*adverse effects]; Ureter [surgery]; Ureteral Obstruction [etiology; prevention & control]; Urinary Tract Infections [etiology]
MeSH check words
* Indicates the major publication for the study