International Journal of Surgery 19 (2015) 83e86
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Best evidence topic
Urinary reconstruction after kidney transplantation: Pyeloureterostomy versus ureteroneocystostomy Joseph M. Norris a, *, Sangeeta Ravi-Shankar b, Stefan G. Klimach c a
University of Cambridge, Cambridge, Cambridgeshire, UK King's College Hospital, London, UK c Brighton and Sussex Medical School, Brighton, East Sussex, UK b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 August 2014 Received in revised form 14 May 2015 Accepted 18 May 2015 Available online 23 May 2015
A best evidence topic in transplant surgery was written according to a structured protocol. The question addressed was: In adults undergoing renal transplantation, does pyeloureterostomy, as compared to ureteroneocystostomy, improve clinical outcomes? A total of 235 articles were identified using the search protocol described, of which six represented the best evidence available to answer the clinical question. The authors, journal, date and country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated. These included one prospective cohort study, three retrospective cohort studies and two case series. The largest of the five cohort studies demonstrated significantly reduced levels of complications with pyeloureterostomy as compared to ureteroneocystostomy. The consensus from the remaining trials was that pyeloureterostomy is a safe but underused technique. However, the majority of the evidence pertaining to pyeloureterostomy and ureteroneocystostomy was archaic, with four of the six dating from pre-1990. Furthermore, the most recent articles (reported in 2010 and 2013) provide only level three and four evidence respectively, and contain important flaws with regard to patient-cohort allocation inherent to the study design. For these reasons we are cautious in recommending pyeloureterostomy over ureteroneocystostomy with the current evidence base, but would like to emphasise that pyeloureterostomy remains a safe surgical option which should form part of the modern transplant surgeon's reconstructive repertoire, particularly when managing patients in which multiple complications are anticipated, or when there is fear of ureteral vascular compromise, such as with cadaveric kidneys. We call for larger scale prospective trials to aid clarification of the roles of pyeloureterostomy and ureteroneocystostomy in renal transplant surgery and to enrich this prescient field with much needed 21st century evidence. © 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
Keywords: Kidney Pyeloureterostomy Transplant Ureteroneocystostomy Urinary reconstruction
1. Introduction
2. Clinical scenario
This BestBET (Best Evidence Topic) in transplant surgery was written according to a structured protocol [1]. The question addressed was: In adults undergoing renal transplantation does pyeloureterostomy, as compared to ureteroneocystostomy, improve clinical outcomes?
Your experienced, but rather traditional, transplant surgery consultant has asked you to take the lead in the next kidney transplantation. When planning the urinary reconstruction, you wonder whether a less favoured technique, the pyeloureterostomy (PU), might actually have an advantage over the more popular approach, the ureteroneocystostomy (UNC). You resolve to consult the literature.
* Corresponding author. Department of Surgery, University of Cambridge, Addenbrooke's Hospital, Cambridge University Hospitals NHS Foundation Trust, Hills Road, Cambridge, Cambridgeshire, CB2 0QQ, UK. E-mail address: [email protected] (J.M. Norris).
3. Three-part question In patients in requiring urinary reconstruction after renal
http://dx.doi.org/10.1016/j.ijsu.2015.05.028 1743-9191/© 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
84
J.M. Norris et al. / International Journal of Surgery 19 (2015) 83e86
Fig. 1. Pyeloureterostomy technique. Align, transect and spatulate the ureters (A). Transect distal donor ureter and begin anastomosis of recipient ureter (B). Complete the lateral and then medial anastomoses (C). (Adapted from: Jaffers et al. 1982.)
transplant surgery, does pyeloureterostomy as compared to ureteroneocystostomy improve clinical outcomes? 4. Search strategy A search strategy using Medline was limited to articles from January 1946 to August 2014, using the PubMed interface: ((pyeloureterostomy OR ureteropyelostomy OR ureteroneocystostomy) AND (kidney OR renal) AND (transplant*)).ti,ab. The search was duplicate filtered and reference lists searched for additional evidence. 5. Search results A total of 235 articles were identified using the search protocol described. Of these 65 were unrelated retrospective cohort studies, 21 were unrelated prospective cohort studies, 20 were unrelated review articles 22 were unrelated basic science or technical articles, 7 were unrelated RCTs, 51 were unrelated case series, 4 were unrelated letters/commentaries, 18 were unrelated case reports and 27 were excluded due to non-English language. Six represented the best evidence available to answer the clinical question. The authors, journal, date, country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated below. 6. Results The key results from the six articles are summarised in Table 1. 7. Discussion Successful renal transplantation depends firstly upon meticulous arterial and venous anastomoses; and then on excellent urinary tract reconstruction. There are several approaches to the surgical reconstruction of the urinary tract during renal transplant surgery, including: ureteroneocystostomy, ureteroureterostomy and pyeloureterostomy (or ureteropyelostomy). The most commonly debated techniques in the literature are ureteroneocystostomy (UNC) and pyeloureterostomy (PU). UNC involves anastomosis of the donor ureter to the recipient urinary bladder, through a submucosal tunnel or with a mucosal anastomosis. In contrast PU involves ipsilateral nephrectomy with preservation of the native ureter followed by spatulated anastomosis of the donor renal pelvis/proximal ureter to the transected distal recipient ureter
before it enters the bladder (Fig. 1). Advantages of both these techniques, over lesser utilised approaches (e.g. ureteroureterostomy), include reduced incidence of early and late obstruction and decreased need for prolonged catheterisation. However anatomical variants, for example duplex donor or recipient systems, add technical difficulty, particularly when performing PU. Furthermore the requirement for ipsilateral nephrectomy could necessitate longer incisions and operative risks. Timsit et al. [2] performed a prospective cohort study of 280 renal transplant patients that underwent UNC (n ¼ 129) or PU (n ¼ 151). Outcome measures were ureteric stenting requirement and overall complications (complex leakage, stenosis and surgical revision). Fewer stents were required for patients undergoing UNC than with those undergoing PU (21.7% versus 68.9%, p < 0.05). The number of overall complications was slightly higher for patients who had UNC compared to those with PU (13.2% versus PU 9.3%, p ¼ 0.15) e however this difference was not statistically significant. This prospective cohort study concluded that PU offers a harmless alternative to UNC. Despite this, patients undergoing PU actually required significantly more intra-operative endourological procedures (ureteral stents) to manage ureteric leak and/or stenosis, suggesting either a more difficult patient subcohort, or technique with more drawbacks. The biggest limitation of this study was the failure to describe how patients were selected to receive UNC or PU. Additionally, intra-operative stents were placed more often in patients receiving PU, potentially confounding the expected incidence of complications between cohorts. Saidi et al. [3] conducted a retrospective cohort study of 1,066 renal transplants that underwent UNC (n ¼ 298) or PU (n ¼ 768). Outcome measures were operative time, surgical complications, urinary complications, urinary obstruction rates and revision/reintervention rates. Operative time was significantly lower in patients undergoing UNC compared with those undergoing PU (182 ± 24 min versus 210 ± 36 min, p < 0.05). Surgical complications were similar between UNC and PU (12.3% versus 9.5%, p ¼ NS). There were no considerable differences shown in urinary complication rates between UNC and PU (5% versus 3.2%, p ¼ NS). In contrast, urinary obstruction occurred significantly more frequently in patients undergoing UNC than those undergoing PU (60% versus 32%, p < 0.05). Furthermore, revision or re-intervention was required more often with UNC compared with PU (46.6% versus 25%, p < 0.05). This large retrospective cohort study concluded that PU is a safe and effective technique with a lower incidence of urinary obstruction and re-intervention. However, the authors note that PU is technically more demanding for the
J.M. Norris et al. / International Journal of Surgery 19 (2015) 83e86
85
Table 1 Best evidence papers. Author, date and country, study type (level of evidence)
Patient group
Outcomes
Key result
Comments
Timsit et al. [2] 2010 France J. Urol. Prospective cohort study (Level 3 evidence)
280 patients underwent urinary reconstruction following renal transplantation UNC n ¼ 129 PU n ¼ 151
Ureteral stenting required (UNC vs. PU)
21.7% vs. 68.9%, p < 0.05
Overall complications (UNC vs. PU)
13.2% vs. 9.3%, p ¼ 0.15
Surgical re-intervention (UNC vs. PU)
5.4% vs. 1.3% (no p value)
The authors utilised a robust study methodology (prospective, dual-centre cohort trial, over a one-year period) to conclude that PU is a safe alternative to UNC. Advantages of PU over UNC include a decreased need for surgical re-intervention (despite an increased requirement for ureteral stenting with PU compared to UNC), and the possibility of performing endourological procedures. A major limitation of the study was the skewed sample size toward the PU arm. NB: The authors defined surgical re-intervention as the need for open surgery, related to a urological complication. This did not include the requirement for ureteral stenting.
Operative time (min) (UNC vs. PU) Surgical complications (UNC vs. PU) Urinary complications (UNC vs. PU) Urinary obstruction (UNC vs. PU) Revision/intervention required (UNC vs. PU)
182 ± 24 vs. 210 ± 36, p < 0.001
Saidi et al. [3] 2013 USA J. Surg. Res. Retrospective cohort study (Level 4 evidence)
1,066 patients underwent urinary reconstruction following renal transplantation UNC n ¼ 298 PU n ¼ 768
12.3% vs. 9.5% 5% vs. 3.2% 60% vs. 32%, p < 0.01 46.6% vs. 24%, p < 0.01 5.3% vs. 1.9%
Again, this article surmised that PU is a safe and effective technique, with higher technical complexity, but with a lower incidence of haematuria, reflux and obstruction, than with UNC. The investigators used a retrospective cohort method over a 15 year period to elucidate complications rates for each technique. This trial was also hampered by very large discrepancies in sample size between the study arms, and by a lack of detail in the exact types of UNC procedure.
Landau et al. [4] 1986 South Africa Br. J. Urol. Retrospective cohort study (Level 4 evidence)
128 patients undergoing urinary reconstruction following renal transplantation UNC n ¼ 76 PU n ¼ 52
Urinary complications (UNC vs. PU)
Jaffers et al. [5] 1982 USA Ann. Surg. Retrospective cohort study (Level 4 evidence)
371 patients underwent urinary reconstruction following renal transplantation UNC n ¼ 111 PU n ¼ 260
Urinary complications (UNC vs. PU)
7.2% vs. 7.7%
This lengthy trial showed that PU is a safe option e with similar rates of morbidity to UNC. The study took a retrospective cohort approach to assess complication rates; the authors used a single method of UNC construction (Politano-Leadbetter), but changed the suture material utilised for PU as the study progressed due to high rates of anastomotic leakage with chromic catgut suture. The evolving nature of this operative approach may alter the impact of the conclusions of this article, as the complications rates do not represent consistent urinary reconstructions. Furthermore, the study arms were also unequal in size.
Donohue et al. [6] 1975 USA J. Urol. Case series (Level 4 evidence)
132 patients underwent urinary reconstruction following renal transplantation UNC n ¼ 88 PU n ¼ 44
Urinary complications (UNC vs. PU)
4% vs. 11%
Greenberg et al. [7] 1977 J. Urol. USA
108 patients underwent urinary reconstruction following renal transplantation PU n ¼ 108
Urinary complications (PU)
11%
In contrast to the other papers, this dated series concluded that UNC may have potential advantages over PU, but that surgeons should be cognisant of the option of PU, for instance in cases of compromised ureteral blood supply. The investigators employed a retrospective case series method, in which the primary outcome was urinary complication rate. The study is heavily marred by a paucity of detail regarding the materials and methodology, and by lack of consideration of additional confounders, such as the higher prevalence of DCD organs in the PU group. The authors of this archaic paper utilised a retrospective case series methodology to investigate complication rates. They demonstrated that PU is a very viable urinary reconstruction
This small study demonstrated that PU has potential advantages over UNC, with regards to urological complications and overall sepsis rates. The authors conducted this retrospective comparative trial to assess urinary complication rates, wound sepsis rates, and the necessity for operative intervention following reconstruction. One drawback of the study included confusion over the different forms of sepsis.
(continued on next page)
86
J.M. Norris et al. / International Journal of Surgery 19 (2015) 83e86
Table 1 (continued ) Author, date and country, study type (level of evidence)
Patient group
Outcomes
Key result
Case series (Level 4 evidence)
Comments
technique. Unfortunately, the conclusions cannot strongly influence the summary of our review, as the authors provide no details on the comparison aspect of their case series, and many other operative aspects have evolved since the article was published.
PU, pyeloureterostomy; UCN, ureteroneocystostomy; VUR, vesicoureteric reflux.
operator and this is reflected in the greater operative times reported for patients undergoing PU. Drawbacks of this study included its retrospective nature and massive discrepancies in sample sizes between cohorts, heavily indicating the more favoured technique at the authors’ centre. Also, the precise detail of the UNC technique performed was not reported. Landau et al. [4] reported a retrospective cohort study of 128 renal transplant patients that underwent UNC (n ¼ 76) or PU (n ¼ 52). Outcome measures were urinary complications (including sepsis). There was an increase in urinary complications in those that underwent UNC compared with those that underwent PU (5.3% versus 1.9%). This small retrospective cohort study demonstrated that PU has significant advantages with regards to urological complications and is a safe, effective technique. Limitations included retrospective nature, small sample size and difficulties classifying septic complications. Jaffers et al. [5] performed a retrospective cohort study of 371 renal transplant patients that underwent UNC (n ¼ 111) or PU (n ¼ 260). Outcome measures were included urinary complications. Complication rates were very similar between UNC and PU (7.2% versus 7.7%). This long-term retrospective cohort study, with data spanning over 13 years, demonstrated that PU is a safe and effective technique for urinary reconstruction for renal transplant recipients. The study was confounded by unequal sample sizes and failure to adequately describe patient selection criteria. Moreover, the surgical techniques employed actually evolved during the data collection period, thus potentially altering the study outcomes in an unquantifiable manner. Donohue et al. [6] reported a case series of 132 renal transplant patients that underwent UNC (n ¼ 88) or PU (n ¼ 44). Outcome measures were urinary complications. There were considerably less urinary complications with UNC than PU (4% versus 11%). In contrast to the other studies described this case series reported more urinary complications in patients undergoing PU compared UNC. However, this series was reported in 1975, the surgical and statistical techniques employed were not adequately described and are likely to have changed substantially since, limiting the applicability of this study to the current clinical question. Greenberg et al. [7] conducted a case series of 108 renal transplant patients that underwent PU (n ¼ 108). Outcome measures were urinary complications. In this group, 11% of patients undergoing PU experienced a urinary complication, of some sort. This case series suggests that PU offers a safe and effective method of urinary reconstruction after renal transplantation. However, it was published in 1977, and since that time a myriad of aspects of transplantation surgery and medicine have changed dramatically,
and so its applicability to todays practice is highly questionable. The trial was limited further by low patient numbers, failure to fully describe surgical techniques, and, perhaps most significantly, the lack of a control cohort for comparison. Therefore, the summary of this publication has not strongly influenced the conclusions of our review. 8. Clinical bottom line The clinical bottom line is that a pyeloureterostomy should form an essential component in every transplant surgeon's toolbox, particularly for difficult patients with multiple anticipated complications, or when there is fear of ureteral vascular compromise, such as with cadaveric kidneys. However, given the archaic nature of the majority of evidence pertaining to this topic, we cannot recommend that centres abandon ureteroneocystostomy in favour of pyeloureterostomy. Clearly, the surgical community has a responsibility to produce high-quality 21st century evidence in this domain so that more definitive guidelines can be produced. A secondary conclusion from our review is that a further, larger scale prospective trial would be immensely useful in this prescient field, to aid clarification to the discrepancies that we have highlighted. Conflict of interest None. References [1] O.A. Khan, J. Dunning, A.C. Parvaiz, R. Agha, D. Rosin, K. Mackway-Jones, Towards evidence-based medicine in surgical practice: best BETs, Int. J. Surg. 9 (8) (2011) 585e588. [2] M.O. Timsit, F. Lalloue, A. Bayramov, M. Taylor, C. Billaut, C. Legendre, et al., Should routine pyeloureterostomy be advocated in adult kidney transplantation? A prospective study of 283 recipients, J. Urology 184 (5) (2010 Nov) 2043e2048. [3] R.F. Saidi, N. Elias, M. Hertl, T. Kawai, A.B. Cosimi, D.S. Ko, Urinary reconstruction after kidney transplantation: pyeloureterostomy or ureteroneocystostomy, J. Surg. Res. 181 (1) (2013 May 1) 156e159. [4] R. Landau, J.R. Botha, J.A. Myburgh, Pyeloureterostomy or ureteroneocystostomy in renal transplantation? Br. J. Urology 58 (1) (1986 Feb) 6e11. [5] G.J. Jaffers, A.B. Cosimi, F.L. Delmonico, M.P. LaQuaglia, P.S. Russell, H.H. Young 2nd, Experience with pyeloureterostomy in renal transplantation, Ann. Surg. 196 (5) (1982 Nov) 588e593. [6] J.P. Donohue, M. Hostetter, J. Glover, J. Madura, Ureteroneocystostomy versus ureteropyelostomy: a comparison in the same renal allograft series, J. Urology 114 (2) (1975 Aug) 202e203. [7] S.H. Greenberg, A.J. Wein, L.J. Perloff, C.F. Barker, Ureteropyelostomy and ureteroneocystostomy in renal transplantation: postoperative urological complications, J. Urology 118 (1 Pt 1) (1977 Jul) 17e19.
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Best evidence topic
Urinary reconstruction after kidney transplantation: Pyeloureterostomy versus ureteroneocystostomy Joseph M. Norris a, *, Sangeeta Ravi-Shankar b, Stefan G. Klimach c a
University of Cambridge, Cambridge, Cambridgeshire, UK King's College Hospital, London, UK c Brighton and Sussex Medical School, Brighton, East Sussex, UK b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 August 2014 Received in revised form 14 May 2015 Accepted 18 May 2015 Available online 23 May 2015
A best evidence topic in transplant surgery was written according to a structured protocol. The question addressed was: In adults undergoing renal transplantation, does pyeloureterostomy, as compared to ureteroneocystostomy, improve clinical outcomes? A total of 235 articles were identified using the search protocol described, of which six represented the best evidence available to answer the clinical question. The authors, journal, date and country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated. These included one prospective cohort study, three retrospective cohort studies and two case series. The largest of the five cohort studies demonstrated significantly reduced levels of complications with pyeloureterostomy as compared to ureteroneocystostomy. The consensus from the remaining trials was that pyeloureterostomy is a safe but underused technique. However, the majority of the evidence pertaining to pyeloureterostomy and ureteroneocystostomy was archaic, with four of the six dating from pre-1990. Furthermore, the most recent articles (reported in 2010 and 2013) provide only level three and four evidence respectively, and contain important flaws with regard to patient-cohort allocation inherent to the study design. For these reasons we are cautious in recommending pyeloureterostomy over ureteroneocystostomy with the current evidence base, but would like to emphasise that pyeloureterostomy remains a safe surgical option which should form part of the modern transplant surgeon's reconstructive repertoire, particularly when managing patients in which multiple complications are anticipated, or when there is fear of ureteral vascular compromise, such as with cadaveric kidneys. We call for larger scale prospective trials to aid clarification of the roles of pyeloureterostomy and ureteroneocystostomy in renal transplant surgery and to enrich this prescient field with much needed 21st century evidence. © 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
Keywords: Kidney Pyeloureterostomy Transplant Ureteroneocystostomy Urinary reconstruction
1. Introduction
2. Clinical scenario
This BestBET (Best Evidence Topic) in transplant surgery was written according to a structured protocol [1]. The question addressed was: In adults undergoing renal transplantation does pyeloureterostomy, as compared to ureteroneocystostomy, improve clinical outcomes?
Your experienced, but rather traditional, transplant surgery consultant has asked you to take the lead in the next kidney transplantation. When planning the urinary reconstruction, you wonder whether a less favoured technique, the pyeloureterostomy (PU), might actually have an advantage over the more popular approach, the ureteroneocystostomy (UNC). You resolve to consult the literature.
* Corresponding author. Department of Surgery, University of Cambridge, Addenbrooke's Hospital, Cambridge University Hospitals NHS Foundation Trust, Hills Road, Cambridge, Cambridgeshire, CB2 0QQ, UK. E-mail address: [email protected] (J.M. Norris).
3. Three-part question In patients in requiring urinary reconstruction after renal
http://dx.doi.org/10.1016/j.ijsu.2015.05.028 1743-9191/© 2015 IJS Publishing Group Limited. Published by Elsevier Ltd. All rights reserved.
84
J.M. Norris et al. / International Journal of Surgery 19 (2015) 83e86
Fig. 1. Pyeloureterostomy technique. Align, transect and spatulate the ureters (A). Transect distal donor ureter and begin anastomosis of recipient ureter (B). Complete the lateral and then medial anastomoses (C). (Adapted from: Jaffers et al. 1982.)
transplant surgery, does pyeloureterostomy as compared to ureteroneocystostomy improve clinical outcomes? 4. Search strategy A search strategy using Medline was limited to articles from January 1946 to August 2014, using the PubMed interface: ((pyeloureterostomy OR ureteropyelostomy OR ureteroneocystostomy) AND (kidney OR renal) AND (transplant*)).ti,ab. The search was duplicate filtered and reference lists searched for additional evidence. 5. Search results A total of 235 articles were identified using the search protocol described. Of these 65 were unrelated retrospective cohort studies, 21 were unrelated prospective cohort studies, 20 were unrelated review articles 22 were unrelated basic science or technical articles, 7 were unrelated RCTs, 51 were unrelated case series, 4 were unrelated letters/commentaries, 18 were unrelated case reports and 27 were excluded due to non-English language. Six represented the best evidence available to answer the clinical question. The authors, journal, date, country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated below. 6. Results The key results from the six articles are summarised in Table 1. 7. Discussion Successful renal transplantation depends firstly upon meticulous arterial and venous anastomoses; and then on excellent urinary tract reconstruction. There are several approaches to the surgical reconstruction of the urinary tract during renal transplant surgery, including: ureteroneocystostomy, ureteroureterostomy and pyeloureterostomy (or ureteropyelostomy). The most commonly debated techniques in the literature are ureteroneocystostomy (UNC) and pyeloureterostomy (PU). UNC involves anastomosis of the donor ureter to the recipient urinary bladder, through a submucosal tunnel or with a mucosal anastomosis. In contrast PU involves ipsilateral nephrectomy with preservation of the native ureter followed by spatulated anastomosis of the donor renal pelvis/proximal ureter to the transected distal recipient ureter
before it enters the bladder (Fig. 1). Advantages of both these techniques, over lesser utilised approaches (e.g. ureteroureterostomy), include reduced incidence of early and late obstruction and decreased need for prolonged catheterisation. However anatomical variants, for example duplex donor or recipient systems, add technical difficulty, particularly when performing PU. Furthermore the requirement for ipsilateral nephrectomy could necessitate longer incisions and operative risks. Timsit et al. [2] performed a prospective cohort study of 280 renal transplant patients that underwent UNC (n ¼ 129) or PU (n ¼ 151). Outcome measures were ureteric stenting requirement and overall complications (complex leakage, stenosis and surgical revision). Fewer stents were required for patients undergoing UNC than with those undergoing PU (21.7% versus 68.9%, p < 0.05). The number of overall complications was slightly higher for patients who had UNC compared to those with PU (13.2% versus PU 9.3%, p ¼ 0.15) e however this difference was not statistically significant. This prospective cohort study concluded that PU offers a harmless alternative to UNC. Despite this, patients undergoing PU actually required significantly more intra-operative endourological procedures (ureteral stents) to manage ureteric leak and/or stenosis, suggesting either a more difficult patient subcohort, or technique with more drawbacks. The biggest limitation of this study was the failure to describe how patients were selected to receive UNC or PU. Additionally, intra-operative stents were placed more often in patients receiving PU, potentially confounding the expected incidence of complications between cohorts. Saidi et al. [3] conducted a retrospective cohort study of 1,066 renal transplants that underwent UNC (n ¼ 298) or PU (n ¼ 768). Outcome measures were operative time, surgical complications, urinary complications, urinary obstruction rates and revision/reintervention rates. Operative time was significantly lower in patients undergoing UNC compared with those undergoing PU (182 ± 24 min versus 210 ± 36 min, p < 0.05). Surgical complications were similar between UNC and PU (12.3% versus 9.5%, p ¼ NS). There were no considerable differences shown in urinary complication rates between UNC and PU (5% versus 3.2%, p ¼ NS). In contrast, urinary obstruction occurred significantly more frequently in patients undergoing UNC than those undergoing PU (60% versus 32%, p < 0.05). Furthermore, revision or re-intervention was required more often with UNC compared with PU (46.6% versus 25%, p < 0.05). This large retrospective cohort study concluded that PU is a safe and effective technique with a lower incidence of urinary obstruction and re-intervention. However, the authors note that PU is technically more demanding for the
J.M. Norris et al. / International Journal of Surgery 19 (2015) 83e86
85
Table 1 Best evidence papers. Author, date and country, study type (level of evidence)
Patient group
Outcomes
Key result
Comments
Timsit et al. [2] 2010 France J. Urol. Prospective cohort study (Level 3 evidence)
280 patients underwent urinary reconstruction following renal transplantation UNC n ¼ 129 PU n ¼ 151
Ureteral stenting required (UNC vs. PU)
21.7% vs. 68.9%, p < 0.05
Overall complications (UNC vs. PU)
13.2% vs. 9.3%, p ¼ 0.15
Surgical re-intervention (UNC vs. PU)
5.4% vs. 1.3% (no p value)
The authors utilised a robust study methodology (prospective, dual-centre cohort trial, over a one-year period) to conclude that PU is a safe alternative to UNC. Advantages of PU over UNC include a decreased need for surgical re-intervention (despite an increased requirement for ureteral stenting with PU compared to UNC), and the possibility of performing endourological procedures. A major limitation of the study was the skewed sample size toward the PU arm. NB: The authors defined surgical re-intervention as the need for open surgery, related to a urological complication. This did not include the requirement for ureteral stenting.
Operative time (min) (UNC vs. PU) Surgical complications (UNC vs. PU) Urinary complications (UNC vs. PU) Urinary obstruction (UNC vs. PU) Revision/intervention required (UNC vs. PU)
182 ± 24 vs. 210 ± 36, p < 0.001
Saidi et al. [3] 2013 USA J. Surg. Res. Retrospective cohort study (Level 4 evidence)
1,066 patients underwent urinary reconstruction following renal transplantation UNC n ¼ 298 PU n ¼ 768
12.3% vs. 9.5% 5% vs. 3.2% 60% vs. 32%, p < 0.01 46.6% vs. 24%, p < 0.01 5.3% vs. 1.9%
Again, this article surmised that PU is a safe and effective technique, with higher technical complexity, but with a lower incidence of haematuria, reflux and obstruction, than with UNC. The investigators used a retrospective cohort method over a 15 year period to elucidate complications rates for each technique. This trial was also hampered by very large discrepancies in sample size between the study arms, and by a lack of detail in the exact types of UNC procedure.
Landau et al. [4] 1986 South Africa Br. J. Urol. Retrospective cohort study (Level 4 evidence)
128 patients undergoing urinary reconstruction following renal transplantation UNC n ¼ 76 PU n ¼ 52
Urinary complications (UNC vs. PU)
Jaffers et al. [5] 1982 USA Ann. Surg. Retrospective cohort study (Level 4 evidence)
371 patients underwent urinary reconstruction following renal transplantation UNC n ¼ 111 PU n ¼ 260
Urinary complications (UNC vs. PU)
7.2% vs. 7.7%
This lengthy trial showed that PU is a safe option e with similar rates of morbidity to UNC. The study took a retrospective cohort approach to assess complication rates; the authors used a single method of UNC construction (Politano-Leadbetter), but changed the suture material utilised for PU as the study progressed due to high rates of anastomotic leakage with chromic catgut suture. The evolving nature of this operative approach may alter the impact of the conclusions of this article, as the complications rates do not represent consistent urinary reconstructions. Furthermore, the study arms were also unequal in size.
Donohue et al. [6] 1975 USA J. Urol. Case series (Level 4 evidence)
132 patients underwent urinary reconstruction following renal transplantation UNC n ¼ 88 PU n ¼ 44
Urinary complications (UNC vs. PU)
4% vs. 11%
Greenberg et al. [7] 1977 J. Urol. USA
108 patients underwent urinary reconstruction following renal transplantation PU n ¼ 108
Urinary complications (PU)
11%
In contrast to the other papers, this dated series concluded that UNC may have potential advantages over PU, but that surgeons should be cognisant of the option of PU, for instance in cases of compromised ureteral blood supply. The investigators employed a retrospective case series method, in which the primary outcome was urinary complication rate. The study is heavily marred by a paucity of detail regarding the materials and methodology, and by lack of consideration of additional confounders, such as the higher prevalence of DCD organs in the PU group. The authors of this archaic paper utilised a retrospective case series methodology to investigate complication rates. They demonstrated that PU is a very viable urinary reconstruction
This small study demonstrated that PU has potential advantages over UNC, with regards to urological complications and overall sepsis rates. The authors conducted this retrospective comparative trial to assess urinary complication rates, wound sepsis rates, and the necessity for operative intervention following reconstruction. One drawback of the study included confusion over the different forms of sepsis.
(continued on next page)
86
J.M. Norris et al. / International Journal of Surgery 19 (2015) 83e86
Table 1 (continued ) Author, date and country, study type (level of evidence)
Patient group
Outcomes
Key result
Case series (Level 4 evidence)
Comments
technique. Unfortunately, the conclusions cannot strongly influence the summary of our review, as the authors provide no details on the comparison aspect of their case series, and many other operative aspects have evolved since the article was published.
PU, pyeloureterostomy; UCN, ureteroneocystostomy; VUR, vesicoureteric reflux.
operator and this is reflected in the greater operative times reported for patients undergoing PU. Drawbacks of this study included its retrospective nature and massive discrepancies in sample sizes between cohorts, heavily indicating the more favoured technique at the authors’ centre. Also, the precise detail of the UNC technique performed was not reported. Landau et al. [4] reported a retrospective cohort study of 128 renal transplant patients that underwent UNC (n ¼ 76) or PU (n ¼ 52). Outcome measures were urinary complications (including sepsis). There was an increase in urinary complications in those that underwent UNC compared with those that underwent PU (5.3% versus 1.9%). This small retrospective cohort study demonstrated that PU has significant advantages with regards to urological complications and is a safe, effective technique. Limitations included retrospective nature, small sample size and difficulties classifying septic complications. Jaffers et al. [5] performed a retrospective cohort study of 371 renal transplant patients that underwent UNC (n ¼ 111) or PU (n ¼ 260). Outcome measures were included urinary complications. Complication rates were very similar between UNC and PU (7.2% versus 7.7%). This long-term retrospective cohort study, with data spanning over 13 years, demonstrated that PU is a safe and effective technique for urinary reconstruction for renal transplant recipients. The study was confounded by unequal sample sizes and failure to adequately describe patient selection criteria. Moreover, the surgical techniques employed actually evolved during the data collection period, thus potentially altering the study outcomes in an unquantifiable manner. Donohue et al. [6] reported a case series of 132 renal transplant patients that underwent UNC (n ¼ 88) or PU (n ¼ 44). Outcome measures were urinary complications. There were considerably less urinary complications with UNC than PU (4% versus 11%). In contrast to the other studies described this case series reported more urinary complications in patients undergoing PU compared UNC. However, this series was reported in 1975, the surgical and statistical techniques employed were not adequately described and are likely to have changed substantially since, limiting the applicability of this study to the current clinical question. Greenberg et al. [7] conducted a case series of 108 renal transplant patients that underwent PU (n ¼ 108). Outcome measures were urinary complications. In this group, 11% of patients undergoing PU experienced a urinary complication, of some sort. This case series suggests that PU offers a safe and effective method of urinary reconstruction after renal transplantation. However, it was published in 1977, and since that time a myriad of aspects of transplantation surgery and medicine have changed dramatically,
and so its applicability to todays practice is highly questionable. The trial was limited further by low patient numbers, failure to fully describe surgical techniques, and, perhaps most significantly, the lack of a control cohort for comparison. Therefore, the summary of this publication has not strongly influenced the conclusions of our review. 8. Clinical bottom line The clinical bottom line is that a pyeloureterostomy should form an essential component in every transplant surgeon's toolbox, particularly for difficult patients with multiple anticipated complications, or when there is fear of ureteral vascular compromise, such as with cadaveric kidneys. However, given the archaic nature of the majority of evidence pertaining to this topic, we cannot recommend that centres abandon ureteroneocystostomy in favour of pyeloureterostomy. Clearly, the surgical community has a responsibility to produce high-quality 21st century evidence in this domain so that more definitive guidelines can be produced. A secondary conclusion from our review is that a further, larger scale prospective trial would be immensely useful in this prescient field, to aid clarification to the discrepancies that we have highlighted. Conflict of interest None. References [1] O.A. Khan, J. Dunning, A.C. Parvaiz, R. Agha, D. Rosin, K. Mackway-Jones, Towards evidence-based medicine in surgical practice: best BETs, Int. J. Surg. 9 (8) (2011) 585e588. [2] M.O. Timsit, F. Lalloue, A. Bayramov, M. Taylor, C. Billaut, C. Legendre, et al., Should routine pyeloureterostomy be advocated in adult kidney transplantation? A prospective study of 283 recipients, J. Urology 184 (5) (2010 Nov) 2043e2048. [3] R.F. Saidi, N. Elias, M. Hertl, T. Kawai, A.B. Cosimi, D.S. Ko, Urinary reconstruction after kidney transplantation: pyeloureterostomy or ureteroneocystostomy, J. Surg. Res. 181 (1) (2013 May 1) 156e159. [4] R. Landau, J.R. Botha, J.A. Myburgh, Pyeloureterostomy or ureteroneocystostomy in renal transplantation? Br. J. Urology 58 (1) (1986 Feb) 6e11. [5] G.J. Jaffers, A.B. Cosimi, F.L. Delmonico, M.P. LaQuaglia, P.S. Russell, H.H. Young 2nd, Experience with pyeloureterostomy in renal transplantation, Ann. Surg. 196 (5) (1982 Nov) 588e593. [6] J.P. Donohue, M. Hostetter, J. Glover, J. Madura, Ureteroneocystostomy versus ureteropyelostomy: a comparison in the same renal allograft series, J. Urology 114 (2) (1975 Aug) 202e203. [7] S.H. Greenberg, A.J. Wein, L.J. Perloff, C.F. Barker, Ureteropyelostomy and ureteroneocystostomy in renal transplantation: postoperative urological complications, J. Urology 118 (1 Pt 1) (1977 Jul) 17e19.
