Pyeloplasty vs. nephrectomy for ureteropelvic junction obstruction in poorly functioning kidneys (differential renal function <20%): a multicentric study.

INTRODUCTION: Management of poorly functioning kidneys with ureteropelvic junction obstruction (UPJO) is controversial, with some recommending direct nephrectomy and others direct pyeloplasty, and others temporary diversion. This study aimed to determine whether pyeloplasty is associated with higher complication rate than nephrectomy under these circumstances and whether it allows for functional recovery. METHODS: A retrospective review of 77 patients undergoing surgery for UPJO associated with a differential renal function (DRF) < 20% at 4 centres between January 2000 and December 2015 was conducted. Short- and long-term complications between pyeloplasties (n = 47) and nephrectomies (n = 16) and the changes in the DRF after pyeloplasty were compared. RESULTS: Patients undergoing nephrectomy had significantly lower median pre-operative DRF (p < 0.001) and were significantly more likely to undergo a minimally invasive approach (p = 0.002) than those undergoing pyeloplasty. No postoperative variable was statistically different between groups. After a mean follow-up of 63 (10-248) months, no statistically significant difference was found in intra-operative, early, late, and overall complications between pyeloplasty and nephrectomy. Pyeloplasty failed in 3% (2/62) of cases. Of the patients undergoing successful pyeloplasty, 36 had a pre-operative and a postoperative renogram, and functional recovery >5% was observed in 13 (36%), whereas the DRF remained unchanged in 16 (45%). Only postnatal diagnosis was associated with a significantly higher chance of functional recovery (odds ratio [OR] = 4.13, p = 0.047). DISCUSSION: Nephrectomy was more commonly performed using a minimally invasive approach and required less-intense follow-up than pyeloplasty. Moreover, complications after pyeloplasty, although occasional, required additional surgery which was never required after nephrectomy. Moreover, it should be considered that conservative treatment might be a third option in some of these patients. CONCLUSIONS: In the study patients, pyeloplasty was not associated with significantly higher morbidity than nephrectomy. Need for deferred nephrectomy seems exceptional in decompressed kidneys even though renal function remains poor. Of the pyeloplasty cases not requiring additional surgery, one-third showed an improvement in DRF and a postnatal diagnosis was the only predictive factor for renal functional recovery.