Endoscopic management of symptomatic caliceal diverticula: A retrospective comparison of percutaneous nephrolithotripsy and ureteroscopy

Background and Purpose: A myriad of minimally invasive options exist for managing symptomatic caliceal diverticula, including shockwave lithotripsy, percutaneous surgery, retrograde ureteroscopy, and laparoscopy. Yet no direct comparisons have been made in the literature of the relative treatment efficacy of ureteroscopy (URS) and percutaneous nephrolithotripsy (PNL). A retrospective review of our patients was performed to determine the most appropriate endoscopic management option for patients with symptomatic caliceal diverticula. Patients and Methods: Between November of 1994 and April 2001, 39 patients presented with symptomatic caliceal diverticula, 37 of which contained calculi. Twenty-two patients (56%) underwent PNL, and 17 patients (44%) were managed by URS. Of the PNL group, 82% required the creation of a neoinfundibulotomy. The stone burden in the PNL group averaged 11.4 X 12.0 mm and that in the URS group 12.7 X 13.0 mm (p > 0.05). Pain, recurrent urinary tract infections, and nausea and vomiting were the presenting complaints in both subgroups of patients, with pain being by far the most common symptom. The average hospital stay was 2.8 days for the PNL group. All the URS procedures were performed on a same-day-surgery basis. Results, including stone-free, symptom-free, and complication rates, were compared for the two groups. Results: Thirty-five percent of the URS group were symptom free at 6 weeks' follow-up, with an additional 29% reporting an improvement in pain, whereas 86% of the PNL group was completely symptom free at 6 weeks' follow-up. Only 19% of the URS group were stone free on follow-up intravenous urography v 78% of those undergoing PNL (three patients failed to return for follow-up imaging). It was not possible to identify the ostium of the stenotic infundibulum in 4 patients (24%) undergoing URS, and 7 patients (41%) eventually went on to PNL with ultimate success. The PNL was statistically better than URS in producing stone-free results for diverticula located in the upper pole and for stones <11 mm (p <0.05). No complications occurred in the URS group; however, complications were identified in four patients after PNL. One patient developed clot urinary retention necessitating Foley catheterization and manual bladder irrigation; one patient experienced significant bleeding necessitating early cessation of the procedure. Two patients sustained intrathoracic complications, one a pneumothorax and the other a pneumohemothorax after supra-11(th) rib access. Both were managed successfully with tube thoracostomy. Conclusions: Our review clearly suggests an advantage of percutaneous management over ureteroscopy for complex posterior symptomatic caliceal diverticula, although with a slightly increased risk of complications. Therefore, PNL should be considered the primary modality for managing these difficult processes. In cases where the stenotic infundibulum cannot be traversed with a guidewire, creation of a neoinfundibulotomy permitted secure access to the collecting system while providing effective results.