Percutaneous Nephrolithotripsy in Patients with Urinary Diversions: A Case-Control Comparison of Perioperative Outcomes

Introduction

C omplications related to urinary-intestinal diversion are numerable. Urinary stone formation is one of the more common adverse events and can develop in the upper tract or within the diversion itself in up to 11% of patients within 3 years after their surgery. ¹

The heightened risk of urinary stone formation can be because of a number of metabolic and anatomic factors. In patients with refluxing ureteral-ileal anastomoses, upper tract dilation is common, leading to urinary stasis and a heightened risk of urinary stone formation. Patients with a continent urinary reservoir have an increase in urinary excretion of calcium, phosphate, and magnesium that might promote urinary calculi formation. ² The use of ileum or colon segments to divert the urinary tract predisposes patients to a hyperchloremic metabolic acidosis from reabsorption of solutes across the intestinal mucosa. ³ The endogenous chronic acid load in these patients arises mainly from reabsorption of ammonium from urine, therefore increasing the burden of urinary titratable acids. ⁴ As a result, phosphate compounds derived from bone mineral and monohydrophosphate then become the dominant titratable acids. ⁵ The chronic acid load in patients with long bowel segments incorporated in the urinary tract and decreased renal function may be so great as to release alkali, bicarbonate, and carbonate from the skeletal bone to aid in maintaining a stable systemic pH and bicarbonate concentration. ^6,7

There is conflicting information in the literature regarding the increased risk of hyperoxaluria and calcium oxalate stone formation in patients with ileal resection, with studies concluding this while others do not find any oxalate hyperabsorption. ⁸

In patients with urinary diversions, the risk of forming stones is also increased because of the higher recurrent urinary tract infection rate seen in this group. Infection with urea-splitting organism results in an elevation in the urinary ammonium from urea breakdown. ⁹ The resulting elevation in urinary pH increases the risk of forming calcium phosphate (CaP) and struvite (magnesium ammonium phosphate [MAP]) stones; usually, CaP stones predominate in this setting, because it is the first salt to crystallize as urinary pH exceeds 6; MAP crystallizes at a pH above 7. ⁹

Certain diversions by the nature of their reconstruction are associated with stone development. The Kock pouch using metal staples for formation of the intussuscepted nipple valve is well known to be associated with stone formation within the pouch. ¹⁰

Percutaneous nephrolithotripsy (PCNL) is often considered the treatment of first choice for patients with urinary diversions and large or complex upper tract calculi. Although shockwave lithotripsy and ureteroscopy are attractive considerations, the altered anatomy in patients with urinary intestinal diversions often make these options less reliable in achieving a stone-free state in this patient population.

In this review, our objective was to determine whether patients with urinary diversions who were undergoing PCNL had greater perioperative complications or inferior efficacy when compared with patients with nondiverted lower urinary tracts.

Patients and Methods

A review of the clinical prospective database of all patients who were undergoing PCNL from 1990 to 2009 performed at our institution by two endourologists (JDD and HR) was performed. Those patients with a urinary-intestinal diversion were identified, and their data were retrospectively reviewed after approval by The University of Western Ontario Health Science Research Ethics Board. Each urinary diversion patient's first PCNL was age matched with four control patients without a urinary diversion who underwent PCNL. Children were age matched with controls who were 18 years old or younger. Adults controls were age (±5 years) and sex matched.

Controls were not matched for stone size or location per se, although all stones were renal in origin and at least 2 cm in greatest dimension for patients in both arms. By matching four control patients with each diversion patient, it was our intention to capture a meaningful comparison of the two cohorts by accounting for a range of clinical scenarios. Controls with a surgery date as close as possible to that of the urinary diverted patient were chosen.

The intraoperative and postoperative outcomes were compared between the diversion and control cohorts. Statistical analysis was performed using GraphPad software (GraphPad Software, Inc, La Jolla, CA), where the noncategorical variables were compared with a paired t test while categorical variables were compared with a Fisher exact test.

Results

Twenty-five patients with a urinary diversion who had undergone 33 PCNLs were identified. One hundred nondiverted patients were selected as the control cohorts based on the criteria outlined above. The types of urinary diversion included: Ileal conduit in 22 (88%) patients and one each of either a Kock pouch, Indiana pouch, or ileal neobladder. The indications for diversion were neurogenic bladder in 15 (60%) patients, invasive bladder cancer in 8 (32%), and benign fistulous disease in 2 (8%). In the urinary diversion group, there were 9 (36%) males and 16 (64%) females with an average age of 49.3 years±20.71 (range 8–85 y).

The control group consisted of 41 (41%) males and 59 (59%) females with an average age of 48.9 years±18.46 (range 4–84 y). Symptomatic presentation was more common in the diversion group (72% vs 82%, P=NS); however, pain was the most common presenting symptom in the control group (32% vs 69%, P<0.0001). Urinary tract infections (64% vs 15%, P<0.0001) and neurologic disease (64% vs 2%, P<0.0001) were identified more frequently in the diversion group (Table 1). History of previous treatment for the same calculus (24% vs 4%, P=0.0044), urinary tract anatomic abnormalities (56% vs 14%, P<0.0001), and patients with a solitary kidney (20% vs 3%, P=0.0081) were noted more commonly for each of these parameters in the diversion group (Table 2). Stones were composed mainly of a >50% proportion of struvite in the diversion group (80% vs 12.5%, P=0.0006), while calcium oxalate calculi were noted more often in the control group (0 vs 32%, P=0.0196) (Table 3).

Regarding intraoperative characteristics, both the diversion and control groups had a similar incidence of a left-sided procedure (46%), as well as bilateral procedures (4%). The mean operative time was similar in both groups; 83.0 minutes±42.22 (43–251 min) in the diversion group and 84.2 minutes±36.55 (34–230 min) in the controls.

Intraoperative percutaneous renal access was facilitated by loopography in 10 (30.3%) patients, retrograde pyelography in 4 (12.1%), antegrade pyelography in 4 (12.1%), and intravenous pyelography in 1 (3.1%). Before 1999, tracts in both groups were dilated exclusively using Amplatz dilators. In the diversion group, 7 (21.2%) procedures had the tract dilated with Amplatz while in 26 (78.7%), balloon dilation was used. In the control group, tracts were dilated with the Amplatz dilator in 32 and with a balloon in 68 (Table 4). (Ultraxx,™ Cook Medical Inc, Bloomington, IN; NephroMax,™ Boston Scientific, Natick MA). Percutaneous renal access was gained by the radiologist using ultrasonography in 40% of the patients in the diversion group, compared with none in the control group (P<0.0001). For patients needing ultrasonography-guided access in radiology, the operative time calculation did not include this time interval.

There was no significant difference in the percentage of patients who had a ureteral stent inserted during PCNL (24% vs 10% in the diversion and control groups, respectively). The number of tracts and their locations are compared in Table 4.

Although the differences in the complication rate between groups were not statistically significant (12% vs 4% in the diversion and control groups, respectively), fever/sepsis was encountered more frequently in the diversion group than in the control group (8% vs 0, respectively, P=0.03). The occurrence of a collecting system perforation (0 vs 2%) and hydrothorax (0 vs 1%) was more commonly encountered in the control group without statistical significance, while the presence of a urine leak was observed more frequently in the diversion group than in the control cohort (4% vs 1%, respectively), also with no statistical significance. No patient in either group needed a blood transfusion.

Mean length of hospital stay for the diversion and control cohorts was 5.5±3.05 and 4.1±2.49 days, respectively, with no statistically significant difference. Although the stone-free rate at discharge was observed more commonly in the control group (83.1% vs 76%), this difference was not significant. Ancillary procedures are listed in Table 5. The postoperative stone-free rate as determined by radiography of the kidneys, ureters, and bladder or nephrostography at discharge from the hospital was 76% for the diversion group and 83% for the controls.

Discussion

The choice of treatment for patients with upper tract urinary stones after urinary diversion is determined by a number of factors, including the stone burden, location, and configuration of the pelvicaliceal system. Extracorporeal shockwave lithotripsy alone is an attractive treatment option for small volume stone burdens with a stone-free rate of up to 81.5%. ¹¹ Ureteroscopy may be an option for small stone burdens, although ureteral access can be challenging. If ureteroscopy is selected, a combination of an antegrade and retrograde approach has been described. ¹²

PCNL is the preferred method used in patients with large renal stones, or in patients with smaller stones when the anatomy does not allow for a retrograde approach. ^12,13 PCNL as monotherapy offers a high stone-free rate with minor morbidity, but requires experience in percutaneous renal access and surgery. ¹³ One of the key technical challenges in performing PCNL in the urinary diversion population is obtaining access to the ureter for retrograde opacification of the collecting system. Identifying and cannulating the ureteral orifice with a ureteral catheter can be difficult if not impossible in some patients with a urinary-intestinal diversion. In patients whose diversions have a refluxing ureteral-enteral anastomosis, filling the conduit or pouch may allow opacification of the upper collecting system to allow percutaneous renal access. Other strategies include using intravenous contrast or direct percutaneous renal puncture to allow antegrade contrast opacification.

For urologists with the skill and equipment, intraoperative ultrasonography-guided percutaneous access is another option. At our center, ultrasonography-directed percutaneous renal access is performed by our interventional radiologist in the radiology department. Although this adds another procedural step for the patient, we found this process did not extend the hospital length of stay.

We have been encouraged to learn that despite the more complex anatomy and potential for greater technical challenges, our patients with urinary diversion tolerate PCNL very well. The heightened risk of postoperative infection and sepsis observed in our series among the diverted patients has led to a change in our clinical care pathway. We now admit all our diversion patients who are undergoing PCNL for 24 hours preoperatively for broad-spectrum intravenous antibiotic coverage.

Conclusions

Our results illustrate that although patients with urinary diversions present a challenge to the endourologist because of higher rates of urinary tract anatomic abnormalities and variable clinical presentation, PCNL should be considered the procedure of choice in patients with a large stone burden and in patients in whom a retrograde approach is not feasible. Ultrasonography-guided percutaneous renal access can be useful if a retrograde approach to the ureter is not possible. Liberal use of ancillary second-look nephroscopy allows for stone-free rates similar to those observed in patients with unaltered anatomy. It has been shown that with appropriate planning and experience, patients with urinary diversions can be treated as safely and effectively by PCNL as nondiverted patients.