Ureteroscopy in Pediatric Patients with Spinal Abnormalities

Introduction

E ndoscopic management of pediatric stone disease has evolved significantly since its beginnings in the late 1980s. The adoption of smaller flexible and semirigid ureteroscopes as well as pediatric caliber accessory devices has allowed for ureteroscopy (URS) to be considered a safe and effective treatment modality for stones in the pediatric population. ^1,2

Particular anatomic abnormalities, however, may preclude certain patients from safe endoscopic ureteral procedures, including those with spinal hardware or other anatomic variants that might deviate the ureter from its expected course or involve the ureter in a fibrotic or inflammatory process within the retroperitoneum. Patients with severe scoliosis or spinal hardware are not uncommon in the pediatric population, and these patients present a challenge to the endoscopist because of difficulty in operative positioning and the possibility of significant ureteral tortuosity. URS has typically been the procedure of choice when shockwave lithotripsy (SWL) or percutaneous nephrolithotomy (PCNL) are not possible because of stone burden or stone location. Our experience shows that URS may not be always the straightforward primary modality in all cases.

We propose that ureteroscopic procedures are not free of risk in abnormally positioned ureters because of spinal hardware or spinal deformities. Therefore, a review of 130 consecutive ureteroscopic procedures at our institution over 8 years was performed to determine if the presence of spinal hardware or severe spinal deformities was associated with increased intraoperative complications or worsened efficacy of treatment.

Patients and Methods

A retrospective chart review was performed, using electronic medical records from 130 ureteroscopic procedures in 102 patients at a single institution—Rady Children's Hospital, San Diego. Medical records, operative logs, and radiographic studies were reviewed for each patient. Variables evaluated included total preoperative stone burden via summed radiographic measurement when possible, stone location, intraoperative complications; and stone-free status based on intraoperative assessment, postoperative imaging, or the need for a secondary stone procedure. The majority of retained stones were discovered by intraoperative assessment and need for a secondary procedure.

We classified patients as having spinal abnormalities if they had spinal hardware in place or had moderate to severe scoliosis, as determined by official radiographic report of spinal curvature greater than 35 degrees, in accordance with the Scoliosis Research Society gradation of moderate or greater scoliosis. ³

Intraoperative complications were defined as gross hematuria preventing URS, ureteral injury, urinary extravasation, or need to abort the procedure due to inability to safely negotiate the ureter or access the stone. These were classified according to Satava grading (grade I, an error without consequence or near miss; grade II, an error with immediate identification and correction; grade III, an error that is unrecognized that leads to a significant consequence). ⁴ Statistical analysis included use of the Fischer exact test using SPSS v 17.0 (Chicago, IL).

Results

Data were collected between 2002 and 2010 on 102 patients undergoing a total of 130 ureteroscopic procedures (Table 1). A total of 45 males and 57 females were included in the study. The mean patient age was 12.4 years. There were 105 URS for stone disease whereas 25 were performed for diagnostic purposes or for treatment of encrusted ureteral stents. Nine patients had spinal hardware or significant spinal deformities, such as severe scoliosis. In our cohort, these patients were those with myelodysplasia, cerebral palsy, or other etiologies of scoliosis (Table 2). Spinal hardware was present in the majority of those patients (6 of 9), with the remainder of patients classified as having mild to moderate scoliosis.

When comparing both the intraoperative complications as well as the stone-free status, there was a difference between those patients with spinal hardware and those without. Of 90 URS performed for stones in patients with normal anatomy, the stone free-rate was 61%, compared with 35.7% in the patients with spinal deformities (P=0.088) (Table 3). The median measured stone burden was higher for patients with spinal abnormalities (10.5 mm vs 6 mm, P=0.065). Neither prestent placement nor balloon dilation was routinely used per individual urologist preference. The majority of the patients in whom stone clearance was not achieved went on to have an additional procedure—namely, SWL or PCNL.

When examining the entire cohort of URS for intraoperative complications, there were 13 total complications, 6 (40%) in patients with spinal deformities compared with a 6% complication rate in the URS in patients with normal anatomy (P=0.001). We included inability to access the ureter as a complication as has been done in previous similar studies. ⁵ Of the six complications in patients with spinal deformity, five were related to ureteral tortuosity and one to bleeding (Table 4).

Discussion

It is not uncommon to encounter patients with marked spinal deformities in the pediatric population. Conditions such as cerebral palsy, muscular dystrophy, and myelomeningocele are associated with congenital neuromuscular and paralytic scoliosis that often necessitate surgical intervention early in life. In 2003, approximately 4000 spinal fusion procedures were performed in pediatric hospitals in the United States. ⁶ A combined anterior and posterior approach to spinal fusion is becoming more common in the treatment of pediatric patients with neuromuscular scoliosis. The anterior approach involves accessing the spine either retroperitoneally, posterior to the Gerota fascia and anterior to the quadratus lumborum and psoas muscles, or transperitoneally, in which the posterior peritoneum is incised near the common iliacs and dissection continues near the point at which the ureters cross the iliac vessels. ⁷

While the surgical approach to the spine may not be the sole basis for ureteral deviation in such patients, the presence of hardware and potential resulting inflammatory reaction may certainly contribute and potentially lead to difficulty with ureteral instrumentation. Retroperitoneal fibrosis after anterior spinal fusion has been reported in the literature, citing either the presence of an organized hematoma near the site of dissection or reactive fibrosis within the retroperitoneal space as potential causes of this inflammatory process within the retroperitoneum. ⁸ In fact, retroperitoneal fibrosis caused by an adjacent abdominal aortic aneurysm has been well described to cause a classic deviation of the ureter medially on imaging, ^9,10 although this has been seen to be a normal variant in a small percentage of cases. It is not unreasonable to assume that a similar process occurs in patients after spinal instrumentation.

We propose that this process has contributed to difficulty during URS in this group of patients because of ureteral deviation and/or altered positioning of the ureter. The incidence of ureteral injury in our series was minimal; however, we did find a higher rate of necessity to abort the endoscopic procedure because of perceived difficulty navigating the ureter, which occurred more often in the patients with spinal abnormalities or spinal hardware.

As far as patients with severe scoliosis who do not have spinal hardware, one need only view cross-sectional imaging or intravenous urography (IVU) to recognize the potential deviation of the ureter in these cases. Reports of as high as 34% incidence of genitourinary abnormalities in patients with congenital (nonmyelodysplastic, nonneuromuscular) scoliosis have been described, including ureteral abnormalities. ¹¹ Congenital forms of scoliosis have also been reported to cause a lateral displacement of structures such as the inferior vena cava. ¹² In these cases, access to and manipulation of the ureter can be severely limited because of the significant curvature of the spine and pelvic tilt (Fig. 1).

OPEN IN VIEWER

FIG. 1.

Radiograph of patient in study cohort with scoliosis and spinal hardware.

We found a lower stone clearance rate in the patients with severe scoliosis and/or spinal hardware. This is likely in part because of the much higher stone burden in these patients as well as the fact that they had a higher proportion of proximally located stones. When comparing the URS in patients with a normal spine performed in the first half of our series with those in the latter half, there was a lower stone-free rate in the initial half (60% vs 75%). This likely relates to physician experience with pediatric URS, which would hold true for both the spinal abnormality and the normal patient groups.

Overall, there were significantly more intraoperative complications in the patients with spinal abnormality. We classified these according to Satava grade, which has been used previously to assess intraoperative complications during pediatric URS. ⁵ We believe that these intraoperative complications and worsened stone clearance rate were the result of both difficulty accessing the ureter as well as a deviation of the normal course of the ureter in these patients. As well described by Goumas-Kartalas and Montanari ¹³ with regard to PCNL in adults with spinal deformities, obtaining percutaneous access, patient positioning, and anesthetic issues become major concerns when operating on this population. They reported a higher rate of renal hemorrhage despite a relatively low stone burden in their series and attributed this to the difficulty faced in percutaneous puncture because of the spinal deformity itself. They also reported a lower stone-free rate in their small series compared with the stone-free rate reported in literature guidelines for PCNL.

As far as SWL procedures, spinal deformities, which prevent optimal patient positioning, are considered a relative contraindication to performing effective and safe SWL. ¹⁴ As a result, patient selection becomes an important characteristic when considering ureteroscopic stone management.

In our cohort of patients, we found that spinal abnormalities contributed to impaired access because of several factors. First, these patients have a contracted body habitus that makes positioning difficult as has been seen in both SWL and PCNL literature. The patients with spinal instrumentation and hardware did appear to have more tortuosity of the ureter than in normal patients. Those who did not have previous instrumentation or hardware but had scoliosis had a somewhat fixed ureter that either was narrowed or was not as pliable as in normal patients. All of these factors made passage of the ureteroscope more arduous than usual. Preoperative IVU or CT urography could be used to predict and better characterize these findings; however, many think it is best to limit radiation exposure in children as much as possible and therefore these radiographic studies were typically not performed in the preoperative setting.

To our knowledge, there are no studies in the literature focusing on spinal deformities in children with regard to URS. Our study is unique in that we have compared several aspects of endoscopic ureteral procedures in patients with anatomic abnormalities with normal controls. We identified intraoperative complications according to Satava grade, which may be more applicable to assessment of endoscopic procedures in comparison with Clavien grading.

There are significant limitations to this study including: (1) The small number of patients with spinal deformities in our entire series, thus potentially limiting widespread clinical applicability; (2) multiple surgeons performed the procedures with surgical experience having a known influence on outcome; (3) the stone burden for the patients with spinal abnormality was much greater than that for the patients with normal anatomy; and (4) stone-free status was not always determined by regimented radiologic follow-up and thus relied more on intraoperative assessment of stone clearance, which is not uncommon in pediatric patients likely in part to keep radiation dose to a minimum. ¹ In addition, we have found that follow-up within our patient demographic is not reliable, and this is certainly reflective in the lack of substantial postoperative imaging. Nevertheless, when evaluating the total complication rates, there was a higher rate of total complications in patients with spinal deformities and a noted observed difficulty in performing the procedure in those patients.

We think that ureteroscopic management of stones can be performed safely as first-line treatment for stones in a spinal abnormality population; however, it may be more difficult and may have a lower success rate. Repeated URS or PCNL is another option that can be used as secondary treatment if first-time URS is not successful.

Conclusion

Abnormal spinal anatomy may be indicative of altered retroperitoneal anatomy in certain pediatric populations, particularly with regard to endoscopic ureteral procedures. Our experience with URS in patients with severe spinal deformities suggests that it may not be as safe or straightforward compared with the general pediatric population and therefore necessitates special attention and caution.