The Effect of Restrictive Bariatric Surgery on Urolithiasis

Introduction

T he obesity epidemic is a leading public health concern in the United States that is associated with significant comorbidity and increased healthcare expenditures. ^1,2 Recent studies suggest that the prevalence of Americans who are morbidly obese (body mass index [BMI] >40) is increasing at a faster rate than those who are moderately obese (BMI >30), reinforcing the need for proven weight-loss solutions. ³ Currently, bariatric surgery is the only available therapy that can result in effective, long-term weight loss as well as the associated reduction of comorbidity and long-term mortality. ^4,5

Among the different options, Roux-en-Y gastric bypass (RYGB) surgery, a mixed malabsorptive/restrictive operation, is currently the most commonly performed bariatric procedure in the United States, representing approximately 70% to 75% of all bariatric procedures performed. ⁶ Both laboratory and clinical studies, however, have found that the malabsorptive component of RYGB is associated with enteric hyperoxaluria and an increased risk of kidney stone disease. ^{7 –9}

Besides RYGB surgery, adjustable gastric banding and sleeve gastrectomy are becoming increasingly popular options for bariatric surgery in the United States. These procedures are purely restrictive and potentially avoid the metabolic complications associated with the malabsorption seen in RYGB. Therefore, it stands to reason that patients undergoing restrictive procedures should not be at an increased risk for stone disease of metabolic etiology.

Studies on this topic so far have demonstrated no increase in urinary stone risk factors after restrictive bariatric procedures. ¹⁰ A previous publication also demonstrated no increased incidence in kidney stones with banding restrictive procedures, but did not include any data on sleeve gastrectomy, a common procedure that has gained increasing popularity as a bariatric surgical option. ^6,11 Therefore, we sought to report the incidence of kidney stone disease after both adjustable gastric banding as well as sleeve gastrectomy performed for bariatric indications.

Patients and Methods

This study is an observational retrospective case series that seeks to describe the incidence of kidney stone disease in patients receiving adjustable gastric banding or sleeve gastrectomy. The Institutional Review Board of the University of California, San Diego (UCSD) granted a waiver of consent and approved the retrospective data analysis of all patients undergoing adjustable gastric banding and sleeve gastrectomy at the UCSD Center for the Treatment of Obesity over a 54-month period from September 2006 to February 2011. Each chart belonging to this treatment period was manually reviewed for all records of urology department visits, emergency department visits, and any positive abdominal CT imaging, CT urography, abdominal or renal ultrasound studies, or reports of radiography of the kidneys, ureters, and bladder. If outside records were available in the UCSD chart, these records were also reviewed; however, we otherwise did not have access to patient charts from outside healthcare systems.

Both Ethicon^© and Allergan^© adjustable gastric banding procedures were combined into the same recipient cohorts. The primary outcome was a composite of either diagnosis of urinary calculi or stone removal procedure after the restrictive bariatric procedure. Secondary outcomes included kidney stone surgery, stone analysis, and 24-hour urine studies. Manual review of all charts along with an electronic search for all International Statistical Classification of Diseases and Related Health Problems (ICD-9) diagnostic and procedural codes was performed (a list of the ICD-9 codes used can be found in Table 1). Patients with any preexisting urinary calculi diagnosis, positive imaging for renal calculi, or treatment for nephrolithiasis before the date of their bariatric procedures were excluded from analysis. Patients without any of these exclusions were assumed to be stone free before their restrictive procedures. Image reports mentioning calcifications were reviewed to rule out stones, and image reports mentioning calculi were reviewed to rule out Randall's plaques.

To account for the variable postsurgery follow-up durations between procedure cases, person-time incidence rates were calculated for the adjustable gastric banding and sleeve gastrectomy cohorts. The calculations assume that the patient population exposed to bariatric procedures is at risk for the investigated outcome of kidney stone diagnosis or treatment, and that the risk remains constant across time. Using person-time rather than just time is useful when the amount of observation time differs between people, such as kidney stones developing after bariatric procedures. ¹² This method also provides a variant and more useful expression of incidence for clinicians.

Results

A total of 425 patients received restrictive bariatric surgery within the 54-month period of observation. Of these patients, 417 did not have any preexisting urinary calculus diagnosis, positive imaging, or treatment before their bariatric procedures. From this inception cohort of 417 subjects that were included in our analysis, 332 patients received adjustable gastric banding and 85 patients received sleeve gastrectomy. Females outnumbered males in both groups approximately 4 to 1. Within the adjustable gastric band cohort, 4 of 332 (1.2%) patients subsequently had an upper tract urinary calculus diagnosis, with a mean follow-up time of 42.5 months, representing 1177.83 person years. This translates to a person-time incidence rate of 3.40 stone diagnoses per 1000 person-years—alternatively expressed: If you were to follow 1000 gastric band cases over 1 year, 3.40 stone diagnoses would occur. Within the sleeve gastrectomy cohort, 1 of 85 (1.18%) patients subsequently had an upper tract urinary calculus diagnosis, with a mean follow-up time of 26.8 months, representing a total of 190.17 person-years. This translates to a person-time incidence rate of 5.25 stone diagnoses per 1000 person-years—alternatively expressed: If you were to follow 1000 gastric band cases over 1 year, 5.25 stone diagnoses would occur.

The majority of patients with a new stone diagnosis after either laparoscopic gastric banding or sleeve gastrectomy were asymptomatic and only 2/5 underwent surgery. Among the patients in whom a new stone developed after laparoscopic gastric banding, 2/4 needed surgery. One of these patients underwent PCNL and the other patient was treated by ureteroscopy with holmium laser lithotripsy. One of these patient's stone was analyzed and was 100% calcium oxalate monohydrate. In the patient who underwent a laparoscopic sleeve gastrectomy, a renal calculus developed that passed spontaneously. None of the five patients in whom stone disease developed underwent 24-hour urine collections.

Discussion

RYGB surgery, currently the most popular bariatric procedure in the United States, achieves weight loss through a combination of restrictive and malabsorptive mechanisms by physically limiting the size of the gastric reservoir and limiting nutrient absorption by bypassing the distal stomach, duodenum, and proximal jejunum. Because the incidence of new stones per 1000 person-years after RYGB is 16.62⁹ and the incidence of new stones per 1000 person-years in obese patients is 11.3, ⁹ our study further substantiates previous work that gastric banding does not increase the risk of nephrolithiasis and also provides new data that sleeve gastrectomy is equally not a risk factor for nephrolithiasis. Such a decrease in stone incidence provides one significant clinical advantage of performing restrictive bariatric procedures instead of traditional RYGB.

Purely restrictive procedures such as gastric banding and sleeve gastrectomy have recently gained popularity, with gastric banding comprising anywhere from 20% to 25% of bariatric procedures in the United States. Studies comparing restrictive procedures with RYGB have indicated shorter operating room times and hospitalizations, lower complication rates, and decreased 30-day mortality in restrictive procedures. ⁶ RYGB surgery, however, consistently results in better long-term weight loss and obesity comorbidity risk reduction. ¹³ Despite the relative outcome differences, the increasing popularity of restrictive procedures merits investigating the risk for stone disease after these surgeries.

As opposed to RYGB surgery, gastric banding and sleeve gastrectomy do not involve any bowel restructuring and avoid the nutrient malabsorption present in RYGB. To date, the published incidence data on stone risk after restrictive bariatric procedures has focused solely on gastric banding, leaving out sleeve gastrectomy in analyses. Nadler and associates ¹⁴ first reported on potential stone complications after gastric banding in a pediatric population: 1 patient out of 73. Semins and colleagues ¹¹ followed with a study that reported no significant differences in kidney stone incidence between patients receiving gastric banding and a control group of obese patients. ¹¹

Our study represents the first evaluation of the incidence of kidney stones after sleeve gastrectomy and demonstrates the minimal risk of nephrolithiasis subsequent to this procedure. In distinct contrast, RYGB surgery carries a 7.65% risk of kidney stones. ⁹ Admittedly, the follow-up is short and potentially underestimates the long-term risk of nephrolithiasis. This study is also unique in that it represents, to our knowledge, the largest series reviewing the incidence of nephrolithiasis after gastric banding and confirms previous published results.

Because our study is a retrospective descriptive case series, limited causality of risk can be inferred between restrictive bariatric surgery and development of kidney stones. Limitations inherent in any case series include the lack of comparable controls necessary for tests of significance and causal inference. There is also the inherent assumption involved with calculating person-time incidence that the risk for development of kidney stones after restrictive bariatric surgery is constant across time. Also, because of institutional limitations, we were only able to use ICD-9 codes and manual chart review for kidney stone diagnosis and treatment identification, and were unable to simultaneously use billing data and Current Procedural Terminology codes, which could have improved the accuracy of this study. Incomplete or erroneous charting, lack of access to outside medical records, and coding may also have limited our ability to capture all outcomes accurately, and with a lack of controls, we are unable to account for the extent of this. Also, our study is a conservative one, given that we make the assumption that patients without exclusionary criteria were free of stone history before their surgeries. This study is also unable to account for any occult or undetected nephrolithiasis because of its reliance on the reports within the medical record. Finally, larger sample sizes, longer follow-up, and controlled prospective studies are necessary to validate the low risk found in our study.

Conclusions

Published data on the effect of restrictive bariatric surgery on kidney stone formation have largely excluded sleeve gastrectomy data. We now provide information on both the minimal risk of nephrolithiasis after sleeve gastrectomy and confirm the low risk of nephrolithiasis after gastric banding. Prospective studies that include longer follow-up are necessary to validate our findings.