Abstract
Introduction:
Although ureteroscopy (URS) has been established as a viable treatment for stones in obese patients, its safety and success has not been fully elucidated. The current study describes the worldwide prevalence of obesity in patients with urolithiasis and examines trends in URS outcomes, safety, and efficacy.
Methods:
This study utilized the Clinical Research Office of the Endourological Society (CROES) URS Global Study, which was a prospective, multicenter study including 11,885 patients treated with URS for urinary stones at 1 of 114 urology departments across 32 countries. The relationship between body mass index (BMI), diabetes, and creatinine, with retreatment, stone-free rates, complications, and long hospital stay, was examined with a multivariate logistic regression analyses.
Results:
Of the 10,099 URS patients with BMI data, 17.4% were obese and 2.2% were super obese. Overall, 86.7% patients were stone free and 16.8% required retreatment. Higher BMI was associated with lower stone-free rates, and any deviation from normal weight was associated with higher retreatment rates. In multivariate analysis controlling for several variables including stone size, the association between BMI and lower stone-free rates with higher retreatment rates persisted. Intraoperative complications occurred in 518 (5.1%) patients, and 343 (3.4%) experienced a postoperative complication. Postoperative complications were more frequent in the underweight and super obese subjects, and there was no relationship between BMI and intraoperative complications.
Discussion:
Although URS for stone disease was found to be an overall safe procedure for obese and super obese patients, efficacy of the procedure may be lower compared with normal-weight subjects and higher retreatment rates may be necessary.
Introduction
T
Stone formation in obesity
Incident stone formation has been associated with obesity. 6,7 The effect of excessive body weight on stone formation is suggested both through direct dietary excess resulting in abnormal urine supersaturations and via decreased insulin sensitivity. 5,6,8 –10 Insulin resistance may be triggered by hyperuricemia; high levels of uric acid in blood secondary to dietary excesses. Hyperuricemia is, therefore, not only regarded as a renal manifestation of the metabolic syndrome 9 but also associated with calcium oxalate and uric acid stone formation. Furthermore, insulin resistance may contribute to kidney stone formation. Insulin helps to drive ammonia into the proximal tubule of the nephron to act as a hydrogen ion buffer forming ammonium. When insulin resistance is significant with resulting diabetes mellitus (DM), there is a reduction in the production and transport of ammonia, resulting in alterations of urine acidity, and lower urine pH. The low pH, subsequently, increases formation of uric acid stones. 9
Whether the link between stone disease and obesity is direct, indirect, or both is not completely clear; however, known urinary lithogenic risk factors that are frequently found in obesity are low urine volume, increased urinary osmolality, acidic pH, hyperoxaluria, hypercalciuria, and insulin resistance. 6,7
Treatment options in obesity
Although both extracorporeal shockwave lithotripsy (SWL) and ureteroscopy (URS) are considered reasonable first-line therapies for symptomatic ureteral calculi, obese subjects are often treated with URS. The increased skin to stone distance secondary to excess adipose deposition observed in obese patients makes SWL inefficient and decreases overall success rates. 6 For renal calculi, URS may be chosen over percutaneous nephrolithotomy (PCNL) in obese patients due to difficulty with prone positioning and limitations of PCNL equipment length, which may not be able to traverse the increased skin-to-stone distance. 6 Because URS can be performed in the supine position and skin-to-stone distance does not influence stone access, URS has become a standard treatment option for urolithiasis in obese patients.
Although URS has been established as a viable treatment option for stones in obese patients, the safety and success may be different when compared with the non-obese patient. The current study describes the worldwide prevalence of obesity in patients with urolithiasis and examines trends in URS outcomes, safety, and efficacy.
Methods
Study population
The Clinical Research Office of the Endourological Society (CROES) URS Global Study was a prospective, observational, international, multicenter study including 11,885 patients with urinary stones, and also including 10,099 cases with known information on body mass index (BMI). These patients were all treated with URS at one of the 114 urology departments with URS expertise across 32 countries. 11 Each of the participating centers included all consecutive patients who were treated during a 1 year period. Starting date for inclusion of the first patient was between January 2010 and October 2011. Eligible patients were those who were candidates for URS, as a primary treatment or after failure of a previous treatment. In the total CROES URS study, there were no specific exclusion criteria.
Data collection
For every patient, baseline characteristics were recorded, and intraoperative, postoperative, and 3-month follow-up data were collected, as previously described. 11 In brief, baseline characteristics recorded included age, gender, American Society of Anesthesiologists (ASA) score, BMI, and medical history regarding diabetes, cardiovascular disease, congenital abnormalities, previous stone treatment, and anticoagulant use. Preoperative urinary creatinine levels were obtained as a surrogate marker of urinary pH. 8 Intraoperative characteristics were duration of the procedure, intraoperative complications, and stone-free status. Stone-free status was determined by postoperative imaging protocols, which differed at each site. Postoperative outcome data collected included: long hospital stay defined as longer than 1 day, retreatment, including readmission within 3 months, and postoperative complications according to the Clavien-Dindo grading system. 12,13
Statistical analyses
The current study examined regional and worldwide trends in outcomes of URS for stone disease in obese patients. The influence of BMI (obesity), DM, and creatinine on retreatment, stone-free rates, complications, and long hospital stay was analyzed. To assess differences between patients based on BMI, patients were categorized into underweight (<18.5 kg/m2), normal weight (18.5–25 kg/m2), overweight (25–30 kg/m2), obese (30–40 kg/m2), and super obese (> 40 kg/m2).
Differences between patients based on BMI were described by using mean (SD) for continuous, normally distributed data, and median (IQR) for continuous, skewed data. For categorical and dichotomous data, actual numbers and percentages were presented numbers (%).
The role of BMI, DM, and creatinine in retreatment, stone-free rates, complications, and long hospital stay was examined by using univariate logistic regression models. Age, gender, previous treatment, congenital abnormalities, preoperative stent placement, and stone burden were examined as potential confounders. Possible effect modification by gender was also examined.
The theoretical relationship of BMI, diabetes, and creatinine, with retreatment, stone-free rates, complications, and long hospital stay, was examined with a multivariate logistic regression analyses. Possible confounders in this relationship were again: gender, age, previous treatment, congenital abnormalities, preoperative stent placement, and stone burden. For the analyses with BMI categories, the normal-weight group was used as reference. Odds ratios represent the relative effect for each of the groups compared with the normal-weight patients.
Results
Baseline characteristics of the 10,099 patients undergoing URS with available information on BMI are presented in Table 1 and Figure 1. The vast majority of the patients were normal or overweight (35.4% and 42.8% respectively), whereas 17.4% were obese and 2.2% were super obese. As BMI increased in patients undergoing URS, the proportion of patients with DM also increased. In the total URS cohort, 63.7% were men and 36.3% were women. Women did not show total higher rates of BMI, but they were a higher percentage of the patients in the underweight and super obese groups. Also, the super obese group proved to be treated with flexible URS more often.
A visual representation of descriptive information in the population as currently used.
Data are mean (SD) or n (%) of patients for whom data were available.
Percentages exclude missing values from denominators. Long hospital stay >1 day.
BMI = body mass index; CROES = Clinical Research Office of the Endourological Society; IOC = intraoperative confirmation; IVU = intravenous urography; KUB = kidney ureter and bladder X-ray; NA = not available; PCNL = percutaneous nephrolithotomy; RP = retrograde pyelogram; SD = standard deviation; SWL = extracorporeal shockwave lithotripsy; URS = ureteroscopy; US = ultrasound.
Intraoperative complications occurred in 518 (5.1%) patients undergoing URS. Median hospital stay was 1 days (1–2). Postoperative complications were noted in 343 (3.4%) patients. Of the 9876 patients with appropriate imaging follow-up, 8560 (86.7%) were found to be stone free after their URS and 1692 (16.8%) required a second stone treatment or retreatment. In Table 2, the results of logistic regression analyses are presented. In univariate analysis, higher BMI, presence of DM, and higher preoperative serum creatinine levels were significantly associated with retreatment, stone-free rates, postoperative complications, and long hospital stay. For long hospital stay and postoperative complications, a significant association is found, but there is no pattern that fits current hypothesis (linear or exponential relationship) in BMI categories compared with normal-weight subjects.
Odds ratio; p-values in parentheses. BMI categories are compared with patients with a normal weight.
p < 0.05, ** p < 0.01, *** p < 0.001.
Table 3 shows the multivariate logistic regression models of BMI with outcomes that were significantly associated with BMI in the univariate analyses. These models are the same as model 1, but then corrected for DM and creatinine (model 2). The relationship of BMI categories with the outcomes of retreatment rates, stone-free rates, postoperative complications, and long hospitalization did not change after controlling for DM and serum creatinine.
Odds ratio; p-values in parentheses. BMI categories are compared with patients with a normal weight.
Model 1 with correction for creatinine and diabetes mellitus.
Model 2 with correction for gender, age, previous treatment, congenital abnormalities, preoperative ureteral stent placement, type of URS, and stone burden.
p < 0.05, ** p < 0.01, *** p < 0.001.
Model 3 (Table 3) shows multivariate modeling controlling for preoperative variables, including gender, age, previous treatment, congenital abnormalities, preoperative stent placement, type of URS, and stone burden. Controlling for the preoperative variables did not change the significance of the relationships between BMI categories and outcomes; however, it did change the nature of the relationships. High BMI remained associated with lower stone-free rates. However, patients from all the BMI categories had a higher retreatment rate compared with normal-weight subjects. The relationship between postoperative complications and BMI was significant when patients were either underweight or super obese. Finally, the length of hospital stay seems to linearly decrease with increasing BMI.
Discussion
This analysis from the CROES URS Global Study aimed at investigating the safety and efficacy of URS in obese patients. In the 10,099 URS patients studied, we found that nearly 20% of the population was either obese (17.4%) or super obese (2.2%). When assessing outcomes, higher BMI was associated with a lower stone-free rate and any deviation from normal BMI was associated with higher retreatment rates even after controlling for stone burden and other confounding factors. The study also found that postoperative complications were higher for underweight and super obese subjects, and there was no relationship between BMI and intraoperative complications. Thus, although it is safe with low intra- and postoperative complications, the efficacy of URS in the obese patient in terms of retreatment and stone-free rates may be decreased compared with normal-weight subjects.
Although the rise in obesity has been frequently cited as a potential cause for the observed increase in incidence of stone disease, the current study demonstrates that only 20% of all URS treated patients at 114 locations were either obese or super obese. This observed low rate of obesity in patients undergoing surgical intervention for stone disease suggests that BMI is not the only lithogenic factor influencing the increase in stone formation and other factors may be at play We did note that DM was more common in the obese and super obese groups compared with the normal-weight patients, and these findings were comparable to rates of DM previously presented by Strohmaier and colleagues. 9
In terms of URS safety, we found no difference in intraoperative complications between the BMI groups in the currently examined study population. Postoperative complication rates were higher in both the super obese and underweight groups of patients. In a recent systematic review of URS in the obese population, Ishii and colleagues reviewed 15 studies. 6 They found an overall complication rate of 9.3%, and morbidly obese patients were at higher risk than obese patients. It should be noted that although the complication rates may be statistically significantly higher compared with normal-weight subjects (3.0%), the complication rates for the entire cohort was low at 5.2% and 7.6% for the underweight and super obese groups, respectively. Thus, this suggests that URS is a safe procedure in all BMI subgroups.
In terms of efficacy, the study found retreatment rates to be higher and stone-free rates to be lower in the obese and super obese patients treated with URS compared with the normal-weight patients undergoing URS. The normal weight group had an overall 88.1% stone-free rate; whereas the stone-free rate dropped to 83.1% in the obese, and 73.0% in the super obese. This finding of decreased stone-free rates was statistically significant even after correcting for other confounding variables, including DM, creatinine, gender, age, previous treatment, congenital abnormalities, preoperative stent placement, type of URS, and stone burden. Also, current results show higher frequency of flexible URS use (36.8%) in super obese patients; this could possibly be explained by increased skin-to-stone distance. Semirigid scopes may, as in PCNL, be limited by the equipment length that is more than flexible URS.
In the systematic review by Ishii and colleagues, they found similar decreases in URS efficacy in the obese patient. 6 The authors found an overall stone-free rate of 82.5%, and this decreased to 80.4% in the morbidly obese group. Furthermore, retreatments were common.
Although underweight subjects are not the target group of the current analyses, this group was unique. BMI showed a linear association with stone-free rates; however, the underweight group, as well as the overweight and obese group had higher retreatment rates. The absence of fat tissue may have a role in outcomes for the underweight group. Fat tissue in the human body plays an important role in thermoregulation, protection of internal organs, and energy storage. 14 Absence of body fat, with a subsequent decreased energy storage and buffer function, possibly explains the increase in postoperative complications for underweight patients compared with normal-weight patients. Also, the urinary tract is more prone to influences from outside the body, as the protection of internal organs is limited in underweight subjects. In addition, the adverse effect of high levels of body weight on postoperative complications may only be apparent when a certain level of BMI (i.e., obese or super obese) is met.
In the CROES URS Global study, women were well represented in both the underweight and super obese groups; whereas the proportion of men was significantly higher in the overweight and obese groups. This is comparable with findings by Negri and colleagues, 8 suggesting a different etiology, or level of BMI, in men and women to have a lithogenic influence. A different etiology is also supported by Duffey and colleagues 15 showing different excretion profiles in men and women.
Another result from the currently presented study is the association of BMI with shorter hospital stay. Since the CROES database involves multiple practitioners from all over the world and does not have the granularity that is necessary to determine the exact reason for hospital admission, we are unable to offer an explanation as to why BMI appears to be protective. It may be that with increased BMI, the frailty of the patient is perceived to be less and the surgeon is more comfortable with discharge. However, this explanation is purely speculative and further research is needed to understand the observed inverse association between BMI and length of hospital stay.
In conclusion, although URS for stone disease was found to be an overall safe procedure for obese and super obese patients, efficacy of the procedure may be lower compared with normal-weight subjects and higher retreatment rates may be necessary.
Footnotes
Acknowledgment
The URS Global Study was supported by an unrestricted educational grant from Boston Scientific.
Author Disclosure Statement
Amy Krambeck: Data Safety Monitoring Board member for Histosonics; Hemendra N. Shah: Speaker Cook; Jean de la Rosette: Consultant Angiodynamics; Consultant Boston Scientific.