The retrieved published studies were mainly one-arm, uncontrolled and observational. We separated the studies into those referring to IBD or bariatric surgery respectively.
IBD and renal stones
Incidence and clinical evaluation: No large epidemiological studies assessing the prevalence of IBD and urolithiasis co-existence were identified.
Small and historical studies initially reported this association along with other urogenital IBD complications (such as fistulation, intrinsic renal disease and obstructive uropathy)[2,7]. From smaller studies it has been estimated that 9%-18% of adult IBD patients may be diagnosed with renal stones at some time in their life[8,9]. In cases when the patients with IBD (Crohn’s) had resection of the terminal ileum, the percentage of associated urolithiasis is much higher (reported up to 28%); this is attributed to metabolic disturbances due to steatorrhoea and bile salt malabsorption. Interestingly, in patients with short bowel syndrome post several resections for complicated Crohn’s, preservation of colon has been associated with higher incidence of renal stones compared with those who had a completion colectomy (24% vs 0% in a study of 52 patients). In a study of UC patients who had a panproctocolectomy, the risk of renal stone disease was compared between conventional ileostomy and ileal J-pouch; the risks of forming uric acid stones were high for both ileostomy and J-pouch patients, but J-pouch reduced the risk of renal stones containing calcium (the relative probability of calcium stone formation was 0.58 in ileostomy group vs 0.18 in the J pouch group). Another study on patients with ileal J-pouch post panproctocolectomy for UC showed that the presence of several extra-intestinal manifestations, no use of antibiotics and low serum bicarbonate level were the most important risk factors for the presence of concurrent urolithiasis; the overall incidence of the latter was 37% in this group.
A study from Japan assessed the risk factors for developing urolithiasis in patients with CD. Renal stones were more frequent in men. The mean time from Crohn’s diagnosis to diagnosis of calculi was 8.8 years (range 0 to 22 years). The commonest type of stone was calcium oxalate. The probability of developing calculi was approximately eight times higher for patients with a urine pH of ≤ 6.0 than for those with a urine pH of ≥ 6.5.
Renal stone disease appears to be less common in the paediatric population with IBD. A report from a national IBD registry in Italy estimated its incidence to be at 0.37%.
In a study when computerised tomography was used to diagnose asymptomatic extra-intestinal manifestations of CD, 4% of patients were found to have concomitant urolithiasis.
With regards to patients’ awareness, only 12% of those included in a cross-sectional survey in Canada reported that they had been aware of the IBD-nephrolithiasis association. They had obtained their information from their gastroenterologist or the internet and they had much better knowledge on other potential extra-intestinal IBD manifestations such as risk of colon cancer (75%), arthritis (77%) and dermatological manifestations (49%).
An interesting observation on patients with IBD and urolithiasis was made in a large study (14352 patients) from emergency departments in the United States. IBD patients with urolithiasis presented with infections, sepsis and renal failure more frequently than non-IBD patients [infections (10.4% vs 9.1%; P < 0.001), sepsis (0.6% vs 0.2%; P < 0.001), and end-organ failure (6.3% vs 1.6%; P < 0.001) respectively]. This was due to the increased occurrence and severity of infected urolithiasis in this group as well as the fact that the urolithiasis was noted to occur more commonly in older patients (with IBD) compared with non-IBD controls. In another study it was noted that recurrent urolithiasis and surgical interventions for its treatment, along with bowel resections are the two independent risk factors for the development of chronic kidney disease in IBD patients; this is generally a rare phenomenon particularly in UC.
Pathogenesis: CD and UC are characterized by recurrent inflammatory involvement of different intestinal segments resulting in malabsorption of bile salts and fatty acids. This causes increased oxalate absorption by increasing oxalate solubility in the intestinal lumen and permeability of the colonic mucosa; this secondary hyperoxaluria is associated with oxalate renal stones[9,21]. In addition to that, the loss of bicarbonate in the liquid stool and the dehydration cause acidosis and subsequent reduced citrate excretion. This can lead to uric acid or mixed stones. Factors such as decreased urinary excretion of other inhibitors of crystallisation (magnesium), dietary parameters and medications (such as aminosalicylates), can also play a role in stone formation in these patients (Figure 2).
Figure 2 Risk factors for stones in patients with bowel disease.
In a study including both paediatric and adult patients with CD, the patients’ 24 h urine oxalate levels and intestinal oxalate absorption were assessed vs the relevant parameters in healthy individuals. The intestinal absorption of oxalate was significantly higher in CD patients; (0.92 ± 0.57 mmol/1.73 m2 per 24 h) compared with those without the disease (0.53 ± 0.13 mmol/1.73 m2 per 24 h) respectively (P < 0.05); this also correlated with hyperoxaluria and risk of developing urolithiasis.
Kumar et al examined the association of intestinal oxalate degrading bacteria Oxalobacter formigenes with the development of hyperoxaluria in IBD. The investigators studied stool samples of IBD patients and controls respectively for the presence of O. formigenes using polymerase chain reaction. Only 10.4% of patients with IBD compared to 56% of controls had positive for O. formigenes stool samples. Patients without O. formigenes had higher urinary oxalate than those with it. They were also more likely to develop renal stones. In this study there was no significant difference in the excretion of citrate. However, in those patients who had hypermagnesiuria (another stone inhibitor) the formation of stones was not common, irrespective of the IBD status. A recent study suggested that oxalate is not only absorbed by the gastrointestinal (GI) tract in different concentrations depending on its permeability, but can also be secreted by the small intestine. This is an interesting observation which provides some evidence that perhaps diseased GI tract can influence the oxalate metabolism and result in hyperoxaluria and subsequent urolithiasis.
Contrary to the above, a study suggested that the formation of renal stones in IBD patients is more likely related to low urinary concentration of magnesium and citrate relative to calcium than the hyperoxaluria. However, only 40 patients with IBD and 17 controls were included in the study with 2 IBD patients developing renal stones.
Evan et al analysed renal biopsies of IBD patients with urolithiasis. The histopathological abnormalities seen, included moderate glomerular sclerosis, tubular atrophy, and interstitial fibrosis. Randall’s plaque was abundant, with calcium oxalate stone overgrowth; as the authors suggested this was similar to that seen in idiopathic calcium oxalate stone formers, and primary hyperparathyroidism. Abundant plaque was compatible with low urine volume and acidic pH.
Interestingly a recent study provided early evidence of genetic background as an independent risk factor for urolithiasis in IBD patients alongside known mechanisms such as malabsorption and medication.
Another risk factor for urolithiasis in IBD patients with intestine failure (short bowel) is the use of vitamin C in parenteral nutrition. This has been shown to increase urinary oxalate excretion and subsequently increase the risk of urinary stone formation.
Ileostomy: The formation of ileostomy post bowel resection is an independent risk factor for the formation of renal stones. The relative dehydration from liquid stool, and the acidic PH from bicarbonate loss are responsible for this. In this acidic environment, the urinary citrate level excretion reduces. The stones most commonly seen in these patients contain uric acid or are of mixed composition. In addition, the risk of calcium containing stones also increases with ileostomy. In a retrospective study comparing IBD patients with ileostomy vs IBD patients with J-pouch vs controls, the relative risk of calcium containing stones formation was significantly higher (0.58 vs 0.18) in those with ileostomy.
Treatment: In our review, no study directly assessing the in vivo effect of a treatment to reduce urolithiasis in IBD patients was identified.
In a study using computer modelling, simulated urine situations based on reported composition values of IBD patients were compared with the relevant ones from normal individuals. In this model, supplementation of calcium would reduce the urinary calcium oxalate supersaturation in IBD patients as their hyperoxaluria is much more prominent compared with non-IBD stone formers. The authors suggested that calcium supplements can help reduce stone risk in those patients, but initial efforts should be directed towards reducing urinary oxalate by reducing dietary oxalate. Citrate therapy that increases both urine pH and urinary citrate could also provide an additional therapeutic benefit.
Several authors have advocated preventative strategies. These include close monitoring of IBD patients with urinalysis and imaging of the upper tracts at regular intervals as well as early surgical intervention even in cases of small renal stone burden. Also the avoidance of low urinary citrate and magnesium levels could prevent the stone formation in these patients[25,27,32]. In UC patients who had panproctocolectomy and formation of ileoanal pouch, close monitoring for renal stone formation and administration of prophylactic oral bicarbonate has been suggested. Urinary alkalinisation along with increased hydration is also advocated in IBD patients receiving aminosalicylates.
Bariatric surgery and renal stones
Incidence: Bariatric operations have traditionally been divided into three groups: (1) restrictive, i.e., procedures that produce weight loss solely by limiting intake (gastric banding, gastric sleeve); (2) malabsorptive, i.e., operations that induce weight loss totally by interference with digestion and absorption (intestinal bypass); and (3) mixed, i.e., procedures that limit intake and produce malabsorption (gastric bypass, duodenal switch).
The most commonly performed bariatric operation is Roux-en-Y gastric bypass (RYGB). It belongs to the “mixed group” in the bariatric operations classification. Nelson et al first described the association of gastric bypass surgery with the formation of calcium oxalate renal stones. Matlaga et al reviewed the insurance claims and records of 4639 patients who had RYGB for obesity and compared them with the relevant ones from other 4639 obese patients over a four-year period. Urolithiasis was diagnosed in 7.65% (355 of 4639) of RYGB patients compared to 4.63% (215 of 4639) of patients in the control group (P < 0.0001). The patients in the treatment group more commonly underwent shock wave lithotripsy and ureteroscopy, making the RYGB surgery a significant predictor of diagnosis and requirement for treatment of urolithiasis in the post-operative period. In a retrospective study of 972 patients who underwent RYGB, 3.2% developed de-novo urolithiasis post-operatively (mean FU was 2.8 years); in those individuals who had renal stones pre-operatively (8.8%), the recurrence of the disease was even more common (32%). The authors concluded that although the RYGB had a significant influence on the development of stones, the combination of pre-operative stone history with this type of bariatric surgery is the most important risk factor for recurrent urolithiasis. Other reviewers have come to the same conclusions. In a large, prospective study of 151 patients undergoing laparoscopic RYGB, Valezi et al reported de novo stone incidence of 8% at 12 mo post-surgery. The relative risk for development of stones increased by approximately 70%. In another study of patients who had RYGB for gastric cancer, the investigators assessed the incidence of renal stones post-operatively. Computed tomography scans from gastric cancer patients who had either distal gastrectomy with Bilroth I anastomosis/RYGB vs total gastrectomy with RYGB reconstruction were reviewed. Patients with total gastrectomy were more likely to have renal stones (25%) than patients with some remaining stomach (7%). The authors hypothesised that total gastrectomy may lead to more fat malabsorption than partial gastrectomy, perhaps exacerbating hyperoxaluria. Although these procedures were not directly done for the purpose of weight loss, the study adds some evidence that the malabsorptive procedures are associated with increased incidence of urolithiasis; in particular, in this study the extent of resection, and not just the bypass itself, was an independent factor for the development of renal stones.
A relative historical bariatric operation, jejuno-ileal bypass, has also been implicated with significant risk of post-operative urolithiasis. In a retrospective study of 56 patients who had undergone this surgery because of morbid obesity, twenty-two (39.3%) were found to have renal calculi in a 16-year mean follow-up. Interestingly, patients who had surgery for reversal of this bypass continued to have high incidence of renal stones despite the decrease in their 24-h urinary oxalate levels. The authors of this study concluded that the persistence of low citrate in the urine is implicated on this phenomenon.
Sleeve gastrectomy and gastric banding are the most commonly done bariatric operations of the restrictive type. Patients who undergo these procedures generally appear to have less likelihood of developing renal stones post-operatively. Chen et al reported an approximate stone incidence rate of 1% in a retrospective review of 332 cases of gastric banding and 85 cases of sleeve gastrectomy respectively; they estimated a person-time incidence rate of 3.40 stone diagnoses per 1000 person-years for gastric banding and 5.25 stone diagnoses per 1000 person-years for sleeve gastrectomy respectively. Semins et al reported no association of gastric banding surgery with increased risk of urolithiasis or renal stone intervention post-operatively. In their case-control study, the authors retrospectively reviewed the insurance claims of 201 morbidly obese patients who had gastric banding surgery and 201 morbidly obese controls through a national database. The incidence of new renal stones was 1.5% in the operated individuals and 6% in the controls; this was the first study which provided evidence that the outcome of restrictive bariatric surgery on the development of urolithiasis is different compared to malabsorptive or mixed procedures.
Pathophysiology: Bariatric surgery has been associated with several metabolic and urinary changes first described in the 1980s. The most common one is the development of hyperoxaluria, first seen post jejuno-ileal bypass surgery.
In a prospective study of 21 patients undergoing RYGB, 24-h urine specimens were analysed pre- and post-operatively; urinary oxalate excretion increased significantly after RYGB (33 ± 9 mg/d pre-operatively vs 63 ± 29 mg/d post-operatively). De novo hyperoxaluria developed in 11 (52%) patients. There was also significant increase of patients with hypocitraturia (from 10% at baseline to 48%) at 2 years. Wu et al performed a similar study; 38 patients undergoing RYGB had serum and urine chemistry assessed preoperatively and 6 mo post-operatively. Urinary changes known to increase the risk of urolithiasis were found: Decrease in total urine volume (from 2 L/d to 1.5 L/d respectively), increase in calcium (from 139 mg/d to 182 mg/d respectively), and increase in oxalate (from 38 mg/d to 48 mg/d respectively); all these changes were statistically significant. Their conclusion was that this study provided evidence that renal stones post RYGB is caused by many biochemical factors, with hyperoxaluria being the main one.
Similar conclusions were drawn by Patel et al in their case-control study of 58 patients who underwent RYGB (52) or duodenal switch (6) procedures. In this study 67.2% of patients who had bariatric surgery developed hyperoxaluria (in their 24-h urine specimens 6 mo post-operatively) compared to almost 34.1% of non-stone formers (controls) and 37% of stone formers with no history of bariatric surgery respectively. Hyperoxaluria and hypocitraturia were also demonstrated in post RYGB patients in a cross-sectional study by Maalouf et al. They also pointed out that in their study the urinary calcium excretion decreased which may be counteracting the effects of the above significant risk factors. Asplin et al provided evidence that hyperoxaluria is a common metabolic finding in patients post RYGB with calcium oxalate being the main driving force for the development of kidney stones; the levels of urinary oxalate were not as high as they found in the jejunoileal bypass cohort.
The hyperoxaluria in post bariatric surgery patients is believed to be caused by increased intestinal oxalate absorption due to the relative fat malabsorption, particularly in the bypass operations. As fat is malabsorbed, fat-soluble vitamins and calcium ions are saponified by intraluminal free fatty acids, leading to steatorrhea. The calcium is less available in the intestinal lumen, which subsequently decreases binding with oxalate. When oxalate is free, it gets absorbed by the intestine and then majority is excreted in the kidney. It may then precipitate with urinary calcium to form insoluble crystals and eventually kidney stones. In addition to the above, permeability of the intestine to oxalate is increased by exposure to unconjugated bile salts and long chain fatty acids, both of which are increased in the GI tract of RYGB patients[49,50].
A study by Froeder et al used the oxalate load test as a direct assessment of intestinal oxalate absorption in patients post gastric bypass surgery. The mean oxaluric response to this load was markedly elevated in post-bariatric surgery patients, suggesting that increased intestinal absorption of dietary oxalate is a predisposing mechanism for enteric hyperoxaluria.
In the restrictive group of bariatric operations (such as gastric banding and sleeve gastrectomy) the metabolic changes are not as common as they are in the malabsorption group. Studies comparing the metabolic profile of patients’ groups who had one procedure from each group concluded that urinary oxalate excretion of the “restrictive” cohorts was significantly less than the “RYGB” cohorts and not significantly different from that of the normal subjects and routine stone-formers[52-55]. This difference is likely related to the lower supersaturation of calcium oxalate, predominantly due to higher urinary volume and lower urinary calcium excretion in restrictive procedures compared to RYGB.
Treatment: Patients submitted to bariatric surgery are at risk of nephrolithiasis and nephropathy. Accurate stone screening, careful monitoring of renal function and diet counselling are strongly encouraged in these patients[6,45]. Close monitoring with renal tract imaging/screening is advocated mainly for patients who undergo gastric bypass surgery and have a previous history of urolithiasis.
Strategies to prevent stones after bariatric surgery are similar to those recommended to all stone formers. They include increased daily fluid intake to achieve urine volumes higher than 2 L/d and low oxalate (< 100 mg/d) intake. Additionally, low sodium and animal protein intake are indicated. Bariatric stone formers should particularly reduce daily fat intake to minimise enteric oxalate absorption and consider calcium supplementation. Citrate salts, like potassium citrate, can be used to correct metabolic acidosis and hypocitraturia[35,44].
A phase II randomised study has assessed the use of potassium calcium citrate (PCC) in post RYGB patients. PCC significantly increased citrate, and pH. The urinary saturation of uric acid decreased significantly. Also calcium oxalate agglomeration was significantly inhibited by PCC. As it has not yet been tested in phase III trials it is not available for clinical use.