Introduction

Obesity is a chronic disease characterized by accumulation of excess fat tissue, which affects body anatomy and physiology, leading to several metabolic, biochemical, and psychosocial complications [1]. It is a global health issue affecting 10.8% of men and 14.9% of women worldwide in 2014 [2], and it is expected to rise by 40% over the next decade [3]. Computerized tomography (CT) is considered to be a reference method for assessing abdominal adiposity [4]. Visceral fat, which has been reported to be strongly associated with albuminuria [5] and future cardiovascular events [6, 7], can be measured independently of the subcutaneous fat on CT images.

Obesity has been demonstrated to be associated with a higher risk of chronic kidney disease (CKD), end-stage renal disease (ESRD), nephrolithiasis, and renal cell cancer [8,9,10,11]. The increased risk for CKD and ESRD can be explained by various mechanisms; obesity-induced hypertension, insulin resistance, renin-angiotensin aldosterone system activation, dysregulation of adipocytokines, and inflammation [12, 13]. These changes result in glomerular hyperfiltration with subsequent albuminuria/proteinuria [14, 15]. With the progression of obesity-induced kidney damage, hyperfiltration is followed by decreasing glomerular filtration rate (GFR) and progression toward ESRD [14, 15].

Bariatric surgery (BS) is considered the most successful treatment for obesity resulting in significant and sustained loss of weight [16]. It has been reported that BS ameliorates many obesity-related comorbidities [17], and has a beneficial effect on the renal functions [18]. However, BS is considered as a risk factor for renal stones. Approximately 20% of patients develop kidney stones 10 years after operation [19]. The key mechanism of renal stones formation is hyperoxaluria and calcium oxalate super-saturation [20].

Most of the data of postoperative hyperoxaluria after BS originated from research on roux-en-Y gastric bypass (RYGB) surgery [21,22,23]. We assumed that there would be different renal effects of sleeve gastrectomy (SG) and one anastomosis gastric bypass (OAGB), particularly on postoperative hyperoxaluria. Accordingly, this research was conducted to determine the short-term (6 months) renal effects of BS and to compare such results between SG and OAGB. In addition, the association of abdominal fat compartments, measured by CT, with renal outcome was investigated.

Methods

Patients

This is a single-center non-randomized prospective observational study. During the period between September 2018 and June 2019, BS patients ≥ 18 years of age were recruited from the Gastrointestinal Surgery Center, Mansoura University, Egypt. Patients with CKD (eGFR < 60 mL/min/1.73 m2), taking medications that may affect GFR or reduce proteinuria as angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) or who are not willing to participate in the study were excluded. Sex, age, and associated comorbidities have been recorded for the included patients. The next assessments were performed prior to BS and repeated 6 months after surgery.

Surgical Interventions

Two procedures, laparoscopic SG (LSG) [24] and laparoscopic OAGB [25], were primarily performed for weight loss. The selection of LSG versus OAGB was based mainly on the patient’s eating habits and lifelong commitment to multivitamin use and follow-up visits (favoring OAGB for sweet eaters and compliant patients). However, patient’s choice was taken into consideration. The biliopancreatic limb in OAGB patients was 200 cm as measured from the ligament of Treitz.

Nutrition After BS

All patients were followed regularly by a nutrition consultant. In the first two weeks after surgery, patients were instructed to take enough fluids at moderate temperatures devoid of sugar or fat. They were later advised to eat well-ground soft food with a high protein content and low sugar, fat, and oxalate content while taking enough fluid.

Anthropometric Measurements

Body mass index (BMI), waist circumference [26], hip circumference [26], waist to hip ratio [27], and neck circumference [28] were measured. Body adiposity [29] and visceral adiposity [30] indices were calculated.

Laboratory Investigations

The following investigations were conducted for all patients: serum creatinine, urine analysis, 24-h urinary creatinine clearance (CLCr), protein and oxalate excretion, lipid profile, and HbA1c.

Abdominal Fat Tissue Analysis

Abdominal obesity was evaluated by CT using a 3D synapse program. Two-dimensional (2D) and three-dimensional (3D) views were assessed. This technique allows assessment of both subcutaneous and visceral fat surface area (cm2) and volume (cm3). The 2D technique used the specific range of Hounsfield units (HU) between − 250 and − 30 at the level of umbilicus or at the level of disc between lumbar 4 and 5 vertebra [31]. The 3D technique used the specific range of HU between − 250 and − 30 all over the abdomen, and the fat volume could be assessed in voxels and then translated into cubic centimeter for both visceral and subcutaneous fats [31].

Sample Size and Statistical Analysis

Sample size estimation was based on figures derived from the study of Chang et al. [32] Using a statistical power of 90% and two-tailed significance level of 5%, the minimum required sample size was 52 patients. Considering a possible dropout of 10%, the sample size was increased to 57 patients. Data were collected, revised, verified, and analyzed using the Statistical Package of Social Sciences (SPSS) version 21 for Windows (SPSS, Inc., Chicago, IL, USA). Medians and interquartile range (IQR) or mean ± standard deviation (SD) were used for all quantitative values, while numbers of cases and percentages (%) were used to describe qualitative variables. The significance of differences between two groups was determined by independent-sample t test for normally distributed variables, Mann-Whitney test for non-parametric variables, and Chi-square or Fisher exact tests for qualitative variables. Multiple logistic regression analysis was done using enter approach. All the potential variables that can predict the postoperative hyperfiltration were analyzed in 5 models. p values of < 0.05 were considered significant for all statistical analyses in this study.

Results

A total of 57 patients suffering from severe obesity who received BS were included. LSG was performed in 47 participants, and OAGB surgery was performed in 10 participants. The mean age was 36.7 years, and 75.4% of patients were females. BS resulted in marked reduction in the studied anthropometric measurements 6 months after surgery, and comparable anthropometric effects were observed in the 2 forms of BS performed (Table 1).

Table 1 Clinicodemographic characteristics and anthropometric measurements
Full size table

Six months after BS, substantial reduction was observed in the 24-h CLCr, urinary protein excretion, and serum triglyceride level. HbA1c percentage and serum cholesterol level were modestly reduced. In the LSG group, the magnitude of HbA1c% and the total cholesterol level reduction was lower (p < 0.001 and = 0.03, respectively). The effect of BS on urinary oxalate excretion was contradictory between the two types of performed BS. LSG resulted in a modest decrease, whereas OAGB resulted in more than one-fold increase in urinary oxalate excretion (p < 0.001) (Table 2).

Table 2 Laboratory investigations before and 6 months after both operations
Full size table

A marked reduction in total, subcutaneous, and visceral abdominal fat tissue surface area and volume, assessed by CT, was observed 6 months after BS. There was no substantial difference of impact of BS type on abdominal fat tissue measurements (Table 3 and Figs. 1, 2, 3, and 4).

Table 3 Radiological abdominal fat tissue analysis
Full size table
Fig. 1
figure 1

Abdominal CT axial view (2D) in a LSG patient: (a) before operation; (b) 6 months after operation. Subcutaneous (blue) and visceral (red) fat surface areas (cm2) are represented in the box at the left lower corner of each image

Full size image
Fig. 2
figure 2

Abdominal CT (3D) in a LSG patient: (a) before operation; (b) 6 months after operation. Subcutaneous (blue) and visceral (red) fat volumes (cm3) are represented in the box with red borders at the right half

Full size image
Fig. 3
figure 3

Abdominal CT axial view (2D) in a laparoscopic OAGB patient: (a) before operation; (b) 6 months after operation

Full size image
Fig. 4
figure 4

Abdominal CT (3D) in a laparoscopic OAGB patient: (a) before operation; (b) 6 months after operation

Full size image

Glomerular hyperfiltration (CLCr > 130 mL/min/1.73 m2) was evident in 54 patients before surgery. The remaining 3 patients with normal preoperative GFR belonged to the OAGB group. In 29 of those 54 patients, glomerular hyperfiltration continued six months after the operation, while the GFR in the remaining 25 patients became in the normal range. Persistent glomerular hyperfiltration was more common in male patients (p = 0.01) and those with obstructive sleep apnea syndrome (p = 0.02), with no major difference between LSG and OAGB. Patients with postoperative glomerular hyperfiltration had higher preoperative body weight and surface area, neck and waist circumferences, and higher total and subcutaneous abdominal fat surface area and volume (p = 0.01–0.03) (Table 4).

Table 4 Comparison between different preoperative data regarding change of creatinine clearance after the surgery
Full size table

Six models of multivariate logistic regression, including mix of demographic, anthropometric, laboratory, and radiological variables, were developed to predict the persistence of glomerular hyperfiltration after BS. Among the studied variables, only the surface area of subcutaneous abdominal fat tissue, assessed by CT, was found to be an independent predictor for the persistence of glomerular hyperfiltration after BS (Table 5).

Table 5 Regression analysis for predictors of postoperative hyperfiltration
Full size table

Discussion

Obesity is a clearly recognized global public health problem [2]. BS is one solution to this worrisome issue that has increased in popularity as it results in persistent weight loss and improvement in hard-end points, such as diabetes, sleep apnea, and even death [33]. Moreover, BS has been proved to have favorable renal effects [20]. SG and RYGB are the two leading procedures worldwide [34]. Over the last few years, OAGB has also gained worldwide acceptance [34] based on several advantages, such as short operating times, low morbidity and mortality rates [35], as well as sustained weight loss [36] or type 2 diabetes remission results equal to or even greater than RYGB [37]. To the best of our knowledge, the disparity in renal effects between SG and OAGB has not been discussed before.

Six months after BS, all anthropometric measurements decreased significantly, with no major variations between the two forms of operations performed. In accordance with these results, Kular and colleagues found in the first 2 years after BS, weight loss was similar in OAGB and LSG. Afterwards, however, the LSG displayed a lower proportion of weight loss than the OAGB operation [38]. Another study showed a greater weight loss 1 year following OAGB operation [39].

In the current study, the mean CLCr significantly decreased from 224 to 145.4 mL/min/1.73m2 and 25 of 54 patients with preoperative glomerular hyperfiltration had normal CLCr 6 months after BS. Other studies in patients with preserved kidney function showed similar findings after BS [40,41,42]. A meta-analysis reported a mean GFR reduction of 25.6 mL/min/1.73 m2 after BS [43]. The higher magnitude of CLCr reduction in the present study can be explained by the clearly higher preoperative CLCr caused by the very high BMI (56.1 ± 9.64 kg/m2), the relatively young age of the study participants (36.7 ± 8.88 years) and the exclusion of patients taking drugs that may affect GFR, such as ACE inhibitors or ARBs. There was no difference between the two operations, either statistical or clinical, regarding the reduction of CLCr. Only subcutaneous abdominal fat tissue surface area was found to be an independent indicator for the GFR response after BS. Contrary to this finding, a study conducted by Lee et al. included 138 patients with severe obesity who had different BS procedures (RYGB, SG, and biliopancreatic bypass with duodenal switch), indicated a better resolution of glomerular hyperfiltration in female gender, lower ages, and lower preoperative BMI [44]. However, Lee and colleagues assessed the kidney function through estimated GFR (eGFR) (CKD-Epidemiology equation) and not the measured creatinine clearance, and they performed different types of BS procedures.

The median urinary protein excretions significantly decreased from 220 to 112 mg/day after BS. Several studies and meta-analyses have shown a favorable effect of BS on reducing proteinuria or albuminuria after surgery. Whether this is a direct effect of weight loss or mediated by better blood pressure and insulin resistance is unclear [18, 45]. There was no statistically relevant difference between the two operations as regards reducing the excretion of urinary protein.

Bariatric surgery has a range of renal complications, including perioperative acute kidney injury (AKI), long-term nephrolithiasis, and oxalate nephropathy [20]. AKI is relatively common after BS, with estimates varying from 2.9 to 8.5% using different definitions of AKI [46, 47]. Risk factors for AKI after BS include higher BMI, lower eGFR, preoperative use of ACE inhibitors, ARBs, and intraoperative hypotension [20]. In view of good hydration in the perioperative period and regular follow-up by a nutrition consultant, no single case of postoperative AKI occurred in this study up to six months after surgery.

The risk of kidney stones can increase following some forms of BS, which tends to be associated with the degree of fat malabsorption achieved [20]. Steatorrhea is believed to induce hyperoxaluria by increasing the formation of calcium fatty acid salts, leading to a decrease in the binding of calcium to oxalate and ultimately an increase in the absorption of oxalate [23, 48]. In the current research, there was a substantial reduction after LSG of 24-h urinary oxalate excretion and a substantial increase in 24 h urinary oxalate excretion following OAGB. It is the only research to the best of our knowledge that associated the OAGB procedure with the excretion of urinary oxalate in humans. The above-mentioned biochemical changes are similar to those reported by DeFoor and his colleagues who postulated that SG may have stone formation protective mechanisms due to lower levels of urinary oxalate excretion compared with RYGB patients and control group with obesity [49]. In addition, OAGB operation was associated with increased urinary oxalate excretion and elevation of the calcium oxalate super-saturation risk in an experimental animal study in rats [50].

This study has limitations. First, the single-center nature of the study. Second, the short duration of follow-up. Third, the different number of operated patients in the two groups, as during this observational study period, the selection criteria for BS type and patient’s choice were in favor of LSG. Fourth, the amount of fat and oxalate in diet was not quantified. However, this is the first comparative research, to the best of our knowledge, between OAGB and SG that focuses primarily on the renal outcome in patients with severe obesity. Moreover, detailed assessment of the abdominal fat amount and distribution by CT and correlating them to the renal outcome are considered strengths of this study.

Conclusion

In conclusion, BS is an effective treatment of severe obesity and can relieve many of the obesity-induced renal disorders. Nevertheless, hyperoxaluria remains a serious problem following BS, particularly in malabsorptive procedures, such as OAGB.