Weight Loss Surgery Increases Kidney Transplantation Rates in Patients With Renal Failure and Obesity

Abstract

Objective:

To describe the outcomes of kidney transplantation (KT) candidates with obesity undergoing sleeve gastrectomy (SG) to meet the criteria for KT.

Patients and Methods:

Retrospective analysis using electronic medical records of KT candidates with obesity (BMI>35 kg/m²) who underwent SG in our institution. Weight loss, adverse health events, and the listing/transplant rates were abstracted and compared to the non-surgical cohort.

Results:

SG was performed in 54 patients; 50 patients did not have surgery. Baseline demographics were comparable at the time of evaluation. Mean±SD BMI of the SG group was 41.7±3.6 kg/m² at baseline (vs. 41.6±4.3 kg/m² for controls), while 36.4±4.1 and 32.6±4.0 kg/m² at two and twelve months after SG (P<.01 for both). In the median follow-up time of 15.5 months (IQR 6.4, 23.9), SG was followed by active listing (37/54 people), and 20/54 received KT over a median follow-up time of 20.9 months (IQR 14.7, 28.3) after SG. In contrast, 14/50 patients in the non-surgical cohort were listed, and 5/50 received a KT (P<.01). Three patients (5.6%) experienced surgical complications. There was no difference in overall hospitalization rates and adverse health outcomes, but the SG cohort experienced a higher risk of clinically significant functional decline.

Conclusions:

SG appears to be effective in KT candidates with obesity, with 37% of patients undergoing KT over the next 18 months (P<.01). Further research is needed to confirm and improve the safety and efficacy of SG for patients with obesity seeking a KT.

INTRODUCTION

Patients with obesity represent the fastest growing population with advanced chronic kidney disease (CKD) defined as stages 4, 5, and 5D (eGFR 15–29 mL/min/1.73 m², eGFR<15 mL/min/1.73 m², and dialysis, respectively¹). Approximately 40% of patients on dialysis in the USA meet the criteria for obesity² (body mass index [BMI] >30 kg/m²)³.

Effective therapy for obesity is understudied in this high-risk population. This is likely because high BMI is generally viewed as beneficial, especially in patients on dialysis, due to the paradoxical association of having better survival with a higher BMI.⁴ Treatment of obesity is therefore generally limited to KT candidates who need to achieve the transplant centers’ prespecified BMI goals for KT waitlisting.

Among the weight loss surgeries, laparoscopic sleeve gastrectomy (SG) has been the most commonly performed procedure in patients seeking KT. SG has been shown to be effective in inducing rapid weight loss, reducing obesity-related comorbidities, and improving access to KT; however, safety concerns remain.⁵ Limited literature exists regarding the risks and benefits of SG in patients with CKD stage 4-5D, mostly focused on the efficacy of weight loss, short-term surgical complications, and an improvement in KT rates.⁶ In contrast, the other assessments including cardiometabolic risks and functional status post SG are underreported.

Therefore, the main goals of this clinical retrospective cohort study were 1) to analyze the outcomes of SG in terms of weight loss, improvement in cardiometabolic risks, and access to KT and 2) to assess the risks of surgical complications, hospitalizations, frailty, and adverse health outcomes following SG in comparison to non-surgical cohort in a high volume transplant center performing SG in KT candidates with class 2 and 3 obesity (BMI 35–39.9 kg/m² and ≥40 kg/m²).⁷

METHODS

Study design and population

This retrospective study was approved by the Mayo Foundation Institutional Review Board. We analyzed data on KT candidates who participated in our Transplant Metabolic (TRANSMET) program from its inception on February 1st, 2020, to August 31st, 2023 and underwent SG at our center. TRANSMET program was developed to offer an accelerated path to KT by integrating weight management into the transplant evaluation and ongoing care model.

In our multidisciplinary treatment program, all patients presenting for KT evaluation who have CKD stage 4-5D and a BMI>35 kg/m2 are automatically offered participation in the TRANSMET program. Patient are referred for kidney transplantation by their nephrologists or are self-referred. Integrated and streamlined pretransplant and weight loss evaluation allows KT candidates to learn the benefits of transplantation and importance of weight loss for KT eligibility and an improvement of metabolic health. The details of our program were previously described.⁷ TRANSMET program includes pre-transplant evaluation, and multidisciplinary weight management consultations (transplant nephrologist, transplant surgeon with bariatric expertise, registered dietician, bariatric endocrinologist and bariatric psychologist). High risk psychosocial candidates are seen by transplant and addiction psychiatrists with expertise in selection and follow up for patients for bariatric surgical procedures. Patients at high risk for frailty (older, with diabetes or on dialysis) are offered physical therapy (PT) consultation. PT assessment in KT candidates includes routine reporting of Short Physical Performance Battery (SPPB) score⁸ and Fried frailty phenotype (FP).⁹ Clinically meaningful change in the frailty assessment was defined as “a change that has clinical or practical importance, has an impact on an individual’s self-perceived health status or quality of life”.⁸ Clinically meaningful change is defined as a decline by more than 1 point on SPPB score⁸, and 0.623 for FP.⁹

SG is offered to all patients who are enrolled in the TRANSMET program unless there are psychiatric (e.g., untreated depression or active substance use disorder) or medical contraindications or the lack of insurance coverage for weight management. Patients who undergo SG have subsequent structured, comprehensive follow-ups by the multidisciplinary team. They are either contacted or seen by the team at 2 weeks (remote), 1 month (remote), 2 months (in-person), 6 months (in-person), 12 months (in-person) post SG and yearly thereafter (in-person). Assessment includes weight, nutrition, physical performance, glucose metabolism, blood pressure control and kidney function in patients not on dialysis. Patients on peritoneal dialysis are converted to hemodialysis prior to surgery and subsequently may be converted back to HD after at least 6 weeks post-surgery.

Demographics and clinical characteristics

Baseline demographics, collected from the electronic medical records at the time of transplant evaluation, included age, sex, weight, BMI, history of T2D and treatment, history of hypertension and treatment and dialysis status. The diagnosis of T2D was verified by electronic chart review and based on clinical diagnosis at the time of referral to the transplant team, the use of antihyperglycemic agents and C-peptide levels. Patients with type 1 diabetes (4 patients post-SG and 4 non-SG) were excluded due to the different pathophysiologic mechanism and response to SG¹⁰ and will be reported separately.

For the SG cohort, follow up data included weight at 2, 6 and 12 months post-SG, medical and surgical complications, hospitalizations rates, antihypertensive and antidiabetic therapy, incidence of hypotension, hypoglycemia (symptomatic hypoglycemia reported by patients such as sweating, shakiness, anxiety, etc. in the presence of blood glucose < 70 mg/dL¹¹).

Initially, the general functional assessment was done at 2 months post-SG and included evaluation of strength or capacity to tolerate loads, core strength, capacity/tolerance for exercise/activity, deconditioning, gait mechanics, and functional mobility. Subsequently, our clinical team implemented a more formal assessment using SPPB score and FP one day prior to and 2 months post-SG.

Non-surgical cohort included patients who opted for a nonsurgical weight-loss approach. Patients were provided with a referral in their home region for ongoing professional support for weight management. They remained in contact with the transplant coordinator for the reassessment of weight loss efforts and KT eligibility. We collected weight and BMI at 1 year post the initial evaluation. In both cohorts, we collected history of five-point major adverse cardiovascular events (MACE; defined as death, myocardial infarction, coronary revascularization, stroke, and heart failure), dialysis related complications, hospitalizations, and infections from the initial evaluation.

Statistical analysis

Data were summarized using mean and standard deviation for continuous variables and count and percent for categorical variables. There were no missing data. Variables with highly skewed data were summarized with median and interquartile range. Differences between baseline groups were tested using analysis of variance for continuous variables or chi-square tests for categorical variables. Non-parametric tests were used when data was not normally distributed.

Longitudinal trends in weight were modeled using a ‘broken-stick’ or linear spline model with a fixed knot at the day of bariatric surgery.¹² Briefly, weight is modeled linearly over time with an interaction term with surgery status. In this way, the overall weight trend is captured as well as the change in the slope once surgery has occurred. Random effects are used for patients to account for within patient correlations. For graphical demonstrations of the effect of SG, we show the slope prior to surgery and then the deviation in this slope post-surgery. Patients were ‘centered’ in time around their date of surgery for those that received a surgery and 130 days post-evaluation for those that didn’t. This represents the mean time from evaluation to surgery. In terms of the frailty reporting, we used descriptive statistics, an absolute change of SPPB scores per patient at the same time points pre- and post-SG when available, as well as broken stick model to collectively study and compare both cohorts.

Time-to-event endpoints were summarized with Kaplan-Meier methods. Bariatric evaluation defined the index date for all time-to-event and longitudinal analyses. SG was treated as a time-dependent covariate along with baseline variables in Cox regression modeling. Since a time-dependent grouping variable was used in the analysis, a direct Kaplan-Meier plot is not appropriate (group belonging is not defined at baseline but changes over time). Instead, the variable effect was demonstrated visually using incidence summaries from Cox models assuming SG occurred at 90 days post-evaluation.

Medication usage (yes/no) and dosage were modeled using generalized linear mixed effects models with binomial and Gaussian linkages, respectively. The STROBE checklist was used to guide the submission¹³ (Supplement).

RESULTS

Study population

There were 106 KT candidates with obesity who participated in the TRANSMET program, of whom 56 patients underwent SG. Of 56 patients, 2 (3.7%) underwent simultaneous SG and KT and were subsequently excluded, resulting in the final cohort of 54 patients in the SG group and 50 in the nonsurgical cohort who declined SG (104 in total). Median follow-up time for both groups was 15.5 months (interquartile range, 6.4 to 23.9 months). Table 1 describes the demographic characteristics of all patients (SG and nonsurgical cohort). Most patients were male (60 [57.7%]) and White (75 [72.1%]), with a mean age of 52.2 +/−11.1 years and on dialysis 63 [60.6%] at initial evaluation. Type 2 diabetes and hypertension were highly prevalent, affecting 58 [55.8%] % and 93 [89.4%] of patients, respectively. There were no significant differences in baseline demographic characteristics or BMI, but the incidence of T2D and lifetime MACEs (Table 2) was numerically higher in the SG cohort.

Table 1:

Baseline demographics, weight change and kidney transplant rates

	No sleeve gastrectomy N=50	Sleeve gastrectomy N=54	Total N=104	P value
Mean age at baseline±SD [years]	52.9±11.7	51.5±10.3	52.2±11	.52
Gender, N (%)				.65
Male	30 (60)	30 (55.6)	60 (57.7)
Female	20 (40)	24 (44.4)	44 (42.3)
Race, N (%)				.66
White	36 (72)	39 (72.2)	75 (72.1)
Black	7 (14)	6 (11.1)	13 (12.5)
Asian	4 (8)	2 (3.7)	6 (5.8)
American Indian or Alaskan Native	1 (2)	5 (9.3)	6 (5.8)
Native Hawaiian	1 (2)	1 (1.9)	2 (1.9)
Other	1 (2)	1 (1.9)	2 (1.9)
Ethnicity, N (%)				.56
Not Hispanic or Latino	46 (92)	50 (92.6)	96 (92.3)
Hispanic or Latino	4 (8)	3 (5.6)	7 (6.7)
Unknown/not reported	0 (0)	1 (1.9)	1 (1)
Cause of ESRD, N (%)				.28
DM	13 (26)	26 (48.1)	39 (37.5)
HTN	10 (20)	3 (5.6)	13 (12.5)
FSGS	5 (10)	3 (5.6)	8 (7.7)
Glomerulonephritis	9 (18)	9 (16.7)	18 (17.3)
IgA nephropathy	6 (12)	5 (9.3)	11 (10.6)
Membranous GN	2 (4)	2 (3.7)	4 (3.8)
SLE	1 (2)	2 (3.7)	3 (2.9)
ADPKD	4 (8)	7 (13)	11 (10.6)
Other	9 (18)	6 (11.1)	15 (14.4)
Dialysis at time of evaluation, N (%)	28 (56)	35 (64.8)	63 (60.6)	.36
HD	19 (67.9)	25 (71.4)	44 (69.8)
PD	9 (32.1)	10 (28.6)	19 (30.2)
Mean time on dialysis at baseline±SD [years]	2.3±2.8	2.2±1.7	2.2±2.2	.81
Comorbidities, N (%)
HTN	46 (92)	47 (87)	93 (89.4)	.41
T2D	25 (50)	33 (61.1)	58 (55.8)	.25
Insulin therapy in T2D	14 (28.6)	25 (48.1)	39 (37.5)	.48
Mean BMI at baseline±SD [kg/m2]	41.5±4.3	41.7±3.6	41.6±3.9	.88
Mean BMI at 12 months±SD [kg/m2]	39.8±4.8	32.6±4.0	35.4±5.6	<.01
Kidney transplant, N (%)	5 (10)	20 (37)	25 (24)	<.01

ADPKD – autosomal dominant polycystic kidney disease, ATN – acute tubular necrosis, DM – diabetes mellitus, ESRD – end-stage renal disease, FSGS – focal segmental glomerulosclerosis, GN – glomerulonephritis, HD – hemodialysis, HTN – arterial hypertension, PD – peritoneal dialysis, SD – standard deviation, SLE – systemic lupus erythematosus, T2D – type 2 diabetes

Table 2:

Complications comparison of no sleeve gastrectomy vs sleeve gastrectomy patients

	No sleeve gastrectomy N=50	Sleeve gastrectomy N=54	Total N=104	P value
Major adverse cardiovascular event over the lifetime, N (%)	14 (28)	19 (35.1)	33 (31.7)	.43
Stroke	4 (8)	6 (11.1)	10 (9.6)
Myocardial infarction	2 (4)	4 (7.4)	6 (5.7)
Heart failure	8 (16)	9 (16.6)	17 (16.3)
Major adverse cardiovascular event after evaluation, N (%)	1 (2)	5 (9.2)	6 (5.7)	.16
Stroke	1 (2)	2 (3.7)	3 (2.9)
Myocardial infarction	0 (0)	2 (3.7)	2 (1.9)
Heart failure	0 (0)	1 (1.9)	1 (1)
Dialysis related, N (%)	3 (6)	2 (3.7)	5 (4.8)	.22
Catheter related infection, Fistula thrombosis, Catheter malfunction
Surgery related, N (%)	NA	3 (5.6)	3 (5.6)	NA
Postoperative bleeding requiring blood transfusion.		2 (3.7)	2 (3.7)
Gastrointestinal perforation with anastomotic leak		1 (1.9)	1 (1.9)
Hospitalization, N (%)	9 (18)	12 (22.2)	21 (20.2)	.62
Cardiovascular complication	1 (2)	5 (9.3)	6 (5.7)
Respiratory complication	1 (2)	0 (0)	1 (1)
Fluid and electrolyte disturbances	0 (0)	1 (1.9)	2 (1.9)
Urinary tract infection	1 (2)	2 (3.7)	3 (2.9)
Anemia/bleeding	1 (2)	2 (3.7)	3 (2.9)
Wound infection	0 (0)	1 (1.9)	1 (1)
Other	5 (10)	1 (1.9)	6 (5.8)
Infection, N (%)	8 (16)	12 (22.2)	20 (19.2)	.73
Urinary tract infection	5 (10)	11 (20.3)	16 (15.4)
Other	3 (6)	1 (1.9)	4 (3.8)

Weight loss

At the time of evaluation, the mean weight of the SG group was 122.5±18.7 kg (median, 120.0 kg [108.2 to 135.0 kg]), and BMI was 41.7±3.6 kg/m² (median, 41.3 kg/m² [39.3 to 43.6 kg/m²]). Mean BMI subsequently decreased to 36.4 ±4.1 kg/m² 34.7±4.1 and 32.6±4.0 kg/m² at 2, 6, and 12 months, respectively, (P<.01). Total mean weight loss reported at 2, 6 and 12 months was 16.1±8.4 kg (13.1 %), 21.3±11.9 kg (17.1 %) and 26.5±12.8 kg (21.3 %) respectively, (P<.01). The median weight loss was 14.3 [11.0; 19.7] kg, 22.0 [15.1; 28.1] kg and 25.1[17.3; 30.8] kg at the same time points. Excess weight loss (calculated using an ideal body weight of BMI 25 kg/m²) at 2, 6 and 12 months was 22.1%, 28.8% and 38% respectively (P<.01). In the non-surgical cohort, the BMI changed from 41.5±4.3 at baseline to 39.8 ±4.8 kg/m² at 1 year post bariatric evaluation.

Figure 1A illustrates the weight trend in patients who did and did not have SG at the last follow-up time point. Based on a linear mixed effects model, average weight decreased by 0.45 kg/month (P<.001) for those that did not have SG (including those that never had surgery as well as those that had it in the future). After having SG, weight on average decreased by 1.1 kg/month (P<.001). Overall, patients with T2D were less likely to lose weight following the SG than patients with no diagnosis of T2D (Figure 1B; P<.001). There was no significant difference in weight loss in the dialysis vs no dialysis group (Figure 1C).

Figure 1: Change in total weight from time of evaluation to last follow-up.

Prior to surgery all patients are considered to be part of the ‘No SG’ group. The effect of surgery is then a change in the slope for those that received surgery compared to that of the overall cohort. For visual demonstration, patients’ timelines were centered on either the time of bariatric surgery for those who received surgery or 130 days post-evaluation for those who didn’t receive bariatric surgery.

A: All patients

B: Patients with type 2 diabetes versus no type 2 diabetes

C: Patients on dialysis versus no dialysis

At the time of evaluation for bariatric surgery, 23 patients were on glucagon-like peptide-1 receptor agonists (GLP1RA) medications (12 that did not have bariatric surgery and 11 that did). GLP1RA was subsequently restarted in 4 patients post-SG at variable time points post-surgery.

Transplant waitlisting and transplant rates post-SG

In total, 51 patients were waitlisted during the study period. Of those, 37 (72%) had SG and 14 (28%) did not have SG. Undergoing SG improved time-to-waitlisting (HR 3.84; 95% CI [1.81, 8.14]; P<.01); (Figure 2A). Similarly, SG also improved propensity of getting a transplant with 20 post-SG patients (37%) receiving a KT versus 5 (10%) of non-surgical cohort (HR 3.03; 95% CI [1.11, 8.26]; P=.01). Mean BMI at transplant in SG group was 33.7±4.5 and 37.6±1.1 kg/m² in no SG group.

Figure 2: Time to adverse health outcomes from evaluation to the last follow-up.

Incidence plots are predicted curves from Cox regression for a hypothetical patient who received surgery at 90 days post-evaluation. These are used as an alternative to the standard Kaplan-Meier plot, given that SG status is not defined at baseline but changes over time.

A: Time to waitlist for a kidney transplant. Incidence is given as curves predicted from a Cox regression model that uses SG status as a time-dependent covariate.

B: Time to first hospitalization

C: Time to the first major adverse cardiovascular event from evaluation to the last follow-up

D: Time to first dialysis complication from evaluation to the last follow-up

Major adverse health outcomes

Surgical complications and other health events are described in Table 2. Figure 2B, 2C and 2D demonstrate incidence of hospitalizations, MACE events and dialysis complications respectively, predicted from Cox models with SG status as a time-dependent covariate.

Post-SG surgical complications

Three patients (5.6%) experienced post-surgical complications, including postoperative bleeding requiring blood transfusion (2 patients) and gastrointestinal perforation with anastomotic leak (1 patient). Postsurgical complications did not affect KT candidacy in patients with bleeding, but delayed listing in patient with anastomotic leak due to prolonged hospitalization and resulting frailty.

Hospitalizations in SG and non-surgical cohorts

The rate of hospitalizations during follow-up time was high, regardless of SG status: 12 (22.2%) and 9 (18%) in patients who did and did not undergo SG, respectively (HR 1.38; 95% CI [0.55, 3.47]; P=.49). Twenty patients (19.2%) experienced one or more infections in total, 8 patients (16%) in non-SG group and 12 patients (22.2%) in SG group, with the most common being urinary tract infection. One patient experienced AV fistula thrombosis and one developed dialysis related infection following SG, while in the non-surgical cohort there were catheter related malfunctions in 2 patients and dialysis related infection in 1 patient.

MACE

The total number of MACE occurring at any time during the lifetime was 19 in SG and 14 in the non-SG cohort. Six patients (5.7%) experienced MACE during the follow-up time since evaluation; five in the SG group in which one event was before surgery while four events were after surgery (with one event, stroke, occurring within 30 days of surgery) and 1 in the non-surgical cohort (P=.16). Most common MACE was stroke (3 patients), followed by myocardial infarction (2 patients) and heart failure (1 patient).

Functional status in patients undergoing SG.

Early in our TRANSMET program, overall functional status was assessed and described by PT at two months post-SG, with no pre-SG assessment. Thirty-four patients underwent PT evaluation at the mean time of 62.5±13.4 days post-SG. Among these patients, 12 (35%) had severe functional impairment/debility, 11 (32%) had moderate functional impairment, and 16 (47%) had no evidence or mild functional impairment. Later, we implemented pre- and post-SG assessments that included SPPB and FP indices. In patients who had pre- and post-SPPB frailty assessment, the score changed from 10.8 (1.8) at evaluation to 8.6 (4.6) at 12 months, indicating clinical significance⁸, but was not statistically significant (P=.0.15). Similarly, FP increased from 0.16±0.4 to 0.75±1.5 indicating clinical significance⁹ without reaching statistical significance (P=.39). Next; we collectively analyzed all SBBP scores and FPs that were ever performed in the SG group and non-surgical controls pre- and post-evaluation. Overall, SPPB score declined, and FP increased in patients post-SG compared to those who did not undergo SG, but both changes did not reach statistical significance. For the SPPB score using all measurements over time, the average patient had an increase in SPPB of 0.0009 per month. After SG, this slope decreased to an average change of −0.0775 per month (p = 0.82). For the FP score, the average patient had a decreasing score of −0.0910 per month. For those who received SG, this frailty index increased to 0.0328 per month on average post-surgery (p = 0.44).

Impact of SG on cardiometabolic profile

T2D

Twenty five of 33 (75.8%) patients with T2D were on insulin, and 32% discontinued insulin at 12 months post-SG. The mean long-acting insulin dose decreased from 12.7±21.6 IU at the time of SG evaluation to 2.8±7.4 IU (P=.05). Symptomatic hypoglycemia in patients with a history of T2D affected 10 patients (30%).

Hypertension

The proportion of patients on antihypertensive medications decreased between the time of evaluation (88.5%) and 12 months post-SG (77.8%, P=.05). Mean number of antihypertensive medications changed from 2.8±1.1 to 1.6±1.2 (P<.01). The subgroup taking 3 or more medications decreased from 61% at the time of evaluation to 26%. Ambulatory orthostatic hypotension was experienced by 19 patients (35.2%) in the first year post-SG. Four of these patients were started on midodrine. During the same follow-up period, dizziness was present in 20 patients (37%).

DISCUSSION

The results from this clinical cohort date-of-treatment matched control study found that SG is an effective weight loss procedure in KT candidates, but we have concerns about the impact of SG on functional status. The mean total weight loss in the SG group was 13.1%, 17.1%, and 21.3% at 2, 6, and 12 months, respectively (median [interquartile range]: 12.6% [9.3% to 16.7%], 16.9% [13.6% to 23.7%], and 18.8% [16.4% to 28.2%] at the same points). Patients on dialysis experienced a similar weight loss as MAYO CLINIC PROCEEDINGS 712 Mayo Clin Proc. n May 2024;99(5):705-715 n https://doi.org/10.1016/j.mayocp.2024.01.017 www.mayoclinicproceedings.org their counterparts with no dialysis. The risk of postsurgical complications was low overall (5.6%), and the rate of hospitalizations and infections were similar to that of the non-SG group. Patients who underwent SG were significantly more likely to be listed on inactive status on the KT list and, most important to receive a KT.

We have previously shown that conservative weight approaches do not adequately induce weight loss in patients with CKD stage 4-5D, as opposed to surgical weight loss and we confirmed these results on a larger number of patients with the more detailed assessment.¹⁴ Majority of weight loss occurs within the initial 2 months post-SG, after which the rate of weight loss decreases, with wide variations between the patients. Patients with T2D were less likely to lose weight post-SG, compared to those with no T2D, similar to previously reported in patients with normal kidney function treated for obesity with surgical^15,16 or medical therapies.^17,18 We believe that patients with suboptimal weight loss may benefit from the early addition of medical therapies, pending further studies.

The risk of postoperative surgical complications was overall very low (3 patients; 5.6%). Hospitalizations affected 12 (22.2%) of patients post-SG, but the risk was not different than in the non-surgical cohort, indicating an overall high risk of adverse events in patients with CKD stage 4-5D. Patients who underwent SG had a numerically higher incidence of MACE during the follow up time. However, this group also had a numerically higher incidence of T2D and the total number of MACE over their lifetime. Therefore, we are unable to draw a conclusion regarding the impact of SG on cardiovascular risks and strokes. Regardless, poor nutritional state and low protein intake have been previously shown to increase the risk of stroke in patients on dialysis.¹⁹ Sheetz and colleagues demonstrated higher risk of mortality at 1 year post-SG compared to well-matched non-surgical counterparts, based on analysis of the United States Renal Data System.²⁰ Patients with CKD stage 4-5D undergoing SG require close attention to managing comorbidities and nutritional status post SG. Importantly, patients who underwent SG were significantly more likely to be listed in active status on the KT list and receive a KT. When performed prior to initiation of dialysis in patients with advanced CKD, bariatric surgery was shown to be associated with decreased mortality, even in individuals who progressed to ESRD.²¹

How SG impacts physical functioning in patients with CKD stage 4–5D has not been adequately studied. In our cohort, moderate or severe functional impairment was highly prevalent in individuals at 12 months post-SG, and there was a clinically meaningful decline in SPPB and FP scores from pre-to post-SG. SG induces rapid muscle catabolism due to severe calorie restriction, leading to the increased prevalence of sarcopenia from 8% at the time of surgery, to 30% at 1 year.²² Cardiorespiratory fitness was reported to decline as early as 6–8 weeks post-procedure in otherwise healthy, free-of-illness, and musculoskeletal pain individuals.^23,24 Some of these important outcomes will e of these important outcomes will need to be addressed by the prospective studies.

Nevertheless, prospective interventional studies to prevent muscle loss and preserve/improve physical fitness should be developed in patients with CKD stage 4-5D treated with SG. These interventions, may need to be tailored to the specific needs of patients with CKD, who are overburdened by chronic diseases and therapies, have high psychosocial complexity²⁵, low adherence to recommendations²⁶, and baseline physical activity level below the 5^th percentile of healthy age-matched individuals.²⁷ Physical therapy and health and wellness coaching are some of the interventions that could be explored in patients with CKD stage 4–5D undergoing SG, similar to those previously performed in patients with normal kidney function²⁸, as well in patients with chronic diseases at risk of frailty ²⁹

Our data support performing SG in patients with CKD stage 4–5D to improve general health and access to KT. In the era of minimally invasive surgery, KT in the candidates with class 3 obesity is possible with acceptable surgical risks. However, medical comorbidities such as insulin resistance, T2D, hypertension, and hyperlipidemia, highly associated with obesity, contribute to higher rates of morbidity and cardiovascular mortality in patients on dialysis.³⁰ Importantly, these comorbidities, if not adequately addressed, may negatively affect post KT outcomes, including the increased risk of recurrent diabetic kidney disease,³¹ graft failure and cardiovascular morbidity and mortality.^32–34 One third of the patients with T2D discontinued insulin post-SG, although the formal diagnosis of diabetes remission (defined as return of HbA1c to less than 6.5%)³⁵ cannot be made due to the incompletely collected HbA1C data and limitations of this test in patients with renal failure.³⁶ Similarly, SG positively affected hypertension rates.

Our study has several important limitations, including the retrospective design and inability to match patients SG with non-surgical cohort based on demographics and comorbidities due to the relatively small number of patients. In addition, the sample was predominantly Caucasian, thus it is currently unclear how these findings would apply to a more diverse population. Eating behaviors, substance use, and mood were not assessed over time, thus the impact of SG on mental health and comorbidities is unknown. Additionally, there was a large variation in the follow up time. The study has several strengths including a uniform approach (SG patients and non-surgical cohort underwent similar transplant and weight loss evaluation) and detailed structured follow-up.

CONCLUSION

We found that SG improved the overall health of patients with CKD stage 4–5D. Future efforts should be made to develop interventional trials to maximize the benefits of the procedure, which in turn may remove barriers causing clinical inertia in treating obesity in patients seeking KT.

Abbreviations

BMI

body mass index

CKD

chronic kidney disease

FP

Fried frailty phenotype

GLP1RA

glucagon-like peptide-1 receptor agonists

HD

hemodialysis

IU

international units

KT

kidney transplant

MACE

major adverse cardiovascular event

PT

physical therapy

SG

Sleeve Gastrectomy

SPPB

Short Physical Performance Battery score

T2D

type 2 diabetes

TRANSMET

Transplant Metabolic program