Obesity Management in Kidney Transplant Candidates: Current Paradigms and Gaps in Knowledge

Abstract

In this review, we discuss evidence of the increasing prevalence of obesity among people with chronic and end-stage kidney disease (ESKD) and implications for kidney transplant (KT) candidate selection and management. Although people with obesity and ESKD receive survival and quality-of-life benefits from KT, studies show that most KT programs maintain strict body mass index (BMI) cut-offs to determine transplant eligibility. However, BMI does not distinguish between visceral adiposity, which confers higher cardiovascular risks and risks of perioperative and adverse post-transplant outcomes, and muscle mass, which is protective in ESKD. Furthermore, requirements for ESKD patients with obesity to lose weight before KT should be balanced with the findings of numerous studies that show weight loss is a risk factor for death among ESKD patients, independent of starting BMI. Data suggest that KT is associated with survival benefits relative to remaining on dialysis for candidates with obesity, although recipients without obesity have higher delayed graft function rates and longer transplant hospitalizations. Research is needed to determine the optimal body composition metrics for KT candidacy assessments and risk stratification. In addition, ESKD-specific obesity management guidelines are needed that will address the neurologic, behavioral, socioeconomic, and physical underpinnings of this increasingly common disease.

Introduction

The prevalence of obesity, defined as a Body Mass Index (BMI) greater than 30 kg/m² by the World Health Organization (WHO), has been steadily increasing over the past several decades in the United States (US)[1]. In 2017, 39.8% of US adults and 18.5% of US children had obesity[2]. Obesity is an independent risk factor for kidney disease, and the worsening worldwide obesity epidemic has dramatically increased the burden of chronic kidney disease[3]. Population trends of obesity are reflected in US kidney transplant (KT) candidates, among whom obesity prevalence increased from 11.6% to 25.1% between 1987 and 2001[4]. Interestingly, many observational studies have shown that higher body mass is protective for people with (end-stage kidney disease) ESKD who are dependent on dialysis. This protection might arise from numerous factors, such as the benefits of having higher nutritional reserves and muscle mass. However, many people with ESKD and obesity are excluded from the benefits of KT on the basis of body mass index (BMI) alone, due to concerns about higher risks of peri-operative and post-operative complications.

In this review, we discuss contemporary practices with respect to management and selection of transplant candidates with obesity, evidence that BMI is an imperfect metric to assess cardiometabolic and surgical risks for transplant candidates, and alternative measures to more accurately assess body composition and fat distribution. We also provide an appraisal of the evidence that has linked obesity and weight loss with adverse outcomes after KT. Finally, we examine gaps in knowledge for ESKD-specific obesity management and the need for research and consensus guidelines to improve the overall care and transplant outcomes of people with ESKD and obesity.

Methodology

We first conducted a comprehensive search of the Pubmed database for studies published between 1999 to 2021 that contained the keywords “obesity,” and “kidney transplantation,” yielding 1,637 articles. We then refined our search to studies published within the past ten years that included the previous key words, “kidney disease” and “weight loss” (n = 142). We screened the resulting article abstracts and included those that examined impacts of weight loss and obesity on post-transplant outcomes and/or explored current obesity management strategies for ESKD patients in the current review.

Origins of BMI and the Reverse Epidemiology of Obesity in ESKD

BMI, a function of height and weight, was first developed by Adolphe Quetelet in On Man and the Development of his Faculties, or Essays on Social Physics (1842) for studying population statistics[5]. BMI was rarely used as a proxy for evaluating body fat percentages until 1972, when Ancel Keys, a nutritional epidemiologist and physician, endorsed the metric as a simple and applicable method for measuring body composition. Subsequent research has shown that obesity, as defined by BMI, has important implications for health outcomes in the general population[5]. According to the World Health Organization (WHO), people with a BMI of 25.0–29.9 kg/m² are considered overweight whereas people with a BMI of 30.0 kg/m² or higher are classified as having obesity. Furthermore, obesity is divided into three categories: class I (BMI 30.0–34.9 kg/m²), class II (BMI 35.0–39.9 kg/m²), and class III (BMI≥40.0 kg/m²). In 2016, the WHO estimated that 13% of the world’s adult population had obesity, a proportion that nearly tripled since 1975[6]. People with class II and III obesity are at a higher risk of death relative to people with normal BMI[5]. Obesity confers health risks to people both directly and indirectly: obesity can lead to joint disorders and immobility, and is a known risk factor for cardiovascular disease, kidney disease, diabetes, and several cancers[6]. As nearly 40% of US adults have obesity, the US Department of Health and Human Services has recognized that reducing the prevalence of obesity in children and adults is a national priority[7].

In contrast to observations of higher mortality with higher BMI in the general population, research in people with chronic conditions such as ESKD, heart failure, acquired immunodeficiency syndrome, and malignancy suggest that higher BMI might be protective. This “obesity paradox,” or reverse epidemiology of obesity, is thought to be related to greater muscle mass, hemodynamic stability, lipoprotein defense against circulating endotoxins, protective cytokine profiles, toxin sequestration by fat mass, and/or anti-oxidative properties of muscle among those with higher BMI[8, 9]. Many epidemiologic studies have shown that among dialysis-dependent people, having a higher BMI is associated with lower mortality than having a lower BMI[10]. A 2010 retrospective study by Kalantar-Zadeh and colleagues analyzed survival outcomes in a five-year cohort of 121,762 patients receiving hemodialysis three times weekly between July 1, 2001 and June 30, 2006 and found a graded and linear association between higher BMI (up to 45 kg/m²) and five-year survival[11]. A 2016 study by Doshi and colleagues used a marginal structural model to analyze associations between BMI and all-cause mortality among 123,624 hemodialysis patients who received treatment between 2001 to 2006. The study found that compared to the reference BMI of 25 to <27.5 kg/m², patients with a BMI of 40 to <45 kg/m² had a 31% lower risk of mortality (Hazard Ratio [HR] 0.69, 95% Confidence Interval [CI] 0.64–0.74). Mortality risk of hemodialysis patients declined linearly with increasing BMI[12]. Other studies using a variety of modeling approaches have also demonstrated the reverse epidemiology of obesity in dialysis patient populations[11, 13–16]. These studies suggest that, in contrast to findings in the general population, higher BMI might signal advantageous body composition in people with chronic disease, and especially those with ESKD.

BMI and Post-Transplant Outcomes

There are numerous risks associated with KT, including skin and soft tissue complications (such as wound infections), anastomotic complications, systemic infections, complications related to allograft function (such as delayed graft function, rejection, and graft failure), and other complications (e.g., hospital readmissions, new-onset diabetes mellitus after transplantation)[17]. Studies examining whether obesity influences the likelihood of these post-transplant adverse outcomes have yielded a mixed picture (Table 1). Excess abdominal fat creates operative challenges that might lead to delayed graft function. However, many experts contend that obesity is not associated with differences in overall recipient survival, surgical site infection, or other surgical complications relative to normal BMI[18–20]. Some studies suggest that it is mainly obesity in the setting of additional comorbidities that confers higher post-transplant risks. In 2017, Schachtner and colleagues investigated the impact of comorbidities of obesity on post-transplant outcomes. In this study, only those KT recipients with obesity who also had pre-existing diabetes mellitus had inferior patient and allograft survival, worse allograft function, delayed graft function, and prolonged hospitalization when compared to kidney transplant recipients without obesity. There were no significant differences in patient or allograft survival, delayed graft function, or hospitalization duration when comparing non-diabetic KT recipients with obesity and both diabetic and non-diabetic KT recipients without obesity[21]. Similarly, in a 2016 study by Kwan and colleagues, although higher BMI was associated with shorter time to graft failure independent of diabetes mellitus status, non-diabetic recipients with a BMI between 30–34.9 kg/m² had similar graft survival to recipients with normal BMI[20]

Table 1.

Recent Findings on BMI and Other Metrics of Fat Distribution as Predictors of Transplant-Related Outcomes

Citation	Study Design and Participants	Transplant-Related Outcomes	Results/Conclusions
Lynch, R. J., et al. (2009) [51]	Retrospective cohort study 869 KT recipients at a single center between 2003 and 2008; 351 with obese BMI (>30 kg/m2)	Graft loss SSI Post-Transplant Mortality	SSI, but not obese BMI, independently associated with increase in risk of graft loss at 3 years [HR] 2.2 (95% [CI] 1.36–3.55) Risk factors for SSI: recipient age, DGF, & BMI >30 kg/m2 Obese BMI was not an independent risk factor for mortality or graft loss
Kovesdy, C. P., et al. (2010) [22]	Prospective cohort study 993 KT recipients in Budapest February and August 2007 until death or June 27, 2010	Mortality Graft Loss	Waist circumference more accurate predictor of mortality than BMI Unadjusted moratlity HRs (95%CI) associated with one SD higher BMI and waist circumference: 0.94 (0.78, 1.13), p=0.50 and 1.20 (1.00, 1.45), p=0.05, respectively Higher BMI associated with lower mortality after adjustment for waist circumference (0.48 [0.34, 0.69], p<0.001); higher waist circumference more strongly associated with higher mortality after adjustment for BMI (2.18 [1.55–3.08], p<0.001)
Luan, F. L., et al. (2010) [19]	Retrospective cohort study 591 non-diabetic KT recipients in one center between 1999 and 2005 who were followed up to 2006	NODAT Incident metabolic syndrome	314 (53.1%) had metabolic syndrome and 90 (15.2%) developed NODAT 84 patients with NODAT also had metabolic syndrome (14.2%) Risk factors for metabolic syndrome: weight gain after transplant associated with metabolic syndrome Risk factors for NODAT: Black race, old age, and HTN-related ESRD Risk factors for both: elevated BMI and fasting glucose levels at transplant
Molnar, M. Z., et al. (2011) [16]	Retrospective cohort study 14,632 KT waitlisted HD patients from 7/2001–6/2007 Used 13-week-averaged BMI & pre-TX serum creatinine as a muscle mass surrogate and their changes over time	Waiting List Mortality	Each kg/m2 increase of BMI was associated with a lower hazard of death (HR 0.96 95%CI: 0.95–0.97) Compared to the lowest serum creatinine quintile, the 4th and 5th quintiles had lower mortality HRs: 0.75 (95 % CI 0.66–0.86) and 0.57 (95% CI 0.49–0.66), respectively Compared to minimal (< +/− 1 kg) weight change over 6 months, those with 3kg – < 5kg and ≥ 5kg weight loss had mortality HRs of 1.31 (1.14–1.52) and 1.51 (1.30–1.75), respectively
Streja E. et al. (2011) [52]	Retrospective cohort study 10,090 HD patients who underwent KT between July 2001 to June 2007	Graft Loss Mortality	Relative to KT recipients with BMI 22–25 kg/m2, recipients with pre-KT BMI >35 kg/ m2 had similar post-KT mortality Higher graft loss in KT recipients with higher BMI Compared with low creatinine and low BMI groups, groups with high creatinine and high BMI had 34% lower adjusted mortality risk (HR: 0.66 [0.49 to 0.88], P < 0.01) 2.2-fold higher risk of death or graft loss with pre-KT serum creatinine <4.0 mg/dL 22% better graft and patient survival with pre-KT serum creatinine ≥14.0 mg/dL Every 1-mg/dL increase of pretransplant serum creatinine associated with 6% lower combined risk of death or graft loss when adjusted for BMI and other covariates (HR: 0.94 [0.91 to 0.97], P <0.001)
Weissenbacher, A., et al. (2012) [53]	Retrospective cohort study 1132 DDKT recipients at a single center, transplanted between 2000 and 2009	DGF, defined as requirement for dialysis within the first week after KT	Recipient BMI is an independent predictor of DGF DGF rates were 25.2%, 29.8%, 40.9%, and 52.6% in recipients with BMI <18.5, 18.5 to 24.9, 25 to 29.9, and ≥30 kg/m2, respectively (P<0.0001)
Gill et al. (2013) [54]	Retrospective cohort study National cohort (n=208,498) waitlisted on HD or transplanted between 1995 and 2007	Mortality after KT relative to remaining on HD	HD patients with an obese BMI (>30 kg/ m2) who receive KT have survival advantage compared to remaining on HD 1 year post KT survival benefit: 48% lower mortality with BMI ≥40 kg/ m2 & ≥66% lower mortality with BMI < 40 kg/m2 Survival benefit of KT relative to HD was not observed in Black race subgroup with BMI ≥ 40 kg/m2
Nicoletto, B. B., et al. (2014) [18]	Meta-analysis 21 studies (9,296 participants) included	DGF Graft Survival Mortality	Obese BMI (>30 kg/m2) was associated with DGF (RR 1.41; 95%CI: 1.26–1.57; I=8%; Pheterogeneity=0.36), but not with acute rejection Conclusion: Graft and patient survival similar between recipients with obese BMI and recipients with healthy BMI (18–25 kg/m2)
Dedinska et al, 2015 [23]	Retrospective cohort study 167 DDKT recipients in the years 2003–2012) at single center with information on waist circumference, BMI and weight gain one year after KT	NODAT Mortality Graft Loss	64 recipients developed NODAT Waist circumference, but not BMI, was independent risk factor for NODAT Higher waist circumference related to higher incidence of NODAT (r=0.1935, [95% CI: 0.01156 to 0.3630], P=0.04) Independent risk factors for NODAT: Age>50 years at time of KT (aHR=2.50, [95% CI: 1.72 to 3.65], P<0.0001) Waist circumference in men >94 cm (aHR=1.95, [95% CI: 1.17 to 3.25], P=0.01) and in women >80 cm (HR=4.50, [95% CI: 1.87 to 10.86], P=0.009
Krishnan, N., et al. (2015) [55]	Retrospective cohort study in United Kingdom including 13,526 patients on KT waitlist who were listed from Jan 1, 2004 to Dec 31, 2010, with follow-up until Dec 31, 2011	1 and 5 Year Mortality	1 and 5 year patient survival significantly better in all BMI groups (<18.5, 18.5-<25, 25-<30, 30-<35, 35-<40, and 40+ kg/m2) with KT relative to remaining on the waiting list (p < 0.0001)
Lafranca JA, et al. (2015) [56]	Retrospective cohort study Meta-analysis of 56 studies (data of more than 209,000 KT recipients)	Acute Rejection Wound Infection and Dehisence Graft Loss Mortality Transplant operation duration NODAT Transplant hospitalization LOS	High BMI (>30 kg/m2) is associated with lower post-KT mortality with an overall HR of 0.93 (CI, 0.89–0.97; P <0.001, I2 0%; P = 0.68) Relative to low BMI (<30 kg/m2) KT recipients, outcomes of recipients with high BMI (≥30 kg/m2): DGF (RR = 1.52) Acute rejection (RR = 1.17) 1-, 2-, and 3-year graft survival (RR = 0.97, 0.95, and 0.97) Wound infection and dehiscence (RR = 3.13 and 4.85) NODAT (RR = 2.24) Operation duration (+0.77 hours) LOS (+2.31 days)
Kwan, J. M., et al. (2016) [20]	Retrospective cohort atudy US Scientific Registry of Transplant Recipients database 191,091 KT recipients from 1987–2013	Post-KT proteinuria Acute rejection Graft Loss DGF	Combined data for DDKT & LDKT recipients: Recipients with obese BMI: increased risk DGF, graft loss, proteinuria and acute rejection relative to normal BMI recipients Non-diabetic recipients with a BMI between 30–34.9 kg/m2 had similar graft survival to recipients with normal BMI Subset analysis of LDKT recipients: Difference in graft loss between lower and higher BMI recipients not significantly different
Schachtner, T., et al. (2017) [21]	Retrospective cohort study 660 KT recipients at single center, received KT between 2005 and 2012	Mortality DGF Graft Loss LOS	Among recipients with obesity (BMI>30 kg/m2), only those who also had pre-existing diabetes mellitus had inferior patient and allograft survival, worse allograft function, DGF, and prolonged hospital LOS (p<0.05) compared to KT recipients without obesity No significant differences in patient or allograft survival, DGF, or hospital LOS when comparing non-diabetic KT recipients with obesity and both diabetic and non-diabetic KT recipients without obesity (p>0.05)
Erturk, T., et al. (2019) [17]	Retrospective cohort study 561 LDKT recipients from a single center in Turkey	LOS Post-TX hospital readmission rate DGF Graft Loss Mortality NODAT	Pre-KT obese BMI >30 kg/m2) associated with: Delayed wound healing (P=0.03), Longer hospitalization LOS (P=0.03) More readmissions (P =0.04), NODAT(P =0.02), and CAD (P=0.03) Increased graft loss (P=0.04) Obesity at one year post-KT associated with: Increased graft loss at years 3 & 5 post-KT (P = .04 and P = .03, respectively) Increased 5-year mortality (P = .03)
Fellmann, M., et al. (2020) [57]	Retrospective cohort study 506 HD patients who received a KT at one center over 11 year period	DGF Graft Loss Mortality	Obese BMI (≥30 kg/m2) associated with DGF: HR 2.60 [95% CI 1.31–5.02], P=0.004) & higher graft loss (HR 1.55 [1.06–2.99], P =0.04) Obese BMI not associated with increased mortality (HR = 1.82 [0.88–3.79], P=0.11)

Abbreviations: HD – hemodialysis; KT – kidney transplantation; DGF—delayed graft function; HTN – hypertension; aHR – adjusted HR; OR – odds ratio; aOR – adjusted OR; CI – confidence interval; DDKT – deceased donor KT; LDKT – living donor KT; RR – Relative risk; HR – Hazard ratio; SD – standard deviation; SSI – surgical site infection; CAD – coronary artery disease; BMI – body mass index; USRDS – United States Renal Data Service; NODAT - new onset diabetes after transplantation; LOS – length of hospital stay; DM – diabetes mellitus; CAD – coronary artery disease.

Research increasingly suggests that fat distribution is more strongly associated with KT outcomes than BMI. In a prospective study of 993 KT recipients with information on waist circumference and BMI, Kovedsky and colleagues found that larger waist circumference was associated with over 2-fold higher post-transplant mortality (adjusted HR 2.18, 95% CI 1.55–3.08), whereas higher BMI was associated with lower post-transplant mortality (adjusted HR 0.48, 95% CI 0.34–0.69)[22]. In a study of 167 deceased donor KT recipients, waist circumference, but not BMI, was an independent risk factor for new onset diabetes after KT [23]. These studies underscore the need to better characterize the specific features of obesity that are most specific for adverse post-transplant outcomes, to improve both the assessment of transplant candidates and the ability to target obesity interventions to the most vulnerable transplant recipients.

Pitfalls of Relying on BMI for Transplant Candidacy Assessments

Despite limitations of BMI to characterize fat distribution and body, most transplant programs rely strongly on BMI when determining KT candidacy[24, 25]. Although there is considerable heterogeneity in practices, a “BMI cut-off” of 35 kg/m² is typical of many transplant programs[26]. Furthermore, wait-listed transplant candidates with a higher BMI are more likely than those with a lower BMI to be inactivated due to their excess weight, and once inactivated, only 50% are likely to be re-activated for transplantation[27, 28]. These practices should be considered in the context of findings that many ESKD patients with a BMI in the “healthy range” have excess abdominal body fat. In a study that compared BMI to skinfold-thickness-estimated body fat among hemodialysis patients, over 55% of those with a BMI <30 kg/m² had obesity by skinfold thickness assessment, and these patients with subclinical obesity had evidence of lower muscle mass and strength than those with a BMI ≥ 30 kg/m² [29]. Therefore, although BMI is convenient to assess, the inconsistent associations between BMI and post-transplant outcomes (Table 1) and the reverse epidemiology of obesity in ESKD are strong motivations to augment transplant candicacy assessments with metrics that directly assess abdominal obesity, body fat distribution, and physical function (Table 2). Such metrics are likely to be helpful when monitoring ESKD patients with a stable BMI in addition to those who are attempting weight loss or weight loss maintenance, as they can help clinicians identify patients who are losing muscle or gaining fat [30]. Direct assessments of muscle function might also serve as more accurate biomarkers of risk for transplant candidates and recipients than BMI [31].

Table 2.

Alternative to BMI: Metrics that Might Augment Kidney Transplant Candidacy Assessments

Metric	Definition	Healthy Definition	Associated Outcomes in ESKD	Pros/Cons
Waist circumference (WC)	Waist measured horizontally, just above the hip bone	<80 cm in women, <94 cm in men [58]	Significant association between larger WC and post-transplant mortality [22]	Pros: Direct assessment of abdominal obesity, low cost, rapid Cons: Operator dependent; requires training to ensure reliability
Waist-to-hip ratio (WHR)	Waist circumference / hip circumference	>0.85 in women, >0.97 in men [59]	Associated with all-cause mortality in ESKD [59]	Pros: Measure of fat distribution, low cost, rapid Cons: Operator dependent; requires training to ensure reliability
Mid-arm muscle circumference (MAMC)	Mid-arm circumference (cm) – 3.142 × TSF (cm). TSF measured with skinfold caliper; surrogate for lean body mass (LBM) [60]	Normal ≥ 90%, Depleted < 90%	Larger MAMC is an independent predictor of survival in hemodialysis patients [60]	Pro: Estimate of nutritional stores and protein reserves Cons: Operator dependent; requires training to ensure reliability
Serum creatinine	Marker measured via standard blood test; surrogate for muscle mass	Higher levels might indicate higher muscle mass for patients with ESKD.	Higher pre-KT serum creatinine levels associated with better graft and patient survival post-transplant [52]	Pro: Assessed routinely; inexpensive Cons: potentially confounded by kidney function and dialysis adequacy
CT and MRI	Laser-based 3D body imaging of visceral adipose tissue (VAT)	Morphometric features of lean abdominal mass	Low psoas mucle cross-sectional area and density on CT associated with waitlist mortality[61]	Pro: Often used in transplant evaluation Cons: Cost, radiation (CT), convenience
Dual-energy X-ray absorptiometry (DEXA)	Whole body fan beam method to determine lean body mass [52]	Normal ≥ −1 Osteopenia < −1.1	Bone mineral density associated with post-transplant fracture risk [62]	Pro: Can directly measure lean muscle mass and fat and bone density Cons: Cost, convenience
Bioelectrical impedance spectroscopy (BIS)	Measures body impedance to an applied alternating electric current; allows for calculation of total muscle mass	Normal BCM cell percentage of lean body mass: For men: 53% – 59% For women: 50% – 56%	Sarcopenia measured by BIS associated with 2–3 fold higher mortality in elderly HD patients after adjustment for covariates[24]	Pro: Can directly measure lean muscle mass, more cost-effective and portable than other imaging modalities Cons: Sensitive to changes in total body water/volume status

Abbreviations: WC – waist circumference; WHR – waist-to-hip ratio; ESKD – end-stage kidney disease; MAMC – mid-arm muscle circumference; TSF – triceps skinfold thickness; KT – kidney transplant; CT – computed tomography; MRI – magnetic resonance imaging; DEXA – dual-energy X-ray absorptiometry; BIS – bioelectrical impedance spectroscopy; BCM – body cell mass

Another important concern about the routine practice of using BMI cut-offs to determine who receives a transplant evaluation is that such policies might exacerbate racial and socioeconomic disparities in transplant access. In the US, obesity prevalence is much higher among non-Hispanic Black adults than among non-Hispanic White adults (38.4% vs 28.6%)[32]. Poorer adults with obesity are more likely to have unhealthy diets and central fat distribution than richer adults with obesity [33]. Racial and socioeconomic differences in obesity prevalence might be driven by unequal access to healthy foods, chronic stress, lower educational attainment, and built environments that limit physical activity [7]. Therefore, it is important to consider impacts on health equity and disparities when considering the role of BMI in transplant candidacy assessments.

Implications of Weight Loss Requirements for Kidney Transplantation

Given well-established transplant program BMI “cut-offs,” patients with obesity and ESKD are very likely to encounter recommendations to lose weight in order to be eligible for a KT. However, transplant centers have varying approaches for integrating BMI into evaluation practices, with likely implications for patient outcomes and access. Some transplant centers will deny patients an in-person assessment if their BMI is above that center’s threshold. Other centers will conduct in-person assessments of all patients but subsequently decline those who are not under their BMI threshold. Some centers might refer patients directly to associated dietary and bariatric programs. At other centers, including ours, there is no formal BMI threshold for evaluation or wait-listing, and decisions about weight loss before KT are based on our assessment of functional status and a surgeon’s assessment of abdominal girth and fat distribution.

In addition to achieving KT access, patients with ESKD and obesity might also be motivated to lose weight due to the adverse impacts of obesity on physical and mental quality-of-life [34]. Among dialysis patients, a higher BMI is associated with a perceived need to lose weight[35]. However, numerous observational studies suggest that weight loss might confer risks for people with ESKD. In 2011 a study by Molnar and colleagues of 14,632 hemodialysis patients from a large dialysis organization who had not yet undergone KT, patients who experienced weight loss (3 kg ≤5 kg and ≥5 kg) over six months had a significantly higher risks for mortality (HR 1.31 [95% CI 1.14–1.52]) and 1.51 [95% CI 1.30–1.75], respectively) compared to hemodialysis patients who experienced little to no weight change (<±1 kg) over six months. Weight loss prior to the initiation of dialysis is also associated with excess risks among dialysis patients. In 2018, Ku and colleagues conducted a study of 770 Chronic Renal Insufficiency Cohort (CRIC) participants who began hemodialysis or peritoneal dialysis during the study period. They showed that a >5% annualized weight loss in late-stage CKD was associated with a 54% higher mortality risk after initiation of dialysis (adjusted HR 1.54, 95% CI 1.17–2.03) relative to participants who had stable weight before dialysis initiation[36]. In a 2019 study of 94,465 deceased donor KT recipients who received KT between December 4, 2004 to December 3, 2014, KT recipients with weight loss ≥10% of their listing weight were at 18% higher risk of death (aHR 1.18, 95% CI 1.11–1.25, p<0.001) and results were consistent in the subgroup of recipients who were listed with a BMI ≥30 kg/m² [28]. Importantly, weight loss was a risk factor in all of these studies independent of starting BMI. However, few studies to date have provided granular nutritional and functional context when assessing associations between weight loss and adverse outcomes.

Some studies have tried to distinguish between total body weight changes in dialysis patients from change in muscle mass using serum creatinine as a muscle mass surrogate. In a nationally representative study from 2001 to 2006 of 121,762 hemodialysis patients over five years, weight loss in the context of stable or increasing serum creatinine conferred lower mortality risk than weight loss with decreasing serum creatinine [11]. Similar associations between serum creatinine level and mortality were observed in a six year study of 14,632 wait-listed hemodialysis patients[16]. Another recent study among 919 kidney transplant recipients found that pre-transplant unintentional, but not intentional, weight loss was associated with more post-transplant weight gain, graft failure, and mortality. Those with unintentional weight loss before transplant were more likely than others to report exhaustion and low physical activity, and findings were robust among recipients who had a BMI in the overweight and obese category before transplant[37]. These studies underscore the heterogeneity of weight loss trajectories in people with ESKD, and the need for obesity management strategies that incorporate assessments of muscle mass, function, quality-of-life, and fat distribution.

Current Obesity Management Practices and Gaps in Knowledge

Several key questions must be answered to improve obesity management paradigms in ESKD (Figure). A fundamental first step is likely for nephrology and transplant professionals to stop defining obesity by BMI alone—a practice that underemphasizes the complexity of obesity as a disease that requires comprehensive and multidisciplinary management. The Obesity Medicine Association defines obesity as “a chronic, relapsing, multifactorial, neurobehavioral disease, wherein an increase in body fat promotes adipose tissue dysfunction and abnormal fat mass physical forces, resulting in adverse metabolic, biomechanical, and psychosocial health consequences”[38]. This definition of obesity underscores the neurologic and behavioral and physical underpinnings of the disease that must be addressed in any weight loss intervention for patients with ESKD. The chronic and relapsing nature of obesity also necessitates the study of weight loss maintenance interventions for people with ESKD.

Figure.

Key Questions to Improve Obesity Management in Kidney Transplant Candidates

Currently, there are no ESKD-specific guidelines to support obesity management and the evidence base is limited because patients with ESKD and obesity have historically been excluded from most weight loss trials [39]. Numerous weight loss strategies are employed in contemporary practice, including behavioral and nutritional interventions, pharmacotherapy, and bariatric surgery (Table 3). Few data exist to compare success rates between approaches in the ESKD and transplant candidate populations, though dialysis patients with obesity might have unique barriers to nutritional weight loss approaches given dietary and fluid intake restrictions, treatment-related fatigue, and hemodynamic effects [40]. Although several pharmacologic agents exist for weight loss that have been approved for people with impaired kidney function [41], they have been understudied in patients with ESKD and obesity[40]. For example, glucagon-like peptide-1 (GLP-1) receptor agonists have been shown to be remarkably effective for weight loss in the general population [42, 43] and to have acceptable safety profiles in earlier-stage CKD [44], but far fewer studies have examined the safety and tolerability of these agents in late-stage CKD and ESKD populations that might be KT-eligible [45]. Bariatric surgery is increasingly utilized for ESKD patients with obesity and is an effective weight loss intervention, although short-term complication rates are higher than among patients without ESKD[46]. A 2019 study by Cohen and colleagues among 1,694 CKD patients and 925 ESKD patients who underwent bariatric surgery found a significantly increased risk of 30-day reoperation (CKD odds ratio [OR] 2.25 95% CI 1.45–3.51; ESKD OR 3.10, 95% CI 1.72–5.61) and readmission (CKD OR 1.98, 95% CI 1.5–2.56; ESKD OR 2.97, 95% CI 2.05–4.31) compared to patients without CKD; mortality risk was elevated in patients with ESKD (OR 11.59, 95% CI 6.71–20.04) but not in those with CKD (OR 1.00, 95% CI 0.32–3.11)[46]. However, although the relative risks of bariatric surgery complications might be higher for ESKD patients, absolute risks are low and might be acceptable to patients who have not been successful with other therapies.

Table 3.

Studies of Weight Loss Interventions among People with ESKD and Obesity and Areas for Future Study

Citation	Study Participants	Intervention	Results	Potential Limitations & Questions For Future Research
Cook et al. (2008) [63]	66 participants with stage 4 CKD and stable or increasing weight (BMI 27.9–47.3 kg/m2): 22 received weight program intervention, 22 received standard care Intervention arm: 8 PD patients, 14 HD patients, 3 KT recipients, 19 in stage 2–4 CKD	A multidisciplinary weight management program with a low fat, reduced energy diet, individual exercise prescription and pharmacotherapy with Orlistat 120 mg total dissolved solids	73% of intervention group attended at least half of appointments Mean body weight reduced by 7.1 % from 102.9 kg to 95.7 kg (P<0.001) Reduction in BMI from 35.7 kg/ m2 at baseline to 33.2 kg/m2 at 12 months Waist circumference decreased by 12.9 cm (P<0.005) at 12 months Significant improvement in functional metrics	Attendance rates suggests needs to explore/address barriers to intensive behavioral interventions
Koshy, A. N., et al. (2008) [64]	Single-center case series 3 HD patients with obese BMI on KT waitlist	LAGB	Surgery enabled pts to lose sufficient weight loss to be eligible for KT No reported complications	Small sample size Potential selection bias
Freeman, C. M., et al. (2015) [65]	Prospective single center study of patients treated with LSG between December 2011 and January 2014 52 KT candidates Average preoperative BMI of 43.0 +/− 5.4 kg/m2 (range 35.8–67.7 kg/m2)	LSG	Weight loss with 6 months medical management before surgery significantly lower than 6 months following LSG (3.6% +/− 2.3% vs 37.6% +/−3.3%, p<0.0001) Mean BMI decreased from 43.0 kg/m2 pre-LSG to 36.4 kg/m2 post LSG No perioperative (<30 days) deaths In 25 months of follow up: three deaths, 2 within the first year 41% decrease in anti-HTN medications (p<0.001) 50% decrease in daily insulin dose (p<0.001) 6/52 participants received KT during follow-up	No information on patients who were deemed inappropriate for LSG-potential selection bias
Idorn, T., et al. (2016) [43]	Randomized placebo-controlled, double-blinded trial 20 patients with T2DM & ESKD (1:1 for liraglutide vs. placebo) 20 pts with T2DM & normal kidney function (1:1)	Liraglutide-titrated more slowly in ESKD participants than in participants with normal kidney function	Dose-corrected plasma trough liraglutide concentration at the final visit 49% higher (95% CI 6–109, P = 0.02) ESKD patients compared to patients with normal kidney function Body weight reduced similarly in both liraglutide-treated groups (from 91.1 +/− 4.9 to 99.7 +/− 5.2 kg) Adverse side effects more common in ESKD patients: nausea and vomiting	Findings support more testing of glucagon-like peptide-1 agonists for efficacy and tolerability in KT-eligible patients with obesity
Cohen J., et al. (2019) [46]	Retrospective cohort study 323,034 individuals in the United States without CKD, 1,694 patients with CKD, and 925 patients with ESKD who underwent bariatric surgery	LSG (primary surgery for ESKD pts) Gastric bypass (primary surgery for CKD and non-renal disease pts)	Post-operative mortality higher in ESKD ([OR] 11.59, 95% CI 6.71–20.04) but not among those with CKD ([OR] 1.00, 95% CI 0.32–3.11) Risk of 30-day reoperation compared to patients without CKD: CKD [OR] 2.25 (95% [CI] 1.45–3.51); ESKD [OR] 3.10 (95% CI 1.72–5.61) Increased risk of readmission, compared to patients without CKD: CKD [OR] 1.98 (95% CI 1.5–2.56); ESKD [OR] 2.97 (95% CI 2.05–4.31)	Different surgery types predominate based on CKD/ESKD status. Data source does not provide granular information on changes in nutrition and function after surgery Focus on short-term (30 day) outcomes
Cohen J., et al. (2019)[47]	Retrospective, single center cohort study 43 patients→pre-KT bariatric surgery 21 patients→post-KT bariatric surgery Propensity-score matched controls	Gastic bypass, SG, LAGB, Vertical banded Gastroplasty	Compared to matched controls, surgery pre- and post-KT associated with decreased graft failure HR among pre-KT surgery 0.31, 95% CI 0.29–0.33; HR among post-KT surgery 0.85, 95% CI 0.85–0.86 Post-KT mortality risk lower among pre- and post-KT surgery groups compared to matched controls	Potential unmeasured confounders, confounding by indication < 20% pre-KT surgery patients were dialysis-dependent at time of surgery-more study needed on outcomes in ESKD population
Bouchard, P., et al (2020)[66]	Retrospective cohort study 32 patients referred by transplant team fo surgical weight loss Average BMI 42.3 (5.2) kg/m2	LSG	One year median BMI change −9.8 (3.7) kg/m2 90 day complication rate: 3% (no deaths) One year after surgery: 63% listed for KT KT performed in 14 patients median of 8 months after surgery	No information on candidates who were not selected for surgical intervention.
Kassam A., et al. (2020)[48]	Retrospective single-center cohort study of patients referred for surgical weight loss by transplant program between 2011–2018 198 patients with ESKD 45 patients with CKD	LSG	Mean weight decreased from 129.4 +/− 26.8 kg to 105.0 +/− 22.3 kg. 71.7% patients achieved a BMI ≤40 kg/m2 (center transplant threshold)	Large loss to follow-up among those who did not receive surgery - of 499 patient referred to clinic, 256 patients did not receive SG (14→ medical therapy and 242 lost to follow-up) Reasons for loss to follow-up among those referred for surgical weight loss need further exploration

Abbreviations: BMI – body mass index; BS – bariatric surgery; HD – hemodialysis; KT – kidney transplantation; aHR – adjusted HR; OR – odds ratio; aOR – adjusted OR; CI – confidence interval; DDKT – deceased donor KT; CKD – chronic kidney disease; GBP – gastric bypass surgery; SG – sleeve gastrectomy; LSG - laparoscopic SG; RYGB – Roux-en-Y gastric bypass procedure; AGB - adjustable gastric banding; BPD–DS - biliopancreatic diversion–duodenal switch; ESKD - end stage kidney disease; RR – Relative risk; HR – Hazard ratio; SD – standard deviation;. GBD – gastric band surgery; T2DM – Type 2 DM.

Benefits of surgical weight loss are not limited to those around KT access, as pre- and post-KT bariatric surgery might also lower the risks of allograft loss and post-transplant mortality[47].Yet, many patients might not prefer or qualify for bariatric surgery. For example, in a study of 499 patients with morbid obesity who were referred for bariatric surgery by a transplant program, less than half (n=198 patients) ultimately underwent the procedure. Furthermore, among 256 patients that did not receive surgery, 94% (n=242) did not subsequently follow up[48]. Acknowledging the challenges of weight loss for ESKD patients, some transplant programs have adopted robotic surgery as an option for transplant candidates with morbid obesity[49]. However, this option remains very limited and innovation in weight loss interventions remains an important priority for the care of most patients with obesity and ESKD who desire kidney transplant.

Ideally, patients with ESKD and obesity who require weight loss for KT should be under the care of health professionals with formal training on obesity pathophysiology and management. In reality, due to variations in obesity management practices, patients are likely to receive guidance from numerous sources, such as transplant nutritionists, dietitians, and commercial weight loss programs that incorporate individual or group interactions. Given the lack of a strong evidence base to support ESKD-specific nutritional weight loss management guidelines, experts have suggested that a reasonable initial approach would be through individualized nutritional guidance by a renal dietitian [39]. This approach offers numerous advantages in particular for facility-based dialysis patients, whose renal dietitians can meet with them during treatments and integrate their weight loss goals with their other dialysis-related nutritional goals. However, many renal dietitians report that their ability to support patients’ weight loss attempts is hampered by “too little time,” “lack of training [in obesity management],” and “too few resources”[50]. Dietitians also cited the unavailability of healthy foods as a major barrier for patients trying to lose weight, underscoring the need for weight loss interventions to consider patients’ social, financial, and environmental contexts. Therefore, there is a need to explore substantial changes to current ESKD care delivery paradigms to provide the comprehensive support needed by patients with obesity, including increasing dedicated time and obesity management training for renal dietiians. Indeed, the added complexities of caring for the ESKD patient with obesity likely requires a team approach with coordinated care from professionals with nephrology, obesity, transplant, exercise, mental health, and social context expertise.

Discussion

In this review, we discussed evidence that obesity is a highly prevalent disease among patients with ESKD. Although higher body mass might confer survival benefits in ESKD, obesity is also a pervasive barrier to KT and a direct cause of numerous health problems and immobility that impact quality-of-life. Nephrology and transplant professionals often rely on BMI to define obesity and set thresholds for transplant eligibility. Current evidence consistently demonstrates that transplant recipients with a BMI of 30 kg/m² or higher are more likely to experience delayed graft function compared to those with a lower BMI. However, most studies suggest that graft and patient survival outcomes are excellent in recipients with a BMI in the obese category and that transplant is associated with a survival benefit in these patients compared to remaining on dialysis. BMI thresholds for transplant eligibility are problematic in that they are applied unevenly across transplant programs and do not capture fat distribution and might systematically disadvantage racial and ethnic minority patients and lower-income patients. Racial and socioeconomic disparities in obesity prevalence should also be an important consideration in the design and assessment of weight loss interventions for patients with ESKD. Overall, important questions to prioritize for future research include the long-term health, function and quality-of-life impacts of existing weight loss interventions, strategies to address weight loss intervention non-adherence or loss to follow-up, and predictors of successful weight loss maintenance in the setting of ESKD and KT.

In summary, obesity is a complex and chronic disease that will be better defined if nephrology and transplant professionals integrate direct metrics of body composition, fat distribution, physical and mental health into clinical assessments. In the absence of a strong evidence base to support nutritional weight loss guidelines for patients with ESKD and obesity, current practices to support patients with ESKD who desire or require weight loss are highly variable, and include diet and lifestyle changes, bariatric surgery, and pharmacological interventions. Research is needed to understand the health impacts of these interventions on people with ESKD before and after transplantation, particularly in light of numerous studies that suggest that loss of muscle mass and function are a risk factors for mortality in ESKD. Assessments of weight loss interventions in ESKD should also include mental health-related quality of life endpoints.

Clinical Summary:

• The prevalence of obesity is rising among people with end-stage kidney disease.

• Although higher body mass might confer survival benefits to patients who are dependent on dialysis, obesity is a pervasive barrier to kidney transplantation.

• The routine use of body mass index “cut-offs” for kidney transplant candidacy assessments is problematic because body mass index is not specific for body composition and oversimplifies obesity as multi-system disease.

• Numerous observational studies show that weight loss is a risk factor for death in end-stage kidney disease, though these studies have rarely accounted for potential changes in functional status or body composition corresponding with weight trajectory.

• There is a need for end-stage kidney disease specific guidelines to support obesity management and more study and innovation in surgical and non-surgical weight loss interventions.