It is more than just size: Obesity and transplantation


  • See Article on Page 907

Address reprint requests to Mark S. Orloff, M.D., Division of Solid Organ Transplant, University of Rochester Medical Center, 601 Elmwood Avenue, Box Surg-Txp, Rochester, NY 14642. Telephone: 585-275-5875; FAX: 585-276-2182; E-mail:

The obesity pandemic is an inescapable problem in health care. A recent survey found nearly 80 million Americans, or 36% of the population older than 20 years, to be obese.[1] Those caring for persons with end-stage liver disease have seen a 5.4% increase in the number of liver transplant candidates with a body mass index greater than 30 kg/m2 in the United States.[2] This concern is reflected in the frequency of publications on this topic, including the work in this issue of Liver Transplantation by Perez Protto et al.,[3] who report equivalent results in a matched retrospective analysis of lean and morbidly obese patients undergoing liver transplantation at a single center and cite excellent outcomes for both groups. However, a large retrospective review of liver transplantation in obese patients from the United Kingdom pointed to higher degrees of morbidity and health care resource utilization in patients who were obese versus a leaner cohort.[4] Inferior graft survival and more postoperative complications have also been cited in a number of other studies.[5-8] Although a high body mass index has been found by some to be an independent predictor of mortality after transplantation,[9] others have demonstrated a survival benefit for obese patients.[10, 11] The issue of superior outcomes versus a survival benefit has resulted in a debate over the rationing of the precious resource of donor organs, with some, including the American Association for the Study of Liver Diseases through its published guidelines, advocating restrictions on the transplantation of obese patients.[12, 13] The superb clinical results described by Perez Protto et al. in this issue of Liver Transplantation are in contrast to the various conclusions of other single-center studies contributing to this unsettled issue[14-18] and thus invite a more critical examination of transplantation for the obese patient.

There is little doubt that obesity poses some risk to health, mostly through its association with cardiovascular disease, type 2 diabetes, osteoarthritis, and some cancers. A number of studies have reported U- or J-shaped correlations of obesity with overall mortality.[19, 20] Although one might postulate that obesity is universally associated with higher morbidity and mortality rates, this is not always the case. Findings in critically ill intensive care patients have revealed a somewhat unexpected protective association with survival, although the basis for this has yet to be defined.[21-24] In fact, many of these studies have shown higher mortality rates for underweight patients. The average American can expect to undergo 7 surgical procedures throughout his or her life[25]; as a result, surgery is being scrutinized with the goal of improving results. The National Surgical Quality Improvement Program of the American College of Surgeons provides a robust reporting system designed to produce risk-adjusted outcome measures for national surgical quality improvement.[26, 27] Studies using National Surgical Quality Improvement Program data have failed to find a direct correlation between obesity and mortality in surgical patients.[28-30] The term obesity paradox is used to describe the unexpected protective effect associated with an increased body mass index.[30] Similarly, retrospective analyses of single-center data sets have found no correlation of obesity with overall complications for general or colorectal cases.[31, 32]

In a retrospective study of renal transplant recipients, Lynch et al.[33] linked surgical site infections to premature graft loss, and although obese patients (body mass index > 30 kg/m2) had a higher rate of surgical site infections, obesity itself was not independently associated with worse outcomes in a multivariate analysis. Findings of increased wound complications (eg, infections, hernias, and dehiscence) in obese patients are pervasive across surgical specialties,[34-38] including solid organ transplantation.[39] In a retrospective review of liver transplants, Schaeffer et al.[40] also found that obesity did not affect survival, but it was associated with wound infections. Thus, it appears that although obese patients may have higher rates of wound complications, their survival with respect to surgery and critical illness is no worse.

The concern with obesity in the general population is related primarily to abdominal obesity and its association with cardiometabolic risk and metabolic syndrome.[41] The cluster of variables for metabolic syndrome include visceral obesity, systemic hypertension, dyslipidemia, and insulin resistance. Obesity is independently associated with cerebrovascular disease,[42, 43] coronary heart disease,[44-46] and an increase in overall mortality from cardiovascular complications.[47-49] Are there risks attributable to obesity that are independent of cardiovascular complications? Studies in patients undergoing surgery for coronary heart disease provide a surrogate answer to this question because cardiovascular risk factors are prevalent in both obese and normal-weight subjects. Although the findings are not universal, the majority of these studies have revealed that with the notable exception of a heightened rate of sternal wound infections, there is no increase in morbidity or mortality for the obese patient. Here again, underweight patients appear to be at the greatest risk for adverse outcomes.[50-53] Underweight patients also fare poorly in studies of dialysis and renal transplant patients.[54, 55]

Although excess fat is linked to mortality primarily through cardiovascular risks, the important question is which individuals in this population are in jeopardy. Epidemiological studies have found that 10% to 40% of obese individuals are metabolically healthy with higher insulin sensitivity, an absence of hypertension, and favorable lipid, inflammation, hormone, and liver enzyme profiles.[56] Conversely, there are normal-weight individuals who behave with a phenotypically obese metabolic profile, including insulin resistance, an increased risk for diabetes, hypertriglyceridemia, and atherosclerosis.[57] The difficulty in identifying risk is primarily related to 2 issues: (1) the lack of a consensus on the optimal definition of corpulence and (2) the lack of a mechanistic understanding of how obesity causes an increase in cardiovascular risk. The anthropometric use of the body mass index is convenient and reproducible but does not distinguish fat from the fat-free contributions of muscle and bone to body mass.[58-61] Obesity also exhibits a spectrum of fat distribution with visceral and subcutaneous adipose tissue. Although excess in either compartment can lead to insulin resistance, their structure, composition, metabolic activity, and functional significance differ with varying adipokine profiles. Visceral adipose tissue tends to be associated with a more adverse metabolic, dyslipidemic, and atherogenic phenotype, but the relationship between obesity and cardiometabolic disorders is more complex. Consequently, the pathophysiology associated with excess fat is not purely a matter of size or even distribution but stems from other parameters related to genetics[62-64] and environmental factors shaping the gut microbiome[65, 66] (ie, the populations of bacteria colonizing the gut that are responsible for energy harvesting), which differs in obese and lean individuals. Undoubtedly, better predictors of cardiovascular risk will be developed that supersede the current inaccurate anthropometric descriptions. These will likely incorporate variables of the fat phenotype and distribution as well as genetic and environmental influences.

Obesity is also associated with the explosive growth of nonalcoholic steatohepatitis, which, in turn, is linked to metabolic syndrome, although the causal mechanisms of this relation are not well understood. Nonalcoholic steatohepatitis is the fastest growing indication for liver transplantation in the United States, and this is a trend that shows no signs of slowing.[67] It is predicted that by 2025, nearly 25 million Americans will develop nonalcoholic steatohepatitis, a fifth of whom could undergo transplantation.[68] Furthermore, nonalcoholic steatohepatitis as an etiology of cirrhosis and as an indication for transplantation will accelerate over time because of increasing rates of childhood obesity. Outcomes for patients undergoing liver transplantation for nonalcoholic steatohepatitis have been comparable to outcomes for patients with other indications.[69-71] However, a large series looking at transplantation in patients with nonalcoholic steatohepatitis recently reported longer operative times, more intraoperative blood loss, and longer lengths of stay after transplantation.[68] The nonalcoholic steatohepatitis group had a significantly increased proportion of obese and morbidly obese patients (nearly 50%) in comparison with patients with other etiologies of liver disease. Although the outcomes were similar, the authors concluded that this growing group of patients stresses health care resources significantly.

Because cardiovascular disease significantly threatens operative success, transplant candidates are highly scrutinized and carefully selected to avoid these major sources of short-term morbidity and mortality. Thus, it is no surprise that obese liver transplant recipients have good short-term outcomes, just as the authors of the current study have confirmed.[3] The longer term risks associated with metabolic syndrome or its components affect overall long-term survival in the liver recipient population.

The prevalence of metabolic syndrome has been reported to be 44% to 58% in those followed for at least 6 months after liver transplantation[72-74] versus 23% in the general population.[75] Components of metabolic syndrome are exacerbated after transplantation by organ-sustaining immunosuppressive agents (particularly calcineurin inhibitors and corticosteroids).[76] It has been shown that 26% of transplant patients develop new-onset diabetes after transplantation, and this does not even account for those with nondiabetic levels of hyperglycemia.[77] Hypertension is also affected by immunosuppressive agents, particularly calcineurin inhibitors, which increase vasoconstriction systemically and in the kidneys. A greater than 50% increase in the prevalence of high blood pressure is seen after liver transplantation.[76, 78] Weight gain after transplantation is also common.[79, 80] Although steroids are often blamed, it has been shown that the strongest predictor of posttransplant weight gain is a high pretransplant body mass index,[72, 81] and patients manifest excessive weight gain even under steroid-sparing regimens.[82] These factors and a number of others all influence the long-term outcomes of liver transplants. A prospective, long-term follow-up of liver transplant recipients based on the National Institute of Diabetes and Digestive and Kidney Diseases database revealed the top causes of death more than 1 year after liver transplantation to be graft failure, malignancies, cardiovascular complications, and renal failure, all of which can be interrelated with metabolic syndrome.[83]

When nonalcoholic steatohepatitis is the indication for transplantation, posttransplant steatohepatitis and fibrosis are more common,[84] and furthermore, the presence of metabolic syndrome is predictive of nonalcoholic steatohepatitis recurrence.[85] This raises cardiovascular morbidity and death rates.[86] Cardiovascular events are seen with greater frequency in posttransplant patients with metabolic syndrome versus liver transplant recipients without metabolic syndrome.[73, 74] Renal insufficiency, commonly seen after transplantation and often attributed to calcineurin inhibitor toxicity, is also affected by diabetes and hypertension. In a study of renal failure after liver transplantation,[87] renal biopsies revealed dysfunction to be nearly equally attributable to calcineurin inhibitor toxicity and hypertensive changes or diabetic nephropathy. Collectively, hypertension, weight gain, type 2 diabetes, and cardiovascular complications (essentially metabolic syndrome and its individual components) are all more prevalent in obese liver transplant recipients versus lean recipients and significantly worsen under immunosuppressive therapy. This constellation of risk factors associated with obesity raises posttransplant mortality, but at later time points in this highly select patient population.

With the notable exception of wound complications, no significant short-term morbidity or mortality risk factors have been clearly linked to obesity in liver transplant recipients. It must be appreciated that liver transplant candidates are carefully selected to avoid the obesity-related health hazards that would lead to perioperative cardiovascular events. The majority of studies on the obese liver transplant population have queried large databases that lack granularity, or they are retrospective, single-center analyses often covering different eras of transplantation, and this consequently limits the strength of their conclusions. To the question whether obesity poses a risk to liver transplant recipients, the answer is emphatically yes! Obesity affects long-term outcomes because of its interaction with immunosuppression as well as the increased potential for the development of fatty liver. Obese patients derive benefits from surgery in general and, when indicated, from transplantation,[10, 88, 89] but at the same time, they invite more scrutiny by practitioners. Consequently, in comparison with normal-weight individuals, obese patients are more often denied the advantages that these treatments afford.[90] Obesity is an accelerating health concern, but the mechanisms associated with the increased risk (particularly the cardiovascular risk) are poorly understood. The identification of obesity as a hazard has almost exclusively relied on anthropometric definitions and established comorbidities such as hypertension and type 2 diabetes. Systems biology, though conceptually promising, has not yielded clinically useful associations that guide therapeutic interventions.

In summary, excellent results following liver transplantation can be obtained for highly select obese patients, as confirmed by the work in this issue of Liver Transplantation. However, obesity will continue to challenge the transplant community because of its impact on the development of liver disease; its stress on resource utilization; and the long-term risks of cardiovascular disease, which are magnified by immunosuppression. Future prospective studies are warranted to gain a better understanding of the complexity, heterogeneity, and impact of obesity, particularly in the liver transplant population.