Pretransplant Obesity: A Weighty Issue Affecting Transplant Candidacy and OutcomesBaylor Regional Transplant Institute, Baylor University Medical Center, Dallas, Texas Correspondence: Jeanette Hasse, PhD, RD, FADA, CNSD, Baylor University Medical Center, Transplant Services, 3500 Gaston Avenue, Dallas, TX 75243. Electronic mail may be sent to jm.hasse{at}sbcglobal.net. Because of the global increase in prevalence of obesity, many more overweight and obese individuals are undergoing evaluation for transplantation than in the past. Although obesity seems to provide a survival benefit in dialysis patients, obesity has traditionally been considered a contraindication for transplantation of most organs. It is theorized that obesity will contribute to worse transplant outcomes, including lower rates of graft and patient survival and higher rates of delayed graft function and infection. This review evaluates the available literature evaluating outcomes of obese patients with end-stage organ failure who undergo transplantation. Obesity seems to be associated with increased rates of wound infection after transplantation. However, other adverse transplant outcomes related to obesity seem to be dependent on the type of organ being transplanted and the degree of obesity. For example, a body mass index (BMI) of 30 kg/m2 may reduce short-term survival in lung transplant recipients; however, obesity does not seem to confer an adverse effect on short- or long-term survival in liver transplant patients until a much higher BMI is reached (such as 35 or 40 kg/m2). Each transplant center must determine weight guidelines and criteria for identifying the level of obesity as a contraindication for transplantation. This must be based on organ type, each center's transplant and complication statistics, and available donor pools. Guidelines must also consider the morbidity and mortality risks of the obese patient with organ failure who does not receive a transplant.
The prevalence of obesity in developed countries continues to increase
exponentially. The Centers for Disease Control and Prevention reported that in
2005, the residents in 46 of the 50 U.S. states had an obesity prevalence of
>20%.1 This rate
of obesity prevalence was nonexistent in 1995, and only 28 states in 2000 had
an obesity prevalence >20%. The same trend is found in transplant patients,
with an increasing proportion of obese patients being referred for
transplantation. In 2003, using the United Network of Organ Sharing (UNOS)
database, Friedman et
al2 analyzed the
weights of >85,000 kidney transplant recipients who were evaluated for
transplantation from 1987 to 2002. Only 11.6% of patients had a body mass
index (BMI)
Obesity has been found to be a protective factor for patients with
end-stage renal disease requiring dialysis. In the 2001 Dialysis Outcomes and
Practice Patterns Study (DOPPS), Leavey et
al3 demonstrated
that dialysis patients with a BMI of >30 kg/m2 had a relative
risk of death of 0.77 compared with patients with a normal reference BMI of
23–24.9 kg/m2. If obesity is a protective factor for
hemodialysis, should obese patients choose to keep receiving dialysis instead
of seeking transplantation? Glanton and
colleagues4 examined
this question and evaluated survival chances for 7521 patients enrolled in the
United States Renal Data System (USRDS) from 1995 to 1999 who had a BMI >30
kg/m2. Those who underwent transplantation still had a survival
benefit if they underwent transplantation vs continuing to receive
dialysis (transplant from deceased donor: hazard risk 0.39, 95% confidence
interval (CI), 0.33–0.47; living donor: hazard risk 0.23, 95% CI,
0.16–0.34). However, survival benefit from transplantation was
eliminated when the BMI exceeded 41 kg/m2 (hazard risk 0.47; 95% CI
0.17–1.25; p = .013). A similar study looked at survival
benefit of liver and kidney transplant recipients who were
obese.5 The
mortality rate of obese patients (BMI >30 kg/m2) registered in
the Scientific Registry of Transplant Recipients who underwent liver or kidney
transplantation was compared with the mortality rate of patients awaiting
transplantation. Obese patients who underwent transplantation had a reduction
in mortality compared with patients who did not receive a transplant. The
relative risk of death was elevated in liver transplant patients with a BMI
Obesity has long been considered a relative contraindication for organ transplantation. Theorized risks of obesity include reduced patient and graft survival, as well as increased rates of acute cellular rejection, delayed graft function, infection, wound problems, cardiovascular disease, and diabetes mellitus. In addition, it has been feared that obesity would prolong posttransplant hospitalization and delay graft function, thus affecting the cost of performing transplants for obese patients. The purpose of this review is to consider the published literature regarding the effect of obesity on the outcomes of solid organ transplantation. Although this review is comprehensive of all types of transplants and studies, there are, unfortunately, significant variations between the studies. There is variability in how obesity is defined, the weight delineations for the comparison groups, the number of subjects in the studies, whether the studies are single- or multicenter reports, and the type and dose of immunosuppressive drugs and prophylactic antibiotics taken by the patients. The results also differ according to the type of organ transplant being considered. However, there is value in looking at these studies to determine if trends exist in which obesity could be considered a contraindication for transplantation. This paper will first look at the studies according to transplant type and then discuss some of the drawbacks and trends reported by these studies.
Over a dozen research studies have been published to evaluate the effect of obesity in kidney transplantation.6–22 Table 1 summarizes the study design and subject characteristics of each of these studies, as well as the main findings. Some of the variables evaluated in these studies include patient and graft survival, rate of delayed graft function, incidence of acute cellular rejection, and occurrence of wound infections.
In kidney transplantation, results from studies evaluating obesity as a risk factor for transplantation are mixed. Three of 5 studies that evaluated patient survival showed that obesity had a significant adverse effect on 5-year patient survival8,11,21; the other 2 studies did not13,14 (Figure 1). Four studies8,9,15,21 showed a significant reduction in 5-year graft survival in patients considered obese vs nonobese; 4 other studies found no effect of obesity on graft survival10,11,13,14 (Figure 2). Delayed graft function was found to be increased in obese kidney transplant recipients in 5 reports6,9,10,16,21; another 6 studies published later found no effect of obesity on delayed graft function11,13-15,18,19 (Figure 3). Acute rejection was not influenced by obesity in any of the studies7,11,12,14,15,18-20 (Figure 4). The incidence of wound infection, however, was increased in obese patients in most6,7,9,10,14,16,18,19 but not all studies (Figure 5).11
Three studies have analyzed data from databases to evaluate effect of obesity in a larger patient population. In 2002, Meier-Kriesche and colleagues23 reported on the effect of weight on transplant outcomes according to the analysis of 51,927 kidney transplant recipients from the USRDS database. In this study, patients were divided into 11 subgroups according to their BMI. There were 11 BMI subgroups from <18 to >36 kg/m2. An inverse bell-shaped curve represented extremes of low and high weight that were associated with increased mortality, graft loss, infectious death, and cardiovascular death compared with a normal reference rate. These data suggest that a BMI  30 kg/m2 is a risk factor for adverse outcomes after
kidney transplantation. However, a BMI of  20–22 kg/m2 was
also considered a risk factor for some adverse outcomes.
A report from the Netherlands evaluated short- and long-term patient and
graft survival in obese vs nonobese renal transplant recipients
according to data in a national database. In the report, 1871 nonobese
patients (BMI <30 kg/m2) were compared with 196 obese patients
(BMI
Gore et al22
analyzed kidney transplant data from the UNOS database to evaluate obesity's
effect on outcomes. One thousand forty-two patients were underweight (BMI
<18.5 kg/m2), 12,089 patients were normal weight (BMI
18.5–24.9 kg/m2), 8765 patients were overweight (BMI
25–29.9 kg/m2), 3891 patients were obese (BMI 30–34.9
kg/m2), and 1590 patients were morbidly obese (BMI
Gore et al22 also used multivariate Cox regression modes to look at the association between BMI and graft survival, controlling for covariates of age, gender, race/ethnicity, comorbidities (coronary artery disease, peripheral vascular disease, diabetes mellitus), human leukocyte antigen matching, cold ischemia time, and donor type (living or deceased donor). The hazard ratio (HR) for graft failure was significant for underweight (HR 1.21 with 95% CI, p = .02) and morbidly obese patients (HR 1.22 with 95% CI, p = .001). Additional reviews on obesity in kidney transplantation have been published24–26 and further discuss the data and impact of these studies.
Although fewer in number than the kidney transplant reports, several studies have evaluated the effect of obesity on liver transplant outcomes. The first case report was published in 1994 by Keeffe et al.27 Included in this evaluation were 276 patients; however, only 5 patients had a BMI of 24.6–33.2 kg/m2 and 13 patients had a BMI of 36–49 kg/m2. There was a 100% survival in the obese patients, with a 1- to 43-month follow-up time. Wound infection was the only significant problem reported in the obese patients, with an occurrence rate of 61%.
The next published report was by Braunfeld et al28 in 1996. Forty liver transplant recipients with a BMI >30 kg/m2 were compared with a group of 62 patients having a BMI <30 kg/m2. There were no significant differences between the groups with regard to surgical time, transfusion requirements, or reoperation for bleeding or wound repair. In addition, there were no differences between the groups for length of stay in the intensive care unit or hospital, or 30-day and 1-year posttransplant survival.
In 1999, a study by Sawyer and colleagues29 compared outcomes according to 3 different body weight levels: 30 liver transplant patients were severely obese (BMI >35 kg/m2), 55 patients were obese (BMI of 30–34 kg/m2), and 217 control patients were overweight or of normal weight (BMI <30 kg/m2). Wound infection and death from multisystem organ failure occurred at a significantly higher rate in the severely obese vs other groups (Figure 6). On the other hand, obesity was not associated with a reduced 1-year graft or patient survival rate.
We investigated our center's statistics on obesity using data from patients who received transplants between 1985 and 1998.30 Liver transplant recipients were divided according to BMI: 128 patients were underweight (BMI <20 kg/m2), 707 patients were of acceptable weight (BMI 20–27.5 kg/m2), 101 patients were overweight (BMI 27.5–32 kg/m2), and 189 patients were severely or morbidly obese (BMI >32 kg/m2). Obesity had no effect on rejection rate, surgical time, or 5-year patient and graft survival rates. Obesity also did not significantly affect rate of infection, although wound infections were slightly more prevalent in the underweight and severely or morbidly obese groups (p = .07). The last single-center report was published in 2005 by a Danish group, who reported a significantly lower survival rate in obese liver transplant recipients compared with nonobese patients (p = .013).31 Three hundred forty-five patients with a BMI <30 kg/m2, transplanted from 1999 to 2003, were compared with 20 recipients having a BMI >30 kg/m2 (mean, 33.7; range, 30.1–46.3 kg/m2). A 65% mortality rate during the first 2 posttransplant years in the obese group was attributed to multiorgan failure (n = 3), septicemia (n = 5), acute myocardial infarction (n = 1), central nervous system–related death (n = 2), and cancer (n = 2). Results of this study could be hard to interpret, given the small number of obese patients. Single-center studies lack statistical strength due to small numbers of transplant recipients in the most extreme weight categories. Multicenter trials or analyses using databases can provide a larger number of patients to evaluate. In 2002, Nair et al32 looked at effects of obesity in liver transplant recipients, using multicenter data from the UNOS database; 18,172 patients were evaluated. Patients were evaluated according to BMI classification: <25 (n = 8382), 25.1–30 (n = 5913), 30.1–35 (n = 2611), 35.1–40 (n = 911), and >40.1–50 kg/m2 (n = 355). Obesity did not have a significant effect on 2-year graft survival or 1-year mortality. On the other hand, patients with a BMI >40 kg/m2 had significantly higher 1-, 2-, and 5-year mortality rates compared with the other groups. The 5-year mortality after transplantation was also elevated in the group with a BMI of 35–40 kg/m2. The results of this study were the basis for the recommendation to consider morbid obesity a contraindication for liver transplantation in the 2005 American Association for the Study of Liver Disease (AASLD) practice guidelines for the evaluation of the patient for liver transplantation.33
A similar study was conducted and reported by Rustgi et
al34 in 2004. Data
from 26,920 liver transplant recipients in the UNOS database were evaluated to
determine effect of weight on outcomes. Patients were divided according to
BMI: 2991 had a BMI <19, 3702 had a BMI of 19–22, 5654 had a BMI of
23–25, 8289 had a BMI of 25–29, 3961 had a BMI of 30–34,
1585 had a BMI of 35–39, and 738 had a BMI of
At least 3 studies looked at the effect of obesity in patients with combined pancreas and kidney transplants.35–37 Bumgardner et al35 compared 28 patients with a BMI of >27 kg/m2 with 212 patients with a BMI 27 kg/m2. There were no differences between the
groups with regard to weight gain, hyperlipidemia, hypertension, rejection, or
patient survival. Actuarial graft survival was lower in the obese compared
with nonobese group (p = .029), with the most common causes of graft
loss 3 years after transplant attributed to patient death (primarily from
cardiovascular disease) and rejection in both groups. A study by Rogers et al36 evaluated 88 consecutive patients who underwent simultaneous pancreas and kidney transplantation at a single center from 1995 to 2001. Fifty-two patients were classified as nonobese (BMI <25 kg/m2) or mildly to moderately obese (BMI >25 kg/m2; range, 25–35.1 kg/m2). There were no differences between the groups in the rates of 1-, 3-, or 5-year patient survival, pancreas graft survival, or kidney graft survival. In addition, obesity did not influence pancreas thrombosis, infection, relaparotomy, or length of hospital stay. There was a significantly higher rate of duodenojejunal anastomotic leakage in the overweight (17%) vs normal weight group (2%; p = .012). It is important to note in this study that the patients in the obese group were only mildly to moderately obese; it was not clear how many patients had a BMI <30 vs >30 kg/m2.
Hanish and
colleagues37
evaluated 145 pancreas transplant recipients (either simultaneous
pancreas-kidney transplantation or pancreas after kidney transplantation). The
obese cohort consisted of 33 patients (23%) with a BMI >30
kg/m2. The obese group had significantly higher rates of wound
dehiscence, ventral hernia, intraabdominal infection, gangrene, necrotizing
fasciitis, repeat laparotomy, graft pancreatitis, and enteric leak than the
nonobese patients. In this study, predictors of posttransplant complications
were age
Traditionally, heart transplant programs have had stricter weight guidelines for transplant candidacy than other organ transplant programs. The first report of the effect of obesity in 193 heart transplant recipients (transplanted between 1984 and 1991) was published in 1996 by Grady et al.38 Obesity was defined as >110% ideal body weight. Those who were overweight (n = 84) had significantly reduced survival (p = .018). A similar study was published by the same authors in 1999, but different criteria were used to define obesity.39 The later study evaluated 4515 heart transplant recipients from 38 institutions during the period of 1990 to 1995. Patients were classified according to percent ideal body weight and BMI. BMI standards were based on those from the National Center Health Statistics at the time. There were 527 under-weight patients (BMI <20.7 kg/m2), 2869 recommended-weight patients (BMI 20.7–27.7 kg/m2), 799 overweight patients (BMI 27.8–31.0 kg/m2), and 320 obese patients (BMI 31.1
kg/m2). BMI was not correlated with any outcomes. However, body
weights <80% or >140% ideal body weight predicted mortality and were
associated with increased infection rates. Another study published in the same year evaluated 584 heart transplant recipients.40 One hundred six patients had a BMI <21 kg/m2, 369 patients had a BMI 21–27 kg/m2, 80 patients had a BMI of 27–30 kg/m2, and only 29 patients had a BMI >30 kg/m2. There was no information about the range of BMI or if there were any patients with a BMI >35 kg/m2. Comparisons were made between patients with a BMI >27 kg/m2 and those with a BMI <27 kg/m2. Wound infection was elevated in patients with a BMI >27 kg/m2 (p = .018), but there was no effect on postoperative survival. Lietz et al41 took another approach to evaluating obesity in their 2001 publication. These researchers looked at 474 heart transplant recipients and used risk ratio (RR) to determine factors that influenced 5-year mortality, 1-year acute rejection rates, and 5-year transplant-related coronary artery disease. Sixty-eight patients with a BMI <20 kg/m2 were classified as cachectic, 238 normal weight (BMI 20–27 kg/m2), 113 overweight (BMI >27–30 kg/m2), and 55 morbidly obese (BMI >30 kg/m2). Five-year mortality rates were nearly double in morbidly obese and cachectic patients vs normal weight recipients (p = .001). Factors that adversely affected 5-year morbidity were cachexia (RR, 2.113), morbid obesity (RR, 2.093), African American descent (RR, 1.623), female gender (RR, 1.437), and donor age >50 years (RR, 1.82). Acute rejection in the first year posttransplant occurred more frequently in patients who were cachectic (RR, 1.4), morbidly obese (RR, 1.795), and <45 years old (RR, 1.423) compared with the other groups. Finally, the 5-year transplant-related coronary artery disease occurred more frequently in patients who were morbidly obese (RR, 1.435), had a donor >40 years old (RR, 1.679), or experienced acute cellular rejection (RR, 1.452) vs those who did not have these risk factors. Conversely, patients who were cachectic before transplantation had a lower risk of coronary artery disease 5 years after transplantation (RR, 0.524). Taking a slightly different approach than other studies, a report from the Cardiac Transplant Research Database evaluated posttransplant weight gain on outcomes.42 Obesity (BMI >30 kg/m2) ina cohort of 3540 heart transplant recipients was more prevalent 1 year posttransplant (38%) vs pretransplant (17%; p < .00001). Patients who were obese or underweight (BMI <18.5 kg/m2) 1 year after transplant had more rejection episodes than normal weight or overweight patients (p = .009) and a trend toward reduced 5-year survival (p = .06). Because this study did not evaluate body composition, it cannot be determined if weight gain was due to an increase in fat tissue, muscle, or fluid.
There are few data available on the influence of obesity on lung transplant outcomes. In 2002, Kanasky et al43 reported results of 85 lung transplant recipients who underwent transplantation from 1994 to 1998. Patients were classified as underweight (BMI <18.5 kg/m2;n = 19), normal weight (BMI 18.5–24.9 kg/m2;n = 37), overweight (BMI 25–29.9 kg/m2;n = 19), or obese (BMI 30 kg/m2; n = 10). Using Kaplan-Meier
survival distribution, the researchers concluded that patients with a BMI
30 kg/m2 had nearly a 3-fold reduction in posttransplant
survival (p < .05). Overweight patients had an equal survival time
compared with normalweight patients.
Culver and
colleagues44
evaluated both BMI and percent ideal body weight as risk factors for short-
and long-term survival. One hundred twenty patients were used as normal-weight
controls (BMI 18.5–24.9 kg/m2). Lung transplant recipients
with a BMI
One of the drawbacks of the studies on transplant obesity is that the field of transplantation has advanced dramatically in the last 2 decades so that criteria that may have applied in the 1980s and 1990s may no longer be strong predictors of poor outcome today and in the future. For example, there have been advances in immunosuppressive drugs, and the regimens that were used 2 decades ago do not match many of the current regimens. Likewise, rejection rates are lower today than they were years ago. In addition, prophylactic antimicrobial treatments have changed in the last 20 years, thus influencing infection rates. Another drawback to these studies is that when obese patients are accepted for transplantation, they are "superselected" meaning that they are usually free from most other contraindications for transplantation. These patients may be obese but are not likely to be debilitated or have multiple other comorbidities such as infection, uncontrolled diabetes, or other chronic organ dysfunctions. One must also consider that most of the studies that have been published include fairly small numbers of patients. Single-center reports may have more evenly matched groups and control better for outside variables, but the multicenter studies have the advantage of larger sample sizes. Conversely, multicenter studies interject more variability among the groups and less control of the confounding factors. Finally, the studies reviewed in this paper compared the risk of obesity with nonobesity in transplant recipients, but they did not always compare obesity with other risk factors for transplant complications. To get a better picture of what degree of obesity is a risk factor for transplantation, one must compare obesity to other factors considered as risks for transplantation. For example, Lietz et al41 also showed that other factors, such as being of African American descent or female or receiving a transplant from a donor >50 years, also conferred an increased risk (RR, 1.623, 1.436, and 1.82, respectively). However, these factors do not preclude patients from transplantation. If one compares the risk of obesity in transplantation with other risks, it may seem that obesity does not have as great an influence on outcome as some of the other risk factors considered acceptable for transplant candidacy.
Obesity and transplant trends influence the selection criteria for transplantation. First, obesity is more prevalent today in organ transplant recipients than in the past. Transplant centers cannot avoid evaluating and accepting transplant candidates who are obese and who may have been excluded in the past. Second, because the availability of deceased donors cannot match the need for transplants, it is incumbent upon the transplant community to allocate scarce organs to those who will best benefit from transplantation. Obesity must be considered a risk factor for transplantation, but the question that remains is at what level is obesity a significant contributor to adverse outcomes after transplantation. The results from the studies presented in this paper suggest that obesity may adversely affect some outcomes after transplant. The most common adverse event in obese patients across all transplant types appears to be wound infections. Most transplant specialists would agree that this complication is surmountable and acceptable, given that these patients have reduced survival or quality of life without transplantation. On the other hand, if transplanting obese patients results in worse survival compared with no transplant, or if it precludes those from receiving transplants who would have better survival, transplant centers must consider obesity a relative contraindication for transplantation. It appears that kidney and liver transplantation conveys a better survival for obese patients than for those not receiving a transplant.4,5 However, these data are not widely available for other transplant types. In addition, the degree of obesity at which complications occur after transplantation is not clear or consistent among the studies or the types of transplants. For example, in liver transplantation, obesity does not seem to create worsened short-term mortality until BMI is >40 kg/m2.32 Even 5-year postliver-transplant mortality is not affected until BMI is >35 kg/m2.32 However, a BMI >30 kg/m2 can influence even short-term mortality in lung transplant recipients.44 These inconsistencies mean that weight guidelines must be specific to each type of transplant. If obesity is to be considered a contraindication for transplantation, one must also evaluate whether extreme underweight should also be considered as a contraindication for transplantation. At least 4 of the studies discussed earlier in this paper showed that being underweight or cachectic had adverse effect upon survival in kidney,23 liver,34 and heart38,41 transplant recipients. There are multiple studies in the transplant literature also showing that malnutrition results in adverse posttransplant outcomes.45–56 Does this mean that if we require obese patients to lose weight before transplantation that we should also require malnourished patients to gain weight in order to qualify for transplantation? This is usually not the case; most transplant specialists argue that most patients cannot successfully gain weight until transplantation. Is the same true for weight loss in obese patients requiring transplantation?
It is incumbent upon transplant centers to determine what features should be considered absolute or relative contraindications for transplantation. It appears that obesity is likely a relative, not absolute, contraindication for organ transplantation. It is up to each transplant center to look at organspecific programs for their own statistics, treatment regimens, and donor pools to determine at what level of obesity transplantation would create a higher risk to an individual patient than to provide no transplant. Second, transplant programs must determine what types of weight loss programs are acceptable for their patients. Another paper in this issue by DiCecco57 evaluates the medical treatments for weight loss available to organ transplant candidates. An earlier article by Alexander and Goodman58 outlined the effects of bariatric surgery on pre- and post–kidney-transplant recipients. Finally, a very important question must be asked: If patients are obese, do their outcomes improve if they lose weight? A study by Schold et al59 attempted to look at this question in renal transplant patients. Using data from the USRDS database, researchers looked at outcomes of patients who lost weight while waiting for a transplant. A reduced BMI did not improve posttransplant survival or graft loss. However, it was unclear if weight loss was associated with worsening of disease condition. In addition, it was not clear if volitional weight loss resulted in loss of subcutaneous fat instead of a more desirable loss of visceral fat. As with many nutrition issues in organ transplantation, there are more questions than there are answers. Because of the prevalence of obesity, transplant programs will be accepting obese candidates for transplantation. However, most experts would agree that it is better to set defined weight criteria for each organ to apply to all patients rather than arbitrarily accepting or declining patients for transplantation according to the opinions of those who happen to be at the transplant selection committee the day a patient is presented to the committee. Certainly, transplantation is an evolving science. Transplant nutrition specialists must be willing to track data, and it is incumbent upon the transplant community to collaborate to obtain large sample groups in order to continually reevaluate data relating to obesity and transplant outcomes.
Nutrition in Clinical Practice, Vol. 22, No. 5,
494-504 (2007) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Top
Kidney Transplantation
Liver Transplantation and...
