Select Utilization of Obese Donors in Living Donor Liver Transplantation: Implications for the Donor Pool


  • J. Moss,

    1. Center for Liver Disease and Transplantation, New York-Presbyterian Hospital, New York, USA
    2. Medicine
    Search for more papers by this author
  • D. Lapointe-Rudow,

    1. Center for Liver Disease and Transplantation, New York-Presbyterian Hospital, New York, USA
    Search for more papers by this author
  • J. F. Renz,

    1. Center for Liver Disease and Transplantation, New York-Presbyterian Hospital, New York, USA
    2. Surgery, Columbia University College of Physicians & Surgeons, New York, USA
    Search for more papers by this author
  • M. Kinkhabwala,

    1. Center for Liver Disease and Transplantation, New York-Presbyterian Hospital, New York, USA
    2. Surgery, Columbia University College of Physicians & Surgeons, New York, USA
    Search for more papers by this author
  • L. M. Dove,

    1. Center for Liver Disease and Transplantation, New York-Presbyterian Hospital, New York, USA
    2. Medicine
    Search for more papers by this author
  • P. J. Gaglio,

    1. Center for Liver Disease and Transplantation, New York-Presbyterian Hospital, New York, USA
    2. Medicine
    Search for more papers by this author
  • J. C. Emond,

    1. Center for Liver Disease and Transplantation, New York-Presbyterian Hospital, New York, USA
    2. Surgery, Columbia University College of Physicians & Surgeons, New York, USA
    Search for more papers by this author
  • R. S. Brown Jr.

    Corresponding author
    1. Center for Liver Disease and Transplantation, New York-Presbyterian Hospital, New York, USA
    2. Medicine
    3. Surgery, Columbia University College of Physicians & Surgeons, New York, USA
    Search for more papers by this author

Corresponding author: Robert S. Brown, Jr.,


Living donor liver transplantation evolved in response to donor shortage. Current guidelines recommend potential living donors (LD) have a body mass index (BMI) <30. With the current obesity epidemic, locating nonobese LD is difficult. From September 1999 to August 2003, 68 LD with normal liver function test (LFTs) and without significant comorbidities underwent donor hepatectomy at our center. Post-operative complications were collected, including wound infection, pneumonia, hernia, fever, ileus, biliary leak, biliary stricture, thrombosis, bleeding, hepatic dysfunction, thrombocytopenia, deep venous thrombosis, pulmonary embolism, difficult to control pain, depression and anxiety. Complication rates for LD with BMI >30 (n = 16) and BMI <30 (n = 52) were compared. The incidence of wound infection increased with BMI, 4% for nonobese and 25% for obese LD (p = 0.024). There were no statistically significant differences for all other complications. No LD died. Recipient survival was 100% with obese LD and 80% with nonobese LD (p = 0.1). Select donors with a BMI >30 may undergo donor hepatectomy with acceptable morbidity and excellent recipient results. Updating current guidelines to include select LD with BMI >30 has the potential to safely increase the donor pool.


Living donor liver transplantation (LDLT) was initially performed in children as a response to a donor shortage that led to disproportionately high waitlist morbidity and mortality (1). The widespread application of this technique directly impacted pediatric transplantation by significantly reducing waitlist morbidity and mortality. Since 1992, the waiting list for liver transplantation has increased by more than sixfold, from just over 2000 in 1992 to over 17 800 patients actively waiting for liver transplantation in July 2005 (2). The median wait time has also increased significantly, from 65 days in 1991 to 514 days in 1998 (3).

Adult-to-adult LDLT was first performed in the United States in 1996 as reported by Wachs et al. (4). To date, over 1700 adult-to-adult LDLT procedures have been performed in the United States; however, the implementation of adult-to-adult LDLT has yet to significantly impact adult waitlist morbidity or mortality (5). In addition, with implementation of the Model for End-stage Liver Disease (MELD) for organ allocation, the waiting list decreased by only 6% in 2002. Despite this, the gap between supply and demand leaves many patients dying, awaiting liver transplantation (6,7). Thus, transplant physicians continue to pursue alternate sources including living donation to increase the donor pool.

Donor exclusion criteria for LDLT remain poorly defined. A frequent reason for donor exclusion is obesity. Current guidelines suggest that donors for LDLT should not be obese. The potential donor with a body mass index (BMI) greater than 30 raises concerns regarding poor recipient outcomes secondary to graft steatosis and a lower graft volume, as well as post-operative complications within the donor. Obesity is associated with liver pathology, including steatosis, fibrosis and cirrhosis. However, data on the incidence of these findings are only beginning to emerge and so far have not consistently defined the relationship between obesity and liver pathology. Rinella et al. found 78% of potential donors with BMI greater than 28 had at least 10% steatosis on liver biopsy (8). However, Ryan et al. showed 73% of donors with BMI greater than 25 demonstrated little or no hepatic fat, while 9% of donors with BMI less than 25 exhibited 10% or greater steatosis (9). The significance of steatosis among LDLT donors remains undefined. Marcos et al. found no functional impairment in grafts containing less than 30% steatosis, although many centers refer to 10% steatosis as their threshold for utilization (10–12). The implication of obesity and steatosis among donors remains controversial.

Obesity, defined as a BMI greater than 30, is epidemic within the United States. The most recent National Health and Nutrition Evaluation Surveys (NHANES) data outline the rapid increase in the prevalence of obesity from 15% in 1976 to 31% in the year 2000 (13). Today, approximately 35% of adults are obese. While the donor pool does not reflect the U.S. adult population, categorical decisions based on BMI will exclude a significant number of potential adult donors. Expansion of these guidelines may increase the application of adult-adult LDLT.

We hypothesized select donors with a BMI between 30 and 40 may undergo donor hepatectomy without a significant increase in the incidence of adverse complications and with equivalent results in the recipient. This retrospective study reports our single center experience utilizing obese donors for adult-adult LDLT.


Donor selection and evaluation

A formal program, which is in compliance with New York State guidelines for LDLT, exists for the evaluation of eligible living donors (LD). This program involves a nurse practitioner who provides education necessary to make informed choice; a hepatologist who is not the primary hepatologist of the recipient; a transplant psychiatrist; an independent social worker who assesses the donor's social support, education level, relationship to recipient, financial status, health insurance and ability to understand impact of donation; a transplant surgeon to inform surgical risk and an independent donor advocate. The independent donor advocate is a board-certified general internist.

All eligible donors evaluated at New York-Presbyterian Hospital between April 1998 and August 2003 were between the ages of 18 and 60 years and met the following criteria: (i) ABO compatibility; (ii) absence of chronic illness, liver disease and severe psychological disorders and (iii) no evidence of coercion or payment. Recipients signed consent to have LD evaluated prior to the beginning of donors' evaluations, and donors were provided with written information that explicitly outlined the risks of surgery and potential complications.

After the above evaluations were completed, donor candidates underwent routine blood testing, chest radiography, electrocardiography and either a CT angiogram or MRI/MRA of the liver. Ultrasonography was not routinely used. Instead, our experience initially favored CT angiogram, but now MRI/MRA is our modality of choice, given the lack of intravenous (i.v.) radiocontrast dye and radiation exposure. Six donor candidates had abdominal ultrasounds. Four of these donor candidates had ultrasonography as their first imaging modality, three subsequently had CT scans and one had an MRI for further evaluation. Two donor candidates had CT scans first and had ultrasonography after the CT scans revealed hypoechogenic areas, suggestive of cysts. When indicated, donor candidates had mammography, PAP smears, pulmonary function tests, exercise stress thallium tests and formal cardiac evaluations. Select donors had pre-donation biopsies. Indications for biopsy were history of abnormal liver function tests (LFTs), positive anti-nuclear antibody (ANA) or the suggestion of fatty liver on cross-sectional imaging. Indications for biopsy did not differ between obese and nonobese individuals. Donors with a BMI >30 but no significant medical history and normal LFTs continued with donation.

Prior to 2003, a bilateral chevron incision was used in donors. In 2003, the standard incision was changed to the right sub-costal incision, with upper midline extension. As a result, approximately 12% (n = 8) of the study group had right sub-costal incisions. One of these subjects had a BMI >30.

This study was reviewed and deemed exempt from consent requirement by the Columbia University Medical Center Institutional Review Board as part of a retrospective review of all LD. All donors identified a health care proxy and signed informed consent.

Outcome measures

Data on post-operative complications in donors and recipients were collected and recorded. Post-operative donor complications were, in general, defined by the need for intervention and included wound infection, pneumonia, hernia, fever, ileus, biliary leak, biliary stricture, vascular thrombosis, bleeding, hepatic dysfunction, thrombocytopenia, deep venous thrombosis, pulmonary embolism, difficult to control pain, anxiety and depression. A wound infection was defined by the need for antibiotics beyond the initial prophylactic antibiotics or a need to reopen the incision. Pneumonia was defined as any clinical pneumonia that required the use of antibiotics. Hepatic dysfunction was defined as hyperbilirubinemia with total bilirubin more than 5 mg/dL. Thrombocytopenia was defined as a platelet count less than 150 000. Ileus was defined by the presence of dilated loops on abdominal x-ray and the need for nasogastric tube. Other complications, including pain, anxiety and depression, were diagnosed clinically, on the basis of chart review of complaints at the time of discharge and at post-operative visits or by the need for narcotics beyond post-operative day 14 or any psychotropic medications. Hepatic dysfunction, biliary leak, biliary stricture, bleeding and vascular thrombosis were considered to be ‘serious’ complications. All other complications were considered to be nonserious. Complication rates for donors with a BMI greater than and that for those with a BMI less than 30 were compared.

Finally, complications were also analyzed according to the standard Food and Drug Administration (FDA) definition of serious adverse event (SAE). According to the FDA, an SAE is defined as any undesirable experience associated with the use of a medical product that involves one of the following outcomes: (i) death; (ii) life-threatening: substantial risk of dying; (iii) hospitalization: initial or prolonged; (iv) disability: resulting in a significant persistent, permanent change/impairment/disruption to a body's function/structure/quality of life/physical activity; (v) congenital anomaly; (vi) requires medical or surgical intervention (14). We extended this definition from ‘use of a medical product’ to ‘use of a medical procedure’, specifically living donor hepatectomy. A complication was deemed life threatening if it resulted in a prolonged stay in the surgical intensive care unit or re-admission to the surgical intensive care unit. Data on hospitalizations and length of stay were collected from the New York Presbyterian Hospital admissions department. We used the FDA's definition of disabilityand interpreted the sixth outcome, ‘requires medical or surgical intervention’, as the requirement of any ‘additional’ medical or surgical intervention, other than what is considered to be the standard of care for LDLT. SAEs for donors with BMIs greater and less than 30 were compared.

Statistical analyses

Proportions were compared using the 2-tailed Fisher's exact test. An additional logistic regression analysis was performed using either total or serious complications as an outcome variable and BMI as a continuous predictor variable. An alpha level of less than 0.05 was considered significant. Statistical analyses were performed using Stata 8.0 (Stata Corporation, Collegetown, TX).


Donor and recipient characteristics

Between April 1998 and August 2003, a total of 192 adult donor candidates were evaluated. Twenty-four potential donors were obese, with two-thirds of potential obese donors successfully completing donation. The 8 obese donor candidates who did not successfully complete donation were excluded for the following reasons: uncontrolled hypertension (n = 1), HCV positivity (n = 1), HBV positivity (n = 1), elevated serum transaminases with hepatomegaly and refusal of biopsy (n = 2) and failure to complete the evaluation (n = 3). One additional individual was not considered to be a potential donor, as she appeared to be super, super obese, defined as BMI >60. Her BMI was calculated to be 80, and she did not have any further medical or laboratory evaluation. For nonobese donor candidates, 52 of 168 (31%) successfully completed donation. The most common reason for not undergoing donation was failure to complete the evaluation; 39 potential nonobese donors (23%) chose not to complete the evaluation. Other common reasons included use of an alternative donor, active alcohol/drug use and ABO incompatibility.

During the study period, 68 adult-to-adult LDLTs were performed. Sixty-five donors had right lobectomies and 3 had left lobectomies (for smaller recipients). The nonobese group consisted of 52 donors who had a BMI less than 30. The obese group (n = 16) included 16 donors who had a BMI greater than 30, and 2 donors who had a BMI greater than 40 (41 & 45). The 2 donors with a BMI greater than 40 were both male, aged 32 and 37. All donors with BMI >30 appeared obese to the hepatologist and surgeon; these donors were described as obese in their medical charts. Median length of follow-up for both groups was 25 months.

Baseline donor characteristics are listed in Table 1. Overall, there were no significant differences in the demographic characteristics of the donors in the two groups. The median age at donation was 37 years for obese donors (mean = 38.4 years) and 39 years for nonobese donors (mean = 38.4 years). Thirty-seven percent of nonobese donors were Caucasian Hispanic, 53% were non-Hispanic Caucasian and 10% were of other race. Of donors with a BMI greater than 30, 25% were Caucasian Hispanic, 68% were non-Hispanic Caucasian and 7% were of other race. Recipient MELD scores ranged from 7 to 30 with a median of 14 (mean = 15.5, standard deviation 5.3) for 67 donors; 1 recipient was of UNOS status 1 at the time of transplantation.

Table 1.  Selected baseline characteristics of living donors; n = 68
Body mass indexBMI <30BMI >30Overall
Mean age (year)38.438.41.0
 Non-Hispanic Caucasian53680.39
 Caucasian Hispanic37250.54

Liver function tests and biopsy results

Overall, 4 obese donor candidates had initial LFT abnormalities. As mentioned above, 2 obese donor candidates were excluded because of these abnormalities, with imaging suggestive of fatty liver. These 2 obese donor candidates did not lose weight and or undergo liver biopsy. The other 2 obese donor candidates, with BMIs of 31 and 41, lost 9 and 5 kg, respectively, resulting in normalization of LFTs prior to donation. In addition, 2 other obese donor candidates, with normal LFTs, lost 2 kg each, prior to donation. Of nonobese donor candidates who underwent laboratory testing, seven (4%) were excluded because of LFT abnormalities.

Six donor candidates, 5 nonobese and 1 obese, with normal LFTs had pre-operative liver biopsies; the obese donor candidate had a BMI of 32. Indications for biopsy included abnormal ultrasound results (n = 3), history of elevated transaminases (n = 1), family history of PSC (n = 1) and positive ANA (n = 1). The obese donor candidate underwent biopsy for abnormal ultrasound results, which suggested fatty infiltration of the liver. Percutaneous biopsy demonstrated rare large droplets of fat, involving less than 5% of parenchyma as well as mild, nonspecific inflammation. Mallory bodies were not seen, and this nonspecific inflammation was interpreted benign steatosis, not compatible with steatohepatitis. This obese donor candidate went on to successful donation. The two other biopsies performed for abnormal ultrasound results revealed focal nodular hyperplasia and lipofuscinosis consistent with age-related changes. All other biopsies were unremarkable. No potential donor candidate was excluded on the basis of biopsy. No liver biopsies were performed intraoperatively at the time of donation, and no livers appeared grossly steatotic at the time of operation.

Donor complication rates

There were no aborted hepatectomies for either the obese or nonobese donor groups. Complications were equal between both groups, with 48% of nonobese donors and 44% of obese donors experiencing at least one complication. Obese donors experienced more wound infections than nonobese donors (25% vs. 4%, p = 0.024). However, obese donors did not develop a significantly increased number of hernias or pneumonias, both of which are known to be associated with obesity (see Figure 1). No statistically significant differences were seen between groups for any of the other complications (See Table 2 and Figure 2). A logistic regression analysis using BMI as a continuous variable against complications showed no association with overall complications as a dichotomous outcome variable, (odds ratio: 1.0, p-value 0.95). Analyses with serious and nonserious complications yielded similar results, with odds ratios of 0.97 and 1.01, respectively. No donors died; nor did any require consideration or listing for emergency transplantation. Recipient outcome was not significantly different between the two groups. Recipient survival was 80% for recipients of nonobese donors and 100% for recipients of obese donors (p = 0.1). This trend toward improved survival in the recipients of obese donors is not statistically significant. In order to have an 80% chance of having a statistically significant difference in survival of this magnitude, we would have needed 44 obese and 44 nonobese subjects.

Figure 1.

Donor complications known to be associated with obesity.

Table 2.  Complication rate by BMI

BMI >30
BMI <30

Type of complicationnnp-value
Wound infection420.024
Biliary leak040.57
Biliary stricture001.0
Hepatic dysfunction391.0
Vascular thrombosis011.0
Deep venous thrombosis001.0
Pulmonary embolism001.0
Figure 2.

Frequency of donor complications by BMI.

One nonobese donor developed a partial portal vein thrombosis. This donor had a trifurcation anatomy and intra-operative revision of the repair. This donor had normal ultrasounds, documenting a patent portal vein, on POD no. 2, 5 and 28. Long-term follow-up revealed asymptomatic late (>1 year) thrombosis of the portal vein with splenomegaly on ultrasound. These were the standard intervals for ultrasound surveillance at that time in the transplant program. This donor has not developed ascites, varices or other signs of portal hypertension and has normal LFTs and functional status.

Donors were classified as having hepatic dysfunction if their total bilirubin rose above 5 mg/dL. Nine nonobese and 3 obese donors had early hepatic dysfunction. Four of these donors only had transient elevations of total bilirubin between 5.0 and 5.5 mg/dL, lasting a maximum of 2 days. For all nonobese and obese donors with hepatic dysfunction, average peak total bilirubin levels were 5.8 and 5.7 mg/dL, respectively. All of these donors were thought to have mild small-for-size syndrome, and liver transplantation was never considered for any of them. All had resolution of hyperbilirubinemia within 7 days (mean 2.9 days).

When the more standard and stringent definitions of the FDA for medicinal products for SAEs are used, the SAE rates for obese and nonobese donors are found to be equivalent; there is no significant difference between these two groups. The SAE rate for nonobese donors remains the same as our serious complication rate, 19%. The SAE rate for obese donors increases minimally when compared to our serious complication rate (27% vs. 31%, p = NS). This minimal increase is not associated with a significant difference between the SAE rates for obese and nonobese donors.

The FDA classified six outcomes as SAEs. Three of these outcomes did not apply to any of our donors. We had no deaths among our obese and nonobese donors. None of our obese and nonobese donors had life-threatening complications, nor were they at risk of dying at any time. Moreover, none of our obese and nonobese donors experienced congenital anomalies. Of note, average hospital LOS post-donation was shorter for obese donors, compared to that for nonobese donors (6.8 days vs. 7.5 days, p = NS).

Donors did experience three outcomes, classified as SAEs. First, 3 obese and 4 nonobese donors had repeat initial hospitalizations for hernia repairs (p = NS). Second, 6 obese and 11 nonobese donors required additional medical or surgical intervention (p = NS). The number of donors requiring additional medical or surgical intervention is smaller than the number of complications because some donors experienced more than one complication. Third, all donors experienced a change in body structure, as a consequence of surgery. However, none of the obese or nonobese donors suffered disability, defined as permanent change to bodily function or physical activity. The single nonobese donor with the portal vein thrombosis has not developed ascites, varices or other signs of portal hypertension. This donor has normal LFTs and functional status, and he continues to be monitored for signs and symptoms of portal hypertension. Quality of life (QOL) is difficult to measure, and follow-up did not include use of validated QOL scales. However, chart review reveals that none of our obese and nonobese donors complained of decreased QOL.


A challenge to the current application of LDLT is expanding guidelines without compromising donor or recipient safety (15). Unfortunately, standard guidelines for reporting complications in the donor evaluation process do not exist, making interpretation of existing data on donor complications extremely difficult (16,17). Beaver et al. reviewed the literature on donor complications after LDLT and noted wide variability among reported events, with an incidence ranging from 0 to 67%, and an overall complication rate of 31% (18). We classified any minor clinical deviation following living donation as a complication. We chose to be inclusive in defining all deviations from expected outcome as a complication as (i) this is more appropriate for donors who do not directly need the surgery, (ii) this would prevent bias in a retrospective review and (iii) it would be the most sensitive manner of detecting differences between the two groups. In this study, 48% of nonobese donors and 44% of obese donors experienced at least one complication. This does not differ significantly from other reports, particularly when minor complications are rigorously defined and reported. Our serious complication rates of 27% for obese donors and 19% for nonobese donors (p = NS), our biliary complication rates of 8% in obese donors and 0% (p = NS) in nonobese donors and our SAE rates of 31% in obese and 19% in nonobese donors are equivalent to most rates reported in both obese and nonobese donors. Obese donors did experience a significantly increased rate of wound infections, a complication known to be associated with obesity (19). However, none of the obese donors with wound infections developed a disability as a result of the infection.

In 2003, our standard donor incision changed from a chevron incision to a right sub-costal incision with upper midline extension or a midline-only incision. Following that change, our hernia rate decreased. Since this approach was adopted in 2003, only eight subjects in this study had right sub-costal incisions with upper midline extensions, with only one developing a hernia during the follow-up period. Since this latter group comprises only 12% of the study group, the hernia rate remains elevated in our analysis. At this time, the rate of hernia appears to be less than the rate with chevron incisions used previously; but additional follow-up is required.

Obese donors did not experience a significantly different rate of any complications other than wound infections compared to nonobese donors. Although none of our subjects developed deep venous thrombosis or pulmonary embolism, it is important to recognize these complications are common in the post-operative setting (20–22). In addition, obesity is a risk factor for deep venous thrombosis and pulmonary embolism. Our subjects received aggressive preventive measures during the post-operative period, including pneumatic compression stockings and physical therapy to assist with early ambulation.

Wound infections and biliary complications deserve specific discussion. While wound infections are treatable, they may require several weeks of i.v. antibiotics and lead to disfigurement, chronic wound care and disability, particularly in the obese. Furthermore, the obese are at higher risk for necrotizing fasciitis. Donors in pediatric LDLT are often parents of their recipients, and they may be more willing to undergo LDLT with a higher complication rate, as they believe they directly benefit from the LDLT experience through the improved health of their children. However, adult-to-adult donation frequently involves donors outside the immediate family. As a result, a complication involving long-term treatment and potential disfigurement threatens overall donor satisfaction, and may adversely affect attitudes toward adult-to-adult LDLT. Therefore, wound infections deserve a prospective study in the future.

In our study, there were no biliary complications among obese donors; perhaps the lack of complications reflects extra precautions taken by surgeons when operating on obese donors. However, it is important to recognize that obesity may directly affect the ability to treat biliary complications, as this treatment frequently involves endoscopic retrograde cholangiopancreatography (ERCP) or interventional radiologic procedures (23). Advanced endoscopy and interventional radiologic procedures may be more difficult to perform on obese donors, thus increasing the level of danger if biliary complications occur. Still, it seems unlikely that ERCP and percutaneous cholangiography would be overly challenging in our population of obese donors, as none were super obese, defined as a BMI >50.

In order to minimize risk, some experts advocate pre-donation weight loss with rigorous diet and exercise programs. In fact, such weight loss programs have the potential to reduce excessive steatosis and expand the donor pool. Hwang et al. showed that small reductions in body weight resulted in marked decrease in microvesicular steatosis (24). However, these benefits are contingent on maintaining a reduced weight after the period of weight loss, which is historically difficult for most obese patients.

This manuscript addresses short-term complications; however, LDLT donors are at risk for long-term complications as well. The development of long-term fibrosis in LDLT donors remains contested, and it is unknown if obese donors have a greater disposition toward progression to fibrosis. Obesity is the most common risk factor for nonalcoholic steatohepatitis (NASH), which can cause progressive fibrosis (25,26). Most people with NASH are asymptomatic, and imaging studies may not reliably detect steatosis below 33% (27). NASH also occurs more frequently in women, and it is not known if female obese donors are at increased risk for fibrosis. Recipients of obese female donors may be at increased risk of hepatic dysfunction as well (28). At this point, our limited understanding of NASH permits only speculation about its long-term effects on donors as well as recipients. Prospective studies are needed to answer these important questions and to investigate the relationship between gender and type of complication. The multi-center NIH-funded Adult-to-Adult Living Donor Cohort (A2ALL) study may also provide insight into these questions (29). However, in our cohort, all of our obese donors had normal LFTs and none had NASH; thus their risk seems minimal unless they gain further weight as steatosis rarely evolves to NASH.

A majority of programs performing LDLT obtain biopsies of select donors but there is tremendous variation with respect to application of this technique (30,31). Liver biopsy is not without risk, though less than 1% of subjects undergoing liver biopsy have a significant complication (32). Even when a biopsy is obtained, there is a risk the specimen will be nonrepresentative, particularly in the obese. In our study, only 1 obese donor had a pre-donation liver biopsy. This donor and others who underwent liver biopsy had good recipient and donor outcomes, with no donor excluded solely on the basis of biopsy results. The utility of biopsy in the evaluation of obese donors with normal LFTs is unclear, but compels the following question, ‘Is biopsy necessary in the evaluation of the obese donor’? The value of liver biopsy, with respect to steatosis, in donors with normal LFTs requires further evaluation. Some centers utilize graft-recipient body weight ratio and graft weight to determine graft acceptability. We do not routinely calculate graft to body weight ratio, and do not have data on graft volumes.

Our survival was 80% among recipients who received livers from donors with a BMI <30 versus 100% for recipients who received livers from donors with a BMI >30. We suspect the outcomes in both groups are equivalent and that this discrepancy is likely due to random variation and sample size. With increasing experience and expertise in LDLT, our donor selection has been expanded to include the obese. Alternatively, it is possible that obese donors donated greater hepatic mass, contributing to improvement in recipient survival.

Our current decision algorithm in the evaluation and acceptance of obese donors is illustrated in Figure 3. It is interesting to note that a higher percentage of potential obese donors eventually donated than did potential nonobese donors, suggesting that much of the selection may occur by the potential donor prior to presenting for evaluation. The obese donors were highly motivated, losing weight and rarely dropping out of the process.

Figure 3.

Decision algorithm for donor evaluation.

The strengths of this study include a single group of transplant professionals at a single center, who made consistent decisions about donor selection, surgical procedure and post-surgical care, as well as detailed follow-up. In addition, our donor and recipient populations included a significant proportion of minorities, a demographic group at greater risk for obesity. Atypically, however, minority donors in our study were less obese than Caucasian donors. The limitations include small sample size and a retrospective analysis. In addition, our assessment of obesity was simple, based on calculation of BMI and the general appearance of the subjects, as described in the medical chart. However, in our cohort, all of these donors appeared to be obese and were described as such in the chart by physicians prior to the decision to undertake this retrospective study. Thus, it is unlikely that the data are biased. For future studies, body fat percentage will be measured more formally.

Our observations support limited expansion of living donation to include select individuals with a BMI greater than 30. This can be done with a minimal increase in donor risk and good recipient outcomes. Since LDLT represents a small fraction of all adult liver transplantations performed, it is difficult to predict how this expansion will affect the gap between adult patients awaiting transplant and available donors. Our data cannot be extrapolated to obese donors with abnormal LFTs or other significant comorbidities. With precise relaxation of donor selection, 24% of our donor population was composed of individuals with a BMI greater than 30, normal LFTs and no significant comorbidities.


BMI guidelines may be expanded for LDLT. While wound infection rates were increased among the obese, this did not correlate with significant morbidity. However, increased risk of wound infections, their potential disfigurement and requirement for long-term care may lead to overall lower satisfaction among obese donors. The management of biliary complications in the obese is also more challenging, with no data from this study to accurately support the efficacy of routinely utilized endoscopic and interventional radiologic procedures in the obese donor. The long-term risk of fibrosis in nonobese and obese donors has yet to be determined, and studies examining the impact of pre-donation biopsy are needed. Select donors with a BMI greater than 30, normal LFTs and no significant comorbidities may safely participate in LDLT.


This study is supported by NIDDK grant # DK02–010 to RSB and JCE.