4.1. What is the Impact of Pre-LT BMI on Post-LT Patient and Graft Survival?
Whether obesity affects patient or graft survival after LT is still a matter of debate. Despite obesity being a well-known risk factor for cardiovascular events and cancer, which are two important causes of death in patients after LT [22
], the association between pre-LT obesity and dismal outcomes is controversial. This may be due to selection bias, because obese patients usually undergo an extensive cardiac workup before getting listed, and because most programs emphasize oncologic screening after LT [22
There are several studies that have found no effect of obesity, severe obesity, or morbid obesity, on short- and mid-term patient or graft survival [21
]. Nair et al. published one of the first studies to link obesity with poor outcomes after LT, and it was based on his publication that the American Association for the Study of the Liver guidelines for LT on 2005 stated morbid obesity as a contraindication for LT [20
]. In their study, based on the United Network for Organ Sharing database, they analyzed 18,172 LTs performed on the pre-MELD era and found that the severely obese had lower five-year survival, and the morbidly obese had lower one-year, two-year, and five-year survival. Excess mortality was attributed to cardiovascular events. One of the cardinal limitations of this study is that BMI was not adjusted for ascites. After the study by Nair, there have been several others linking obesity, specifically severe and morbid obesity, with decreased survival after LT [1
]. Of note, most studies that have adjusted for concomitant comorbidities, and especially for diabetes, by either multivariate analysis [18
] or through propensity score matching [86
], have failed to show an independent role of obesity on patient´s outcomes. Whereas morbid obesity and severe obesity may have a negative impact on outcomes, overweight and mild obesity are associated with better survival [98
There are data suggesting that the interaction between BMI and NAFLD should be considered when evaluating post-LT outcomes. A recent study that interrogated the ELTR database (2002–2016) to study the outcomes of patients with NAFLD undergoing LT found, through multivariate analysis, that morbid obesity, MELD > 23, and age were independent predictors of death. However, BMI was not corrected for ascites, which limits the interpretation of the findings given the operational definition of NAFLD cirrhosis that was used (i.e., cirrhosis of unknown origin in association with BMI > 30 kg/m2
). Also, the multivariate analysis did not include other comorbidities that run along with NAFLD and that may influence survival such as the metabolic syndrome, coronary artery disease, or chronic kidney disease [5
]. Another study looking at patients undergoing LT for NAFLD cirrhosis found that patients that died early after LT tended to be older, more obese, and with a higher frequency of diabetes and hypertension. In fact, patients that had all these four characteristics had a one-year survival of only 50%, but no multivariate analysis was performed to assess the specific role of obesity. Nonetheless, as the authors pinpointed, careful pre-LT assessment is warranted before offering LT to these high-risk patients [103
At least three systematic reviews and meta-analyses have been performed looking at the role of pre-LT BMI on LT outcomes. The first of them is the meta-analysis by Saab et al. that comprised 13 studies involving 76,620 patients (72,212 non-obese and 2275 obese) and found no association between different thresholds of BMI (i.e., 25, 30, 35, 40) and mortality. Sensitivity analysis also showed no association in studies that adjusted BMI for ascites, but did find a lower survival in obese patients in a subgroup analysis according to etiology of liver disease [104
]. However, the control group in this meta-analysis was criticized because it excluded overweight patients and included malnourished patients, which are known to have a poor prognosis. A second meta-analysis published by Barone et al. that took care of the limitations of the previous meta-analysis, found that patients with morbid obesity had a higher one-, two-, and five-year mortality rates [87
]. Finally, the study by Beckmann et al. showed no differences in patient or graft survival rates at specific time points (e.g., 30-day, one-year, five-year) between obese and non-obese LT recipients. However, it was found that patients with obesity had shorter overall patient and graft survival, though there was marked heterogeneity between the studies, limiting the validity of the results. Also, no meta-regression was done. Importantly, a sub-analysis based on year of publication showed that the more recent the study, the longer the survival reported, reflecting improvement in LT care [105
]. In this case, meta-analyses, though useful, are not capable of answering whether increased mortality seen in some studies is due to the comorbidities associated with obesity, or to obesity itself.
In conclusion, due to the heterogeneity of the results, it is impossible to draw firm conclusions about whether obesity by itself is risk factor for poor LT outcomes. Nonetheless, as Thuluvath states, in real life, obesity comes as a full package [106
], and therefore, to recapitulate the good outcomes seen in some studies, it is essential to do a thorough pre-LT assessment to offer LT only to those ideal obese patients [107
]. Based on the more consistent findings on the different studies, morbid obesity, and probably severe obesity, may have some degree of impact on LT outcomes, particularly in patients with NAFLD cirrhosis and/or with diabetes. The role of overweight or mild obesity is even less clear, and they may even have protective effects. Most of the studies showing good outcomes in obese patients have evaluated short and mid-term outcomes, and there is some paucity of data on long-term survival, which may be affected by the burden of cardiovascular complications [108
], so more data are needed regarding long-term follow-up.
4.2. What is the Impact of Pre-LT BMI on Other Post-LT Outcomes?
Pre-LT BMI and weight gain after LT, which is particularly accelerated during the first two years, are risk factors for development of metabolic syndrome, NAFLD, and new onset diabetes after transplant [105
]. Therefore, weight should be closely monitored after LT, especially in those patients at increased risk of weight gain such as those who are older than 50, those with NAFLD as underlying liver disease, and those with pre-LT obesity [110
Patients with obesity have a higher rate of comorbidities associated with the metabolic syndrome when compared with patients with lower BMI [86
], with a subsequent higher cardiovascular risk. The CAR-OTL score was developed using pre-LT variables to predict the risk of cardiovascular disease within one year after LT. Though it still needs further validation, it may be used to identify patients at increased risk of this outcome in order to keep a strict control of the modifiable risk factors. Of note, BMI was not included in the model because it was not associated with adverse outcomes [115
]. A small retrospective study of 170 patients that underwent LT also showed no difference in the rate of cardiovascular complications between obese and normal-weight recipients after a mean follow up of 5 years [116
Pre-LT obesity is a risk factor for HCC recurrence. A study of 159 patients undergoing LT for HCC found that recurrence of HCC was twice as frequent in patients with overweight/obesity when compared with normal-weight individuals, and they had a shorter time-to-recurrence [117
]. Similarly, in a study of 342 patients that underwent LT for HCC, an association was found between obesity, microvascular invasion, and poor survival [118
], probably because adipokines promote angiogenesis. A study that evaluated the impact of skeletal muscle mass to visceral fat area ratio in LDLT recipients due to HCC found that a low ratio, but not BMI, was associated with a lower recurrence-free and overall survival [119
] reinforcing that BMI may not be the best tool to define obesity.
The role of obesity as a risk-factor for vascular complications after LT has been controversial. However, a meta-analysis that pooled the results of six retrospective cohort studies found no association between obesity and vascular or biliary complications. The same results were obtained considering different BMI cutoff points, and when considering an ascites-adjusted BMI [121
Finally, in terms of financial costs, a study in 700 LTs found that overweight and obesity had no impact on monetary costs [122
], but obesity does seem to delay improvement in physical quality of life initially after LT [123
4.3. How should Obesity be Treated after LT?
The implications and treatment of obesity after LT are outside the scope of this review, but in general, management should include the same steps as in other patients: Diet and exercise, pharmacologic therapy, and BS. Orlistat was evaluated in the post-LT setting and seems to be safe as long as immunosuppression levels are closely monitored [124
], but there are no data regarding its efficacy. BS is also possible after LT, but may be more technically demanding, and associated with increased morbidity when compared with non-LT patients. It is reasonable to wait at least one year after the LT to reduce the risk of rejection from modifications of immunosuppression, and to reduce the risk of infectious complications associated with immunosuppression, which is highest during the first year [67
]. Also, use of steroids has been associated with increased 30-day post-BS morbidity and mortality in the general population [125
], and most patients will receive steroids at least during the first 3–6 months after LT. The safety and feasibility of SG has been well reported [126
], and there are also reports of RYGB, intragastric balloon [129
], and robot-assisted minimally invasive BS [130
]. Laparoscopic adjustable gastric banding (LAGB) remains poorly explored in this setting, probably due to the potential for side effects in terms of infection of a foreign body in the context of a patient under immunosuppression. Considering the immunosuppression needed after LT, pharmacokinetic studies have shown that SG does not affect the kinetics of tacrolimus or mycophenolic acid, which is convenient. In opposition to this, the pharmacokinetics of tacrolimus, sirolimus, and mycophenolic acid do get significantly altered in the RYGB population, which can result in the need for higher doses [131
]. Table 2
includes pros and cons of the different surgical techniques and of the endoscopic intragastric balloon as well (RYGB, SG, LAGB, endoscopic intragastric balloon).