Increased risk of rejection in liver transplant recipients with a history of malabsorptive bariatric surgery

Abstract

BACKGROUND

Malabsorptive bariatric surgery, including Roux-en-Y gastric bypass and duodenal switch, are known to be more metabolically effective than restrictive surgery. However, the permanent alteration of gastrointestinal anatomy from these operations has been shown to alter the kinetics of drug absorption and may make subsequent surgeries more technically challenging.

AIM

To evaluate perioperative liver transplant outcomes and rates of acute cellular rejection in recipients with prior malabsorptive bariatric surgery.

METHODS

Patients who underwent liver transplantation at a single institution between 2005-2024 with a history of malabsorptive bariatric surgery were identified. Matched controls were selected based on age, sex, listing model for end-stage liver disease (MELD), and primary liver diagnosis.

RESULTS

A total of 12 liver transplant patients with prior malabsorptive surgery and 25 controls were included. The mean age in the malabsorptive group was 50.5 years at the time of transplant and 92% were female. The mean MELD at the time of transplant was 27.6 and mean body mass index was 28. There were no significant differences in length of stay, post operative complications, or 1 year survival between the controls and malabsorptive patients. However, the malabsorptive group was significantly more likely to experience biopsy-proven and clinically treated acute cellular rejection than the controls (24% vs 66.7%, P = 0.012), more frequent rejection episodes (0.28 ± 0.53 vs 1.0 ± 0.91, P = 0.006), and earlier time to first rejection episode (P = 0.002).

CONCLUSION

Previous malabsorptive bariatric surgery in liver transplant recipients did not increase the risk of perioperative complications or mortality but significantly increased the rate and frequency of acute cellular rejection.

Key Words: Liver transplantation; Acute cellular rejection; Malabsorption; Bariatric surgery; Roux-en-Y gastric bypass; Duodenal switch

Core Tip: This study compares perioperative outcomes and episodes of acute cellular rejection between liver transplant recipients with a history of malabsorptive bariatric surgery and controls. Patients with malabsorptive anatomy have higher rates, frequency, and earlier rejection, but no differences in perioperative or 1 year survival outcomes. The mechanism of increased acute cellular rejection is unknown but may be due to altered drug absorption following intestinal bypass. Therefore, this cohort may require meticulous drug-level monitoring or specified postoperative protocols to reduce rejection risk.

INTRODUCTION

Obesity, which is estimated to affect 40% of adults in the United States, is a major cause of morbidity, which has prompted a surge in pharmacological and surgical approaches to weight loss[1]. The number of bariatric surgeries performed in the United States has rapidly expanded over the last two decades with approximately 280000 bariatric procedures performed in 2022[2,3]. Some procedures such as laparoscopic sleeve gastrectomy assist in weight loss through a restrictive mechanism[4-6]. Others, including the Roux-en-Y gastric bypass (RYGB) and duodenal switch (DS), are both restrictive and malabsorptive by reducing gastric volume and rerouting portions of the small intestines to decrease nutrient absorption[4-6]. As a result, the permanent modified gastrointestinal anatomy from these operations can have adverse malabsorptive effects such as micronutrient deficiencies and altered drug absorption[7].

Due to the surge in bariatric surgery, the population of patients with malabsorptive anatomy can be seen across all specialties, including transplantation. The number of liver transplants performed per year in the United States has been steadily rising, exceeding 10000 for the first time in 2023[8]. The current leading causes of liver failure requiring transplant in the United States are alcohol-associated liver disease (ALD) and metabolic dysfunction-associated steatohepatitis (MASH), which is a consequence of rising rates of obesity[9,10]. While bariatric surgery attempts to halt progression of MASH, the prevalence of MASH cirrhosis is expected to increase by 91% from 2020 to 2050, creating demand for 6000 Liver transplants[11].

Allograft rejection can lead to adverse long-term consequences including chronic rejection requiring re-transplantation[12]. Advancements in immunosuppressive medications have markedly decreased rejection rates[12,13]. However, drug dosing and pharmacokinetics greatly impact the successful maintenance of immunosuppression[14]. The impact of malabsorptive bariatric surgery on immunosuppression maintenance and rates of rejection has not yet been examined in the liver transplant population. Our study aims to address this gap in the literature by comparing liver transplant outcomes between patients with prior malabsorptive bariatric surgery and propensity-score matched control recipients. We hypothesize that patients with malabsorptive bariatric anatomy who undergo liver transplantation are at higher risk for acute cellular rejection.

MATERIALS AND METHODS

This is a single center retrospective cohort study. Patients with a history of malabsorptive bariatric surgery including RYGB and DS who underwent orthotopic liver transplantation (OLT) at our institution between 2005 and 2024 were identified. Matched controls who underwent liver transplantation in the same study period were selected based on age, sex, listing model for end-stage liver disease (MELD), and primary liver diagnosis. Transplant dieticians are integrated into our group and take an active role during every stage of transplantation. All patients were assessed pre-operatively by transplant dieticians in the clinic who utilized nutrition focused physical exam to diagnose nourished/moderate/severe malnutrition based on Academy of Nutrition and Dietetics/American Society for Parenteral and Enteral Nutrition criteria. Patients at risk for malnutrition were supported with careful monitoring, calorie counts, and dietary supplementation with nutritional shakes. All grafts were procured from deceased liver donors. All OLTs were performed with the standard piggyback technique by the same small group of surgeons whose practice remained comparable over the study period. Study variables were collected via retrospective chart review. Postoperative complications were graded based on the Clavien-Dindo (CD) classifications[15]. Episodes of rejection were defined as biopsy proven acute cellular rejection that were subsequently clinically treated. Patients were maintained with tacrolimus and mycophenolate mofetil in accordance with our institutional standard drug regimen unless contraindicated due to a patient’s medical status or drug toxicity. The χ², Students t-test, propensity score matching (PSM), and Kaplan-Meier curves were used for analysis. P < 0.05 was considered significant. R Studio (Ames, IA, United States) and Stata 18SE (College Station, TX, United States) were utilized for statistical analyses.

RESULTS

Twelve patients with a history of malabsorptive bariatric surgery who underwent liver transplantation during the study window were identified. Eleven of these patients had a history of RYGB, and one patient had a history of DS. The mean time between malabsorptive surgery and liver transplant was 10.5 ± 5.2 years. Twenty-five liver transplant recipient controls were also identified. There were no differences in the mean age, sex, race and ethnicity, body mass index at listing, or comorbidities between the controls and the malabsorptive groups (Table 1). The mean listing MELD was 26.5 ± 6.7 in the control group and 27.6 ± 8.1 in the malabsorptive group. The most common primary listing diagnoses was ALD, followed by MASH, hepatitis C virus, hepatitis B virus, and autoimmune hepatitis in the malabsorptive groups which was reflected in the control group.

Table 1 Background and demographic information between liver transplant recipients with malabsorptive anatomy and controls, n (%).

Variable	Control (n = 25)	History of malabsorptive bariatric surgery (n = 12)	P value
Age	50.2 ± 8.9	50.5 ± 6.8	0.93
Female sex	22 (88)	11 (92)	0.97
Race
White	17 (68)	7 (58.3)	0.71
Black	6 (24)	3 (25)
Other	2 (8)	2 (8)
Hispanic ethnicity	5 (20)	3 (25)
Comorbidities
Hypertension	5 (20)	1 (8.3)	0.37
Diabetes	6 (24)	1 (8.3)	0.26
Heart failure	1 (4)	1 (8.3)	0.97
Chronic kidney disease	3 (12)	1 (8.3)	0.74
Primary listing diagnosis
Alcohol-associated liver disease	15 (60)	7 (58.3)	0.92
Metabolic dysfunction-associated steatohepatitis	3 (12)	2 (16.7)	0.70
Hepatitis C virus	3 (12)	1 (8.3)	0.74
Hepatitis B virus	0	1 (8.3)	0.14
Autoimmune hepatitis	2 (8)	1 (8.3)	0.97
Body mass index (kg/m2)	26.5 ± 6.8	28.0 ± 6.5	0.49
Model for end-stage liver disease	26.5 ± 6.7	27.6 ± 8.1	0.68

There were no differences in peri-operative or long-term outcomes between the control and malabsorptive groups. The cold ischemia time in the control group was 6.8 ± 2.5 hours compared to 7.4 ± 2.2 hours in the malabsorptive group (P = 0.54). The mean length of stay was 20.3 ± 18.0 days in the control group compared to 25.3 ± 22.4 days in the malabsorptive group. There were no significant differences in CD complications between the groups, and no differences in 1-year survival (96% in controls vs 100% in the malabsorptive group) (Table 2). There were no issues with vascular anastomoses in either group.

Table 2 Perioperative and long-term outcomes between liver transplant recipients with malabsorptive anatomy and controls, n (%).

Variable	Control (n = 25)	History of malabsorptive bariatric surgery (n = 12)	P value
Cold ischemia time (hours)	6.8 ± 2.5	7.4 ± 2.2	0.54
Length of stay	20.3 ± 18.0	25.3 ± 22.4	0.48
Clavien-Dindo complication
Grade II	3 (12)	0	0.21
Grade IIIa	3 (12)	1 (8.3)	0.74
Grade IIIb	7 (28)	7 (58.3)	0.08
Grade IV	2 (8)	0	0.31
Grade V	0	0	-
1 year survival	24 (96)	12 (100)	0.48

After PSM on age, sex, listing MELD score, and primary listing diagnosis, the malabsorptive group had a significantly higher risk of rejection compared to matched controls with an average of 41.7% (95%CI: 3.0%-80.3%, P = 0.035). The patients with a history of malabsorptive bariatric surgery were significantly more likely to be diagnosed with and treated for acute cellular rejection compared to the control patients (24% vs 66.7%, P < 0.05), and they also experienced more frequent episodes of acute cellular rejection with a mean of 0.28 ± 0.53 episodes in the control group compared to 1 ± 0.91 episodes in the malabsorptive group (P < 0.01) (Table 3). Figure 1 is a Kaplan-Meier curve illustrating rejection-free probability post OLT which demonstrates that OLT recipients with a history of malabsorptive bariatric surgery are also at higher risk for early acute cellular rejection compared to the control group (P < 0.01).

Figure 1 Kaplan-Meier curve representing rejection free probability after liver transplantation in patients with history of malabsorptive bariatric surgery (solid line) vs propensity matched controls (dotted line). This demonstrates that recipients with malabsorptive anatomy experience rejections significantly earlier and more frequently than controls (P = 0.002).

Table 3 Graft rejection outcomes between liver transplant recipients with malabsorptive anatomy and controls, n (%).

Variable	Control (n = 25)	History of malabsorptive bariatric surgery (n = 12)	P value
Acute cellular rejection	6 (24)	8 (66.7)	0.012a
Number of rejection episodes	0.28 ± 0.53	1 ± 0.91	0.006b
1 year survival	24 (96)	12 (100)	0.48

^aP < 0.05.

^bP < 0.01.

DISCUSSION

Our study adds to the limited existing body of literature on the impact of malabsorptive bariatric surgeries on liver transplant outcomes. Our findings show no difference in perioperative CD graded complications as well as 1-year mortality rates. However, our data does support our initial hypothesis that patients with malabsorptive surgical anatomy that undergo liver transplants have a higher rate of acute cellular rejection compared to their PSM controls. Episodes of acute cellular rejection also happen earlier in the postoperative course and more frequently for malabsorptive patients compared to controls. While the mechanism for this increased risk remains unknown, there were no differences in other factors that may impact rejection rate such as the operative technique, complications involving the vascular anastomoses, or graft ischemia time between groups. Thus, it is possible that malabsorptive anatomy impacts the pharmacokinetics of maintenance immunosuppressant drugs, potentially decreasing bioavailability.

Malabsorptive surgical anatomy has been shown to alter the pharmacokinetics of various drugs in other studies[16,17]. However, the data on immunosuppressant drugs is scarce. One commonly used maintenance immunosuppressant is tacrolimus, a calcineurin-inhibitor that is slowly absorbed in the small intestine, then metabolized and excreted by the liver[18,19]. One study in the post-kidney transplant population showed decreased drug plasma concentration levels of tacrolimus following RYGB[20], but this has not been replicated for the liver transplant population. Current studies comparing the efficacy of daily vs twice daily dosing of tacrolimus[21] should also be reproduced in this population to fully understand and adequately maintain immunosuppression for patients with expedited gastrointestinal transit. The mechanism behind our finding of increased rejection in malabsorptive patients after liver transplantation is likely complex and multifactorial. The nutritional status and eating habits of the patients is another factor that could contribute to this observation. Studies on healthy subjects have shown that higher fat content of meals lowers the bioavailability of tacrolimus[22]. However, we are not able to draw any inferences in this regard as there was no available data on the patients’ eating habits or how long before or after meals they took their immunosuppression due to the retrospective nature of the study.

The volume of bariatric surgery performed in the United States continues to increase to address the ongoing obesity epidemic[2]. Malabsorptive surgeries have seen a recovery in popularity since 2018, and almost one-quarter of bariatric surgeries performed in 2022 were primary RYGB or DS[2]. Thus, the proportion of patients undergoing liver transplant with remote bariatric surgery is likely to increase, which is reflected in our cohort who underwent RYGB an average of 10 years prior to liver transplant. Our findings support existing literature that prior-RYGB does not increase risk of perioperative morbidity or 1-year mortality[23]. While our study demonstrates an increased risk of acute cellular rejection, the pathogenesis of chronic rejection and its association with acute cellular rejection is not well understood[24]. Additional longitudinal studies are needed to assess long term survival outcomes, risk of chronic rejection and graft loss in this population.

Our study has several limitations. The prevalence of liver transplant patients with prior bariatric surgery is still relatively low, thus our sample size is small. Future multicenter studies would improve the study power. Additionally, this is a single center retrospective cohort study, limiting the quality of the data and external validity. Our center did not have adequate sample size to include patients with prior restrictive surgeries, such as sleeve gastrectomy. Additional studies may utilize this comparison to further support the hypothesis about malabsorptive anatomy underpinning the increased risk of acute cellular rejection. Furthermore, the retrospective nature of our study limited our ability to investigate the pharmacokinetics of tacrolimus in our cohort. Prospective studies with protocolized serial collections of drug-levels will be necessary to test our hypothesis on malabsorption.

CONCLUSION

This study highlights considerations in this subset of patients undergoing liver transplantation. Patients with malabsorptive anatomy who undergo liver transplant may experience higher rates, frequency, and earlier rejection episodes. The mechanism of this observation is unknown but may be related to decreased drug absorption related to intestinal bypass. Our data suggests that patients with post-bariatric malabsorptive anatomy may require more diligent drug-level monitoring or specialized postoperative protocols to prevent rejection. Further studies on the complex interactions of these physiologically demanding surgeries are needed to optimize care and outcomes for this population of liver transplant patients.

ACKNOWLEDGEMENTS

The authors would like to acknowledge Mathias Aagaard Christensen, MD, PhD for his expertise in statistical analysis, guidance and contributions to the creation of the figures in this manuscript.