Bariatric Surgery in the Peritransplant Period

Watch a video presentation of this article

Watch the interview with the author

Abbreviations

BMI

body mass index

BS

bariatric surgery

%EWL

percent estimated weight loss

LAGB

laparoscopic adjustable gastric banding

LRYGB

laparoscopic Roux‐en‐Y gastric bypass

LSG

laparoscopy sleeve gastrectomy

LT

liver transplantation

MELD

Model for End‐Stage Liver Disease

NAFLD

nonalcoholic fatty liver disease

NASH

nonalcoholic steatohepatitis

N/A

not applicable

NC

no change

P

prospective

R

retrospective

RYGB

Roux‐en‐Y gastric bypass

SG

sleeve gastrectomy

%TBWL

percent total body weight loss

Nonalcoholic fatty liver disease (NAFLD) is the most common liver disorder, with a prevalence rate in the United States as high as 46%. ¹ , ² Nonalcoholic steatohepatitis (NASH), the inflammatory phenotype of NAFLD that progresses to end‐stage liver disease, has a biopsy‐based prevalence rate of approximately 3% to 5% in the United States, with approximately 20% of these patients progressing to end‐stage liver disease. ¹ , ³ The mainstay of treatment for NAFLD is sustained weight loss and intensive lifestyle interventions; however, even with enhanced weight‐loss counseling and lifestyle interventions, only approximately one‐third of obese patients achieve long‐term, clinically meaningful weight loss. ⁴ , ⁵ Individuals who continue to progress despite these interventions progress to decompensated NAFLD cirrhosis, at which time transplantation considerations arise.

Bariatric surgery has proved effective at reducing obesity, insulin resistance, type 2 diabetes mellitus, metabolic syndrome, and mortality. ⁶ , ⁷ , ⁸ , ⁹ In patients with NAFLD, including NASH, with no to minimal fibrosis, bariatric surgery has not been associated with increased rates of adverse events or mortality. ¹⁰ , ¹¹ After bariatric surgery in patients with NASH, sustained regression of steatosis and inflammation has been shown in prospective studies. ¹² , ¹³ , ¹⁴ Per the US Organ Procurement and Transplantation Network, in 2018, 17% of patients listed for liver transplant had a body mass index (BMI) ≥ 35 kg/m², with the proportion of candidates with a BMI ≥ 40 kg/m² continuing to increase (Fig. 1). ¹⁵ We focus in this review on the role of weight‐loss surgery as an intervention in the peritransplant period associated with decompensated liver disease caused by NAFLD.

FIG 1.

Distribution of adults waiting for LT by BMI. Rates of patients with BMI >35 kg/m² and >40 kg/m² continue to increase. Reproduced with permission from American Journal of Transplantation. ¹⁵ Copyright 2020, American Society of Transplantation and the American Society of Transplant Surgeons.

Methods of Bariatric Surgery

Currently, bariatric surgery is performed almost exclusively laparoscopically, increasing safety and feasibility in complex patients. ¹⁶ Laparoscopy sleeve gastrectomy (LSG) is the most commonly performed bariatric surgery in the United States and consists of the creation of a tubular stomach by removing the majority of the greater curvature of the stomach, including the fundus ¹⁶ (Fig. 2, right). ¹⁷ The remaining stomach has decreased capacity with resistance to expansion, making LSG a restrictive procedure. In addition, postoperatively LSG has been shown to uniquely decrease fasting and postprandial ghrelin levels, leading to appetite suppression. ¹⁸ Laparoscopic adjustable gastric banding (LAGB), where an inflatable prosthetic band is placed around the upper stomach leading to a restrictive gastric pouch, has largely been replaced by LSG. ¹⁶ , ¹⁹

FIG 2.

RYGB (left) and SG (right) anatomy. Reproduced with permission from Annals of Laparoscopic and Endoscopic Surgery. ¹⁷ Copyright 2017, AME Publishing Company.

Laparoscopic Roux‐en‐Y gastric bypass (LRYGB) results in the creation and separation of a small proximal gastric pouch from the distal stomach; this gastric pouch is then anastomosed to an alimentary limb (Roux limb) of small bowel, creating a gastrojejunostomy and jejunojejunostomy ¹⁹ (Fig. 2, left). The <30‐mL gastric pouch is restrictive, whereas the small‐bowel reconfiguration leads to mild malabsorption. Expected weight loss is slightly greater after LRYGB compared with LSG; however, LRYGB has a higher rate of postoperative complications and need for reoperations. ²⁰ , ²¹ The malabsorption component of the LRYGB has also been shown to lead to alterations in immunosuppression pharmacokinetics, although in clinical practice, additional immunosuppression modification and severe graft dysfunction or rejection have not been observed when compared with LSG. ²² , ²³ , ²⁴ In a high‐risk liver transplant population, selection of bariatric surgery options must aim to maximize safety and reduce complications. As such, LSG is the preferred surgical approach, with a decreased operative time, reduced need for reoperations, lack of malabsorptive effects, and preserved gastrointestinal tract with easier access to the biliary tree along with similar weight‐loss outcomes. ²⁰ , ²¹ , ²⁵ , ²⁶

Lower‐risk endoscopic approaches, such as endoscopic sleeve gastroplasty and endoscopic gastrointestinal bypass devices, are currently being investigated as alternatives to laparoscopic procedures but are out of the scope of this review.

Safety of Bariatric Surgery in Patients With Advanced Fibrosis and Cirrhosis

Bariatric surgery for patients with advanced fibrosis and cirrhosis has been an area of continued investigation. In retrospective tertiary care single‐center studies of small numbers of patients with advanced fibrosis or well‐compensated cirrhosis who are undergoing bariatric surgery, similar improvements in weight loss and metabolic comorbidities were seen without additional complications compared with those with minimal fibrosis. ²⁷ , ²⁸ , ²⁹ However, other retrospective studies have reported increased complications in patients with cirrhosis, including acute kidney injury, wound infections, bleeding complications, anastomotic leak, and pneumonias. ³⁰ , ³¹ In the majority of studies, LRYGB was the most common procedure performed. ²⁷ , ³⁰ , ³¹ In an analysis of the Nationwide Inpatient Sample, bariatric surgery in patients with no liver disease had lower mortality than those with compensated and decompensated cirrhosis (0.3% versus 0.9% versus 16.3%; P = 0.0002). ³² Importantly, however, 40% of patients with decompensated cirrhosis underwent bariatric surgery at a low‐volume center performing less than 50 bariatric surgeries per year. Postoperative mortality rate was 41% at low‐volume centers versus 0% at high‐volume centers. ³² The safety of bariatric surgery in Child‐Pugh class B and C has not been thoroughly investigated with limited cases reported in the literature. Patients with advanced fibrosis and compensated cirrhosis who are considering bariatric surgery should be referred to a high‐volume center or enrolled in prospective studies.

Bariatric Surgery Related in the Peritransplant Period

Bariatric surgery, especially sleeve gastrectomy (SG), is being investigated in all phases of liver transplantation (LT). Currently, American Association for the Study of Liver Diseases 2013 Guidelines include class 3 obesity (BMI ≥ 40) as a relative contraindication to LT (2‐B). ³³ Obese patients are less likely to be listed for LT or to undergo transplantation once on the wait list and have a higher mortality rate on the wait list. ³³ , ³⁴ , ³⁵ In the pretransplantation period, bariatric surgery may improve candidacy, especially with BMI cutoffs for transplantation, but the surgical risk typically depends on the degree of liver decompensation, as discussed earlier. Several studies have shown significant reductions in BMI below relative cutoffs for transplantation with improvement in metabolic syndrome comorbidities (Table 1). The majority of patients had mild portal hypertension or were classified as Child‐Pugh class A, and those with increased complications, including superficial wound infections, encephalopathy, acute kidney injury, and bleeding typically occurred in patients with higher Model for End‐Stage Liver Disease (MELD) scores >11. ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ In two retrospective matched cohort studies, patients with a history of bariatric surgery who underwent LT had no difference in hospital length of stay, complication rate, rejection rate, or 1‐ and 3‐year survival compared with patients who underwent LT alone. ⁴² , ⁴³ In the study by Idriss et al., ⁴³ although patients with a history of BS were more likely to be listed for transplant than their matched counterparts, the authors found a higher rate of delisting or death on the wait list, attributed to complication from their BS. However, in both studies, RYGB was the most commonly performed procedure, as well as the most likely to have post‐LT complications. ⁴² , ⁴³

TABLE 1.

Literature Review of Pre‐LT, Simultaneous, and Post‐LT Bariatric Surgery

Reference	N	Study Design	Surgery Type	Follow‐up (months)	MELD/Child‐Pugh	%EWL	Comorbidities	Complication Rate	Complications
Pre‐LT
Shimizu et al. (2013) 37	23	R case series	LRYGB (n = 14)	12 (n = 18)	Child‐Pugh class A: 22	12 months: 67.40%	Type 2 diabetes mellitus: remission 66.7%, improved 20%, NC or worsened 13.3% (12‐month follow‐up)	34.80%	LRYGB: 1 gastrojejunal anastomotic leak, 2 gastrojejunal strictures, and 1 infected hematoma
			LSG (n = 8)	37 (n = 15)	Child‐Pugh class B: 1	37 months: 67.70%	Dyslipidemia: remission 25%, improved 41.7%, NC or worsened 33.3% (12‐month follow‐up)		LSG: 1 staple line leakage, 1 stricture, 1 case of pneumonia
			LAGB (n = 1)				Hypertension: remission 18.7%, improved 50%, NC or worsened 31.3% (12‐month follow‐up)		LAGB: 1 postoperative hemorrhage (requiring transfusion)
Lin et al. (2013) 37	20	R case series	LSG	6‐48	Mean MELD score: 11 (range: 6‐21)	12 months: 50%	Liver disease: 35% of patients went on to receive LTs (mean time between LSG and transplant was 16.6 months)	25%	2 superficial wound infections, 1 staple line leak, 1 postoperative bleeding requiring transfusion, 1 transient encephalopathy, 1 temporary renal insufficiency
						24 months: 66%	Type 2 diabetes mellitus: 61.5% improved, with 87.5% of improved patients showing complete remission
Rebibo et al. (2014) 29	13	R matched case‐control	SG, LSG	6‐35 (median of 18 months)	Median MELD score: 7 (range: 7‐8)	12 months: 73.40%	Information not available	7.70%	Intra‐abdominal hematoma
					Child‐Pugh class A: 13				Conversion to laparotomy was required in 1 patient due to injury to left gastroepiploic vein, bleeding
Pestana et al. (2015) 28	14	R case series	LSG (n = 11)	12 months (n = 12)	MELD range 6‐9	%TBWL 12 months: 24.80%	Type 2 diabetes mellitus: remission 60% (12‐month follow‐up)	0%	No perioperative or postoperative bleeding or surgical complications
			LRYGB (n = 3)		Child‐Pugh class A: 14
Hanipah et al. (2018) 39	13	R case series	LSG (n = 10)	12 months (n = 11)	Median MELD score: 9 (range 7‐17)	12 months: 37.50%	Type 2 diabetes mellitus: improved 100%, remission 80% (12‐month follow‐up)	23%	1 wound infection requiring debridement, 1 subcutaneous hematoma managed by drainage, 1 intra‐abdominal hematoma
			LRYGB (n = 3)	24 months (n = 9)	All with portal hypertension (determined by endoscopy, imaging, or intraoperative findings)	24 months: 49%	Dyslipidemia: improved 80% (12‐month follow‐up)
							Hypertension: improved 86% (12‐month follow‐up)
Sharpton et al. (2019) 40	32	R cohort study	LSG	12 months (n = 24)	Median MELD score: 12 (range 7‐18)	12 months: 52.4%	Liver disease: 88% deemed eligible to be actively listed	9.40%	1 renal insufficiency requiring albumin, 1 transient encephalopathy secondary to medications (Child‐Pugh class B)
					Child‐Pugh class A: 15		Type 2 diabetes mellitus: improved 64%, remission 36%		1 gastric staple line leak caused by retained orogastric tube (Child‐Pugh class A)
					Child‐Pugh class B: 17		Hypertension: 83% had reduction in antihypertensive medication use		No reoperations or perioperative death
García‐Sesma et al. (2019) 41	8	R cohort study	SG (n = 1)	12 (n = 7)	Child‐Pugh class A: 6	12 months: 76.3%	Liver disease: 2 patients underwent LT within 12 months	0%	No postoperative morbidity or mortality by Dindo‐Clavien criteria
			LSG (n = 7)	60 (n = 3)	Child‐Pugh class B: 2
Takata et al. (2008) 36	6	P cohort study	LSG	3‐18 (66.7% were followed ≥9 months)	Child‐Pugh class A: 4	≥9 months: 33%	Diabetes, hypertension, and obstructive sleep apnea improved or resolved in all patients with ≥6 months of follow‐up	33.30%	1 patient developed bleeding and later; 1 patient with hepatic encephalopathy secondary to a urinary tract infection
						Child‐Pugh class B: 2
Jung et al. (2017) 38	9	P cohort study	LSG	4.7 ± 4.2	Median MELD score: 7 (range 6‐14)	Mean: 35.3% ± 23.6%	Type 2 diabetes mellitus: remission 66.6%, improved 33.3%	0%	No complications or acute liver decompensation during follow‐up
							Hypertension: improved 71.4%, NC or worsened 28.6%
With LT
Nesher et al. (2017) 46	3	R case series	SG	13 months (range 3‐24 months)	Median MELD: 24	%TBWL: 27.90%	Type 2 diabetes mellitus: remission in 66.6%	33.30%	One patient with a biliary leak treated with reoperation, as well as acute renal failure secondary to overdiuresis, both of which resolved
							Hypertension: remission in 66.6%
Heimbach et al. (2013) 45	7	P cohort study	SG	8‐33 (mean of 17 months)	Mean biological MELD	Post‐LT BMI ≥ 35: LT+SG 0% versus LT alone 60% (P = 0.001)	Type 2 diabetes mellitus post‐LT: LT+SG 0% versus LT alone 34% (P = 0.03)	28.50%	Late hepatic artery thrombosis, excess weight loss; steroid‐resistant rejection; early graft dysfunction, leak from gastric staple line
					LT+SG: 32 (range: 11‐40) versus LT alone: 19 (range: 8‐35) (P < 0.001)
Zamora‐Valdes et al. (2018) 44	29	P cohort study	SG	56.28 ± 28.2	Biological MELD	%TBWL 36 months: LT+SG 34.8 ± 17.3 versus LT 3.9 ± 13.3 (P < 0.001)	LT+SG versus LT at last follow‐up:	44.80%	Late hepatic artery thrombosis, excess weight loss; steroid‐resistant rejection; leak from gastric staple line; reoperation for bleeding; 2 hemodialysis requirements; severe reflux; recurrent hepatocellular carcinoma
					LT+SG: 32.0 ± 9.5 versus LT alone 18.9 ± 8.1 (P < 0.001)
							Type 2 diabetes mellitus: 30.8% versus 58.3% (P = 0.114)
							Dyslipidemia: 53.8% versus 69.4% (P = 0.178)
							Hypertension: 23.1% versus 63.9% (P = 0.021)
							Metabolic syndrome: 23.1% versus 52.8% (P = 0.104)
							Hepatic steatosis: 23.1% versus 66.7% (P = 0.01)
Post‐LT
Tsamalaidze et al. (2018) 51	12	R case series	LSG	25.3 ± 5.1	N/A	12 months: 50%	Complete resolution of:	33%	Sleeve dilation for poor intake, late drain removal, gastrostomy tube for poor intake
							Type 2 diabetes mellitus: 44%		Length of stay was 3.1 days in LT group versus 1.7 days in no‐LT group (P < 0.001)
							Cardiac disease: 25%
							Dyslipidemia: 43%
							Hypertension: 27%
Osseis et al. (2018) 50	6	R case series	SG, LSG	13‐101 (median of 37.2 months)	N/A	12 months: 76%	Hypertension (12 months): 50% had resolution	33.30%	Leak from gastric staple line requiring reintervention by RYGB followed by death 19 months after LSG
							Obstructive sleep apnea (6 months): 66.7% had resolution		Wound infection of mesh
Ayloo et al. (2020) 49	2	R case series	LSG	16 months postoperative	N/A	12 months: 65%	Type 2 diabetes mellitus: 50% complete remission, 50% improved	0%	None
						16 months: 63%	Hypertension: 100% improved
							Dyslipidemia: 50% improved

Zamora‐Valdes et al. ⁴⁴ provided updates and long‐term outcomes to an earlier experience published in Heimbach et al. ⁴⁵ of combined liver transplant (LT) with concurrent SG in obese patients who did not lose weight prior to LT. Patients who underwent simultaneous LT and SG had higher biological MELD scores than those undergoing LT alone. Increases in operative time and additional postoperative complications, including staple line leak and excessive weight loss, occurred. Three years posttransplant, all simultaneous LT and SG maintained >10% loss in total body weight versus only 29.4% of the LT alone cohort (P < 0.001) (Fig. 3). Patients with simultaneous LT and SG also maintained a higher percentage of total body weight loss, with a lower prevalence of hepatic steatosis and improved insulin resistance (Table 1). ⁴⁴ , ⁴⁵ , ⁴⁶ Preliminary findings from small prospective studies suggest SG simultaneous with LT leads to more durable weight loss and improved metabolic risk factors, but increased complications and adaptations to two major surgeries need to be considered.

FIG 3.

Comparison of percentage of total body weight loss among patients who underwent medical therapy followed by LT versus those who underwent LT plus SG. Reprinted with permission from Hepatology. ⁴⁴ Copyright 2018, American Association for the Study of Liver Diseases.

Recurrent NAFLD and NASH are seen in 88.2% and 41.2% of posttransplant patients, respectively, with up to 20.6% having bridging fibrosis on liver biopsies from 2 to 8 years posttransplant. ⁴⁷ Recurrent NAFLD is earlier in onset, more severe, and less reversible when compared with de novo NAFLD posttransplantation. ⁴⁸ Bariatric surgery after LT is considered when signs of recurrent steatosis occur, but safety in the posttransplant population has been limited to retrospective studies with appreciable variation in the preoperative evaluation of the patients by center. Case studies of LSG post‐LT have shown improvements in steatosis, as well as obesity and its associated comorbidities, without adverse effects to graft function; however, postoperative complications remain a concern. ⁴⁹ , ⁵⁰ The largest retrospective case‐control study of 12 post‐LT patients undergoing LSG found similar changes in BMI and resolution of comorbid conditions. Although patients post‐LT had longer hospital stays, there was no increase in morbidity, change in doses of immunosuppressive medications, or difference in postoperative complications. Increased frequency of poor oral intake and malnutrition necessitating placement of a gastrostomy tube, however, was seen. ⁵¹

Although these studies suggest bariatric surgery is likely beneficial in the peritransplant period, additional prospective studies with sustained follow‐up and direct comparisons of surgical timing with respect to LT are needed to develop updated, evidence‐based guidelines. As obesity‐related liver diseases such as NAFLD increase in the proportion of total liver transplants, effective intervention for sustained weight loss should be optimized to prevent the development of advanced liver disease in many and the recurrence of disease in others.