Adipose Tissue Can Have a Protective Effect on Outcome After Simultaneous Pancreas–Kidney Transplantation in Non-Obese Recipients

Introduction

The prevalence of overweight and obese individuals is increasing at an alarming rate in developed and developing countries throughout the world [1,2]. In Western countries, overweight and obesity are highly prevalent [3], and obesity is considered to be among the most significant threats to human health [4]. The prevalence of overweight and obesity has been increasing in dialysis patients and kidney transplant candidates. In 2001, 60% of patients with end-stage renal disease (ESRD) at the time of kidney transplant were obese or overweight, and the proportion of obese recipients increased from 11.6% in 1987 to 25.1% in 2001 [5]. Several studies suggest that obesity impairs the outcome after kidney, pancreas, or simultaneous pancreas–kidney transplantation (SPK), which affects both patient and graft survival [6–10]. The assessment of obesity is conducted based on the body mass index (BMI) according to the definition of the World Health Organization. The quantity and distribution of adipose tissue are not included in this assessment, which could affect the accuracy of obesity assessment, especially in patients with renal edema due to end-stage renal disease on the waiting list for simultaneous pancreas–kidney transplantation. Furthermore, some epidemiologic studies with large samples of chronic kidney disease (CKD) and end-stage renal disease patients have reported that worse survival was observed with a lower body mass index (BMI), and higher values of BMI reflecting overweight or obesity seemed to be associated with better survival. This phenomenon has been referred to as the obesity paradox [11,12], which may play a role in patients after simultaneous pancreas–kidney transplantation.

The aim of this study was to evaluate the effect of obesity, considering the quantity and distribution of adipose tissue, on the outcome after simultaneous pancreas–kidney transplantation.

Material and Methods

ETHICS AND STUDY PROTOCOL:

The study was registered and approved by the local ethics committee of Friedrich-Schiller University, Jena (registration number 4428-05/15), and written informed consent was obtained from the individual patients. All research was conducted in compliance with the ethics standards of the Declaration of Helsinki and the Declaration of Istanbul. Data collection and analysis were conducted retrospectively. We included 40 patients who underwent a simultaneous pancreas–kidney transplantation between 2007 and 2021 in Jena University Hospital. All patients were transplanted due to end-stage renal disease caused by type 1 diabetes mellitus (DM). Although there were no strictly defined selection criteria for recipients regarding obesity, patients with obesity-induced multimorbidity were not eligible for simultaneous pancreas–kidney transplantation SPK. This explains the small number of included obese recipients. We observed the distribution of body tissue composition and its impact on the outcome after transplantation. Since the measurement of body tissue composition had to be performed based on computed tomography (CT), 24 patients who did not undergo a CT during the evaluation procedure were excluded. This is because performing CT was not standard in the evaluation of recipients for SPK before 2015. Otherwise, there were no exclusion criteria. A standardized surgical approach and postoperative immunosuppression were performed according to the center protocol. The arterial vascularization of the pancreatic graft was prepared back-table. The superior mesenteric and splenic arteries were reconstructed using grafts from the iliac bifurcation of the donor. A medial laparotomy was performed in all cases. A sequential pancreas–kidney transplantation was performed. The arterial anastomoses were performed on the right and left common iliac arteries, and the venous anastomoses were performed on the vena cava and left common vein. The pancreatic graft was transplanted en bloc with the donor’s duodenum in the right abdomen for duodeno-jejunostomy. The kidney was transplanted in the left iliac fossa in the preperitoneal space. An antirefluxplasty of the ureter was always performed. A quadruple immunosuppression regimen was performed postoperatively using induction with anti-thymocyte globulin (ATG) for 7 days, followed by maintenance with calcineurin-inhibitor (tacrolimus), mycophenolic acid (mpa), and steroids (prednisolone). Somatostatin analogs were administered to reduce the pancreas secretions for 3 days after surgery.

:

The body tissue composition assay was performed based on CT images at the venous phase with special computer software (Synapse 3D, FUJIFILM, Tokyo, Japan). The margins of the study area are standardized at the tenth thoracic vertebra as the upper margin and the pubic symphysis as the lower margin. Depending on a standard Hounsfield unit (HU) threshold of the adipose tissue (−200 to −50 HU), the subcutaneous and visceral fat mass are calculated automatically. For a better exploration of the adipose tissue, 2 ratios were included in the analysis: overall adipose tissue of body stem to stem volume, defined as percent adipose tissue (AT%), and visceral adipose tissue of body stem to cavity volume, defined as percent visceral adipose tissue (VT%). The WHO BMI classification was used to determine the overweight/obesity state (normal weight: BMI (kg/m²) >18 to <25, overweight: ≥25 to <30, obesity class 1: ≥30 to <35, class 2: ≥35 to <40, and class 3: ≥40).

ENDPOINTS (POSTOPERATIVE OUTCOME):

Complications according to Clavien-Dindo classification and overall survival (OS) were considered as primary endpoints. Severe complications were defined as Clavien-Dindo >IIIa [13]. The following functional parameters were considered as secondary endpoints: recurrence of DM defined as de novo need for insulin therapy, de novo need for hemodialysis, glomerular filtration rate (GFR) (ml/min) (reference >90), C-peptide (ng/ml) (reference 1.1–4.4), and glycated hemoglobin (HbA1c%) (reference <6.5%).

DONOR CHARACTERISTICS:

The following donor-specific variables and their potential impact on postoperative outcome were analyzed: age, BMI, P-PASS (Pre-procurement Pancreas Allocation Suitability Score) [14], and presence of extended donor criteria (EDC) according to the official guidelines issued through the German Medical Association. EDC include the following parameters: presence of malignancy in the donor’s history, drug abuse, viral hepatitis, sepsis with positive blood culture, and meningitis.

STATISTICAL ANALYSIS:

The incidence of complication generally and severe complications specifically was considered the dependent variable. Age, BMI, sex, time between first diagnosis of DM and transplantation, adipose tissue, and psoas mass were considered independent variables. BMI was included as a commonly used parameter for estimating obesity, while the other measurements are novel, nonstandard factors for this aim. To avoid collinearity between BMI and other body composition-related factors, BMI was analyzed through bivariate Pearson’s correlation and excluded from the multivariate analysis. Box plots and the 2-sided Mann–Whitney U test were used to compare the distribution of continuous variables between independent groups. The median and range are provided for each group. Spearman’s correlation was used to examine the potential correlation between the adipose tissue and psoas mass of the patients and complication grade. To determine the effect size, we referred to Cohen’s classification (1992) as follows:

A multivariate analysis was performed through a backward stepwise binary logistic regression model. In this regression model, complications or severe complications were considered the dependent variable. Independent variables were as follows: donor’s and recipient’s age, sex, time interval between first diagnosis of DM and transplantation, AT%, VT%, and psoas mass. Kaplan-Meier analysis with log-rank test and stepwise Cox regression with the same independent variables used in the binary logistic regression were applied for survival.

The significance level was set at p<0.05. All statistical analyses were performed with IBM SPSS Statistics 25 software (IBM, Armonk, New York, USA).

Results

DESCRIPTIVE STATISTICS:

Basic cohort characteristics and descriptive statistics of the investigated variables are provided in Table 1. In total, only 6 (15%) patients were obese, while 25 (62.5%) were normal weight, and the other 9 (22.5%) patients were overweight.

Complications occurred in 26 (65%) patients (Table 2) – 17 (42.5%) of them were >IIIa according to the Clavien-Dindo classification, and 13 (32.5%) of them had to be reoperated on. Other severe complications included sepsis with multiorgan failure and necrotizing graft pancreatitis. The only patient who died in the early course after surgery due to necrotizing graft pancreatitis had a BMI of 37.1 kg/m2. The BMI correlated significantly with AT% (r=0.72, p<0.001) and VT% (r=0.58, p<0.001), but not with psoas (r=0.04, p=0.82). Males showed significantly greater AT% (p=0.03) and psoas mass (p<0.001). There was no significant correlation between age and body tissue components.

CENTRAL TENDENCY AND CORRELATION STUDIES:

Table 3 provides an overview of the univariate statistical analyses. In general, patients who developed complications and severe complications specifically showed a significant tendency to have lower total body fat (Figure 2). The subcutaneous adipose tissue was significantly correlated with fewer complications. Furthermore, AT% correlated significantly with complication grade, so the lower AT% was the higher complication grade (r=−0.35, p=0.03). There was no similar tendency or correlation regarding psoas mass. However, overall survival in obese patients (BMI ≥30 kg/m2) was significantly worse than in others (p<0.001) (Figure 3).

Recurrence of DM was not significantly different in the obese patients (p=0.43). De novo need for hemodialysis occurred in just 1 patient (BMI 25.8 kg/m²) 5 years after transplantation.

DONOR CHARACTERISTICS:

The median age of the donors was 44.5 years (range, 6–58). The median BMI was 23.5 kg/m2 (range, 15–29). The median P-PASS was 17 (range, 10–22), and 14 donors (35%) had a P-PASS <17. There was only 1 donor with 1 extended donor criteria (meningitis). The univariate correlation studies showed donor age as the only significant factor influencing the incidence of severe complications (p=0.006) (Figure 4). The other characteristics showed no significant correlation.

MULTIVARIATE ANALYSIS:

Tables 4 and 5 present an overview of regression models between morbidity and body composition. There was a similar significant negative correlation between AT% and incidence of complications (β=−0.07, p=0.015) and correlation of severe complications with AT% (β=−0.06, p=0.053). The other included independent variables in the regression analysis showed no significant correlation and were excluded stepwise backward. Cox regression analysis showed no correlation between the included dependent variables, especially body composition, and overall survival or recurrence of DM. When the 6 obese patients were excluded, a similar correlation remained between AT% and incidence of complications as the dependent variable (β=−0.10, p=0.01).

VISCERAL FAT AND GLYCATED HEMOGLOBIN:

There was generally no relationship between body composition and functional laboratory parameters in the early and long-term course after transplantation. However, at 5-year follow-up, a significant correlation between glycated hemoglobin and visceral fat was detected (r=0.53, p=0.003), VT% (r=0.47, p=0.01) (Figure 5).

Discussion

Obesity is thought to impair outcomes after simultaneous pancreas–kidney transplantation affecting graft and patient survival. Sampaio et al [8] investigated the impact of obesity on post-transplant complications, patient death, and kidney and pancreas allograft failure in 5725 type 1 diabetic SPK recipients. They found that obesity in SPK recipients was associated with an increased risk of post-transplant complications, pancreas and kidney graft loss, and patient death. However, there have been few studies on this topic, and most of these are derived from single-center analysis. Further, the quantity and distribution of adipose tissue and its impact on outcome after SPK has not been investigated yet. In our study, we aimed to consider the body composition in the assessment of obesity and its impact on outcome after simultaneous pancreas–kidney transplantation in patients with end-stage renal disease due to type 1 diabetes mellitus. To our knowledge, this is the first clinical study describing this relationship, since the current literature addresses the relationship with obesity without regarding the adipose tissue itself as a body component. We found a potentially protective effect of adipose tissue on patients who undergo a SPK. The incidence of complications and severe complications was significantly lower in patients with higher fat proportion. Percent muscle mass did not affect the outcome after transplantation. However, these recent results refer more to normosthenic patients since most of the study cohort (85%) were not obese. On the other hand, the only 2 patients who died after transplantation had obesity class II (both with BMI=37.1 kg/m2). So, the overall survival in obese patients in our study was significantly inferior to the other study group. The only patient who died in the early postoperative course had the highest visceral fat proportion in the study (VT%=37). The P-PASS (Pre-procurement Pancreas Allocation Suitability Score) was implemented in 2009 in the Eurotransplant region to facilitate recognition of suitable pancreas donors. A combination of 9 clinical parameters available at the time of donor reporting – age, BMI, ICU stay, cardiac arrest, sodium, amylase, lipase, (nor)adrenaline, and dobutamine/dopamine are calculated, providing a P-PASS score between 9 and 27 for each donor. Vinkers et al showed that consideration of a combination of pre-procurement factors can help identify a suitable pancreas donor. Therefore, we recommend that a pancreas donor score be calculated for each potential pancreas donor, and all donors with a P-PASS <17 should be considered for pancreas donation. However, there was no effect of P-PASS on the outcome in our study. Nevertheless, donor age but not BMI was another significant independent variable affecting the postoperative outcome after SPK. This matches the results of Almeida et al [15]; no effect of donor’s BMI on outcome could be detected since all donors had a BMI <29 kg/m2. The same thing made exploration of the effect of P-PASS in that study meaningful since most donors had a low P-PASS. Another visceral fat-related factor was found in the long-term follow-up of glycated hemoglobin, in which there was a significant linear correlation between visceral fat proportion and HbA1c. This point represents a new correlation, which is not yet explored in the current literature. On the other hand, we found that subcutaneous adipose tissue was in general associated with lower morbidity. This is important, since both fat quantity and distribution (subcutaneous/visceral) can affect BMI assessment. For example, in one of our recent studies on liver regeneration, we found a positive effect of subcutaneous fat distribution on liver volume gain [16]. To better investigate the role of visceral distribution on HbA1c and possibly on the recurrence of diabetes mellitus, further follow-up body composition measurements should be done; however, radiation exposure should be taken into consideration since these measurements were performed based on computed tomography in this study. There are other methods that can be used to investigate body composition that can be performed without radiation exposure. For example, bioelectrical impedance analysis (BIA) is a validated method for noninvasive assessment of body composition [17].

Frailty is a common problem in solid organ transplant candidates who are in the process of being listed for a transplant, as well as after transplantation. Frail patients have increased risk of morbidity and mortality after pancreas–kidney transplantation [18–20]. In our analysis, however, percent muscle mass was not associated with outcome after SPK. This could be due to the absence or scarcity of pathological values of psoas mass in the study cohort. The lack of predefined cutoff points for psoas mass and for fat proportion for our measurements impeded the statistical analysis and is a study limitation. Further studies on large cohorts with normal distribution are required to define cutoff points for sarcopenia using our recently proposed tool. Another limitation was the small sample size of obese patients, which was due to the eligibility criteria for SKP in our center.

Although most patients with type 1 DM are not obese, the results of this study show the potential effect of adipose tissue in non-obese recipients, independent of the level of obesity. Assessment of body composition and its potential consequences on patients undergoing evaluation for SPK is not yet part of routine clinical practice and should be further analyzed in the future.

Conclusions

We found a protective effect of adipose tissue on outcome after simultaneous pancreas–kidney transplantation, although the small sample size limits the strength of our results. Obesity may have a negative effect on outcome after SPK. Visceral fat distribution is associated with de novo diabetes mellitus.