EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Key words : End-stage liver disease, Liver transplantation, Morbidity, Muscle strength, Skeletal muscle mass

Sarcopenia is an important metabolic disorder associated with end-stage liver disease (ESLD) and is defined as the loss of muscle strength and skeletal muscle mass that is exacerbated with aging.^1,2 Causes of sarcopenia in ESLD include increased protein catabolism, inadequate nutrient intake, chronic inflammatory state, and decreased physical activity. The prevalence of sarcopenia in ESLD varies between 22% and 70%.³ The most effective treatment for ESLD is liver transplant (LT). Presently, the Model for End-Stage Liver Disease (MELD) and MELD sodium (MELD-Na) scores are used to predict mortality for patients on the LT waiting list. The MELD score primarily evaluates liver and kidney functions, so the nutritional status of the patient is ignored. Therefore, it is insufficient to predict mortality. Sarcopenia in LT candidates is an important risk factor for (1) increased mortality in patients awaiting transplant, (2) posto-perative complications (especially infections), and (3) posttransplant mortality.^4,5 The European Working Group on Sarcopenia in Older People 2018 Revised Guidelines divided sarco-penia into 3 groups for staging: probable sarcopenia, sarcopenia, and severe sarcopenia. Various methods are used to measure muscle mass.^6,7 Pretransplant liver computed tomography (CT) is routinely used in the evaluation of ESLD.⁸ Computed tomography is a good and reliable method to evaluate adipose tissue distribution. Subcutaneous fat tissue, visceral fat tissue, and skeletal muscle are measured using CT values specific to these tissues.⁹ Recent studies have shown that the increase in intramuscular fat content ratio (IMAC) with aging is a potential contributor to decreased muscle strength and muscle quality.¹⁰ Low pretransplant skeletal muscle mass, poor skeletal muscle quality, and visceral adipose tissue are independent risk factors for death in LT.^11,12 We evaluated the effect of 3 pretransplant indicators (muscle mass, muscle quality, and visceral adipose tissue) on mortality, morbidity, and prognosis after LT.

Materials and Method

From January 2015 to February 2020, 65 patients who underwent LT and had pretransplant liver CT images in their medical records were included in the study. All patients were older than 18 years of age and had undergone LT in the past. Liver transplants were from both deceased donors and living donors. Living donors were first-degree and second-degree relatives of the respective LT recipients. Our study was approved by Ankara Bilkent City Hospital Scientific Research Evaluation and Ethics Committee (decision No. 72300690-799). Patient histories and medical records were obtained from patient files and digital records of our hospital.

Images were obtained on a computed tomography machine (GE Revolution 512 Slice). Muscle and adipose tissue measurements were performed at a workstation with Advantage Workstation software (version 4.7, GE Medical systems) at the L3 vertebral level on axial tomography images. To determine the muscle tissue area, -29 to 150 Hounsfield units (HU) were used as cutoff values, and transversus abdominis, erector spina, psoas, quadratus lumborum, and the external and internal oblique muscle groups were included in the measurement.6 Similarly, -190 to -30 HU and -150 to -50 HU were set as cutoff values to determine the subcutaneous and visceral adipose tissue area, respectively. For each patient, we calculated skeletal muscle mass index (SMI, calculated as muscle tissue area in centimeters squared divided by height in meters squared), visceral-to-subcutaneous fat ratio (VSR, calculated as visceral adipose tissue area divided by subcutaneous adipose tissue area; an indicator of visceral adiposity), and intramuscular adipose tissue content ratio (IMAC, calculated as multifidus muscle mass divided by the subcutaneous adipose tissue mean density; an indicator of muscle quality). In the study, 54 healthy donors who had abdominal CT data in the system for the period 2013-2018 were assigned as a control group. Cutoff values were calculated with median data. The normal median cutoff values for SMI, VSR, and IMAC were 54.64, -0.48, and 0.24 for women and 68.59, -0.52, and 0.645 for men, respectively.

Analysis parameters

First, VSR, SMI, and IMAC values were analyzed with patient age, lymphocyte count, body mass index (BMI, calculated as weight in kilograms divided by height in meters squared), the MELD score, the MELD-Na score, and the Child-Pugh score. Patients were divided into groups according to skeletal muscle mass (low SMI vs normal SMI), skeletal muscle quality (normal IMAC vs high IMAC), and visceral adiposity (high VSR vs normal VSR). Overall survival probability after LT was evaluated according to SMI, VSR, and IMAC groups.

Statistical analyses

The normality of the distribution of continuous variables was assessed via the Kolmogorov-Smirnov test. Normally distributed continuous variables were defined as mean values (±SD) and were compared via the t test, whereas nonnormally distributed continuous variables were given as median values (with IQR) and were analyzed with the Mann-Whitney U test. Cate-gorical variables were expressed as frequency (with percentage) and were compared via the chi square test or the Fisher exact test, as appropriate. A 2-tailed Spearman correlation was used to analyze correlations among the 3 pretransplant indicators (SMI, IMAC, and VSR) versus age, the MELD score, time to transplant from diagnosis, the MELD-Na score, the Child-Pugh score, operation time, pretransplant BMI, length of postoperative ICU stay, and neutrophill-ymphocyte ratio. The results are given as correlation coefficients (R) and P values. Independent factors were evaluated by univariate Cox regression analysis. Multivariate Cox regression analysis including the variables asso-ciated with P < .25 in univariate analyses was used to analyze the independent predictors of mortality. Survival analyses were performed with the Kaplan-Meier methods and the log-rank test. A 2-tailed test with P < .05 was considered significant. We used IBM SPSS software (version 25.0, for Windows) for statistics.

A total of 65 patients were included in the study. The median age was 55 years (IQR, 45-63 years), and 48 (73.8%) of the patients were men. About two-thirds of the transplants were from living donors. The most common comorbidity was diabetes mellitus in all study groups and in the survival group. In the deceased group, diabetes mellitus and asthma/chronic obstructive pulmonary disease were the most common comorbidities. Approximately 40% of the patients were in the Child-Pugh class C category. The median MELD score was 24 (IQR, 18-24), and the median MELD-Na score was 24 (IQR, 20-24). Forty-seven (72.3%) patients developed postoperative complications. The mean follow-up period was 47.58 ± 5.57 months, which was significantly longer in the survival group versus the deceased group (P < .001). The deceased patient group had a significantly higher pretransplant VSR versus the survival group (P = .046). For laboratory variables, living patients had statistically significantly higher preoperative hemoglobin and lymphocyte values and lower C-reactive protein (CRP) values versus deceased patients (P = .010, P = .033, and P = .007, respectively) (Table 1). Although preoperative low hemoglobin and lymphocyte values and preoperative high creatinine and CRP values were found to be statistically significant predictors of mortality in univariate Cox regression analysis, no variable was found to be a statistically significant independent predictor of mortality in multivariate regression analysis (P > .05 for all parameters)(Table 2). (Table 3) shows baseline characteristics and perio-perative data for pretransplant SMI (low/normal subgroups), pretransplant IMAC (high/normal subgroups), and pretransplant VSR (high/normal subgroups). When the low and normal pretransplant SMI groups were compared, only the pretransplant BMI value was statistically significantly higher in the normal pretransplant SMI group (P = .007). When the pretransplant IMAC subgroups were compared, the group with high pretransplant IMAC was signi-ficantly older (P = .040) and had higher MELD score, MELD-Na score, and pretransplant BMI versus the group with normal pretransplant IMAC (P = .023, P = .001, and P = .010, respectively). In addition, this group had a statistically significantly higher rate of hepatocellular carcinoma (P = .010). On the other hand, the rate of viral hepatitis was statistically significantly higher in the group with pretransplant normal IMAC compared with the group with pretransplant high IMAC (P = .039); no statistically significant difference was found in other data (P > .05, for all parameters). There were statistically significant differences in MELD score, CRP, and postoperative intensive care unit (ICU) length of stay between the high and normal pretransplant VSR groups, which were higher in the normal pretransplant VSR group (P = .028, P = .048, and P = .027, respectively). During the follow-up period, 53 (81.5%) patients in the study group survived, with a mean survival time of 71.73 ± 3.81 months. There was no statistically significant difference in survival in the pretransplant SMI (low/normal subgroups) (Figure 1A), pretran-splant IMAC (high/normal subgroups) (Figure 1B ), and pretransplant VSR (high/normal subgroups) (Figure 1C) subgroups (P = .732 for SMI, P = .181 for IMAC, and P = .094 for VSR). For continuous variables with pretransplant SMI, pretransplant IMAC, and pretransplant VSR, there was a weak positive correlation between pretransplant SMI and age (R = 0.251, P = .043) (Figure 2A) and a moderate positive correlation with pretransplant BMI (R = 0.573, P < .001) (Figure 2B). There was a moderate positive correlation between pretransplant IMAC and age (R = 0.529, P ? .001) (Figure 2C) and a weak positive correlation with pretransplant BMI (R = 0.361, P = .003) (Figure 2D). Furthermore, pretransplant VSR showed a weak positive correlation with age (R = 0.306, P = .013) (Figure 2E) and a weak negative correlation with MELD score (R = -0.301, P = .016) (Figure 2F). No statistically significant difference was found in other correlation analyses (P > .05 for all analyses) (Table 5). (Figure 3) shows the patient groups according to all 3 sarcopenia indicators (only 1 indicator positive, or 2 indicators positive, or 3 indicators positive) as Kaplan-Meier survival curves. There was 100% survival in the group positive for 1 indicator (9 patients), 86.2% survival for 2 indicators (29 patients), and 70.4% survival for 3 indicators (27 patients); although there was a numerical decrease in survival, no statistically significant difference was found among the groups (log-rank chi square, 4.681; P = .096).

Liver transplant is the curative treatment that provides a chance of long-term survival in patients with ESLD, hepatocellular carcinoma, or acute liver failure.^4,13 However, despite such advances, some patients still exhibit poor prognosis after LT. Identification, prevention, and treatment of poor prognostic factors after LT will reduce mortality and morbidity and improve quality and duration of life in LT patients. End-stage liver disease is associated with sarcopenia, and sarcopenia causes increased mortality in candidates awaiting LT and is also an independent predictive factor for mortality after transplant.⁵ Hamaguchi and colleagues reported lower survival and more complications in patients with sarcopenia in a retrospective study of 250 LT recipients.¹² Similarly, other studies have reported that sarcopenia caused increased morbidity and mortality after LT and negatively affected survival.^14,15 In our study, pretransplant VSR values were significantly higher in patients who developed mortality during follow-up versus the survival group, but we did not detect a significant effect of this on survival. We also found that SMI was higher and IMAC was lower in the survival group in terms of numerical and percentage ratios, but no statistical difference was found compared with the mortality group. In our study, we did not find a significant effect of sarcopenia on survival, contrary to studies in the literature. This result could be due to the small number of patients and the fact that all of our patients had sarcopenia according to the pretransplant CT result. Therefore, we suggest that prospective, randomized controlled studies with a large population and with patients without sarcopenia may be enlightening. The respective cytokine production profiles of subcutaneous adipose tissue and visceral adipose tissue are different. Subcutaneous adipose tissue secretes anti-inflammatory adiponectin, whereas visceral adipose tissue secretes proinflammatory tumor necrosis factor ? and interleukin 6. However, increased adipose tissue causes abnormal cytokine secretion from adipocytes and macrophages, leading to chronic low-grade inflammation and subsequent activation of proinflammatory signaling pathways. Inflammatory parameters have been shown to be higher in patients with sarcopenia.^16,17 In another study, the relationship between VSR and pretransplant inflammatory state was shown.¹² In our study, we found that high CRP was associated with high VSR, and again, high CRP had an effect on posttransplant mortality. Various studies have reported that sarcopenia increases the ICU length of stay. In a study conducted by Kalafateli and colleagues with 232 LT candidates, sarcopenia was associated with longer ICU stay.² In our study, ICU length of stay was significantly higher in patients with high VSR, but we did not find a relationship between this difference and SMI and IMAC. We suggest that VSR may be a more important parameter on ICU length of stay and survival versus IMAC and SMI in patients with sarcopenia. With aging, subcutaneous adipose tissue decreases and visceral adipose tissue increases, and the differentiation pathway of preadipocytes into mature adipocytes becomes dysfunctional. As a result, inflammation increases with age. Hamaguchi and colleagues reported a significant correlation between age and SMI, IMAC, and VSR.¹² In our study, we also found a statistically significant correlation with age for SMI, IMAC, and VSR. The MELD scoring system was designed to help predict posttransplant mortality. Although many studies found no association between MELD score and sarcopenia, in a study by Masuda and colleagues with 204 patients, MELD was found to be statistically significantly higher in patients with sarcopenia.¹⁸ Similarly, in a study conducted by Montano-Loza and colleagues with 669 patients, MELD was found to be statistically significantly higher in patients with sarcopenia.¹⁹ Similarly, in our study there was a statistically significant relationship between IMAC and MELD and between VSR and MELD. We believe that prospective and randomized studies are needed to clarify the relationship between MELD and sarcopenia. In our study, survival was 100% if 1 of the sarcopenia indicators detected on CT in the pretransplant period was positive, whereas survival decreased to 70% if 3 of these indicators were positive. Similarly, in another study, mortality rate increased as the number of positive parameters increased.¹² Therefore, the study of parameters indicating sarcopenia with objective measurements in the pretransplant period allows the identification of patients with a higher probability of survival after transplant. In addition, patients with severe sarcopenia may receive pretransplant nutrition and physical therapy. Our study has several limiting factors. Most importantly, the study was a single-center retros-pective study with a small number of patients. Another limiting factor was that all patients included in the study had sarcopenia.

Pretransplant sarcopenia affects mortality and morbidity in posttransplant follow-up. Computed tomography provides valuable information in the evaluation of sarcopenia in patients with cirrhosis. As the parameters indicating pretransplant sarco-penia increase, the posttransplant mortality rate increases. However, VSR, among other sarcopenic parameters, is a more valuable indicator than IMAC and SMI to predict mortality and morbidity in LT patients.

Volume: 21 Issue: 2 February 2023

FULL TEXT