Description of the Study Cohort

The study cohort consisted of 181 LT recipients that included 66 (36%) females and 140 (77%) non-Hispanic whites (Table 1). The mean age of the study cohort was 60 ± 12 years. The etiology of liver disease requiring LT was MASH (45%), followed by alcohol (25%), and hepatitis C (13%). Prevalence of metabolic diseases was 44% for diabetes, 84% for hypertension, 54% for obesity, and 68% for dyslipidemia. Most patients were managed on tacrolimus (171 or 95%) as the primary immunosuppressant, followed by cyclosporine (10 or 6%). The median time from LT to lab draw was 36 (25th, 75th percentile 13, 95) months.

Table 1 Clinical characteristics of the liver transplant recipientsImpact of LT on MVX and Its Components

Compared to both lean- and BMI-matched controls, LT recipients had higher MVX, IVX, and MMX scores (Fig. 1a). No significant differences in the MVX, IVX, or MMX scores or the concentrations of their components were noted between the control groups (lean vs. BMI-matched controls). The MMX components are presented in Fig. 1b which demonstrates a significant reduction in leucine and valine concentration in LT recipients when compared to non-LT controls. Compared to lean controls, isoleucine levels were significantly higher in LT recipients however, no difference between BMI-matched controls and LT recipients was noted (Fig. 1b). The IVX components are presented in Fig. 1c and show a significant increase in GlycA levels in LT recipients when compared to controls, whereas there were no significant differences in S-HDL-P concentrations noted across groups. Even when LT and matched controls were stratified for co-morbid conditions such as diabetes or hypertension, similar trends were observed (Supplemental Fig. 2). Time from LT surgery did not impact MVX, IVX, MMX or their components (Supplementary Table 1).

Fig. 1

a Liver transplant (LT) recipients have significantly higher metabolic vulnerability index (MVX), inflammation vulnerability index (IVX) and metabolic malnutrition index (MMX) scores when compared to obese controls (matched for age-, gender- and BMI) or lean controls (matched for age- and gender, but BMI < 25 kg/m2). b Significant differences in components of MMX are noted in LT recipients compared to matched controls that contribute to higher MMX scores. c LT recipients have significantly higher serum GlycA levels but no significant differences in S-HDL-P levels

Regression analysis performed to evaluate the relationship between clinical parameters and MVX, IVX, and MMX are presented as univariate and multivariate adjusted (age, gender and presence of CAD) univariate and multivariate analysis (Supplemental Table 2). In the adjusted multivariate analysis, MVX correlated directly with serum creatinine and female gender and inversely with dyslipidemia. A trend between post-LT MASLD and MVX was noted, however, this did not reach the threshold for statistical significance (p = 0.06). IVX was positively associated with BMI and tacrolimus use and inversely with diagnosis of dyslipidemia in adjusted multivariate analysis. MMX correlated positively with female gender and inversely with statin use and post-LT MASLD diagnosis (Supplemental Table 2).

To determine if liver donor characteristics could influence MVX, IVX, or MMX, a multiple backward linear regression was performed with donor variables (diabetes, hypertension, BMI, gender, ethnicity and age). No statistically significant relationship between donor variables and MVX, IVX, or MMX were noted (Supplemental Table 3).

Relationship Between MASH Cirrhosis and MVX in LT Recipients

Among LT recipients, patients transplanted for MASH cirrhosis had lower MMX scores (57 ± 9 umol/L vs. 60 ± 7 umol/L, p = 0.035), which were related to higher concentrations of BCAA (Supplemental Fig. 3A). While IVX scores were similar between patients transplanted for MASH vs. non-MASH cirrhosis, patients transplanted for MASH cirrhosis had significantly higher concentrations of GlycA (456 ± 78 vs. 432 ± 80, p = 0.04). Overall, MVX was similar between the two groups. When the characterization was broadened to compare the three most common indications for LT (MASH, HCV, and alcohol), no significant differences between MVX, IVX, or MMX were noted (Supplemental Fig. 3B).

Development of post-LT MASLD either as recurrence (i.e. originally transplanted for MASLD cirrhosis who develop graft steatosis) or de novo (non-MASLD indication developing graft steatosis) on MVX was evaluated (Supplemental Fig. 4). No significant differences in MVX, MMX, or IVX were noted in patients who developed post-LT MASLD compared to those who did not (Fig. 2a). However, patients who developed MASLD after LT had higher serum valine (199 ± 56 umol/L vs. 180 ± 45 umol/L, p = 0.014), isoleucine (66 ± 22 umol/L vs. 58 ± 21 umol/L, p = 0.02), and GlycA (458 ± 89 umol/L vs. 425 ± 76 umol/L, p = 0.002) concentrations compared to patients without evidence of MASLD in the transplanted liver (Fig. 2b and 2c).

Fig. 2

a No significant differences in metabolic vulnerability index (MVX), inflammatory vulnerability index (IVX), and malnutrition vulnerability index (MMX) scores were noted in patients who developed metabolic dysfunction associated steatotic liver disease (MASLD) after liver transplant (LT). b The components of metabolic malnutrition index were significantly higher among patients who developed MASLD following LT. c Serum GlycA was significantly elevated in patients with post-LT MASLD

Interaction Between Metabolic Diseases and MVX in LT Recipients

Presence of diabetes, hypertension, or obstructive sleep apnea following LT did not significantly affect MVX, IVX or MMX scores. While MVX and IVX were similar in obese and non-obese patients, a trend towards a lower MMX value was noted in obese LT recipients, however, this did not reach statistical significance (57.7 ± 8.5 vs. 59.9 ± 7.8; p = 0.07). Compared to non-obese LT recipients, obese LT recipients had significantly higher concentrations of GlycA (456 ± 80 umol/L vs. 428 ± 77 umol/L, p = 0.017), valine (202 ± 57 umol/L vs. 178 ± 43 umol/L, p = 0.002), leucine (105 ± 39 umol/L, 90 ± 29 umol/L, p = 0.004), and isoleucine (67 ± 23 umol/L, 58 ± 20 umol/L, p = 0.009).

Tacrolimus use when compared to cyclosporine was associated with higher IVX scores (53.5 ± 10.1 vs. 42.1 ± 7.8, p = 0.004) but not MMX scores. The MVX scores approached significance with a value of 57.2 ± 10.1 in patients on tacrolimus compared to 50.9 ± 7.9 in patients on cyclosporine (p = 0.06).

Impact of Gender on MVX and Its Components in LT Recipients

A trend towards higher MMX and MVX scores was noted in female LT recipients compared to males, however, this did not reach statistical significance (Fig. 3a). The components of MVX including serum leucine, isoleucine, and valine levels were significantly lower among females when compared to males (Fig. 3b). In contrast, females had higher concentrations of GlycA and S-HDL-P than males (Fig. 3c).

Fig. 3

a A trend towards higher metabolic vulnerability index (MVX) and metabolic malnutrition index (MMX) scores was noted in females. b Female liver transplant (LT) recipients had lower serum concentrations of branched chain amino acids compared to males. c Females LT recipients had higher serum concentrations of small high-density lipoprotein particles (S-HDL-P) and GlycA when compared to males