Abstract
Background
Pre–liver transplant (LT) sarcopenia is associated with poor survival. Methods exist for measuring body composition with use of CT scans; however, it is unclear which components best predict post-LT outcomes.
Purpose
To quantify the association between abdominal CT–based body composition measurements and post-LT mortality in a large North American cohort.
Materials and Methods
This was a retrospective cohort of adult first-time deceased-donor LT recipients from 2009 to 2018 who underwent pre-LT abdominal CT scans, including at the L3 vertebral level, at Johns Hopkins Hospital. Measurements included sarcopenia (skeletal muscle index [SMI] <50 in men and <39 in women), sarcopenic obesity, myosteatosis (skeletal muscle CT attenuation <41 mean HU for body mass index [BMI] <25 and <33 mean HU for BMI ≥25), visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), and VAT/SAT ratio. Covariates in the adjusted models were selected with use of least absolute shrinkage and selection operator regression with lambda chosen by means of 10-fold cross-validation. Cox proportional hazards models were used to quantify associations with post-LT mortality. Model discrimination was quantified using the Harrell C-statistic.
Results
A total of 454 recipients (median age, 57 years [IQR, 50–62 years]; 294 men) were evaluated. In the adjusted model, pre-LT sarcopenia was associated with a higher hazard ratio (HR) of post-LT mortality (HR, 1.6 [95% CI: 1.1, 2.4]; C-statistic, 0.64; P = .02). SMI was significantly negatively associated with survival after adjustment for covariates. There was no evidence that myosteatosis was associated with mortality (HR, 1.3 [95% CI: 0.86, 2.1]; C-statistic, 0.64; P = .21). There was no evidence that BMI (HR, 1.2 [95% CI: 0.95, 1.4]), VAT (HR, 1.0 [95% CI: 0.98, 1.1]), SAT (HR, 1.0 [95% CI: 0.97, 1.0]), and VAT/SAT ratio (HR, 1.1 [95% CI: 0.90, 1.4]) were associated with mortality (P = .15–.77).
Conclusions
Sarcopenia, as assessed on routine pre–liver transplant (LT) abdominal CT scans, was the only factor significantly associated with post-LT mortality.
© RSNA, 2022
See also the editorial by Ruehm in this issue.
Summary
Among the abdominal CT–based body composition measurements typically available before liver transplant, sarcopenia was the only binary factor significantly associated with posttransplant outcomes.
Key Results
■ In this retrospective cohort of 454 adults who underwent deceased donor liver transplant (LT), sarcopenia based on pre-LT abdominal CT (including the third lumbar vertebral level) was associated with post-LT mortality (hazard ratio [HR] after adjustment, 1.6; Harrell C-statistic: 0.64; P = .02).
■ Every 5-cm2/m2 reduction in skeletal muscle index was associated with a 1.1-fold increased risk of death (HR after adjustment, 1.1; C-statistic: 0.64; P = .01).
■ There was no evidence that myosteatosis was associated with mortality (HR after adjustment, 1.3; C-statistic: 0.64; P = .21).
Introduction
The term body composition refers to the proportions of fat, water, bone, and muscle within the body. The amount and distribution of body fat and the amount and composition of muscle mass, in particular, substantially affect health outcomes (1). The most common method to predict outcomes in liver transplant (LT) is the Model for End-Stage Liver Disease (MELD) score. Although it is a good predictor of pre-LT outcomes, the MELD score is considered a weak predictor of posttransplant outcomes (2). We do not have any imaging markers available to predict outcomes before transplant. Therefore, having an objective method to predict outcomes is essential in the LT recipient population.
In the past decade, efforts to measure body composition, specifically sarcopenia and myosteatosis, in patients with chronic diseases and cancers have been rapidly expanding as noninvasive and objective radiographic measurement tools develop (3). Among patients awaiting LT, loss of muscle mass is common and is associated with higher mortality rates and inferior post-LT outcomes (4). A valid and objective measure of body composition is necessary in this patient population to guide pre-LT planning and counseling.
Several radiographic methods can be used to measure body composition, including MRI, CT, dual-energy x-ray absorptiometry, and US (5). Among these methods, MRI and CT are the most valid methods for measuring sarcopenia, and both are performed routinely as part of the pre-LT evaluation (6). Sarcopenia, a loss of muscle mass resulting in reduced muscle function (6), is strongly associated with physical disability, falls, and mortality. Myosteatosis, which is the infiltration of fat into muscle and reduced muscle radiodensity, is also associated with negative outcomes in various diseases, including post-LT outcomes (7,8); however, the association between myosteatosis and poor post-LT outcomes is still unclear.
CT-based methods can be used to assess body composition (eg, sarcopenia, sarcopenic obesity, myosteatosis, visceral adipose tissue [VAT], subcutaneous adipose tissue [SAT], and VAT/SAT ratio). These data can provide objective measurements of muscle and fat components in patients undergoing LT. Patients undergoing LT typically have severe ascites and edema; therefore, body mass index (BMI) is not a good indicator of prognosis. Other studies of LT recipients have analyzed one or a few body composition components, but studies that measure all components and compare their associations with post-LT outcomes remain lacking (8–10). Our purpose was to quantify the associations between abdominal CT-based measurements (sarcopenia, sarcopenic obesity, myosteatosis, VAT, SAT, and VAT/SAT ratio) of body composition and post-LT outcomes in a large North American cohort.
Materials and Methods
The institutional review board (no. 00224066) approved this study, which was conducted in compliance with the Health Insurance Portability and Accountability Act and the current version of the Declaration of Helsinki. Because this study was retrospective research, the institutional review board waived the requirement for written consent forms for all patients.
Study Sample
We performed a retrospective cohort study of our electronic medical records for all patients at our academic tertiary hospital who underwent LT between January 2009 and December 2018 and found 893 consecutive patients. We included 454 adult first-time deceased-donor LT recipients who had a pre-LT abdominal CT scan available for review. We excluded recipients who lacked a CT scan obtained within 1 year before LT, underwent a living donor transplant, were younger than 18 years, had undergone a previous LT, had severe ascites or anasarca, had a status of 1A (acute liver failure) at the time of LT, or whose abdominal CT scan did not include the third lumbar vertebra (Fig 1). For clinical and methodologic homogeneity, we excluded living-donor recipients because clinically they are treated differently, and they are phenotypically distinct from deceased-donor recipients.
Figure 1:
Flowchart shows selection for the study sample of 454 adult first-time deceased-donor liver transplant (LT) recipients at Johns Hopkins Hospital who underwent LT between January 2009 and December 2018. EMR = electronic medical record, PACS = Picture Archiving and Communication System.
Data Sources
We linked electronic medical record data from our center to data from the Scientific Registry of Transplant Recipients to ascertain recipient characteristics (including race and ethnicity), CT-based body composition measurements, and post-LT mortality. We collected race and ethnicity data because LT outcomes are strongly associated with these factors. The Scientific Registry of Transplant Recipients data system comprises data on all donors, waitlisted transplant candidates, and transplant recipients in the United States, as submitted by the members of the Organ Procurement and Transplantation Network. The Health Resources and Services Administration, an agency of the U.S. Department of Health and Human Services, provides oversight for the activities of the Organ Procurement and Transplantation Network and Scientific Registry of Transplant Recipients contractors (11).
CT Imaging and Segmentation
A single cross‐sectional axial CT image at the third lumbar vertebral level was analyzed per recipient; this level is the most commonly used section to measure total body muscle, visceral fat area, and SAT (12–14). We used CT scans obtained during the year before LT and selected the oldest available scan from that period to minimize the severity of illness and the incidence of severe ascites that is more likely to occur immediately before LT. We used noncontrast CT scans when available (243 of 454 patients [54%]) and early arterial phase scans when noncontrast CT was unavailable (211 of 454 patients [46%]). Most CT scans (264 of 454 patients [58%]) were acquired by using a Siemens Somatom Definition Flash scanner. Other CT scanner vendors included GE Medical Systems, Toshiba, and Philips. The technical parameters for CT imaging were as follows: the tube voltage ranged from 90 to 140 (mode, 120 kVp), mean x-ray tube current–second product was 395 mAs, and the section thickness ranged from 0.75 to 10 mm (mode, 3 mm). The segmentation of skeletal muscle and adipose tissue was performed using OsiriX software (version 11.0.1, Pixmeo) to draw a closed polygon (delineating the interior border of the abdominal muscle wall; yellow outlines in Figures 2 and 3) and to use the region of interest tool to define thresholds for segmentation of muscle and fat in the OsiriX environment. Segmentation was performed by a trained radiology research fellow (O.S.) independently and reviewed by an abdominal radiologist with more than 13 years of experience (A.Z.). Both were blinded to recipients’ clinical data and outcomes. Skeletal muscle area, which includes the psoas major, erector spinae, quadratus lumborum, transversus abdominis, external and internal obliques, and rectus abdominis, was identified and quantified by thresholding using attenuation values of −30 to 150 HU. Visceral fat area was quantified using attenuation values of −150 to −50 HU. To identify SAT, we used attenuation values of −190 to −30 HU (6–8) (Figs 2, 3).
Figure 2:
Noncontrast axial abdominal CT scan (third lumbar vertebra) obtained 288 days before liver transplant in a 65-year-old man with nonalcoholic steatohepatitis cirrhosis with hepatocellular carcinoma (A) before and (B) after segmentation with OsiriX software (version 11.0.1, Pixmeo). Body mass index (calculated as weight in kilograms divided by height in meters squared) was 30, and skeletal muscle index (calculated as skeletal muscle area in centimeters squared divided by height in meters squared) was 66, with mean muscle attenuation of 44 HU, which was consistent with no sarcopenia or myosteatosis, respectively. Yellow: subcutaneous adipose tissue; red: skeletal muscle; green: visceral fat. The yellow outline represents the closed polygon drawn using the region of interest tool around the interior border of the abdominal muscle wall.
Figure 3:
Axial contrast-enhanced abdominal CT scan (third lumbar vertebra) obtained 276 days before liver transplant in a 57-year-old woman with hepatitis C cirrhosis (A) before and (B) after segmentation with OsiriX software (version 11.0.1, Pixmeo). Body mass index (calculated as weight in kilograms divided by height in meters squared) was 22, and skeletal muscle index (calculated as skeletal muscle area in centimeters squared divided by height in meters squared) was 39, with mean muscle attenuation of 32 HU, which was consistent with sarcopenia and myosteatosis, respectively. Yellow: subcutaneous adipose tissue; red: skeletal muscle; green: visceral fat. The yellow outline represents the closed polygon drawn using the region of interest tool around the interior border of the abdominal muscle wall.
Body Composition Measurements
To determine sarcopenia, we calculated skeletal muscle area in centimeters squared. Because recipients with larger body size typically have greater skeletal mass, we normalized skeletal muscle area by dividing it by the square of recipient’s height (in meters) to calculate the skeletal muscle index (SMI) in centimeters squared divided by meters squared. Sarcopenia was defined as SMI under 50 in men and SMI under 39 in women (8,15). Myosteatosis (low skeletal muscle CT attenuation) was determined by calculating the mean muscle attenuation in Hounsfield units for the entire muscle area. We defined myosteatosis as under 41 mean HU in recipients with a BMI (calculated as weight in kilograms divided by height in meters squared) less than 25 and under 33 mean HU in those with a BMI of 25 or higher (8,10). Finally, we defined sarcopenic obesity as concurrent obesity (BMI ≥30) and sarcopenia.
Statistical Analysis
We compared recipient characteristics between those who had sarcopenia and myosteatosis and those who did not by using Pearson χ2 tests for categorical variables, analysis of variance for normally distributed continuous variables, and Kruskal-Wallis tests for nonnormally distributed continuous variables.
Survival time was measured from the time of LT until mortality or the date of administrative censoring (April 1, 2020). We estimated cumulative incidence of mortality with use of the Kaplan-Meier method. Log-rank tests were used to separately compare unadjusted survival curves by sarcopenia, myosteatosis, and sarcopenic obesity. Proportional hazard assumptions were confirmed by visual inspection of log-log plots. We then estimated the hazard ratios (HRs) for both outcomes by BMI, sarcopenia, myosteatosis, sarcopenic obesity, VAT, SAT, and VAT/SAT ratio with use of the same Cox proportional hazards models. Sarcopenia and myosteatosis were treated as both continuous (SMI and mean skeletal muscle CT attenuation) and binary (yes or no) variables. BMI, SMI, and skeletal muscle CT attenuation were divided by five units to make the results more interpretable. To ensure that the thresholds of sarcopenia and myosteatosis identified in other populations were relevant to our population, we also estimated the nonlinear hazards of mortality according to SMI and mean skeletal muscle CT attenuation, respectively, using cubic spline terms and Cox proportional hazards models. Respective thresholds defining the nodes of the cubic spline were generated by default at 10%, 50%, and 90% of the corresponding continuous variables. The covariates used in the adjusted models, which were selected from the least absolute shrinkage and selection operator with lambda chosen by 10-fold cross-validation, were recipient race, age at LT, life support use (whether patients were on life support at the time of LT), MELD score at LT, donor age, and binary cold ischemia time (≤6 hours or >6 hours). Additionally, BMI was adjusted for models of mean skeletal muscle CT attenuation and myosteatosis. We calculated the Harrell C-statistic to compare the model discrimination of adjusted models of each CT measurement.
Analyses were performed using Stata/SE software (version 16, StataCorp). Two-sided P < .05 was considered to indicate a statistically significant difference.
Missing Data
In the Scientific Registry of Transplant Recipients, missing data only existed when centers failed to report certain factors. Of 454 LT recipient records, seven (1.5%) were missing pre-LT diabetes status, 24 (5.3%) were missing the previous abdominal surgery indicator, and one was missing portal vein thrombosis status. No covariates in survival analyses were missing.
Results
Demographic and Clinical Characteristics of the Study Sample
Among the 893 patients who underwent LT between January 2009 and December 2018, we excluded recipients who lacked a CT scan acquired within 1 year before LT (n = 207), underwent a living donor transplant (n = 126), were younger than 18 years (n = 47), had undergone a previous LT (n = 22), had severe ascites or anasarca (n = 14), had a status of 1A (acute liver failure) at the time of LT (n = 12), or whose abdominal CT scan did not include the third lumbar vertebra (n = 11) (Fig 1). Of 893 patients, 454 LT recipients were included in our analysis. Of these, 294 (65%) were men, 88 (19%) were Black, 341 (75%) were White, and 25 (6%) were Asian, Hispanic, and/or multiracial. The median age of recipients was 57 years (IQR, 50–62 years), and the mean BMI was 29 ± 5.7 (SD). The median time between CT scan and LT was 153 days (IQR, 45–246 days). Body composition measurements are summarized in Table 1.
Table 1:
Radiographic Measurements at LT Admission among LT Recipients in the Scientific Registry of Transplant Recipients according to Sex
Overall, 136 of 454 recipients (30%) were defined as having sarcopenia and 276 (61%) as having myosteatosis. The proportion of women was smaller among recipients with sarcopenia (31 of 136 [23%]) than among those without sarcopenia (129 of 318 [41%]) (P < .001). Mean BMI was lower among recipients with sarcopenia (26 ± 5.4 [SD]) than those without sarcopenia (30 ± 5.6) (P < .001). We found no evidence of differences in age (P = .10), primary diagnosis (P = .16), diabetes status (P = .67), race or ethnicity (P = .09), or MELD score (patients without sarcopenia, 21.4 ± 10.5 [SD] vs those with sarcopenia, 21.4 ± 11.0; P = .98) according to sarcopenia status.
The proportion of women was larger among recipients with myosteatosis (113 of 276 [41%]) than those without myosteatosis (47 of 178 [26%]) (P = .002). Recipients with myosteatosis had a higher mean MELD score (24.3 ± 10.6) than those without myosteatosis (16.9 ± 9.2) (P < .001). Recipients with myosteatosis were older (P = .047) and less likely to be Black (P = .04). We found no evidence of differences in BMI (patients without myosteatosis, 28 ± 4.9 vs those with myosteatosis, 29 ± 6.2; P = .39) and diabetes status (36 of 174 patients [21%] vs 55 of 273 patients [20%]; P = .89) according to myosteatosis status (Tables 2, 3).
Table 2:
Baseline Recipient Characteristics Overall and by Abdominal CT-based Body Composition at LT Admission among LT Recipients in the Scientific Registry of Transplant Recipients, 2009–2018
Table 3:
Baseline Donor Characteristics Overall and by Recipient Abdominal CT–based Body Composition at LT Admission among LT Donors in the Scientific Registry of Transplant Recipients, 2009–2018
Mortality
There were 109 deaths after LT per 1967 person-years. The median follow-up time was 3.7 years (range, 0.003–11.1 years). Among 454 LT recipients, 345 (76%) were right-censored. Among sarcopenic recipients, the median survival time was 9.7 years. In contrast, the median survival time for nonsarcopenic recipients was over 11 years (beyond the scope of our data). For sarcopenia (cumulative incidence at 10 years after LT: patients without sarcopenia, 0.35 vs those with sarcopenia, 0.56; P = .004), myosteatosis (cumulative incidence at 10 years after LT: patients without myosteatosis, 0.35 vs those with myosteatosis, 0.42; P = .03), and sarcopenic obesity (cumulative incidence at 10 years after LT: nonobese, 0.39 vs obese, 0.46; P = .03), recipients with each condition had a higher cumulative incidence of mortality compared with those without the condition (Fig 4). We found no evidence that SMI was associated with unadjusted nonlinear hazard of mortality (Fig 5A). Higher skeletal muscle CT attenuation was negatively associated with unadjusted nonlinear hazard of mortality (Fig 5B).
Figure 4:
Survival graphs show the unadjusted cumulative incidence of mortality according to the CT-based body composition measures of (A) sarcopenia, (B) myosteatosis, and (C) sarcopenic obesity at liver transplant (LT) admission among 454 LT recipients in the Scientific Registry of Transplant Recipients. In each patient, a single cross‐sectional abdominal CT image at the level of the third lumbar vertebra was analyzed. The segmentation of skeletal muscle and adipose tissue was performed to quantify total skeletal muscle area and skeletal muscle CT attenuation. Sarcopenia was defined as skeletal muscle index (calculated as skeletal muscle area in centimeters squared divided by height in meters squared) under 50 for men and under 39 for women. Myosteatosis was defined as skeletal muscle CT attenuation under 41 mean HU for recipients with a body mass index (calculated as weight in kilograms divided by height in meters squared) less than 25 and under 33 mean HU for recipients with body mass index of 25 or higher. Sarcopenic obesity was defined as body mass index of 30 or higher and concomitant sarcopenia.
Figure 5:
Histograms and line graphs show unadjusted, nonlinear associations between abdominal CT-based (A) skeletal muscle index (SMI) and (B) skeletal muscle CT attenuation at liver transplant (LT) admission and post-LT mortality among 454 LT recipients in the Scientific Registry of Transplant Recipients. In each patient, a single cross‐sectional abdominal CT image at the level of the third lumbar vertebra was analyzed. Segmentation of skeletal muscle and adipose tissue was performed to quantify total skeletal muscle area and skeletal muscle CT attenuation. SMI was calculated by dividing skeletal muscle area in centimeters squared by the square of height in meters. Skeletal muscle CT attenuation was determined using Hounsfield units. Bars represent frequency of Hounsfield units, the line represents the hazard ratio (HR), and the shaded area represents the 95% CI for the HR.
When we treated pre-LT SMI and skeletal muscle CT attenuation as continuous variables, smaller magnitudes of each were associated with higher risk of mortality after adjustment. For pre-LT sarcopenia (SMI), the HR was 1.1 per 5-cm2/m2 reduction (95% CI: 1.0, 1.2; C-statistic: 0.64; P = .01), and for pre-LT myosteatosis (skeletal muscle CT attenuation), the HR was 1.2 per 5-HU reduction (95% CI: 1.0, 1.3; C-statistic: 0.64; P = .03). However, when pre-LT SMI and skeletal muscle CT attenuation were treated as binary variables, only sarcopenia was associated with higher risk of mortality after adjustment (HR, 1.6 [95% CI: 1.1, 2.4]; C-statistic: 0. 64; P = .02). We found no evidence of an association for binary myosteatosis with risk of mortality (HR, 1.3 [95% CI: 0.86, 2.1]; C-statistic: 0.64; P = .21). There was no evidence of an association for BMI (P = .15), sarcopenic obesity (P = .15), VAT (P = .31), SAT (P = .77), or VAT/SAT ratio (P = .27) with risk of mortality after adjustment for potential confounders (Table 4). Harrell C-statistics were similar across all models.
Table 4:
Unadjusted and Adjusted HRs of Post-LT Mortality according to Abdominal CT–based Body Composition at LT Admission among 454 LT Recipients in the Scientific Registry of Transplant Recipients
Discussion
Body composition varies across liver transplant (LT) recipients, and certain phenotypes are associated with worse post-LT outcomes. Although many body composition measures can be acquired from pretransplant abdominal CT scans, it is unclear which measures best predict post-LT risk. We showed that sarcopenia was associated with a higher adjusted hazard of post-LT mortality (adjusted hazard ratio [HR], 1.6 [95% CI: 1.1, 2.4]; P = .02). Our data demonstrated that a reduction in skeletal muscle index was associated with a higher hazard of mortality (adjusted HR, 1.1; C-statistic: 0.64; P = .01). We found no evidence that body mass index (P = .15), sarcopenic obesity (P = .15), visceral adipose tissue (VAT) (P = .31), subcutaneous adipose tissue (SAT) (P = .77), and VAT/SAT ratio (P = .27) were associated with mortality. We did not find any evidence that myosteatosis was associated with mortality after adjustment for confounders (HR, 1.3 [95% CI: 0.86, 2.1]; C-statistic: 0.64; P = .21).
We analyzed various CT-based body composition measures and their associations with post-LT outcomes in one of the largest North American cohorts to date. One of the strengths of our study was the large patient sample. We obtained all available body composition measures (sarcopenia, sarcopenic obesity, myosteatosis, VAT, SAT, and VAT/SAT ratio) and determined associations between the measures and post-LT outcomes.
We believe that measuring these body composition parameters can help identify which patients will benefit from the transplant, and we can use this to address low muscle density and quality by means of preoperative rehabilitation using nutritional support and exercise (4).
Our finding that a reduction in SMI is associated with higher hazard of mortality is consistent with previous research. Kalafateli et al (16) found that sarcopenia was associated with early post-LT mortality and morbidity, which is consistent with the findings of other studies (17,18). Multiple studies also showed that sarcopenia was associated with higher mortality rates among patients on the LT waitlist (19–21). Meza-Junco et al (22) found that approximately one-third of patients (35 of 116) had sarcopenia, and that sarcopenia was an independent and strong risk factor for mortality in patients with concurrent cirrhosis and hepatocellular carcinoma. The association between sarcopenia and mortality in patients with cirrhosis has been established in multiple studies (23–25). However, the prognostic value of sarcopenia is unclear. For instance, Montano-Loza et al (26) found that sarcopenia was common in patients with cirrhosis and was predictive of longer hospitalization and higher risk of bacterial infection after LT compared with LT recipients without sarcopenia; however, sarcopenia was not associated with mortality. Bhanji et al (10) found that sarcopenia was associated with duration of hospital stay but not mortality. Valero et al (27) found that although sarcopenia was an independent predictor of postoperative complications, it was not associated with long-term survival.
In our study, we found no evidence that BMI, sarcopenic obesity, VAT, SAT, or VAT/SAT ratio were associated with mortality. Higher skeletal muscle CT attenuation was associated with a lower hazard of mortality. However, we did not find any evidence that myosteatosis was associated with mortality after adjustment for confounders. This finding was consistent with that of Bhanji et al (10), who found no evidence of an association between myosteatosis and mortality. Shenvi et al (28) evaluated the prognostic value of MRI-assessed myosteatosis in 180 LT recipients and found no evidence of an association between myosteatosis and mortality, acute transplant rejection, or longer hospital stay. However, several studies have reported that myosteatosis was associated with mortality and graft loss after LT (8,9,29). The differences in findings are likely related to using cutoff values based on an oncologic population, as well as the presence of peripheral edema and anasarca in these severely ill patients.
The use of CT-based methods to measure muscle and fat can give an objective assessment and direct visualization of body composition. In patients with end-stage liver disease and cirrhosis, BMI is often inflated and a poor reflection of body composition because of peripheral edema, ascites, and fluid overload in the abdomen (30–32). Two common approaches are used to measure muscle mass area and quality at the third lumbar vertebra: measuring psoas muscle area and measuring total skeletal muscle area. Measuring total skeletal muscle area is the most common and recommended method because the psoas is a minor muscle and measuring psoas muscle area does not represent overall sarcopenia (9,33). However, there is disagreement about the value of this method (6).
Our study had several limitations. First, only 19% of LT recipients were Black and 35% were women, which may limit generalizability and the ability to detect subtle interactions between sarcopenia and race and sex. Approximately 75% of our patient sample was White, and the prevalence of sarcopenia was higher among White patients. Second, artifacts limited our ability to define the borders of muscle and fat, potentially reducing the accuracy of our CT-based body composition segmentation. Most patients who underwent LT had severe anasarca and cirrhosis, which reduced the visibility of segment muscle, VAT, and SAT. In patients with severe obesity, some parts of the axial view of the third lumbar vertebra were outside the field of view, which reduced visibility of subcutaneous adipose tissue. Third, we analyzed noncontrast CT scans for most patients and early arterial phase scans when noncontrast CT scans were unavailable. The intravenous contrast agent might have caused a beam-hardening artifact, as well as higher muscle attenuation, which could have led to a false-negative myosteatosis diagnosis. However, the effect of intravenous contrast material is small (34). In some cases, when fluid was inside the abdomen, we could not use the currently accepted threshold for visceral fat area (−150 to −50 HU) because a large proportion of the visceral area would be missing from the visual field. By changing the threshold to −150 to −10 HU, we were able to detect visceral adipose tissue more accurately. Fourth, our study was retrospective, which might have caused selection bias (ie, sicker patients and patients with underlying cancers [eg, hepatocellular carcinoma] underwent more CT examinations before LT). Fifth, our study sample included only patients who underwent transplant. The prevalence of sarcopenia and myosteatosis may be much higher among all waitlisted patients. Sixth, the current thresholds for sarcopenia and myosteatosis are based on healthy populations (35) or on patients with cancer (36). The current understanding of clinically meaningful cut points in the LT population needs to be further validated.
In addition, segmentation and image evaluation of fat and muscle tissue takes approximately 30 minutes with the current semiautomated CT number thresholding method, which is time-consuming in the analysis of a large data set. An algorithm model using machine learning approaches is necessary to shorten this timeframe and improve accuracy (34).
In conclusion, pre–liver transplant (LT) body composition measurements can be obtained from available abdominal CT scans and used as a prognostic tool to predict post-LT outcomes. Sarcopenia was the measure most strongly associated with poorer post-LT outcomes. CT-based measurement of sarcopenia can be used to augment individualized risk prediction for LT recipients. Pre-LT CT is a valuable tool for identifying recipients at elevated risk. Transplant surgeons can use the results of this investigation for predicting outcomes before LT as a guide for their decisions and management.
Acknowledgments
Acknowledgments
For their editorial assistance, we thank Rachel Box, MS; Denise Di Salvio, MS; and Sandy Crump, MPH, and the Editorial Services group of The Johns Hopkins Department of Orthopaedic Surgery.
O.S. and Y.L. contributed equally to this work.
M.M.D. and C.R.W. are co–senior authors.
R.M.S. is supported by the Intramural Research Program of the National Institutes of Health Clinical Center. This research was supported by grants F32DK113719 (K.R.J.), T32DK007713 (B.J.B.), and R01DK120518 (M.M.D.) from the National Institute of Diabetes and Digestive and Kidney Diseases; grant T32EB006351-12 (M.A.L. and F.Y) from the National Institutes of Health; and grant K24AI144954 (D.L.S.) from the National Institute of Allergy and Infectious Diseases.
Disclosures of conflicts of interest: O.S. No relevant relationships. Y.L. No relevant relationships. K.R.J. No relevant relationships. J.D.M. No relevant relationships. B.J.B. No relevant relationships. M.A.L. Grant from the National Institutes of Health. F.Y. No relevant relationships. A.K. No relevant relationships. E.A.K. No relevant relationships. A.Z. No relevant relationships. R.M.S. Royalties from patents and/or software licenses from iCAD, Philips, ScanMed, PingAn, and Translation Holdings; research support to laboratory from PingAn. D.L.S. No relevant relationships. M.M.D. Grant from the National Institutes of Health. C.R.W. Grants to institution from Siemens Healthcare, Boston Scientific Corporation, Guerbet, Medtronic, and Johns Hopkins Technology Ventures; consulting fees from Siemens Healthcare, Boston Scientific Corporation, and Medtronic; material support from Guerbet and Boston Scientific Corporation; board positions: chair of the Grants and Education Committee of the Society of Interventional Radiology and the Society of Interventional Radiology Foundation; executive council and steering council member for Society of Interventional Radiology and Society of Interventional Radiology Foundation; deputy editor of Radiology.
Abbreviations:
- BMI
- body mass index
- HR
- hazard ratio
- LT
- liver transplant
- MELD
- Model for End-Stage Liver Disease
- SAT
- subcutaneous adipose tissue
- SMI
- skeletal muscle index
- VAT
- visceral adipose tissue
References
- 1. Lee DH , Keum N , Hu FB , et al . Predicted lean body mass, fat mass, and all cause and cause specific mortality in men: prospective US cohort study . BMJ 2018. ; 362 : k2575 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Kanwal F , Dulai GS , Spiegel BMR , Yee HF , Gralnek IM . A comparison of liver transplantation outcomes in the pre- vs. post-MELD eras . Aliment Pharmacol Ther 2005. ; 21 ( 2 ): 169 – 177 . [DOI] [PubMed] [Google Scholar]
- 3. Brown JC , Cespedes Feliciano EM , Caan BJ . The evolution of body composition in oncology-epidemiology, clinical trials, and the future of patient care: facts and numbers . J Cachexia Sarcopenia Muscle 2018. ; 9 ( 7 ): 1200 – 1208 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Lai JC , Sonnenday CJ , Tapper EB , et al . Frailty in liver transplantation: an expert opinion statement from the American Society of Transplantation Liver and Intestinal Community of Practice . Am J Transplant 2019. ; 19 ( 7 ): 1896 – 1906 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lee K , Shin Y , Huh J , et al . Recent issues on body composition imaging for sarcopenia evaluation . Korean J Radiol 2019. ; 20 ( 2 ): 205 – 217 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Cruz-Jentoft AJ , Bahat G , Bauer J , et al . Sarcopenia: revised European consensus on definition and diagnosis . Age Ageing 2019. ; 48 ( 1 ): 16 – 31 . [Published correction appears in Age Ageing 2019;48(4):601.] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Bhanji RA , Moctezuma-Velazquez C , Duarte-Rojo A , et al . Myosteatosis and sarcopenia are associated with hepatic encephalopathy in patients with cirrhosis . Hepatol Int 2018. ; 12 ( 4 ): 377 – 386 . [DOI] [PubMed] [Google Scholar]
- 8. Czigany Z , Kramp W , Bednarsch J , et al . Myosteatosis to predict inferior perioperative outcome in patients undergoing orthotopic liver transplantation . Am J Transplant 2020. ; 20 ( 2 ): 493 – 503 . [DOI] [PubMed] [Google Scholar]
- 9. Amini B , Boyle SP , Boutin RD , Lenchik L . Approaches to assessment of muscle mass and myosteatosis on computed tomography: a systematic review . J Gerontol A Biol Sci Med Sci 2019. ; 74 ( 10 ): 1671 – 1678 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Bhanji RA , Takahashi N , Moynagh MR , et al . The evolution and impact of sarcopenia pre- and post-liver transplantation . Aliment Pharmacol Ther 2019. ; 49 ( 6 ): 807 – 813 . [DOI] [PubMed] [Google Scholar]
- 11. Massie AB , Kucirka LM , Segev DL . Big data in organ transplantation: registries and administrative claims . Am J Transplant 2014. ; 14 ( 8 ): 1723 – 1730 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Yip C , Dinkel C , Mahajan A , Siddique M , Cook GJR , Goh V . Imaging body composition in cancer patients: visceral obesity, sarcopenia and sarcopenic obesity may impact on clinical outcome . Insights Imaging 2015. ; 6 ( 4 ): 489 – 497 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Noumura Y , Kamishima T , Sutherland K , Nishimura H . Visceral adipose tissue area measurement at a single level: can it represent visceral adipose tissue volume? Br J Radiol 2017. ; 90 ( 1077 ): 20170253 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Schweitzer L , Geisler C , Pourhassan M , et al . What is the best reference site for a single MRI slice to assess whole-body skeletal muscle and adipose tissue volumes in healthy adults? Am J Clin Nutr 2015. ; 102 ( 1 ): 58 – 65 . [DOI] [PubMed] [Google Scholar]
- 15. Carey EJ , Lai JC , Wang CW , et al . A multicenter study to define sarcopenia in patients with end-stage liver disease . Liver Transpl 2017. ; 23 ( 5 ): 625 – 633 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kalafateli M , Mantzoukis K , Choi Yau Y , et al . Malnutrition and sarcopenia predict post-liver transplantation outcomes independently of the Model for End-stage Liver Disease score . J Cachexia Sarcopenia Muscle 2017. ; 8 ( 1 ): 113 – 121 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Esser H , Resch T , Pamminger M , et al . Preoperative assessment of muscle mass using computerized tomography scans to predict outcomes following orthotopic liver transplantation . Transplantation 2019. ; 103 ( 12 ): 2506 – 2514 . [DOI] [PubMed] [Google Scholar]
- 18. Kuo SZ , Ahmad M , Dunn MA , et al . Sarcopenia predicts post-transplant mortality in acutely ill men undergoing urgent evaluation and liver transplantation . Transplantation 2019. ; 103 ( 11 ): 2312 – 2317 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Tandon P , Ney M , Irwin I , et al . Severe muscle depletion in patients on the liver transplant wait list: its prevalence and independent prognostic value . Liver Transpl 2012. ; 18 ( 10 ): 1209 – 1216 . [DOI] [PubMed] [Google Scholar]
- 20. van Vugt JLA , Levolger S , de Bruin RWF , van Rosmalen J , Metselaar HJ , IJzermans JNM . Systematic review and meta-analysis of the impact of computed tomography-assessed skeletal muscle mass on outcome in patients awaiting or undergoing liver transplantation . Am J Transplant 2016. ; 16 ( 8 ): 2277 – 2292 . [DOI] [PubMed] [Google Scholar]
- 21. D’Arcangelo F , Zanetto A , Aliberti C , et al . The impact of sarcopenia on the outcome of patients with cirrhosis with and without hepatocellular carcinoma who undergo liver transplantation . Hepatoma Res 2021. ; 7 : 4 . [Google Scholar]
- 22. Meza-Junco J , Montano-Loza AJ , Baracos VE , et al . Sarcopenia as a prognostic index of nutritional status in concurrent cirrhosis and hepatocellular carcinoma . J Clin Gastroenterol 2013. ; 47 ( 10 ): 861 – 870 . [DOI] [PubMed] [Google Scholar]
- 23. Montano-Loza AJ , Meza-Junco J , Prado CMM , et al . Muscle wasting is associated with mortality in patients with cirrhosis . Clin Gastroenterol Hepatol 2012. ; 10 ( 2 ): 166 – 173, 173.e1 . [DOI] [PubMed] [Google Scholar]
- 24. Paternostro R , Lampichler K , Bardach C , et al . The value of different CT-based methods for diagnosing low muscle mass and predicting mortality in patients with cirrhosis . Liver Int 2019. ; 39 ( 12 ): 2374 – 2385 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Durand F , Buyse S , Francoz C , et al . Prognostic value of muscle atrophy in cirrhosis using psoas muscle thickness on computed tomography . J Hepatol 2014. ; 60 ( 6 ): 1151 – 1157 . [DOI] [PubMed] [Google Scholar]
- 26. Montano-Loza AJ , Meza-Junco J , Baracos VE , et al . Severe muscle depletion predicts postoperative length of stay but is not associated with survival after liver transplantation . Liver Transpl 2014. ; 20 ( 6 ): 640 – 648 . [DOI] [PubMed] [Google Scholar]
- 27. Valero V 3rd , Amini N , Spolverato G , et al . Sarcopenia adversely impacts postoperative complications following resection or transplantation in patients with primary liver tumors . J Gastrointest Surg 2015. ; 19 ( 2 ): 272 – 281 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Shenvi SD , Taber DJ , Hardie AD , Botstein JO , McGillicuddy JW . Assessment of magnetic resonance imaging derived fat fraction as a sensitive and reliable predictor of myosteatosis in liver transplant recipients . HPB (Oxford) 2020. ; 22 ( 1 ): 102 – 108 . [DOI] [PubMed] [Google Scholar]
- 29. Montano-Loza AJ , Angulo P , Meza-Junco J , et al . Sarcopenic obesity and myosteatosis are associated with higher mortality in patients with cirrhosis . J Cachexia Sarcopenia Muscle 2016. ; 7 ( 2 ): 126 – 135 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. DiMartini A , Cruz RJ Jr , Dew MA , et al . Muscle mass predicts outcomes following liver transplantation . Liver Transpl 2013. ; 19 ( 11 ): 1172 – 1180 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Cruz RJ Jr , Dew MA , Myaskovsky L , et al . Objective radiologic assessment of body composition in patients with end-stage liver disease: going beyond the BMI . Transplantation 2013. ; 95 ( 4 ): 617 – 622 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. McHugh PP , Shah SH , Johnston TD , Gedaly R , Ranjan D . Predicting dry weight in patients with ascites and liver cirrhosis using computed tomography imaging . Hepatogastroenterology 2010. ; 57 ( 99-100 ): 591 – 597 . [PubMed] [Google Scholar]
- 33. Rutten IJG , Ubachs J , Kruitwagen RFPM , et al . The influence of sarcopenia on survival and surgical complications in ovarian cancer patients undergoing primary debulking surgery . Eur J Surg Oncol 2017. ; 43 ( 4 ): 717 – 724 . [DOI] [PubMed] [Google Scholar]
- 34. Perez AA , Pickhardt PJ , Elton DC , Sandfort V , Summers RM . Fully automated CT imaging biomarkers of bone, muscle, and fat: correcting for the effect of intravenous contrast . Abdom Radiol (NY) 2021. ; 46 ( 3 ): 1229 – 1235 . [DOI] [PubMed] [Google Scholar]
- 35. van der Werf A , Langius JAE , de van der Schueren MAE , et al . Percentiles for skeletal muscle index, area and radiation attenuation based on computed tomography imaging in a healthy Caucasian population . Eur J Clin Nutr 2018. ; 72 ( 2 ): 288 – 296 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Mourtzakis M , Prado CMM , Lieffers JR , Reiman T , McCargar LJ , Baracos VE . A practical and precise approach to quantification of body composition in cancer patients using computed tomography images acquired during routine care . Appl Physiol Nutr Metab 2008. ; 33 ( 5 ): 997 – 1006 . [DOI] [PubMed] [Google Scholar]