Radiol Oncol 2020; 54(2): 237-246. doi: 10.2478/raon-2020-0016 237 research article Sarcopenia and myosteatosis at presentation adversely affect survival after esophagectomy for esophageal cancer Matevz Srpcic1,2, Taja Jordan3, Karteek Popuri4, Mihael Sok1,2 1 Department of thoracic surgery, Surgical clinic, University Medical Centre Ljubljana, Slovenia 2 Faculty of Medicine, University of Ljubljana, Slovenia 3 Institute of radiology, University Medical Centre Ljubljana, Slovenia 4 Simon Fraser University, Burnaby, Canada Radiol Oncol 2020; 54(2): 237-246. Received 7 January 2020 Accepted 3 March 2020 Correspondence to: Matevž Srpčič, M.D., Department of Thoracic Surgery, Surgical Clinic, University Medical Centre Ljubljana, Zaloška 7, SI-1000 Ljubljana, Slovenia. Phone: +386 1 522 3813; fax +386 1 522 2485; E-mail: matevz.srpcic@kclj.si Disclosure: MSr, TJ and MSo declare that they have no competing interests. KP is a co-founder of and actively directs Voronoi Health Analytics Incorporated, a Canadian corporation that sells commercial licenses for the ABACS (Automated Body Composition Analyzer using Computed tomography image Segmentation) software. Background. Esophageal cancer remains a disease with poor survival and many complications. Measuring muscle mass and quality can identify patients with diminished muscle mass (sarcopenia) and muscle fat infiltration (myostea- tosis). We studied the impact of sarcopenia and myosteatosis in resectable esophageal cancer on overall survival and complications. Patients and methods. 139 patients received a radical esophagectomy. Skeletal muscle area (SMA) and muscle attenuation (MA) in CT images at L3 level were recorded and groups with and without sarcopenia and myosteatosis were compared for overall survival (OS), perioperative mortality, conduit complications, pleuropulmonary complica- tions, respiratory failure requiring mechanical ventilation and other significant complications. Results. Prevalence of sarcopenia and myosteatosis at presentation was 16.5% and 51.8%, respectively. Both were associated with decreased OS. Median survival was 18.3 months (CI 5.4–31.1) vs. 31.0 months (CI 7.4–54.6) for sarco- penia/no sarcopenia (log rank p = 0.042) and 19.0 months (CI 13.3–24.7) vs. 57.1 months (CI 15.2–99.0) for myosteatosis (log rank p = 0.044), respectively. A relationship between sarcopenia and myosteatosis and other negative outcomes after esophagectomy could not be established. Conclusions. Sarcopenia and myosteatosis before esophagectomy are associated with decreased overall survival but not with more frequent perioperative complications. Identification of patients at risk can guide therapeutic deci- sions and interventions aimed at replenishing muscle reserves. Key words: sarcopenia; myosteatosis; esophagectomy; survival; esophageal cancer; muscle depletion Introduction Constant gradual improvements of operative tech- niques and perioperative care have reduced the dangers of esophagectomy, the cornerstone of rad- ical treatment of resectable esophageal cancer, but it remains a major procedure burdened with high morbidity and mortality.1 Overall 5-year survival in resectable esophageal cancer has improved in recent years by about 2–3 fold.2 This improvement was attributed to centralization of surgical treat- ment and introduction of neoadjuvant chemoradi- otherapy.3 Advances were also made in periopera- tive care and better understanding and prevention of the detrimental effects of muscle depletion so typical of esophageal malignancies.4 Further improvement in outcomes can be achieved by tailoring the treatment to patients’ ability to withstand the trauma of surgery and to return to a functional life after treatment. Adequate Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy238 fitness for treatment has traditionally been assessed from various performance scores, risk scores as well as more basic patients’ characteristics like age and body mass index.5 Body mass index (BMI) at pres- entation has proven to be an inaccurate predictor of outcomes since it does not correspond to body composition well.6,7 Better methods to assess the most important parameter of body composition, the skeletal muscle content, have been introduced. They include functional tests like muscle strength measurements and measurements of muscle mass with dual energy x-ray imaging (DEXA), bioim- pedance analysis or cross-sectional imaging (CT or MRI).8 Cross-sectional imaging (or planimetry) has the advantage of being readily available in cancer patients for staging purposes. This has en- couraged many studies to examine the relationship between overall muscle mass, its quality and their effect on outcomes. A reliable relationship between planimetrically determined muscle mass and qual- ity and its function, determined by other methods available, has been established. Muscle area at the level of 3rd lumbal vertebra, normalized for height (skeletal muscle index (SMI)) is highly correlated with total body skeletal muscle mass.9 Estimating survival chances for a patient pre- senting with resectable esophageal cancer is im- portant in planning appropriate treatment strate- gies and interventions aimed at improving sur- vival and quality of life. Pronounced weight loss is a hallmark of malignant disease, especially pro- nounced in digestive tract tumors, among them in esophageal and pancreatic cancers in particular.10 In their seminal work, the team from University of Alberta have shown that skeletal muscle depletion (sarcopenia and low muscle attenuation) is the real negative predictor of survival regardless of overall body weight in cancer patients.6 Sarcopenia is defined by the European Working Group on Sarcopenia in Older people as the pres- ence of low muscle mass (under the 5th percen- tile) and low muscle function (strength or per- formance)11 typically presenting in advanced age but also in cancer and other diseases. It is a well established predictor of poor survival and treat- ment outcomes in cancer patients.6 Myosteatosis is defined as abnormal fat infiltration in skeletal mus- cle. It is negatively associated with muscle strength and quality and is brought on by aging12, diabetes13, obesity14 and malignant disease.6,15,16 Radiodensity of human muscle on CT scan (or muscle attenua- tion, MA) correlates well with its triglyceride con- tent.14 Measuring the attenuation values of muscle tissue corresponds well to the extent of myosteato- sis, which is a sign of muscle wasting and again a predictor of poor outcome.17 By assessing muscle mass and quality before treatment an individualized risk assessment for overall survival and complications during treat- ment can be improved, patients at risk identified and appropriate interventions (mainly directed towards maintaining and gaining muscle mass) undertaken.18 Our aim was to study the impact of muscle depletion (sarcopenia and myosteatosis) on outcomes (overall survival [OS], perioperative mortality and rate of complications) in resectable esophageal cancer. Patients and methods Study population All patients who received an esophagectomy with curative intent for esophageal or esophago-gastric junction cancer at Clinical Department of Thoracic Surgery at University Medical Centre Ljubljana were eligible for inclusion in the study. Patients received either upfront surgery or neoadjuvant chemoradiotherapy followed by esophagectomy according to national guidelines. All patients re- ceived individualized nutritional support and counselling according to ESPEN best practice guidelines19 and in all patients a catheter feeding jejunostomy was placed during esophagectomy. Clinical parameters were recorded prospectively in a database since 2003. Out of the 162 patients operated on consecutively between 2008 and 2018 CT images suitable for analysis of muscle mass and quality were available for 139 patients which were included in the study. Requirements for adequate images were the inclusion of L3 level and availabil- ity of non-contrast images for attenuation analysis. Only images recorded at presentation before the initiation of any treatment were considered. Our study design was approved and the need for obtaining informed consent from participants waived by the Slovenian National medical ethics committee (approval number 0120–301/2016–2). Definitions We grouped complications into following groups. Conduit complications included clinically silent fis- tulae seen on esophagograms and/or CT scans, clinically important leaks that required interven- tions and frank gastric necroses. Respiratory compli- cations included respiratory failure requiring me- chanical ventilation and pneumonia, defined as the Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy 239 presence of new infiltrates on chest radiography and a positive culture result from bronchoalveolar lavage or sputum requiring antibiotics. Respiratory failure requiring mechanical ventilation was re- corded separately as well. Other complications were defined as other seri- ous complications (Dindo Clavien 2 or greater)20 requiring intervention (i.e. early reoperation, cardi- oversion, endoscopic intervention) or directoscopi- cally proven laryngeal nerve paralysis.21 OS was defined as the time interval between esophagectomy and death of any cause. Patients alive on 1.10.2018 as reported by Cancer registry of Slovenia were censored at that date. BMI was calculated as patient weight [kg]/ height [m]2, recorded at admission one day before surgery. CT body composition analysis (planimetry) Pre-operative abdominal CT or whole body PET- CT scans were obtained. In each patient a single slice at the level of the 3rd lumbar vertebra (L3) was selected for automatic segmentation. CT scans were analyzed using the “Automated Body Composition Analyzer using Computed tomography image Segmentation” (ABACS) software22,23, which uses a priori information about the skeletal muscle shape in the L3 region and predefined Hounsfield units (HU) values to recognize different tissues. HU values used to assess the total cross-sectional area for muscular tissue (SMA – skeletal muscle area) were −29 to +150 HU. Muscle attenuation (MA) was assessed by averaging HU of skeletal muscle. Additionally, SMI was calculated using the follow- ing formula: (SMA [cm2])/(patient height [m2]). All abdominal CT and PET-CT scans were analyzed by one blinded independent radiologist. The following planimetry data were reported: number of days between CT and esophagectomy, SMA (skeletal muscle area) reported in cm2, SMI (skeletal muscle index) is SMA corrected for height (i.e. divided by height squared) and expressed in cm2/m2. MA (muscle attenuation) was reported in Hounsfield units. Previously defined muscle index cut-off values for sarcopenia in a healthy non-elderly Caucasian population were used to define limits for SMI in men at less than 43.1 cm2/m2and less than 32.7 cm2/ m2 in women. Cutoff values for myosteatosis from the same study were used with myosteatosis de- fined as MA of less than 30.9 HU in men and 24.8 HU in women.24 Outcomes and statistical analysis Standard descriptive statistics of demographic and clinical characteristics for patients with and without sarcopenia and myosteatosis were sum- marized. Differences in demographic and clinical characteristics between groups (sarcopenia/no sar- copenia and myosteatosis/no myosteatosis) were evaluated with Pearson’s Chi-square tests for cat- egorical and t-tests for parametric variables. Primary outcome studied was overall survival. It was reported in each group with the Kaplan- Meier curve and the survival of groups with/with- out sarcopenia and with/without myosteatosis was compared using the log rank Mantel Cox test. Secondary outcomes of interest were the inci- dences of complications in groups with/without sar- copenia and with/without myosteatosis. They were compared with Pearson’s Chi-square test. P value of < 0.05 was considered significant. All statistical anal- yses were performed using Statistical Package for the Social Sciences (SPSS, version 22.0, Armonk NY). Results Patient characterictics One hundred and thirty-nine patients underwent esophagectomy with primary reconstruction with curative intent. Overall demographic, clinical and complication characterictics are summarized in Table 1. Mean BMI was 26.3 ± 4.8 with only 7 (5.0%) having a BMI less than 18.5. As many as 46 (33.1%) patients reported having lost 10% or more of their normal body weight prior to esophagectomy. Average time between CT and esophagectomy was 76.9 ± 52.3 days with a much shorter time in those receiving primary resection compared to those with neoadjuvant treatment. Sarcopenia was present in 23 (16.5%) patients and myosteatosis in 72 (51.8%). Surgery and pathology Eighty-seven (62.6%) patients received an open esophagectomy and 52 (37.4%) had a hybrid or completely minimally invasive procedure. Type of procedure data, radicality rates, numbers of lymph nodes harvested and histology and staging data are given in Table 1. Complications and survival 9 patients died after esophagectomy during the ini- tial hospitalization (in hospital mortality of 6.5%). Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy240 D emographic and preoperative data Age at Surgery (mean ± SD) [years] 63.9 ± 9.5 min-max 30–83 Gender (N, % female) 22 (15.8%) BMI (mean ± SD) [kg/ m2] 26.3 ± 4.8 Weight loss > 10% (N, %) 46 (33.1%) Neoadjuvant therapy (N, %) 74 (53.2 %) Planimetry data Days between CT and esophagectomy all (mean ± SD) 76.9 ± 52.3 min-max 6–192 median 84 Neoadjuvant (mean ± SD) 115.2 ± 36.0 min-max 14–192 median 125 No neoadjuvant (mean ± SD) 33.5 ± 28.5 min-max 6–141 median 23 SMA [cm2] (mean ± SD) male 157.6 ± 28.0 female 103.9 ± 16.3 SMI [cm2/m2] (mean ± SD) male 52.1 ± 9.5 female 39.8 ± 6.8 Muscle attenuation [HU] (mean ± SD) male 31.2 ± 8.3 female 27.8 ± 8.7 sarcopenia (N, %) 23 (16.5%) myosteatosis (N, %) 72 (51.8%) Procedure data Surgical approach (N, %) open 87 (62.6%) MIE 52 (37.4%) Type of esophagectomy (N, %) Ivor-Lewis 109 (78.4%) McKeown 26 (18.7%) Transhiatal 4 (2.9%) Radicality (N, %) R0 130 (93.5%) R1 5 (3.6%) R2 4 (2.9%) Lymph nodes (mean ± SD) (N, %) 23.4 ± 12.3 min-max 0–76 median 21 Cancer type (N, %) Adenocarcinoma 74 (53.2%) Squamous cell carcinoma 64 (46.0%) GIST 1 (0.7%) Pathological Stage (AJCC 2017) (N, %) I 51 (36.7%) II 27 (19.4%) III 36 (25.9%) IVA 23 (16.5%) IVB 2 (1.4%) Complications (N, %) In hospital mortality 9 (6.5%) Any complication 65 (46.8%) Conduit complications 21 (15.1%) Pleuropulmonary complications 37 (26.6%) Respiratory failure 26 (18.7%) Any other complications 42 (30.2%) Median survival [months] 26.8 (95% CI 8.1–45,7) 1 year survival 73.7% 3 year survival 45.1% 5 year survival 40.3% TABLE 1. Demographic, preoperative, procedure and outcome data in all patients (N = 139) AJCC = American joint committee on cancer; BMI = body mass index; CI = confidence interval; CT = computed tomography; GIST = gastrointestinal stromal tumor; HU = Hounsfield units; MIE = minimally invasive esophagectomy; SD = standard deviation; SMA = skeletal muscle area; SMI = skeletal muscle index Almost half or 65 patients (46.8%) experienced a complication of Dindo-Clavien grade 2 sever- ity or greater20 after the procedure. Rates of other complications and survival rates are shown in Table 1. Survival is shown as a Kaplan-Meier curve in Figure 1. Median follow up was 18.1 months (range 0–115). 72 patients (51.8%) died during the observation period and 67 (48.1%) were censored. Sarcopenia and myosteatosis subgroups Demographic and clinical data was compared be- tween patients with and without sarcopenia and with and without myosteatosis (Table 2). Patients with myosteatosis were significantly older than pa- tients without it whereas in patients with or with- out sarcopenia age difference didn’t reach statisti- cal significance. BMI was significantly lower in sar- copenic patients but significantly higher in patients with myosteatosis. Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy 241 There was no statistically significant differ- ence in sex distribution, days between CT and es- ophagectomy, weight loss, neoadjuvant therapy, cancer type, pathological stage, lymph nodes har- vested or surgical approach between sarcopenia/ no sarcopenia and myosteatosis/no myosteatosis groups. Complications and survival were compared be- tween sarcopenia/no sarcopenia and myosteatosis/ no myosteatosis groups as shown in Table 3 and Figure 2 and 3. No statistically significant difference in in hos- pital mortality, any complications, pleuropulmo- nary complications, respiratory failure or any oth- er complications was found between sarcopenia/ no sarcopenia and myosteatosis/no myosteatosis groups. Conduit complications were however sig- nificantly less common in the myosteatosis group (5/72 (6.9%) vs. 16/67 (23.9%) in patients without myosteatosis (OR 0.238 (0.082–0.692), p = 0.005). Survival for sarcopenia/no sarcopenia and my- osteatosis/no myosteatosis is given in two Kaplan Meier plots in Figures 2 and 3. Survival curves were compared with the log rank Mantel Cox test and differences in survival between each pair were statistically significant (p = 0.042 for sarcopenia/no sarcopenia and p = 0.044 for myosteatosis/no my- osteatosis). Discussion Our prospective cohort study shows that dimin- ished muscle reserves, measured as sarcopenia (loss of muscle mass) and myosteatosis (infiltration of muscle with fat), are associated with decreased overall survival in patients receiving esophagecto- my as part of radical esophageal cancer treatment. A relationship between sarcopenia and myosteato- sis and other negative outcomes after esophagec- tomy (perioperative mortality and incidence of complications) could not be established. Effects of muscle mass loss have been studied in numerous other malignancies as well as non malig- nant diseases25-27 but studies reporting myosteato- sis as well as sarcopenia are still rare.28 Prevalence of sarcopenia in studies on correlation between muscle area and survival in esophageal cancer can range widely from 16%–80%.29-31 Choosing the right cutoff values for defining sarcopenia and my- osteatosis can be challenging. In keeping with the definition of sarcopenia as absolute muscle mass below the 5th percentile of the population32 we chose recently published cutoff values for a popu- lation closely resembling ours. Van der Werf et al. have published sex specific percentiles for SMI and MA for a healthy Caucasian population.24 They FIGURE 1. Cumulative survival Kaplan-Meier curve. FIGURE 2. Kaplan-Meier survival curves for sarcopenia. FIGURE 3. Kaplan-Meier survival curves for myosteatosis. Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy242 TABLE 2. Demographic, preoperative, pathological and procedure data compared between sarcopenia/no sarcopenia and myosteatosis/no myosteatosis groups Sarcopenia (N = 23 (16.5%)) No Sarcopenia (N = 116 (83.5%)) p Myosteatosis (N = 72 (51.8%)) No Myosteatosis (N = 67 (48.2%)) p Age at Surgery (mean ± SD) 67.1 ± 7.8 63.3 ± 9.7 0.076 67.1 ± 7.7 60.5 ± 10.0 < 0.001 Female sex (n (%)) 3 (13.0%) 19 (16.4%) 0.689 10 (13.9%) 12 (17.9%) 0.516 BMI (mean ± SD) 23.8 ± 5.9 26.7 ± 4.4 0.006 27.3 ± 4.9 25.2 ± 4.4 0.006 Days between CT and esophagectomy (mean ± SD) 81.4 ± 57.6 76.1 ± 51.4 0.654 78.8 ± 52.8 75.0 ± 52.1 0.666 Weight loss > 10% (n (%)) 11 (47.8%) 35 (30.2%) 0.100 25 (34.7%) 21 (31.3%) 0.672 Neoadjuvant Therapy (n (%)) 14 (60.9%) 60 (51.7%) 0.422 34 (47.2%) 40 (59.7%) 0.141 Cancer Type (n (%)) 0.864 0.500 Adenocarcinoma 13 (56.6%) 61 (52.6%) 37 (51.4%) 37 (55.2%) Squamous cell carcinoma 10 (43.4%) 54 (46.6%) 35 (48.6%) 29 (43.3%) GIST 1 (0.8%) 1 (1.5%) Pathological Stage (AJCC 2017) (n (%)) 0.650 0.546 I 8 (34.8%) 43 (37.1%) 26 (36.1%) 25 (37.3%) II 6 (26.1%) 21 (18.1%) 11 (15.3%) 16 (23.9%) III 4 (17.4%) 32 (27.6%) 21 (29.2%) 15 (22.4%) IVA 4 (17.4%) 19 (16.4%) 12 (16.7%) 11 (16.4%) IVB 1 (4.3%) 1 (0.8%) 2 (2.8%) 0 Lymph nodes (mean ± SD) 28.8 ± 10.5 23.9 ± 12.6 0.266 24.4 ± 11.1 22.4 ± 13.5 0.337 Surgical approach 0.258 0.167 open 12 (52.2%) 75 (64.7%) 49 (68.1%) 38 (56.7%) MIE 11 (47.8%) 41 (35.3%) 23 (31.9%) 29 (43.3%) AJCC = American joint committee on cancer; BMI = body mass index; CT = computed tomography; GIST-gastrointestinal stromal tumor; HU = Hounsfield units; MIE = minimally invasive esophagectomy; SD = standard deviation; SMA = skeletal muscle area; SMI = skeletal muscle index proposed using the 5th percentile for cutoff val- ues for SMI and MA in non-elderly (age 20–60) to avoid age related muscle loss. These values (SMI 43.1 cm2/m2 for men and 32.7 cm2/m2 for women) are markedly lower then ones used in most pre- vious studies. Consequently, the prevalence of sarcopenia in our study (16.5%) is also lower than 26–75% reported in other studies in resectable es- ophageal cancer. Mean SMA and SMI was 157.6 ± 28.0 cm2 and 52.1 ± 9.5 cm2/m2 in males and 103.9 ± 16.3 cm2 and 39.8 ± 6.8 cm2/m2 in females (both sig- nificantly different between sexes with p < 0.001) which correlates well with studies in similar popu- lations. We believe that choosing the right popula- tion with which patients are compared is crucial in determining the real prevalence of sarcopenia (e.g., the study by Nishigori et al. in Japanese esophageal cancer patients33 used the cutoff points obtained in Canadian obese patients34 and reported sarcopenia in 75% of patients). Defining myosteatosis is even more difficult, since the term is not used much yet and reports are scarcer. We chose cutoffs according to the same principle, i.e. at the 5th percentile of a healthy pop- ulation. We did not find a statistically significant difference in muscle attenuation between males and females (31.2 ± 8.3 HU vs. 27.8 ± 8.7 HU, p = 0.082), but with small numbers in our groups and the availability of sex-specific cutoff values for at- tenuation we opted for those. Myosteatosis was present in 51.8% of our patients and there was no significant relationship between sarcopenia and myosteatosis (OR 1.256 (CI 0.510–3.093, p = 0.620)). This is in contrast with the study by Stretch et al. where the proportions of patients with sarcopenia and myosteatosis were inverse (40.7% vs. 25.2%) but they similarly reported no correlation between muscle mass and muscle radiodensity. A possible reason for this are the higher cutoffs they used for sarcopenia (40th percentile of their patients or 47.7 cm2/m2 and 36.5 cm2/m2).28 On univariate analysis sarcopenia and myostea- tosis were associated with lower overall survival in our study group (Kaplan Meier log rank p = 0.042 Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy 243 and p = 0.044, respectively). For sarcopenia this is in accordance with previously published data and for myosteatosis this is one of the first published reports. Dijksterhuis et al. have published a report on body composition, survival and toxicity in ad- vanced esophagogastric cancer patients receiving palliative chemotherapy where they used BMI- specific cutoff values to define myosteatosis (< 41 HU in non obese (BMI < 25) and < 33 HU in over- weight patients). Prevalence of myosteatosis in their group was 50% and they found a lower risk of grade III and IV toxicity in patients with higher muscular density but no association between sar- copenia or myosteatosis and survival was found.35 Tamandl et al. published a study with 200 patients receiving an esophagectomy. They stratified pa- tients in low- and high-muscle attenuation groups with a cutoff of 40HU in a population similar to ours. Average MA was 36 HU (31–41) and patients with MA < 40 HU had significantly poorer over- all survival.36 The percentage of patients with MA over and under 40 HU is not given, so we cannot compare the prevalence to our results but this defi- nition of reduced muscle attenuation uses a cutoff considerably higher than ours. On the other hand, a study by Gabiatti et al. in patients with locally advanced esophageal cancer receiving definitive chemoradiotherapy demon- strated favorable progression free survival and overall survival in a subgroup of patients with my- osteatosis but without systemic inflammation.37 Sarcopenia has been studied extensively as a predictive factor in esophageal cancer. A recently published meta-analysis by Boshier et al. reviewed TABLE 3. Complication and survival data compared between sarcopenia/no sarcopenia and myosteatosis/no myosteatosis groups Sarcopenia (N = 23 (16.5%)) No Sarcopenia (N = 116 (83.5%)) Odds Ratio (OR. 95% CI) p Complications (n (%)) In hospital mortality 1 (4.3%) 8 (6.9%) 0.614 (0.073–5.158) 0.650 Any complication 11 (47.8%) 54 (46.6%) 1.052 (0.430–2.578) 0.911 Conduit complications 4 (17.4%) 17 (14.7%) 1.226 (0.371–4.049) 0.738 Pleuropulmonary complications 8 (34.8%) 29 (25.0%) 1.600 (0.615–4.160) 0.332 Respiratory failure 5 (21.7%) 21 (18.1%) 1.230 (0.410–3.689) 0.711 Any other complications 4 (17.4%) 38 (32.8%) 0.432 (0.137–1.359) 0.143 Median survival [months] 18.3 (CI 5.4–31.1) 31.0 (CI 7.4–54.6) 0.042 1 year survival 50.8% 78.5% 3 year survival 32.9% 47.7% 5 year survival 32.9% 42.2% myosteatosis (N = 72 (51.8%)) no myosteatosis (N = 67 (48.2%)) odds ratio (OR. 95% CI) p Complications (n (%)) In hospital mortality 7 (9.7%) 2 (3.0%) 3.500 (0.701–17.486) 0.107 Any complication 32 (44.4%) 33 (49.3%) 0.824 (0.423–1.607) 0.570 Conduit complications 5 (6.9%) 16 (23.9%) 0.238 (0.082–0.692) 0.005 Pleuropulmonary complications 17 (23.6%) 20 (30.0%) 0.726 (0.341–1.545) 0.406 Respiratory failure 14 (19.4%) 12 (17.9%) 1.066 (0.453–2.510) 0.884 Any other complications 24 (33.3%) 18 (26.9%) 1.361 (0.656–2.822) 0.407 Median survival [months] 19.0 (CI 13.3–24.7) 57.1 (CI 15.2–99.0) 0.044 1 year survival 64.2% 84.0% 3 year survival 36.9% 53.7% 5 year survival 33.9% 46.9% CI = confidence interval; OR = odds ratio Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy244 29 studies with 3193 patients (38% sarcopenic) in which various methods were used to diagnose sarcopenia.38 Sarcopenic patients had more pul- monary complications and lower overall survival. A similar meta-analysis by Deng et al. reviewed 11 cohort studies including 1520 patients (52.3% sarcopenic). Patients with sarcopenia had lower 3-year and 5-year survival after resection.39 Complications and perioperative mortality were compared in our study between sarcopenia/no sar- copenia and myosteatosis/no myosteatosis groups and no statistically significant negative effect of muscle depletion was found. This is in concord- ance with most other studies who failed to show a connection even in studies who showed differ- ences in long term survival.29,40 Insufficient statisti- cal power in most studies including ours to detect a potential difference in complication rates is no doubt a strong factor. For conduit complications however, the incidence in our cohort was signifi- cantly lower in the myosteatosis group (5/72 (6.9%) vs. 16/67 (23.9%) in patients without myosteatosis, (OR 0.238 (0.082–0.692), p = 0.005). It is difficult to explain the reason for this observation. A higher BMI in patients with myosteatosis could indicate a better nutritional status at presentation. Despite the lower incidence of this dangerous complication perioperative mortality in patients with myostea- tosis was not different than in patients without it. General clinical data in our cohort does not dif- fer significantly from similar published series in re- sectable esophageal cancer. Patients with myostea- tosis were significantly older than patients without it (67.1 ± 7.7 vs. 60.5 ± 10.0 (p < 0.001)) whereas in patients with or without sarcopenia age difference didn’t reach statistical significance (67.1 ± 7.8 vs. 63.3 ± 9.7 (p = 0.076)). BMI was significantly lower in sarcopenic patients (23.8 ± 5.9 vs. 26.7 ± 4.4 (p = 0.006)) but significantly higher in patients with myosteatosis (27.3 ± 4.9 vs. 25.2 ± 4.4 (p = 0.006)). 13 patients (9.4%) had both sarcopenia and myoste- atosis, their BMI was 25.5 ± 6.1 (range 18.1–37.1). 33.1% of our patients lost 10% or more of their body weight but this did not confer a greater risk of having sarcopenia (OR 2.12 (CI 0.855–5.266), p = 0.100) or myosteatosis (OR 1.165 (CI 0.574–2.366), p = 0.672). As suggested elsewhere28 sarcopenia and myosteatosis are probably two separate entities with different causes and effects reflecting differ- ent disturbances in metabolic processes. Underlying causes of sarcopenia and myostea- tosis are most likely overlapping to some extent. Possible mechanisms, through which they nega- tively affect survival, are various. Diminished food intake due to dysphagia and loss of appetite as well as a chronic inflammation state in esopha- geal cancer lead to sarcopenia. This in turn causes diminished mobility and rehabilitation after sur- gery41, respiratory complications33, inferior wound healing42 and diminished tolerance of chemo and radiotherapy.35 Skeletal muscle has been described as an endocrine organ43 and it is the derangement of this function that is also a possible cause of infe- rior survival. Carefully designed studies are need- ed to corroborate this hypothesis. The inclusion of myosteatosis assessment is in our opinion a strength of our study. We see that myosteatosis is more prevalent than sarcopenia and is a more sensitive marker of muscle degra- dation which precedes muscle mass and overall body mass loss. It is nevertheless at least as detri- mental to prognosis as sarcopenia. Our study also uses recently published cut-off values that in our opinion assess the incidence of sarcopenia better than previous studies. However, this hinders the comparability of our results with others. It is not without weaknesses either. All CT images were re- corded at staging with approximately half the pa- tients going straight to resection and the other half receiving neoadjuvant treatment first. No repeat CT images were taken after neoadjuvant treatment if there were no clinical signs of progression accord- ing to our group’s guidelines. The distribution of intervals from CT to esophagectomy is therefore bimodal and the planimetric data reflects patients’ muscle reserves at beginning of any treatment and not necessarily at esophagectomy. This is a short- coming when assessing the impact on periopera- tive mortality and complications since muscle mass loss is a well known process during neoadjuvant therapy.44-47 The large variation in times between CT and esophagectomy should in our opinion how- ever not be regarded as a weakness when assessing the impact on overall survival of radical esophageal cancer treatment. Our study also lacks statistical power to detect a potential difference in mortal- ity and complications, an issue that has fraught all previous studies as well. With growing numbers of cases in which CT images are available for analysis and with potential pooling of data these statistical issues can be overcome in the future. Lastly, due to the univariate nature of our anal- ysis no causal effect between survival and muscle depletion markers can be established, but the asso- ciation shown can serve as an incentive for further research. Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy 245 Conclusions In a prospective cohort study from a dedicated da- tabase on esophagectomies we studied the associa- tion of sarcopenia and myosteatosis with outcomes after curative esophagectomies with or without neoadjuvant chemoradiotherapy. Prevalence of sarcopenia and myosteatosis at presentation was 16.5% and 51.8%, respectively. Both sarcopenia and myosteatosis were associated with decreased overall survival. For sarcopenia this is in accord- ance with previously published data and for my- osteatosis this is one of the first published reports. Identifying novel predictors of outcomes can be beneficial for tailoring treatment options in pa- tients with esophageal cancer as well as for plan- ning intervention strategies targeted at improving functional body reserves. Authors’ contributions MSr and MSo designed the study. MSr collected, analyzed and interpreted the data and was the ma- jor contributor in writing the manuscript. TJ col- lected the imaging data and was a minor contribu- tor in writing the manuscript. TJ and KP analyzed and interpreted the imaging data. MSo designed the data collecting database. All authors read and approved the final manuscript. Acknowledgements The authors would like to sincerely thank Professor Vickie Baracos, Ph.D., for her invaluable advice and guidance in designing this study. Funding was provided by University Medical Centre Ljubljana, Slovenia. The funding body had no immediate role in the design of the study and collection, analysis, and interpretation of data and in writing the manu- script. References 1. Low DE, Alderson D, Cecconello I, Chang AC, Darling GE, DʼJourno XB, et al. International consensus on standardization of data collection for complications associated with esophagectomy: esophagectomy complica- tions consensus group (ECCG). Ann Surg 2015; 262: 286-94. doi: 10.1097/ SLA.0000000000001098 2. Miller KD, Siegel RL, Lin CC, Mariotto AB, Kramer JL, Rowland JH, et al. Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin 2016; 66: 271-89. doi: 10.3322/caac.21349 3. van Putten M, de Vos-Geelen J, Nieuwenhuijzen G, Siersema PD, Lemmens VEPP, Rosman C, et al. Long-term survival improvement in oesophageal cancer in the Netherlands. Eur J Cancer 2018; 94: 138-47. doi: 10.1016/j. ejca.2018.02.025 4. Soma D, Kawamura YI, Yamashita S, Wake H, Nohara K, Yamada K, et al. Sarcopenia, the depletion of muscle mass, an independent predictor of respiratory complications after oncological esophagectomy. Dis Esophagus 2019; 32. pii: doy092. doi: 10.1093/dote/doy092 5. Markar SR, Low DE. Physiology, not chronology, dictates outcomes after esophagectomy for esophageal cancer: outcomes in patients 80 years and older. Ann Surg Oncol 2013; 20: 1020-6. doi: 10.1245/s10434-012-2703-x 6. Martin L, Birdsell L, MacDonald N, Reiman T, Clandinin MT, McCargar LJ, et al. Cancer cachexia in the age of obesity: skeletal muscle depletion is a powerful prognostic factor, independent of body mass index. J Clin Oncol 2013; 31: 1539-47. doi: 10.1200/JCO.2012.45.2722 7. Gonzalez MC, Correia MITD, Heymsfield SB. A requiem for BMI in the clini- cal setting. Curr Opin Clin Nutr Metab Care 2017; 20: 314-21. doi: 10.1097/ MCO.0000000000000395 8. Di Sebastiano KM, Mourtzakis M. A critical evaluation of body composition modalities used to assess adipose and skeletal muscle tissue in cancer. Appl Physiol Nutr Metab 2012; 37: 811-21. doi: 10.1139/h2012-079 9. Shen W, Punyanitya M, Wang Z, Gallagher D, St-Onge M-P, Albu J, et al. Total body skeletal muscle and adipose tissue volumes: estimation from a single abdominal cross-sectional image. J Appl Physiol Bethesda Md (1985) 2004; 97: 2333-8. doi: 10.1152/japplphysiol.00744.2004 10. Muscaritoli M, Lucia S, Farcomeni A, Lorusso V, Saracino V, Barone C, et al. Prevalence of malnutrition in patients at first medical oncology visit: the PreMiO study. Oncotarget 2017; 8: 79884-96. doi: 10.18632/on- cotarget.20168 11. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, Boirie Y, Cederholm T, Landi F, et al. Sarcopenia: European consensus on definition and diagnosis. Age Ageing 2010; 39: 412-23. doi: 10.1093/ageing/afq034 12. Delmonico MJ, Harris TB, Visser M, Park SW, Conroy MB, Velasquez-Mieyer P, et al. Longitudinal study of muscle strength, quality, and adipose tissue infiltration. Am J Clin Nutr 2009; 90: 1579-85. doi: 10.3945/ajcn.2009.28047 13. Miljkovic I, Kuipers AL, Cvejkus R, Bunker CH, Patrick AL, Gordon CL, et al. Myosteatosis increases with aging and is associated with incident diabetes in African ancestry men. Obes Silver Spring Md 2016; 24: 476-82. doi: 10.1002/oby.21328 14. Goodpaster BH, Theriault R, Watkins SC, Kelley DE. Intramuscular lipid con- tent is increased in obesity and decreased by weight loss. Metabolism 2000; 49: 467-72. doi: 10.1016/s0026-0495(00)80010-4 15. Sabel MS, Lee J, Cai S, Englesbe MJ, Holcombe S, Wang S. Sarcopenia as a prognostic factor among patients with stage III melanoma. Ann Surg Oncol 2011; 18: 3579-85. doi: 10.1245/s10434-011-1976-9 16. Antoun S, Lanoy E, Iacovelli R, Albiges-Sauvin L, Loriot Y, Merad-Taoufik M, et al. Skeletal muscle density predicts prognosis in patients with metastatic renal cell carcinoma treated with targeted therapies. Cancer 2013; 119: 3377-84. doi: 10.1002/cncr.28218 17. Aubrey J, Esfandiari N, Baracos VE, Buteau FA, Frenette J, Putman CT, et al. Measurement of skeletal muscle radiation attenuation and basis of its biological variation. Acta Physiol Oxf Engl 2014; 210: 489-97. doi: 10.1111/ apha.12224 18. Jordan T, Mastnak DM, Palamar N, Kozjek NR. Nutritional therapy for patients with esophageal cancer. Nutr Cancer 2018; 70: 23-9. doi: 10.1080/01635581.2017.1374417 19. Arends J, Baracos V, Bertz H, Bozzetti F, Calder PC, Deutz NEP, et al. ESPEN expert group recommendations for action against cancer-related malnutrition. Clin Nutr Edinb Scotl 2017; 36: 1187-96. doi: 10.1016/j. clnu.2017.06.017 20. Dindo D, Demartines N, Clavien P-A. Classification of surgical complications. Ann Surg 2004; 240: 205-13. doi: 10.1097/01.sla.0000133083.54934.ae 21. Seder CW, Raymond DP, Wright CD, Gaissert HA, Chang AC, Clinton S, et al. The Society of Thoracic Surgeons General Thoracic Surgery Database 2017 update on outcomes and quality. Ann Thorac Surg 2017; 103: 1378-83. doi: 10.1016/j.athoracsur.2017.02.073 Radiol Oncol 2020; 54(2): 237-246. Srpčič M et al. / Sarcopenia and myosteatosis impair survival after esophagectomy246 22. Popuri K, Cobzas D, Esfandiari N, Baracos V, Jägersand M. Body composition assessment in axial CT images using FEM-based automatic segmentation of skeletal muscle. IEEE Trans Med Imaging 2016; 35: 512-20. doi: 10.1109/ TMI.2015.2479252 23. Chung H, Cobzas D, Birdsell L, Lieffers J, Baracos V. Automated segmentation of muscle and adipose tissue on CT images for human body composition analysis. Med Imaging 2009; 7261: 72610K. doi: 10.1117/12.812412 24. van der Werf A, Langius JAE, de van der Schueren MAE, Nurmohamed SA, van der Pant KAMI, Blauwhoff-Buskermolen S, et al. Percentiles for skeletal muscle index, area and radiation attenuation based on computed tomog- raphy imaging in a healthy Caucasian population. Eur J Clin Nutr 2018; 72: 288-96. doi: 10.1038/s41430-017-0034-5 25. Deng H-Y, Hou L, Zha P, Huang K-L, Peng L. Sarcopenia is an independent unfavorable prognostic factor of non-small cell lung cancer after surgical resection: a comprehensive systematic review and meta-analysis. Eur J Surg Oncol 2019; 45: 728-35. doi: 10.1016/j.ejso.2018.09.026 26. Levolger S, van Vugt JLA, de Bruin RWF, IJzermans JNM. Systematic review of sarcopenia in patients operated on for gastrointestinal and hepatopancrea- tobiliary malignancies. Br J Surg 2015; 102: 1448-58. doi: 10.1002/bjs.9893 27. Waduud MA, Wood B, Keleabetswe P, Manning J, Linton E, Drozd M, et al. Influence of psoas muscle area on mortality following elective abdominal aortic aneurysm repair. Br J Surg 2019; 106: 367-74. doi: 10.1002/bjs.11074 28. Stretch C, Aubin J-M, Mickiewicz B, Leugner D, Al-Manasra T, Tobola E, et al. Sarcopenia and myosteatosis are accompanied by distinct biological profiles in patients with pancreatic and periampullary adenocarcinomas. PloS One 2018; 13: e0196235. doi: 10.1371/journal.pone.0196235 29. Paireder M, Asari R, Kristo I, Rieder E, Tamandl D, Ba-Ssalamah A, et al. Impact of sarcopenia on outcome in patients with esophageal resection fol- lowing neoadjuvant chemotherapy for esophageal cancer. Eur J Surg Oncol 2017; 43: 478-84. doi: 10.1016/j.ejso.2016.11.015 30. Elliott JA, Doyle SL, Murphy CF, King S, Guinan EM, Beddy P, et al. Sarcopenia: prevalence, and impact on operative and oncologic outcomes in the multimodal management of locally advanced esophageal cancer. Ann Surg 2017; 266: 822-30. doi: 10.1097/SLA.0000000000002398 31. Järvinen T, Ilonen I, Kauppi J, Salo J, Räsänen J. Loss of skeletal muscle mass during neoadjuvant treatments correlates with worse prognosis in esopha- geal cancer: a retrospective cohort study. World J Surg Oncol 2018; 16: 27. doi: 10.1186/s12957-018-1327-4 32. Fearon K, Strasser F, Anker SD, Bosaeus I, Bruera E, Fainsinger RL, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol 2011; 12: 489-95. doi: 10.1016/S1470-2045(10)70218-7 33. Nishigori T, Okabe H, Tanaka E, Tsunoda S, Hisamori S, Sakai Y. Sarcopenia as a predictor of pulmonary complications after esophagectomy for thoracic esophageal cancer. J Surg Oncol 2016; 113: 678-84. doi: 10.1002/jso.24214 34. Prado CMM, Lieffers JR, McCargar LJ, Reiman T, Sawyer MB, Martin L, et al. Prevalence and clinical implications of sarcopenic obesity in pa- tients with solid tumours of the respiratory and gastrointestinal tracts: a population-based study. Lancet Oncol 2008; 9: 629-35. doi: 10.1016/ S1470-2045(08)70153-0 35. Dijksterhuis WPM, Pruijt MJ, van der Woude SO, Klaassen R, Kurk SA, van Oijen MGH, et al. Association between body composition, survival, and toxicity in advanced esophagogastric cancer patients receiving palliative chemotherapy. J Cachexia Sarcopenia Muscle 2019; 10: 199-206. doi: 10.1002/jcsm.12371 36. Tamandl D, Paireder M, Asari R, Baltzer PA, Schoppmann SF, Ba-Ssalamah A. Markers of sarcopenia quantified by computed tomography predict ad- verse long-term outcome in patients with resected oesophageal or gastro- oesophageal junction cancer. Eur Radiol 2016; 26: 1359-67. doi: 10.1007/ s00330-015-3963-1 37. Gabiatti CTB, Martins MCL, Miyazaki DL, Silva LP, Lascala F, Macedo LT, et al. Myosteatosis in a systemic inflammation-dependent manner predicts favorable survival outcomes in locally advanced esophageal cancer. Cancer Med 2019; 8: 6967-76. doi: 10.1002/cam4.2593 38. Boshier PR, Heneghan R, Markar SR, Baracos VE, Low DE. Assessment of body composition and sarcopenia in patients with esophageal cancer: a systematic review and meta-analysis. Dis Esophagus 2018; 31. doi: 10.1093/ dote/doy047 39. Deng H-Y, Zha P, Peng L, Hou L, Huang K-L, Li X-Y. Preoperative sarcopenia is a predictor of poor prognosis of esophageal cancer after esophagectomy: a comprehensive systematic review and meta-analysis. Dis Esophagus 2019; 32. doi: 10.1093/dote/doy115 40. Tsukioka T, Nishiyama N, Izumi N, Mizuguchi S, Komatsu H, Okada S, et al. Sarcopenia is a novel poor prognostic factor in male patients with pathologi- cal Stage I non-small cell lung cancer. Jpn J Clin Oncol 2017; 47: 363-8. doi: 10.1093/jjco/hyx009 41. JR, Bathe OF, Fassbender K, Winget M, Baracos VE. Sarcopenia is associated with postoperative infection and delayed recovery from colorectal cancer resection surgery. Br J Cancer 2012; 107: 931-6. doi: 10.1038/bjc.2012.350 42. Achim V, Bash J, Mowery A, Guimaraes AR, Li R, Schindler J, et al. Prognostic indication of sarcopenia for wound complication after total laryngectomy. JAMA Otolaryngol Head Neck Surg 2017; 143: 1159-65. doi: 10.1001/ jamaoto.2017.0547 43. Pedersen BK, Febbraio MA. Muscles, exercise and obesity: skeletal muscle as a secretory organ. Nat Rev Endocrinol 2012; 8: 457-65. doi: 10.1038/ nrendo.2012.49 44. Guinan EM, Doyle SL, Bennett AE, O’Neill L, Gannon J, Elliott JA, et al. Sarcopenia during neoadjuvant therapy for oesophageal cancer: charac- terising the impact on muscle strength and physical performance. Support Care Cancer 2018; 26: 1569-76. doi: 10.1007/s00520-017-3993-0 45. Reisinger KW, Bosmans JWAM, Uittenbogaart M, Alsoumali A, Poeze M, Sosef MN, et al. Loss of skeletal muscle mass during neoadjuvant chemora- diotherapy predicts postoperative mortality in esophageal cancer surgery. Ann Surg Oncol 2015; 22: 4445-52. doi: 10.1245/s10434-015-4558-4 46. Yip C, Goh V, Davies A, Gossage J, Mitchell-Hay R, Hynes O, et al. Assessment of sarcopenia and changes in body composition after neoadjuvant chemo- therapy and associations with clinical outcomes in oesophageal cancer. Eur Radiol 2014; 24: 998-1005. doi: 10.1007/s00330-014-3110-4 47. Awad S, Tan BH, Cui H, Bhalla A, Fearon KCH, Parsons SL, et al. Marked changes in body composition following neoadjuvant chemotherapy for oe- sophagogastric cancer. Clin Nutr Edinb Scotl 2012; 31: 74-7. doi: 10.1016/j. clnu.2011.08.008