Pre-operative Psoas Muscle Size Combined With Radiodensity Predicts Mid-Term Survival and Quality of Life After Fenestrated-Branched Endovascular Aortic Repair

1 TITLE PAGE – Original Article Preoperative Psoas Muscle Size and Radiodensity Predict Mid-Term Survival and Quality of Life after Fenestrated-Branched Endovascular Aortic Repair Jussi M. Kärkkäinen MD PhD1, Emanuel R. Tenorio MD PhD1, Niku Oksala MD PhD DSc(med)2, Thanila A. Macedo MD1, Sen Indrani MD1, Bernardo C. Mendes MD1, Randall R. DeMartino MD1, Michael J. Jacobs MD PhD3, Barend Mees MD PhD3 and Gustavo S. Oderich MD1 * From the 1Advanced Endovascular Aortic Research Program, Division of Vascular and Endovascular Surgery, Mayo Clinic, Rochester, MN, USA; 2Department of Surgery, Faculty of Medicine and Health Technology, University of Tampere, Finland and Division of Vascular Surgery, Tampere University Hospital, Tampere, Finland and Finnish Cardiovascular Research Center, Tampere, Finland; and 3Maastricht University Medical Center, Maastricht, Netherlands, European Vascular Center Aachen-Maastricht, Germany and The Netherlands. The abstract of this study has been submitted to European Society for Vascular Surgery (ESVS) 2019 congress in Hamburg. * Corresponding author: Gustavo S. Oderich, Gonda Vascular Center, Mayo Clinic, 200 1st Street SW, Rochester, MN 55902, USA. E-mail: [email protected] Abstract: 298/300 words Body: 4070/4000 words, references are included in the word count Tables/figures: 7/7 tables and figures (3 tables, 4 figures) Supplements: 5 supplementary tables, 3 supplementary figures This is the accepted manuscript of the article, which has been published in European Journal of Vascular and Endovascular Surgery. 2020, 59(1), 31-39. https://doi.org/10.1016/j.ejvs.2019.06.021 2 Abstract (298/300 words) Objective: To investigate the association of psoas muscle area (PMA) and density (PMD) with survival and quality of life (QOL) after fenestrated-branched endovascular aortic repair (F-BEVAR). Methods: The study included 244 consecutive patients enrolled in a prospective study to investigate outcomes of F-BEVAR. Preoperative computed tomography angiography was used to measure PMA (cm2) and PMD (Hounsfield unit, HU) at L3-level. Lean psoas muscle area (LPMA) was calculated (PMA×PMD). Patients were divided into two groups using LPMA cut-point based on Cox hazard model. Group A was defined as LPMA≥350 (n=79) and group B as LPMA<350 cm2×HU (n=165). QOL was assessed at baseline and 12 months using SF-36 survey. Results: Patients in Group A were younger (mean age, 72±8 vs 76±7 years, P<.001), more often males (95% vs 59%, P<.001) and had higher body mass index (30±6 vs 27±5 kg/m2, P=.001). There were no major differences in comorbidities, aneurysm extent and procedural measures between the groups. Thirty-day mortality (0% vs 0.6%, P=1.00) and major adverse event rates (15% vs 24%, P=.18) were similar in Group A and B. At 3 years, patient survival was 94±3% in Group A and 75±4% in Group B (hazard ratio [HR] 0.20, 95% confidence interval [CI] 0.07-0.56, P=.002). The 3-year survival difference was even more prominent in patients aged ≥75 years: 100% for Group A and 72±5% for Group B (HR 0.12, 95% CI 0.02-0.86, P=.035). Group A patients had significantly higher QOL scores at baseline and 12 months. LPMA was the strongest independent predictor of survival during the follow-up in multivariable analysis (adjusted HR 0.59 per one standard deviation, 95% CI 0.40-0.87, P=.008). 3 Conclusions: A high LPMA was independently and strongly associated with better mid-term survival and quality of life after F-BEVAR. LPMA may help to identify best candidates for F-BEVAR among elderly patients. Keywords: psoas, sarcopenia, frailty, fenestrated, branched, endovascular aortic repair What this paper adds: (99/100 words) This study investigated a novel surrogate measure of sarcopenia called lean psoas muscle area (LPMA) and its feasibility as a predictor of survival, outcomes and quality of life after fenestratedbranched endovascular aortic repair (F-BEVAR). LPMA was calculated by multiplying psoas muscle area with radiodensity measured from a single axial preoperative computed tomography angiography slice at L3 level. LPMA proved to be the strongest independent preoperative predictor of mid-term survival after F-BEVAR, especially in the elderly. In addition, high LPMA was associated with better quality of life after repair. Thus, LPMA can be used to identify suitable candidates for F-BEVAR. 4 Introduction Fenestrated-branched endovascular aortic repair (F-BEVAR) has allowed treatment of pararenal and thoracoabdominal aortic aneurysms (TAAAs) in elderly and fragile patients who would otherwise be unfit for open surgical repair (1). The primary treatment goal is prevention of aneurysm rupture and aortic-related death, prolonging overall survival. However, postoperative complications, re-interventions and the requirement for life-long surveillance may compromise the patient's quality of life (2). The question for many elderly patients with complex aneurysms is not whether the procedure is technically feasible, but rather if repair should be indicated in patients with relatively short life expectancy or when treatment may compromise the patient’s ability to live independently. In many patients, operative risk and life expectancy can be difficult to determine. Comorbidities, larger aneurysm size and more extensive aneurysmal disease have been strongly associated with higher mortality, whereas age alone is a poor predictor of survival after F-BEVAR (3,4). Due to the disparity between chronological and biological age, there is a need for better tools to determine operative risk and life expectancy. Frailty is a complex process of age-associated decline in overall physiologic reserve and functioning (5,6). Frailty appears to be superior to many conventional anesthesiologic risk scores in estimating survival after surgical procedures (7). However, frailty can be exhaustively difficult to determine; there are currently more than 70 assessment tools and no agreement on how to measure it (8). Sarcopenia is a component of frailty characterized by loss of skeletal muscle mass and it has been associated with an increased risk of all-cause mortality and functional decline (9). Although there is currently no consensus on the definition, measurement of cross-sectional psoas muscle area (PMA) from axial computed tomography (CT) images has been shown to be a reproducible and convenient surrogate for sarcopenia (10-13). A recent systematic review of 24 studies involving 5267 patients undergoing abdominal surgery for various conditions showed that presence of 5 sarcopenia, assessed by peri-operative CT, is associated with worse long-term survival and significant increase in major post-operative complications and 30-day mortality (14). In recent years, several authors have published promising results suggesting that PMA could be used as a novel prognostic tool for patient survival after open and endovascular aortic repair of abdominal aortic aneurysms (AAAs) (13, 15-18). The initial enthusiasm was shadowed by two recent studies that failed to replicate these results (19,20). However, all these previous studies evaluated only muscle size as a predictor. This does not take the importance of muscle quality over quantity into account. Indeed, Lindström and colleagues demonstrated that PMA alone was not sufficient to predict mortality among 301 patients who underwent elective open AAA repair. Their group discovered that lean psoas muscle area (LPMA), a composite of psoas muscle size and radiodensity, was independently associated with patient survival in multivariable analysis (12). The association of psoas muscle measurements, including PMA and psoas muscle density (PMD), with treatment outcomes and survival has not yet been studied in patients with complex aneurysms undergoing F-BEVAR. Moreover, association of sarcopenia with quality of life after aneurysm repair has not been investigated. The aim of this study was to examine the association of PMA and PMD, measured from preoperative CT angiography (CTA), with mid-term survival (primary end point), in-hospital outcomes and quality of life (secondary end points) after F-BEVAR of pararenal aneurysms and TAAAs. Methods The study cohort included patients enrolled in a prospective non-randomized single-center study approved by the Institutional Review Board. Participation required informed consent. The FBEVAR was performed using manufactured patient-specific or off-the-shelf Cook Zenith (Cook Medical, Inc., Bloomington, Ind) fenestrated and branched stent grafts under physician sponsored investigational device exemption protocols (numbers G130030 and G130266). A total of 244 6 consecutive patients were enrolled between November 2013 and March 2018. Patient characteristics, cardiovascular risk factors, operative data, length of hospital stay, and 30-day outcomes were collected prospectively and stored in MEDIRAVE database. Study subjects were scored based on Short Form-36 (SF-36) Quality of Life Questionnaire at baseline and 12 months postoperatively; the SF-36 scores were divided into eight subscales. Deaths were retrieved from the medical records, and the survival status of the study patients was verified utilizing the Accurint® database in September 2018. CTA analysis was performed retrospectively in a standardized fashion by one experienced vascular surgeon. Preoperative CTA was used to measure PMA and PMD from a single axial slice using freehand drawing tool of the image display software (QReads). The slice was chosen at the level of L3 vertebrae where the lateral tips of both transverse processes were visible (Figure 1). The regions of interest (ROIs), hence, the left and right psoas muscles, were carefully drawn with the freehand tool according to the anatomical boundaries. If both transverse processes could not be visualized in one axial image due to oblique orientation of the spine (scoliosis), the left and right psoas muscles were drawn in two separate slices where the corresponding transverse processes were most clearly visible. The area (cm2) and average radiodensity (Hounsfield unit, HU) of the ROIs in each side were registered. PMA and PMD were defined as the mean value of left and right psoas muscle measurements (PMA = PMALEFT + PMARIGHT / 2; PMD = PMDLEFT + PMDRIGHT / 2). Lean psoas muscle area (LPMA, cm2×HU) was calculated by multiplying PMA and PMD (LPMA = PMA × PMD). Psoas muscle index (PMI, cm2/m2) was defined as PMA / Height2. If the preoperative imaging was more than 6 months old, postoperative CTA, obtained within one week of the index procedure, was used instead. Preoperative imaging protocol included CTA of the chest, abdomen and pelvis with contrast bolus tracking; the threshold was usually set at 150 HU at the descending thoracic aorta. The contrast agents used were Omnipaque 350 or 300 (Iohexol), or Isovue 300 (Iopamidol). The amount 7 of contrast bolus was between 80 to 150 ml depending on patient’s weight followed by 30 mL saline flush at a rate of 4-6 ml/s. The amount of contrast was reduced, if necessary, in patients with glomerular filtration rate less than 30 ml/min. CTA slices were reconstructed to the thickness of 1-3 mm (typically 2 mm). Any deviation in the imaging protocol was registered as well as any difficulty in assessing the correct level for the psoas muscle measurements (Supplementary Table I). The time from contrast injection to a point when the concentration in the descending thoracic aorta reached 150 HU was registered (hereafter referred to as "CTA bolus-tracking time"). End Points and Statistical Analysis The primary end point of the study was all-cause mortality during the follow-up. Secondary outcome end points included 30-day mortality, major adverse events, length of hospital stay, discharge status (discharge to home versus transfer to another institution such as skilled nursing facility) and significant decrease (≥10 points) in SF-36 subscale scores between baseline and 12 months. The studied CT variables (PMA, PMD, LPMA and PMI) were first tested for association with the primary end point in univariable Cox regression model and the strongest predictor (LPMA) was chosen for grouping of the patients. An optimal cut point value for LPMA was first estimated based on receiver operator characteristics (ROC) analysis (Supplementary Figure 1) and then confirmed by adjusted Cox hazard model using the time-dependent primary end point (Figure 2). Based on the optimal cut point, the study patients were divided in two groups; group A – high muscle mass (LPMA≥350 cm2×HU) and group B – low muscle mass (LPMA<350 cm2×HU). Differences between the study groups were analyzed using Fisher's exact test for categorical variables and Mann-Whitney U test for continuous variables. The categorical variables were expressed as numbers and percentages, and continuous variables as mean ± standard deviation (SD) and median with interquartile range (IQR) when appropriate. P-values <.05 were considered statistically significant. For univariable and multivariable analyses, LPMA was standardized using z-scoring. Cox regression univariable analysis was performed for all preoperative variables to 8 determine significant predictors of survival during the follow-up. All variables with P<.10 in the univariable analysis were included in multivariable Cox model 1, and those variables, that were significantly different (P<.05) between groups A and B, were included in multivariable Cox model 2. The multivariable models were used to determine independent preoperative predictors of survival during the follow-up. Body mass index (BMI) was chosen to represent conventional body mass measures in the multivariable analysis and LPMA for the novel muscle mass measures. The results of the Cox regression analyses were expressed as hazard ratios (HRs) with 95% confidence intervals (CIs). All statistical analyses were performed using SPSS version 24.0 (IBM Corp. Armonk NY, USA). Results Patient Characteristics and Preoperative Risk Assessment There were 244 consecutive patients included in the study. The optimal LPMA cut point was determined at 350 cm2×HU; one-third of the patients (n=79, 32%) had high muscle mass (group A) based on the cut point whereas two-thirds (n=165, 68%) had low muscle mass (group B). The mean age of the patients was 75±8 years, half were 75 years or older and 71% were male. Patients in group A were younger (72±8 vs. 76±7, P<.001) and more often male (95% vs. 59%, P<.001) compared to group B. There were no statistically significant differences in the prevalence of comorbidities between the groups (Table I). Patients in group A had a higher mean BMI than patients in group B (30±6 vs. 27±5 kg/m2, P<.001). Consequently, body surface area and estimated lean body mass were also higher in group A (P<.001). Patients in group A were more often obese; 43% of patients had BMI>30 kg/m2 in group A compared to 22% in group B (P=.001). Out of 241 patients who had cardiac stress test performed, 21% had a positive test with no differences between the groups. Ejection fraction was measured in 236 patients with no difference in mean values between the groups. American Society 9 of Anesthesiologists (ASA) scores were similar between the study groups; 28% had score 3 or higher; only three patients had score 4. There were no major differences in aneurysm size and extent between the groups, although 41% of the aneurysms in group B were extent I-III TAAAs compared to 28% in group A (P=.05). The mean PMA was 8.3±2.7 cm2 (median 8.0 [IQR 6.4-9.8]) and mean LPMA 297±130 cm2×HU (median 279 [IQR 199-372], Supplementary Figure 2). PMA, PMI, PMD and LPMA were all significantly (P<.001) higher in group A (Table II). Procedural Characteristics F-BEVAR was done using patient-specific devices in 222 (91%) patients and off-the-shelf device was used in 22 (9%). There were no major differences in procedural characteristics between the groups (Supplementary Table II). However, the mean operation time was longer in group B (227±71 min vs. 265±87 min, P=.002), and percutaneous femoral access was used slightly more often in group A compared to group B patients (87% vs. 76%, P=.04). Implantation of the aortic stent-graft and all target vessel components was successful in 242 patients (99%). Primary End Point Mid-term survival was significantly higher in group A patients (Figure 3). At 3 years, survival was 94±3% in Group A and 75±4% in Group B (Log Rank P=.001; HR 0.20, 95% CI 0.070.56, P=.002). The survival difference did not change when adjusted for the age, gender and BMI differences between the groups (adjusted HR 0.23, 95% CI 0.08-0.66, P=.006). The study population was divided in half based on the patients’ age for further survival analysis using 75 years as a cut point (Supplementary Figure 3). The 3-year survival difference was even greater in patients aged ≥75 years: 100% for Group A and 72±5% for Group B (Log Rank P=.011; HR 0.12, 95% CI 0.02-0.86, P=.035). The mean follow-up time was 2.1±1.3 years. Secondary End Points The 30-day mortality and major adverse event rates were similar between the study groups. There was only one 30-day death (0.4%) and seven patients (3%) suffered from paraplegia 16 15. Lee JS, He K, Harbaugh CM, Schaubel DE, Sonnenday CJ, Wang SC, et al. Frailty, core muscle size, and mortality in patients undergoing open abdominal aortic aneurysm repair. J Vasc Surg 2011;53(4):912-7. 16. Newton DH, Kim C, Lee N, Wolfe L, Pfeifer J, Amendola M. Sarcopenia predicts poor long-term survival in patients undergoing endovascular aortic aneurysm repair. J Vasc Surg 2018;67(2):453-459. 17. Drudi LM, Phung K, Ades M, Zuckerman J, Mullie L, Steinmetz OK, et al. Psoas Muscle Area Predicts All-Cause Mortality After Endovascular and Open Aortic Aneurysm Repair. Eur J Vasc Endovasc Surg 2016;52(6):764-769. 18. Shah N, Abeysundara L, Dutta P, Christodoulidou M, Wylie S, Richards T, et al. The association of abdominal muscle with outcomes after scheduled abdominal aortic aneurysm repair. Anaesthesia 2017;72(9):1107-1111. 19. Indrakusuma R, Zijlmans JL, Jalalzadeh H, Planken RN, Balm R, Koelemay MJW. Psoas Muscle Area as a Prognostic Factor for Survival in Patients with an Asymptomatic Infrarenal Abdominal Aortic Aneurysm: A Retrospective Cohort Study. Eur J Vasc Endovasc Surg 2018;55(1):83-91. 20. 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 [Epub ahead of print] 21. Kirkland JL, Tchkonia T, Pirtskhalava T, Han J, Karagiannides I. Adipogenesis and aging: does aging make fat go MAD? Exp Gerontol 2002;37(6):757-67. 22. Hamrick MW, McGee-Lawrence ME, Frechette DM. Fatty Infiltration of Skeletal Muscle: Mechanisms and Comparisons with Bone Marrow Adiposity. Front Endocrinol (Lausanne) 2016;7:69. 23. Reinders I, Murphy RA, Brouwer IA, Visser M, Launer L, Siggeirsdottir K, et al. Muscle Quality and Myosteatosis: Novel Associations With Mortality Risk: The Age, Gene/Environment Susceptibility (AGES)-Reykjavik Study. Am J Epidemiol 2016;183(1):53-60. 24. Kays JK, Liang TW, Zimmers TA, Milgrom DP, Abduljabar H, Young A, et al. Sarcopenia is a Significant Predictor of Mortality After Abdominal Aortic Aneurysm Repair. J Cachexia Sarcopenia Muscle Clinical Reports 2018;3(1):e53. 17 Table I. Patient demographics and comorbidities Group A Group B All patients (n=244) High Muscle Mass (n=79) Low Muscle Mass (n=165) P value Mean age, years 75 ± 8 72 ± 8 76 ± 7 <.001 Age ≥ 75 years 127 (52) 29 (37) 98 (59) .001 Male gender 172 (71) 75 (95) 97 (59) <.001 Cigarette smoking 212 (87) 69 (87) 143 (87) 1.00 Hypertension 220 (90) 70 (89) 150 (91) .65 Hypercholesterolemia 201 (82) 66 (84) 135 (82) .86 Coronary artery disease 126 (52) 41 (52) 85 (52) 1.00 Chronic obstructive pulmonary disease 89 (37) 22 (28) 67 (41) .07 Chronic kidney disease (stages III-V) 49 (20) 14 (18) 35 (21) .61 Congestive heart failure 26 (11) 7 (9) 19 (12) .66 Peripheral artery disease 52 (21) 11 (14) 41 (25) .07 Diabetes mellitus 36 (15) 12 (15) 24 (15) 1.00 Stroke 24 (10) 5 (6) 19 (12) .25 Malignancy 53 (22) 16 (20) 37 (23) .74 Prior aortic repair 95 (39) 32 (41) 63 (38) .78 Data are presented as n (%), mean ± standard deviation or median (interquartile range). 18 Table II. Operative risk assessment and CTA measures Group A Group B All patients (n=244) High Muscle Mass (n=79) Low Muscle Mass (n=165) P value Positive cardiac stress test 50/241 (21) 18/78 (23) 32/163 (20) .61 Ejection fraction, % 58 ± 11 57 ± 10 58 ± 11 .98 Ejection fraction < 30 % 5/236 (2) 1/75 (1) 4/161 (2) 1.00 Baseline GFR 61 ± 19 63 ± 19 60 ± 19 .32 Baseline GFR < 30 10 (4) 4 (5) 6 (4) .73 ASA score 2.2 ± 0.6 2.1 ± 0.6 2.2 ± 0.6 .22 ASA score ≥ 3 68/242 (28) 19/79 (24) 49/163 (30) .36 Maximum aneurysm diameter, mm 66 ± 11 65 ± 11 67 ± 12 .39 Aneurysm type Pararenal 97 (37) 35 (44) 56 (34) .12 TAAA extent I-III 90 (37) 22 (28) 68 (41) .05 TAAA extent IV 63 (26) 22 (28) 41 (25) .64 Conventional body mass measures Body mass index, kg/m2 28±6 30 ± 6 27 ± 5 <.001 Body mass index > 30 kg/m2 70 (30) 34 (43) 36 (22) .001 Body surface area, m2 2.0 ± 0.3 2.1 ± 0.3 1.9 ± 0.3 <.001 Estimated lean body mass (eLBM) 58 ± 11 55 ± 10 64 ± 10 <.001 Novel muscle mass measures Psoas muscle area, cm2 8.3 ± 2.7 10.9 ± 2.3 7.1 ± 1.8 <.001 Psoas muscle index, cm2/m2 2.8 ± 0.8 3.5 ± 0.8 2.4 ± 0.6 <.001 Psoas muscle density, HU 35.7 ± 8.9 41.4 ± 6.1 23.9 ± 8.7 <.001 Lean psoas muscle area, cm2×HU 298 ± 130 446 ± 101 226 ± 67 <.001 CTA bolus-tracking time to 150 HU 20.5 ± 4.2 20.8 ± 4.3 20.3 ± 4.2 .47 Data are presented as n (%) or mean ± standard deviation. CTA, computed tomography angiography; GFR, glomerular filtration rate; ASA, American Society of Anesthesiologists; TAAA, thoracoabdominal aortic aneurysm; HU, Hounsfield unit For males: eLBM = 0.407 × Weight(kg) + 0.267 × Height(cm) - 19.2 For females: eLBM = 0.252 × weight(kg) + 0.473 × height(cm) - 48.3 19 Table III. Multivariable analysis of preoperative risk factors for long-term mortality. Cox Regression Multivariable Analysis Hazard ratio 95 % confidence interval P value Lower Upper Multivariable model 1 Body mass index 0.93 0.88 0.99 .03 Congestive heart failure 2.33 1.04 5.20 .04 ASA score ≥3 2.19 1.10 4.36 .03 Lean psoas muscle area (per 1 SD) 0.62 0.42 0.94 .02 Multivariable model 2 Lean psoas muscle area (per 1 SD) 0.59 0.40 0.87 .008 ASA, American Society of Anesthesiologists; SD, standard deviation Model 1 included variables that were considered significant (P<.10) in the univariable analysis. Model 2 included preoperative variables that were significantly different in group A compared to group B; age, gender, body mass index and lean psoas muscle area. 20 Supplementary Table I. Details of the computed tomography angiography (CTA) protocols CTA available for analysis 244 (100%) Preoperative CTA used for analysis (preop CTA ≤6 months old) 230 (94%) Mean time from CTA to surgery 111±57 days Postoperative CTA used for analysis (preop CTA >6 months old) 14 (6%) Mean time from surgery to CTA 3±3 days CTA scan done in the study institution 209 (86%) CTA scan done elsewhere 35 (14%) Imaging area; chest, abdomen, pelvis 220 (90%) Imaging area; abdomen, pelvis 24 (10%) 1-3 mm axial slice thickness 238 (98%) 5 mm axial slice thickness 6 (2%) Trigger threshold for bolus tracking 150 Hounsfield units 191 (78%) 120 Hounsfield units 7 (3%) 100 Hounsfield units 3 (1%) Information unavailable 43 (18%) Both transverse processes visible in the same axial slice 226 (93%) Oblique vertebra (psoas muscles measured at two different levels) 18 (7%) Minor technical challenges in psoas muscle measurement 4 (2%) Fused lumbar vertebrae 2 (1%) Extra lumbar vertebra 1 (0.5%) Severe degeneration of the lumbar spine 1 (0.5%) 21 Supplementary Table II. Procedural characteristics Group A Group B All patients (n=244) High Muscle Mass (n=79) Low Muscle Mass (n=165) P value Fenestrated-branched device type Off-the-self (t-Branch®) 22 (9) 6 (8) 16 (10) .81 Patient-specific 222 (91) 73 (92) 149 (90) Number of incorporated target vessel 3.8 ± 0.6 3.8 ± 0.6 3.8 ± 0.6 .79 Fenestrations 2.6 ± 1.5 2.9 ± 1.5 2.4 ± 1.6 .013 Branches 1.4 ± 1.5 0.8 ± 1.4 1.3 ± 1.3 .015 General anesthesia 244 (100) 79 (100) 165 (100) 1.00 Cerebrospinal fluid drain 166 (68) 50 (63) 116 (70) .31 Neuromonitoring 170 (70) 51 (65) 119 (73) .23 Percutaneous femoral access 193 (79) 68 (87) 125 (76) .04 Upper extremity access 220 (90) 69 (87) 151 (92) .36 Contrast volume, ml 155 ± 56 153 ± 62 156 ± 54 .31 Operation time, min 252 ± 83 227 ± 71 265 ± 87 .002 Estimated blood loss, ml 464 ± 553 419 ± 518 487 ± 570 .47 Technical success 242 (99) 79 (100) 163 (99) 1.00 Any reintervention before discharge 24 (10) 4 (5) 20 (12) .11 Data are presented as n (%) or mean ± standard deviation. 22 Supplementary Table III. Secondary outcome end points Group A Group B All patients (n=244) High Muscle Mass (n=79) Low Muscle Mass (n=165) P value Major adverse event 51 (21) 12 (15) 39 (24) .18 30-day or in-hospital death 1 (0.4) 0 (0) 1 (0.6) 1.00 Estimated blood loss > 1000 ml 21 (9) 6 (8) 15 (9) .81 Acute kidney injury (RIFLE) 26 (11) 3 (4) 23 (14) .02 Risk (↓ GFR > 25 %) 23 (9) 3 (4) 20 (12) .04 Injury/Failure (↓ GFR > 50 %) 3 (1) 0 (0) 3 (2) .55 Myocardial infarction 10 (4) 2 (3) 8 (5) .51 Respiratory failure 8 (3) 0 (0) 8 (5) .06 Paraplegia 7 (3) 0 (0) 7 (4) .10 Stroke 7 (3) 0 (0) 7 (4) .10 Bowel ischemia 3 (1) 0 (0) 3 (2) .55 Hospital length of stay, days 7.6 ± 10.1 4.7 ± 2.9 9.0 ± 12.0 <.001 Discharge to home 200 (82) 70 (89) 130 (79) .08 Data are presented as n (%) or mean ± standard deviation. RIFLE, risk-injury-failure classification; GFR, glomerular filtration rate 23 Supplementary Table IV. Quality of life outcomes Group A Group B All patients (n=244) High Muscle Mass (n=79) Low Muscle Mass (n=165) P value SF-36 questionnaires completed At baseline 237 (97) 75 (95) 162 (98) .22 At 12 months 161 (66) 63 (80) 98 (59) .002 At baseline and 12 months 157 (64) 61 (77) 96 (58) .004 Individuals with significant decrease (≥10 points) in SF-36 scores between baseline and 12 months Physical Functioning 64 (41) 19 (31) 45 (47) .07 Role Physical 62 (40) 21 (34) 41 (43) .32 Role Emotional 41 (26) 8 (13) 33 (34) .003 Vitality 69 (44) 21 (34) 48 (50) .07 Mental Health 35 (22) 9 (15) 26 (27) .08 Social Functioning 45 (29) 11 (18) 34 (35) .02 Bodily pain 62 (40) 24 (39) 38 (40) 1.00 General Health 63 (40) 27 (44) 36 (38) .41 Data are presented as n (%). 24 Supplementary Table V. Univariable analysis of preoperative risk factors for mortality during the follow-up. Variables with P<.10 and those of special interest are included in the table. Cox Regression Univariable Analysis Hazard ratio 95 % confidence interval P value Lower Upper Patient demographics and comorbidities Age 1.04 0.99 1.09 .10 Gender, male 0.64 0.33 1.21 .17 Hypercholesterolemia 0.47 0.23 0.95 .04 Chronic obstructive pulmonary disease 1.86 0.99 3.48 .054 Chronic kidney disease (stages III-V) 1.76 0.89 3.48 .10 Congestive heart failure 2.38 1.09 5.17 .03 Operative risk assessment Positive cardiac stress test 0.58 0.23 1.49 .26 Ejection fraction 1.01 0.98 1.05 .40 Ejection fraction < 30 % 4.01 0.95 16.85 .06 ASA score 1.68 1.01 2.80 .04 ASA score ≥ 3 2.04 1.05 3.95 .04 Maximum aneurysm diameter (per 1 mm) 1.02 1.00 1.04 .054 Aneurysm type: TAAA extent I-III 1.42 0.75 2.69 .29 Conventional body mass measures Body mass index 0.93 0.89 0.98 .009 Body mass index > 30 kg/m2 0.34 0.13 0.86 .02 Body surface area 0.34 0.13 0.93 .04 Estimated lean body mass 0.97 0.94 1.00 .05 Novel muscle mass measures Psoas muscle area 0.90 0.79 1.02 .11 Psoas muscle index 0.71 0.47 1.08 .11 Psoas muscle density 0.96 0.92 0.99 .02 Lean psoas muscle area (per 1 cm2×HU) 0.996 0.993 0.999 .008 Lean psoas muscle area (per 1 SD) 0.59 0.40 0.87 .008 CTA bolus-tracking time to 150 HU 0.99 0.91 1.08 .84 TAAA, thoracoabdominal aortic aneurysm; ASA, American Society of Anesthesiologists; SVS, Society for Vascular Surgery; HU, Hounsfield unit; SD, standard deviation 25 Figure 1. Illustration showing the anatomical landmarks for choosing the correct CT slice at L3 level. The axial slice, where the lateral tips of both transverse processes are best visualized, is chosen for the psoas muscle area and density measurements. The upper CT slice on the right is from a patient with large psoas muscle, and the lower is an example of atrophied psoas muscle. By permission of Mayo foundation for Medical education and research. All rights reserved.