2. 2 Vascular Medicine
(KT). The summation of these muscle groups represented
the total skeletal muscle area (cm2) used for establishing the
presence or absence of sarcopenia. Sarcopenia was defined
as having a skeletal muscle area <114.0 cm2 (men) or <89.8
cm2 (women); this definition was originally derived from
data in donor liver transplantation patients and later used to
assess sarcopenia in CLI patients.2,3 Sarcopenia is not yet a
well-defined entity; as such, we performed a quartile analy-
sis on the distribution of muscle mass in EVAR patients to
evaluate the definition of sarcopenia used in the literature.
Data collection and statistical analysis
Data collection included patient demographics and comor-
bidities, including: age, sex, race, diabetes (DM), smok-
ing status, coronary artery disease (CAD), hypertension
(HTN), obesity, chronic obstructive pulmonary disease
(COPD), and AAA diameter (cm). The presence of CAD
was defined as having an abnormal electrocardiogram,
prior myocardial infarction, prior coronary bypass, or
prior percutaneous coronary intervention. Hypertension
was defined as having a resting blood pressure greater
than 140/90 mmHg or requiring antihypertensive medica-
tions. Obesity was defined as having a body mass index
(BMI) greater than 30.0. COPD was defined as having a
history of pulmonary obstructive disease or required use
of home oxygen secondary to lung disease. Data were
analyzed according to the last available follow-up visit.
Social security death index data were helpful in determin-
ing the date of death, which was the primary endpoint.
Follow up was defined as the date of the procedure to the
last date of observation.
Baseline descriptive statistics were used to determine the
study population, followed by bivariate analysis to determine
the differences in patient demographics. The differences in
continuous variables were analyzed using parametric t-tests
and the differences in discrete variables were analyzed with
chi-squared or, in the case of small sample sizes (n<5),
Fisher’s exact tests. Differences in Kaplan–Meier survival
curves were analyzed using a log-rank test. Statistically sig-
nificant covariates in the bivariate analyses were incorpo-
rated into a multivariate logistic regression. Resulting odds
ratios and 95% confidence intervals were calculated for each
covariate in the multivariate analysis. Data analysis was
performed using R statistical software (version 3.1.3; R
Foundation for Statistical Computing, Vienna, Austria).
Results
Patient characteristics and analysis
The original abdominal CT scans on 200 patients withAAA
who subsequently underwent EVAR by the vascular sur-
gery service at Greenville Health System were analyzed.
Overall demographic information is presented in Table 1.
There were 175 men and 25 women in the study. Mean age
at the time of treatment was 74 ± 7.5 years. The overall
mortality rate was 51%, with a median follow up of 8.4
years (interquartile range, 5.3–11.7) and median time to
death of 5.4 years (interquartile range, 3.0–8.4). Five major
postoperative complications were noted: two ruptures, one
graft migration, one graft infection, and one acute aneu-
rysm expansion.
The mean skeletal muscle area for all patients was 138.2
± 27.9 cm2. The majority of the EVAR sarcopenia patients
were in the lowest quartile of total muscle mass. The break-
down of the quartile analysis on skeletal muscle mass is
described in Table 2.
From the 200 AAA patients, 25 had sarcopenia and 175
did not. Demographic differences between patients with
sarcopenia and without are also described in Table 1.
Patients with sarcopenia tended to be older (77.9 vs 73.0
years; p=0.002), female (32% vs 9.7%; p=0.005), and have
a slightly larger aneurysm at the time of repair (5.87 cm vs
5.53 cm; p=0.080). Furthermore, patients with sarcopenia
had a significantly higher mortality rate during follow-up
than those without (76% vs 48%; p=0.016) (Figure 1).
Figure 1 shows the Kaplan–Meier life table analysis sur-
vival probabilities. The survival curves start to separate
between years 3 and 4, with a statistical difference that
strengthens over time (log-rank test, p=0.016).
When comparing living with deceased patients, three
important statistical demographic differences were seen
in those who died: increased age at time of EVAR (75.9
vs 71.2 years; p<0.001), presence of hypertension (92.2%
vs 81.4%; p=0.040), and presence of sarcopenia (18.5% vs
6.19%; p=0.016) (Table 1). Logistic regression analysis
was used to compare these three significant variables; sex
was also included in the analysis, as it was strongly associ-
ated with sarcopenia (Table 3). The odds ratios and 95%
confidence intervals shown in Table 3 demonstrate the sig-
nificance of sarcopenia (OR 3.17, 95% CI 1.20–9.54),
hypertension (OR 2.73, 95% CI 1.17–7.11), and advanced
age (76–85 years) (OR 0.97, 95% CI 1.04–3.58) on mortal-
ity following EVAR. Sex did not play a significant role in
mortality, nor did it act as a confounding factor with the
other three variables.
Discussion
These data confirm sarcopenia as an independent predictor
of long-term mortality in patients treated for AAA with
EVAR. Although risk prediction models for elective aneu-
rysm repair have been created, they primarily center on
30-day mortality, as opposed to long-term mortality (i.e.
Glasgow aneurysm score, Leiden Score, Hardman Index).4–6
However, in 2013, the Vascular Surgical Group of New
England (VSGNE) did identify four major and four minor
risk factors for assessing long-term survival (5 years) fol-
lowing AAA repair.7 Their major risk criteria included:
unstable angina or recent myocardial infarction, age >80
years, oxygen-dependent COPD, and estimated glomerular
filtration rate <30 ml/min/1.73 m2. The minor risk criteria
included: age 75–79 years, prior myocardial infarction, sta-
ble angina, and not taking aspirin or statins. Based on these
criteria, patients were evaluated and stratified as ‘low risk’
(no major risk factors, 1–2 minor risk factors), ‘medium
risk’ (1 major, 1–3 minor), or ‘high risk’ (1–2 major, >3
minor). The VSGNE ‘high risk’ group showed a mortality
rate of 43% at 5 years; this compares favorably to our group
of EVAR patients with sarcopenia who had a 5-year
at HIGH POINT UNIV on April 23, 2016vmj.sagepub.comDownloaded from
3. Hale et al. 3
mortality rate of 40%. These data suggest that preoperative
risk assessment of long-term survival could potentially be
simplified to measuring sarcopenia. Complex risk predic-
tion models tend to lose calibration over time, so the sim-
pler the model, the less recalibration is needed.8
In our study, patients with sarcopenia had a signifi-
cantly higher long-term mortality rate during follow up
than those without (76% vs 48%; p=0.016). However,
noticeable separation between the two survival curves in
Figure 1 is not seen until year 4, with the gap continuing to
increase over time. As future studies on other patient popu-
lations are conducted, longer follow-up (5+ years) may be
necessary to accurately assess the long-term impact sarco-
penia has on patient outcomes, specifically mortality. The
measurement technique and cutoff values we used to
determine and define sarcopenia were previously utilized
in a study that evaluated mortality in patients with CLI.2
The cutoff values for skeletal muscle mass (<114.0 cm2 for
men and <89.8 cm2 for women) represent patients below
the fifth percentile of the standard value in healthy adults.3
We believe this is representative of our AAA population,
Table 1. Patient characteristics.
All Sarcopenia Mortality
Yes No p-value Deceased Living p-value
n 200 25 175 103 97
Age, years, mean (SD) 74 (7.5) 77.9 (7.5) 73.0 (7.4) 0.002 75.9 (7.2) 71.2 (7.1) <0.001
Race, n (%) 0.873 0.878
White 188 (94) 24 (96.0) 164 (93.7) 97 (94.2) 91 (93.8)
African American 11 (5.5) 1 (4.0) 10 (5.7) 6 (5.83) 5 (5.15)
Other 1 (0.5) 0 (0) 1 (0.57) 0 (0) 1 (1.03)
Sex, n (%) 0.005 0.789
Male 175 (87.5) 17 (68.0) 158 (90.3) 89 (86.4) 86 (88.7)
Female 25 (12.5) 8 (32.0) 17 (9.7) 14 (13.6) 11 (11.3)
Smoking status, n (%) 0.815 0.162
Former 109 (54.5) 15 (60.0) 94 (53.7) 62 (60.2) 47 (48.5)
Never 33 (16.5) 4 (16.0) 29 (16.6) 17 (16.5) 16 (16.5)
Current 58 (29) 6 (24.0) 52 (29.7) 24 (23.3) 34 (35.0)
ESRD, n (%) 0.267 0.107
Renal insufficiency 30 (15) 4 (16.0) 26 (14.9) 19 (18.5) 11 (11.3)
No 168 (84) 1(4.0) 1 (0.57) 82 (79.6) 86 (88.7)
Yes 2 (1) 20 (80.0) 148 (54.6) 2 (1.9) 0 (0)
DM, n (%) 42 (21.0) 5 (20.0) 37 (21.1) 1.000 25 (24.3) 17 (17.5) 0.319
CAD, n (%) 123 (61.5) 17 (68.0) 106 (60.6) 0.621 68 (66.0) 55 (56.7) 0.163
HTN, n (%) 174 (87) 22 (88.0) 152 (86.9) 1.000 95 (92.2) 79 (81.4) 0.040
Hyperlipid, n (%) 120 (60) 12 (48.0) 108 (61.7) 0.275 59 (57.3) 61 (62.9) 0.507
Obese, n (%) 33 (16.5) 2 (8.0) 31 (17.7) 0.349 16 (15.5) 17 (17.5) 0.850
COPD, n (%) 52 (26) 11 (44.0) 41 (23.4) 0.051 32 (31.1) 20 (20.6) 0.128
Dementia, n (%) 8 (4) 1 (4.0) 7 (4.0) 1.000 4 (3.88) 4 (4.12) 1.000
Sarcopenia, n (%) 25 (12.5) 19 (18.5) 6 (6.19) 0.016
Men 17 (9.7) 17 (68.0) 0 (0) 13 (12.6) 4 (4.12) 0.049
Women 8 (32.0) 8 (32.0) 0 (0) 6 (5.83) 2 (2.06) 0.378
Total skeletal area,
cm2 mean (SD)
138.2 (27.9) 96.2 (14.0) 144.2 (24.0) <0.001 133.6 (27.1) 143.0 (28.2) 0.017
AAA diameter, cm,
mean (SD)
5.76 (0.91) 5.87 (1.13) 5.53 (0.87) 0.080 5.65 (1.04) 5.50 (0.73) 0.233
Mortality 103 (51.5) 19 (76.0) 84 (48.0) 0.016 103 (100.0)
Time till death, years
(median)
5.43 4.94 5.66 0.532 5.91
ESRD, end-stage renal disease; DM, diabetes mellitus; CAD, coronary artery disease; HTN, hypertension; COPD, chronic obstructive pulmonary
disease;AAA, abdominal aortic aneurysm.
Table 2. Total skeletal muscle area.
Size cm2 Died, n (%)
Total area
Mean ± SD 138.2 ± 27.9
Quartiles, men
Q1, n=44 < 124.3 28 (63.6)
Q2, n=44 124.4–140.1 24 (54.5)
Q3, n=43 140.2–159.6 19 (43.2)
Q4, n=44 >159.7 18 (40.9)
Quartiles, women
Q1, n=7 < 86.7 5 (71.4)
Q2, n=6 86.8–101.8 3 (50.0)
Q3, n=6 101.9–118.7 3 (50.0)
Q4, n=6 >118.8 3 (50.0)
Quartiles, combined
Q1, n=51 mixeda 33 (64.7)
Q2, n=50 mixed 27 (54.0)
Q3, n=49 mixed 22 (44.9)
Q4, n=50 mixed 21 (42.0)
aMixed by sex differences.
at HIGH POINT UNIV on April 23, 2016vmj.sagepub.comDownloaded from
4. 4 Vascular Medicine
as our 5-year survival rate in those without sarcopenia was
78.3%, compared to 77.5% in their CLI patients without
sarcopenia (Figure 1). The sarcopenic CLI population had
a lower 5-year survival rate (23.5%) than our AAA sarco-
penic population (40%). Sarcopenia is a consistent marker
of reduced survival, particularly in patients with CLI.
A similar measurement technique was used by Lee et al.
in a study that evaluated 262 patients who underwent open
AAA repair.9 They measured psoas muscle area on CT at
the L4 vertebra, as opposed to L3 in our study. At the 2.3-
year mean follow-up, 55 (21%) of their patients had died.
As psoas muscle area decreased, mortality increased; this
relationship was logarithmic and non-linear. Lee et al. used
the word ‘frailty’ to describe this patient-centered charac-
teristic and suggested that the measurement of psoas
muscle volume was an excellent predictor of mortality. Our
study builds upon these data by measuring more muscle
groups to define a threshold of low muscle mass. By defin-
ing sarcopenia to be a natural aspect of aging that involves
the reduction of skeletal muscle tissue, mass, and function,
we feel this is a more specific term than frailty.
A BMI greater than 30 has been shown to increase AAA
mortality risk. Using the National Surgical Quality
Improvement Program database (NSQIP), Giles et al.
showed a twofold increase in mortality in the morbidly
obese compared to the non-obese.10 We analyzed total mus-
cle area in quartiles and saw no relationship to mortality as
long as patients were ‘above the threshold’ for sarcopenia
(Table 2). With that said, the morbidly obese patient may
also be sarcopenic, as BMI is not equivalent to muscle mass
and should be measured and analyzed separately.11
In our study, women with AAA were more likely than
men to have sarcopenia (32.0% versus 9.7%; p=0.005). The
literature suggests that sex does not play a significant role
in long-term survival following aneurysm repair; however,
sarcopenia was not assessed in these studies.12,13 Although,
in our study, the number of women was low (n=25), sarco-
penia may be helpful in assessing risk based on sex. Further
investigation using a larger population of women would
strengthen, or refute, the association.
Lim et al. provided a contemporary study to the EVAR 2
Trial by evaluating high-risk EVAR patients.14,15 Using a
Figure 1. Survival curves of patients with and without sarcopenia who underwent EVAR.
Table 3. Logistic regression results comparing all significant
demographic variables in patients (living vs deceased)
undergoing EVAR.
Odds ratio 95% Confidence interval
(Intercept) 0.31 (0.12, 0.72)
Age 76–85 0.97 (1.04, 3.58)
Hypertension 2.73 (1.17, 7.11)
Sarcopenia 3.17 (1.20, 9.54)
Sex 0.97 (0.38, 2.44)
at HIGH POINT UNIV on April 23, 2016vmj.sagepub.comDownloaded from
5. Hale et al. 5
backward stepwise logistic regression analysis, they identi-
fied five prognostic indicators for post-EVAR death; these
included: age, chronic kidney disease stages 4 and 5, con-
gestive heart failure, home oxygen use, and current cancer
therapy. Mortality at 4 years in their trial was 35%, in the
EVAR 2 it was 36%, and in our sarcopenic patients it was
28% (Figure 1).
Elderly patients (>80 years) are more frequently being
treated with EVAR than open repair since the perioperative
risk is lower.16,17 Our data showed that patients >76 years of
age were at an approximately twofold risk for mortality
(Table 3). However, when we modeled multiple variations
of age, no arrangement of age with sarcopenia was found to
significantly impact mortality. This conclusion appears to
confound logic, as both age and sarcopenia were deter-
mined to be independent risk factors (Table 3). However,
this unexpected outcome is likely due to the combination of
our small study population (only 12.5% of patients had sar-
copenia) and the low age difference (4.9 years) between
those with sarcopenia and without. A larger study would
likely resolve this Type II error and provide an effect strong
enough to reach statistical significance.
In this study, the median follow up until death or the end
of the study was 8.4 years, with an interquartile range of
5.3–11.7 years. Our short-term mortality rate (3.5% at >6
months), as well as our 5-year mortality rate (20.5%), were
comparable to the EVAR population reported in the EVAR
1 Trial (4.1% at 6 months, 20.8% at 4 years).18 To our
knowledge, this paper represents the longest follow up, in
terms of mortality, on AAA patients undergoing EVAR in
the literature (Table 4).
Limitations
This study has all the inherent limitations of a retrospective
study performed from a prospective database. The entire
AAA population was not studied as most open AAA
patients fall out of follow-up and do not frequently require
CT scans at follow-up. However, considering we were
focusing on the long-term mortality and not perioperative
events, those who returned for their annual office visits
comprised the study population.
Lost to follow up can be a significant limitation in any
retrospective study. We had previously reported on sur-
vival in EVAR patients who participated in a clinical trial
(mandated follow up) as compared to those treated outside
of a trial (real-world follow-up). The long-term mortality
(5 years) was remarkably consistent between these two
patient populations, with a mortality rate of 36–39%.19
This compares quite favorably to our overall study popula-
tion where we noted a 40% mortality rate at 5 years. This
suggests that a sensitivity analysis on those lost to follow-
up is unnecessary.
Women and minorities are underrepresented in this
study population, which makes the data less generalizable.
We did not measure the use of aspirin or statins in our
patients, which may alter long-term outcomes. Lastly,
acquiring archived CT scan images performed prior to 2004
was challenging; this represented a major limitation to
evaluating all EVAR patients in this study.
Conclusion
The presence of sarcopenia on a CT scan is an important,
patient-specific, risk factor for long-term mortality in AAA
patients treated with EVAR. Pending further study, these data
suggest that sarcopenia may aid in the pre-procedural long-
term survival assessment of patients undergoing EVAR.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
References
1. Arya S, Kim SI, Duwayri Y, et al. Frailty increases the risk of
30-day mortality, morbidity, and failure to rescue after elec-
tive abdominal aortic aneurysm repair independent of age
and comorbidities. J Vasc Surg 2015; 61: 324–331.
2. Matsubara Y, Matsumoto Aoyagi Y, et al. Sarcopenia is a
prognostic factor for overall survival in patients with critical
limb ischemia. J Vasc Surg 2015; 61: 945–950.
3. Masuda T, Shirabe K, Ikegami T, et al. Sarcopenia is a
prognostic factor in living donor liver transplantation. Liver
Transpl 2014; 20: 401–407.
4. Hardman DT, Fisher CM, Patel MI, et al. Ruptured abdomi-
nal aortic aneurysms: Who should be offered surgery? J Vasc
Surg 1996; 23: 123–129.
5. Steyerberg EW, Kievit J, de Mol Van Otterloo JC, et al.
Perioperative mortality of elective abdominal aortic aneu-
rysm surgery. A clinical prediction rule based on literature
and individual patient data. Arch Intern Med 1995; 155:
1998–2004.
6. Baas AF, Janssen KJ, Prinssen M, et al. The Glasgow
Aneurysm Score as a tool to predict 30-day and 2-year
mortality in the patients from the Dutch Randomized
Endovascular Aneurysm Management trial. J Vasc Surg
2008; 47: 277–281.
7. DeMartino RR, Goodney PP, Nolan BW, et al. Optimal selec-
tion of patients for elective abdominal aortic aneurysm repair
based on life expectancy. J Vasc Surg 2013; 58: 589–595.
8. Hickey GL, Grant SW, Murphy GJ, et al. Dynamic trends in
cardiac surgery: Why the logistic EuroSCORE is no longer
suitable for contemporary cardiac surgery and implications
Table 4. Patient deaths according to time since EVAR.
All patients Sarcopenia No sarcopenia
(n=200) (n=25) (n=175)
Mortality
All, n (%) 103 (51.5) 19 (76.0) 84 (48.0)
Time since procedure
0–6 months 7 (3.5) 0 (0) 7 (4.0)
6 months
– 5 years
41 (20.5) 10 (40.0) 31 (17.7)
5–10 years 39 (19.5) 5 (20.0) 34 (19.4)
10 years 16 (8.0) 4 (16.0) 12 (6.9)
at HIGH POINT UNIV on April 23, 2016vmj.sagepub.comDownloaded from
6. 6 Vascular Medicine
for future risk models. Eur J Cardiothorac Surg 2013; 43:
1146–1152.
9. Lee JS, He K, Harbaugh CM, et al. Michigan Analytic
Morphomics Group (MAMG). Frailty, core muscle size, and
mortality in patients undergoing open abdominal aortic aneu-
rysm repair. J Vasc Surg 2011; 53: 912–917.
10. Giles KA, Wyers MC, Pomposelli FB, et al. The impact of
body mass index on perioperative outcomes of open and
endovascular abdominal aortic aneurysm repair from the
National Surgical Quality Improvement Program, 2005–
2007. J Vasc Surg 2010; 52: 1471–1477.
11. Wannamethee SG, Atkins JL. Muscle loss and obesity: The
health implications of sarcopenia and sarcopenic obesity.
Proc Nutr Soc 2015; 27: 1–8.
12. Chung C, Tadros R, Torres M, et al. Evolution of gender-
related differences in outcomes from two decades of endo-
vascular aneurysm repair. J Vasc Surg 2015; 61: 843–852.
13. Gloviczki P, Huang Y, Oderich GS, et al. Clinical presenta-
tion, comorbidities, and age but not female gender predict
survival after endovascular repair of abdominal aortic aneu-
rysm. J Vasc Surg 2015; 61: 853–861.
14. Lim S, Halandras PM, Park T, et al. Outcomes of endovascu-
lar abdominal aortic aneurysm repair in high-risk patients.
J Vasc Surg 2015; 61: 862–868.
15. United Kingdom EVAR Trial Investigators,Greenhalgh
RM, Brown LC, Powell JT, et al. Endovascular repair of
aortic aneurysm in patients physically ineligible for open
repair. N Engl J Med 2010; 362: 1872–1880.
16. Schwarze ML, Shen Y, Hemmerich J, et al. Age-related
trends in utilization and outcome of open and endovascular
repair for abdominal aortic aneurysm in the United States,
2001–2006. J Vasc Surg 2009; 50: 722–729.
17. Schermerhorn ML, Buck DB, O’Malley AJ, et al. Long-term
outcomes of abdominal aortic aneurysm in the Medicare
population. N Engl J Med 2015; 373: 328–338.
18. United Kingdom EVAR Trial Investigators,Greenhalgh
RM, Brown LC, Powell JT, et al. Endovascular versus open
repair of abdominal aortic aneurysm. N Engl J Med 2010;
362: 1863–1871.
19. Jones WB, Taylor SM, Kalbaugh CA, et al. Lost to follow-
up: A potential under-appreciated limitation of endovascular
aneurysm repair. J Vasc Surg 2007; 46: 434–441.
at HIGH POINT UNIV on April 23, 2016vmj.sagepub.comDownloaded from
