J Thromb Thrombolysis DOI 10.1007/s11239-017-1508-y
Comparison of body mass index, waist circumference, and waistheight ratio in predicting functional outcome following ischemic stroke Kyusik Kang1 · Wong‑Woo Lee1 · Jung‑Ju Lee1 · Jong‑Moo Park1 · Ohyun Kwon1 · Byung Kun Kim1
© Springer Science+Business Media New York 2017
Abstract Although a positive association between body mass index (BMI) and stroke incidence has been reported, having a higher BMI is known to be advantageous in surviving and recovering from stroke. The association between adiposity and stroke incidence is more evident for measures of abdominal obesity than for general obesity. The aim of our study was to compare BMI, waist circumference, and waist-height ratio (WHR) as predictors of 3-month functional outcome in stroke patients. The BMI, waist circumference, and WHR of acute stroke patients were divided into sex-specific quartiles. A total of 605 female and 727 male patients were included. For BMI, male patients in the second quartile were more likely to have a favorable functional outcome compared with those in the lowest quartile (adjusted OR 1.64, 95% CI 1.02–2.62). For waist circumference (adjusted OR for top quartile vs. lowest quartile 1.79, 95% CI 1.14–2.82) and WHR (adjusted OR for second quartile vs. lowest quartile 1.99, 95% CI 1.22–3.25), male patients in the two top quartiles were more likely to have a favorable functional outcome compared with those in the respective lowest quartile. BMI and WHR showed similar relationships to a favorable functional outcome, with a favorable functional outcome occurring most often among male patients in the second quartiles. In women, however, obesity was not related to functional outcome. In conclusion, general obesity measured by BMI and abdominal obesity measured by WHR showed similar effects on the functional outcome after stroke in men. * Byung Kun Kim
[email protected] 1
Department of Neurology, Nowon Eulji Medical Center, Eulji University, 68 Hangeulbiseok‑ro, Nowon‑gu, Seoul 01830, Republic of Korea
Keywords Functional outcome · Ischemic stroke · Obesity · Prognosis
Introduction Mounting evidence indicates a graded positive association between obesity and stroke incidence independent of vascular risk factors [1, 2]. A meta-analysis of 45,235 Japanese patients aged 40–89 years found a positive relation of elevated body mass index (BMI) to both ischemic and hemorrhagic stroke in both men and women [2]. A study of 234,863 Korean men aged 40–64 years revealed an adjusted hazard of 11% for ischemic stroke for each 1-point increase in BMI [1]. However, there is a paradox with an elevated BMI and stroke, and that is, an elevated BMI, a risk factor for stroke, transforms into a protective agent against an adverse prognosis after an occurrence of a stroke [3–6]. The BMI is the most frequent measure of overall adiposity [7]. Although BMI is easy to calculate, it does not accurately address the distribution and proportion of body fat, muscle, and bone [3]. For example, athletes may have a high BMI because of increased muscle mass instead of increased fat mass. Waist circumference is closely associated with visceral fat mass and can be used as a surrogate indicator for abdominal obesity [8]. Because waist circumference is associated with body-frame size, waist-height ratio (WHR) is often used to control for body-frame size [7, 8]. Abdominal obesity is a stronger predictor of stroke risk than elevated BMI [9]. However, there are few studies that evaluated the impact of abdominal obesity on functional outcomes following ischemic stroke. The aim of the study was to compare BMI, waist circumference,
13
Vol.:(0123456789)
and WHR as predictors of 3-month functional outcome in acute ischemic stroke patients.
Methods Patients This was a retrospective study based on prospectively collected data. We collected the records of first-ever ischemic stroke patients with an onset less than 7 days before presentation at Nowon Eulji Medical Center from April 2008 to May 2015. Ischemic stroke was defined as suddenly developed focal neurological deficits attributable to occlusion of cerebral vessels that lasted ≥24 h and supported by brain images. An event that lasted <24 h but demonstrated evidence of acute brain infarction on diffusion-weighted magnetic resonance image was also regarded as a stroke. We selected first-ever ischemic stroke patients on whom we held data on anthropometric indices and 3-month modified Rankin Scale (mRS) score. The study protocol has been approved by the Institutional Review Board of the hospital (EMCIRB 17-12). Data collection Details of each patient’s medical history, risk factors for stroke, neurological examination and results of investigations were prospectively recorded and transferred to a computerised database [10]. All patients had their height, weight, and waist circumference measured on admission. Their waist circumference was measured at the level of the umbilicus using a measuring tape. The BMI was calculated as weight in kilograms divided by height in meters squared. WHR was calculated by dividing the waist circumference by the height [7]. The BMI, waist circumference, and WHR were divided into sex-specific quartiles (Table 1). These quartile versions were used as predictor variables in modeling. Baseline stroke severity was categorized as mild [National Institutes of Health Stroke Scale (NIHSS) score 0 to 7] and moderate to severe (NIHSS score ≥8) [4]. All patients were assessed at 3 months using the modified Rankin Scale (mRS) [11]. For patients who died, mRS was defined as six. Functional outcome was dichotomized as unfavorable versus favorable based on the mRS at 3 months. Favorable functional outcome was defined as mRS <2. Anemia was defined as hemoglobin <135 g/L in men and <120 g/L in women.
13
K. Kang et al. Table 1 Quartiles of anthropometric measures Q4 Male Body mass index, kg/m2 Quartile cutoffs <21.6 No. of cases 172 Waist circumference, cm Quartile cutoffs <80.0 No. of cases 178 Waist to height ratio Quartile cutoffs <0.471 No. of cases 166 Female Body mass index, kg/m2 Quartile cutoffs <21.2 No. of cases 150 Waist circumference, cm Quartile cutoffs <75.0 No. of cases 129 Waist to height ratio Quartile cutoffs <0.489 No. of cases 151
Q3
Q2
Q1
21.6–23.4 175
23.5–25.5 196
≥25.6 184
80.0–83.9 147
84 0.0–89.9 182
≥90 220
0.471–0.502 197
0.503–0.540 176
≥0.541 188
21.2–23.5 149
23.6–26.1 150
≥26.2 156
75.0–81.9 170
82 0.0–88.9 154
≥89 152
0.489–0.530 150
0.531–0.579 151
≥0.580 153
Statistical methods All analyses were stratified by sex. The relationship between categorical variables was evaluated using the χ2 test or the Fisher exact test. The Student t test and the Mann–Whitney test were applied to compare continuous variables versus binary groups. Multivariable analyses were conducted using binary logistic regression for dichotomous measures. To evaluate the role of possible confounding factors, each model included one of the sets of anthropometric index quartiles plus other covariates associated with 3-month functional outcome (p < 0.2 after bivariate analysis). The included covariates are listed in the legend of Fig. 1. BMI and waist circumference were analyzed separately because of their colinearity. We reported odds ratios (OR) with 95% confidence intervals (CI). If a patient was missing information on educational level (n = 16), we assigned them to the median value. The level of significance was set at p < 0.05 for all statistical analyses. The statistical analysis was performed with PASW Statistics version 18 (IBM Corporation, New York, United States).
Results Of 1586 ischemic stroke patients with an onset less than 7 days before presentation in the database, 183 were missing anthropometric measurements, and 71 were lost to
Comparison of body mass index, waist circumference, and waist-height ratio in predicting…
A Male
B Male
C Male
Q1
Q1
Q1
Q2
Q2
Q2
Q3
Q3
Q3
Q4
Q4
Q4
Odds ratio
D Female
Odds ratio
E Female
Odds ratio
F Female
Q1
Q1
Q1
Q2
Q2
Q2
Q3
Q3
Q3
Q4
Q4
Q4
Odds ratio
Odds ratio
Odds ratio
Fig. 1 Association between admission anthropometric index quartiles versus a 3-month functional outcome. The figure shows OR and 95% confidence limits of a favorable 3-month functional outcome in male (a–c) and female (d–f) patients according to quartiles of BMI (a, d), waist circumference (b, e), and WHR (c, f). Analyses in male patients were adjusted for age, educational level, hypertension, diabe-
tes, atrial fibrillation, anemia, systolic blood pressure, stroke subtype, and stroke severity (a–c). Analyses in female patients were adjusted for age, educational level, hypertension, diabetes, atrial fibrillation, coronary artery disease, anemia, stroke subtype, stroke severity, and recanalization therapy (d–f)
follow-up. Thus, a total of 1332 ischemic stroke patients were included in this study. The sex-specific characteristics of 605 female and 727 male participants are shown in Table 2. Male patients with an unfavorable functional outcome were significantly older, more likely to have a history of atrial fibrillation and diabetes, and more likely to have anemia and moderate-to-severe stroke than those with a favorable functional outcome in bivariate analysis (Table 3). Female patients with an unfavorable functional outcome were significantly older, more likely to have a history of atrial fibrillation and coronary artery disease, more likely to have anemia and moderate-to-severe stroke, and had lower level of education than those with a favorable functional outcome in bivariate analysis (Table 3). Next, we used multivariable analyses to determine the effects of anthropometric indices on 3-month functional outcome. For BMI, male patients in the second quartile were more likely to have a favorable functional outcome compared with those in the lowest quartile (adjusted OR 1.64, 95% CI 1.02–2.62) (Fig. 1a). For waist circumference (adjusted OR for top quartile vs. lowest quartile 1.79, 95% CI 1.14–2.82) (Fig. 1b) and WHR (adjusted OR for
second quartile vs. lowest quartile 1.99, 95% CI 1.22–3.25) (Fig. 1c), male patients in the two top quartiles were more likely to have a favorable functional outcome compared with those in the lowest quartile. BMI and WHR showed similar relationships to a favorable functional outcome, with a favorable functional outcome occurring most often among male patients in the second quartiles (Fig. 1a, c). However, there were no significant differences with regard to the functional outcome among quartiles of any anthropometric indices in female (Fig. 1d–f).
Discussion In our study, BMI and WHR showed similar relationships to functional outcome in men. A retrospective study conducted in Korea showed that a favorable functional outcome occurred most often among stroke patients with a BMI of 24.6–26.2 kg/m2 [4]; a prospective registry study in China demonstrated that a favorable functional outcome occurred most often among stroke patients with a BMI of 23–27.4 kg/m2 [6]. These findings are in accordance with that observed in our study, which revealed that male
13
K. Kang et al.
Table 2 Baseline characteristics of stroke patients by sex Variables
Male (n = 727) Female (n = 605)
Age, years Educational level, years 0–3 4–6 7–9 10–12 ≥13 Hypertension Diabetes Hyperlipidemia Atrial fibrillation Previous coronary artery disease Anemia Current smoker Systolic blood pressure, mmHg Diastolic blood pressure, mmHg Stroke subtype Small vessel occlusion Large artery atherosclerosis Cardioembolism Other determined or undetermined Recanalization therapy Height, cm Weight, kg Body mass index, kg/m2 Abdominal circumference, cm Waist to height ratio NIHSS score <8 ≥8 Favorable 3-month functional activity
64 ± 13
72 ± 12
48 (6.6)a 168 (23.3)a 160 (22.2)a 206 (28.5)a 140 (19.4)a 483 (66.4) 304 (41.8) 159 (21.9) 103 (14.2) 57 (7.8) 165 (22.7) 374 (51.4) 149 ± 25 85 ± 16
191 (32.2)b 217 (35.5)b 77 (13.0)b 87 (14.6)b 22 (3.7)b 457 (75.5) 265 (43.8) 172 (28.4) 142 (23.5) 48 (7.9) 152 (25.1) 55 (9.1) 150 ± 28 83 ± 15
106 (14.6) 329 (45.3) 103 (14.2) 189 (26)
76 (12.6) 235 (38.8) 118 (19.5) 176 (29.1)
88 (12.1) 167.4 ± 5.8 66.3 ± 10.1 23.6 ± 3.1 84.6 ± 8.6 0.506 ± 0.051 3 (1, 6) 585 (80.5) 142 (19.5) 426 (58.6)
75 (12.4) 153.6 ± 6.1 56.0 ± 10.3 23.7 ± 3.9 82.1 ± 10.3 0.535 ± 0.068 4 (2, 8) 444 (73.4) 161 (26.6) 241 (39.8)
Data are mean ± SD, median (interquartile range) or number (%) NIHSS National Institutes of Health Stroke Scale
a
The sample size is 722 because of missing data
b
The sample size is 594 because of missing data
patients with a BMI of 23.5–25.5 kg/m2 had best functional outcomes 3 months after stroke. A prospective registry study in Mexico showed that a U-shaped relationship existed between WHR and mortality with elevated mortality among the leanest and the heaviest of the patients with ischemic stroke or transient ischemic attack [12]. Although the mechanism of mortality and that of functional recovery are different, it is interesting that our study showed similar pattern. For waist circumference, male patients in the two top quartiles were more likely to have a favorable functional
13
outcome compared with those in the lowest quartile in our study, which means that male patients with a waist circumference of ≥84 cm had better functional outcome compared with those with a waist circumference <80 cm. Interestingly, men presenting with a waist circumference of ≥85 cm are categorized as having abdominal obesity in Japan [13] and China [14]. There are few studies that evaluated the impact of abdominal obesity on functional outcomes following ischemic stroke. Although BMI and WHR showed similar relationships to a favorable functional outcome, the OR for a favorable functional outcome was higher for WHR (adjusted OR for second quartile vs. lowest quartile 1.99) than for BMI (adjusted OR for second quartile vs. lowest quartile 1.64). There can be two possible explanations for this difference. First, BMI does not provide any information on the regional distribution of fat mass. Abdominal fat is more metabolically active than subcutaneous fat and secrets a lot of free fatty acids and cytokines [15, 16]. Second, BMI may not be an accurate measure of fat mass in the elderly because lean body mass decline with aging [8]. Even though the BMI remains stable or may even decrease, fat mass can increase with aging [8]. Because strokes occur predominantly in elderly individuals, this may contribute to the weaker association between BMI and functional outcome. There are few satisfactory explanations for the differential effect of sex on the relationship between obesity and short-term outcomes among stroke survivors. Like our study, results from the Virtual International Stroke Trials Archive database showed that functional outcome benefits associated with general obesity measured by BMI were restricted to men [5]. Abdominal obesity is also termed android obesity and is observed more commonly in men; subcutaneous obesity is also referred to as gynoid obesity and is more commonly seen in women [16]. The association between sex and the fat distribution might explain the sex differential effect of obesity on functional outcome after stroke. Our study shows that, for men, being overweight or obese can be advantageous in recovering from ischemic stroke. Several hypotheses can be proposed to explain the association between a higher level of adiposity and a better functional outcome. First, patients with the lowest BMI or waist circumference quartile may have lost weight because of medical or psychiatric illness, neglect, malnutrition, or lower socioeconomic status [3]. Because anemia is related to poor nutritional status and a poor outcome after stroke [17, 18] and education is a surrogate for socioeconomic status [19], we adjusted our analyses for anemia and education. However, adjustment for anemia and education did not alter our results. Second, the acute stroke patients exist in a hypercatabolic state with frequent infections, high fevers, sympathetic activation, and insulin sensitivity dysfunction
Comparison of body mass index, waist circumference, and waist-height ratio in predicting… Table 3 Bivariate statistics for 3-month functional outcome by sex Variable
Age, years Educational level, years 0–3 4–6 7–9 10–12 ≥13 Hypertension Diabetes Hyperlipidemia Atrial fibrillation Previous coronary artery disease Anemia Current smoker Systolic blood pressure, mmHg Diastolic blood pressure, mmHg Stroke subtype Small vessel occlusion Large artery atherosclerosis Cardioembolism Other determined or undetermined Recanalization therapy Body mass index, kg/m2 Q1 Q2 Q3 Q4 Abdominal circumference, cm Q1 Q2 Q3 Q4 Waist to height ratio Q1 Q2 Q3 Q4 NIHSS score <8 ≥8
Men
Women Favorable functional outcome (n = 241)
P
67 ± 12
23 (5.4) 94 (22.1) 91 (21.4) 122 (28.6) 96 (22.5) 273 (64.1) 164 (38.5) 94 (22.1) 44 (10.3) 33 (7.7) 77 (18.1) 214 (50.2) 147 ± 24 85 ± 15
<0.001 74 ± 11 0.05 141 (38.7) 126 (34.6) 47 (12.9) 42 (11.5) 8 (2.2) 0.11 285 (78.3) 0.03 170 (46.7) 0.88 108 (29.7) <0.001 106 (29.1) 0.91 37 (10.2) <0.001 107 (29.4) 0.44 221 (91.7) 0.13 151 ± 29 0.22 83 ± 16
<0.001 <0.001
81 (19) 193 (45.3) 45 (10.6) 107 (25.1) 48 (11.3) 24.0 ± 2.9 80 (18.8) 107 (25.1) 126 (29.6) 113 (26.5) 85.8 ± 8.2 83 (19.5) 85 (20) 115 (27) 143 (33.6) 0.513 ± 0.048 79 (18.5) 107 (25.1) 119 (27.9) 121 (28.4) 2 (1, 4) 393 (92.3) 33 (7.7)
<0.001 33 (9.1) 141 (38.7) 84 (23.1) 106 (29.1) 0.41 57 (15.7) <0.001 23.5 ± 4.2 0.003 108 (29.7) 78 (21.4) 89 (24.5) 89 (24.5) <0.001 82 ± 11 0.001 81 (22.3) 102 (28.0) 85 (23.4) 96 (26.4) <0.001 0.535 ± 0.072 <0.001 98 (26.9) 81 (22.3) 90 (24.7) 95 (26.1) <0.001 6 (3, 12) <0.001 217 (59.6) 107 (40.4)
43 (17.8) 94 (39) 34 (14.1) 70 (29) 18 (7.5) 24.1 ± 3.5 42 (17.4) 71 (29.5) 61 (25.3) 67 (27.8) 82.3 ± 8.9 48 (19.9) 68 (28.2) 69 (28.6) 56 (23.2) 0.536 ± 0.059 53 (22) 69 (28.6) 61 (25.3) 58 (24.1) 2 (0, 4) 227 (94.2) 8 (5.8)
Unfavorable functional outcome (n = 301)
Favorable functional outcome (n = 426)
P
68 ± 12
62 ± 13
25 (8.3) 74 (24.6) 74 (24.6) 84 (27.9) 44 (14.6) 210 (69.8) 140 (46.5) 65 (21.6) 59 (19.6) 24 (8.0) 88 (29.2) 160 (53.2) 150 ± 27 84 ± 17 25 (8.3) 136 (45.2) 58 (19.3) 82 (27.2) 40 (13.3) 23.1 ± 3.3 92 (30.6) 68 (22.6) 70 (23.3) 71 (23.6) 83.0 ± 9.0 95 (31.6) 62 (20.6) 67 (22.3) 77 (25.6) 0.496 ± 0.052 87 (28.9) 90 (29.9) 57 (18.9) 67 (22.3) 6 (3, 10.5) 192 (63.8) 109 (36.2)
Unfavorable functional outcome (n = 364)
50 (20.7) 91 (37.8) 41 (17) 45 (18.7) 14 (5.8) 172 (71.4) 95 (39.4) 64 (26.6) 36 (14.9) 11 (4.6) 45 (18.7) 20 (8.3) 148 ± 27 83 ± 14
0.05 0.08 0.41 <0.001 0.01 0.003 0.58 0.2 0.89 0.002
0.003 0.04 0.004
0.64 0.47
0.84 0.26
<0.001 <0.001
Data are mean ± SD, median (interquartile range) or number (%) NIHSS National Institutes of Health Stroke Scale
resulting in weight loss [3]. A higher level of adiposity does not mean good nutrition, but may be protective against the hypercatabolic state following acute stroke leading to a
better functional outcome [3, 20]. Third, a positive relationship between leptin concentrations and sepsis survival has been reported suggesting that leptin could be protective in
13
patients with critical illness [21]. Last, abdominal obesity is positively correlated with plasma concentrations of triglycerides and of total cholesterol. Interestingly, higher concentrations of triglycerides and total cholesterol have been linked to better functional outcomes following ischemic stroke [20, 22, 23]. In conclusion, general obesity measured by BMI and abdominal obesity measured by WHR showed similar effects on the functional outcome after stroke. However, the functional outcome benefit associated with obesity was restricted to men, whereas the benefit was not apparent in women. Funding This study was supported by a grant of the Korea Health 21 R&D Project, Ministry of Health, Welfare and Family Affairs, Republic of Korea (HI10C2020). Compliance with ethical standards Conflict of interest The authors declare that they have no conflict of interest. Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent For this type of study formal consent is not required.
References 1. Song YM, Sung J, Davey Smith G, Ebrahim S (2004) Body mass index and ischemic and hemorrhagic stroke: a prospective study in Korean men. Stroke 35:831–836. doi:10.1161/01. STR.0000119386.22691.1C 2. Yatsuya H, Toyoshima H, Yamagishi K, Tamakoshi K, Taguri M, Harada A, Ohashi Y, Kita Y, Naito Y, Yamada M, Tanabe N, Iso H, Ueshima H (2010) Body mass index and risk of stroke and myocardial infarction in a relatively lean population: meta-analysis of 16 Japanese cohorts using individual data. Circ Cardiovasc Qual Outcomes 3:498–505. doi:10.1161/ CIRCOUTCOMES.109.908517 3. Katsnelson M, Rundek T (2011) Obesity paradox and stroke: noticing the (fat) man behind the curtain. Stroke 42:3331–3332. doi:10.1161/STROKEAHA.111.632471 4. Kim Y, Kim CK, Jung S, Yoon BW, Lee SH (2015) Obesitystroke paradox and initial neurological severity. J Neurol Neurosurg Psychiatry 86:743–747. doi:10.1136/jnnp-2014-308664 5. Saini M, Saqqur M, Shuaib A (2014) Body mass index and acute ischemic stroke outcomes. Int J Stroke 9:618–623. doi:10.1111/ ijs.12168 6. Zhao L, Du W, Zhao X, Liu L, Wang C, Wang Y, Wang A, Liu G, Xu Y (2014) Favorable functional recovery in overweight ischemic stroke survivors: findings from the China National Stroke Registry. J Stroke Cerebrovasc Dis 23:e201–e206. doi:10.1016/j.jstrokecerebrovasdis.2013.10.002
13
K. Kang et al. 7. Page JH, Rexrode KM, Hu F, Albert CM, Chae CU, Manson JE (2009) Waist-height ratio as a predictor of coronary heart disease among women. Epidemiology 20:361–366. doi:10.1097/ EDE.0b013e31819f38f1 8. Walker SP, Rimm EB, Ascherio A, Kawachi I, Stampfer MJ, Willett WC (1996) Body size and fat distribution as predictors of stroke among US men. Am J Epidemiol 144:1143–1150. doi:10.1093/oxfordjournals.aje.a008892 9. Bodenant M, Kuulasmaa K, Wagner A, Kee F, Palmieri L, Ferrario MM, Montaye M, Amouyel P, Dallongeville J (2011) Measures of abdominal adiposity and the risk of stroke: the MOnica Risk, Genetics, Archiving and Monograph (MORGAM) study. Stroke 42:2872–2877. doi:10.1161/ STROKEAHA.111.614099 10. Kim BJ, Han MK, Park TH, Park SS, Lee KB, Lee BC, Yu KH, Cha JK, Kim DH, Lee J, Lee SJ, Ko Y, Park JM, Kang K, Cho YJ, Hong KS, Cho KH, Kim JT, Kim DE, Lee JS, Jang MS, Broderick JP, Yoon BW, Bae HJ (2014) Current status of acute stroke management in Korea: a report on a multicenter, comprehensive acute stroke registry. Int J Stroke 9:514–518. doi:10.1111/ijs.12199 11. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J (1988) Interobserver agreement for the assessment of handicap in stroke patients. Stroke 19:604–607. doi:10.1161/01. STR.19.5.604 12. Chiquete E, Ruiz-Sandoval JL, Murillo-Bonilla L, Leon-Jimenez C, Ruiz-Madrigal B, Martinez-Lopez E, Roman S, Panduro A, Ramos A, Cantu-Brito C (2013) Central adiposity and mortality after first-ever acute ischemic stroke. Eur Neurol 70:117–123. doi:10.1159/000350762 13. Examination Committee of Criteria for ‘Obesity Disease’ in Japan, Japan Society for the Study of Obesity (2002) New criteria for ‘obesity disease’ in Japan. Circ J 66:987–992. doi:10.1253/circj.66.987 14. Zhou BF (2002) Predictive values of body mass index and waist circumference for risk factors of certain related diseases in Chinese adults-study on optimal cut-off points of body mass index and waist circumference in Chinese adults. Biomed Environ Sci 15:83–96 15. Einstein FH, Atzmon G, Yang XM, Ma XH, Rincon M, Rudin E, Muzumdar R, Barzilai N (2005) Differential responses of visceral and subcutaneous fat depots to nutrients. Diabetes 54:672– 678. doi:10.2337/diabetes.54.3.672 16. Boyett D, Magnuson T, Schweitzer M (2016) Metabolic changes following bariatric surgery. In: Cameron JL, Cameron AM (eds) Current surgical therapy. 12th edn. Elsevier, Philadelphia, PA, pp 802–806 17. Ryu WS, Lee SH, Kim CK, Kim BJ, Yoon BW (2010) Effects of low serum triglyceride on stroke mortality: a prospective follow-up study. Atherosclerosis 212:299–304. doi:10.1016/j. atherosclerosis.2010.05.006 18. Tanne D, Molshatzki N, Merzeliak O, Tsabari R, Toashi M, Schwammenthal Y (2010) Anemia status, hemoglobin concentration and outcome after acute stroke: a cohort study. BMC Neurol 10:22. doi:10.1186/1471-2377-10-22 19. Petitti DB, Freedman DA (2005) Invited commentary: how far can epidemiologists get with statistical adjustment? Am J Epidemiol 162:415–418 (discussion 419–420). doi:10.1093/aje/ kwi224 20. Vauthey C, de Freitas GR, van Melle G, Devuyst G, Bogousslavsky J (2000) Better outcome after stroke with higher serum cholesterol levels. Neurology 54:1944–1949. doi:10.1212/ WNL.54.10.1944 21. Bornstein SR, Licinio J, Tauchnitz R, Engelmann L, Negrao AB, Gold P, Chrousos GP (1998) Plasma leptin levels are increased in survivors of acute sepsis: associated loss of diurnal rhythm, in
Comparison of body mass index, waist circumference, and waist-height ratio in predicting… cortisol and leptin secretion. J Clin Endocrinol Metab 83:280– 283. doi:10.1210/jcem.83.1.4610 22. Weir CJ, Sattar N, Walters MR, Lees KR (2003) Low triglyceride, not low cholesterol concentration, independently predicts poor outcome following acute stroke. Cerebrovasc Dis 16:76–82. doi:10.1159/000070119
23. Terry RB, Wood PD, Haskell WL, Stefanick ML, Krauss RM (1989) Regional adiposity patterns in relation to lipids, lipoprotein cholesterol, and lipoprotein subfraction mass in men. J Clin Endocrinol Metab 68:191–199. doi:10.1210/jcem-68-1-191
13