Secondary Logo

Journal Logo

Featured Articles: Original Clinical Research Report

Association Between High Body Mass Index and Mortality Following Myocardial Injury After Noncardiac Surgery

Lee, Seung-Hwa MD*; Yang, Kwangmo MD; Park, Jungchan MD; Lee, Jong Hwan MD, PhD; Min, Jeong Jin MD, PhD; Kwon, Ji-hye MD; Yeo, Junghyun MD; Kim, Jihoon MD*; Hyeon, Cheol Won MD*; Choi, Jin-ho MD, PhD*,§; Lee, Sang-Chol MD, PhD*; Gwon, Hyeon-Cheol MD, PhD*; Kim, Kyunga PhD∥,¶; Ahn, Joonghyun MB§; Lee, Sangmin Maria MD, PhD

Author Information
doi: 10.1213/ANE.0000000000005303

Abstract

See Article, p 957

KEY POINTS

  • Question: Does overweight, defined by body mass index, have a clinical impact among patients diagnosed with myocardial injury after noncardiac surgery?
  • Findings: Overweight was associated with lower mortality following myocardial injury after noncardiac surgery.
  • Meaning: Preoperative body mass index may need to be taken into account when managing patients with myocardial injury after noncardiac surgery.

Obesity, a major health issue in the modern era, has been implicated as one of the risk factors for the development of several cardiovascular conditions such as hypertension, heart failure, and ischemic heart disease.1 Although mortality of the general population is lowest at a body mass index (BMI) within the normal range,2 numerous studies have reported an “obesity paradox,” where obese patients demonstrate better clinical outcomes in various clinical settings.1,3,4 Particularly, obese patients with established ischemic heart disease, such as stable angina or acute coronary syndrome, showed lower mortality than normal or underweight patients despite obesity itself being one of the risk factors for the development of these diseases.5–10

In the fourth universal definition of myocardial infarction, any cardiac troponin (cTn) elevation above the 99th-percentile upper reference is defined as myocardial injury regardless of ischemic symptoms,11 and recent studies have suggested that myocardial injury after noncardiac surgery (MINS) is significantly associated with cardiovascular events and mortality up to the first 2 years after surgery, with a 5-fold increase from the baseline risk.12–16 The diagnostic criteria for MINS include the occurrence of postoperative cTn concentrations above the 99th-percentile upper reference limit within 30 days postsurgery resulting from myocardial ischemia without the requirement of an ischemic feature.17 The high prevalence of MINS, reported to be above 20%, makes it the leading postoperative condition correlated with mortality.18,19 In previous studies, obese patients with myocardial infarction have shown better clinical outcomes,6–8 but this phenomenon has never been evaluated in the context of MINS. In this study, we enrolled adult MINS patients, defined as those with peak postoperative cTn I above the 99th-percentile upper reference limit of 40 ng·L−1 within 30 days using TnI-Ultra immunoassay,11,17 and hypothesized that high BMI is associated with mortality also following MINS. The primary end point was all-cause death during 1 year of follow-up, while the mortality during 30 days was also compared.

METHODS

This study was approved by our Institutional Review Board (SMC 2019-08-048), who waived the need to gather individual written informed consent before access to the registry since the entire dataset was initially extracted in deidentified form. The cohort was registered before patient enrollment at https://cris.nih.go.kr (KCT0004244).

Study Population and Data Curation

This was an observational study conducted using data from the Samsung Medical Center Troponin in Noncardiac Operation (SMC-TINCO) registry, which is a large, single-center, deidentified cohort consisting of 43,019 consecutive patients. Eligible patients included those assessed from January 2010 to June 2019 who had at least 1 measured cTn I during the preoperative evaluation period or within 30 days after noncardiac surgery at Samsung Medical Center, Seoul, Korea. From this registry, adult patients with postoperative cTn data were identified. To select patients with MINS, we initially identified patients with peak postoperative cTn elevation above the 99th-percentile upper reference limit within 30 days after surgery. Those with evidence of a nonischemic etiology such as sepsis, pulmonary embolus, atrial fibrillation, cardioversion, or chronic elevation were excluded.11,12,17 the study population was stratified into normal, low BMI, and high BMI groups according to preoperative BMI.

All data from the SMC-TINCO registry were extracted using the “Clinical Data Warehouse Darwin-C” of Samsung Medical Center, which is an electronic system built for investigators to search and retrieve deidentified medical records from the institutional electronic archive system. Extracted preoperative evaluation sheets and discharge notes were manually reviewed by trained investigators to organize baseline characteristics and clinical events during the hospital stay. For mortality statistics at institutions other than ours, data in this system are consistently validated with and updated according to the National Population Registry of the Korea National Statistical Office using a unique personal identification number.

Definitions

High BMI was defined as a preoperative BMI ≥25 kg·m−2, and a BMI of <18.5 kg·m−2 was defined as low BMI. In the high BMI group, patients with BMI ≥30 kg·m−2 were further divided as obese patients, according to the Centers for Disease Control and Prevention (CDC) guidelines. Medical history and underlying disease of patients were organized based on the International Classification of Diseases code and by manually reviewing preoperative evaluation sheet. Charlson comorbidity index was calculated using an updated method of weighting.20

Study End points

The primary study end point was all-cause mortality during the first year. As a secondary end point, all-cause mortality within 30 days was assessed. Mortalities during 1 year and 30 days after surgery were classified as cardiovascular or noncardiovascular in nature, with cardiovascular mortality defined as death due to myocardial infarction, cardiac arrhythmia, heart failure, stroke, or vascular causes. All deaths without an undisputed noncardiovascular cause were considered cardiovascular deaths.21

Perioperative Management and cTn I Measurement

Perioperative management followed institutional protocols based on current guidelines. Routine perioperative cTn I measurement was recommended for moderate- or high-risk surgery or for patients with at least one of the major cardiovascular risk factors such as a history of ischemic heart disease, heart failure, stroke including transient ischemic attack, diabetes mellitus on insulin therapy, or chronic kidney disease based on current guidelines.22 In patients with minor risk factors, cTn was measured at the discretion of the attending clinician, considering old age or recently suspected symptoms of ischemic disease. An automated analyzer (Advia Centaur XP, Siemens Healthcare Diagnostics, Erlangen, Germany) with TnI-Ultra immunoassay was used. The lowest limit of detection was 6 ng·L−1, and 40 ng·L−1 of the 99th-percentile upper reference limit was provided by the manufacturer.

Statistical Analysis

The data are presented as numbers with incidence for categorical variables and means ± standard deviations (SD) or medians with interquartile ranges (IQRs) for continuous variables. The baseline characteristics were compared using analysis of variance or Kruskal-Wallis test. To evaluate the hypothesis that obesity, according to BMI, is associated with mortality, we compared normal, low BMI, and high BMI groups on mortality in a pairwise fashion. As stated before, the primary end point was all-cause mortality during 1 year, while 30-day mortality was also compared.

To compare mortalities during 1 year and 30 days, we conducted rigorous adjustments using weighted Cox proportional-hazards regression models with an inverse probability of treatment weighting (IPTW) to maximize the study power by balancing variables between the 3 groups. The inverse probability weights were defined as the reciprocal of propensity scores that were calculated via the generalized boosted model for estimating the average treatment effects. The generalized boosted regression was used to estimate the probability of being in each of the 3 study groups instead of a multinomial logistic regression model.23 Variables retained in the adjustment included age, male, days to peak cTn, smoking alcohol, cardiac comorbidities, such as hypertension, coronary artery disease, arrhythmia, and valve disease, Charlson comorbidity index, preoperative medications and management, operation type, and operative variables, such as general anesthesia, emergency operation, operation duration, intraoperative use of inotropics, and red blood cell transfusion. We maintained the α level at .05 for the primary outcome by adopting a post hoc Bonferroni’s correction, with a significance criterion of 0.0167 (0.05/3) when comparing the 3 groups, and reported the risks of mortality as hazard ratio (HR) with 98.3% confidence intervals (CIs). Kaplan-Meier estimates were used to construct survival curves of the 3 groups, which were compared with the log-rank test. We also constructed smoothing HR plots according to BMI. For sensitivity analysis, we assumed unmeasured confounders with the incidence of 40% and estimated the change of HR and CI according to the associations of unmeasured confounders with BMI and mortality.24

Before the formal analysis, we assessed the statistical power of this study in which HR was tested to compare between each of 3 pairs of BMI groups based on the Cox regression.25 An anticipated event (ie, all-cause mortality during 1 year) rate was 0.2 and a significance level of 0.05 was considered with Bonferroni’s multiplicity correction. We assumed a coefficient of multiple determination (R2) of 0–0.5 on the other covariates in the Cox regression. The expected power ranged 0.71–0.73, 0.74–0.77, and >0.99 when the HR was 1.4, 0.7, and 0.6 between the normal and low BMI groups, between the normal and high BMI groups, and between the low and high BMI groups, respectively. All statistical analyses were performed with R version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria), and twang’ package in R programming.23

RESULTS

Patient Characteristics

Table 1. - Baseline Characteristics According to Preoperative Body Mass Index
Normal (n = 3246) Low BMI (n = 425) High BMI (n = 1962) Before IPTW After IPTW
P ASD P ASD
BMIa 22.17 (±1.68) 16.92 (±1.31) 27.53 (±2.50)
Agea 65.4 (±13.98) 68.9 (±16.01) 64.4 (±13.14) <.001 0.204 .68 0.031
Peak cTn I levela 2560 (±18,404) 3058 (±24,996) 3034 (±23,249)
Days to peak cTn I levelb 1.22 (0.43–3.58) 1.35 (0.50–3.98) 1.27 (0.44–3.40) .61 0.022 .92 0.017
Male 1861 (57.3) 236 (55.5) 1271 (64.8) <.001 0.126 .69 0.026
Smoking 270 (8.3) 37 (8.7) 199 (10.1) .08 0.042 .85 0.018
Alcohol 416 (12.8) 40 (9.4) 361 (18.4) <.001 0.175 .67 0.033
Cardiac comorbidity
 Hypertension 2077 (64.0) 260 (61.2) 1402 (71.5) <.001 0.146 .20 0.065
 Coronary artery disease 712 (21.9) 88 (20.7) 511 (26.0) .001 0.84 .66 0.029
 Arrhythmia 316 (9.7) 48 (11.3) 205 (10.4) .50 0.34 .54 0.033
 Valve disease 58 (1.8) 15 (3.5) 42 (2.1) .05 0.73 .72 0.016
Charlson comorbidity index 2.40 (±1.97) 2.17 (±1.86) 2.37 (±1.95) .08 0.79 .91 0.01
 Heart failure 114 (3.5) 23 (5.4) 69 (3.5)
 Dementia 128 (3.9) 24 (5.6) 61 (3.1)
 Chronic pulmonary disease 186 (5.7) 40 (9.4) 120 (6.1)
 Rheumatic disease 30 (0.9) 9 (2.1) 10 (0.5)
 Mild liver disease 348 (10.7) 34 (8.0) 238 (12.1)
 Diabetes with complication 1473 (45.4) 191 (44.9) 909 (46.3)
 Hemiplegia 19 (0.6) 7 (1.6) 15 (0.8)
 Renal disease 452 (13.9) 69 (16.2) 247 (12.6)
 Any malignancy 1761 (54.3) 187 (44.0) 1054 (53.7)
 Moderate to severe liver disease 61 (1.9) 4 (0.9) 32 (1.6)
 Metastatic solid tumor 109 (3.4) 8 (1.9) 61 (3.1)
 Human immunodeficiency virus 1 (0.0) 0 0
Preoperative medication
 Beta blocker 934 (28.8) 111 (26.1) 632 (32.2) .01 0.09 .96 0.026
 Calcium channel blocker 1099 (33.9) 141 (33.2) 787 (40.1) <.001 0.096 .66 0.023
 RAAS inhibitor 1215 (37.4) 140 (32.9) 915 (46.6) <.001 0.188 .30 0.058
 Statin 1030 (31.7) 97 (22.8) 805 (41.0) <.001 0.264 .27 0.061
 Antiplatelet agent 1213 (37.4) 164 (38.6) 819 (41.7) .01 0.060 .47 0.043
 Direct oral anticoagulant 51 (1.6) 6 (1.4) 47 (2.4) .08 0.048 .70 0.027
 Warfarin 229 (7.1) 38 (8.9) 110 (5.6) .02 0.086 .19 0.068
Preoperative management
 Intensive care unit 331 (10.2) 76 (17.9) 202 (10.3) <.001 0.148 .31 0.049
 ECMO 0 (0.0) 0 (0.0) 1 (0.1) .39 0.021 .53 0.019
 CRRT 27 (0.8) 3 (0.7) 32 (1.6) .02 0.058 .60 0.032
 Ventilator 68 (2.1) 14 (3.3) 54 (2.8) .15 0.05 .73 0.017
Data are presented as n (%). ASD was defined as absolute difference in means divided by pooled standard deviation. ASD is the average of pairwise absolute standardized mean difference across the 3 groups, and ASD <0.1 was deemed to suggest a successful balance between the 3 groups.
Abbreviations: ASD, absolute standardized mean difference; BMI, body mass index; CRRT, continuous renal replacement therapy; cTn, cardiac troponin; ECMO, extracorporeal membrane oxygenation; IPTW, inverse probability of treatment weighting; RAAS, renin–angiotensin–aldosterone system.
aMean (±standard deviation).
bMedian (interquartile range).

Table 2. - Operative Variables According to Preoperative Body Mass Index
Normal (n = 3246) Low BMI (n = 425) High BMI (n = 1962) Before IPTW After IPTW
P ASD P ASD
Operation type <.001 36.2 .03 22.3
 Vascular 342 (10.5) 45 (10.6) 187 (9.5)
 Orthopedic 428 (13.2) 110 (25.9) 303 (15.4)
 Neuro 425 (13.1) 22 (5.2) 240 (12.2)
 Breast or endo 68 (2.1) 10 (2.4) 39 (2.0)
 Plastic, otolaryngeal, or eye 107 (3.3) 12 (2.8) 86 (4.4)
 Transplantation 298 (9.2) 31 (7.3) 184 (9.4)
 Gynecology or urology 224 (6.9) 12 (2.8) 86 (4.4)
 Gastrointestinal 802 (24.7) 117 (27.5) 431 (22.0)
 Noncardiac thoracic 533 (16.4) 65 (15.3) 289 (14.7)
 Others 19 (0.6) 1 (0.2) 12 (0.6)
Operative variables
 General anesthesia 2867 (88.3) 369 (86.8) 1628 (83.0) <.001 10.2 .12 7.5
 Emergency operation 922 (28.4) 156 (36.7) 505 (25.7) <.001 15.9 .12 7.5
 Operation durationa, h 3.49 (±2.67) 3.07 (±2.68) 3.54 (2.88) .01 11.5 .96 0.6
 Intraoperative inotropics 1353 (41.7) 181 (42.6) 784 (40.0) .39 3.6 .41 4.9
 Intraoperative transfusion 494 (15.2) 68 (16.0) 283 (14.4) .62 2.9 .16 7.0
Data are presented as n (%). ASD was defined as absolute difference in means divided by pooled standard deviation. The ASD is the average of pairwise absolute standardized mean difference across the 3 groups, and ASD <0.1 was deemed to suggest a successful balance between the 3 groups.
Abbreviations: ASD, absolute standardized mean difference; BMI, body mass index; IPTW, inverse probability of treatment weighting.
aMean (±standard deviation).

F1
Figure 1.:
Flowchart of study patient selection. BMI indicates body mass index; SMC-TINCO, Samsung Medical Center Troponin in Noncardiac Operation.

From the total 43,019 patients in the cohort, we excluded 1154 from the registry who were <18 years at the time of surgery and 6596 without postoperative cTn measurements. Among 35,296 adult patients with postoperative cTn measurements, 29,394 patients had normal cTn levels, and 242 patients showed cTn elevation attributed to nonischemic causes, so the incidence of MINS was 16.0% (5633 of 35,296). Finally, a total of 5633 adult patients with MINS were identified and divided into the 3 groups according to BMI as follows: 3246 (57.6%) patients in the normal group, 425 (7.5%) patients in the low BMI group, and 1962 (34.8%) patients in the high BMI group (Figure 1). Preoperative cTn was available in 2.3% (75 of 3246) in the normal group, 1.2% (5 of 425) in the low BMI group, and 10.3% (202 of 1962) in the high BMI group. The incidence of type I myocardial infarction that was angiographically identified was 2.3% (75 of 3246) in the normal group, 1.2% (5 of 425) in the low BMI group, and 10.3% (202 of 1962) in the high BMI group. Baseline characteristics are summarized in Table 1 and operative variables in Table 2. After IPTW, absolute standardized mean difference <10% suggested a well balance between the groups. Separately, the median days to peak postoperative cTn from surgery were 1.22 (IQR: 0.43–3.58) days in the normal group, 1.35 (IQR: 0.50–3.98) days in the low BMI group, and 1.27 (IQR: 0.44–3.40) days in the high BMI group (P = .61). The high BMI group tended to be younger and show higher frequencies of male sex, previous coronary artery disease, and prescriptions for other cardiovascular medications before surgery as well as lower frequencies of emergency operation and general anesthesia. The median follow-up duration was 365 (IQR: 147–365) days.

Clinical Outcomes

Among the 5633 patients, the mortality during the first year after surgery was 19.2% (1079 of 5633). Following IPTW adjustment, the mortality during 1 year of the high BMI group was lower than those of the normal (14.8% vs 20.9%; HR: 0.75; 95% CI, 0.66-0.85; P < .001) and low BMI (14.8% vs 25.6%; HR: 0.56; 95% CI, 0.48-0.66; P < .001) groups (Table 3). The mortality of the high BMI group during the first year after surgery was significantly lower regardless of cardiovascular or noncardiovascular death compared with the normal (3.8% vs 5.7%; HR: 0.71; 95% CI, 0.53-0.96; P = .01 for cardiovascular and 11.0% vs 15.2; HR: 0.76; 95% CI, 0.64-0.91; P < .001 for noncardiovascular death) and the low BMI (3.8% vs 8.5%; HR: 0.48; 95% CI, 0.33-0.70; P < .001 for cardiovascular and 11.0% vs 17.2; HR: 0.60; 95% CI, 0.47-0.76; P < .001 for noncardiovascular death) groups (Table 3). Survival curves during the 1 year of follow-up are shown in Figure 2, while survival curves for 30 days follow-up are presented in Supplemental Digital Content 1, Figure 1, https://links.lww.com/AA/D269. The high BMI group was further divided into overweight and obesity patients, defined as BMI ≥30 kg·m−2 (Supplemental Digital Content 2, Table 1, https://links.lww.com/AA/D270). The mortality of obese patients was numerically lower compared with overweight patients, but it was not statistically significant.

Table 3. - Mortalities of the 3 Groups
Normal (n = 3246) Low BMI (n = 425) High BMI (n = 1962) Low BMI versus high BMI
1-y follow-up
 Mortality, no. (%) 680 (20.9) 109 (25.6) 290 (14.8)
  Unadjusted HR (CI) 1 [reference] 1.34 (1.10-1.65) 0.69 (0.60-0.79) 0.52 (0.42-0.65)
    P value .004 <.001 <.001
  IPTW adjusted HR (CI) 1.33 (1.15-1.54) 0.75 (0.66-0.85) 0.56 (0.48-0.66)
    P value .004 <.001 <.001
 Cardiovascular mortality, no. (%) 186 (5.7) 36 (8.5) 75 (3.8)
  Unadjusted HR (CI) 1 [reference] 1.65 (1.15-2.35) 0.65 (0.50-0.85) 0.39 (0.26-0.59)
    P value .006 .002 <.001
  IPTW adjusted HR (CI) 1.49 (1.06-2.11) 0.71 (0.53-0.96) 0.48 (0.33-0.70)
    P value .01 .01 <.001
 Noncardiovascular mortality, no. (%) 494 (15.2) 73 (17.2) 215 (11.0)
  Unadjusted HR (CI) 1 [reference] 1.23 (0.96-1.58) 0.71 (0.60-0.83) 0.58 (0.44-0.76)
    P value .10 <.001 <.001
  IPTW adjusted HR (CI) 1.27 (1.02-1.58) 0.76 (0.64-0.91) 0.60 (0.47-0.76)
   P value .01 <.001 <.001
30-d follow-up
 Mortality, no. (%) 254 (7.8) 34 (8.0) 127 (6.5)
  Unadjusted HR (CI) 1 [reference] 1.04 (0.72-1.48) 0.82 (0.66-1.02) 0.79 (0.54-1.16)
    P value .84 .07 .23
  IPTW adjusted HR (CI) 1.08 (0.79-1.48) 0.84 (0.66-1.08) 0.78 (0.56-1.10)
    P value .56 .10 .08
 Cardiovascular mortality, no. (%) 67 (2.1) 6 (1.4) 30 (1.5)
  Unadjusted HR (CI) 1 [reference] 0.69 (0.30-1.59) 0.74 (0.48-1.13) 1.07 (0.45-2.57)
    P value .39 .17 .88
  IPTW adjusted HR (CI) 0.58 (0.24-1.39) 0.79 (0.49-1.28) 1.35 (0.55-3.35)
    P value .14 .24 .42
 Noncardiovascular mortality, no. (%) 187 (5.8) 28 (6.6) 97 (4.9)
  Unadjusted HR (CI) 1 [reference] 1.16 (0.78-1.73) 0.85 (0.67-1.09) 0.73 (0.48-1.12)
    P value .46 .20 .15
  IPTW adjusted HR (CI) 1.26 (0.89-1.78) 0.86 (0.65-1.15) 0.69 (0.47-0.99)
    P value .11 .11 .02
Values are n (%). We maintained the α level at .05 for the primary outcome by adopting a post hoc Bonferroni correction, with a significance criterion of 0.0167 (0.05/3), and reported 98.3% CIs.
Abbreviations: CI, confidence interval; BMI, body mass index; HR, hazard ratio; IPTW, inverse probability of treatment weighting.

F2
Figure 2.:
Kaplan-Meier curves of mortality during 1 y for (A) overall death, (B) cardiovascular death, and (C) noncardiovascular death. CI indicates confidence interval; HR, hazard ratio.
F3
Figure 3.:
Smooth plot of HRs for mortality during 1 year according to BMI. BMI indicates body mass index; CI, confidence interval; HRs, hazard ratios.

Finally, change of HR for the mortality during the first year after surgery according to BMI is shown in Figure 3.

Sensitivity Analysis

The sensitivity analysis was conducted assuming that the prevalence of an unmeasured confounder on the observed association was 40% and revealed that the result between the normal and the high BMI groups could become insignificant if odds ratio (OR) with exposure is at least 0.3 and with outcome is at least 2.5 (Supplemental Digital Content 2, Table 2, https://links.lww.com/AA/D270). For the comparisons between the normal and low BMI groups and the low BMI and high BMI groups, an unmeasured confounder with OR stronger than 4.0 with the outcome and stronger than 0.70 with exposure would be needed to change the conclusions for the association with mortality.

DISCUSSION

This study compared mortality following MINS by stratifying patients according to BMI and demonstrated an association between high BMI and lower mortality during 1 year of follow-up after MINS despite a higher incidence of postoperative myocardial infarction in the high BMI group. These findings suggest that the previously reported associations between high BMI and lower mortality in various diseases may be also valid in patients with a diagnosis of MINS.1–10

Robust evidence has consistently indicated poor prognosis for underweight patients,2 while higher BMI has exhibited associations with better outcomes in various clinical situations.1–10 This association has been termed as the “obesity paradox” because obesity is generally considered as a risk factor for health issues, contributing to the development of various diseases.2 The proposed mechanisms to explain this phenomenon in patients with developed pathology are as follows. First, various cytokines released by adipose tissue may play key roles by providing a protective effect against inflammation—that is, they regulate inflammation and endovascular homeostasis and also neutralize tumor necrosis factor-α.26 Second, obesity may also protect patients from the adverse effects of malnutrition and energy expenditures related to their surgical procedures.6 Finally, the prevalence of cardiovascular risk factors relevant to obesity has declined during the modern era,27 and obesity may even prevent malignant ventricular arrhythmias, thereby decreasing the risk for sudden cardiac death.6

The obesity paradox has been particularly well-demonstrated to date in patients with ischemic heart disease, one of the major health issues that obesity contributes to.6,8 In this study, the incidence of postoperative myocardial infarction was also overwhelmingly higher in the high BMI group, but the mortality during 1 year was lower when compared with the normal or low BMI groups. Previously suggested explanations for the obesity paradox in ischemic heart disease include younger age at presentation, greater metabolic reserve, less cachexia, high blood pressure allowing for more cardiac medication, increased muscle mass and muscular strength, and implications regarding cardiorespiratory fitness.9 The high BMI group of our study also matched some of these explanations such as younger age and more cardiac medications after discharge. Despite a rigorous statistical adjustment, these factors might have affected the observed association. In addition, there was also a previous study that demonstrated a smaller infarct size after ST-segment elevation myocardial infarction in obese patients.3 This finding may be applicable to the amount or degree of injury in our patients, which could have affected their survivals.3

The previously reported obesity paradox in surgical patients was mostly present during the long-term follow-up after surgery, not impacting the short-term mortality.28,29 Our results also suggested the significant results during the 1-year follow-up, and the short-term mortality within 30 days did not have a significant difference. For postoperative mortality after cardiac and noncardiac surgeries, a high BMI within the moderate range generally presented an inverse relationship, while extremes of BMI (both underweight and overweight) showed worse outcomes immediately and lasted during the long-term follow-up.26 This study evaluated the association between mortality during the first year after surgery and BMI and found an inverse relationship existed between the 2, but the extremes of BMI showed too low incidence to be analyzed. This may be related to the ethnic difference. Most of our study population consist of Asian, and our patients in the high BMI group could be all defined as obesity according to the 2018 Korean Society for the Study of Obesity Guidelines.30 Within the high BMI group, we further compared overweight patients to obese patients following the CDC guideline and showed numerically lower mortality in obese patients without statistical significance.

To explain the relationship between BMI and postoperative mortality, mechanisms such as increased lean body mass, peripheral body fat, and reduced inflammatory response were suggested.26,28,31 Additionally, patients with higher BMI tend to possess increased muscle mass and show a decreased prevalence of sarcopenia, which is an important risk factor for frailty, mortality, and worse surgical outcomes.32 Increased awareness on both the parts of the surgeon and the anesthesiologist for obesity-associated hazards may also contribute to improved outcomes.31

The clinical implication of this study is that BMI could be also considered when predicting mortality following MINS, which has emerged as one of the most common postoperative complications associated with fetal outcome.12–15 Especially in underweighted patients, the active detection of MINS seems important, and an early intensification of cardiovascular medications, such as aspirin, β-blockers, statin, and/or direct oral anticoagulants, should be considered.33,34 However, the clinical efficacy of weight control among patients scheduled for surgery remains unclear and requires further investigation.

This study had some limitations. First, as it was a single-center, observational investigation, our findings might have been affected by confounding factors. Despite rigorous statistical adjustments, unmeasured variables could not be corrected for. Detailed evaluations of coronary vasculature were not performed in all patients, so the role of the coronary vasculature status on the observed association remains unclear, while noncardiovascular factors related to postoperative mortality might also have affected the results. In addition, this study involved all types of noncardiac surgery, and our results might be different according to particular types of surgery. Second, perioperative cTn measurement was not performed routinely at our institution. It was generally measured in patients with cardiovascular risk according to the institutional protocol; consequently, the possibility of selection bias and a significant difference between the patients with and without preoperative cTn may exist. Moreover, using a period of 30 days postoperatively to detect MINS according to the current definition may impart a large degree of variability, which may have affected the observed association. Third, most of the enrolled patients were Asian. The diagnostic criteria for obesity differ by ethnicity, and the guidelines suggest different cutoff values of BMI for defining obesity. Hence, our results may not be applicable in Western countries. Although the effect of BMI was evaluated in a quantitative manner, due to the small number of patients with morbid obesity, the effect could not be accurately compared. Finally, by defining all undisputed deaths as cases of cardiovascular mortality, the cardiac death rate might have been overestimated. Despite these limitations, this is the first study to show an association between BMI and mortality following MINS. The results of this study may affect future clinical investigations and daily practices.

In summary, in patients who underwent MINS, high BMI appeared to be associated with lower mortality during 1 year. Larger cohort studies are needed to confirm this finding.

DISCLOSURES

Name: Seung-Hwa Lee, MD.

Contribution: This author helped draft the manuscript, interpret the data, and finally approve the manuscript.

Name: Kwangmo Yang, MD.

Contribution: This author helped draft the manuscript, interpret the data, and finally approve the manuscript.

Name: Jungchan Park, MD.

Contribution: This author helped draft the manuscript, interpret the data, and finally approve the manuscript.

Name: Jong Hwan Lee, MD, PhD.

Contribution: This author helped draft the manuscript, interpret the data, and finally approve the manuscript.

Name: Jeong Jin Min, MD, PhD.

Contribution: This author helped draft the manuscript, interpret the data, and finally approve the manuscript.

Name: Ji-hye Kwon MD.

Contribution: This author helped interpret the data and finally approve the manuscript.

Name: Junghyun Yeo, MD.

Contribution: This author helped interpret the data and finally approve the manuscript.

Name: Jihoon Kim, MD.

Contribution: This author helped interpret the data and finally approve the manuscript.

Name: Cheol Won Hyeon, MD.

Contribution: This author helped interpret the data and finally approve the manuscript.

Name: Jin-ho Choi, MD, PhD.

Contribution: This author helped conceive or design the manuscript, interpret the data, and finally approve the manuscript.

Name: Sang-Chol Lee, MD, PhD.

Contribution: This author helped conceive or design the manuscript, interpret the data, and finally approve the manuscript.

Name: Hyeon-Cheol Gwon, MD, PhD.

Contribution: This author helped conceive or design the manuscript, interpret the data, and finally approve the manuscript.

Name: Kyunga Kim, PhD.

Contribution: This author helped acquire, analyze, interpret the data, and finally approve the manuscript.

Name: Joonghyun Ahn, MB.

Contribution: This author helped acquire, analyze, interpret the data, and finally approve the manuscript.

Name: Sangmin Maria Lee MD, PhD.

Contribution: This author helped conceive or design the manuscript, interpret the data, and finally approve the manuscript.

This manuscript was handled by: Tong J. Gan, MD.

    REFERENCES

    1. Lavie CJ, Milani RV, Ventura HO. Obesity and cardiovascular disease: risk factor, paradox, and impact of weight loss. J Am Coll Cardiol. 2009;53:1925–1932.
    2. Di Angelantonio E, Bhupathiraju Sh N, et al.; The Global BMI Mortality Collaboration. Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. Lancet. 2016;388:776–786.
    3. Sohn GH, Kim EK, Hahn JY, et al. Impact of overweight on myocardial infarct size in patients undergoing primary percutaneous coronary intervention: a magnetic resonance imaging study. Atherosclerosis. 2014;235:570–575.
    4. Gruberg L, Weissman NJ, Waksman R, et al. The impact of obesity on the short-term and long-term outcomes after percutaneous coronary intervention: the obesity paradox? J Am Coll Cardiol. 2002;39:578–584.
    5. Shah M, Patil S, Patnaik S, et al. Outcomes in cardiogenic shock from acute coronary syndrome depending on severity of obesity. Am J Cardiol. 2019;123:1267–1272.
    6. Angerås O, Albertsson P, Karason K, et al. Evidence for obesity paradox in patients with acute coronary syndromes: a report from the Swedish Coronary Angiography and Angioplasty Registry. Eur Heart J. 2013;34:345–353.
    7. Das SR, Alexander KP, Chen AY, et al. Impact of body weight and extreme obesity on the presentation, treatment, and in-hospital outcomes of 50,149 patients with ST-segment elevation myocardial infarction results from the NCDR (National Cardiovascular Data Registry). J Am Coll Cardiol. 2011;58:2642–2650.
    8. Romero-Corral A, Montori VM, Somers VK, et al. Association of bodyweight with total mortality and with cardiovascular events in coronary artery disease: a systematic review of cohort studies. Lancet. 2006;368:666–678.
    9. Lavie CJ, McAuley PA, Church TS, Milani RV, Blair SN. Obesity and cardiovascular diseases: implications regarding fitness, fatness, and severity in the obesity paradox. J Am Coll Cardiol. 2014;63:1345–1354.
    10. Fukuoka S, Kurita T, Dohi K, et al. Untangling the obesity paradox in patients with acute myocardial infarction after primary percutaneous coronary intervention (detail analysis by age). Int J Cardiol. 2019;289:12–18.
    11. Thygesen K, Alpert JS, Jaffe AS, et al. Fourth universal definition of myocardial infarction (2018). Eur Heart J. 2019;40:237–269.
    12. Khan J, Alonso-Coello P, Devereaux PJ. Myocardial injury after noncardiac surgery. Curr Opin Cardiol. 2014;29:307–311.
    13. Puelacher C, Lurati Buse G, Seeberger D, et al.; BASEL-PMI Investigators. Perioperative myocardial injury after noncardiac surgery: incidence, mortality, and characterization. Circulation. 2018;137:1221–1232.
    14. Devereaux PJ, Xavier D, Pogue J, et al.; POISE (PeriOperative ISchemic Evaluation) Investigators. Characteristics and short-term prognosis of perioperative myocardial infarction in patients undergoing noncardiac surgery: a cohort study. Ann Intern Med. 2011;154:523–528.
    15. Devereaux PJ, Biccard BM, et al.; Writing Committee for the Vision Study Investigators. Association of postoperative high-sensitivity troponin levels with myocardial injury and 30-day mortality among patients undergoing noncardiac surgery. JAMA. 2017;317:1642–1651.
    16. Devereaux PJ, Yang H, et al.; Poise Study Group. Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery (POISE trial): a randomised controlled trial. Lancet. 2008;371:1839–1847.
    17. Devereaux PJ, Szczeklik W. Myocardial injury after non-cardiac surgery: diagnosis and management. Eur Heart J. 2020;141:3083–3091.
    18. Smilowitz NR, Redel-Traub G, Hausvater A, et al. Myocardial injury after noncardiac surgery: a systematic review and meta-analysis. Cardiol Rev. 2019;27:267–273.
    19. Sessler DI, Devereaux PJ. Perioperative troponin screening. Anesth Analg. 2016;123:359–360.
    20. Quan H, Li B, Couris CM, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol. 2011;173:676–682.
    21. Hicks KA, Tcheng JE, Bozkurt B, et al.; American College of Cardiology; American Heart Association. 2014 ACC/AHA key data elements and definitions for cardiovascular endpoint events in clinical trials: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards (Writing Committee to Develop Cardiovascular Endpoints Data Standards). Circulation. 2015;132:302–361.
    22. Kristensen SD, Knuuti J. New ESC/ESA guidelines on non-cardiac surgery: cardiovascular assessment and management. Eur Heart J. 2014;35:2344–2345.
    23. McCaffrey DF, Griffin BA, Almirall D, Slaughter ME, Ramchand R, Burgette LF. A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Stat Med. 2013;32:3388–3414.
    24. Groenwold RH, Nelson DB, Nichol KL, Hoes AW, Hak E. Sensitivity analyses to estimate the potential impact of unmeasured confounding in causal research. Int J Epidemiol. 2010;39:107–117.
    25. Schoenfeld DA. Sample-size formula for the proportional-hazards regression model. Biometrics. 1983;39:499–503.
    26. Valentijn TM, Galal W, Tjeertes EK, Hoeks SE, Verhagen HJ, Stolker RJ. The obesity paradox in the surgical population. Surgeon. 2013;11:169–176.
    27. Gregg EW, Cheng YJ, Cadwell BL, et al. Secular trends in cardiovascular disease risk factors according to body mass index in US adults. JAMA. 2005;293:1868–1874.
    28. Tjeertes EK, Hoeks SE, Beks SB, et al. Obesity--a risk factor for postoperative complications in general surgery? BMC Anesthesiol. 2015;15:112.
    29. Davenport DL, Xenos ES, Hosokawa P, Radford J, Henderson WG, Endean ED. The influence of body mass index obesity status on vascular surgery 30-day morbidity and mortality. J Vasc Surg. 2009;49:140–147147.e1.
    30. Seo MH, Lee WY, Kim SS, et al.; Committee of Clinical Practice Guidelines, Korean Society for the Study of Obesity (KSSO). 2018 Korean Society for the Study of Obesity Guideline for the Management of Obesity in Korea. J Obes Metab Syndr. 2019;28:40–45.
    31. Klasen J, Junger A, Hartmann B, et al. Increased body mass index and peri-operative risk in patients undergoing non-cardiac surgery. Obes Surg. 2004;14:275–281.
    32. Friedman J, Lussiez A, Sullivan J, Wang S, Englesbe M. Implications of sarcopenia in major surgery. Nutr Clin Pract. 2015;30:175–179.
    33. Foucrier A, Rodseth R, Aissaoui M, et al. The long-term impact of early cardiovascular therapy intensification for postoperative troponin elevation after major vascular surgery. Anesth Analg. 2014;119:1053–1063.
    34. Devereaux PJ, Duceppe E, Guyatt G, et al.; MANAGE Investigators. Dabigatran in patients with myocardial injury after non-cardiac surgery (MANAGE): an international, randomised, placebo-controlled trial. Lancet. 2018;391:2325–2334.

    Supplemental Digital Content

    Copyright © 2020 International Anesthesia Research Society