Abstract
Background:
Older adults undergoing non-cardiac surgery have a high risk of major adverse cardiovascular events (MACE). This study aims to estimate the magnitude of increased perioperative risk, and examine national trends in perioperative MACE following in-hospital non-cardiac surgery in older adults compared to middle aged adults.
Design:
Time-series analysis of retrospective longitudinal data
Setting:
The United States Agency for Healthcare Research and Quality National Inpatient Sample (NIS)
Participants:
Hospitalizations for major non-cardiac surgery among adults age ≥45 years between January 2004 to December 2014
Measurements:
Inpatient perioperative MACE was defined as a composite of in-hospital death, myocardial infarction (MI), and ischemic stroke. In hospital death was determined from the NIS discharge disposition. Myocardial infarction and ischemic stroke were defined by International Classification of Diseases, Ninth Revision codes.
Results:
Of an estimated 55,349,978 surgical hospitalizations, 26,423,039 (47.7%) were for adults age 45–64, 14,231,386 (25.7%) age 65–74, 10,621,029 (19.2%) age 75–84 years, and 4,074,523 (7.4%) age ≥85 years. MACE occurred in 1,601,022 surgical hospitalizations (2.9%). Adults 65–74 (2.8%; aOR 1.16, 95% CI 1.14–1.17), 75–84 years (4.5%; aOR 1.30, 95% CI 1.28–1.32) and ≥85 years (6.9%; aOR 1.55, 95% CI 1.52–1.57) had greater risk of MACE than those 45–64 years (1.7%). From 2004–2014, MACE declined among adults 65–74 (3.1% to 2.5%, p<0.001), 75–85 years (4.9% to 3.9%, p<0.001), and ≥85 years (7.7% to 6.1%, p<0.001), but was unchanged for adults age 45–64. Declines in MACE were driven by decreased MI and mortality despite increased stroke.
Conclusion:
Older adults accounted for half of hospitalizations, but experienced the majority of MACE. Older adults had greater adjusted odds of MACE than younger individuals. The proportion of perioperative MACE declined over time, despite increases in ischemic stroke. These data highlight risks of non-cardiac surgery in older adults that warrant increased attention to improve perioperative outcomes.
Keywords: Mortality, Myocardial Infarction, Noncardiac, Older adults, Operative, Outcomes, Stroke
Introduction:
Individuals ≥65 years old are projected to account for 20% of the United States population by the year 2030, with the greatest population growth anticipated among adults ≥85 years old.1 Older adults already constitute a large proportion of those undergoing non-cardiac surgery, and are at greater risk of perioperative adverse outcomes compared with younger surgical candidates in smaller series.2–4 However, the magnitude of increased perioperative risk in older adults is not well characterized.5–7 Recent studies have not reported major adverse cardiovascular events (MACE) by age from a large, representative cohort in the United States (US). Although the overall rate of perioperative MACE after non-cardiac surgery declined over the past decade, trends in older adults are unknown.8 With the aging US population, perioperative care of older adults will become even more common. Estimation of perioperative risk in older adults may inform shared decision-making, targeted research efforts, and health policy. This study aims to estimate the magnitude of increased risk of perioperative MACE following in-hospital non-cardiac surgery in older adults compared to middle aged adults, and report national trends in perioperative MACE stratified by age.
Methods:
Study population:
Adults ≥45 years of age undergoing in-hospital non-cardiac surgery between 2004–2014 were identified using the US Agency for Healthcare Research and Quality (AHRQ) National Inpatient Sample (NIS). The NIS is comprised of discharge level data from a nationally representative stratified sample of hospitalizations in the US. Prior to 2012, the NIS sampled all discharges from 20% of participating hospitals, but starting in 2012, the NIS changed its methodology to include data from 20% of discharges from all participating hospitals. Adults undergoing in-hospital non-cardiac surgery were identified using principal International Classification of Diseases, Ninth Revision (ICD-9) codes for a major therapeutic operating room procedure (HCUP procedure class 4). Principal Clinical Classifications Software (CCS) procedure codes were used to stratify surgical procedures by subtype. Patients were excluded when the principal procedure was cardiac surgery, bone marrow transplant, ophthalmological surgery, radiation therapy, dental surgery or a non-operating room procedure. The remaining hospitalizations were clustered into surgical subtypes, including endocrine, general, genitourinary, gynecologic, neurosurgery, orthopedic, otolaryngology, skin and breast, thoracic, non-cardiac solid organ transplant, and vascular surgeries. Emergent, urgent and elective surgeries were included in the analysis. As per the US Agency for Healthcare Research and Quality (AHRQ) guidance, clinical comorbidities were defined using AHRQ Elixhauser Comorbidity Software (Version 3.7) standard definitions, CCS codes, or individual ICD9 codes (Supplemental File S1).
Modified Revised Cardiac Risk Index
A modified version of the Revised Cardiac Risk Index (RCRI) was calculated for each hospitalization based upon the presence of ICD-9 codes for ischemic heart disease, heart failure, prior transient ischemic attack or stroke, chronic kidney disease, and diabetes. ICD-9 codes used to define these conditions are outlined in Supplemental File S2.9 High risk surgery was defined as vascular or thoracic surgery. Since the ICD-9 diagnosis code for prior stroke or transient ischemic attack was introduced in 2007, modified RCRI scores were calculated for the years 2008–2014.
Outcomes:
The primary outcome was perioperative MACE, a composite of all-cause in-hospital death, acute myocardial infarction (MI), and acute ischemic stroke during hospitalization for non-cardiac surgery. All-cause in-hospital death was determined from the NIS discharge disposition. Myocardial infarction was defined based on ICD-9 diagnosis code for acute ST-segment elevation myocardial infarction (ICD-9 diagnosis codes 410.01 to 410.61, 410.81, and 410.91) and non-ST-segment elevation myocardial infarction (ICD-9 diagnosis code 410.71), as previously described.10–12 Acute ischemic stroke was defined by ICD-9 diagnosis codes 433.x, 434.x, 436.x, 437.0, and 437.1.13, 14 Perioperative MACE was evaluated in subgroups stratified by age at the time of surgery: 45–64, 65–74, 75–85 years, and ≥85 years old. Trends in perioperative MACE over time were evaluated by age group.
Statistical analysis:
Continuous variables were reported as means with standard deviations. Categorical variables were compared using chi-square tests. Multivariable logistic regression was used to estimate the odds of MACE by age group. Partially adjusted models included race, sex, urgent or emergent hospitalization, and subtype of non-cardiac surgery as covariates for adjustment. Fully adjusted models included all covariates from the partially adjusted models with the addition of comorbidities listed in Table 1. Sampling weights were applied to determine national estimates unless otherwise noted. Statistical analyses were performed using SPSS 25 (IBM SPSS Statistics, Armonk, NY). Two-sided P-values <0.05 were considered to be statistically significant. The NIS is a publicly available, de-identified dataset, and as such, did not require institutional board review.
Table 1:
Characteristics of adults undergoing major in-hospital non-cardiac surgery between 2004–2014
| 45–65 years (n=26,423,039) | 65–75 years (n=14,231,386) | 75–84 years (n=10,621,029) | 85+ years old (n=4,074,523) | |
|---|---|---|---|---|
| Age (SD) | 54.97 (5.7) | 69.31 (2.8) | 79.10 (2.8) | 88.38 (3.0) |
| Female Sex | 14,743,759 (56.0%) | 7,654,627 (53.8%) | 6,077,877 (57.3%) | 2,689,827 (66.0%) |
| Race/ethnicity | ||||
| White non-Hispanic | 16,330,958 (61.8%) | 9,694,456 (68.1%) | 7,443,974 (70.1%) | 2,956,151 (72.6%) |
| Black non-Hispanic | 2,687,067 (10.2%) | 973,690 (6.8%) | 559,153 (5.3%) | 188,535 (4.6%) |
| Hispanic | 1,782,276 (6.7%) | 743,889 (5.2%) | 485640 (4.6%) | 155919 (3.8%) |
| Other | 1,104,980 (4.2%) | 540,915 (3.8%) | 367156 (3.5%) | 128215 (3.1%) |
| Unknown | 4,517,757 (17.1%) | 2,278,436 (16%) | 1765107 (16.6%) | 645703 (15.8%) |
| Expected source of payment | ||||
| Medicare | 4,076,055 (15.5%) | 11,907,913 (83.8%) | 9763975 (92.0%) | 3819586 (93.8%) |
| Medicaid | 2,513,930 (9.5%) | 170,376 (1.2%) | 85394 (0.8%) | 27208 (0.7%) |
| Private insurance | 16,749,750 (63.6%) | 1,868,494 (13.1%) | 640225 (6.0%) | 186487 (4.6%) |
| Self-pay | 1,212,206 (4.6%) | 72,502 (0.5%) | 36776 (0.3%) | 11374 (0.3%) |
| No charge | 162,378 (0.6%) | 7,925 (0.1%) | 3889 (0.0%) | 1183 (0.0%) |
| Other | 1,642,398 (6.2%) | 185,550 (1.3%) | 79129 (0.7%) | 24676 (0.6%) |
| Clinical co-morbidities | ||||
| Tobacco (current or former) | 6,549,744 (24.8%) | 3,197,586 (22.5%) | 1767050 (16.6%) | 412325 (10.1%) |
| Obesity | 4,044,554 (15.3%) | 1,635,190 (11.5%) | 603956 (5.7%) | 76982 (1.9%) |
| Hypertension | 12,684,975 (48.0%) | 9,401,837 (66.1%) | 7398474 (69.7%) | 2858690 (70.2%) |
| Dyslipidemia | 6,135,299 (23.2%) | 5,185,610 (36.4%) | 3793777 (35.7%) | 1130853 (27.8%) |
| Diabetes mellitus | 5,570,423 (21.1%) | 3,893,759 (27.4%) | 2657001 (25.0%) | 763574 (18.7%) |
| Coronary artery disease | 2,580,282 (9.8%) | 3,103,852 (21.8%) | 3001612 (28.3%) | 1180219 (29.0%) |
| Previous PCI | 631,308 (2.4%) | 707,842 (5.0%) | 575769 (5.4%) | 172610 (4.2%) |
| Previous CABG | 576,497 (2.2%) | 867,858 (6.1%) | 865742 (8.2%) | 267001 (6.6%) |
| Atrial fibrillation or flutter | 733,346 (2.8%) | 1,257,097 (8.8%) | 1787303 (16.8%) | 1008190 (24.7%) |
| Prior venous thromboembolism | 577,804 (2.2%) | 408,450 (2.9%) | 335719 (3.2%) | 124929 (3.1%) |
| Prior cerebrovascular accident | 404,027 (1.5%) | 445,872 (3.1%) | 458919 (4.3%) | 236264 (5.8%) |
| Chronic kidney disease | 1,452,719 (5.5%) | 1,186,705 (8.3%) | 1158497 (10.9%) | 576729 (14.2%) |
| End stage renal disease | 677,511 (2.6%) | 365,980 (2.6%) | 238922 (2.2%) | 61664 (1.5%) |
| Heart failure | 826,656 (3.1%) | 905,767 (6.4%) | 1157262 (10.9%) | 761686 (18.7%) |
| Valvular disease | 585,827 (2.2%) | 576,780 (4.1%) | 748922 (7.1%) | 443993 (10.9%) |
| Peripheral vascular disease | 1,218,759 (4.6%) | 1,239,229 (8.7%) | 1133826 (10.7%) | 433070 (10.6%) |
| Chronic pulmonary disease | 3,868,538 (14.6%) | 2,741,243 (19.3%) | 34723 (19.6%) | 11241 (16.9%) |
| Malignancy | 1,209,989 (4.6%) | 876,736 (6.2%) | 705838 (6.6%) | 246266 (6.0%) |
| Anemia | 3,256,814 (12.3%) | 2,109,863 (14.8%) | 2065964 (19.5%) | 1076330 (26.4%) |
| Coagulopathy | 651,502 (2.5%) | 443,401 (3.1%) | 432573 (4.1%) | 213956 (5.3%) |
| Modified RCRI score | ||||
| 0 | 11009255 (64.4%) | 4549997 (48.3%) | 2641941 (40.4%) | 964047 (36.9%) |
| 1 | 3993783 (23.4%) | 2744444 (29.2%) | 2019471 (30.9%) | 842877 (32.3%) |
| 2 | 1351999 (7.9%) | 1320730 (14.0%) | 1150032 (17.6%) | 509408 (19.5%) |
| 3 | 534308 (3.1%) | 565647 (6.0%) | 517914 (7.9%) | 221882 (8.5%) |
| 4 | 173485 (1.0%) | 191592 (2.0%) | 174196 (2.7%) | 64812 (2.5%) |
| 5 | 32187 (0.2%) | 36932 (0.4%) | 32008 (0.5%) | 9543 (0.4%) |
| 6 | 2106 (0.0%) | 2438 (0.0%) | 2062 (0.0%) | 576 (0.0%) |
| Teaching Hospital | ||||
| Rural | 2,162,540 (8.2%) | 1,363,226 (9.6%) | 1123738 (10.6%) | 495942 (12.2%) |
| Urban nonteaching | 9,910,937 (37.7%) | 5,767,325 (40.7%) | 4569075 (43.2%) | 1820909 (44.9%) |
| Urban teaching | 14,220,539 (54.1%) | 7,045,694 (49.7%) | 4890348 (46.2%) | 1743090 (42.9%) |
| Emergency Hospitalization | ||||
| No | 16,869,815 (64.0%) | 9,439,545 (66.5%) | 5860829 (55.3%) | 1241341 (30.6%) |
| Yes | 9,474,994 (36.0%) | 4,752,979 (33.5%) | 4731367 (44.7%) | 2821611 (69.4%) |
| Surgery Type | ||||
| General | 6,193,184 (23.4%) | 2,776,714 (19.5%) | 2089477 (19.7%) | 755237 (18.5%) |
| Endocrine | 389,144 (1.5%) | 150,148 (1.1%) | 70884 (0.7%) | 10181 (0.2%) |
| Genitourinary | 1,870,013 (7.1%) | 1,117,003 (7.8%) | 732010 (6.9%) | 250568 (6.1%) |
| Gynecological | 2,323,071 (8.8%) | 485,189 (3.4%) | 241962 (2.3%) | 44150 (1.1%) |
| Neurosurgery | 1,630,775 (6.2%) | 816,177 (5.7%) | 568191 (5.3%) | 131299 (3.2%) |
| Orthopedic | 9,713,369 (36.8%) | 6,112,327 (42.9%) | 4705626 (44.3%) | 2142644 (52.6%) |
| Otolaryngology | 248,902 (0.9%) | 80,692 (0.6%) | 57336 (0.5%) | 23757 (0.6%) |
| Skin/Breast | 1,311,142 (5.0%) | 483,223 (3.4%) | 372044 (3.5%) | 176837 (4.3%) |
| Thoracic | 555,273 (2.1%) | 387,026 (2.7%) | 237872 (2.2%) | 39169 (1.0%) |
| Transplant | 134,239 (0.5%) | 33,542 (0.2%) | 3020 (0.0%) | 38 (0.0%) |
| Vascular | 2,009,095 (7.6%) | 1,789,335 (12.6%) | 1542609 (14.5%) | 500643 (12.3%) |
Standard deviation
Percutaneous coronary intervention
Coronary artery bypass graft
Results:
Study Population:
We identified 10,581,621 hospitalizations (unweighted) between January 2004 to December 2014 for non-cardiac surgery among adults age ≥45 years old, which represented an estimated 55,349,978 hospitalizations after applying sampling weights. Of these surgical hospitalizations, 47.7% were for adults 45–64 years, 25.7% for adults 65–74 years, and 19.2% for adults 75–84 years and 7.4% for adults ≥85 years. The number of annual hospitalizations for non-cardiac surgery peaked among individuals age 65 and then declined with increasing age (Figure 1). Trends in age during surgical hospitalizations over time are shown in Supplemental Figure S1. Demographic and clinical characteristics of the study population are displayed in Table 1. Overall, among surgical hospitalizations, 56.4% of patients were women, and 65.8% were of non-Hispanic white race/ethnicity.
Figure 1. Annual hospitalizations for non-cardiac surgery and the incidence of major adverse cardiovascular events by age.
The number of annual hospitalizations for non-cardiac surgery (green line) decreased with age (45 years old: 987,467 hospitalizations per year vs. 90 years old: 636,542 hospitalizations per year), while MACE per 100,000 non-cardiac surgeries increased with age (45 years old: 921 MACE per 100,000 surgeries vs. 90 years old: 7,171 MACE per 100,000 surgeries).
Older adults, particularly those ≥75 years old, had a high burden of cardiovascular risk factors, such as coronary artery disease and heart failure. Adults ≥75 were most likely to have an urgent or emergent surgical hospitalization (45–64 years: 36.0% vs. 64–75 years: 33.5% vs. 75–84 years: 44.7% vs. ≥85 years: 69.4%). Surgical subtypes performed by age group are shown in Table 1. Overall, the most common surgical procedures were orthopedic surgery (41.0%), general surgery (21.4%), and vascular surgery (10.6%). Proportions of surgical admissions for orthopedic surgery (45–64 years: 36.8% vs. 65–74 years: 42.9% vs. 75–84 years: 44.3% ≥85 years: 52.6%) and vascular surgery (45–64 years: 7.6% vs. 65–74 years: 12.6% vs. 75–84 years: 14.5% ≥85 years: 12.3%) increased with age.
Perioperative Major Adverse Cardiovascular Events by Age:
Perioperative MACE occurred in 2,893 per 100,000 hospitalizations among adults ≥45 years old undergoing non-cardiac surgery. The proportion of MACE during surgical hospitalizations increased with age (45–64 years: 1,658 per 100,000 surgeries, [1.7%], 95% CI: 1,653–1,663 per 100,000 surgeries vs. 65–74 years: 2,825 per 100,000 surgeries [2.8%], 95% CI: 2,816–2,833 per 100,000 surgeries vs. 75–84 years: 4,513 per 100,000 surgeries, [4.5%], 95% CI: 4,500–4,525 ≥85 years: 6,913 per 100,000 surgeries [6.9%], 95% CI: 6,889–6,938 per 100,000 surgeries). The relationship between age and perioperative MACE is displayed in Figure 1. Similar relationships were observed between age and perioperative mortality, MI, and stroke (Supplemental Figures S2a–c). Adults age 65–74 years (adjusted OR [aOR] 1.48, 95% CI 1.47–1.50), 75–84 years (aOR 2.29, 95% CI 2.26–2.31), and ≥85 years (aOR 4.23, 95% CI 4.16–4.30) had greater odds of perioperative MACE than adults age 45–64 after adjusting for race/ethnicity, sex, urgent or emergent hospitalization, and subtype of non-cardiac surgery. After adjusting for the full complement of clinical comorbidities, adults age 65–74 (aOR 1.16, 95% CI 1.14–1.17), 75–84 years (aOR 1.30, 95% CI 1.28–1.32) and ≥85 years (aOR 1.55, 95% CI 1.52–1.57) had greater odds of perioperative MACE compared with adults age 45–64 (Figure 2). Older adults had a greater adjusted odds of perioperative MACE than younger individuals across all sex and race/ethnicity categories (Supplemental Table S1 & S2).
Figure 2. Incidence and adjusted odds of perioperative MACE among older adults undergoing in-hospital non-cardiac surgery by age group.
The crude incidence of MACE is shown by age category. MACE was highest among patients ≥85 years old (6,913 per 100,000 surgeries) and 75–64 years (4,513 per 100,000 surgeries) versus 45–64 years old (1,658 per 100,000 surgeries). The odds of perioperative MACE are shown after adjustment for age, sex, race/ethnicity, urgent/emergent hospitalization, subtype of non-cardiac surgery and clinical comorbidities. Surgical patients age 45–64 years serve as the reference group.
Trends in Perioperative Major Adverse Cardiovascular Events:
Between 2004 to 2014, the incidence of perioperative MACE declined among adults 65–74 years (from 3,116 per 100,000 surgical hospitalizations in 2004 [95% CI: 3,085–3,148 per 100,000 surgeries] to 2,459 per 100,000 in 2014 [95% CI: 2,433–2,485 per 100,000 surgeries], p<0.001), 75–84 years (4,880 per 100,000 in 2004 [95% CI 4,838–4,922] to 3,918 per 100,000 surgeries in 2014 [95% CI 3,877–3,960], p<0.001) and ≥85 years (from 7,652 per 100,000 surgical hospitalizations in 2004 [95% CI: 7,563–7,740 per 100,000 surgeries] to 6,055 per 100,000 in 2014 [95% CI: 5,975–6,134 per 100,000 surgeries], p<0.001) but was unchanged among adults 45–64 years (from 1,630 per 100,000 surgical hospitalizations in 2004 [95% CI: 1,613–1,647 per 100,000 surgeries] to 1,580 per 100,000 in 2014 [95% CI 1,563–1,596 per 100,000 surgeries]) (Figure 3). Declines in MACE were driven by decreased perioperative mortality and MI, while the incidence of perioperative ischemic stroke increased in all age groups (Supplemental Figure S3). After adjustment for demographics and clinical comorbidities, the odds of MACE declined over time in all age groups (Supplemental Figure S4a–b).
Figure 3. Perioperative MACE among adults undergoing in-hospital non-cardiac surgery between 2004–2014 by age group.
Trends in perioperative MACE over time are shown stratified by age group. The proportion of patients with MACE declined between 2004 and 2014 among adults ≥85 years (from 7,652 per 100,000 surgical hospitalizations in 2004 to 6,055 per 100,000 in 2014, p<0.001), 75–84 years (4,880 per 100,000 in 2004 to 3,918 per 100,000 surgeries in 2014, p<0.001) and 65–74 years (2004: 3,116 per 100,000 hospitalizations to 2,459 per 100,000 in 2014, p<0.001), but not among adults 45–64 years old (1,630 per 100,000 surgical hospitalizations in 2004 to 1,580 per 100,000 in 2014).
Perioperative Outcomes by Surgical Subtype
Adverse cardiovascular events were most likely to occur in adults age ≥65 years undergoing vascular surgery (8,054 per 100,000 surgeries [8.1%]), 95% CI: 8,027–8,081 per 100,000 surgeries) thoracic surgery (7,645 per 100,000 surgeries [7.6%], 95% CI: 7,581–7,708 per 100,000 surgeries) and general surgery (5,954 per 100,000 surgeries [6.0%], 95% CI: 5,934–5,973 per 100,000 surgeries). Among all surgical subtypes, older adults age ≥65 years had a greater risk of MACE compared to individuals age 45–65 years. The frequency of adverse events stratified by age category and surgical subtype are shown in Figure 4.
Figure 4. Perioperative MACE by surgical subtype, stratified by age group.
The incidence of perioperative MACE was compared by age category across surgical subtypes. Older adults had the highest incidence of MACE across all surgical subtypes. Among adults age ≥65 years, thoracic (8,054 MACE events per 100,000 surgeries), vascular (7,645 MACE events per 100,000 surgeries), and general surgery (5,954 MACE events per 100,000 surgeries) were associated with the greatest risks of MACE.
Risk Indices and Perioperative Major Adverse Cardiovascular Events in Older Adults
A high-risk RCRI (≥2) was present in 12.2% of individuals age 45–65, 22.5% age 65–75, and 28.76% age 75–84 years, and 30.9% in age ≥85 years old. The incidence and adjusted odds of perioperative MACE was higher in patients with an elevated RCRI across all age groups, with stepwise increases in risk observed with additional RCRI factors (Supplemental Figure S5a & S5b). Among all age groups, RCRI ≥2 was associated with a greater odds of perioperative MACE (45–65 years: aOR 2.25, 95% CI 2.18–2.33; 65–74 years: aOR 2.17, 95% CI 2.11–2.24; 75–84 years: aOR 1.98, 95% CI 1.93–2.04, ≥85 years: aOR 1.87, 95% CI 1.82–1.92) compared to patients with low-risk RCRI scores <2.
Discussion:
In a large retrospective study representing 55 million surgical hospitalizations in the US, older adults had a greater risk of MACE compared with younger adults. Perioperative MACE occurred during 1 in 59 hospitalizations of adults 45–64 years, 1 in 36 hospitalizations of adults 65–74 years, 1 in 22 hospitalizations of adults 75–84 years, and 1 in 14 hospitalizations of adults age ≥85 years. Adults ≥65 years accounted for half of surgical hospitalizations in this analysis but experienced 72.6% of perioperative MACE events. Even after full adjustment for clinical covariates, adults age 75–84 years had 30% greater odds and adults age ≥85 years had a 55% greater odds of in-hospital perioperative MACE than adults age 45–64 years. From 2004 to 2014, we observed declines in perioperative MACE over time among adults ≥65 years old, driven by decreases in perioperative MI and mortality. Vascular surgery, thoracic surgery, and general surgery were associated with the highest incidence of MACE among older adults. High risk RCRI was associated with greater risk of MACE among younger and older adults.
This study is one of the largest to estimate outcomes of older adults undergoing non-cardiac surgery in the modern era. Our findings reaffirm that older adults frequently undergo in-hospital non-cardiac surgery that is associated with a significant risk of perioperative MACE. In this analysis, the number of surgical hospitalizations among adults age ≥65 years was roughly equivalent to the number of surgical hospitalizations among those 45–64 years, but older adults accounted for 72.6% of all perioperative MACE. As the population continues to age the numbers of surgical hospitalizations for older adults and perioperative cardiovascular events are anticipated to increase.
Despite the high prevalence of perioperative MACE in older adults, we identified encouraging declines in the rate of in-hospital adverse events over time. There are a number of possible reasons for this favorable trend, including a coordinated, nationwide focus on improvement in surgical outcomes through quality improvement programs.15, 16 Improved surgical techniques, including use of minimally invasive surgery, and improved intraoperative and post-operative monitoring may also contribute to improved outcomes. Furthermore, large randomized trials have informed clinical practice guidelines on optimal patient selection and perioperative medical management to minimize cardiovascular risks.3, 17 Many of these guidelines stress the importance of weighing treatment options in older patients, thoughtful patient selection for surgery, and expanded use of non-operative therapies.18–20 A rising proportion of lower risk patient presentations for inpatient surgery may also explain the favorable trends observed in recent years.
Despite the decline in overall perioperative MACE, we observed an increase in perioperative ischemic stroke over time. Potential explanations for the increased incidence of ischemic stroke in our analysis include increases in atrial fibrillation and chronic kidney disease, perioperative hypotension in the setting of increased use of angiotensin-converting enzyme inhibitors, or perhaps due to ascertainment bias from more frequent use of magnetic resonance and perfusion computed tomographic imaging.
Despite the decline in MACE among older adults undergoing non-cardiac surgery over time, opportunities exist to improve perioperative outcomes in this vulnerable population. Given the excess of perioperative risks compared with younger individuals, pre-operative care in older adults should emphasize shared decision-making.21–23 Development and refinement of risk stratification tools that incorporate multimorbidity, cardiovascular risk, and frailty relevant to older adults are also necessary.6 This analysis, in combination with other efforts to elucidate factors associated with perioperative MACE in older adults, may identify promising areas of investigation and intervention.
Limitations:
There are limitations to the present study. First, the present analysis is limited to in-hospital non-cardiac surgeries, and the findings are not applicable to ambulatory or outpatient surgical procedures. Similarly, our analysis captured in-hospital MACE, but events that occurred after hospital discharge were not reported. Second, our analysis relies on ICD-9 coding data, which may be influenced by accuracy of coding and reporting bias, and lacks granular information such as perioperative laboratory testing, imaging, and medication use. Our modified RCRI may overestimate risk relative to the traditional RCRI, as we were unable to assess insulin use in patients with diabetes mellitus and creatinine clearance in patients with renal disease; instead we assigned points for any kidney disease and any diagnosis of diabetes mellitus. Third, since cardiac biomarkers were not reported, myocardial injury after non-cardiac surgery could not be estimated. Therefore, our composite endpoint of MACE may underestimate the true frequency of cardiovascular complications after surgery as myocardial injury after non-cardiac surgery increases short and long term mortality.24 Fourth, we were unable to adjudicate cause of death from these administrative data, and cardiovascular and non-cardiac mortality could not be differentiated. Estimations of post-operative all-cause mortality may overestimate post-operative cardiovascular mortality. Fifth, the timing of MI and ischemic stroke relative to non-cardiac surgery could not be confirmed based on administrative data from NIS, although the majority of events are likely to have occurred postoperatively, since major surgery is contraindicated immediately following MACE. Finally, the NIS does not track patients across hospitalizations, and therefore all hospitalizations were analyzed independently.
Conclusion:
In this analysis of a large national United States database, older adults accounted for half of surgical hospitalizations. Adults age 65–74 years, 75–84 years and ≥85 years had greater odds of perioperative MACE than adults age 45–64 years, and events in adults age ≥65 accounted for 72.6% of all perioperative MACE. Overall, the proportion of patients with perioperative MACE declined over time from 2004 to 2014, but increases in perioperative ischemic stroke warrant further study. These data highlight risks of non-cardiac surgery among older adults that warrant increased attention to improve perioperative outcomes.
Supplementary Material
Key Points:
Adults ≥65 years accounted for half of surgical hospitalizations in the United States and experienced 72.6% of major adverse cardiovascular events (MACE).
Adults 75–84 (4.5%) and ≥85 years (6.9%) had greater risks of MACE than adults 65–74 years (2.8%) and 45–64 years (1.7%).
Perioperative MACE among patients ≥65 years declined over time.
Why Does This Matter:
Older adults experience the majority of MACE after non-cardiac surgery. Risks of non-cardiac surgery in older adults warrant increased attention to improve perioperative outcomes.
Acknowledgements
Funding / Disclosures: Dr. Smilowitz is supported, in part, by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number K23HL150315. Dr. Dodson is supported by a Patient Oriented Career Development Award (K23 AG052463) from the National Institute of Aging. Dr. Berger is funded, in part, by the National Heart and Lung Blood Institute of the National Institute of Health (R01HL139909 and R35HL144993). The remaining authors have no disclosures to report.
Sponsor’s Role: None
Footnotes
Conflict of interests: None
References
- 1.Vespa J, Medina L, Armstrong D. Demographic Turning Points for the United States: Population Projections for 2020 to 2060. Current Population Reports 2020. [Google Scholar]
- 2.Rich MW, Chyun DA, Skolnick AH, et al. Knowledge Gaps in Cardiovascular Care of the Older Adult Population: A Scientific Statement From the American Heart Association, American College of Cardiology, and American Geriatrics Society. Circulation. 2016;133:2103–2122. [DOI] [PubMed] [Google Scholar]
- 3.Fleisher LA, Fleischmann KE, Auerbach AD, et al. 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. J Am Coll Cardiol. 2014;64:e77–137. [DOI] [PubMed] [Google Scholar]
- 4.Deiner S, Westlake B, Dutton RP. Patterns of surgical care and complications in elderly adults. Journal of the American Geriatrics Society. 2014;62:829–835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chow WB, Rosenthal RA, Merkow RP, et al. Optimal preoperative assessment of the geriatric surgical patient: a best practices guideline from the American College of Surgeons National Surgical Quality Improvement Program and the American Geriatrics Society. Journal of the American College of Surgeons. 2012;215:453–466. [DOI] [PubMed] [Google Scholar]
- 6.Alrezk R, Jackson N, Al Rezk M, et al. Derivation and Validation of a Geriatric-Sensitive Perioperative Cardiac Risk Index. Journal of the American Heart Association. 2017;6:e006648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mohanty S, Rosenthal RA, Russell MM, Neuman MD, Ko CY, Esnaola NF. Optimal Perioperative Management of the Geriatric Patient: A Best Practices Guideline from the American College of Surgeons NSQIP and the American Geriatrics Society. J Am Coll Surg. 2016;222:930–947. [DOI] [PubMed] [Google Scholar]
- 8.Smilowitz NR, Gupta N, Ramakrishna H, Guo Y, Berger JS, Bangalore S. Perioperative Major Adverse Cardiovascular and Cerebrovascular Events Associated With Noncardiac Surgery. JAMA cardiology. 2017;2:181–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation. 1999;100:1043–1049. [DOI] [PubMed] [Google Scholar]
- 10.Smilowitz NR, Gupta N, Guo Y, Berger JS, Bangalore S. Perioperative acute myocardial infarction associated with non-cardiac surgery. European Heart Journal. 2017;38:2409–2417. [DOI] [PubMed] [Google Scholar]
- 11.Smilowitz NR, Beckman JA, Sherman SE, Berger JS. Hospital Readmission After Perioperative Acute Myocardial Infarction Associated With Noncardiac Surgery. Circulation. 2018;137:2332–2339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Metcalfe A, Neudam A, Forde S, et al. Case definitions for acute myocardial infarction in administrative databases and their impact on in-hospital mortality rates. Health Serv Res. 2013;48:290–318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Brinjikji W, Rabinstein AA, Kallmes DF, Cloft HJ. Patient outcomes with endovascular embolectomy therapy for acute ischemic stroke: a study of the national inpatient sample: 2006 to 2008. Stroke. 2011;42:1648–1652. [DOI] [PubMed] [Google Scholar]
- 14.McCormick N, Bhole V, Lacaille D, Avina-Zubieta JA. Validity of Diagnostic Codes for Acute Stroke in Administrative Databases: A Systematic Review. PLoS One. 2015;10:e0135834. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cohen ME, Liu Y, Ko CY, Hall BL. Improved Surgical Outcomes for ACS NSQIP Hospitals Over Time: Evaluation of Hospital Cohorts With up to 8 Years of Participation. Ann Surg. 2016;263:267–273. [DOI] [PubMed] [Google Scholar]
- 16.Ingraham AM, Richards KE, Hall BL, Ko CY. Quality improvement in surgery: the American College of Surgeons National Surgical Quality Improvement Program approach. Adv Surg. 2010;44:251–267. [DOI] [PubMed] [Google Scholar]
- 17.Smilowitz NR, Berger JS. Perioperative Cardiovascular Risk Assessment and Management for Noncardiac Surgery: A Review. Jama. 2020;324:279–290. [DOI] [PubMed] [Google Scholar]
- 18.Colburn JL, Mohanty S, Burton JR. Surgical Guidelines for Perioperative Management of Older Adults: What Geriatricians Need to Know. J Am Geriatr Soc. 2017;65:1339–1346. [DOI] [PubMed] [Google Scholar]
- 19.Ma M, Zhang L, Rosenthal R, Finlayson E, Russell MM. The American College of Surgeons Geriatric Surgery Verification Program and the Practicing Colorectal Surgeon. Semin Colon Rectal Surg. 2020:100779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chow WB, Rosenthal RA, Merkow RP, Ko CY, Esnaola NF. Optimal preoperative assessment of the geriatric surgical patient: a best practices guideline from the American College of Surgeons National Surgical Quality Improvement Program and the American Geriatrics Society. J Am Coll Surg. 2012;215:453–466. [DOI] [PubMed] [Google Scholar]
- 21.Lilley EJ, Cooper Z, Schwarze ML, Mosenthal AC. Palliative Care in Surgery: Defining the Research Priorities. J Palliat Med. 2017;20:702–709. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Arterburn D, Wellman R, Westbrook E, et al. Introducing decision aids at Group Health was linked to sharply lower hip and knee surgery rates and costs. Health Aff (Millwood). 2012;31:2094–2104. [DOI] [PubMed] [Google Scholar]
- 23.Taylor LJ, Nabozny MJ, Steffens NM, et al. A Framework to Improve Surgeon Communication in High-Stakes Surgical Decisions: Best Case/Worst Case. JAMA Surg. 2017;152:531–538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.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. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.