Revascularization Versus Medical Therapy in Patients Aged 80 Years and Older With Acute Myocardial Infarction

Derek Q. Phan, MD; Ara H. Rostomian, MD, MBA, MPH; Franz Schweis, MD; Joanie Chung, MPH; Bryan Lin, MS; Ray Zadegan, MD; Ming-Sum Lee, MD, PhD


J Am Geriatr Soc. 2020;68(11):2525-2533. 

In This Article


Demographic and Clinical Characteristics

A total of 1,433 patients (median age = 83.5 years (IQR = 81.5–86.2 years); 66% male) were included in the analysis (Figure 1), with 714 (49.8%) undergoing PCI, 176 (12.3%) undergoing CABG, and 543 (37.9%) receiving medical therapy.

Table 1 shows baseline characteristics of the three groups. Patients treated with PCI were less likely to have diabetes mellitus, type II, but more likely to have had prior PCI, and higher hemoglobin levels. Those treated with CABG were slightly younger, and more likely to have left main involvement. Overall, those treated with medical therapy alone were more likely to be Black, had lower LVEF, had one or more CTO in any vessel, and had higher brain natriuretic peptide and creatinine laboratory values. They also had more comorbidities, including atrial fibrillation/flutter, prior stroke, CAD, prior CABG, chronic obstructive pulmonary disease, and peripheral vascular disease; and were already on more medications before ICA, including beta-blockers, nitrates, statins, clopidogrel, and angiotensin-converting enzyme inhibitors/angiotensin receptor blockers.

Given the concern for confounding by indication leading to selection bias, whereby the medical therapy group represents a "sicker" population compared with the revascularization group, we adjusted baseline characteristics with IPTW. We used IPTW instead of propensity score matching to avoid omitting subjects without a match and utilize the data set in its entirety. After weighted adjustment using a propensity score derived from a logistic regression model including 29 different covariates, baseline clinical characteristics between groups were well balanced (Supplementary Table S1).

Kaplan-Meier Estimates of Outcomes

At a median follow-up of 2.6 years (IQR = 1–4.8 years), death occurred in 319 of 714 (45%) patients undergoing PCI, 72 of 176 patients (41%) undergoing CABG, and 321 of 543 (59%) patients undergoing medical therapy. In addition, there were 317 nonfatal MIs and 77 repeated revascularizations. We performed Kaplan-Meier survival analysis among the three treatment groups. Compared with PCI or CABG, medical therapy had the highest rate of all-cause mortality (P < .001 between all group) (Figure 2A) and nonfatal MI (P < .001 between all groups) (Figure 2B). Those treated with PCI had the highest rate of repeated revascularization compared with CABG or medical therapy (P < .05 between all groups) (Figure 2C).

Figure 2.

Kaplan-Meier curves stratified by outcomes among groups: all-cause mortality (A), nonfatal myocardial infarction (MI) (B), and repeated revascularization (C). CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention.

Cox Proportional Hazards Regression Analysis of Outcomes

Cox regression was performed to compare treatment groups and outcomes. Univariate, multivariate, and IPTW adjusted Cox regression models were performed to account for confounding (Table 2). Revascularization (PCI or CABG) was associated with a lower risk of all-cause mortality (IPTW hazard ratio (HR) = 0.66; 95% confidence interval (CI) = 0.60–0.73) and nonfatal MI (IPTW HR = 0.68; 95% CI = 0.58–0.78), but a higher risk for repeated revascularization (IPTW HR = 1.60; 95% CI = 1.15–2.23), compared with medical therapy (Table 2). Likewise, PCI was associated with a lower risk of all-cause mortality (IPTW HR = 0.70; 95% CI = 0.63–0.78) and nonfatal MI (IPTW HR = 0.71; 95% CI = 0.61–0.83), but a higher risk of repeated revascularization (IPTW HR = 1.77; 95% CI = 1.26–2.51), compared with medical therapy. CABG was associated with a lower risk of all-cause mortality (IPTW HR = 0.58; 95% CI = 0.50–0.67) and nonfatal MI (IPTW HR = 0.51; 95% CI = 0.41–0.65), but no difference in repeated revascularization, compared with medical therapy (IPTW HR = 0.97; 95% CI = 0.55–1.72). When comparing PCI with CABG, there were no differences in all-cause mortality (IPTW HR = 1.08; 95% CI = 0.94–1.24), but increased nonfatal MI (IPTW HR = 1.61; 95% CI = 1.28–2.03) and repeated revascularization (IPTW HR = 2.85; 95% CI = 1.77–4.58) with PCI.

Subgroup Analysis

Subgroup analysis comparing revascularization with medical therapy with regard to all-cause mortality was performed using IPTW adjusted Cox regression models (Figure 3). Among all subgroups, revascularization was associated with lower mortality compared with medical therapy, except in Black patients and those with prior CABG.

Figure 3.

Forest plot of inverse probability of treatment weighting Cox proportional hazards regression analysis for all-cause mortality by subgroups. CABG, coronary artery bypass grafting; CI, confidence interval; HR, hazard ratio; PCI, percutaneous coronary intervention.

Rationale for Medical Therapy

Among those treated with medical therapy after ICA, the two most common reasons were (1) poor candidacy for either PCI or CABG due to suboptimal coronary anatomy, medical comorbidities, frailty, or other reasons at the discretion of the treating physician at the time (38.9%) and (2) significant obstructive CAD but high risk-benefit ratio favoring a trial of medical therapy (36.3%) first with the option for intervention "as needed" if fails. About 15% of patients declined any invasive treatment (Supplementary Table S2).