Sex Disparity in Outcomes of Ruptured Abdominal Aortic Aneurysm Repair Driven by In-Hospital Treatment Delays

Linda J. Wang, MD, MBA; Satinderjit Locham, MD; Hanaa Dakour-Aridi, MD; Keith D. Lillemoe, MD; Bryan Clary, MD; Mahmoud B. Malas, MD, MHS


Annals of Surgery. 2019;270(4):630-638. 

In This Article


Study Population

There were 3719 operative repairs for rAAA performed over the study period: 2922 in men (M) and 797 in women (W). Just over half (n=1964, 53%) underwent EVAR, whereas 47% (n=1755) had OSR. Median age was 73 years (IQR 66–80) and women tended to present at an older age than men (W 76 yrs [70–83] vs M 72 yrs [65–79], P<0.001). Women were more likely than men to have hypertension (82% vs 79%, P=0.04) and COPD (37% vs 30%, P<0.001). Women were less likely than men to have a smoking history (72% vs 82%, P<0.001) and CAD (25% vs 33%, P<0.001). Regarding anatomic features, women were less likely to have concomitant iliac aneurysmal disease (11% vs 27%, P<0.001) and tended to present with rupture at smaller AAA diameters (median, 68 mm vs 78 mm, P<0.001). There was no difference in preoperative medication use (Table 1).

Periprocedural Variables

The majority (60%) of patients were transferred from an outside facility for rAAA treatment, with no difference in transfer status between sexes (M 60% vs W 61%, P=0.66). When comparing operative approaches, similar proportions of men and women underwent EVAR (M 53% vs W 53%, P=0.93). Procedural variables were similar between sexes, notably procedural time (median, M 2.8 h vs W 2.7 h, P=0.23) and estimated blood loss (M 1.0 L vs W 0.8 L, P=0.59) did not differ (Table 2).

Perioperative Complications

When perioperative complications were reviewed, there were no differences by sex with respect to stroke, MI, acute kidney injury, mesenteric ischemia, or unplanned return to the operating room. Women were more likely than men to have lower extremity ischemia requiring operative intervention (7.3% vs 5.1%, P=0.02). There were no differences in overall major complication rates (M 52.6% vs W 52.3%, P=0.89). Overall, 30-day mortality was 27.1% (n=1006) with higher rates seen in women compared with men (W 32.8% vs M 25.5%, P<0.001) (Table 3).

Predictors of 30-day Mortality

When significant factors on univariate analysis were assessed simultaneously in multivariable analysis, 7 remained significant predictors of all-cause 30-day mortality: age ≥60 years, female sex, Caucasian race, obesity, maximum AAA diameter ≥90 mm, facility transfer, and open surgical repair (Table 4). Transfer from another facility was associated with a 26% reduction in 30-day mortality [odds ratio (OR) 0.74, 95% confidence interval (CI), 0.63–0.87, P<0.001]. Female sex was associated with a 48% increased risk of perioperative death (OR 1.48, 95% CI, 1.23–1.78, P<0.001). When the overall population was stratified by type of operative repair, female sex was associated with a 42% increased risk of perioperative mortality after EVAR (OR 1.42, 95% CI, 1.09–1.86, P=0.01) and a 50% increased risk after OSR (OR 1.50, 95% CI, 1.17–1.94, P=0.002).

Time-delay Cohorts

The impact of in-hospital treatment delays was analyzed. One hundred forty-six patients ruptured after admission and 349 patients were missing one or more time data point, leaving a total of 3224 patients to include in this subanalysis (Figure 1). Women and men had similar delays in time from symptom onset to admission at treatment facility (median, M 5.5 h vs W 6.0 h, P=0.18). Despite similar presentation delays, women had longer in-hospital delays, with longer time from admission to operative repair (M 1.2 h vs W 1.5 h, P=0.047). Women were also more likely than men to have treatment delays >1.5 h (M 44% vs W 49%, P=0.01) (Table 2).

Figure 1.

Flow diagram of patients undergoing time-delay subgroup analysis.

30-day Mortality

Multivariable logistic models were rerun based on door-to-intervention times of ≤90 and >90 minutes. Among patients treated within 90 minutes of admission, no increased risk of perioperative mortality was noted among women; this finding persisted when the cohort was stratified by operative repair type (model 2, Table 5). When looking at patients who were treated beyond the 90-minute benchmark, women were found to have increased risk of mortality; adjusted odds of 30-day mortality was 1.77 (P<0.001) in women treated after 90 minutes of presentation. Stratifying by operative repair, this increased risk persisted after both EVAR (OR 1.52, P=0.03) and OSR (OR 2.00, P<0.001) (model 3, Table 5).

A dose–response was also analyzed. In 726 patients intervened upon within 30 minutes of admission, 247 (34%) died within 30 days and 60% of these deaths occurred within 24 hours of intervention. After risk adjustment, women had increased risk of 30-day death over men (OR 2.02, P=0.001) in this subgroup. In patients treated within 30 to 90 minutes, no difference in mortality was observed between sex cohorts (OR 0.97, P=0.86). In patients treated at the 90- to 120-minute mark, a near doubling of risk was observed in women (OR 1.93, P=0.046).

One-year Survival

All-cause mortality at 1-year was reviewed. At 1-year, survival was 66% in men and 56% in women (P=0.001) (Figure 2). Regression performed with Cox proportional hazards modeling found age ≥80 [hazard ratio (HR) 2.7, standard error (SE) 0.4, P<0.001], CHF (HR 1.4, SE 0.1, P=0.001), COPD (HR 1.4, SE 0.1, P<0.001), ESRD (HR 1.8, SE 0.4, P=0.02), and major complication (HR 3.1, SE 0.3, P<.001) to be predictive of 1-year mortality. Smoking history was protective (HR 0.7, SE 0.1, P<.001).

Figure 2.

Kaplan–Meier curves and life tables for 1-year survival after repair of ruptured abdominal aortic aneurysm, stratified by sex.