• endocarditis
• cardiac surgical procedures
• risk factors

In recent years, there has been increasing evidence of sex-related disparities in diagnostic investigations, medical therapies, referrals for invasive care as well as health services, and outcomes across a range of cardiovascular conditions.1–3 While several factors including sex-specific differences in age, physiological differences and comorbidities might contribute to these differences, adjusting for these variables often leaves sex as an independent predictor of treatments as well as outcomes; women are underdiagnosed, under referred and undertreated, and suffer worse outcomes across several cardiovascular conditions.

Relatively little is known about sex differences in clinical profile, surgical referrals and outcomes in infective endocarditis (IE). IE appears to occur in men more commonly than women, with sex differences in the microbial profile, aetiology and comorbidities of patients with IE.4–6 Some of these differences may translate to differences in outcomes, but evidence in this regard has varied.

The decision around both the receipt of surgical intervention and its timing in IE is complex and depends on many factors (figure 1), which include the size of the vegetation, the nature of the infective organism, the presence of paravalvular infection particularly with evidence of heart block or annular/periaortic abscesses, embolic phenomena, presentation with valve dysfunction resulting in symptoms or signs of heart failure, patient age, comorbidities and overall surgical risk, but importantly should not include sex per se.7 It is recommended by international guidelines that decisions around both the indication and timing of surgical intervention should be led by a multispecialty team with involvement of cardiology, cardiac surgery, imaging and infectious diseases.7 Patterns of surgical referral and uptake in men versus women with IE are unknown, and in light of known sex-based disparities in other cardiovascular conditions merit investigation.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Factors associated with infective endocarditis outcomes

In the current issue, Varela et al 8 analysed sex differences in clinical presentation, treatment recommendations, management and prognosis of IE through analysis of a prospective observational national registry (the GAMES registry; ‘Grupo de Apoyo al Manejo de la Endocarditis Infecciosa en Espa a’ or ‘Spanish Collaboration on Endocarditis’) that includes all consecutive patients with a diagnosis of definite IE in Spain. Of the 3451 patients included, a minority (32.0%) were women, who on average were older than men (68.4 vs 64.5 years; p<0.001). Native valves accounted for two-thirds of cases, with significant differences between the sexes (women 66.2%, men 68.7%; p=0.14), with the aortic valve most frequently affected overall, but less commonly in women than in men (48.7% vs 50.6%; p<0.001). The mitral valve was the next most frequently affected valve, more commonly affected in women than in men (48.7% vs 29.5%; p<0.001). Staphylococcus aureus, the most common pathogen, was more common in women than in men (30.1% vs 23.1%; p<0.001).

Among those recommended for surgical treatment by local multidisciplinary panels, a greater proportion of women than men did not undergo surgery (31.3% vs 25.7%). The reasons for surgical turndown despite surgical referral were not captured, or whether there were sex differences and whether such decisions were medically appropriate or a manifestation of sex-based systems bias. After propensity score matching for age and estimated EuroSCORE II surgical risk, women had 26% lower odds of surgical treatment than men (OR 0.74; 95% CI 0.59 to 0.91; p=0.05). The sex-disaggregated results of those who were offered and underwent surgery versus those who did not undergo surgery were not provided.

The observed overall in-hospital mortality was 27.9% and was consistently higher among women than men (32.8% vs 25.7%) regardless of type of valve involved (aortic, mitral, native, prosthetic). After adjustment for clinical covariates associated with mortality, women had 25% greater odds of dying than men (adjusted OR 1.25; 95% CI 1.07 to 1.47; p<0.01). Of note, hospital-level clustering was not accounted for in any of the analyses. Such adjustment is particularly important as referral patterns, surgical volumes and expertise, postsurgical care quality, and socioeconomic demographics that determine health outcomes vary across hospitals. Nonetheless, the present study does provide evidence that among candidates for surgical treatment of IE, women have lower odds of receiving surgical treatment and greater odds of dying in hospital than men, even when receipt of surgical intervention is adjusted for.

The findings of the present study are consistent with prior studies on sex differences in patient risk profile, endocarditis management and outcomes. The association between sex and outcomes has varied across studies. In one prior study,9 female sex, non-streptococcal aetiology and acute kidney failure were associated with a higher risk of cardiogenic shock in those with endocarditis, and among those with cardiogenic shock female sex was associated with a higher risk of in-hospital mortality. A population-based study from Finland demonstrated similar in-hospital and 1-year mortality in men and women following admission with IE, but found 5-year and 10-year mortality to be higher in women compared with men.6 Yet another study reported that, while female sex was associated with a significantly higher 30-day and 1-year mortality, the association did not persist after multivariable adjustment for underlying comorbidities.10

What is clear from several groups is that women undergoing cardiac surgery for valvular heart disease, irrespective of the aetiology (degenerative or infective), fare worse than men. While premorbid conditions such as advanced age, loss of oestrogen as a protective mechanism and diabetes may be responsible for some differences in outcomes, these findings are inconsistent across studies. What is consistent, however, is that women have lower odds of undergoing surgery as demonstrated by Varela et al,8 and that surgery confers better outcomes.11 Biases and delays in referral may contribute to some disparities in receipt of surgery; while data are limited in IE, there is evidence in other conditions that women are referred late in the disease course for invasive care.1 12 These factors may perpetuate sex differences in postsurgical outcomes, conferring higher preoperative surgical risk scores when validated risk calculators such as the Society of Thoracic Surgeons and EuroSCORE II are used, confirming the biases in many operators’ minds that women have higher surgical risk and do worse. This, in turn, may lead to reticence on the part of surgical teams to offer intervention, contributing to a vicious cycle of undertreatment.

Disparities in referral and receipt of surgical intervention, along with differences in aetiology, microbiology and comorbidities, may be responsible for the higher risk of mortality in women than in men with IE. Ultimately, awareness of these issues should prompt a self-evaluation of biases on the part of clinicians such that objective, timely surgical referrals are made and interventions are offered regardless of demographic group. While the biology is not modifiable, the biases and care disparities are.