Sex Differences in Native-Valve Infective Endocarditis in a Single Tertiary-Care Hospital




The aim of this study was to assess whether the clinical characteristics, management, and outcomes of infective endocarditis differ in women and men through a prospective observational cohort study at a single tertiary care teaching hospital. From January 2000 to December 2008, 271 new cases of infective endocarditis were diagnosed (183 in men, 88 in women) according to modified Duke criteria, and patients were followed for 1 year. Women were older than men (mean age 63 ± 16 vs 58 ± 18 years, p = 0.006); more women were taking immunosuppressants (14% vs 3%, p = 0.006) and had mitral valve involvement (52% vs 36%, p = 0.02). However, more men had human immunodeficiency virus infection than women. There were no gender differences in Charlson index, regurgitation severity, culprit pathogens, or major complications. When surgery was indicated, women were less likely to undergo the procedure (26% vs 47%, relative risk [RR] 0.4, 95% confidence interval [CI] 0.2 to 0.7), p = 0.001). Mortality tended to be higher in women in the hospital (32% vs 23%, RR 1.58, 95% CI 1 to 2.5, p = 0.05) and at 1 year (38% vs 26%, RR 1.7, 95% CI 1.0 to 2.9, p = 0.04). Surgical treatment was a protective factor against death in the hospital (RR 0.18, 95% CI 0.04 to 0.77, p = 0.02) and at 1 year (RR 0.12, 95% CI 0.03 to 0.48, p = 0.03) after adjustment for age, gender, Charlson index, infection by Staphylococcus aureus , severity at presentation, heart failure, acute renal failure, stroke, and the ejection fraction. In conclusion, women with infective endocarditis were slightly older than men but showed similar co-morbidities. Women underwent surgery less frequently and consequently had worse prognosis than men.


Differences in clinical characteristics, treatment choices, and outcomes between men and women continue to be a subject of debate in the published research. Previous studies have shown that infective endocarditis (IE) is more frequent in men than in women, and the possibility that estrogens may protect women against endothelial damage has been suggested. Women, however, have usually been found to have a higher incidence of co-morbid conditions, such as diabetes mellitus and renal failure, and this can contribute to worse outcomes. Some investigators have suggested that a gender-based treatment bias may explain why women have significantly different management and outcomes than men in other cardiovascular diseases. Several studies have documented less aggressive management of such diseases in women, a difference that is especially true in coronary disease. However, few studies have looked at the impact of gender in IE. Recently, Aksoy et al reported that preexisting and coexisting conditions were the most important determinants of the different gender-related outcomes in IE, but it is possible that bias in management could also influence worse prognosis in women. The aim of our study was to determine gender-related differences in clinical presentation, management, and early and 1-year mortality in patients admitted to the hospital with IE.


Methods


This was a prospective observational cohort study at a single tertiary care hospital. The center is a 1,250-bed teaching hospital that includes all major medical and surgical departments and is a referral center for cardiac surgery. All consecutive patients aged ≥18 years diagnosed with confirmed or possible IE according to the modified Duke classification who were treated from January 2000 to December 2008 were enrolled. Patients were identified by the cardiology or infectious diseases department on the basis of positive follow-up culture, echocardiographic disturbances suggesting IE, or referral from other facilities.


All demographic, clinical, diagnostic, management, and outcome variables were recorded prospectively during the admission period. Follow-up was clinical and echocardiographic at 1 year after discharge. Charts were reviewed to record all causes of death.


IE was defined according to the modified Duke classification. The Charlson index was used at hospital admission to stratify patients according to overall morbidity. This co-morbidity index predicts 1-year mortality for a patient who may have a range of conditions, such as heart disease, acquired immune deficiency syndrome, or cancer. IE was defined as associated with health care according to a published definition.


Surgical indications were established according to the guidelines of the European Society of Cardiology and the American Heart Association and American College of Cardiology. All surgical interventions were performed by the same team during the study period. When surgery was indicated but not performed, the main reason for the decision was recorded. Only inpatient cardiac surgery was considered.


IE complications were defined as (1) persistent fever (present 7 days after the start of treatment), (2) heart failure, (3) intracardiac abscess (diagnosed by echocardiography or during surgery), (4) new conduction abnormality, (5) stroke, (6) systemic embolism other than stroke, and (7) acute renal failure (defined as a 50% increase from the baseline creatinine concentration).


In-hospital mortality was defined as death from any cause during hospitalization. One-year cumulative mortality was defined as death from any cause within the year after the diagnosis of IE.


Patients were assessed in the outpatient clinic on days 30, 90, and 365 after hospital discharge. Blood samples were obtained for culture 2 days after completion of antimicrobial treatment and 30 and 90 days after hospital discharge. Echocardiographic examinations were performed during admission, at the end of treatment, and at least once again during the first year of follow-up.


Continuous variables are expressed as mean ± SD. Categorical variables are described using absolute numbers and relative frequencies (percentages) of patients in each category. The distribution of categorical variables was compared using the chi-square test or Fisher’s exact test as appropriate; Student’s t test was used to compare continuous variables. A 2-tailed p value <0.05 was considered significant. Kaplan-Meier actuarial analysis was used to determine the probability of mortality during follow-up, and the resulting survival curves were compared using the Mantel-Cox log-rank test. For the multivariate analysis, the effects on in-hospital and 1-year mortality of variables that were determined to be clinically important and/or statistically significant in the univariate analysis (p <0.10) were analyzed by logistic regression. Vegetation size ≤10 mm was not included in the multivariate analyses, because this variable was missing for 170 patients. When models were constructed for the subset of patients who did have this variable recorded, no changes were seen in the levels of significance of variables accounting for the gender-specific differences in either in-hospital or 1-year mortality. Statistical analyses were performed using SPSS version 15.0 (SPSS, Inc., Chicago, Illinois).




Results


A total of 271 episodes of IE in 264 patients were studied; 183 (67.5%) occurred in men and 88 (32.5%) in women. The mean age of patients was 57 ± 18 years. Differences in baseline clinical characteristics, echocardiographic findings, and clinical course are listed in Table 1 .



Table 1

Differences in clinical and echocardiographic characteristics of infective endocarditis between men and women










































































































































































































































































Variable Overall Women Men p Value
(n = 271) (n = 88) (n = 183)
Baseline characteristics
Age (years) 57 ± 18 63 ± 16 58 ± 18 0.006
Co-morbidities
Type 2 diabetes mellitus 50 (18%) 18 (20%) 32 (18%) 0.67
Hemodialysis 15 (6%) 9 (10%) 6 (3%) 0.05
HIV infection 20 (7%) 1 (1%) 19 (11%) 0.01
Cancer 38 (14%) 12 (14%) 26 (14%) 0.9
Immunosuppression therapy 18 (7%) 12 (14%) 6 (3%) 0.006
General condition
Charlson index 2.02 ± 2.6 1.80 ± 1.85 2.2 ± 2.46 0.19
Affected valve
Aortic 112 (41%) 27 (31%) 85 (46%) 0.02
Mitral 111 (41%) 46 (52%) 65 (36%) 0.02
Tricuspid, pulmonary 48 (18%) 15 (17%) 33 (18%) 0.8
Culprit pathogens 0.93
Staphylococcus aureus 74 (27%) 23 (26%) 51 (28%)
CoNS 19 (7%) 7 (8%) 12 (7%)
Enterococci 28 (10%) 11 (13%) 17 (9%)
Streptococcus viridians 62 (23%) 19 (22%) 43 (24%)
Negative culture 11 (4%) 3 (3%) 8 (4%)
Other 77 (28%) 25 (28%) 52 (28%)
Health care associated 72 (22%) 26 (24.1) 46 (21%) 0.5
Major complications 227 (84%) 74 (84%) 153 (84%) 1
Heart failure 128 (47%) 41 (47%) 87 (48%) 0.98
Acute renal failure 76 (28%) 23 (26%) 53 (29%) 0.7
Stroke 50 (19%) 10 (11%) 40 (21%) 0.05
Peripheral or visceral embolism 98 (36%) 34 (39%) 64 (35%) 0.6
Disturbance of conduction 38 (14%) 15 (17%) 23 (12%) 0.4
Provisional pacemaker (in patients with conduction disturbances) 6 (16%) 4 (27%) 2 (9%) 0.1
Definitive pacemaker (in patients with conduction disturbances) 8 (21%) 1 (7%) 7 (30%) 0.1
Echocardiographic findings
Underlying valve disease 61 (19%) 26 (24%) 35 (16%) 0.09
Moderate to severe aortic regurgitation (when aortic valve was affected ) 83 (74%) 20 (74%) 63 (74%) 1.0
Moderate to severe mitral regurgitation (when mitral valve was affected) 75 (68%) 30 (65%) 45 (69%) 0.81
Ejection fraction (%) 60.5 ± 10.3 62.9 ± 8.1 59.5 ± 10.9 0.52
Vegetation 237 (88%) 77 (87%) 160 (87%) 1.0
Vegetation diameter (mm) 12 ± 9 12.3 ± 12 11.6 ± 8 0.53
Paravalvular complications 105 (39%) 23 (26%) 82 (45%) 0.003
Valve perforation 46 (17%) 7 (8%) 39 (21%) 0.006
Abscess 28 (10%) 8 (9%) 20 (11%) 0.41
Valve rupture 33 (12%) 7 (8%) 26 (14%) 0.17
Fistulae 7 (3%) 2 (2%) 5 (2.7%) 1.0

Data are expressed as mean ± SD or as number (percentage).

CoNS = coagulase-negative staphylococci; HIV = human immunodeficiency virus.

Results were compared using the chi-square test, with the exception of HIV infection, pacemaker use (provisional and definitive), and presentation of fistulae, which were compared using Fisher’s exact test.


Percentages are of a subgroup of 38 patients.


Percentages are of a subgroup of 112 patients.



Women were older than men, and more women than men were taking immunosuppressants, although more men had human immunodeficiency virus infections than women. Charlson index on admission, rate of nosocomial IE, culprit pathogens, and rate of negative culture were similar in men and women. There were no differences in underlying cardiac disease. Infection more frequently involved the mitral valve in women than in men. The frequency of use of transesophageal echocardiography was similar in men and women (p = 0.46). Men were more likely to have paravalvular complications due to valve perforation.


Major complications were common: 227 patients (83.8%) experienced ≥1 complication during hospitalization. The incidences of heart failure, peripheral embolism, acute renal failure, stroke, persistent fever, and conduction disturbance were similar between women and men ( Table 1 ).


Conservative nonsurgical treatment was considered the first-choice management option for 141 patients (52%). In the remaining patients, surgery was indicated because of the presence of complications. Cardiac valve surgery was less often indicated and less often performed in women than in men (26% vs 47%, relative risk [RR] 0.4, 95% confidence interval [CI] 0.2 to 0.7, p = 0.001), even after adjustment for age (RR 0.98, 95% CI 0.9 to 1.0, p = 0.10), severe mitral regurgitation (RR 1.2, 95% CI 0.54 to 2.58, p = 0.70), persistent fever (RR 1.5, 95% CI 0.5 to 4.4, p = 0.40), ejection fraction ≤50% (RR 1.4, 95% CI 0.1 to 18.2, p = 0.04), Charlson index (RR 0.2, 95% CI 0.08 to 0.5, p = 0.001), periannular complications (RR 3.1, 95% CI 1.2 to 7.7, p = 0.01), severe aortic regurgitation (RR 3.52, 95% CI 1.60 to 7.57, p <0.001), and heart failure (RR 9.74, 95% CI 4.48 to 21.18, p <0.0001). There were no significant gender differences in reasons for declining surgery, and operative mortality was similar in men and women ( Table 2 ).



Table 2

Gender differences in treatment of infective endocarditis and overall outcomes






























































































































































































Variable Overall Women Men p Value
(n = 271) (n = 88) (n = 183)
Treatment
Medical treatment
First choice 141 (52%) 57 (64%) 84 (46%) 0.003
Because surgery was ruled out 19 (7%) 7 (8%) 12 (7%) 0.8
Surgery
Indicated 130 (48%) 32 (36%) 98 (54%) 0.008
Performed 111 (41%) 24 (26%) 87 (47%) 0.001
Surgical delay (days) 10.9 ± 9 11.9 ± 8.6 10.74 ± 11 0.6
Reasons for not performing surgery
Elevated surgical risk 13 (68%) 6 (86%) 7 (58%) 0.33
Heart failure 2 (11%) 1 (14%) 1 (8%) 0.68
Hemodynamic instability 3 (16%) 2 (29%) 1 (8%) 0.52
Death 2 (11%) 1 (14%) 1 (8%) 0.68
Mortality
Overall
In hospital 70 (26%) 28 (32%) 42 (23%) 0.05
1 year 80 (29%) 34 (38%) 47 (26%) 0.04
Medical treatment
Overall
In hospital 47 (29%) 21 (34%) 25 (27%) 0.29
1 year 57 (36%) 27 (40%) 30 (31%) 0.1
First choice
In hospital 37 (26%) 17 (30%) 19 (18%) 0.06
1 year 47 (33%) 23 (39%) 24 (29%) 0.09
Because surgery was ruled out §
In hospital 10 (53%) 4 (57%) 6 (50%) 0.7
1 year 10 (53%) 4 (57%) 6 (50%) 0.7
Surgery performed
In hospital 24 (22%) 7 (30%) 17 (19%) 0.2
1 year 24 (22%) 7 (30%) 17 (19%) 0.2

Data are expressed as mean ± SD or as number (percentage).

Results were compared using the chi-square test, with the exception of heart failure, hemodynamic instability, and death (under reasons for not performing surgery) and decision to provide medical treatment because surgery was ruled out, which were compared using Fisher’s exact test.


Percentages were calculated for a subgroup of 160 patients.


Percentages were calculated for a subgroup of 141 patients.


§ Cases in which surgery was ruled out for medical reasons or life-threatening disease. Percentages were calculated for a subgroup of 19 patients.


Percentages were calculated for a subgroup of 111 patients.



Seventy patients (26%) died before hospital discharge, and this mortality rate tended to be higher in women than in men (32% vs 23%, p = 0.05). At 1 year, 80 patients (29%) had died; this mortality rate was significantly higher in women (38% vs 26%, p = 0.05). The mean survival rates were 81 ± 4% in women and 85 ± 3% in men at 30 days and 62 ± 5% in women and 74 ± 3% in men at 1 year (log-rank p = 0.05; Figure 1 ). Causes of death are shown in Figure 2 . More women than men died from congestive heart failure, which was the main cause of death in women.




Figure 1


Differences in cumulative survival in men and women.



Figure 2


Comparison of causes of death between men and women.


In the univariate analysis for the entire cohort, female gender approached significance as a predictor of in-hospital mortality and was clearly an independent predictor of 1-year mortality. However, the multivariate logistic regression model did not indicate that gender was a predictor of in-hospital mortality after adjustment for age, co-morbid conditions, infection by Staphylococcus aureus or coagulase-negative staphylococci, persistent fever, underlying valve disease, presentation severity, the ejection fraction, and major complications. The presence of heart failure, renal failure, or stroke was associated with in-hospital mortality ( Table 3 ).



Table 3

Univariate and multivariate analysis of predictors of in-hospital mortality for the entire cohort of patients with infective endocarditis























































































































































































Variable Univariate Analysis Multivariate Analysis
RR 95% CI p Value RR 95% CI p Value
Female gender 1.58 1.00–2.52 0.05 0.85 0.22–3.20 0.8
Age ≥70 years 1.44 0.82–2.50 0.20 1.08 0.30–3.90 0.7
Type 2 diabetes mellitus 1.29 0.65–2.5 0.47
Hemodialysis 2.0 0.68–5.80 0.50
Intravenous drug use 0.88 0.29–2.30 0.90
HIV infection 0.95 0.30–2.70 0.90
Cancer 1.6 0.77–3.3 0.23
Charlson index ≥3 2.53 1.43–4.49 0.02 2.83 0.65–12.23 0.16
Staphylococcus aureus or CoNS 2.46 1.40–4.30 0.002 0.50 0.12–2.01 0.33
Persistent fever 1.8 0.90–3.60 0.06 4.70 0.84–19.78 0.08
Underlying valve disease 1.85 1.00–3.30 0.04 0.72 0.17–2.90 0.65
Moderate to severe mitral regurgitation 1.4 0.77–2.53 0.20 1.34 0.41–4.40 0.6
Moderate to severe aortic regurgitation 0.72 0.39–1.33 0.30
Ejection fraction <50% 2.33 0.40–13.2 0.30 0.32 0.04–2.37 0.4
Vegetation size ≥10 mm 2.2 1.20–4.90 0.03
Heart failure 3.6 2.01–6.46 <0.001 19.5 3.51–108.2 0.001
Acute renal failure 5.57 3.08–10.07 <0.001 3.69 1.08–12.6 0.03
Stroke 5.42 2.82–10.40 <0.001 5.50 1.20–25.05 0.02
Embolism 0.82 0.40–1.40 0.50
Paravalvular complications 0.98 0.52–1.60 0.77
Surgery performed 0.65 0.37–1.15 0.15 0.18 0.04–0.77 0.02

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Dec 22, 2016 | Posted by in CARDIOLOGY | Comments Off on Sex Differences in Native-Valve Infective Endocarditis in a Single Tertiary-Care Hospital

Full access? Get Clinical Tree

Get Clinical Tree app for offline access