Author + information
- Received January 19, 2013
- Accepted March 7, 2013
- Published online July 2, 2013.
- A. Carla Zomer, MD, PhD∗,†,‡,
- Raluca Ionescu-Ittu, PhD§,
- Ilonca Vaartjes, PhD†,
- Louise Pilote, MD, MPH, PhD⋮,¶,
- Andrew S. Mackie, MD#,
- Judith Therrien, MD§,
- Maurice M. Langemeijer, MSc‡,
- Diederick E. Grobbee, MD, PhD†,
- Barbara J.M. Mulder, MD, PhD∗,‡ and
- Ariane J. Marelli, MD, MPH§∗ ()
- ∗Department of Cardiology, Academic Medical Center, Amsterdam, the Netherlands
- †Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, the Netherlands
- ‡Interuniversity Cardiology Institute of the Netherlands, Utrecht, the Netherlands
- §McGill Adult Unit for Congenital Heart Disease Excellence, McGill University Health Center, Montreal, Quebec, Canada
- ⋮Division of Clinical Epidemiology, McGill University Research Center, Montreal, Quebec, Canada
- ¶Division of Internal Medicine, McGill University Health Centre, Montreal, Quebec, Canada
- #Division of Pediatric Cardiology, University of Alberta, Alberta, Canada
- ↵∗Reprint requests and correspondence:
Dr. Ariane Marelli, McGill Adult Unit for Congenital Heart Disease (MAUDE unit), McGill University Health Centre, 687 Pine Avenue, Montreal, Quebec H3A 1A1, Canada.
Objectives The study objectives were to analyze sex differences in hospital mortality of adult patients with congenital heart disease (CHD) and to determine the impact of health services associated with pregnancy on outcomes in women.
Background The determinants of sex differences in the demographic distribution of CHD are poorly understood.
Methods The Quebec CHD database and the Dutch CONCOR (CONgenital CORvitia) registry were used to identify patients with CHD aged 18 to 65 years who were hospitalized between 1996 and 2005. Regression analyses were used to compare 30-day in-hospital mortality in men versus women and in women aged 18 to 45 years with versus without a pregnancy history, after adjustment for age, CHD severity, comorbidities, and admission diagnosis.
Results Of 39,776 patients followed for 259,741 patient years, 19,099 patients (48%) had 54,195 admissions (62% among women). In those aged 18 to 45 years, 30-day in-hospital mortality was higher in men compared with women with nonpregnancy admissions (adjusted rate ratio: 1.36; 95% confidence interval: 1.02 to 1.81). The adjusted rate ratio for 30-day in-hospital mortality in women with a pregnancy history compared with those without was 0.49 (95% confidence interval: 0.24 to 0.99). A history of pregnancy was not associated with an overall increase in medical encounters.
Conclusions We demonstrated a protective effect of sex on in-hospital mortality in women with CHD of reproductive age that did not correlate with increased medical surveillance. Future studies need to explore other mechanisms to account for our observations. Understanding the determinants of the sex distribution of adults with CHD is important for our ability to predict demographic changes in the population with CHD.
An increasing number of patients with congenital heart disease (CHD) have reached adulthood because of the improved developments in pediatric and adult cardiology (1). The demographic distribution of the population with CHD is changing, and the number of adults with CHD now exceeds the number of children with CHD (2). There are at least 1 million adults with CHD currently living in the United States and more than 1.2 million in Europe (3,4).
The impact of sex on outcomes and mortality has received little attention in the adult population with CHD relative to other cardiovascular diseases (5–8). Data suggest that women are more prevalent than men in the adult population with CHD (2) and that mortality rates are different in female and male adults and children (9–15). Factors affecting outcome related to sex differences are poorly understood. Marelli et al. (14) recently proposed a conceptual model with potential pathways and factors contributing to the sex differences in the adult population with CHD (Fig. 1). Primary biological pathways, CHD surgical and medical outcomes, and intervening pathways, such as health care behavior, were considered.
We focused our attention on sex differences in hospital mortality in adults with CHD as a possible contributing factor to the sex distribution with female predominance observed in the adult population with CHD. We used health services use as a proxy for health behavior and hypothesized that health services associated with pregnancy may confer a protective effect against adverse outcomes in women.
Our aims were thus 3-fold: 1) to assess differences in hospital admissions in adult male and female patients with CHD; 2) to investigate sex differences in all-cause in-hospital mortality of adults with CHD; and 3) to determine the impact of care associated with pregnancy on hospital mortality in women with CHD.
Quebec CHD Database
In Quebec (Canada), every individual is assigned a unique health care number at birth that is systemically linked to all health-related diagnoses and health services until death. A population-based CHD Quebec database was created by linking two databases to each patient: the physician's services and claims database (Régie de l'Assurance Maladie du Québec [RAMQ]) from 1983 to 2005 and the hospital discharge database (Med-Echo) from 1987 to 2005 (2). Diagnoses conform to the International Classification of Disease, Ninth Revision, which has 24 diagnostic codes to designate CHD lesions. Patients were identified with CHD if they had at least one International Classification of Disease-Ninth Revision diagnostic code of a CHD lesion or a procedure code for a CHD operation billed by selected specialists (2). Patients were assigned one or two CHD diagnoses with the use of a previously defined hierarchical algorithm (16). All information was cross-referenced between outpatient and inpatient data sources. The resulting database includes 71,513 patients with CHD of all ages and contains comprehensive longitudinal demographic, diagnostic, and therapeutic data over the period 1983 to 2005 for each patient.
CONCOR (CONgenital CORvitia) Database
From November 2001, patients with CHD aged 18 years or more have been recruited in tertiary and secondary medical centers and included in the CONCOR Dutch national registry database (17). Clinical data, such as diagnosis, clinical events, and procedures—classified using the European Paediatric Cardiac Code Short List coding scheme (18)—and patient and family history were obtained from medical records. A prespecified hierarchical scheme (19) was used to select the CHD diagnosis with the worst prognosis and assign it as the main CHD diagnosis. After entry, data on major events during follow-up were systematically recorded from the patients' medical letters written by their cardiologist. More than 13,500 patients currently are included from 107 Dutch hospitals, including all eight tertiary referral centers from which 65% of patients originate.
Study population and data collection
The study included all patients with CHD in the Quebec CHD database who were between 18 and 65 years of age, between January 1, 1996, to March 31, 2005 (n = 34,339), and all adult patients with CHD included in CONCOR from January 1, 2002, to December 31, 2005 (n = 6,403). Given that the 30-day hospital mortality rate (our outcome of interest) was similar in the CHD databases of Quebec and the Netherlands, the databases were combined for all multivariable analyses.
For patients in Quebec, data on hospital admissions and in-hospital mortality were obtained from Med-Echo, and data on outpatient visits were obtained from RAMQ. The start of follow-up for Quebec patients was January 1, 1996, or the patient's 18th birthday (if the 18th birthday occurred in the period 1996 to 2005). The end of follow-up was March 31, 2005, the patient's 66th birthday, or death, whichever came first.
For Dutch patients, data on hospital admissions and in-hospital mortality were obtained by merging the CONCOR registry with the Dutch national hospital discharge and national mortality registries. Hospitalizations and deaths occurring between CONCOR inclusion date and December 31, 2005, were comprehensively recorded for 85% of patients in the CONCOR database (n = 5,437); 15% could not be linked because of missing or erroneously registered ZIP codes. The start of follow-up was the date of inclusion; the end of follow-up was December 31, 2005, the patient's 66th birthday, or death.
We conducted observational cohort analyses with designs tailored to each objective. To assess differences in hospital admissions in adult male and female patients with CHD, we compared all-cause hospital admissions during follow-up in the two groups in terms of diagnosis at admission, length of stay, and in-hospital mortality. To investigate sex differences in all cause in-hospital mortality, we compared 30-day in-hospital mortality of the men and women hospitalized during follow-up with any admission diagnosis. If a patient had multiple admissions during follow-up, each admission was included as a separate observation, accounting for within-patient correlation. These analyses were first stratified by age at admission, so that women at reproductive age (18 to 45 years) were singled out, and then replicated including only nonpregnancy admissions. Although we could have performed stratification by different age groups, we chose to analyze the reproductive years separately to be consistent with our objective of analyzing the impact of care associated with pregnancy. To further test our hypothesis that health services associated with pregnancy confer protection to women, we also compared 30-day in-hospital mortality in women at reproductive age (18 to 45 years of age) with and without pregnancy history. To determine the impact of care associated with pregnancy on hospital mortality in women with CHD, we selected for each woman aged 18 to 45 years the most recent nonpregnancy admission during follow-up and documented all cardiac, pregnancy-related, and other outpatient medical encounters in the 5 years before this admission. Cardiac surveillance was then compared between women with and without a pregnancy history. These analyses were restricted to patients from the Quebec CHD database because the CONCOR database does not contain outpatient data.
Hospital admissions were classified on the basis of the admission diagnosis as pregnancy related and nonpregnancy related. Nonpregnancy admissions were further categorized as cardiovascular and noncardiovascular.
In-hospital deaths are accurately reported in both databases. In the Netherlands, the attestation of death must be sent to the community registrar of civil status and the Central Bureau of Statistics. In Quebec, the attestation of death must be sent to the registrar of civil status who then forwards this information to the Quebec Health Insurance Board (20).
Pregnancy history was defined as being admitted for a pregnancy-related diagnosis in the 5 years before the index admission. Outpatient visits in the 5 years before the index admission were measured in the RAMQ Quebec database and classified as cardiac, pregnancy related, and other, on the basis of the specialty of the physician. Age was categorized in two categories with the cutoff value set at 45 years to single out the reproductive age. Additional subgroups within the two categories were created for the purpose of adjustment: 18 to 25 years, 26 to 35 years, and 36 to 45 years for category one and 46 to 55 years and 56 to 65 years for category two. Severe defects were defined on the basis of anatomic diagnosis and included tetralogy of Fallot, truncus arteriosus, endocardial cushion defects, transposition complex, univentricular heart, and hypoplastic left heart syndrome. Septal defects included atrial and ventricular septal defect. Valvular defects included congenital tricuspid, pulmonary, mitral, or aortic valve disease. All other lesions were classified as “other defects.” The Charlson comorbidity index was used to reflect overall health status and the burden of diagnoses other than CHD. Weights (1 to 6) are assigned to myocardial, vascular, pulmonary, neurologic, endocrine, renal, gastrointestinal, hepatic, immune, and malignant conditions (16,21–23). A comorbidity score, using the Charlson comorbidity index, was calculated for every admission on the basis of the main admission diagnosis and three secondary diagnoses.
For the purpose of this study, data from the Quebec CHD database and CONCOR database were aggregated at the group level, because regulations from both countries prohibited any disclosure of individual data. Descriptive statistics included proportions and rates. Differences between groups were tested using the chi-square test. To compare 30-day hospital mortality rates in both databases, the mortality rate ratio (RR) with the corresponding 95% confidence interval (CI) was calculated. Mortality rates were calculated using the number of deaths in the numerator and the number of follow-up in-hospital days in the denominator.
To investigate sex differences in hospital mortality, we assessed the adjusted association between sex (men vs. women and women with a pregnancy history vs. women without) and all-cause 30-day in-hospital mortality. To combine the data from both databases to include in one model, we created in each database 240 categories resulting from the cross-classification of the determinant of interest (sex, 2 categories) and the confounders (age at admission: 5 categories, CHD severity: 4 categories, comorbidity: 3 categories and admission diagnosis: 2 categories). For each cross-classified category, we counted the number of 30-day in-hospital deaths (the outcome of interest) and determined the total number of admission days (maximum 30 days per admission). The Quebec and CONCOR databases were then combined so that the death/admission day counts in each category were summed. Because the assumptions of Poisson regression models are not satisfied, we analyzed these count data with negative binomial regression models from which we report RRs and CIs (24).
To determine the impact of care associated with pregnancy, we assessed the adjusted association between women with and without a pregnancy history and the number of outpatient visits. Again, negative binomial regression models were used from which we report RRs and CIs. Because this analysis was based on the Quebec CHD database only, data do not need to be cross-classified and the unit of analysis was the individual. Rates were calculated using the number of outpatient visits as the numerator and the person-time at risk as the denominator. Patients were not considered to be at risk for outpatient visits during the time spent in-hospital during follow-up. The models were adjusted for age at admission and CHD severity (four categories).
All hypotheses were tested at a two-tailed 0.05 significance level and performed using SPSS version 18.0 (SPSS Inc., Chicago, Illinois).
Study population characteristics
In total, 39,776 adult patients with CHD were included: 34,339 patients from the Quebec CHD database and 5,437 patients from the CONCOR database. There were less women (48% vs. 55%, p < 0.05) and more patients with severe defects (28% vs. 10%, p < 0.05) in the CONCOR database than in the Quebec database. Thirty-day hospital mortality rates did not differ (RR: 1.2; 95% CI: 0.8 to 1.5) between the two databases. Overall, 19,099 patients (48%) were hospitalized during a median follow-up period of 9.3 years (total period of 259,741 patient years), 55% of female and 40% of male patients (Table 1). After excluding pregnancy-related admissions, the hospitalization rate was still higher for women than for men (42% vs. 40%, p < 0.05). More than one of five women were hospitalized for a pregnancy-related diagnosis. However, if we presume that the pregnancy-related admissions were in women aged 18 to 40 years at the start of follow-up and followed for at least 5 years (n = 8,898), one of two women were admitted for a pregnancy-related diagnosis. Thirty-day in-hospital mortality rates in women and men were 2.0 and 2.8 per 1,000 admission days, respectively (Table 1). After excluding pregnancy-related admissions, the 30-day in-hospital mortality rate in women was 2.4 per 1,000 admission days.
Of 54,195 admissions, 33,592 (62%) were in women, of which 9,359 admissions (28%) were pregnancy-related. Among the 24,233 nonpregnancy admissions in women, 5,571 (23%) were for cardiovascular diagnoses versus 44% (9,065 of 20,603) in men. The most relevant cardiovascular admission diagnoses in women and men with severe defects are shown in Figure 2. The most frequent cardiovascular admission diagnosis in both women and men with severe defects was arrhythmia/conduction disturbances (Figs. 2A and 2B). There were no large differences in cardiovascular diagnoses for women and men with severe defects who were age 18 to 45 years. In the age group 46 to 65 years, women with severe defects were significantly more often admitted for pacemaker or prosthetic implant-related diagnoses (7% vs. 2%) and valve disease (8% vs. 3%), and less often admitted for coronary artery disease (CAD) (7% vs. 15%) compared with men. The most common cardiovascular admission diagnosis for patients with disease not classified as severe CHD was valve disease. Female patients with disease not classified as severe CHD were overall significantly more often admitted for pulmonary hypertension and less often for CAD and aortic complications, including aneurysm, rupture, and bleeding. For both severe and all other CHD defects, CAD was an important admission diagnosis in patients aged 46 to 65 years.
The three most common noncardiovascular/pregnancy diagnoses in the total study population were genitourinary disease (13%), gastrointestinal disease (8%), and neoplasms (7%) in women and gastrointestinal disease (9%), external injury (6%), and the musculoskeletal system (5%) in men.
The effect of sex on 30-day in-hospital mortality is presented in Figure 3. Although the adjusted 30-day in-hospital mortality for all admissions was not significantly different in men versus women age 18 to 65 years (Fig. 3A), mortality was higher in men than in women in patients age 18 to 45 years (adjusted RR: 1.58; 95% CI: 1.08 to 2.30). Likewise, when excluding pregnancy admissions, the adjusted RR for 30-day in-hospital mortality remained higher in men versus women age 18 to 45 years (adjusted RR: 1.36; 95% CI: 1.02 to 1.81). When the two databases were analyzed separately, in-hospital mortality was higher for men age 18 to 45 years compared with women of the same age and follow-up in both databases. When only nonpregnancy admissions were considered in those 18 to 45 years old (Fig. 3B), a history of pregnancy conferred a protective effect even after adjustment for age at admission, severity of defect, comorbidity, and admission diagnosis (adjusted RR for women with a pregnancy history vs. without = 0.49; 95% CI: 0.24 to 0.99). Finally, in those age 18 to 45 years and admitted for nonpregnancy diagnoses only, risk of death was higher in men compared with women (adjusted RR: 1.29), but the CI just crosses 1.0 (95% CI: 0.92 to 1.68). Table 2 presents the characteristics of nonpregnancy admissions of women with and without a pregnancy history and men age 18 to 45 years.
In total, 195 patients (90 women and 105 men) age between 18 and 45 years died in hospital within 30 days of a nonpregnancy admission. The median age at death was 35.4 years (interquartile range: 25.8 to 41.2) for women and 32.8 years (interquartile range: 24.5 to 39.1) for men. The direct cause of death was cardiovascular, noncardiovascular, and unknown in 50%, 43%, and 7% of the women, respectively, and in 47%, 41%, and 12% of the men, respectively (Fig. 4). Women died less often from arrhythmia but more often from heart failure. Within the noncardiovascular causes of death, women died more often of neoplasm (13.3% vs. 2.9%) and less often of infection/sepsis (4.4% vs. 9.5%) and neurologic diseases (0% vs. 10.5%) compared with men.
After excluding the pregnancy-related visits, women with a pregnancy history visited the outpatient clinic equally as often as those without a pregnancy history, after adjustment for age and CHD severity (adjusted RR: 0.97; 95% CI: 0.93 to 1.02). Although women with a pregnancy history visited the cardiac specialist less often, visits to the general practitioner occurred more often than in women without a pregnancy history (adjusted RR: 0.89; 95% CI: 0.80 to 0.99 and 1.14; 95% CI: 1.08 to 1.21, respectively).
Sex differences in the population with CHD with respect to demographics or outcomes have received relatively little attention. In Quebec, in more than 45,000 adults with CHD, women accounted for 57% of patients, a proportion that was significantly higher than the predominance of women observed in the general population (2). Consistent with these findings, using death registry data in 11,040 adults in the United States, the Centers for Disease Control and Prevention demonstrated lower mortality rates in women with CHD compared with men (25). Potential causes of a shift in demographics toward a predominance of women in the adult CHD population are shown in Figure 1 (14). These include a difference in sex distribution of CHD at birth, milder lesions in female subjects born with CHD, or differences in mortality related to CHD surgical or medical outcomes (14,26). In a previous study using Healthcare Cost and Utilization Project data in the United States, more than 33,000 admissions for CHD in children in the United States were analyzed (14). Although 55% of all children undergoing surgery were male, and male patients were more likely to have high-risk procedures, when female infants did undergo high-risk surgery, they had a higher in-hospital mortality (14).
In this study, we turn our attention to nonsurgical in-hospital mortality in adults with CHD. Two large population-based adult CHD databases with comprehensive health services data from two countries were used to analyze sex differences admissions and mortality, and to study the impact of health services associated with pregnancy on hospital outcomes. Overall, 48% of 39,776 patients were admitted to the hospital during a median follow-up of 9.3 years. Although women were more often admitted to the hospital than men, even after exclusion of pregnancy-related hospitalizations, for those age 18 to 45 years the adjusted 30-day in-hospital mortality was lower in women than in men and lowest in women with a pregnancy history even after adjustment for age, disease severity, admission diagnosis, and comorbidities. The fact that mortality in women age 46 to 65 years was no longer lower than in men might be explained in part by the loss of the cardioprotective estrogen during the menopausal transition (27,28). Consistent with these findings, men were more often admitted for cardiovascular disease compared with women. Higher cardiovascular complication rates in male adults with CHD also could contribute to the latter finding (29). Although we expected that health surveillance in women during the reproductive years may confer protection by increased rates of health services use, this is not what we found, suggesting that other mechanisms might be implicated.
Sex disparities have been documented extensively in cardiovascular disease (5–8). In non-CHD, such as acute myocardial infarction, CAD, and atrial fibrillation, mortality rates differ between men and women (5–8). It is likely that these observations like ours are at least partly explained by biological drivers and by health care behavior on the part of patients themselves, as well as by a differential response of health providers toward a predominantly male population with noncongenital cardiovascular disease. Data on sex differences in adults with CHD are scant, although this question has received more attention recently (13,29). Population-based studies from the Netherlands showed that male adult patients with CHD had a higher risk of having endocarditis and arrhythmias, of undergoing surgery in adulthood, and of death (13,29,30).
Health care use in adults with CHD has been studied by our research groups and others (12,16,31,32). However, this is the largest study demonstrating that women with CHD were more often admitted, even when pregnancy-related admissions were excluded, and more frequently admitted for noncardiovascular disease than men. This latter could be partly explained by the fact that women were more frequently admitted for genitourinary diseases (e.g., breast disease, menorrhagia, urinary tract infection) compared with men. Billett et al. (12) found a higher hospitalization rate in male patients than in female patients of all ages, but they only studied admissions with CHD as the primary diagnosis, whereas we included all-cause admissions and focused only on adult patients.
In this study, one of five women were admitted for a pregnancy-related admission. By focusing only on women age 18 to 40 years at the start of follow-up, 1 of 2 women were admitted for a pregnancy-related diagnosis. This number is similar to that found in a previous study of ours (33). In that study, women with a mean age of 40 years were asked if they had children and more than 50% answered yes (A. Zomer, unpublished data, June 2012).
We found a relatively large proportion of noncardiovascular causes of death in comparison with what is known in the literature (11,34). Moreover, arrhythmia and heart failure were not as often reported as the cause of death as in previous studies (11,34). However, those studies focused on overall mortality and not only on hospital mortality as done in this study.
To test our hypothesis that health services associated with pregnancy might be protective, we analyzed health services use in women who became pregnant compared with those who did not, accounting for disease severity and age. In this study, we could not show that women with a pregnancy history were more often seen in the outpatient clinic by specialists other than the obstetrician. It seems that the self-selection of relatively healthy women becoming pregnant has a stronger effect on hospital mortality than the possible protective effect of health services associated with pregnancy. Moreover, the physiological changes women go through during pregnancy may lead to earlier symptomatic disease and therefore early treatment, which could also contribute to better outcomes for pregnant women with CHD.
One might speculate that the female predominance in the adult population with CHD and the better outcomes for women, especially for those with a history of pregnancy, may result in increased transmission rates of milder forms of CHD, thus diluting severe CHD from one generation to the next. This could be considered consistent with biology's natural intelligence.
Our findings need to be interpreted in light of the limitations of the study design. The recruitment of patients with CHD was different in the two databases that were combined in this study. Thus, the characteristics (e.g., severity of defect) of patients were different. However, we adjusted for those characteristics in our analyses, and therefore bias is assumed to be minimal. Results were largely based on the Quebec database because it was larger and had longer follow-up than the CONCOR database. Nevertheless, we think that the CONCOR database had additional value, especially in terms of generalizability. We used administrative databases, which are prone to misclassification of diagnoses, to collect information on hospital admissions and mortality. However, this information bias is expected to be nondifferential between men and women, minimizing the impact on the interpretation of our results.
We were unable to identify women who delivered at home. However, in both countries all women with CHD are advised to give birth at the hospital. Therefore, we do not expect this potential confounder to significantly affect our results. Even so, because the reported home delivery rates are expected to be similar in both jurisdictions, we would expect this bias to be nondifferential. Access to clinical information was not available, but clinical parameters, such as type of defect and comorbidity, were comparable between women with and without a pregnancy history. However, there could be unmeasured confounding related to within-CHD severity category variation in functional status. Although we accounted for disease burden by adjusting for age, CHD severity and comorbidity, and admission diagnosis in our mortality analysis, there is likely residual and unmeasured confounding that we could not capture accounting for the protective effect that the ability to become pregnant confers in terms of mortality. Within-group variation in functional status cannot be measured with our data.
For patients with CHD age 18 to 45 years, 30-day in-hospital mortality for nonpregnancy admissions was lower in women than in men and lowest in women with a pregnancy history. However, the low hospital mortality in women with a pregnancy history did not correlate with higher cardiac surveillance as demonstrated by a higher use of health services. Our study suggests that other biological, primary, or intervening pathways are implicated. Understanding the determinants of the sex distribution of adults with CHD is important in that it informs our ability to predict demographic changes in the population with CHD.
Drs. Marelli and Pilote are funded by the Fond de la Recherche en Santé du Québec. Dr. Marelli is funded by the Heart and Stroke Foundation of Canada and the Canadian Institute of Health Research. All other authors have reported they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- coronary artery disease
- congenital heart disease
- confidence interval
- Régie de l'Assurance Maladie du Québec
- rate ratio
- Received January 19, 2013.
- Accepted March 7, 2013.
- American College of Cardiology Foundation
- Warnes C.A.
- Marelli A.J.,
- Mackie A.S.,
- Ionescu-Ittu R.,
- Rahme E.,
- Pilote L.
- Pilote L.,
- Dasgupta K.,
- Guru V.,
- et al.
- Daly C.,
- Clemens F.,
- Lopez Sendon J.L.,
- et al.
- Boneva R.S.,
- Botto L.D.,
- Moore C.A.,
- Yang Q.,
- Correa A.,
- Erickson J.D.
- Verheugt C.L.,
- Uiterwaal C.S.,
- van der Velde E.T.,
- et al.
- Billett J.,
- Majeed A.,
- Gatzoulis M.,
- Cowie M.
- Zomer A.C.,
- Verheugt C.L.,
- Vaartjes I.,
- et al.
- Marelli A.,
- Gauvreau K.,
- Landzberg M.,
- Jenkins K.
- Chang R.K.,
- Chen A.Y.,
- Klitzner T.S.
- van der Velde E.T.,
- Vriend J.W.,
- Mannens M.M.,
- Uiterwaal C.S.,
- Brand R.,
- Mulder B.J.
- Warnes C.A.,
- Liberthson R.,
- Danielson G.K.,
- et al.
- Khairy P.,
- Ionescu-Ittu R.,
- Mackie A.S.,
- Abrahamowicz M.,
- Pilote L.,
- Marelli A.J.
- Gilboa S.M.,
- Salemi J.L.,
- Nembhard W.N.,
- Fixler D.E.,
- Correa A.
- Bhupathy P.,
- Haines C.D.,
- Leinwand L.A.
- Donaldson C.,
- Eder S.,
- Baker C.,
- et al.
- Verheugt C.L.,
- Uiterwaal C.S.,
- van der Velde E.T.,
- et al.
- Verheugt C.L.,
- Uiterwaal C.S.,
- van der Velde E.T.,
- et al.