|
|
||||||||
Ann Thorac Surg 2006;81:1637-1643
© 2006 The Society of Thoracic Surgeons
a Department of Thoracic and Cardiovascular Surgery, Jean-Minjoz Hospital, Besançon, France
b Clinical and Biological Research Center, Saint-Jacques Hospital, Besançon, France
c Department of Biostatistics and Epidemiology, University of Franche-Comté Medical School, Besançon, France
Accepted for publication December 1, 2005.
* Address correspondence to Dr Falcoz, Department of Thoracic and Cardiovascular Surgery, Hôpital Jean-Minjoz, Boulevard Fleming, 25000 Besançon, France (Email: pierre-emmanuel.falcoz{at}wanadoo.fr).
| Abstract |
|---|
|
|
|---|
METHODS: From July 2000 to July 2002, 590 elective patients were included in this study. Baseline and follow-up QOL surveys were obtained for 439 patients (307 males and 132 females). The QOL scores were compared by gender, by analysis of variance, and by the Student t test. Factors influencing two-year cardiac functional status were determined by logistic regression.
RESULTS: The comparison of baseline and follow-up scores showed a significant improvement (a sharp increase between baseline and year one, then stabilization) in all dimensions of the SF36, two years after surgery in all patients. However, QOL was significantly lower in women than in men in all but two dimensions; at baseline and during follow-up. When compared with the normal population, men and women over 75 had a similar QOL. The best independent predictive factor of two-year cardiac functional status in women was the physical component summary score and in men, the mental component summary score.
CONCLUSIONS: The benefit of open heart surgery at two-year follow-up is equivalent in both genders in terms of QOL, although women had lower baseline QOL scores.
| Introduction |
|---|
|
|
|---|
Two previous articles have been published by our institution using the SF36 questionnaire in the field of cardiac surgery and QOL [9, 12]. To the best of our knowledge, a gender analysis after open heart surgery, at two-year follow-up with QOL as the main criterion, assessed by the SF36 questionnaire, has never been reported in the literature. Thus, we felt it would be worthwhile to design a specific study dealing with gender differences in QOL after open heart surgery. The aim of this prospective study, based on the iterative completion of the SF36 questionnaire (proposed before and at years 1 and 2) was twofold: to evaluate the changes in QOL scores (over time by gender and in comparison with scores from a normal European population) and to determine whether there were gender differences in two-year cardiac functional status in this sample of patients.
| Material and Methods |
|---|
|
|
|---|
The SF36 is a self-administered 36-item tool which covers eight dimensions of health, including limitations in physical functioning, usual role activities, social functioning related to health problems, and vitality. It also includes a global evaluation of health. Each dimension is scored on a scale from 0 to 100, with higher scores indicating better health. The number of possible responses per item varies from 2 to 6. Two summary scores are also calculated to summarize the patient's physical and mental state of health [13]: a physical component summary score and a mental component summary score. The SF36 has received wide validation in English [14, 15]. The French version, used here, was adapted by forward and back translation, iterative revision, and consensus by experts [16].
The assessment of angina pectoris and dyspnea by self-administered questionnaire, given the day before open heart surgery and at 1 and 2 years, was considered valid in view of the very satisfactory agreement between the coding of the patient and medical coding (New York Heart Association [NYHA] and Canadian Cardiovascular Society classification [CCS]) [17]: kappa = 0.816 for angina pectoris and kappa = 0.768 for dyspnea [9].
The variables recorded were the following: sociodemographics (age, sex, family situation, level of study, area of residence); angina pectoris status according to the CCS; dyspnea class according to the NYHA classification (class II was divided into two subgroups: II-mild for patients not troubled by shortness of breath when walking up a slight hill at a normal pace and II-severe for patients who had to stop for breath when walking up a slight hill at their own pace); ejection fraction; left ventricular wall motion; surgical preoperative risk-estimation scores with the EuroSCORE [18]; comorbid diseases; type of heart operation; and operative complications.
Statistical Analyses
After having described the sample and its main characteristics, we explored the variations in QOL subsequent to open heart surgery. Qualitative variables were expressed in percentage and quantitative variables and QOL scores as mean ± standard deviation. All tests were two-sided. The SF36 scoring rules were applied for the questionnaire [19].
The preoperative and postoperative (year-1 and year-2) scores for each QOL section, and the Physical and Mental Component Summary score scales were compared over time and by gender, using an analysis of variance. The eight QOL dimensions of the SF36 were also compared with those of a normal European population, adjusted on age and gender [20], using the Student t test.
To assess two-year cardiac functional status, patients were divided into two groups: "satisfactory" when the classification of angina pectoris was I and dyspnea was I or II-mild, and "unsatisfactory" for all other patients. For each dimension of preoperative QOL, we transformed the continuous scale into several classes. The choice of the most relevant classes, with regard to two-year cardiac functional status, was determined by receiver operating characteristic (ROC) curves, which revealed the discriminate cut points [21]. We performed logistic regression, adjusted on the preoperative status of angina pectoris and dyspnea, first by univariate analysis and then by multivariate analysis, to model two-year cardiac functional status. Sociodemographic, clinical, and QOL dimension variables with a level of significance equal to 0.30 or less in the univariate analysis were included in the multivariate model, which was analyzed by gender, using a stepwise logistic regression.
Data analysis was anonymous. All statistical analyses were performed with SAS software, version 8.02 (SAS Institute Inc, Cary, NC).
| Results |
|---|
|
|
|---|
The main characteristics of the study population are summarized in Table 1. The sample included 307 men (70%), aged 33 to 88 years (mean 66.0; SD 9.5) and 132 women (30%), aged 14 to 84 years (mean 67.0; SD 12). The subdivision of class II dyspnea was as follows: 131 (30%) patients (108 men and 23 women) in class II-mild and 119 (27%) patients (71 men and 48 women) in class II-severe. Beating heart surgery was performed in 40 patients, 27 men and 13 women; respectively, 17% and 27% of the coronary artery bypass grafts performed in men and women (p = 0.14). Compared with men, more than twice as many women lived alone: 50 women (37.9%) versus 51 men (16.6%), p < 0.0001. In terms of preoperative scores, the patients were divided as follows: 207 men (67.4%) and 80 women (60.6%) were EuroSCORE A or B; 100 men (32.6%) and 52 women (39.4%) were C or D (p = 0.18).
|
|
|
Univariate analysis for men showed the following variables to be statistically linked to the two-year status: 6 sociodemographic and clinical variables (family situation, place of residence, level of study, angina pectoris, comorbid diseases, and postoperative course) and all QOL dimensions (eight baseline dimensions and the two summary scores). The same analysis in women showed five sociodemographic and clinical variables (dyspnea, comorbid diseases, left ventricular wall motion, EuroSCORE category, and postoperative course) as well as five out of the eight dimensions plus the Physical Component Summary score for the QOL to be statistically linked to the two-year status.
The multivariate model, which combined sociodemographic data, clinical variables, and all of the QOL dimensions isolated five predictive variables of two-year status in men (Table 3): the Mental Component Summary score for the QOL dimensions and four sociodemographic and clinical variables (angina pectoris, family situation, comorbid diseases, and postoperative course). Concerning women, the multivariate analysis (combining also sociodemographic data, clinical variables, and the selected QOL dimensions) isolated the three following predictive variables of their two-year status (Table 4): the Physical Component Summary score for the QOL dimensions and two of the clinical variables (comorbid diseases and postoperative course).
|
|
| Comment |
|---|
|
|
|---|
Evolution of QOL Scores After Open Heart Surgery
In a previous study [12], we examined the evolution of SF36 scores at one year in all patients having undergone surgery in our department. This study showed an improvement in all but three dimensions of the SF36 scales, Physical and Mental Component Summary scores included. In the present investigation, patients showed a significant and positive change in health over time, with a sharp increase for both genders in all dimensions of the SF36 between baseline and one-year follow-up. These results confirm the findings of previous studies done with the Nottingham Health Profile (NHP) questionnaire, which indicated QOL to be improved in all sections [7, 24, 25]. The stability of results between years 1 and 2 is also an important issue in our study; it is equivalent irrespective of gender. Soderlind and colleagues [26], in a study dealing with QOL after complicated open heart operations, also found no deterioration in results during the first year. However, a two-year follow-up is probably not sufficient to assess the stability of results. A longer follow-up would be likely to show deterioration in QOL, due partially to aging and partially to deterioration of the surgical results such as graft occlusion or valve-related complications.
Our investigation emphasizes the fact that baseline and follow-up SF36 scores for women were significantly lower than those of men in all but two dimensions of the SF36 scale. This is consistent with a recent study by Koch and colleagues [4], who found similar results with another QOL questionnaire, the Duke Activity Status Index. Furthermore, it should be noted that, irrespective of gender, the evolutionary profile in the present study is the same; despite the lower QOL scores in women, the improvement brought about by surgery is equivalent. The preoperative differences in scores between men and women are proportional to those found at years 1 and 2. Other authors have published contradictory results. Koch [4], as well as Sjoland and colleagues [24], found that women showed greater improvement in QOL after coronary artery bypass graft because they start at low baseline scores, so have more opportunity for improvement.
Our study also demonstrates that patients over age 75 having undergone an open heart surgery, be they men or women, have a QOL similar to that of the normal population at two-year follow-up, as measured by the SF36. The two-year scores of patients under age 75 are, on the whole, lower than those of the normal population irrespective of gender. Our results are supported by those of Chocron and colleagues [27] who found two-year postoperative scores to be similar to expected scores in patients over age 70 in all but two dimensions of the NHP. These results are in partial contradiction with a recently published study [28], which showed men and women (mean age for both genders, 65) undergoing coronary bypass grafting to have a QOL, assessed by the SF36, comparable with or better than that of the normal population.
Factors Influencing Two-Year Cardiac Functional Status
The fact that QOL dimensions were predictive factors of two-year cardiac functional status shows that they give more accurate information than most of the clinical data taken into account in our study. In our present study, univariate analysis selected all QOL dimensions (eight baseline dimensions and the two summary scores) in men and five of the eight dimensions plus the Physical Component Summary score in women. The multivariate model highlighted the Mental Component Summary score in men and the Physical Component Summary score in women. The fact that the two summary scores were selected in the final analysis is not surprising, as several advantages of these scores over the original eight dimensions of the SF36 have been reported [13].
Numerous studies have found improvements in both physical and mental functional health status after coronary revascularization procedures [11, 24, 29, 30]. Our study demonstrates an improvement in Mental and Physical Component Summary scores in the follow-up, as well. Moreover, our analysis shows these scores to be more effective than the eight baseline dimensions in identifying two-year cardiac functional status; it emphasizes the dominant role of the Physical Component Summary score in women and the Mental Component Summary score in men. The fact that a low score (< 40) in the Mental Component Summary score is associated with an "unsatisfactory" two-year cardiac functional status in men may not be surprising. This finding shows that the Mental Component Summary score is important to consider in men, as previously shown by Rumsfeld and colleagues [31] in a Veterans Affairs cohort study. It adds to the increasing evidence of a link between mental health status and outcome in cardiac patients, as found in other studies [3234]. The fact that Mental and Physical Component Summary scores are highlighted in multivariate analysis shows that they summarize the QOL in patients with a certain degree of accuracy. In addition, the fact that the Mental Component Summary score is selected in men and the Physical Component Summary score, in women shows that the concerns are different according to gender, as if men and women assessed their own QOL as to what they deem important in their life.
Study Limitations
Two limitations do need to be mentioned. First, the study population comes from a single institution study. Second, 334 patients (35%) operated on during the study period were not included; these were patients who required unscheduled operations (urgent or emergent) and were not likely to be able to concentrate on the questionnaire before surgery. This restricted the scope of the results to scheduled patients.
In conclusion, our study shows at two-year follow-up: (1) the significant improvement brought about by open heart surgery is equivalent in both genders in terms of QOL, although women had lower baseline QOL scores. Quality of life in patients over age 75 is similar to that of the normal population irrespective of gender; (2) the best independent predictive factor of two-year cardiac functional status in women is the baseline Physical Component Summary score and in men, the baseline Mental Component Summary score.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Loponen, M. Luther, J.-O. Wistbacka, K. Korpilahti, J. Laurikka, H. Sintonen, H. Huhtala, and M. R. Tarkka Quality of life during 18 months after coronary artery bypass grafting Eur. J. Cardiothorac. Surg., July 1, 2007; 32(1): 77 - 82. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |