Improvement in quality of life differs between women and men after coronary artery bypass surgery

Quality of life

The patients completed three self-administered questionnaires for the assessment of QoL: the Physical Activity Score, the Nottingham Health Profile and the Psychological General Well-being Index. These questionnaires have been carefully validated and tested for reliability [17–20].

The Physical Activity Score represents one dimension of an angina-specific questionnaire [17], the Angina Pectoris Quality of Life questionnaire, which contains six questions, for the self-estimation of physical abilities and limitations. Each response is graded from 1 to 6 and the mean value for all six questions is calculated. The higher the total value, the greater is the degree of disability.

The Nottingham Health Profile is divided into two parts [18, 19]. Part I consists of 38 statements which convey limitations of activity or aspects of distress in six dimensions: physical mobility, pain, sleep, energy, social isolation and emotional reactions. Patients are required to indicate by a yes/no answer which of the problems they are experiencing at the time they complete the questionnaire. A score ranging from 0 to 100 can be calculated for each dimension of this part of the profile; the higher the score, the greater the limitations in activity or the distressing social and emotional problems. Part II lists the seven aspects of life which are found to be most affected by a person’s state of health: occupation, ability to perform jobs around the house, social life, home relationships, sexual life, hobbies and holidays; a yes/no answer indicates which areas are affected by the respondent’s present state of health.

The Psychological General Well-being Index contains 22 questions, dealing with six sections of well-being: anxiety, depressed mood, vitality, general health, self-control and well-being [20]. The response format is graded from 1 to 6 (total score range 22–132), with the highest value corresponding to superior well-being.

Statistical methods

Groups were compared (i.e. responders versus nonresponders and women versus men) using Mann–Whitney U-test for continuous/ordered variables and the chi-squared test for proportions. Interaction in Table 2 was analysed using the difference in log likelihood between logistic regression models with and without the interaction term in question. The Cochran–Mantel–Haenzel method was used for comparison between males and females when adjusting for age. Within-group comparisons over time were performed using Wilcoxon’s signed rank test and the sign test for continuous/ordered and dichotomous variables, respectively. We did not use the standard repeated measurement model because of the loss of information this would entail, as not all responders answered the questionnaire at all four occasions and therefore would be excluded from the analysis, thus further reducing the number of patients.

Table 2. Clinical characteristics at operation amongst responders and nonresponders in relation to gender

All P-values are two-sided and considered significant if below 0.01 in Tables 3,4,5,6,7,8.

Table 3. Mean Physical Activity Score prior to and after coronary bypass surgery and changes from before surgery, in relation to sex

Table 4. Mean Nottingham Health Profile Part I total score prior to and after coronary bypass surgery and changes from before surgery, in relation to sex

Table 5. Mean scores for Nottingham Health Profile Part I dimensions prior to and 2 years after coronary bypass surgery, and changes from before surgery, in relation to sex

Table 6. Nottingham Health Profile Part II scores prior to and 2 years after coronary bypass surgery and changes from before surgery, in relation to sex

Table 7. Mean Psychological General Well-being Index total score prior to and 2 years after coronary bypass surgery and changes from before surgery, in relation to sex

Table 8. Mean scores for Psychological General Well-being Index dimensions prior to and 2 years after coronary bypass surgery and changes from before surgery, in relation to sex

Measurements of quality of life

Choice of questionnaires

We used a battery of questionnaires for evaluation of QoL in order to evaluate disease-specific and general impact of coronary disease and CABG in the patients operated on. Thus, the Physical Activity Score represents one dimension of a disease-specific questionnaire for the estimation of physical capacity in angina pectoris [17]. The Nottingham Health Profile and the Psychological General Well-being Index constitute general questionnaires. The Nottingham Health Profile is an instrument for measurement of health-related QoL and is most useful amongst patients with chronic and pronounced symptoms [21] and in the detection of treatment effects [22]. The Psychological General Well-being Index addresses the impact of symptoms on well-being and is applicable both for healthy and patient populations.

General results

In the general population women report more disease symptoms and worse QoL than men [14, 15]. Hence, worse results for QoL in women prior to and after an intervention are to be expected as a result of gender. Elderly people have lower exercise capacity [23] and have also been observed to report worse QoL than younger people in a healthy population [14]. Therefore, measurements of QoL in women and men undergoing CABG cannot be justly evaluated unless one takes age into account when exploring the comparative degree of improvement. Also, diabetes mellitus, renal dysfunction and obesity, as seen in a higher frequency in the female patients, have all been observed to influence QoL in a negative direction [24–26]. A higher proportion of concomitant diseases and higher age amongst female patients operated on are constant findings in patients undergoing CABG [8, 10, 27], and are found also in our study. These differences between men and women may account for better QoL and better physical abilities in males than in females at all points in time. However, the overall improvement in QoL in females after coronary bypass surgery tend to be equal to or greater than that in males. If our results are confirmed, they may have impact on referral patterns and timing of coronary bypass surgery in females.

The overall proportion of females operated on in our study was approximately 20%. As compared to recently published data from a number of databases and comprehensive studies this proportion may appear somewhat low [8, 10, 28, 29]. One must question whether this difference reflects variances in the underlying populations studied, or whether a selection bias in favour of allocating male patients to CABG was present in our study. The design of the study does not allow for an evaluation of this issue. However, if a selection bias was present, either at the level of preoperative investigation or at the level of selection for CABG, this would probably partly account for the increased symptoms, higher age, and comorbidity in the female patients, as compared to the males. All of these items would negatively affect QoL. However, with regard to prognostic factors, the proportional underutilization of CABG in women may not be disadvantageous but reflect good clinical judgement, since a number of studies have shown a higher overall mortality after CABG in women than in men [8, 10, 29, 30], and a comparatively better prognosis in women than men suffering from angina pectoris only [9, 27]. Also in our study, despite more severe chest pain, women that were operated on suffered from less advanced coronary disease and had better left ventricular function, which should be considered [8, 10].

Physical activity score

We have previously demonstrated a good correlation between exercise capacity on bicycle ergometer and the Physical Activity Score (r = – 0.56, P < 0.0001) after CABG [31]. Thus, the present results correspond with our previous finding of greater improvement in exercise capacity on bicycle ergometer after CABG in males than in females [32]. It is logical that younger age and less concomitant diseases (being found in men) may not only be associated with better physical ability but also with a greater potential for improvement. Also, male patients have generally been observed to be more likely to participate in rehabilitation for cardiovascular diseases than females, which may affect the outcome in terms of physical abilities [33, 34].

Nottingham Health Profile

According to the Nottingham Health Profile I women appear to have gained more in terms of health-related QoL from CABG at all time-points. This is important information about the benefit of CABG in women, since the majority of previous reports have stressed the inferior symptomatic outcome amongst women [10, 11, 35]. It is possible that the worse symptomatic outcome in women can mainly be explained by differences already prior to surgery. The dominance of pre- and postoperative householding problems in females, as well as their lesser improvement, may reflect the greater domestic responsibilities and demands on female patients in this respect.

Psychological General Well-being Index

As expected, men reached consistently superior scores. However, these results do not appear to be explained by higher age in women since our group have found that the Psychological General Well-being Index is better in elderly patients as compared with younger after CABG [36]. Other investigators have also found better mental health [37] and less psychological distress in elderly patients after CABG [38]. The improvement tended to be greater in women during the first year. This may indicate a tendency toward more rapid recovery after CABG in women.

Limitations

Responders differed from nonresponders in several clinical characteristics in a way that indicated that nonresponders generally suffered from more advanced cardiovascular disease. Furthermore, a lesser proportion of females responded to the questionnaire. The response rate was considerably lower prior to surgery, which was primarily due to administrative reasons or the fact that emergency surgery was performed, precluding the possibility to make an assessment at short notice. We have previously found that patients who are most symptomatic tend to improve their QoL the most after bypass surgery [32]. Therefore, we do not believe that the comparatively low preoperative response rate has overestimated the improvement in QoL.

One may argue that patients who did not respond to the questionnaires did so because they were the most symptomatic in terms of angina pectoris and suffered from more concomitant diseases ( Table 1). Their failing health would thus preclude their participation, and we would systematically misinterpret the benefits of the procedure. This makes it difficult to generalize our results to an unselected population of patients undergoing CABG. It has previously been reported that in QoL evaluations the sickest patients tend to be excluded [39] and with a longer follow-up more patients tend to be excluded [40, 41]. Similar results were found in this survey. Thus we do believe that our results can be generalized to a CABG population, where the sickest were excluded. However, this will always be an issue in every study demanding the personal participation, motivation and mental skills of the patient. Also, we were unable to distribute reminders, which would be likely to have increased the participation in the study.

Received 7 October 1997; accepted 28 October 1998.