Author + information
- Received June 7, 2004
- Revision received September 16, 2004
- Accepted October 4, 2004
- Published online January 18, 2005.
- Joseph Kim, PhD*,* (, )
- Robert A. Henderson, FRCP†,
- Stuart J. Pocock, PhD*,
- Tim Clayton, MSc*,
- Mark J. Sculpher, PhD‡,
- Keith A.A. Fox, FESC§,
- RITA-3 Trial Investigators
- ↵*Reprint requests and correspondence:
Dr. Joseph Kim, Medical Statistics Unit, London School of Hygiene & Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom
Objectives We sought to compare the effects of an early interventional strategy (IS) versus a conservative strategy (CS) on health-related quality of life (HRQOL) in patients with non–ST-segment elevation acute coronary syndromes (ACS).
Background The third Randomized Intervention Trial of unstable Angina (RITA-3) evaluated early IS (n = 895) versus CS (n = 915). We report one-year results of the RITA-3 trial concerning HRQOL.
Methods The patients' HRQOL was assessed with the Short Form-36 (SF-36) and Seattle Angina Questionnaire (SAQ) at four-month and one-year follow-up, and the EuroQOL Visual Analogue Scale (EQ-VAS) and EuroQOL 5-Dimensional Classification (EQ-5D) also measured at baseline. Analysis was performed using the two-sample ttest and analysis of co-variance.
Results Mean changes from baseline EQ-VAS scores were better for IS than for CS at four months (treatment difference of 3.0, p < 0.001) and one year (2.3, p < 0.01). The EQ-5D utility scores were also higher for IS at four months (treatment difference: 0.036, p < 0.01) and at one year (0.016, p = 0.20). For SF-36, IS scored significantly better at four months for physical function, physical role function, emotional role function, social function, vitality, and general health. The SAQ scores for exertional capacity, anginal stability and frequency, treatment satisfaction, and disease perception were better for IS at four months. These treatment differences were present but attenuated by one-year follow-up. Improvements in HRQOL for IS could be attributed to improvements in anginal symptoms.
Conclusions In patients with non–ST-segment elevation ACS, an early IS provides greater gains in HRQOL, as compared with CS, mainly due to improvements in angina grade.
The use of an early interventional strategy in the management of patients with unstable coronary artery disease has been controversial (1,2). The third Randomized Intervention Trial of unstable Angina (RITA-3) was performed to compare an early interventional strategy (IS) with a conservative strategy (CS) in the treatment of patients with unstable angina or non–ST-segment elevation myocardial infarction (MI) (3). At four-month follow-up, patients randomized to an early IS had a lower risk of developing the combined end point of refractory angina, nonfatal MI, or death. At one year, the treatment groups were comparable with respect to the combined end point of death or nonfatal MI.
Although the clinical benefits of an early IS appear to be clear, its effects on health-related quality of life (HRQOL) remain uncertain. The HRQOL is of particular concern in studies, such as the RITA-3 trial, for which there is a marked difference in the invasiveness of the therapeutic options being considered. In this report, we present our findings from the RITA-3 trial with regard to the effects of an early IS on HRQOL.
The RITA-3 trial
Details of the design and protocol of the RITA-3 trial are described elsewhere (3). Briefly, patients were recruited from 45 centers across England and Scotland from November 12, 1997 to October 2, 2001. Patients were eligible for the study if they experienced chest pain at rest and had documented electrocardiographic or previous arteriographic evidence of coronary artery disease. Patients were randomized either to an early IS (i.e., maximal medical therapy plus early coronary arteriography with possible myocardial revascularization) or to a more CS (i.e., maximal medical therapy plus ischemia- or symptom-provoked angiography and revascularization). Among patients randomized to CS, 94 (10%) of 915 patients had a revascularization procedure during the index admission, compared with 397 (44%) of 895 patients in the IS group (3). Within one year, 28% and 57% of patients in the CS and IS groups were revascularized, respectively.
In all eligible patients, the optimal treatment strategy was uncertain. The co-primary end points of the study were a composite of death, nonfatal MI, or refractory angina at four-month follow-up and a composite of death or nonfatal MI at one-year follow-up.
Health-related quality-of-life instruments
The HRQOL was assessed using both generic and disease-specific self-administered questionnaires: the EuroQOL Visual Analogue Scale (EQ-VAS), EuroQOL 5-Dimensional Classification (EQ-5D), Short Form-36 (SF-36), and Seattle Angina Questionnaire (SAQ) (4–6). Four-month and one-year assessments were performed on all HRQOL instruments, and the two components of the EuroQOL questionnaire were also completed at baseline. The results of HRQOL are confined to those patients who survived to the end of the study period.
The EuroQOL is a two-part, preference-based, generic, self-reported questionnaire composed of the EQ-5D health classification and the EQ-VAS (7). The EQ-VAS is a 20-cm visual analogue scale ranging from 0 (worst imaginable health state) to 100 (best imaginable health state) on which patients are asked to grade their current state of health. An advantage of the EQ-VAS is its relative ease of use and interpretation, although this advantage is offset by the lack of specificity concerning the particular aspect of HRQOL being measured.
The EQ-5D assesses five dimensions of HRQOL: mobility, self-care, usual activities, pain or discomfort, and anxiety or depression. Each dimension is measured on a 3-point ordinal scale where a higher score corresponds to a worse health state (no limitation, some limitation, and greatest limitation in HRQOL). A strength of the EQ-5D is that through their response, patients can effectively locate themselves into one of 245 possible health states for which a global value (i.e., a “utility score”) is assigned. The utility score has been previously determined based on the preferences of a sample of 3,395 members of the United Kingdom general population (8). A utility score of 1 corresponds to the highest degree of HRQOL, and score of 0 is compatible with a level of HRQOL equivalent to death. A negative score indicates a state of HRQOL that is worse than death. The EuroQOL questionnaire has been previously validated in various settings (9–12).
The SF-36 is a generic HRQOL instrument consisting of 36 items grouped into eight dimensions: physical functioning (10 items), physical role functioning (4 items), emotional role functioning (3 items), social functioning (2 items), bodily pain (2 items), mental health (5 items), vitality (4 items), general health perceptions (5 items), and the patient's change in health as compared with one year ago (not considered for this study) (6). Each dimension is scaled on a continuum and ranges from 0 (worst imaginable health state) to 100 (best imaginable health state). The SF-36 has been previously validated in patients with ischemic disease and in the British general population (13–16).
The SAQ is a 19-item, disease-specific questionnaire designed to assess five dimensions of HRQOL related to angina: exertional capacity, angina stability, angina frequency, treatment satisfaction, and disease perception (5). Each component is scaled on a continuum from 0 (worst possible health state) to 100 (best possible health state). An advantage of the SAQ is the high degree of specificity for which it captures HRQOL related to angina (5). The SAQ has been previously validated and applied in patients with ACS (5,17–21).
Although individuals with nonresponses to entire questionnaires were excluded from the analysis, we included those with partial responses where possible. A nonresponse to a particular question did not contribute to the calculation of the component scores.
Angina grade was assessed according to the Canadian Cardiovascular Society (CCS) classification at baseline and all subsequent follow-up visits (22). We modified the CCS scheme for this study to include 0 to indicate the absence of angina. Breathlessness was evaluated on a 6-point ordinal scale: not breathless, breathless while climbing hills, breathless while hurrying on the level, breathless while walking at own pace, breathless while dressing or washing, and breathless at rest (3). Physical activity level was measured on a five-point scale: inactive, sedentary, mild activity level, moderate activity level, and vigorous activity level.
Analysis of covariance was used to assess possible treatment differences in change from baseline scores for the EQ-VAS score and EQ-5D global utility score. The two-sample ttest was used to assess treatment differences for the SF-36 and SAQ. For the EQ-5D, a test of linear trend was performed to evaluate a change from baseline scores between the treatment groups. A p value of <0.05 was considered to be statistically significant. Unless stated otherwise, Bonferroni adjustment was applied in the calculation of treatment differences for each component score in order to correct for having two repeated measures performed at four-month and one-year follow-up. Data are presented as the mean value ± SE, unless otherwise stated. All statistical analyses were performed using STATA (Version 8.2, College Station, Texas).
A total of 1,810 patients were randomized to early IS (n = 895) or CS (n = 915). Response rates for all HRQOL questionnaires were high, as well as being comparable between treatment groups (Table 1).Responses to individual HRQOL questions ranged from 95% to 99% complete.
Baseline EQ-VAS scores were comparable between IS and CS patients (Fig. 1).At four months, patients in both the IS group and CS groups experienced a significant increase in mean VAS scores from baseline (mean differences of 9.2 from baseline, 95% confidence interval [CI] 8.0 to 10.4; and 6.2, 95% CI 5.0 to 7.4, respectively). The improvement at four months was significantly greater in the IS group compared with the CS group (mean difference of 3.0, 95% CI 1.3 to 4.7; p < 0.001). Results for one-year follow-up showed a further increase in mean VAS scores in both treatment groups, but the treatment difference was slightly less (mean difference of 2.3, 95% CI 0.6 to 4.1; p < 0.01).
Table 2shows the frequency of individuals who had a change in EQ-5D score from baseline at four months and at one year (reported p values for trend were not corrected for multiple comparisons). For the most part, there was no difference in change from baseline scores between the treatment groups. However, at both follow-up times, a larger percentage of patients in the CS group had a worsening of HRQOL related to performing usual daily activities. More patients in the CS group also had poorer HRQOL due to anxiety at four months.
With regard to the EQ-5D global utility score (Table 3),patients in both treatment groups experienced a highly significant increase in HRQOL from baseline to four months, as well as from baseline to one year. Although the change in the global utility score was significantly better among patients in the IS group at four months (treatment mean difference of 0.036, p = 0.005), this difference was reduced by one year (treatment mean difference of 0.016, p = 0.20).
All of the mean SF-36 component scores were better for patients in the IS group than in the CS group (Table 4).At four months, all of the treatment differences were statistically significant, with the exception of bodily pain and mental health. At one year, these differences were attenuated but remained statistically significant for physical role function, social function, vitality, and general health. At both follow-up times, the largest gains in HRQOL were made with regard to physical role function and general health.
The mean SAQ component scores were significantly better in the IS group than in the CS group (Table 5).The treatment differences for one-year follow-up were attenuated, as compared with the four-month results, but remained highly significant. The largest gains in HRQOL were related to anginal stability, anginal frequency, and disease perception.
Relation of angina grade to HRQOL
As previously reported (3), there were substantial treatment differences in angina grade. At four months, the prevalences of grade 2 or higher were 26.3% and 36.3% in the IS and CS groups, respectively (difference of 10%, p < 0.0001); at one year, this difference was 7.6% (p = 0.0006). The frequency of angina grade among those who completed the EQ-VAS questionnaire is shown in Table 6.
To evaluate whether the effect of treatment on HRQOL was explained by treatment differences in angina grade, we compared the mean EQ-VAS scores between treatment groups after classifying patients according to current angina grade (Fig. 2).The figure shows a marked decline in HRQOL with increasing severity of angina in both treatment groups. However, at four months and one year, patients with the same angina grade in their respective treatment groups had similar HRQOL. Thus, the gain in HRQOL achieved with an early IS appears to be largely explained by the improvement of angina grade.
Similar results were obtained with the SF-36 (results not shown), where all of the treatment differences became comparable after stratification by angina grade. The only exceptions were observed at four months among those with CCS angina grade 0, who had significant treatment differences for vitality (mean treatment difference of 4.0, p = 0.006) and general health (mean treatment difference of 4.5, p = 0.002).
Our findings suggest that in patients with unstable angina or non–ST-segment elevation MI, an early IS provides greater improvements in HRQOL than does a more CS. The gains achieved with early IS appear to be explained at least in part by improvements in angina grade. We found that treatment differences diminished over time; however, this apparent attenuation may be explained by the use of nonrandomized intervention in some patients assigned to the CS group. It is likely that the observed attenuation in HRQOL scores at one year reflects those severely symptomatic patients in the conservative arm who crossed over and underwent revascularization and hence, were likely to have benefits in HRQOL.
Response rates for all HRQOL instruments used in this study were high (at least 93%) and comparable across treatment groups. Only one patient was lost during the entire course of follow-up. Thus, it is unlikely that our findings were due to bias resulting from differential response rates or losses to follow-up. Overall, baseline clinical characteristics were comparable between patients with and without follow-up of HRQOL scores (results not shown).
An advantage of using a generic HRQOL instrument, such as the EQ-VAS, is ease of use (23). It also provides information on a patient's global state of well-being and allows comparability of relative health states across different diseases (6). However, disease-specific instruments provide greater sensitivity for detecting small changes in HRQOL and give further details on the influence of various components of a particular disease on HRQOL (24). The results of our study were in line with these expectations, where the largest treatment differences in HRQOL were found with the SAQ.
Comparison of results against previous studies
Overall, our findings are consistent with those of previous studies that compared an interventional therapy (i.e., percutaneous transluminal coronary angiography [PTCA], stenting, or coronary artery bypass graft surgery [CABG]) against a more conservative therapy (i.e., medical therapy or symptom-provoked revascularization). Among studies that used the SF-36, it was reported that patients who received an interventional therapy had a significantly better HRQOL related to physical functioning (25–27), physical role functioning (27), general health (25,28), and vitality (25,28). Some studies also found that an interventional therapy provided greater improvements in HRQOL related to bodily pain (28) and mental health (26). In studies where baseline scores were measured, the authors reported significant improvements in at least one of the SF-36 components over time (25,26,28).
With respect to the SAQ, Rinfret et al. (29) reported better HRQOL in patients who underwent stenting compared with those who underwent PTCA with regard to anginal frequency, bodily pain, and disease perception. However, Weaver et al. (30) found no treatment difference in HRQOL between patients who underwent stenting versus balloon angioplasty. The Stent or Surgery Trial (31) compared HRQOL in patients randomized to CABG or PTCA and found that patients in the CABG group had greater improvements in exercise capacity, angina frequency, and disease perception.
In a retrospective cohort study, Sollano et al. (32) compared HRQOL using the EuroQOL in patients who underwent CABG with those who received standard medical therapy. It was reported that the CABG group had better EQ-5D global utility scores than did the medical therapy group. However, in a randomized study that compared stenting to CABG, Serruys et al. (33) found no treatment difference in EQ-VAS or in the EQ-5D global utility index at six-month and one-year follow-up. At one-year follow-up, however, the CABG group had better EQ-5D scores related to usual activities and anxiety.
The results of our study were also consistent with those of a validation study conducted by Dougherty et al. (24), who compared the relative merits of the SAQ (a disease-specific measure), the SF-36 (a global measure), and the Quality of Life Index-Cardiac version III (a more global measure) for capturing CCS angina grade. The authors found the same rank order of sensitivity (i.e., disease-specific, generic, and more generic) as that in our study for detecting changes in HRQOL across the three HRQOL instruments.
Our findings suggest that an early IS provides small but clinically relevant improvements in HRQOL. Based on the criterion specified by Spertus et al. (34,35), which states that a clinically meaningful difference in SAQ scores ranges between 5 to 8 points, the observed four-month treatment differences for exertional capacity, anginal stability, anginal frequency, and disease perception would be considered clinically important.
In addition, we found that for each 1-unit increase in angina grade, there was about a 5-unit decline in the EQ-VAS score (p < 0.001). At four months, patients in the IS group had, on average, scored three points higher on EQ-VAS than those in the CS group, or about 60% of a 1-unit change in angina grade. Likewise, a 1-unit increase in angina grade corresponded to a 0.068-point decline in the EQ-5D utility score (p < 0.001). We found that patients in the interventional arm, on average, scored 0.036 higher on the EQ-5D utility score, or 53% of a 1-unit change in angina grade.
Clinical relevance of our HRQOL results can also be inferred from the observed difference in results between the disease-specific and generic instruments. We found that treatment differences measured using the SAQ were larger than those obtained with the EuroQOL and SF-36. For example, at four months, we found a 7.7-point difference in HRQOL related to disease perception with the SAQ, compared with a 4.6-point difference in HRQOL related to general health using the SF-36. Moreover, all of the SAQ showed highly significant differences in favor of the IS, whereas only some of the components of the generic questionnaires showed such a treatment benefit. The consistency of these findings with the reduction in the occurrence of refractory angina and CCS angina grade reported previously (3) suggests that the observed differences in HRQOL are clinically significant.
Patients with the severest form of angina or ischemia were excluded from this study because the inclusion criteria required continued medical therapy to be a viable treatment option. Thus, our findings may not be generalizable to those with advanced cardiac disease. In addition, baseline scores for the SF-36 and SAQ were not measured and therefore were not adjusted for in the analysis. However, baselines scores for the EuroQOL were available and found to be highly comparable, suggesting that the randomization procedure provided a balance of factors that determine HRQOL between the treatment groups.
The main advantage of early intervention is the reduction of refractory angina at four-month and one-year follow-up (3). An early intervention conferred no advantage over a conservative strategy for death or MI. The effect of an early intervention on HRQOL was small, and the benefits related to HRQOL appeared to be attributable to improvements in angina grade. Thus, the public policy implications of this study are that an early IS should be recommended to reduce the occurrence of refractory angina and possibly to improve angina-related HRQOL in patients with non–ST-segment elevation ACS. However, this benefit must be balanced against economic cost and clinical risk of performing an early intervention on all patients.
The findings of this study, coupled with the main one-year results of the RITA-3 trial, provide evidence in favor of an early IS. However, further research is needed to evaluate the cost-effectiveness of IS compared with CS and to evaluate these results against existing studies (21,36) in order to fully assess the public health implications of these alternative treatment strategies.
In the management of patients with non–ST-segment ACS, an early IS provides greater improvements in HRQOL as compared with a more CS. Much of the gains in HRQOL appear to be attributable to improvements in angina grade.
We gratefully acknowledge the dedicated efforts of the investigators, the coordinators, and the volunteer patients who participated in the RITA-3 trial.
RITA-3 was funded by a competitive grant from the British Heart Foundation (London), and the British Heart Foundation received a donation from Aventis Pharma (Strasbourg, France). Additional governmental support (Culyer) was obtained to partially reimburse interventional center costs of percutaneous coronary intervention catheters and stents. The authors have no conflicts of interest with respect to the RITA-3 trial. However, Dr. Henderson has received honoraria from Schering Plough, as well as travel grants from Boston Scientific and Medtronic AVE. Dr. Fox or his institution has received unrestricted grant funding and/or honoraria from Bristol Myers Squibb, Sanofi-Synthelabo, Aventis, Merck, and Bayer.
- Abbreviations and acronyms
- acute coronary syndrome
- conservative strategy
- Canadian Cardiovascular Society
- health-related quality of life
- interventional strategy
- EuroQOL 5-Dimensional Classification
- EuroQOL Visual Analogue Scale
- myocardial infarction
- third Randomized Intervention Trial of unstable Angina
- Seattle Angina Questionnaire
- Short Form-36
- Received June 7, 2004.
- Revision received September 16, 2004.
- Accepted October 4, 2004.
- American College of Cardiology Foundation
- Manhapra A.,
- Borzak S.
- Fox K.A.,
- Poole-Wilson P.A.,
- Henderson R.A.,
- et al.
- Spertus J.A.,
- Winder J.A.,
- Dewhurst T.A.,
- et al.
- Kind P.
- Dolan P.,
- Kind P.,
- Williams A.
- Dorman P.J.,
- Waddell F.,
- Slattery J.,
- Dennis M.,
- Sandercock P.
- Dorman P.,
- Slattery J.,
- Farrell B.,
- Dennis M.,
- Sandercock P.,
- the United Kingdom Collaborators in the International Stroke Trial
- Kiebzak G.M.,
- Pierson L.M.,
- Campbell M.,
- Cook J.W.
- Brazier J.E.,
- Harper R.,
- Jones N.M.,
- et al.
- Spertus J.A.,
- Jones P.,
- McDonell M.,
- Fan V.,
- Fihn S.D.
- Weintraub W.S.,
- Culler S.D.,
- Kosinski A.,
- et al.
- Dempster M.,
- Donnelly M.
- Pocock S.J.,
- Henderson R.A.,
- Clayton T.,
- Lyman G.H.,
- Chamberlain D.A.
- Mortensen O.S.,
- Madsen J.K.,
- Haghfelt T.,
- et al.
- Rinfret S.,
- Grines C.L.,
- Cosgrove R.S.,
- et al.
- Zhang Z.,
- Mahoney E.M.,
- Stables R.H.,
- et al.