Quality of Life, Employment Status, and Anginal Symptoms After Coronary Angioplasty or Bypass Surgery
3-Year Follow-up in the Randomized Intervention Treatment of Angina (RITA) Trial
Background The Randomized Intervention Treatment of Angina (RITA) trial compares initial policies of percutaneous transluminal coronary angioplasty (PTCA) and coronary artery bypass graft surgery (CABG) in 1011 patients with angina. This report assesses the impact of these revascularization procedures on angina, quality of life (according to the Nottingham Health Profile), and employment over 3 years of follow-up.
Methods and Results Both interventions produced marked improvement in all quality-of-life dimensions (energy, pain, emotional reactions, sleep, social isolation, and mobility) and seven aspects of daily living. Patients with angina at 2 years had more quality-of-life impairment than angina-free patients, whose perceived health was similar to population norms. This reflects the close link at baseline between angina grade and quality of life. The slightly greater impairment of quality of life in PTCA compared with CABG patients is a result of their significantly higher chances of having angina, especially after 6 months. Employment status was investigated mainly for men ≤60 years old. PTCA patients returned to work sooner (40% at 2 months compared with 10% of CABG patients), but the latter caught up by 5 months. After 2 years, 22% and 26% of CABG and PTCA patients, respectively, were not working for cardiac reasons. Patients with angina at 2 years were much more likely to be unemployed than those without.
Conclusions The impact of angina on quality of life and unemployment is greatly alleviated by PTCA or CABG. Angina is avoided more successfully with CABG, but PTCA offers a speedier return to work. Both intervention strategies then produce similar benefits for quality of life and employment over several years.
Percutaneous transluminal coronary angioplasty and CABG are both effective treatments for selected patients with symptomatic coronary artery disease. However, the relative effects of these myocardial revascularization procedures on anginal symptoms and other aspects of health have not previously been quantified in strictly comparable patient groups.
The first RITA trial is comparing the long-term effects of initial treatment strategies of PTCA and CABG both in patients with single-vessel and in those with multivessel disease. Interim results over a mean follow-up of 2.5 years demonstrated no difference in prognosis between the two treatment strategies,1 and similar results have recently been reported from other studies.2 3 4 5 6 In this report, we describe the effects of the two revascularization strategies on angina, employment status, and perceived health (using the NHP)7 8 9 10 over 3 years since randomization.
The RITA trial design has been described in detail.1 11 In brief, patients with arteriographically proven coronary artery disease were eligible for randomization if equivalent revascularization was judged feasible by either PTCA or CABG. Patients with left main stem disease, hemodynamically significant valve disease, previous myocardial revascularization, or a noncardiac illness likely to limit long-term prognosis were excluded.
Between March 1988 and November 1991, 1011 patients were assigned to initial treatment strategies of PTCA or CABG. Patients were followed 1, 6, and 12 months after the revascularization procedure and subsequently at yearly intervals after randomization. At each follow-up visit, the patient's clinical and employment status were assessed. Angina was graded by the Canadian Cardiovascular Society classification,12 and at randomization, 75% of patients had severe angina (grade 3 or 4). A small proportion had no angina or mild angina (grade 1) at randomization: 9% and 10%, respectively. Employment status was classified into one of three categories: at work (full time or part time), not working for cardiac reasons (whether unemployed, retired, or unable to work although still having an employer), or not working for other reasons (similarly).
Perceived health status was assessed at baseline, 6 months after revascularization, and 2 years after randomization by the NHP,7 8 9 10 a standardized, self-administered questionnaire that consists of two parts. Part 1 includes 38 statements describing levels of physical, social, or emotional distress, which are grouped into six dimensions: energy (3 statements), pain (8), emotional reactions (9), sleep (5), social isolation (5), and physical mobility (8). None of the pain statements refer specifically to chest pain or angina. To complete part 1 of the profile, patients indicate which of the 38 problems they are experiencing by providing a yes/no response. Scores are calculated for each of the six health dimensions by summing the number of positive (yes) responses: the higher the score, the greater the impairment of health. Since the statements in part 1 of the protocol differ in terms of their impact on health, the scores in each category can be adjusted with an optional weighting system.7 10 In the present study, weighted and unweighted scores produced very similar results, but weighted scores are reported throughout to allow comparison with other studies. In Fig 2⇓, for each dimension, patients were classified into four categories: none, mild, moderate, or severe impairment: severe, both ≥3 and ≥50% yes responses; moderate, ≥2 yes responses but not severe; mild, 1 yes response; and none, all no responses.
Part 2 of the NHP assesses whether an individual's health is causing problems with seven aspects of daily life: work, tasks around the home, social life, home relationships, sex life, hobbies and interests, and holidays. To complete part 2 of the NHP, patients identify each aspect of daily life that is affected by their state of health with a yes/no response. For both parts 1 and 2, NHP weighted mean scores are compared with population norms of the same age and sex derived from a general community survey.7
The intention-to-treat principle is adopted for all analyses except those concerning time of first return to work after the randomized procedure.
The prevalence of angina in the CABG and PTCA groups during the first 3 years after randomization is shown in Fig 1⇓. At all times, there was a significantly greater risk of angina in the PTCA group, but the magnitude of treatment difference became less marked over time. However, this needs to be set against the much higher prevalence of angina at baseline, when 82% had grade 2 or worse. For the CABG group, there was a steadily increasing prevalence of angina over time, from 1.4% grade ≥2 at 1 month after the procedure to 16.4% at 3 years after randomization. In the PTCA group, the prevalence of grade ≥2 angina remained steady at ≈20%, but this was achieved because some PTCA patients underwent further procedures: 108 patients (21%) required CABG and an additional 77 patients (15%) required further PTCA within 3 years of randomization. The 3-year reintervention rates in the CABG group were much lower: 4 (1%) required additional CABG, and a further 17 (3%) underwent PTCA. After 3 years of follow-up, there was no difference in mortality (18 and 17 deaths in the PTCA and CABG groups, respectively) and nonfatal myocardial infarction (34 and 27 in the PTCA and CABG groups, respectively).
Self-Reported Health Status
Fig 2⇓ shows the prevalence of impaired health-related quality of life as measured by the six dimensions of the NHP (part 1). For both treatment groups, assessments at 6 months and 2 years of follow-up reveal a marked improvement compared with the preintervention assessment in all domains: energy, pain, emotional reactions, sleep, social isolation, and physical mobility. There was not a substantial difference between treatment groups or between 6 months and 2 years, although at 6 months, patients in the CABG group are slightly better than those in the PTCA group in all six dimensions. When all 38 items in part 1 of the NHP were combined, the mean reductions at 6 months compared with baseline in the number of items affected were 4.83 and 6.04 in the PTCA and CABG groups, respectively. The mean difference in favor of CABG is thus 1.21 items (95% CI, 0.33 to 2.09; P=.007). At 2 years of follow-up, this treatment difference was reduced to 0.79 item (P=.10).
Although these are notable average improvements in self-reported health status, this does not apply to every patient. For instance, 42 of 282 patients (15%) with no energy impairment at baseline were impaired at 6 months compared with 170 of 456 (37%) of those with mild or moderate baseline energy impairment and 114 of 212 (54%) of those with severe baseline impairment. Of those with mild or moderate baseline impairment, 3% had severe impairment at 6 months. Similar patterns occurred in the other five dimensions of perceived health status.
Fig 3⇓ shows the patients' perceptions of whether their health was causing problems in seven aspects of living at baseline, 6 months, and 2 years based on part 2 of the NHP. There was marked improvement in both PTCA and CABG patients at both 6 months and 2 years compared with baseline. In most of these life aspects, the proportion of patients impaired by their health was reduced by substantially more than half by 6 months, and there was then little change from 6 months to 2 years. The reduced impact of health problems on work is somewhat less marked at 6 months, but by 2 years a further improvement had occurred.
For all seven life aspects, the prevalence of health-related problems was slightly greater in the PTCA group than the CABG group, both at 6 months and at 2 years. Analyzed separately, these differences are mostly nonsignificant, but when summarized by the mean number of life aspects affected (averaged over the 6-month and 2-year visits), the pattern for a slightly greater life impairment in the PTCA group is significant: means, 0.87 and 1.07 aspects for CABG and PTCA groups, respectively; difference, 0.22; 95% CI, 0.00 to 0.39; P=.05.
Perceived Health Status and Angina
At baseline (ie, before randomized intervention), the patient's grade of angina has a marked relationship with every aspect of perceived health status in both parts 1 and 2 of the NHP (Fig 4⇓). In each case, there is a clear trend whereby the higher the angina grade, the greater the impairment in each domain (part 1) and the greater the impact on life aspects (part 2).
Two years later, the great majority of patients did not have angina, and it is interesting to compare their quality of life with the minorities with mild (grade 1), moderate (grade 2), and severe (grades 3 to 4) angina (Fig 4⇑). The latter grades are combined because of the small proportions with severe angina (4% and 3% with grades 3 and 4 angina, respectively). In addition, the PTCA and CABG groups are combined, there being no evidence of treatment differences in quality of life in any given angina category. Patients with angina fare significantly worse in all aspects of quality of life compared with patients without angina, even if their anginal symptoms are only slight, ie, grade 1. It is interesting to note that for all aspects of perceived health status, patients without angina at 2 years have mean scores very similar to the expected mean population norms for the same age and sex.7
For women and for older men, there are problems in investigating the influence of angina and its treatments on patients' employment status because their numbers are small and it is difficult to determine their employment intentions; ie, many would not wish to work or have the opportunity to work regardless of their health status. Brief results on employment by age and sex are presented at the end of this section. Hence, we confine our attention mainly to men ≤60 years old at baseline, and Fig 5⇓ shows their employment status during 3 years of follow-up.
At baseline, 47% of patients were not working for reasons related to their coronary disease, whereas 39% were employed. The remaining 14% were not working for a variety of other reasons. One month after angioplasty, 27% of patients had returned to work, whereas, not surprisingly, only 2% of surgical patients had returned to work. From 6 months onward, there was a marked improvement in employment status, with fewer than one third of patients not working for cardiac reasons. CABG patients tended to have a lower proportion not working for cardiac reasons, especially at 1 and 2 years. There is a steady increase in the proportions not working for other reasons, possibly linked to the national unemployment rates and to aging. Thus, although there is a decline in the proportion not working for cardiac reasons (more rapid in the CABG group), the proportion in employment stays fairly steady at slightly more than half.
Fig 6⇓ shows the cumulative distributions of time first back at work after randomized procedure for the PTCA and CABG groups. The time lapse is from start of procedure to first reemployment, the few patients who did not receive their randomized procedure are not included (unlike all other analyses), and the results are again for men ≤60 years old at randomization. A sizable minority of PTCA patients returned to work quite quickly, with 25% and 39% returning to employment by 1 month and 2 months, respectively, after their randomized procedure. The CABG group returned to work more slowly, with only 9% back in employment by 2 months. However, by 5 months after the procedure, the PTCA and CABG groups are very similar, with 52% and 50%, respectively, having returned to work. The percentages who never returned to employment within 3 years after the procedure were 38% and 35% in the PTCA and CABG groups, respectively (P=.5). Note that in Fig 5⇑, the percentages not working at 3 years are rather higher (51.8% and 47.7%, respectively), because they include some who returned to work and then became unemployed again by 3 years.
There is a strong link between baseline and subsequent employment status. For instance, for the 218 patients employed initially, 77% remained so at 2 years, and only 11% were not working for cardiac reasons. For the 265 patients not working for cardiac reasons, initially 45% returned to work at 2 years; two thirds of the remainder were not working because of their coronary disease. Of the 77 patients initially not working for other reasons, 13% were employed at 2 years; one quarter of the remainder were not working for cardiac reasons.
In a search for other baseline predictors of employment status at 2 years, a stepwise logistic regression was performed for men ≤60 years old at baseline. After baseline employment status, the other highly significant (inverse) predictor of being at work at 2 years was age (P=.006), mainly because of the decreased prospects of reemployment from the mid-50s onward.
The influences of age and sex on employment rates involve both intentions and opportunities that are not directly related to a patient's angina and its treatment, but it is still useful to describe the age-sex patterns. For no age-sex subgroup was there a discernible difference between PTCA and CABG groups in employment rates at 2 years, and hence, results are presented for both groups combined. For men ≤50, 51 to 55, 56 to 60, 61 to 65, and >65 years old at randomization, the numbers employed 2 years later were 106 of 174 (60.9%), 98 of 159 (61.6%), 76 of 178 (42.6%), 31 of 156 (19.8%), and 9 of 113 (8.0%), respectively. The corresponding rates for women ≤60 and >60 years old at randomization were 40 of 106 (38.5%) and 6 of 81 (7.4%), respectively.
Employment Status and Angina
The Table⇓ shows the relationship of employment status to angina grade at 2 years after randomization for men ≤60 years old at randomization. For patients with grade 2 to 4 angina, the risk of being out of work for cardiac reasons is three times that of patients without angina at 2 years. Even patients with grade 1 angina have a significant increase in their risk of not working for cardiac reasons. On this basis, the observed higher rate of not working for cardiac reasons in the PTCA group at 2 years can be statistically attributed to their higher prevalence of angina compared with the CABG group.
In any discussion with patients and their families about the relative merits of CABG and PTCA, several issues are likely to be raised. One is the perception that the period of incapacity is shorter after PTCA than after CABG, resulting in a speedier return to active life, including return to work. In addition, the PTCA candidate is spared the possible complications of surgery, such as a risk of sternal and leg wound infection and pain that may mar an otherwise good result. Set against this is the need for further interventions after angioplasty. In this report, data on prevalence of angina, return to work, and perceived health status over several years of follow-up in RITA provide valuable information on which to base opinions and advice about the relative merits of the two myocardial revascularization procedures.
We found little difference between the two alternative treatment strategies, as judged by the two outcomes of return to work and perceived health status. We know that incomplete revascularization and early restenosis for those initially managed by angioplasty resulted in a higher proportion of this group requiring further procedures to relieve recurrent angina. However, relief of angina largely determined both return to work and perceived health status, rather than the nature of the intervention.
In reports published between 1977 and 1983, return to work was used extensively as an indicator of the success of coronary surgery in restoring the coronary patient to a full life.13 14 15 16 17 18 19 20 However, as the average age at which patients were operated on went up to >60 years in many clinical series, return to work as an index of success became less pertinent. In particular, two arguments mitigate against return to work as a simple and robust measure of success. One is that in times of high unemployment, a 60-year-old cardiac patient may remain out of work even if angina is relieved. It is partly for this reason that we have paid particular attention to men <60 years old at randomization in this analysis. The other factor is that retirement a little below the usual age may well be sought by patients, often with the support of their medical advisors,20 as a means of maintaining and enjoying their newly restored health.
The issue of return to work has resurfaced with the increasing use of coronary angioplasty to treat angina.21 The fact that a patient can be out of hospital and “back at work” within days of PTCA is seen as an advantage over the longer recuperation necessary after median sternotomy. Furthermore, it can be argued that the less time patients have off work, the more likely they are to retain their jobs and maintain the pattern of work. In a prospective randomized trial, these plausible but unproven arguments can be tested. In RITA, the time to first return to work was earlier in the angioplasty group, but at 2 years and beyond, there was no significant difference in employment rate between the two treatment groups.
It would have been interesting to compare the total work time of the two groups, ie, taking into account work missed because of repeat procedures and setting that against the early advantage of PTCA. However, such continuous employment records were not kept.
Quality-of-life assessment is becoming increasingly important for clinical trial research, although controversies and methodological difficulties still remain to be resolved.22 23 The term “quality of life” has suffered from repeated and casual use. The originators of the NHP used the term “perceived health,”7 but we and others have mostly preferred to use the expression “health-related quality of life.”24 What these subjective measures have in common is that it is the patient's perception, not the practitioner's view of the technical success, that is being measured. They should include inquiries that will reveal the impact on life of any anticipated negative effects or incomplete success of the treatment under evaluation. The instrument chosen should be adequately sensitive to change and discriminate within the range of reported experience of the group of patients under investigation. NHP was chosen for use in RITA because it has been used in other cardiac studies,25 and since RITA was set up, Caine et al26 have reported striking improvement in NHP scores after elective coronary surgery. Furthermore, the functional status after myocardial infarction, as measured by NHP, has been found to correlate strongly with New York Heart Association classification,27 a simple subjective score of cardiac function that is not diagnosis- or technique-specific.
The NHP has the great advantage of being easy to use, although, like many other quality-of-life instruments, it can be cumbersome to analyze and interpret, partly because it generates a large number of data. In total, NHP parts 1 and 2 comprise 13 components, and there are theoretical objections to pooling them. The six domains of part 1 comprise 3 to 9 questions each, with a total of 38 responses. The original data published as weighted means by age and sex are derived solely from subjects in the Nottingham and Derby areas, and the system of weighting (although not liable to affect conclusions) is not well validated.10 28 For all its shortcomings, we know that the NHP does discriminate between different diseases,29 that it correlates with cardiac symptoms,27 and that it is responsive to change.26
At the outset of RITA, it was unknown which initial revascularization policy, PTCA or CABG, would be the more successful in the relief of angina. At baseline, the two randomized groups were similar, and at 3 years, there was relatively little difference between the two groups in the prevalence of angina sufficient to limit ordinary physical activities. However, their experience in the meantime was very different. In those who had angioplasty, although the initial treatment was compatible with a much earlier return to work (Fig 6⇑), the need for further intervention for relief of recurrent angina was much greater (33.1% versus 3.4% after 1 year). It is in making value judgments about the relative burden of these treatment consequences that we look to a measure of perceived health status. Within the case mix of RITA, almost all of the impairment in health appears to be dominated by the symptom of angina. With relief of angina, not only are pain and mobility improved but so are all the other measures of well-being included in the NHP. Furthermore, even mild angina (grade 1) manifests such impairments compared with patients who are free of angina. This raises the possibility that the extra effort to collect quality-of-life data, which may be cumbersome, can be avoided in future trials of angina. This would break down in trials of older or more complex cases, in which the effects of comorbidity may impair quality of life. It also removes the opportunity to check on the negative effects of a treatment such as stroke, wound pain, or other new disability.
The value of this new information is that the choice of PTCA or CABG can be determined largely by how amenable the individual's coronary artery disease is to one or the other of the two technical procedures. It is the relief of angina that is important: provided that this is achieved, the consequences of the two treatments over several years are not markedly different in terms of their beneficial effect on quality of life and return to work.
RITA Trial Participants
Guy's Hospital, London: E. Sowton*, A.K. Yates, P.V.L. Curry, P.B. Deverall, G. Jackson, S. Karani. St George's Hospital, London: C.W. Pumphrey, D.J. Parker, T. Treasure, J. Pepper, J. Smith, D.E. Ward, D. Glen. Wythenshawe Hospital, Manchester: N. Brooks, H. Moussalli, D. Bennett, C. Bray, C. Campbell, A.K. Deiraniya, M. Jones, R. Lawson, A. Rahman, C. Ward, T. Coppinger. London Chest Hospital: R. Balcon, P. Magee, C. Layton, A.D. Timmis, J.E.C. Wright, S. Craythorne, S. Glenn. Middlesex Hospital, London: R.H. Swanton, W. Pugsley, E. Firman. Groby Road Hospital, Leicester: P.J.B. Hubner, R.K. Firman, A.H. Gershlick, T. Spyt, M. Tegg. Belfast City Hospital: D.B. O'Keefe, H. O'Kane, J. Cleland, D.J. Gladstone, P. Morton, J.G. Murtagh, M.E. Scott, D. Fitzsimmons. Royal Infirmary, Glasgow: S.M. Cobbe, D.J. Wheatley, F.G. Dunn, I. Hutton, A.R. Lorimer, A.P. Rae, W.S. Hillis, E. Keith, H. Young. St Bartholomew's Hospital, London: D.S. Dymond, S.J. Edmondson, S.O. Banim, D.W. Davies, A.W. Nathan, G. Reece, R.A.J. Spurrell, A. Powe. Northern General Hospital, Sheffield: D.C. Cumberland, G.H. Smith, G.D.G. Oakley, C. Wales. King's College Hospital, London: D.E. Jewitt, J.R.W. Keates, C.A. Bucknall, A.T. Forsyth, A. Jacob. Brook Hospital, London: R. Wainwright, F.P. Shabbo, J.B. O'Riorden, J. Foran. Royal Infirmary of Edinburgh: P. Bloomfield, E.W.J. Cameron, D. de Bono, H. Peterson. Royal London Hospital: M.T. Rothman, A.J. Wood, A.H. MacDonald, P.G. Mills, T. Lewis, I. McCool. St Mary's Hospital, London: R.A. Foale, R. Standbridge, E. Skinner. University Hospital of Wales, Cardiff: K.A.A. Fox, I.M. Breckenridge, U. Hebden, L. Luddington.
Executive Committee: Prof J.R. Hampton (chairman), Dr R.A. Henderson, Prof D.G. Julian, Mr J. Parker, Prof S.J. Pocock, Dr E. Sowton*, Mr J. Wallwork.
Data Monitoring Committee: Dr D.A. Chamberlain (chairman), Mr J.F. Dark, Dr M.D. Joy.
Health Economic Collaborators: Prof M.J. Buxton, M.J. Sculpher.
Trial Coordinator: B. Youard.
*Died July 30, 1994.
The RITA trial is supported by grants from the British Heart Foundation, the British Cardiac Society, and the Department of Health. Additional financial support has been provided by Advanced Cardiovascular Systems Ltd UK (Basingstoke, Hants), Medtronic Ltd (Watford, Herts), and Schneider (Staines, Middlesex). We are grateful to Martin Buxton and Mark Sculpher for their advice on the use and reporting of the NHP. We thank the 1011 patients who accepted randomization. The Royal Free Hospital coronary care unit provided the randomization service.
Selected Abbreviations and Acronyms
|CABG||=||coronary artery bypass graft surgery|
|NHP||=||Nottingham Health Profile|
|PTCA||=||percutaneous transluminal coronary angioplasty|
|RITA||=||Randomized Intervention Treatment of Angina|
*Participants are listed in the “Appendix.”
- Received August 22, 1995.
- Revision received January 3, 1996.
- Accepted January 15, 1996.
- Copyright © 1996 by American Heart Association
Rodriguez A, Boullon F, Perez-Balino N, Paviotti C, Liprandi MIS, Palacios IF. Argentine Randomised Trial of Percutaneous Transluminal Coronary Angioplasty Versus Coronary Artery Bypass Surgery in Multi-Vessel Disease (ERACI): in-hospital results and 1-year follow-up. J Am Coll Cardiol. 1993;22:1060-1067.
Hunt SM, McEwen J, McKenna SP. Perceived health: age and sex comparisons in a community. J Epidemiol Community Health. 1984;38:156-160.
Hunt M, McEwen J. The development of a subjective health indicator. Sociol Health Illness. 1980;2:231-246.
Backett EM, McEwen J, Hunt SM. Report to the Social Science Research Council. Health and Quality of Life. London, UK: Social Science Research Council; 1981.
McKenna SP, Hunt SM. Weighting the seriousness of perceived health problems using Thurstone's method of paired comparisons. Int J Epidemiol. 1981;10:93-97.
Henderson RA. The Randomised Intervention Treatment of Angina (RITA) trial protocol: a long term study of coronary angioplasty and coronary artery bypass surgery in patients with angina. Br Heart J. 1989;62:411-414.
Wallwork J, Potter B, Caves PK. Return to work after coronary artery surgery for angina. Br Med J. 1978;2:1680-1681.
Westaby S, Sapsford RN, Bentall HH. Return to work and quality of life after surgery for coronary artery disease. Br Med J. 1979;2:1028-1031.
Oberman A, Wayne JB, Kouchoukos NT, Charles ED, Russell RO Jr, Rogers WJ. Employment status after coronary artery bypass surgery. Circulation. 1982;65:115-119.
Smith HC, Hammes N, Gupta S, Vlietstra RE, Elveback L. Employment status after coronary artery bypass surgery. Circulation. 1982;65:120-125.
Almeida D, Bradford JM, Wenger NK, King SP, Willis Hurst J. Return to work after coronary bypass surgery. Circulation. 1983;68:205-213.
McGee HM, Graham T, Crowe B, Horgan JH. Return to work following coronary artery bypass surgery or percutaneous transluminal coronary angioplasty. Eur Heart J. 1993;14:623-628.
O'Brien BJ, Buxton MJ, Ferguson BA. Measuring the effectiveness of heart transplant programmes: quality of life data and their relationship to survival analysis. J Chronic Dis. 1987;40:137S-158S.
Caine N, Harrison SCW, Sharples LD, Wallwork J. Prospective study of quality of life before and after coronary artery bypass grafting. BMJ. 1991;302:511-516.
Keilen M, Treasure T, Schmidt U, Treasure J. Quality of life measurements in eating disorders, angina, and transplant candidates: are they comparable? J R Soc Med. 1994;87:441-444.