Cardiac rehabilitation after coronary artery bypass surgery ...
Cardiac rehabilitation after coronary artery bypass surgery:10-year results on mortality, morbidity and readmissions tohospitalBo Hedback b, Joep Perk a , Mikael Hornblada and Ulf Ohlsson a ¨ ¨Aim To evaluate the long-term secondary preventive effect of a Introductioncomprehensive rehabilitation programme after coronary artery Over recent decades, coronary artery bypass graftingbypass grafting (CABG). ŽCABG. has become one of the common treatment modalities for patients with ischaemic heart diseaseMethods The study group included 49 consecutive patientswho underwent bypass surgery and were then offered a ŽIHD.. After CABG, cardiac rehabilitation programmesrehabilitation programme consisting of education in risk-factor are recommended as deﬁned by the World Healthcontrol, a physical training programme and regular follow-up at Organization ŽWHO.: ‘The sum of activities required toa post-CABG clinic. ensure the best possible physical, mental and social conditions, so that the cardiac patient may resume asThe control group ( n = 98), consisting of two well-matchedCABG patients for each study patient, was offered the usual normal a place as possible in the life of the community’care with no access to a cardiac rehabilitation programme. The w1x.two groups were followed for 10 years and the resultsregarding cardiovascular mortality, morbidity, total cardiac According to the guidelines of the Working Group onevents and readmissions to hospital were compared. Cardiac Rehabilitation of the European Society of Car- diology, a cardiac rehabilitation programme should com-Results The total mortality (study group 8.2%, control group20.4%) and cardiovascular mortality (8.2versus 15.3%) after prise smoking cessation, exercise training, diet and lipid10 years did not differ signiﬁcantly between the groups. In the management, psychosocial components, that is, thestudy group, nine patients (18.4%) had 10 cardiac events (four management of stress, anxiety and depression, occupa-cardiovascular deaths, ﬁve non-fatal myocardial infarctions andone CABG) compared to 34 patients (34.7%, P - 0.05) tional assessment and counselling w2x.suffering 46 ( P - 0.01) cardiac events (15 cardiovasculardeaths, 18 non-fatal myocardial infarctions, eight CABG, ﬁve Programmes based on exercise training reduce mortalitycoronary angioplasties) in the control group. The number of and morbidity, as shown in two major meta-analysisreadmissions to hospital (2.1 versus 3.5 per patient) andlength of admissions (11 versus 26 days per patient) was studies w3,4x. The beneﬁcial effect of secondary preven-signiﬁcantly lower in the study group ( P - 0.01). tion, including lifestyle adaptations, is convincingly documented w5 7x. Psychosocial intervention afterConclusion A comprehensive cardiac rehabilitation programme myocardial infarction is effective in attenuating theoffered to patients after coronary artery bypass surgery will long-term outcome in IHD w8x. Even reports on health-improve the long-term prognosis and reduce the need for economic aspects show that cardiac rehabilitation is ahospital care. J Cardiovasc Risk 2001, 8:153–158 2001Lippincott Williams & Wilkins. justiﬁable use of the health-related budget w9 11x. However, access to cardiac rehabilitation is often limited. In the USA less than 20% of eligible patients attend cardiac rehabilitation programmes w12x.Journal of Cardiovascular Risk 2001, 8:153–158Keywords: cardiac rehabilitation, coronary artery disease, coronary Secondary prevention for CABG patients is of majorartery bypass surgery, secondary prevention, risk factors importance, as 10 to 12 years after CABG almost half ofa Department of Internal Medicine and Rehabilitation, Oskarshamn the patients will have developed a recurrent cardiacHospital, Oskarshamn, Sweden and b Department of Cardiology, event Žcardiac death, myocardial infarction, revascular- ¨University Hospital, Linkoping, Sweden. ization. w13,14x, one-third of the population has died ¨Correspondence and requests for reprints to Bo Hedback, MD, PhD, w15x and among the surviving patients 40% of the graftsDepartment of Cardiology, Heart Center, University Hospital, 58185 are occluded w16x. ¨Linkoping, Sweden.Tel: +46 13 222000; fax: +46 13 222171;e-mail:Bo.Hedback@lio.se Thus, long-term studies of the effect of comprehensive rehabilitation after CABG are needed. In a previousReceived 11 September 2000 Revised 20 December 2000 10-year follow-up we have shown the beneﬁcial effectAccepted 27 March 2001 of cardiac rehabilitation on mortality, morbidity and return to work in a non-selected post-myocardial infarc- tion population w17x.1350-6277 2001 Lippincott Williams & Wilkins
154 Journal of Cardiovascular Risk 2001, Vol 8 No 3The aim of this 10-year follow-up report is to describe cerebral stroke.. The study group consisted of the re-the results on mortality, morbidity and readmissions to maining 49 patients Ž39 men and 10 women.. Preopera-hospital of a consecutive post-CABG study group of- tive coronary artery angiography and exercise testingfered participation in the programme and compared ¨ were performed at the University Hospital in Linkoping.with a matched control population not having access to All patients were followed up at the post-CABG clinic,comprehensive rehabilitation. and were invited to participate in the rehabilitation programme. Forty-three patients entered the training programme. Six patients Žﬁve men, one woman. de- clined to join, mainly from lack of motivation. NoMaterial and methods patient was excluded because of cardiac contraindica-Rehabilitation programme tions. The 43 participants attended approximately 80%The Oskarshamn Hospital situated, in the south-eastern of all possible training sessions during the 3-monthregion of Sweden, has since 1978 provided a compre- programme. No cardiac complications or other adversehensive cardiac rehabilitation programme for post- events occurred during the training. A minority Ž20%.myocardial infarction and post-CABG patients, consist- continued physical training in the community pro-ing of standardized medical follow-up, physical training gramme, the majority preferring to lead an activeand psychosocial support. lifestyle.The programme, as described in more detail in our The results of the study are based on the entire studyreports w18,19x, consists of: population according to the intention-to-treat principle. Thus, the six patients who did not join the trainingŽ1. Follow-up at the post-CABG clinic. This includes programme were included in the evaluation. visits at 6 weeks and 4, 8, and 12 months after surgery and thereafter annually for 5 years. The visits include physical examination and education Control group in risk-factor control. Patients with high total serum The patients in the control group were drawn from the cholesterol levels andror overweight are referred 1-year survivors in the CABG register from the to a dietician. ¨ Linkoping University Hospital. All CABG patients fromŽ2. Outpatient-based exercise training. Exercise tests the Swedish south-eastern region were examined pre- are performed at 6 weeks and at 4 and 12 months operatively at this hospital. The results were compiled after CABG. All patients are encouraged to partici- in a regional CABG register. Data from the 1-year pate in the outpatient training programme, which post-CABG follow-up at the local community hospitals starts directly after the ﬁrst exercise test. The were reported to this register according to a standard- sessions consist of 30 to 40 min high-intensity, ized protocol. dynamic interval training concluded with 5 10 min For each patient in the study group, two matched of relaxation twice weekly for 3 months. controls were chosen. In order to avoid investigator bias,Ž3. Home training programme. Each patient is given matching was done by an independent cardiologist at an individually adapted home training programme the University Hospital with access only to preoperative and instructed to exercise three to ﬁve times data of the patients. Matching criteria included age Ž"5 weekly. In the maintenance phase patients are years., gender, year of CABG Ž"2 years., previous referred to a community-based training programme myocardial infarction, smoking, hypertension, maximal if they wish to continue group training. workload at the pre-CABG exercise test Ž"30 W. andŽ4. Psychosocial support. During the training sessions employment status before surgery. repeated health education is given by both physi- cian and physiotherapist. Considerable informal Control group patients were followed up at the depart- support is provided by the patients within the ment of medicine of their local hospital at varying groups. intervals during the ﬁrst year after surgery. Thereafter, most patients were referred to primary health care physicians. They had no access to comprehensive car-Study group diac rehabilitation. As the University Hospital inBetween January 1980 and October 1985, 51 patients ¨ Linkoping carried out a quality control study, all patientsfrom Oskarshamn underwent CABG. Two patients died underwent a standardized exercise test on a bicycleduring the ﬁrst year Žone of heart failure, one from ergometer 1 year after surgery at their local hospitals.
¨ Rehabilitation after bypass surgery Hedback et al. 155Table 1 Comparison of the study and control groups Study group ( n = 49) Control group ( n = 98)Women/ men 10/ 39 20/ 78Mean age (years) 57.0" 7.4 (43–68) 57.3" 7.3 (42–71)Year of operation 1982, 9" 1, 6 1983, 3" 1, 1Previous infarction (%) 63.3 63.3Smokers (%) 26.5 32.6History of hypertension (%) 20.4 20.4Maximum workload preop. 102 W " 33 107 W " 29Employed, non-retired (%) 75.5 65.3Diabetes (%) 6.1 9.1Preoperative angiography:One-vessel disease (%) 30.6 29.6Two-vessel disease (%) 38.8 45.9Three-vessel disease (%) 30.6 24.5Main stem stenosis (%) 10.2 8.2No signiﬁcant differences were found between the groups. W, WaHs.Mortality, morbidity and readmissions to hospital Coronary risk factorsAll data on mortality, morbidity, readmissions and the In our earlier report of the 1-year results of the pro-total duration of hospital stay were collected from hospi- gramme we found the following modiﬁcation of coro-tal ﬁles and patient interviews. The cause of death was nary risk factors. Smoking was signiﬁcantly lower in theclassiﬁed as cardiovascular, or death due to other causes. study group compared to the control group Ž6 versusIf the patient had died outside hospital the death 17%, P - 0.05.. The systolic and diastolic blood pres-certiﬁcate was checked. sures were also signiﬁcantly lower in the study group Ž135r79 versus 142r85, P - 0.01. w19x. We have alsoThe number of patients who had developed a cardiac reported that, on average 38 months after CABG, 66%event wdeath, non-fatal myocardial infarction, repeat of the patients in the study group trained regularlyCABG or PTCA Žpercutaneous transluminal coronary compared with 46% in the control group Ž P s 0.05. andangioplasty.x were calculated after 5 and 10 years in rated their work capacity higher than their matchedeach group as well as the time to the ﬁrst cardiac event. controls Ž P - 0.05.. The use of anxiolytic drugs was atThe number of readmissions and the total duration of that time 9% in the study group, 30% in the controlthe hospital stay at the department of medicine or group Ž P - 0.01. w18x.cardiology for each patient during 10 years after CABG Mortality, morbiditywere counted. In the study group, four patients Ž8.2%. died, all from cardiovascular causes ŽTable 2.. Five patients had aStatistics non-fatalFor statistical analysis, paired Student’s t-tests and the myocardial infarction and one underwent CABG.exact version of the McNemar test were used, whereappropriate. Two-sided tests were used throughout. In the control group, 20 patients died Ž20.4%., of which 15 Ž15.3%. were cardiovascular deaths. The other causes were: tumours Žthree., car accident Žone., alcohol intoxi- cation Žone.. Fifteen patients suffered in total 18 non-Results fatal myocardial infarctions and 13 patients underwentThe two groups were well matched as to age, hyperten- revascularization Žeight CABG, ﬁve PTCA.. Three ofsion, history, and previous infarction. In 42 patients, the the control patients had ﬁrst a myocardial infarction andmatching for smoking was satisfactory, but in seven of later CABG or PTCA and two patients died from car-the control pairs, one of the controls did not fulﬁl the diovascular causes after an earlier cardiac event.matching criteria for smoking. This resulted in a slight,but statistically non-signiﬁcant, difference between the The difference between the groups concerning thegroups ŽTable 1.. same type of cardiac event was not statistically signiﬁ- cant.Although not included in the matching criteria, thenumber of patients with diabetes Ž6.1 versus 9.1%. and Total cardiac eventspre-CABG angiography showed no statistically signiﬁ- During the ﬁrst 5 years post-CABG, the number ofcant differences between the groups ŽTable 1.. patients with a cardiac event Žcardiovascular death, myo-
156 Journal of Cardiovascular Risk 2001, Vol 8 No 3Table 2 Mortality, morbidity, total cardiac events and readmissions Study group ( n = 49) Control group ( n = 98)Total mortality 4 (8.2%) 20 (20.4%) NSCardiovascular mortality 4 (8.2%) 15 (15.3%) NSNumber of non-fatal MI 5 (10.2%) 18 (18.3%)a NSNumber of CABG 1 (2.0%) 8 (8.2%) NSNumber of PTCA 0 5 (5.1%) NSNumber of patients with a cardiac event 9 (18.4%) 34 (34.7%) P - 0.05Total number of cardiac events 10 (20.4%) 46 (46.9%) P - 0.01Mean time to the ﬁrst cardiac event (months) 82 66 P - 0.05Readmissions to hospital 103 342 P - 0.01Total hospital stay (days) 541 2556 P - 0.01a 15 patients. MI, myocardial infarction; CABG, coronary artery bypass grafting; PTCA, percutaneous transluminal coronary angioplasty.cardial infarction, CABG or PTCA. was one in the Discussionstudy group and 14 control patients Ž2.0% vs. 14.3%, This study shows a signiﬁcant reduction in cardiacP - 0.05. ŽTable 2.. events and readmissions to hospital, which can be re- lated to participation in a comprehensive cardiac re-Over the entire 10-year follow-up, nine patients in the habilitation programme. All outcome measures were instudy group experienced 10 cardiac events as compared a favourable direction, although some failed to reach ato 34 controls experiencing 46 events Ž20.4 versus 46.9%, level of statistical signiﬁcance, possibly because of theP - 0.01.. The mean time to a ﬁrst cardiac event was 82 relatively small size of the population. As the study wasmonths in the study population, compared with 66 not designed to analyse the different elements of themonths in the control group Ž P - 0.05.. programme contributing to the beneﬁcial effect, no speciﬁc explanations for the reduction of cardiac eventsThe percentage of event-free surviving patients during can be given. However, several suggestions can bethe 10-year time period in both groups is shown in made, as follows.Figure 1. The study group patients were strongly encouraged to lead an active lifestyle and detailed home training pro-Readmissions grammes were provided. On average, 38 months post-The study group patients had 103 readmissions Ž2.1 per CABG the study group patients trained more regularlypatient. to departments of cardiology or internal and rated their work capacity higher than the controlmedicine and stayed 541 days Ž11 days per patient. in group w18x. Physical training may reduce mortality andhospital compared to 342 readmissions Ž3.5 per patient. morbidity in coronary patients w3,4x and training 5 6and 2556 days Ž26 per patient. for the matched controls hoursrweek is associated with a regression of the coro-ŽTable 2.. The difference in numbers and length of nary atheromatous lesion in some patients w7x.admissions was statistically signiﬁcant Ž P - 0.01.. The programme included a focus on lifestyle changes, including dietary modiﬁcation and smoking cessation, as applied in different secondary preventive studies w5,6x. We could not compare the serum cholesterol values, as we lack data from the control group but it is highly possible that the patients in the study group improved their cholesterol levels compared with the control group, as many of them discussed their diet with a dietician and other team members and were physically more active. The main part of the study period was before the large studies on secondary prevention with lipid- lowering therapy Žmainly statins. were published. At 1-year post-CABG no patients in the two groups usedPercentage of event-free surviving patients during 10 years. statins w19x and only a few patients in the study group were on this kind of therapy at the end of the study.
¨ Rehabilitation after bypass surgery Hedback et al. 157This makes it unlikely that differences in the use of Table 3 Ten-year follow-up in different CABG populationsstatins could explain the results. Cardiac mortality (%) CABG, PTCA (%) Weintraub et al. At 1 year after surgery we found a positive modiﬁcation ( n = 3480) 22 19of risk factors; signiﬁcantly fewer smokers and lower Bathgate and Irving blood pressure in the study group w19x. ( n = 102) Oskarshamn 26.5 12.7 study group ( n = 49) 8.2 2.0No psychologist was engaged in the programme on a control group ( n = 98) 15.3 13.3regular basis but stress management techniques were Total mortality (%)taught by the staff of the programme. An average 38 Yusuf et al. months post-surgery there was a striking difference in ( n = 1324) 26.4 Oskarshamnthe use of anxiolytic drugs. However, no reliable long- control group ( n = 98) 20.4term data on smoking or anxiolytic medication beyond Alderman et al.  CASS study ( n = 390) 18the 1-year post-CABG visit were available. The patientsin the control group had on average 26 days of readmis- CABG, coronary artery bypass grafting; PTCA, percutaneous transluminal coronary angioplasty.sion to hospital compared to 11 days per patient in thestudy group. This contributes to a cost reduction ofapproximately 23 000 SEK per patient and can be com- Alderman et al. w20x and the overview of 10-year post-pared to a programme cost of 3000 SEK per patient w10x. CABG results by Yusuf et al. w15x ŽTable 3..The long-term outcome on cardiac events of this study Our study has several methodological shortcomings: as the control population was obtained in the early yearshas a remarkable similarity with the 10-year follow-up post-surgery, only the long-term follow-up has aof our previously reported study on a consecutive post- prospective design. For practical reasons a randomizedmyocardial infarction population participating in the study protocol could not be chosen. However, allsame programme as the CABG patients w17x. As in the patients were examined by the same medical teampost-myocardial infarction cardiac rehabilitation study, before surgery and all follow-up at 1-year post-CABGwe observed even in the CABG study a widening occurred in an identical manner over the entire region.difference in cardiac events over the 10-year observa-tion period, which might be an indication of the sec- As we lack consistent longitudinal data in the controlondary preventive potential of lifestyle changes achieved group we were unable to report differences in riskafter joining a comprehensive rehabilitation programme. factor management between the groups. Therefore, we restricted the endpoints of this study to mortality, car-To our knowledge no other controlled studies have diac morbidity and readmissions to departments of in-been reported on the 10-year results in post-CABG ternal medicine or cardiology. In this aspect there werepatients participating in comprehensive cardiac rehabili- no missing data.tation. A few other studies have addressed the 10-yearoutcome after CABG where no cardiac rehabilitation With reservation for the design of the study and theprogrammes were used ŽTable 3.. The long-term prog- relatively small size of the population, we conclude thatnosis in the control group in our study could be com- the multifactorial rehabilitation programme after CABGpared with the outcome in these studies, as the patients may have contributed to a signiﬁcant improvement inwere operated on in the same period, 1978 1985. the long-term prognosis of this consecutive group of coronary patients, well comparable to the outcome ofThe study by Weintraub et al. w13x reported a 22% cardiac rehabilitation for patients after myocardial in-cardiac mortality at 10 years post-CABG and 19% re- farction.newed CABG or PTCA. In this consecutive populationfrom the West Lothian Hospital, Scotland, a similar Referencesnumber of patients had to undergo renewed coronary 1 WHO. The rehabilitation of patients with cardiovascular diseases.procedures as in our control group; however the cardiac Report on a seminar. EURO 0381. Copenhagen: World Healthmortality was higher in the Scottish population w14x Organization, regional ofﬁce for Europe, Copenhagen; 1969. 2 Working Group on Rehabilitation of the European Society of Cardiology.ŽTable 3.. Long-term comprehensive care of cardiac patients. Eur Heart J 1992; 13 (suppl. C).Even total mortality rate among controls was of the 3 Oldridge NB, Guyatt GH, Fischer MS, Rimm AA. Cardiac rehabilitation after myocardial infarction: combined experience of randomized clinicalsame level as in other long-term outcome reports by trials. JAMA 1988; 260:945 – 950.
158 Journal of Cardiovascular Risk 2001, Vol 8 No 34 O’Connor GT, Buring JE, Yusuf S, Goldhaber SZ, Olmstead EM, Paffenbarger Jr RS, Hennekens CH. An overview of randomized trials of rehabilitation with exercise after myocardial infarction. Circulation 1989; 80:234 – 244.5 Ornish D, Brown SE, Scherwitz LW, Billings JH, Armstrong WT, Ports TA, et al. Can lifestyle changes reverse coronary heart disease? Lancet 1990; 336:129 – 133.6 Haskell WL, Alderman EL, Fair JM, Maron DJ, Mackey SF, Superko HR, et al. Effects of intensive multiple risk factor reduction on coronary atherosclerosis and clinical cardiac events in men and women with coronary artery disease. Circulation 1994; 89:975 – 990.7 Niebauer J, Hambrecht R, Velich T, Hauer K, Marburger C, Kalberer B. Attenuated progression of coronary artery disease after 6 years of multifactorial risk intervention. Circulation 1997; 96:2534 – 2541.8 Linden W, Stossel C, Maurice J. Psychosocial interventions for patients with coronary artery disease. Arch Intern Med 1996; 156:745 – 752.9 Oldridge N, Furling W, Feeny D. Economic evaluation of cardiac rehabilitation soon after acute myocardial infarction. Am J Cardiol 1993; 72:154 – 161. ˚ ¨10 Levin LA, Perk J, Hedback B. Cardiac rehabilitation – a cost analysis. J Intern Med 1991; 230:427 – 434.11 Ades PA, Pashkow F, Nestor JR. Cost-effectiveness of cardiac rehabilitation after myocardial infarction. J Cardiopulm Rehabil 1997; 17:222 – 231.12 Thomas RJ, Miller NH, Lamendola C, Berra K, Hedback B, Durstine JL, Haskell W. National Survey on Gender Differences in Cardiac Rehabilitation Programs. Patient characteristics and enrollment patterns. J Cardiopulm Rehabil 1996; 16 (Pt 6 ):402 – 412.13 Weintraub WS, Jones EL, Craver JM, Guyton RA. Frequency of repeat coronary bypass or coronary angioplasty after coronary artery bypass graft using saphenous vein grafts. Am J Cardiol 1994; 73:103 – 112.14 Bathgate AJ, Irving JB. Ten year follow up for patients referred for coronary artery bypass grafting from a single district general hospital. Heart 1997; 78:584 – 586.15 Yusuf S, Zucker P, Peduzzi P, Fisher L, Takara T, Kennedy J, et al. Effect of coronary artery bypass graft surgery on survival: overview of 10-year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Lancet 1994; 344:563 – 570.16 Bourassa MG, Enjalbert M, Campeau L, Lesperance J. Progression of atherosclerosis in coronary arteries and bypass graft: ten years later. Am J Cardiol 1984; 53:102C – 107C. ¨17 Hedback B, Perk J, Wodlin P. Long-term reduction of cardiac mortality after myocardial infarction: 10-year results of a comprehensive rehabilitation programme. Eur Heart J 1993; 14:831 – 835. ¨18 Perk J, Hedback B, Engvall J. Effects of cardiac rehabilitation after coronary artery bypass grafting on readmissions, return to work and physical ﬁtness. Scand J Soc Med 1990; 18:45 – 51. ¨19 Hedback B, Perk J, Engvall J, Areskog N. Cardiac rehabilitation after coronary artery bypass grafting: effects on exercise performance and risk factors. Arch Phys Med Rehab 1990; 71:1069 – 1073.20 Alderman E, Bourassa M, Cohen L, Davis K, Kaiser G, Killip T et al. Ten-year follow-up of survival and myocardial infarction in the Randomized Coronary Artery Surgery Study. Circulation 1990; 82:1629 – 1646.