Journal of Human Hypertension (2002) 16, 691–697 & 2002 Nature Publishing Group All rights reserved 0950-9240/02 $25.00 www.nature.com/jhh Are we negating the benefits of CABG by forgetting secondary prevention?

PR Belcher1, A Gaw3, M Cooper2, M Brown4, DJ Wheatley1 and GM Lindsay2 1Department of Cardiac Surgery, University of Glasgow, Glasgow, UK; 2Nursing and Midwifery School, University of Glasgow, Glasgow, UK; 3Clinical Trials Unit, North Glasgow University NHS Trust, Glasgow, UK; 4Pollokshaws Medical Centre, Pollokshaws, Glasgow, UK

The objective of the study was to examine medically patients (6.5%) received after operation vs none managed secondary prevention at one year after preoperatively. Two took digoxin (0.97%) preoperatively coronary bypass grafting (CABG). In all, 214 and 14 (7.7%) postoperatively (P ¼ 0.001) for chronic consecutive patients undergoing isolated elective CABG . One of these took warfarin. Long- seen four weeks preoperatively and one year post- acting nitrate use fell from 63.4% to 15.8% (P o0.0001). operatively. Preoperative systolic pressure Short-acting nitrate use fell similarly (Po0.0001). Pre- averaged 135 7 20 mmHg, which increased to operatively 37 patients (17.9%) took ACE inhibitors vs 44 148 7 25 mmHg (Po0.0001) as did diastolic pressure postoperatively (24.2%); 39 had not received them (81 7 12 to 87 7 13 mmHg; Po0.0001). Anginal symp- before. Preoperatively 48 (23.2%) took vs 30 toms were reported by 45.1% (Po0.0001) although (16.5%) postoperatively (P ¼ 0.127); 24 had not pre- median severity scored lower (4.0 [3.0–5.4] vs 0 [0–2.0]; viously taken diuretics. More patients took HMGCoA Po0.0001). Breathlessness decreased from 93% to 64% inhibitors postoperatively (P ¼ 0.0068) and total choles- (Po0.0001) and was scored less severely (4.0 [2.0–5.0] terol was significantly reduced with a concomitant vs 2.0 [0–4.0]; Po0.0001. In all, 88% with postoperative increase in HDL fraction. Smoking habit was virtually angina reported dyspnoea against 44% of those without unchanged from 17.8% to 15.1% (P ¼ 0.5023). In conclu- (Po0.0001). Calcium antagonist use was more common sion: angina was common. Apart from prescrib- in patients with angina (27.2% vs 5.1%; Po0.0001), but ing, postoperative secondary prevention measures were not nitrates (P ¼ 0.8695), diuretics (P ¼ 0.4218), digoxin poorly applied, less widespread and less effective than (P ¼ 0.2565), b-blockers (P ¼ 0.0820), or ACE inhibitors preoperatively. The implications are disturbing. (P ¼ 0.7256). Preoperatively 166 patients (80.2%) took Journal of Human Hypertension (2002) 16, 691–697. vs 69.2% afterwards (P ¼ 0.0131). Twelve doi:10.1038/sj.jhh.1001469

Keywords: hypertension; coronary; surgery; risk factor; angina

Introduction CABG has been an accepted treatment for angina pectoris for more than 25 years, with In this prospective observational study, we exam- improved immediate results in terms of mortality ined the prescription of anti-anginal preopera- and morbidity2 as well as improved long-term tively compared to at one year after operation and survival. Following CABG, in most circumstances, the extent to which medically managed secondary it is usual practice to discontinue anti-anginal prevention measures, including control of hyperten- therapy as it is anticipated that symptoms will be sion and hyperlipidaemia, were achieved in this relieved as a result of the procedure. However, patient group. several classes of anti-anginal therapy have an We have previously shown a relatively high antihypertensive effect and indeed may have been incidence of self-reported cardiovascular symptoms prescribed for their antihypertensive properties. including angina (45.1%) and dyspnoea (63.7%), Therefore, the removal of this therapy may reveal following coronary artery bypass grafting (CABG); elevated levels of blood pressure that exceed there was also a disturbingly high rate of persistent 3,4 1 target levels. This may be particularly proble- cigarette smoking at 23%. matic if there is no follow-up monitoring of blood pressure or previously documented history of hypertension. Hypertension is a major modifiable risk factor for the development and progression Correspondence: GM Lindsay, Nursing and Midwifery School, of the underlying disease process, athero- University of Glasgow, 68–70 Oakfield Avenue, Glasgow G12 8LS, sclerosis, and has been shown to increase mortality UK. E-mail: [email protected] 5 Received 30 January 2002; revised 6 July 2002; accepted 15 July at two years post-CABG and significantly lower 2002 quality of life.6 Negating the benefits of CABG? PR Belcher et al

692 We therefore examined the medically adminis- comfort and restriction to life’, was constructed for tered secondary prevention measures adopted by the self-reporting of angina and breathlessness the local health-care workers and the management symptoms. Respondents were asked to indicate the of these symptoms as they occurred. We compared severity of their symptoms at the time of interview pre- and postoperative prescribing and looked by placing a mark on the scale. This is different from for evidence of a structured or rational strategy. the CCC8 and NYHA9 classifications but it provides a means of documenting patients’ perceived symp- tom severity. Methods

Sample and size Statistical analysis A consecutive sample of patients (n ¼ 214) was Data are expressed as the mean 7 s.d. unless other- recruited over a six month period from the surgical wise stated. Dependent upon distribution, paired waiting list for CABG at one cardiac surgical centre comparisons were made using paired Student’s t- according to the following inclusion criteria: tests or Wilcoxon signed rank tests (as each patient 1. Isolated CABG procedure. was their own control), by the use of Arcus 2. Elective operation. Quickstat Biomedical software (Addison Wesley 3. Residence within approximately 50 miles of the Longman trading as Research Solutions). Nominal hospital. data were compared by Fisher exact tests. 4. Expected date for operation was estimated to be approximately four weeks after preoperative assessment for the study. Results Patient descriptors All patients invited to participate in the study agreed to take part and written informed consent The average age was 58.2 7 7.7 years; 79% were was provided. Demographic details (age, sex, post- men. A prior diagnosis of definite MI was known in code) were collected from the patients’ medical 63.2% of the study group, while 33% had no history records. Patients were interviewed prior to opera- and 3.8% uncertain diagnosis. NYHA grading was tion (4 weeks) and postoperative (1 year). grade 1 in 0%, grade 2 in 40.7%, grade 3 in 42.6%, The median number of grafts [interquartile range] grade 4 in 1.9%. Of the patients, 12.6% were was 3 [3–4] per patient. diabetic (an incidence of 11.2% in males and 18.2% in females), 40.1% had systolic blood pressure greater than or equal to 140 mmHg with CHD risk factors 11.9% exceeding 160 mmHg, 30.0% had diastolic blood pressure greater than or equal to 90 mmHg Self-reported tobacco smoking habit was recorded. with 9.0% having diastolic pressure greater than or Blood pressure was measured in accordance with 3 equal to 100 mmHg at preoperative assessment. the British Hypertension Society guideline by the As previously reported smoking habit was vir- same person (GL) on both occasions. Plasma tually unchanged.1 cholesterol level was measured from a venous blood sample at the Institute of Biochemistry, Glasgow Royal Infirmary NHS Trust, using standardised Blood pressure control protocols and internationally agreed quality assur- ance procedures. The patients were weighed in light Before operation the mean systolic blood pressure clothes without shoes to the nearest 0.1 kg and averaged 135 7 20 mmHg. At follow-up, this had height was measured in cm to the nearest 0.5 cm. risen to 148 7 25 mmHg (Po0.0001). Mean diastolic Body mass index (kg/m2) was derived from these pressure before operation was 81 7 12 mmHg but measurements. by one year had risen similarly to 87 7 13 mmHg (Po0.0001); see Figure 1. Postoperative blood pressure was analysed in Drugs terms of antihypertensive therapy as follows: the data set was split between those prescribed the Patients were asked to complete a form indicating specified drug type at follow-up vs those not taking the that they were currently prescribed. that class of agent. For b-blockers, average systolic blood pressure on follow-up, in those treated, was 7 7 Self-reported levels of symptoms 147 25 mmHg against 150 23 mmHg in un- treated patients (P ¼ 0.6410). For patients taking A visual analogue scale7 consisting of a horizontal calcium channel blockers, mean systolic blood continuous line ranging from 0 to 7, where 0 pressure was 145 7 21 mmHg vs 148 7 25 mmHg represents ‘no effect on overall well-being and for those patients not taking the drug (P ¼ 0.6300). health’ and 7 represents ‘complete disability, dis- For patients receiving ACE-inhibitors blood

Journal of Human Hypertension Negating the benefits of CABG? PR Belcher et al

693 Table 1 Patterns of drug prescribing before and after CABG (%)

Pre % P Post %

ACE 17.9 0.172 24.2 Aspirin 80.2 0.013 69.2 Nitrates 88.5 o0.0001 16.4 b-Blocker 58.5 o0.0001 17.0 Calcium antagonist 77.8 o0.0001 14.3 23.2 0.13 16.5 Digoxin 0.97 0.001 7.7 HMG-CoA 16.3 0.0068 27.9 Warfarin 0 0.0001 6.5

ACE: angiotensin converting enzyme inhibitor; HMG-CoA: HMG-CoA reductase inhibitors (); Pre: preoperative; Post: postoperative. P values from Fisher exact tests.

with blanket removal of anti-anginal agents post- operatively. b-blockers. These drugs were taken by only 58.5% of patients preoperatively and by 17.0% at one year afterwards (Po0.0001). It is of interest that patients on b-blockers preoperatively spent less time on the waiting list (192 7 109 vs 232 7 115 days; P ¼ 0.012) than those not receiving these agents. The associa- tion of b-blockade with shorter pre-operative wait- ing times suggests that these patients were more Figure 1 Pre- and postoperative blood pressure. likely to have symptoms of and were thus regarded as more urgent. Nitrates. The use of long-acting nitrates preopera- pressure averaged 145 7 22 mmHg as opposed to tively was 63.4% and postoperatively was 15.8% 148 7 25 mmHg in untreated patients (P ¼ 0.4725). (Po0.0001). Short-acting nitrate use fell similarly from 88.5% to 16.4% (Po0.0001). Thus, the medi- cally identified and treated recurrent angina rate at one year was around 16%, as these drugs are used Vasoactive and anti-anginal drugs specifically to relieve symptoms of myocardial ischaemia. Postoperatively, 97 patients (53%) took no form of ACE inhibitors. Before operation 37 patients antihypertensive or anti-anginal agent. Seventy took (17.9%) were taking an ACE inhibitor and 44 a single agent, 15 took two drugs and one patient patients (24.2%) received a drug of this class took triple therapy. Mean blood pressure in un- afterwards. However, only five of those taking these treated patients was 145 7 22 mmHg, in patients drugs preoperatively were represcribed them post- receiving monotherapy it was 149 7 27 mmHg, and operatively; thus new prescriptions account for the in those taking two drugs it was 153 7 21 mmHg remainder (see Figure 2). (P ¼ 0.4899 by anovar). Those whose systolic blood KATP channel inhibitors. These agents were only pressure was 141 mmHg or higher represent 54% of used in 3.3% of preoperative and 2.7% of post- the sample, whereas for diastolic blood pressure operative patients and have been excluded from 91 mmHg and over this proportion was 34%. If we general analysis. look at 140 or 90 mmHg and over, these proportions change to 60% and 43%, respectively. These figures represent significant increases upon the proportion Diuretic use of patients known to require antihypertensive treat- ment preoperatively both for systolic pressure In total, 48 patients (23.2%) were taking diuretics (P ¼ 0.01; Fisher exact) and for diastolic pressure preoperatively compared to 30 (16.5%) postopera- (P ¼ 0.01; Fisher exact). tively (P ¼ 0.127), of whom 24 had not previously The individual drug treatments are presented taken diuretics (see Table 1). Because of the known below and in Table 1. association between ankle swelling and L-type L-type calcium channel blockers. The use of these calcium antagonists, we looked at those patients agents fell from 77.8% before operation to 14.3% in whom calcium antagonists were the only (Po0.0001; Fisher exact test) afterwards, in keeping anti-anginal drug vs patients exclusively taking

Journal of Human Hypertension Negating the benefits of CABG? PR Belcher et al

694 Cardiac symptoms Anginal pain (see Table 2) was reported to us by 45% of patients at one year follow-up compared to 100% preoperatively (Po0.0001), although the median self-rated score was reduced from that recorded preoperatively (4.0 [3.0–5.4] vs 0 [0–2.0], Po0.0001). Breathlessness had been reported to a greater or lesser degree by 93% of patients pre- operatively and this had been improved to 64% at follow-up (Po0.0001) and again the median self- rated score was reduced (4.0 [2.0–5.0] vs 2.0 [0–4.0], Po0.0001); see Table 2. Only eight patients (4.5%) reported breathlessness as a new symptom and 34.3% with persistent dyspnoea were improved. The presence of postoperative breathlessness was reported by 88% of patients with postoperative angina but in only 44% of those without chest pain (Po0.0001). Systolic and diastolic blood pressure were virtually identical in both these groups of patients (P ¼ 0.5915 and 1, respectively). Drug therapy at follow-up was examined accord- ing to the presence or absence of anginal symptoms Figure 2 Changes in ACE-inhibitor prescription before and after operation. The left-hand column illustrates the cessation of the and dyspnoea. Calcium channel blockers were preoperative ACE-inhibitor drug in individual patients. It can be prescribed more frequently in patients with angina seen from the horizontal lines that five patients continued the (27.2% vs 5.1%; P o0.0001). Significant differences drug through the intervention into the postoperative period. The in prescribing could not be detected with respect to right-hand column indicates de novo prescriptions for these agents. angina for antiplatelet agents (P ¼ 0.8695), diuretics (P ¼ 0.4218), digoxin use (P ¼ 0.2565), b-blockers (P ¼ 0.0820), or ACE inhibitors (P ¼ 0.7256). Of patients reporting postoperative angina, 63% were b-blockers but no significant associations with receiving no form of anti-anginal therapy. diuretic use could be detected. The reason we were The association of postoperative breathlessness unable to associate diuretic use as a result of with different drug therapies was examined. Cal- ingestion of calcium antagonists is possibly because cium channel blockers were more frequently pre- the diuretic had been prescribed for hypertension in scribed (19.3% vs 7.7%; P ¼ 0.0494) in those the first place. patients who reported dyspnoea. However, no such association could be made for antiplatelet drugs (P ¼ 0.3094), diuretic use (P ¼ 0.2990), digoxin (P ¼ 1), b-blockade (P ¼ 0.6854), or ACE inhibitors Use of antiplatelet agents or (P ¼ 0.5905). Preoperatively, 166 patients (80.2%) took aspirin; one took persantin and in five it was unknown whether antiplatelet treatment was being given. Lipid lowering agents and cholesterol targets Forty patients (18.9%) took no antiplatelet drug. Postoperatively 69.2% took aspirin while 30.8% Patients taking these drugs (all classes) numbered 34 were receiving no antiplatelet drug (P ¼ 0.0131; preoperatively (16.3%). Two took a whereas Fisher exact test). No patient took warfarin before the rest were prescribed a statin. The proportion of operation but 12 (6.5%) were recorded as taking the patients taking a HMG-CoA inhibitor rose to 27.9% drug one year after operation; eight of these also took aspirin (see Table 1). Table 2 Symptoms

Pre P Post Treatment of arrhythmias Angina 100% o0.0001 45% Digoxin was taken by two patients (0.97%) Dyspnoea 93% o0.0001 64% preoperatively and 14 (7.7%) postoperatively +angina 88% Àangina 44% (P ¼ 0.001; Fisher exact test) for chronic atrial fibrillation. This is approximately equal to the 10 Pre: preoperative; Post: postoperative. Angina severity score from 0 to number taking warfarin postoperatively but only [7]7. Dyspnoea severity with angina (+angina) vs without angina one patient took both drugs (see Table 1). (Àangina).

Journal of Human Hypertension Negating the benefits of CABG? PR Belcher et al

695 at one year postoperative (P ¼ 0.0068), thus indicat- disease is clear.22 Although target values are not ing a new intervention in these patients. Preopera- being met universally, others are doing better than tively 73.4% of patients had total cholesterol we are.23 It is becoming evident that statins also concentrations exceeding target levels of 5.0 mM/l. have protective effects that are unrelated to levels of Postoperatively, this had improved significantly and cholesterol.23–25 was accompanied by a significant rise in HDL The recurrent angina rate, as determined by fraction. nitrate use, is lower than the self-reported presence of symptoms postoperatively but, as stated in the Methods section, symptom severity levels were Discussion assessed on a subjective seven-point scale which is not the same as the Canadian Cardiovascular The main finding of this study was the poor Criteria8 scale. We are unclear as to the level of application of medically managed secondary pre- severity that would cause the patient to seek vention measures postoperatively. It was also noted medical advice. This may be a problem of low that patients’ perceptions of recurrence of angina expectations in that complete relief of angina was and associated dyspnoea1 were more common than not expected. The presence of postoperative breath- usually reported.11,12 lessness was reported by 88% of patients with Preoperatively, patients may have had angina for postoperative angina but in only 44% of those several years. Many of the drugs used to treat angina without chest pain (Po0.0001). It is recognised that are also those used for the treatment of hyperten- these symptoms commonly occur together and that sion. Normal preoperative blood pressure either dyspnoea is a consequence of myocardial ischae- indicates normotension or successful treatment of mia.26 Diuretic use was not associated with dys- hypertension. Therefore, the true incidence of pnoea. The indication for diuretic therapy in this preoperative hypertension is probably unknown group of patients preoperatively was control of although two-fifths of the patients were diagnosed hypertension but they are also prescribed in an as hypertensive preoperatively. This proportion was attempt to control ankle oedema secondary to the significantly higher at follow-up. Therefore, unrec- use of calcium antagonists, especially amlodipine, ognised hypertension may develop and be revealed although proof of the efficacy of diuretics in this by the practice of blanket withdrawal of anti-anginal situation is missing.27 The decreased use of calcium therapy postoperatively. It is thus not surprising that antagonists postoperatively may possibly explain blood pressure rises were recorded but disappoint- the decreased use of diuretics although we were ing that they had escaped the notice of primary care unable to demonstrate this statistically. physicians, cardiac surgeons, cardiologists and Overall the picture revealed by this study does rehabilitation schemes. Hypertension is a major not compare favourably with other follow-up modifiable risk factor for the development and studies.28–30 Improvements in present practice might progression of the underlying disease process, be achieved by a better monitoring of blood pressure atherosclerosis,10,13 and has been shown to increase after operation. We recommend that blood pressure mortality at two years post-CABG5 and significantly be monitored more frequently than once at the six lower the quality of life.6 The adequate treatment week surgical postoperative visit. Improvements in of hypertension is associated with increased well- the present practice could be made by adherence to being14 and reduction in ,15 myocardial infarc- reportedly successful treatment regimens such as tion,16 and left ventricular hypertrophy.17 the HOT study16 and the British Hypertension It was surprising that significantly fewer patients Society Guidelines: these stipulate that for hyper- were receiving antiplatelet therapy after operation tensives with additional risk factors (such as than before. This was far below the EURASPIRE ), the target blood pressure targets18,19 and also represents a failure of super- should be 140/90 mmHg or below. The HOT study vision. The use of anticoagulants is more complex: has also shown that fewer than 50% potential indications for these drugs include end- of hypertensive patients will be controlled on arterectomy, detection of intraventricular , monotherapy and one-third will require three or deep vein and atrial fibrillation. It is more drugs;16 we found only one patient on triple thus a matter of concern that of the 12 patients being therapy. The most disappointing aspect of this study treated with digoxin for chronic atrial fibrillation, was that antihypertensive treatment was inade- only one was receiving warfarin. This again falls quate, continuity of prescribing was virtually absent well short of published guidelines.20,21 although indications for some drugs, such as Improvement in the management of lipids diuretics, preoperatively were unclear. One striking was noted. While moderately encouraging, this feature was that nearly all the patients who were still represents a shortfall in prescribing because taking ACE inhibitors preoperatively were not 61.2% of patients postoperatively had total represcribed them postoperatively. Most units cholesterol concentrations exceeding target levels routinely stop ACE inhibitors before operation of 5.0 mM/l. The effect of statin drugs on improv- because of their profound hypotensive effects ing survival in subjects with coronary artery during cardiopulmonary bypass. Possibly because

Journal of Human Hypertension Negating the benefits of CABG? PR Belcher et al

696 hypertension does not manifest itself in the first 9 Criteria Committee for the New York Associa- few days after operation, it is easy to overlook tion. Nomenclature and Criteria for Diagnosis of this important part of secondary prevention. How- Diseases of the Heart and Blood Vessels, 7th Edn. ever, it is curious that a similar proportion of Little Brown and Company: Boston, 1974. patients were started de novo on these drugs post- 10 Castelli WP. Epidemiology of coronary heart disease: the Framingham study. Am J Med 1984; 76 (Suppl 2A): operatively. 4–12. Nurse-led programmes of shared care for patients 11 CABRI Trial. Participants first-year results of CABRI 31 awaiting bypass surgery, and secondary preven- (Coronary Angioplasty vs Bypass Revascularisation tion clinics for patients with CHD in the commu- Investigation). Lancet 1995; 346: 1179–1184. nity32 can effectively improve CHD risk factors 12 Pocock SJ et al. Quality of life, employment status, and including smoking status. Such models of care anginal symptoms after coronary angioplasty or bypass could be considered for the postoperative manage- surgery. 3-year follow-up in the Randomized Interven- ment of patients following CABG. Although cardiac tion Treatment of Angina (RITA) Trial. Circulation rehabilitation programmes have a remit for second- 1996; 94: 135–142. ary prevention, attendance rates may be low and 13 Shaper AG et al. Risk factors for ischaemic heart disease: the prospective phase of the British Regional outcome data for secondary prevention targets are Heart Study. J Epidemiol Commun Health 1985; 39: not routinely available. 197–209. We conclude that our present management of 14 Wiklund I, Halling K, Ryden-Bergsten T, Fletcher A. secondary prevention in post-coronary artery bypass Does lowering the blood pressure improve the mood? patients is poor. More considered replacement of Quality-of-life results from the Hypertension Optimal anti-anginal drugs with concomitant antihyperten- Treatment (HOT) study. Blood Press 1997; 6: 357–364. sive effects is mandated. There is little excuse for the 15 Collins R et al. Blood pressure, stroke, and coronary decreased use of antiplatelet agents and lack of heart disease. Part 2, Short-term reductions in blood anticoagulation for those in chronic atrial fibrilla- pressure: overview of randomised drug trials in their tion. Better follow-up and continuity of care are epidemiological context. Lancet 1990; 335: 827–838. 16 Hansson L et al. Effects of intensive blood-pressure required, probably by the use of nurse-led pro- lowering and low-dose aspirin in patients with grammes. Until risk factors are properly controlled, hypertension: principal results of the Hypertension long-term outcomes will remain jeopardised and Optimal Treatment (HOT) randomised trial. HOT cost–benefit analysis of these procedures will be Study Group. Lancet 1998; 351: 1755–1762. adversely affected. 17 Zabalgoitia M et al. Effect of regression of left ventricular hypertrophy from systemic hypertension on systolic function assessed by midwall shortening (HOT echocardiographic study). Am J Cardiol 2001; 88: References 521–525. 18 Enbergs A et al. Secondary prevention of coronary 1 Lindsay GM, Smith LN, Hanlon P, Wheatley DJ. The heart disease: ideal versus reality in the Munster influence of general health status and social support EURASPIRE study population. Z Kardiol 1997; 86: on symptomatic outcome following coronary artery 284–291. bypass grafting. Heart 2001; 85: 80–86. 19 Brady AJB, Oliver MA, Pittard JB. Secondary preven- 2 National Institute of Health Consensus. Development tion in 24,431 patients with coronary heart disease: Conference Statement: coronary artery bypass surgery: survey in primary care. Br Med J 2001; 322: 1463. scientific and clinical aspects. N Engl J Med 1981; 304: 20 Stroke Prevention in Atrial Fibrillation Investigators. 680–684. Warfarin versus aspirin for prevention of thrombo- 3 Sever P et al. Management guidelines in essential embolism in atrial fibrillation: stroke prevention in hypertension: report of the second working party of atrial fibrillation II study. Lancet 1994; 343: 687–691. the British Hypertension Society. Br Med J 1993; 306: 21 Atrial Fibrillation Investigators. Risk factors for stroke 983–987. and efficacy of therapy in atrial 4 Ramsay LE et al. British Hypertension Society guide- fibrillation. Analysis of pooled data from five rando- lines for hypertension management 1999: summary. Br mized controlled trials. Arch Intern Med 1994; 154: Med J 1999; 319: 630–635. 1449–1457. 5 Herlitz J et al. Mortality, risk indicators, mode and 22 Scandinavian Simvastatin Survival Group. Rando- place of death and symptoms of angina pectoris in the mised trial of cholesterol lowering in 4444 patients five years after coronary bypass grafting in patients with coronary heart disease: the Scandinavian with and without a history of hypertension. Blood Simvastatin Survival Study (4S). Lancet 1994; 344: Press 1999; 8: 200–206. 1383–1389. 6 Sjoland H et al. Improvement in various estimates of 23 Brown WV. Cholesterol lowering in atherosclerosis. quality of life after coronary artery bypass grafting in Am J Cardiol 2000; 86 (4B): 29H–32H. patients with and without a history of hypertension. 24 Koh KK et al. Non-lipid effects of statin on hyper- J Hypertens 1997; 15: 1033–1039. cholesterolemic patients established to have coronary 7 Kee KA, Kieckhefer GM. Measuring human responses artery disease who remained hypercholesterolemic using visual analog scales. Western J Nurs Res 1989; while eating a step-II diet. Coronary Artery Dis 2001; 11: 128–132. 12: 305–311. 8 Campeau L. Grading of angina pectoris [Letter]. 25 Flaker GC et al. Pravastatin prevents clinical events Circulation 1975; 52: 522. in revascularised patients with average cholesterol

Journal of Human Hypertension Negating the benefits of CABG? PR Belcher et al

697 concentrations. Cholesterol and Recurrent Events 29 Delacretaz E et al. Secondary prevention of ischemic CARE Investigators. J Am Coll Cardiol 1999; 34: heart disease after coronary revascularization. Schweiz 106–112. Runsch Med Prax 1998; 87: 1497–1503. 26 Cook DG, Shaper AG. Breathlessness, angina pectoris 30 Oldenburg B, Pierce J, Sicree R, Ross D. Coronary and coronary artery disease. Am J Cardiol 1989; 63: risk factor outcomes following coronary artery bypass 921–924. surgery. Aust NZ J Med 1989; 19: 234–240. 27 Oparil S et al. Efficacy, tolerability, and effects on 31 McHugh F et al. Nurse-led shared care for patients on quality of life of losartan, alone or with hydrochloro- the waiting list for CABG: a randomised controlled thiazide, versus amlodipine, alone or with hydrochloro- trial. Heart 2001; 86: 317–323. thiazide, in patients with essential hypertension. 32 Campbell NC et al. Secondary prevention in Clin Ther 1996; 18: 608–625. coronary heart disease: a randomised trial of 28 Skinner JS et al. Patient-related outcomes five years nurse led clinics in primary care. Heart 1998; 80: after coronary artery bypass graft surgery. Q J Med 447–452. 1999; 92: 87–96.

Journal of Human Hypertension