Effect of 3 Months of Antimicrobial Treatment With Clarithromycin in Acute Non–Q-Wave Coronary Syndrome
Background— Coronary artery disease, an inflammatory disease, may be caused by infection. We investigated whether the antibiotic clarithromycin would reduce morbidity and mortality in patients with acute non–Q-wave coronary syndrome.
Methods and Results— Altogether, 148 patients with acute non–Q-wave infarction or unstable angina were randomly assigned to receive double-blind treatment with either clarithromycin or placebo for 3 months. The primary end point was a composite of death, myocardial infarction, or unstable angina during treatment; the secondary end point was occurrence of any cardiovascular event during the entire follow-up period (average 555 days, range 138 to 924 days). There was a trend toward fewer patients meeting primary end-point criteria in the clarithromycin group than in the placebo group (11 versus 19 patients, respectively; risk ratio 0.54, 95% CI 0.25 to 1.14; P=0.10). By the end of the entire follow-up, 16 patients in the clarithromycin group and 27 in the placebo group had experienced a cardiovascular event (risk ratio 0.49, 95% CI 0.26 to 0.92; P=0.03).
Conclusions— Clarithromycin appears to reduce the risk of ischemic cardiovascular events in patients presenting with acute non–Q-wave infarction or unstable angina. No signs of this effect diminishing were observed during follow-up.
Received October 23, 2001; revision received January 28, 2002; accepted January 28, 2002.
Myocardial infarction and unstable angina derive from inflammation, which causes plaque vulnerability.1 Increasing attention has recently been paid to infections as one of the causes of inflammation.2 Seroepidemiological, histopathological, and clinical studies in humans and animals have shown that several infections are associated with coronary artery disease (CAD).3–5⇓⇓
Because only a few studies of antibiotic use in CAD patients have been published, the effect of treatment of infections on the prognosis of CAD is unclear.6–9⇓⇓⇓ Furthermore, studies of patients with stable CAD have given discrepant results.6,8⇓ This may be due to the magnitude of inflammation being different in patients with stable versus unstable CAD.10 Patients with acute coronary syndromes appeared to benefit from a 1-month treatment with a macrolide antibiotic; however, the treatment advantage was lost in 6 months.11 We studied the effect of a 3-month clarithromycin treatment on cardiovascular events in patients who presented with acute non–Q-wave infarction or unstable angina pectoris.
We conducted a randomized, double-blind, placebo-controlled study of patients with unstable angina or non–Q-wave myocardial infarction. The patients were recruited between September 1998 and December 2000 from 9 hospitals in different parts of Finland. The ethics committees of the participating hospitals approved the study, and all patients gave written informed consent.
Patients of either sex aged 18 to 80 years who entered the hospital with prolonged chest pain together with clearly documented ST-T-wave changes indicating either unstable angina or non–Q-wave myocardial infarction were eligible for the study. Patients must have had an episode of angina within the 48 hours preceding randomization and presented with at least 1 of the following clinical entities of anginal pain: (1) accelerating pattern, (2) prolonged (>20 minutes) pain with minimal effort, of (3) pain with minimal exertion (>20 minutes) occurring >48 hours after an acute Q-wave myocardial infarction. Patients also had to have ECG evidence of myocardial ischemia: (1) new, persistent, or transient ST-segment depression >0.1 mV (0.08 seconds after the J-point) in at least 2 extremity leads or 3 precordial leads; (2) transient (<20 minutes) ST-segment elevation >0.1 mV in at least 2 extremity leads or 3 precordial leads; or (3) new or transient T-wave inversion >0.3 mV in at least 3 extremity leads or 3 precordial leads (excluding V1). Patients who met the anginal pain inclusion criteria but none of the ECG criteria were eligible to enter the trial if their creatine kinase-MB mass, creatine kinase-B, troponin-T, or troponin-I fraction at the time of enrollment was consistent with the occurrence of myocardial infarction.
Exclusion criteria were thrombolysis within the previous 48 hours; coronary angioplasty within 6 months or CABG within 3 months, or these procedures already planned; angina precipitated by obvious provoking factors (eg, tachycardia); ST-T-segment elevation (>20 minutes); inability to interpret ST-T-segment changes on ECG; long QTc (>470 ms); severe renal or hepatic failure; and ongoing antibiotic therapy of any duration.
Randomization, Treatment, and Follow-Up
Randomization was done separately in each participating center. Randomization lists were based on a table of random numbers,12 and an independent hospital pharmacist of Helsinki University Central Hospital assigned patient numbers for each center (in blocks of 6) and delivered study medication.
Patients received one 500-mg clarithromycin tablet (Klacid 500 mg OD, Abbott Laboratories) or a matching placebo once daily for 85 days. The randomization codes were sealed in closed envelopes, which could be opened only if considered necessary for the treatment of severe adverse events.
The study treatment was in addition to patients’ normal antianginal medication, which typically consisted of aspirin, β-blocker, nitrate, and low-molecular-weight heparin. The choice of this medical therapy and the decision to perform coronary angiography and revascularization, if indicated, was left to the discretion of the treating physician. Revascularization (PTCA, stenting, or CABG) was not considered an end point, but study medication was discontinued for the duration of this procedure. If off-protocol antibiotics were necessary, the study medication was discontinued for the course and reintroduced thereafter.
Medical history was taken and physical examination, laboratory evaluation, and ECG recording were performed at randomization and at the second (days 5 to 8, usually on the day of discharge), third (days 85 to 95; end of study medication), and fourth (day 365±15) visits. Thereafter, information about the defined end points was obtained from hospital records and telephone inquiry. Baseline characteristics were taken from patients’ medical histories. Treatment compliance was assessed by pill count.
The primary end point was occurrence of the composite of death of any cause, myocardial infarction, or unstable angina within the 3-month treatment period. Originally, the secondary end point was occurrence of the same composite of end points as in the primary grouping plus ischemic strokes within 1 year after randomization. Because the targeted number of patients could not be reached (see Statistical Analyses), it was decided to include all cardiovascular events (death, myocardial infarction, unstable angina, ischemic stroke, or critical peripheral ischemia [demanding operation]) occurring during the entire follow-up (to the end of April 2001) in the secondary-end-point grouping. Myocardial infarction after randomization was defined as a new episode of chest pain, with new Q-wave (>0.03 second in duration in 2 or more leads), or both, and elevated cardiac enzyme levels consistent with occurrence of myocardial infarction.
After randomization but during the initial hospital stay, unstable angina was defined as stated in inclusion criteria. However, this new episode had to lead to some additional intervention, such as thrombolytic therapy or revascularization. After discharge from initial hospitalization, unstable angina was also defined as in the inclusion criteria, but here the patient had to be rehospitalized for at least 24 hours.
Two cardiologists (J.S., M.S.N.) examined all patient files to determine whether the cardiovascular events recorded fulfilled criteria of either primary or secondary end points. This was done before the treatment code was opened.
Patient Populations Formed
Patients were eligible for the intention-to-treat population if they fulfilled all inclusion criteria and none of the exclusion criteria, were successfully randomized, and took at least 1 tablet of study medication and if information about occurrence of end points was available until the end of follow-up.
Clinical event rate estimates were based on the ROXIS pilot study, 7 which reported a reduction in events from 9% to 2% in 1 month. If a placebo event rate of 20% is assumed,13,14⇓ 100 patients per group provided ≥90% power to detect a decrease from 20% to 4% with active treatment using a 2-sided alpha of 0.05.12
The recruitment of patients who matched our inclusion criteria proved difficult. It was therefore decided to extend the follow-up until the end of April 2001 and to include all cardiovascular events as secondary end points. The treatment code was opened only after all patients had completed the entire 3-month treatment, all data checking had been done, and all cardiovascular events reported had been evaluated with regard to meeting or not meeting end-point criteria.
Proportions were compared with χ2 test or Fisher’s exact test, and quantitative data were compared with t test or Mann-Whitney test. Kaplan-Meier curves were produced to describe event-free survival, and the difference between groups was analyzed with the log-rank test.
The independent contribution of the potential determinants of event-free survival was analyzed with the Cox model, with the occurrence of an end point as the dependent variable and the treatment code and other potential determinants (age, smoking, sex, diabetes, body mass index, hypercholesterolemia, and hypertension) as independent variables. The risk ratios and their 95% CIs were also obtained from the Cox model. All calculations were made on an intention-to-treat basis with the SPSS 10.0 statistical package (SPSS Inc).
A total of 152 patients were randomized. Of these, 4 were excluded; 148 patients (74 in each group) started the medication (Figure 1).
Baseline characteristics of the 2 study groups are given in Table 1. The groups were well balanced with regard to demographic features, cardiovascular risk factors, previous treatment procedures, and medications. All patients were white. The reason for inclusion was non–Q-wave infarction for 69% and 73% of the patients in placebo and clarithromycin groups, respectively; the rest of the patients had unstable angina (P=NS between groups).
Overall, 41% of the patients used statins at the time of the first visit (Table 1). At the 1-year visit, the rate had increased to 75% in the placebo group and 65% in the clarithromycin group (P=NS). Aspirin was used by 95% of patients in both groups at the first visit. At the 1-year visit, the frequency was lower but still similar in both groups: 79% in the placebo group and 90% in the clarithromycin group (P=NS).
The use of ACE inhibitors and angiotensin II receptor antagonists was low, with ≈20% of patients using either throughout the study; no significant differences between groups were observed. The use of β-blockers was common, with 85% of patients using them similarly in both groups throughout the trial.
No significant differences were found between groups regarding the use of off-study antibiotics (Table 2). In most cases, the antibiotic was used as a prophylactic for CABG (usually 1 dose of cephalosporin, with or without vancomycin). The use of off-study antibiotics had no influence on results.
Treatment Compliance and Study Withdrawals
Of the patients, 55 (75% of survivors) in the placebo group and 57 (77% of survivors) in the clarithromycin group took at least 80% of the study medication (P=NS between groups). No patients were lost during the follow-up, but 14 patients refused to come to the 1-year visit; they were interviewed by phone. The hospital records of all patients were available for end-point analysis.
The event rates during the 3-month medication period and the entire follow-up are presented in Table 3.
Primary End Point
During the first 3 months, a trend was seen toward events occurring less frequently in patients treated with clarithromycin compared with placebo (11 versus 19 patients; risk ratio 0.54, 95% CI 0.25 to 1.14; P=0.10).
Secondary End Point
When all cardiovascular events throughout the follow-up (median 555 days) were included, clarithromycin significantly reduced the rate of cardiovascular events. Sixteen patients in the clarithromycin group and 27 in the placebo group met end-point criteria (risk ratio 0.49, 95% CI 0.26 to 0.92; P=0.03). There were 4 deaths in the clarithromycin group, all caused by ischemic heart disease, and 1 death in the placebo group caused by cancer (which was not considered a cardiovascular end point). In subgroup analyses, the greatest difference in the number of end points between groups was observed in myocardial infarctions, with 5 and 14 patients experiencing myocardial infarctions in the clarithromycin and placebo groups, respectively. Unstable angina occurred in 5 and 11 patients, respectively. Strokes were more rare; only 2 patients met this end point in both groups. No peripheral critical ischemia was observed during the trial.
The event-free survival curves (Figure 2) show that the 2 treatment groups started to separate during the first 3 months, and the difference between them increased up to ≈300 days, remaining approximately the same thereafter. The reduction in the relative risk of events offered by clarithromycin was 41%. The absolute risk reduction was 14.9%, which indicates that 7 patients would need to be treated to prevent 1 cardiovascular event during the follow-up.
Coronary angiography was performed in 62% of all patients (72% in the placebo group and 53% in the clarithromycin group, P=0.02 between groups), a finding consistent with the higher incidence of end points in the placebo group. During the first 3 months, significantly fewer revascularizations occurred in the clarithromycin group than in the placebo group (17 versus 26 patients; risk ratio 0.54, 95% CI 0.291 to 0.997; P=0.05). A similar finding was made during follow-up, with the corresponding figures being 27 and 38 patients (risk ratio 0.58, 95% CI 0.35 to 0.95; P=0.03). Detailed analyses of revascularizations are shown in Table 4.
Adverse events (excluding deaths) were reported in 31 patients: 18 in the placebo group (liver enzyme rise [10 patients], allergic skin reaction , diarrhea , stomach pains , headache , fever , long-QTc , and worsening of asthma ) and 13 in the clarithromycin group (liver enzyme rise , allergic skin reaction , and supraventricular arrhythmia ; P=0.41 between groups). None of the adverse events were classified as serious, but 19 patients discontinued study medication. Premature opening of treatment codes was not necessary for any patient.
The results of this CLARIFY (Clarithromycin in Acute Coronary Syndrome Patients in Finland) study suggest that 3 months’ treatment with clarithromycin may reduce cardiovascular events in patients who present with acute unstable angina or non–Q-wave myocardial infarction. Clarithromycin reduced the risk of events by 41% during an average follow-up of 555 days. The shapes of the event-free survival curves suggest that the beneficial effect starts during treatment and increases thereafter, with no signs of loss of effect occurring during follow-up. Because this treatment was given in addition to the well-managed cardiovascular medical therapy given to participants, the magnitude of the effect was substantial and comparable to that of cholesterol-lowering agents.15
Only 1 study with clinical end points and antibiotic treatment in patients with unstable angina has been published previously.7 In the ROXIS study (Roxithromycin Ischemic Study), patients appeared to benefit from 1-month macrolide treatment. However, no benefit was seen 6 months later. According to the investigators, this could have been because of an insufficiently long treatment or reinfection with Chlamydia pneumoniae.11 Compared with the ROXIS trial, the present study included more patients with first-time angina pectoris and fewer patients with previous myocardial infarctions or revascularizations. The patient population in the present study was also slightly older and included more women and more hypercholesterolemic individuals. The primary end points in these 2 studies were similar; however, the ROXIS trial excluded patients using antibiotics or who had a revascularization done during the trial. The treatment period and follow-up period in the present study were 3 times longer. When the numbers of end points that occurred during the first 3 months in the placebo groups are compared (26% in the present study versus 13% in the ROXIS trial), our study population probably was more unstable. Furthermore, differences were present in the patients’ normal medication: only 60% of the ROXIS population but 95% of patients in the present study used aspirin. Together, these variances may have contributed to the difference in results.
Two studies with clinical end points and antibiotic treatment in patients with stable CAD have been published. The first showed that short-term treatment could decrease episodes of unstable angina or myocardial infarction.6 However, the second, larger study could not reproduce this result.8 One can speculate that because the patients in these studies had stable CAD, the atherosclerotic plaques in the arteries were also inactive, and there may have been less infectious activity in these patients. In addition, these studies included only patients with elevated C pneumoniae antibody titers. The correlation between C pneumoniae antibodies and its occurrence in coronary arteries is poor.3,16⇓
We consider the antibacterial effect to be the most likely mechanism of action of clarithromycin in the present study. Animal models suggest that persistent C pneumoniae infection of vessels can lead to inflammatory changes, and hypercholesterolemia may accelerate lesion progression.17 C pneumoniae is by no means the only infection associated with CAD. Several respiratory tract infections, viruses, and dental infections3,18,19⇓⇓ may also play a role. Therefore, the drug may have influenced other infections as well. However, clarithromycin also has an anti-inflammatory effect,20 which may have contributed to the beneficial effect seen in the present study.
The time course of the treatment effect may have important implications for future studies. In acute infections, the benefit of antibiotics is usually seen within a few days, in contrast to chronic infections, including C pneumoniae and periodontitis, in which healing or suppression of infection may be much slower. In the present study, the benefit of the 3-month treatment was sustained for at least 1.5 years. These results are consistent with the reduction in an inflammatory parameter, C-reactive protein, reported to take place only after 3 months of macrolide therapy in coronary heart disease patients.21 Another anti-inflammatory and plaque-stabilizing product, the statins, show a delayed impact on CAD, beginning at ≈1 year of therapy.15
With regard to potential sources of bias in the present study, the definition of the secondary end point was revised during the study. The difference in the event rates between the clarithromycin and placebo groups was statistically significant only for this end point. However, the revision of end-point criteria and all assessments of coronary events were done before treatment codes were opened. Moreover, the study groups were well balanced with regard to factors that could influence results. Substantial confounding by these variables is therefore unlikely to explain the results. We believe our findings reliably reflect the true biological effect of clarithromycin. The result is clear: the reduction in the risk of events was substantial, and only 7 patients would need to be treated to prevent 1 cardiovascular event.
Thus, patients with acute non–Q-wave coronary syndromes appear to benefit from antibiotic treatment. This observation justifies and warrants further studies in this field.
The following investigators and institutions, listed in alphabetical order according to city (all located in Finland), participated in the CLARIFY study (number of patients enrolled in each center is shown in parentheses): Jorvi Hospital, Espoo—S. Isosomppi (12); Helsinki University Central Hospital, Helsinki—J. Sinisalo, H. Vuorinen-Markkola (32); Malmi City Hospital, Helsinki—A. Kohvakka (10); Maria City Hospital, Helsinki —J. Sinisalo (6); Jyväskylä Central Hospital, Jyväskylä—J. Melin (32); Kajaani Central Hospital, Kajaani—J. Juvonen, P. Alatalo (20); Lahti Central Hospital, Lahti—O. Anttonen (30); Lappeenranta Central Hospital, Lappeenranta—J. Hartman (1); Turku University Central Hospital, Turku—J. Lund, M. Pietilä, L.-M. Voipio-Pulkki (5).
M.S. Nieminen (chair), J. Sinisalo (principal investigator), K. Mattila (project statistician), V. Valtonen, J. Juvonen, J. Melin, O. Anttonen, H. Vuorinen-Markkola, S. Asikainen, P. Saikku.
J. Sinisalo, K. Mattila, V. Valtonen, J. Juvonen, J. Melin, O. Anttonen, H. Vuorinen-Markkola, M.S. Nieminen.
This study was supported in part by the Aarno Koskelo Foundation and the Finnish Foundation for Cardiovascular Research. Abbott Laboratories provided the trial medication and partially funded patient visit costs. The authors greatly appreciate the technical assistance of Mervi Pietilä.
- ↵Fuster V, Fayad ZA, Badimon JJ. Acute coronary syndromes: biology. Lancet. 1999; 353 (suppl II): 5–9.
- ↵Libby P, Egan D, Skarlatos S. Roles of infectious agents in atherosclerosis and restenosis: an assessment of the evidence and need for future research. Circulation. 1997; 96: 4095–4103.
- ↵Mattila K, Valtonen VV, Nieminen MS, et al. Role of infection as a risk factor for atherosclerosis, myocardial infarction and stroke. Clin Infect Dis. 1998; 26: 719–734.
- ↵Saikku P, Leinonen M, Mattila K, et al. Serological evidence of an association of a novelChlamydia TWAR, with chronic heart disease and acute myocardial infarction. Lancet. 1988; 2: 983–986.
- ↵Zhu J, Nieto FJ, Horne BD, et al. Prospective study of pathogen burden and risk of myocardial infarction or death. Circulation. 2001; 103: 45–51.
- ↵Gupta S, Leatham EW, Carrington D, et al. ElevatedChlamydia pneumoniae antibodies, cardiovascular events, and azithromycin in male survivors of myocardial infarction. Circulation. 1997; 96: 404–407.
- ↵Gurfinkel E, Bozovich G, Beck EB, et al. Randomised trial of roxithromycin in non-Q-wave coronary syndromes: ROXIS pilot study. Lancet. 1997; 350: 404–407.
- ↵Muhlestein J, Anderson JL, Carlquist JF, et al. Randomized secondary prevention trial of azithromycin in patients with coronary artery disease: primary clinical results of the ACADEMIC study. Circulation. 2000; 102: 1755–1760.
- ↵Neumann F-S, Kastrati A, Miethke T, et al. Treatment ofChlamydia pneumoniae infection with roxithromycin and effect on neointima proliferation after coronary stent placement (ISAR-3): a randomised, double-blind, placebo-controlled trial. Lancet. 2001; 357: 2085–2089.
- ↵Ross R. Mechanisms of disease: atherosclerosis—an inflammatory disease. N Engl J Med. 1999; 340: 115–121.
- ↵Gurfinkel E, Bozovich G, Beck E, et al. Treatment with the antibiotic roxithromycin in patients with acute non-Q-wave coronary syndromes: the final report of the ROXIS study. Eur Heart J. 1999; 20: 121–127.
- ↵Machin D, Campbell MJ. Statistical Tables for the Design of Clinical Trials. Oxford, UK: Blackwell Scientific Publications; 1987.
- ↵Cohen M, Demers C, Gurfinkel EP, et al. A comparison of low-molecular-weight heparin with unfractionated heparin for unstable coronary artery disease. N Engl J Med. 1997; 337: 447–452.
- ↵FRISC Study Group. Low-molecular-weight heparin during instability in coronary artery disease. Lancet. 1996; 347: 561–568.
- ↵Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994; 344: 1383–1389.
- ↵Thomas M, Wong Y, Thomas D, et al. Relation between direct detection of Chlamydia pneumoniae DNA in human coronary arteries at postmortem examination and histological severity (Stary grading) of associated atherosclerotic plaque. Circulation. 1999; 99: 2733–2736.
- ↵Campbell L, Blessing E, Rosenfeld M, et al. Mouse models of C. pneumoniae infection and atherosclerosis. J Infect Dis. . 2000; 181 (suppl 3): S508–S513.
- ↵Meier C, Jick SS, Derpy LE, et al. Acute respiratory-tract infections and risk of first-time acute myocardial infarction. Lancet. 1998; 351: 1467–1471.
- ↵Mattila K. Dental infections as a risk factor for acute myocardial infarction. Eur Heart J. 1993; 14: 51–53.
- ↵Labro MT. Anti-inflammatory activity of macrolides: a new therapeutic potential? J Antimicrob Chemother. 1998; 41: 37–46.
- ↵Anderson J, Muhlestein JB, Carlquist J, et al. Randomized secondary prevention trial of azithromycin in patients with coronary artery disease and serological evidence for Chlamydia pneumoniae infection: the Azithromycin in Coronary Artery Disease: Elimination of Myocardial Infection with Chlamydia (ACADEMIC) study. Circulation. 1999; 99: 1540–1547.