The opinions and conclusions expressed in this manuscript are solely those of the authors.
Azithromycin is a broad-spectrum macrolide antibiotic that is frequently used to prevent or treat a wide range of bacterial infections, including upper and lower respiratory infections, skin and soft-tissue infections, as well as some sexually transmitted infections.1,2 Azithromycin was introduced in the early 1950s as the first member of a new subclass of macrolides known as azalide.1 Azithromycin was approved by the US Food and Drug Administration (FDA) in late 1991 as the first once-daily antibiotic with a 5-day course of therapy indicated for acute bacterial exacerbations of chronic obstructive pulmonary disease (COPD), community-acquired pneumonia, pharyngitis/tonsillitis, uncomplicated skin and skin structure infections, urethritis and cervicitis, genital ulcer disease in men, and acute otitis media in children.1,2
The off-label use of azithromycin for the prevention of coronary events has been found in the recent literature, and there may be an association between Chlamydia pneumoniae infection and atherogenesis.3,4
Compared with its parent macrolide erythromycin, azithromycin is more effective against gram-negative pathogens (eg, Haemophilus influenzae, Haemophilus parainfluenzae, Moraxella catarrhalis, and Bordetella pertussis) and has better pharmacokinetic properties.1
Azithromycin has an attractive safety profile. It has been shown to be relatively free of cardiotoxic effects, thus providing an advantage over other drugs in the same class.5 As a result, azithromycin has become one of the 15 most prescribed drugs and one of the best-selling antibiotics in the United States.6 According to the IMS Institute for Healthcare Informatics, approximately 48.6 million patients in the United States received a prescription for azithromycin in 2013.6
By contrast, in March 2013 an alert was issued by the FDA that warranted serious attention.7 The alert was in response to the results of an observational study by Ray and colleagues indicating that the use of azithromycin was associated with 2-fold and 3-fold increased risks for cardiovascular (CV) death compared with placebo and amoxicillin, respectively, in patients with a high baseline risk.8 That study involved patients in Tennessee with Medicaid coverage who had a high baseline risk of CV events and showed an increased risk for such events among patients who received azithromycin therapy.8
Generalizing this association to patients with a lower baseline risk of CV disease was uncertain until a nationwide historic cohort study was conducted to examine the safety of azithromycin in Danish adults aged 18 to 64 years over a period from 1997 through 2010.5 The study showed that treatment with azithromycin was not associated with an increased risk of death from CV causes among young and middle-aged patients.5
Both studies agreed that patients who had the greatest baseline risk for CV disease were at highest risk for the disease from taking azithromycin.5,8 Any antibiotic is going to have risks and benefits; however, both studies were observational, and even the most carefully executed observational study can be misleading.
A double-blind, placebo-controlled trial is the best way to eliminate such potentially misleading information.9 Therefore, we conducted this current meta-analysis of all English-language published and unpublished, randomized, controlled, short- and long-term trials involving high-risk patients to investigate whether azithromycin is associated with increased CV death among this patient population.
Methods
We conducted comprehensive searches of the MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials databases using online-based search engines (OVID, PubMed) for human studies published in English between 1990 and September 2013. The search terms that were used to maximize the search specificity and sensitivity included “azithromycin,” “macrolide,” “antibiotic,” “cardiovascular diseases,” and “cardiovascular events.” We searched and evaluated all reference lists of articles that
met the inclusion criteria; online resources, such as CV and infectious disease conference abstracts; as well as clinicaltrials.gov to ensure that we included all published and unpublished studies.
Eligibility Criteria
To be included in this analysis, studies had to be prospective, randomized, placebo-controlled clinical trials involving humans and comparing azithromycin therapy with placebo or with standard of care. The population of interest was clearly defined as any patients who have been identified in the literature review as high risk for CV complications, such as patients with a history of heart problems, established coronary artery disease (CAD), myocardial infarction (MI), unstable angina, or peripheral artery disease (PAD); COPD; or critically ill patients (eg, patients with severe sepsis).
The selected clinical trials included azithromycin for the treatment of infection or for the secondary prevention of coronary events. All studies had to have data on clinical cardiac end points and had to report the risk ratio (RR) of CV events that were associated with the use of azithromycin. Studies that did not report RR and confidence intervals (CIs) but reported enough information to calculate them were also included. Recent studies were included in the analysis if their data were duplicated in more than 1 study.
Data Extraction
Data were extracted and the study characteristics were recorded, including the author(s), year of publication, study design (randomized or observational, placebo or active-controlled, blinded or open-label), study sample size, duration of follow-up, and the dosing regimen. All studies reporting CV outcomes were collected and were stratified into 3 subgroups—mortality, hospitalization, and coronary intervention.
The first study subgroup, mortality, included all studies that reported patients who died from CV causes, such as death resulting from arrhythmias, severe congestive heart failure, cardiogenic shock, or MI, or any other underlying cause of death consistent with a CV cause. The hospitalization subgroup included studies that reported the number of hospitalized patients as a result of nonfatal MI, stroke, or unstable angina, or any cardiac condition that required hospitalization (eg, resuscitated cardiac arrest). The final subgroup, coronary intervention, included all studies that reported patients who had coronary artery bypass grafting surgery or percutaneous coronary intervention. The RRs of CV events associated with the use of azithromycin and its 95% CI were collected or were calculated for each study.
Statistical Analysis
In the analysis, methods based on RRs were used. RRs were calculated using a random effects model (because we assumed that the treatment effect in all the studies included are not identical), and RRs and 95% CIs were used for each outcome. Statistical heterogeneity scores were assessed with the Higgins I2 test (a value of <40% was regarded as “heterogeneity might not be important” and a value of >75% as “considerable heterogeneity”).
A sensitivity analysis was performed by removing some trials, recalculating the combined RRs for the remaining studies, and examining whether there were any serious changes in the overall results. Publication bias was assessed by visual inspection of a funnel plot of the logarithm of effect size versus the standard error for each trial. We looked at adequacies of the blinding of randomization, the blinding of treatment assignment, and the completeness of follow-up in each study to assess the quality of the study by using a scoring system developed by Jadad and colleagues.10 Meta-analyses were performed using the Comprehensive Meta-Analysis software (Biostat, Englewood, NJ) and SAS software version 9.3 for Windows (SAS Institute Inc, Cary, NC).
Results
A total of 6232 studies were identified after removing duplicate publications. Of the initial qualifying studies, 6110 articles were excluded, because they were not randomized clinical trials (these were reviews, editorials, letters to the editor, case reports, or meta-analyses). A total of 25 articles were excluded because they did not report clinical end points (Figure 1).
Characteristics of Studies Included in the Analysis
A total of 12 randomized controlled trials published between 1997 and 2013 met the inclusion criteria for this meta-analysis.11-22 These randomized clinical trials involved a total of 15,588 patients who received active treatment or placebo. The details of these studies are summarized in Table 1.11-22
Of these 12 studies, 9 were double-blinded trials and 3 were open-label trials. A total of 5 studies randomized patients with stable CAD, 3 trials randomized patients presenting with COPD, and 2 trials randomized patients with acute coronary syndrome. Patients with PAD or with severe sepsis were involved in 2 of the studies.
The baseline characteristics of the patients are presented in Table 2. All 12 studies included patients aged ≥65 years. The mean duration of follow-up ranged from 18 weeks to 6 years. Most trials were rated as being of good methodologic quality (randomized, double-blinded, and placebo-controlled studies); however, 1 study was only available as an abstract, and the full text could not be retrieved for full evaluation.22
Treatment
The majority of the trials included 2 study groups, a treatment group and a placebo group. The treatment courses in the 12 trials varied in duration from 3 days to 1 year. Azithromycin was the only drug included in the treatment groups in most of the studies, except the STAMINA study (which included azithromycin combined with metronidazole and a proton pump inhibitor [omeprazole]) and the studies by Blasi and colleagues, Dogra, and Faverio and colleagues (which included azithromycin combined with standard of care).13,18,20,22
Outcome Analyses
The mortality outcomes were presented in 11 trials. Figure 2 shows the RRs and the 95% CI of death associated with the use of azithromycin in each study, as well as for all of the studies combined. Compared with patients who had not taken azithromycin, patients who had taken azithromycin had an overall RR of death of 0.877 (95% CI, 0.752-1.024; P = .097). The combined mortality rate was 3.7% among 7769 treated patients versus 4.2% of 7723 patients in the placebo groups. To assess the potential impact of the results of a poorer quality study, we performed a sensitivity analysis by calculating the RR using all the studies, and then excluding the 1 poorer quality study,22 as shown in Table 3. No change in the overall results of the analysis was detected.
As for hospitalization, 9 trials reported hospitalization outcomes (Figure 3). No significant correlation between the use of azithromycin and the risk of hospitalization resulting from CV causes was observed in these studies (RR, 1.005; 95% CI, 0.922-1.094; P = .915) and no heterogeneity was found (I2 = 0%). The overall hospitalization rate was 7.6% among the 7498 patients receiving active treatment compared with a 10.1% rate among the 7478 patients receiving placebo.
Regarding coronary intervention, 5 of the trials reported outcomes of coronary intervention (Figure 4). The analysis showed no relationship between azithromycin use and coronary intervention rate (RR, 0.999; 95% CI, 0.896-1.114; P = .984).
Higgins I2 showed no heterogeneity among the studies in any of the subgroups (I2 = 0%). In the sensitivity analyses, no change in the overall estimate was observed after removing the poorer quality study and recalculating the combined RRs for the remaining studies. On visual inspection of the funnel plot, the plot appeared roughly symmetrical, suggesting that the likelihood of publication bias is relatively low (Figure 5).
Discussion
The findings from this meta-analysis of prospective studies show that either azithromycin therapy decreases CV and cardiac events or there are no differences in CV events and clinical cardiac outcomes compared with the use of placebo in these patient populations. All the trials included generally failed to produce convincing evidence to prove that azithromycin could cause CV events among high-risk patients.
To our knowledge, this is the first meta-analysis conducted after the FDA alert was issued in March 20137 that has investigated whether azithromycin is associated with increased risk for CV events in the long-term. Our findings appear to conflict with the study by Ray and colleagues.8 There are several important differences between the observational studies included in the study by Ray and colleagues8 and the selected randomized controlled trials included in the present meta-analysis, which may, to some extent, explain the different findings.
First, Ray and colleagues assessed the risk for CV death in patients receiving azithromycin, amoxicillin, ciprofloxacin, levofloxacin, or no antibiotics, based on retrospective data from the Tennessee Medicaid claims.8 Ray and colleagues specifically focused on the risk associated with the typical 5-day duration of the administration of azithromycin.8 It is unclear what was the disease state for which the patients with the Tennessee Medicaid coverage were receiving azithromycin therapy.8 The present meta-analysis specifically selected all randomized clinical trial data with azithromycin used for the treatment or for the secondary prevention of coronary diseases.
Recently, several studies have found a substantial incidence of CV events, such as congestive heart failure or cardiac arrhythmia, occurring during or soon after a hospital admission for pneumonia.23 Therefore, it is difficult to determine the potential cause for these CV events. Many studies have suggested that macrolides have an advantage over other antibacterial agents in treating patients who have pneumonia that requires a minimum of 5 days of therapy.24,25
It is difficult to detect or to examine the benefits of therapy with azithromycin after the fifth day of treatment in the study by Ray and colleagues.8 Their study included patients with infections that required only 5 days of therapy with azithromycin, and infections such as pneumonia that require a longer duration of therapy might have been excluded. Therefore, the benefits of azithromycin in reducing mortality as mentioned above24,25 compared with nonmacrolide therapies in patients with pneumonia were not examined in the study by Ray and colleagues.8
Finally, the population of the Tennessee Medicaid beneficiaries may have a higher prevalence of coexisting conditions and higher mortality rates compared with patients in the randomized clinical trials included in the present meta-analysis.5,8
It has been proven that patients with preexisting heart disease face the highest risk for CV events, even before taking azithromycin, as a result of the inflammation caused by some infectious pathogens.26 Some studies suggest that CV events could be a result of inflammation that contributes to the pathogenesis of CV disease, such as atherosclerosis.27 These studies have investigated the possibility of infectious pathogens, such as Herpesviridae, influenza virus, Cytomegalovirus, enteroviruses, Helicobacter pylori, periodontal pathogens, and C. pneumoniae in causing negative CV effects.27,28 For example, Saikku and colleagues reported that C. pneumoniae is associated with the development of CAD as well as MI.29
In addition, previous studies also show that seasonal influenza-like illnesses may cause acute thrombotic vascular events in patients with previously stable CAD and cerebrovascular disease.30 In another study, a statistically significant association was discovered between evidence of inflammation and coronary events.31 This study used 2 markers, C-reactive protein and serum amyloid A, to predict the risk of first MI and other atherosclerotic events in healthy patients. These markers appear to predict the risks for coronary events and ischemic complications among high-risk patients.31
Our meta-analysis has several strengths. All the studies included in this analysis have a prospective design, which eliminates the possibility of recall and selection bias. Studies with a long duration of follow-up were included to assess the long-term effects of azithromycin on the development of CV events. The characteristics of patients in the primary studies were always comparable. Little evidence of publication bias was found in the studies.
Limitations
Some limitations exist in the present study. The trials selected are varied in follow-up durations, ranging from 6 months to 6 years; however, most trials that examined azithromycin therapy followed the patients between 1 and 2 years.
The primary CV end points also vary between the studies. Some of the selected studies reported all of the CV end points, whereas some did not.
Similarly, some of the studies specified the causes of hospitalization, and other studies did not. We were therefore unable to distinguish if patients were hospitalized for acute angina or for MI. The treatment regimen and duration of azithromycin therapy were used either for treatment or for secondary prevention.
Finally, the results of the present study may not be applicable for all other populations.
Conclusions
The findings in this meta-analysis indicate that there is no increased risk for mortality or for CV events with the use of azithromycin therapy among patients enrolled in the treatment or prevention clinical trials included in this analysis. Future research considerations are warranted to study a larger number of patients with preexisting bacterial infection and CV risk factors, and longer follow-up duration of trials is needed to detect any potential long-term negative or positive effects, including CV effects, of azithromycin therapy.
Acknowledgment
The authors would like to thank K. Lindsay Foltz for her English-language editing in the early version of the manuscript.
Author Disclosure Statement
Mr Almalki and Dr Guo reported no conflicts of interest related to the content of this article. No funding was provided for this study.
Mr Almalki is a Graduate Student, and Dr Guo is Professor of Pharmacoeconomics and Pharmacoepidemiology, The James L. Winkle College of Pharmacy, University of Cincinnati Academic Health Center, OH.
The opinions and conclusions expressed in this manuscript are solely those of the authors.
References
- Hoepelman IM, Schneider MM. Azithromycin: the first of the tissue-selective azalides. Int J Antimicrob Agents. 1995;5:145-167.
- Zithromax (azithromycin) tablets and oral suspension [product information]. New York, NY: Pfizer Inc; June 2014.
- Mawhorter SD, Lauer MA. Is atherosclerosis an infectious disease? Clev Clin J Med. 2001;68:449-458.
- Belland RJ, Ouellette SP, Gieffers J, Byrne GI. Chlamydia pneumoniae and atherosclerosis. Cell Microbiol. 2004;6:117-127.
- Svanström H, Pasternak B, Hviid A. Use of azithromycin and death from cardiovascular causes. N Engl J Med. 2013;368:1704-1712.
- IMS Institute for Healthcare Informatics. Medicine use and shifting costs of healthcare: a review of the use of medicines in the United States in 2013. April 2014. www.imshealth.com/deployedfiles/imshealth/Global/Content/Corporate/IMS%20Health%20Institute/Reports/Secure/IIHI_US_Use_of_Meds_for_2013.pdf. Accessed August 25, 2014.
- US Food and Drug Administration. Azithromycin (Zithromax or Zmax). Drug safety communication: risk of potentially fatal heart rhythms. March 12, 2013. www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm343350.htm. Accessed August 25, 2014.
- Ray WA, Murray KT, Hall K, et al. Azithromycin and the risk of cardiovascular death. N Engl J Med. 2012;366:1881-1890.
- Noseworthy JH, Ebers GC, Vandervoort MK, et al. The impact of blinding on the results of a randomized, placebo-controlled multiple sclerosis clinical trial. Neurology. 1994;44:16-20.
- Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17:1-12.
- Gupta S, Leatham EW, Carrington D, et al. Elevated Chlamydia pneumoniae antibodies, cardiovascular events, and azithromycin in male survivors of myocardial infarction. Circulation. 1997;96:404-407.
- Anderson JL, 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.
- Stone AF, Mendall MA, Kaski J-C, et al. Effect of treatment for Chlamydia pneumoniae and Helicobacter pylori on markers of inflammation and cardiac events in patients with acute coronary syndromes: South Thames Trial of Antibiotics in Myocardial Infarction and Unstable Angina (STAMINA). Circulation. 2002;106:1219-1223.
- O’Connor CM, Dunne MW, Pfeffer MA, et al; for the Investigators in the WIZARD Study. Azithromycin for the secondary prevention of coronary heart disease events: the WIZARD study: a randomized controlled trial. JAMA. 2003;290:1459-1466.
- Cercek B, Shah PK, Noc M, et al; for the AZACS Investigators. Effect of short-term treatment with azithromycin on recurrent ischaemic events in patients with acute coronary syndrome in the Azithromycin in Acute Coronary Syndrome (AZACS) trial: a randomised controlled trial. Lancet. 2003;361:809-813.
- Grayston JT, Kronmal RA, Jackson LA, et al; for the ACES Investigators. Azithromycin for the secondary prevention of coronary events. N Engl J Med. 2005;352: 1637-1645.
- Vainas T, Stassen FR, Schurink GW, et al. Secondary prevention of atherosclerosis through Chlamydia pneumoniae eradication (SPACE Trial): a randomised clinical trial in patients with peripheral arterial disease. Eur J Vasc Endovasc Surg. 2005; 29:403-411.
- Blasi F, Bonardi D, Aliberti S, et al. Long-term azithromycin use in patients with chronic obstructive pulmonary disease and tracheostomy. Pulm Pharmacol Ther. 2010; 23:200-207.
- Albert RK, Connett J, Bailey WC, et al; for the COPD Clinical Research Network. Azithromycin for prevention of exacerbations of COPD. N Engl J Med. 2011; 365:689-698. Erratum in: N Engl J Med. 2012;366:1356.
- Dogra J. Oral azithromycin in extended dosage schedule for chronic, subclinical Chlamydia pneumoniae infection causing coronary artery disease: a probable cure in sight? Results of a controlled preliminary trial. Int J Gen Med. 2012;5:505-509.
- Berkhof FF, Doornewaard-ten Hertog NE, Uil SM, et al. Azithromycin and cough- specific health status in patients with chronic obstructive pulmonary disease and chronic cough: a randomised controlled trial. Respir Res. 2013;14:125.
- Faverio P, Arango A, Anzueto A, et al. Azithromycin-related cardiovascular events and deaths in severe septic patients. Chest. 2013;144(4_MeetingAbstracts):422A.
- Perry TW, Pugh MJ, Waterer GW, et al. Incidence of cardiovascular events after hospital admission for pneumonia. Am J Med. 2011;124:244-251.
- Martin-Loeches I, Lisboa T, Rodriguez A, et al. Combination antibiotic therapy with macrolides improves survival in intubated patients with community-acquired pneumonia. Intensive Care Med. 2010;36:612-620.
- Rodríguez A, Mendia A, Sirvent JM, et al; for the CAPUCI Study Group. Combination antibiotic therapy improves survival in patients with community-acquired pneumonia and shock. Crit Care Med. 2007;35:1493-1498.
- Clayton TC, Thompson M, Meade TW. Recent respiratory infection and risk of cardiovascular disease: case-control study through a general practice database. Eur Heart J. 2008;29:96-103.
- 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.
- Leinonen M, Saikku P. Evidence for infectious agents in cardiovascular disease and atherosclerosis. Lancet Infect Dis. 2002;2:11-17.
- Saikku P, Leinonen M, Tenkanen L, et al. Chronic Chlamydia pneumoniae infection as a risk factor for coronary heart disease in the Helsinki Heart Study. Ann Intern Med. 1992;116:273-278.
- Smeeth L, Thomas SL, Hall AJ, et al. Risk of myocardial infarction and stroke after acute infection or vaccination. N Engl J Med. 2004;351:2611-2618.
- Ridker PM, Rifai N, Pfeffer MA, et al. Inflammation, pravastatin, and the risk of coronary events after myocardial infarction in patients with average cholesterol levels. Circulation. 1998;98:839-844.