CSIRO PUBLISHING Azithromycin: more lethal than chloramphenicol? Ivan Stratov A,C, Justin Denholm B and Stephen J. Kent A
ADepartment of Microbiology and Immunology, University of Melbourne, Melbourne, Vic. 3010, Australia. BVictorian Infectious Diseases Service, Royal Melbourne Hospital, and Department of Medicine, University of
Melbourne, Parkville, Vic. 3010, Australia. Abstract. Azithromycin is commonly used in sexual health and respiratory medicine, often when the diagnosis is presumptive. A recent article by Ray et al. reported that 1 out of 20 000 courses of low-dose azithromycin was associated with (sudden) cardiovascular death (including 1 out of 4000 courses in high-risk cardiovascular patients), ascribing these deaths to azithromycin itself. Here, we critique the actual study and examine conflicting data from randomised control trials, animal studies and observational data. Additional keywords: cardiovascular health, drugs, mortality rate, presumptive diagnosis.
Received 7 August 2012, accepted 18 November 2012, published online 1 March 2013
Introduction
azithromycin, presumably the standard Z-Pak (Pfizer; 500 mg
Azithromycin is commonly used in sexual health medicine. The
on day 1, then 250 mg on days 2–5; a total of 1.5 g; peak serum
Melbourne Sexual Health Centre has dispensed over 24 700
concentration: 0.31–0.41 mg L–1,although results would
prescriptions for azithromycin to ambulatory patients since 2004
reasonably apply to chlamydia treatment doses (1 g stat;
(including 4069 in 2012), primarily to treat nongonococcal
peak serum concentration 0.82–1.07 mg L–1,) used by
urethritis, although an azithromycin responsive organism
sexual health physicians, as the authors discuss a potential
(e.g. Chlamydia trachomatis or Mycoplasma genitalium) is
only confirmed in ~30% of such Furthermore,
concentrations and the risk of cardiovascular death.,The
authors build a narrative around cardiac arrhythmia and
clinicians have not assessed, a practice used extensively
torsades de pointes (TdP) stating: ‘Our study was prompted
internationally and supported by randomised control trials.
by evidence that azithromycin is pro-arrhythmic. . .’ Seven case
However, a recent report by Ray et al. indicated oral
reports (none of which resulted in death) are cited with disparate
azithromycin resulted in 47 additional acute cardiovascular
ndings: QT prolongation with no TdP, QT prolongation with
(mostly sudden) deaths per million courses.The death rate
TdP and TdP without QT In five cases, the
(1 out of 21 277 courses) exceeds that ascribed to high-dose oral
cumulative dose was 1 g or less, one case occurred at day 7 (total:
chloramphenicol (1 out of 21 671),now essentially abandoned
3.5 g azithromycin) and one case was in a foreign language.
because of toxicity. Alarmingly, the death rate amongst high-
Issues such as sepsis, concurrent use of proarrhythmic drugs,
risk cardiovascular patients was five-fold higher, and estimated
hypokalaemia, congenital QT prolongation and underlying
at 245 excess cardiovascular deaths per million courses or 1 out
cardiomyopathy and dysrhythmia, however, confound the
of 4082 patients, with confidence limits stretching to 1 out of
1736 courses. On face value, this should be of great concern to
similarities suggestive of plagiarism, partially undermining
sexual health physicians. However, we suggest caution be
the basis upon which the study was predicated (this issue has
exercised interpreting their findings, based on a critique of
been brought to the attention of the relevant journal). For
the published report, and conflicting evidence arising from
example, Matsunaga et al. in described the following
multiple randomised control trials, animal data, observational
studies and extensive clinical experience. Given the potential
‘Upon presentation, the patient’s blood
impact of the association, we feel it appropriate to detail some of
pressure was 126/72 mmHg, heart rate 90bpm, respiratory rate 26 pm, temperature38.1C, and SpO2 was 93% on room air. HeBasis for the study had left basal crackles but no signs of right
The lead author has published numerous articles investigating
ventricular failure. Chest X-ray revealed new
the relationships between medications and sudden cardiac
left basal opacity. White blood cell count was
death.The current article studied 5-day courses of oral
16.1/mm3, 93% of which were neutrophils.Arterial blood gas showed pH of 7.49, pCO2 35,
category) was not actually associated with prolonged QTc
pO2 61, HCO3 26, SaO2 93% on room air.’
intervals (36% v. 31%, P = 0.18). Rather, QTc prolongationwas associated with sepsis, among other things (P = 0.001).
Russo et al. in described their patient with exactly the
Thus, azithromycin may be a surrogate marker for sepsis-
induced arrhythmia, rather than the cause. Significantly, no
‘Upon presentation, the patient’s blood
episode of TdP occurred among these 251 high-risk cardiac
pressure was 126/72 mmHg, heart rate 90
patients admitted with prolonged QTc, despite comorbidities
bpm, respiratory rate 26 pm, temperature
including hypokalaemia, hypomagnesaemia, sepsis and ongoing
38.1C, and SpO2 was 93% at room air. He
use of QT-prolonging drugs. Further, a prospective study of 47
presented left basal crackles but no signs of
healthy adults taking azithromycin (3 g over 5 days) found no
right ventricular failure. Chest X-ray revealed
significant increase in the QTc At the maximal limits
new left basal opacity. White blood cell count
of these observations, the following theoretical mathematical
was 16.1/mm3, 93% of which were neutrophils.
progression could be formulated: (a) 1 in 50 patients on
Arterial blood gas showed pH of 7.49, pCO2 35
azithromycin get prolonged QTc, (b) 1 in 250 patients with
mmHg, pO2 61 mmHg, HCO3 26 mmol/L, SaO2
prolonged QTc get TdP and (c) 1 in 20 patients with TdP
progress to sudden death. Hence, it would take at least250 000 courses of azithromycin to get one sudden death,
To our knowledge, there is only one report of arrhythmic death
more than ten-fold above that reported by Ray et al.
associated with azithromycin (anoxic brain injury afterventricular tachycardia);the patient was hypokalaemic and
Clinical experience
it was not clear if the patient actually took the prescribedazithromycin.
High-dose IV azithromycin is extensively used in monitoredintensive care unit patients with multiple comorbidities where
Randomised control trials
potentially lethal ventricular arrhythmias (including TdP) shouldbe more common than in community settings. Considering that
Importantly, Ray et al. do not discuss six randomised placebo-
very few episodes of TdP (perhaps 1 in 20) result in the
controlled trials in high-risk cardiovascular patients, all showing
data from Ray et al. suggest that ~1 in 1000 patients prescribed
that azithromycin reduced mortality (albeit nonsignificantly,
azithromycin would develop TdP, and ~1 in 200 high-risk
odds ratio: including sudden death.The largest
cardiovascular patients. It seems likely that this would have
(WIZARD) utilising azithromycin at a dosage of
been observed and reported more widely in routine azithromycin
1800 mg over 3 days, followed by 600 mg azithromycin
weekly for 11 weeks, indicated the benefit was greatest early(a 30% reduction in mortality and recurrent acute myocardial
Critique of the study by Ray et al
infarction; P < 0.05) and favoured those patients with the highestongoing cardiovascular risk: smokers, males and diabetics. The
Several problems relating to the study design should be
compromised by the 25% error rate acknowledged by theauthors,
Mammalian studies
developed and validated in-house.The use of International
In at least four mammalian models, TdP could not be induced
Classification of Diseases 9 (ICD9) codes 798.2 (‘unexplained’
by azithromycin,–despite specific efforts to do so
death), 798.8 (‘unattended’ death) and 799.9 (‘unknown’ death)
including supratherapeutic high-dose intravenous infusions,
raises doubts over the exact nature of the deaths. Given that there
coadministration of chloroquine and use of experimental
were only 112 cardiovascular deaths across the three main
atrioventricular block and hypokalaemia. Indeed, Milberg
cohorts (azithromycin 29, amoxicillin 42 and null group 41),
et al. characterised a mechanism that potentially explains
physician review of deaths seems feasible, making the data more
why azithromycin did not induce TdP even in the presence
robust and helping to mitigate concerns raised by researchers
about the accuracy of ‘validation’ committees.,Second, it
azithromycin actually suppressed TdP previously provoked
seems unavoidable (given the large size of the cohorts) to assign
‘presumptive’ diagnoses, for which antibiotics were prescribedbased on (1) a predetermined algorithm of ‘likely’ diagnoses and(2) recent Medicaid encounters. However, the data would be
Observational cohort data
more reliable if the authors had excluded the ~30% of encounters
The issue of drug induced corrected QT (QTc) prolongation
(over 720 000 instances) where clinical data were absent and no
leading to TdP is the subject of anecdotal case reports and long
antibiotic indication could be assigned, primarily due to
lists of implicated drugs have been compiled from adverse drug
telephone prescribing. Exhaustive propensity scoring for
reaction reports, with an ascribed mortality of A
underlying comorbidities cannot adequately compensate for
prospective 12-month observational study reported that 251
out of 900 patients admitted to advanced cardiac units had
respiratory rate and blood pressure. This is underscored by a
prolonged QTc Interestingly, the use of ‘QT-
higher rate of respiratory pathology (bronchitis, pneumonia,
prolonging medication’ (azithromycin was included in that
respiratory symptoms, chronic obstructive pulmonary disease
Azithromycin – more lethal than chloramphenicol?
and other respiratory problem) among patients prescribed
Conflicts of interest
azithromycin (44.9%) compared with amoxicillin (27.4%),
the clinical import of which is acknowledged by the authorsand substantiated by higher use of b-agonists (themselves
Acknowledgements
implicated in sudden cardiac With propensityscoring adjustment, respiratory pathology is calculated at
We thank Ms C Forrester and Dr G Flaker for helpful contributions.
45.1% versus 44.8%, respectively. However, across suchlarge cohorts, even a small 0.3% difference accounts for over
References
1000 high-risk respiratory patients, where some (or perhaps
1 Bradshaw CS, Tabrizi SN, Read TR, Garland SM, Hopkins CA, Moss
most) of the 18 excess deaths (allowing for the 1 : 4 ratio) may
LM, et al. Etiologies of nongonococcal urethritis: bacteria, viruses,
clearly be attributable to the underlying illness rather than the
and the association with orogenital exposure. J Infect Dis 2006; 193:
azithromycin itself. Third, one way to evaluate clinically how
well these cohorts were matched at baseline is to observe the
2 Cameron ST. Patient-delivered partner therapy for chlamydia: a
realistic public health measure in the UK. BJOG 2009; 116: 345–6.
early noncardiovascular mortality, which presumably relates to
antibiotic indication and reasonably reflects the severity of the
3 Ray WA, Murray KT, Hall K, Arbogast PG, Stein CM. Azithromycin
illnesses for which the antibiotic was prescribed. Curiously,
and the risk of cardiovascular death. N Engl J Med 2012; 366:
during the initial 48 h, no patients in the azithromycin group
suffered a noncardiovascular death, compared with 11 and 15
4 Wallerstein RO, Condit PK, Kasper CK, Brown JW, Morrison FR.
for the amoxicillin and no antibiotic groups, respectively; at
Statewide study of chloramphenicol therapy and fatal aplastic anemia.
72 h, the noncardiovascular deaths were 1, 19 and 28,
respectively. Again, allowing for the 1 : 4 patient ratio, one
5 Graham DJ, Campen D, Hui R, Spence M, Cheetham C, Levy G, et al.
would expect at least three or four early azithromycin deaths
Risk of acute myocardial infarction and sudden cardiac death in
to be noncardiovascular in nature. How, then, might the data be
patients treated with cyclo-oxygenase 2 selective and non-selectivenon-steroidal anti-inflammatory drugs: nested case–control study.
clinically explained? It is improbable that the azithromycin
group was actually healthier than the null group at baseline,
6 Habel LA, Cooper WO, Sox CM, Chan KA, Fireman BH, Arbogast
as the groups were matched, except for antibiotic indication.
PG, et al. ADHD medications and risk of serious cardiovascular events
Also, it would be contentious to conclude that azithromycin
in young and middle-aged adults. JAMA 2011; 306: 2673–83.
is potent enough to reduce noncardiovascular mortality so
quickly, although azithromycin potency may explain the
7 Ray WA, Chung CP, Murray KT, Hall K, Stein CM. Atypical
remarkably low noncardiovascular mortality seen after day 4
antipsychotic drugs and the risk of sudden cardiac death. N Engl J
of therapy (four azithromycin deaths v. 50 amoxicillin deaths).
Rather, the most plausible explanation for the minimal early
8 Ray WA, Meredith S, Thapa PB, Hall K, Murray KT. Cyclic
noncardiovascular mortality is that some of the early
antidepressants and the risk of sudden cardiac death. ClinPharmacol Ther 2004; 75: 234–41.
cardiovascular deaths were incorrectly assigned. Fourthly, no
9 Ray WA, Meredith S, Thapa PB, Meador KG, Hall K, Murray KT.
allowance is made for compliance, which is a major issue when
Antipsychotics and the risk of sudden cardiac death. Arch Gen
assigning adverse drug events in population-based studies. Psychiatry 2001; 58: 1161–7. doi:
Under research conditions, compliance with the 5-day
10 Ray WA, Murray KT, Meredith S, Narasimhulu SS, Hall K, Stein CM.
azithromycin z-pak is reported to be 82–An
Oral erythromycin and the risk of sudden death from cardiac causes. N
estimate of 10% noncompliance would, pro rata, reassign
Engl J Med 2004; 351: 1089–96. doi:
11 Pfizerlabs. Zithromax product information. New York: Pfizer Inc.;
(41 deaths), an issue that is not relevant to the amoxicillin
group, as the death rate was equal to those who took no
12 Wildfeuer A, Laufen H, Leitold M, Zimmermann T. Comparison of
antibiotics. It would be interesting to know what effect a net
the pharmacokinetics of three-day and five-day regimens ofazithromycin in plasma and urine. J Antimicrob Chemother 1993;
shift of six deaths would have on the hazard ratios. Fifthly, we
note in passing that their critical azithromycin mortality
13 Bergan T, Jorgensen NP, Olszewski W, Zhang Y. Azithromycin
cumulative graph (see fig. 1 of Ray et appears to be
pharmacokinetics and penetration to lymph. Scand J Infect Dis
incorrectly plotted – flat-lining at day 5 instead of day 6.
14 Höffler D, Koeppe P, Paeske B. Pharmacokinetics of azithromycin in
normal and impaired renal function. Infection 1995; 23: 356–61.
Good science is about meticulous observation. Retrospective
15 Schoenenberger RA, Haefeli WE, Weiss P, Ritz RF. Association of
research predicated on poorly scrutinised references and based
intravenous erythromycin and potentially fatal ventricular tachycardia
around patient cohorts for whom antibiotic indications and the
with Q-T prolongation (torsades de pointes). BMJ 1990; 300: 1375–6.
cause of death are imprecisely known is hazardous, particularly
16 Simkó J, Csilek A, Karaszi J, Lorincz I. Proarrhythmic potential of
when it contradicts prospectively collected data. We suggest
antimicrobial agents. Infection 2008; 36: 194–206. doi:
that further study of the association between azithromycin and
sudden death is warranted. In the meantime, azithromycin
17 Arellano-Rodrigo E, Garcia A, Mont L, Roque M. Torsade de pointes
should be used appropriately to control sexually transmissible
and cardiorespiratory arrest induced by azithromycin in a patient with
congenital long QT syndrome. Med Clin (Barc) 2001; 117: 118–9.
18 Huang BH, Wu CH, Hsia CP, Yin Chen C. Azithromycin-induced
32 Vainas T, Stassen FR, Schurink GW, Tordoir JH, Welten RJ, van den
torsade de pointes. Pacing Clin Electrophysiol 2007; 30: 1579–82.
Akker LH, et al. Secondary prevention of atherosclerosis through
chlamydia pneumoniae eradication (SPACE trial): a randomised
19 Kezerashvili A, Khattak H, Barsky A, Nazari R, Fisher JD.
clinical trial in patients with peripheral arterial disease. Eur J Vasc
Azithromycin as a cause of QT-interval prolongation and torsade
Endovasc Surg 2005; 29: 403–11.
de pointes in the absence of other known precipitating factors. J
33 Ohtani H, Taninaka C, Hanada E, Kotaki H, Sato H, Sawada Y, et al. Interv Card Electrophysiol 2007; 18: 243–6. doi:
Comparative pharmacodynamic analysis of Q-T interval prolongation
induced by the macrolides clarithromycin, roxithromycin, and
20 Kim MH, Berkowitz C, Trohman RG. Polymorphic ventricular
azithromycin in rats. Antimicrob Agents Chemother 2000; 44:
tachycardia with a normal QT interval following azithromycin. Pacing Clin Electrophysiol 2005; 28: 1221–2.
34 Milberg P, Eckardt L, Bruns HJ, Biertz J, Ramtin S, Reinsch N, et al.
Divergent proarrhythmic potential of macrolide antibiotics despite
21 Matsunaga N, Oki Y, Prigollini A. A case of QT-interval prolongation
similar QT prolongation: fast phase 3 repolarization prevents early
precipitated by azithromycin. N Z Med J 2003; 116: U666.
afterdepolarizations and torsade de pointes. J Pharmacol Exp Ther
22 Russo V, Puzio G, Siniscalchi N. Azithromycin-induced QT
prolongation in elderly patient. Acta Biomed 2006; 77: 30–2.
35 Thomsen MB, Beekman JD, Attevelt NJ, Takahara A, Sugiyama A,
23 Samarendra P, Kumari S, Evans SJ, Sacchi TJ, Navarro V. QT
Chiba K, et al. No proarrhythmic properties of the antibiotics
prolongation associated with azithromycin/amiodarone combination.
moxifloxacin or azithromycin in anaesthetized dogs with chronic-
Pacing Clin Electrophysiol 2001; 24: 1572–4. doi:
AV block. Br J Pharmacol 2006; 149: 1039–48.
24 Del Rosario ME, Weachter R, Flaker GC. Drug-induced QT
36 Fossa AA, Wisialowski T, Duncan JN, Deng S, Dunne M.
prolongation and sudden death. Mo Med 2010; 107: 53–8.
Azithromycin/chloroquine combination does not increase cardiac
25 Baker WL, Couch KA. Azithromycin for the secondary prevention of
instability despite an increase in monophasic action potential
coronary artery disease: a meta-analysis. Am J Health Syst Pharm
duration in the anesthetized guinea pig. Am J Trop Med Hyg 2007;
26 Cercek B, Shah PK, Noc M, Zahger D, Zeymer U, Matetzky S, et al.
37 Darpo B. Spectrum of drugs prolonging QT interval and the incidence
Effect of short-term treatment with azithromycin on recurrent
of torsades de pointes. Eur Heart J Suppl 2001; 3: K70–80.
ischaemic events in patients with acute coronary syndrome in the
Azithromycin in Acute Coronary Syndrome (AZACS) trial: a
38 Tisdale JE, Wroblewski HA, Overholser BR, Kingery JR, Trujillo TN,
Kovacs RJ. Prevalence of QT interval prolongation in patients
admitted to cardiac care units and frequency of subsequent
27 Grayston JT, Kronmal RA, Jackson LA, Parisi AF, Muhlestein JB,
administration of QT interval-prolonging drugs: a prospective,
Cohen JD, et al. Azithromycin for the secondary prevention of
observational study in a large urban academic medical center in the
coronary events. N Engl J Med 2005; 352: 1637–45. doi:
28 Gupta S, Leatham EW, Carrington D, Mendall MA, Kaski JC, Camm
39 Strle F, Maraspin V. Is azithromycin treatment associated with
AJ. Elevated Chlamydia pneumoniae antibodies, cardiovascular
prolongation of the Q-Tc interval? Wien Klin Wochenschr 2002;
infarction. Circulation 1997; 96: 404–7.
40 Chung CP, Murray KT, Stein CM, Hall K, Ray WA. A computer case
definition for sudden cardiac death. Pharmacoepidemiol Drug Saf
29 Muhlestein JB, Anderson JL, Carlquist JF, Salunkhe K, Horne BD,
Pearson RR, et al. Randomized secondary prevention trial of
41 Gottlieb SS. Dead is dead – artificial definitions are no substitute.
azithromycin in patients with coronary artery disease: primary
clinical results of the ACADEMIC study. Circulation 2000; 102:
42 Cazzola M, Santus P, D’Adda A, Pizzolato S, Di Marco F, Centanni S.
Acute effects of higher than standard doses of salbutamol and
30 O’Connor CM, Dunne MW, Pfeffer MA, Muhlestein JB, Yao L,
ipratropium on tiotropium-induced bronchodilation in patients with
Gupta S, et al. Azithromycin for the secondary prevention of
stable COPD. Pulm Pharmacol Ther 2009; 22: 177–82.
coronary heart disease events: the WIZARD study: a randomized
controlled trial. JAMA 2003; 290: 1459–66.
43 Devasia RA, Jones TF, Collier B, Schaffner W. Compliance with
azithromycin versus erythromycin in the setting of a pertussis
31 Stone AF, Mendall MA, Kaski JC, Edger TM, Risley P, Poloniecki J,
outbreak. Am J Med Sci 2009; 337: 176–8. doi:
et al. Effect of treatment for Chlamydia pneumoniae and Helicobacterpylori on markers of inflammation and cardiac events in patients
44 Martinez SM, Kemper CA, Haiduven D, Cody SH, Deresinski SC.
with acute coronary syndromes: South Thames Trial of Antibiotics in
Azithromycin prophylaxis during a hospitalwide outbreak of a
Myocardial Infarction and Unstable Angina (STAMINA). Circulation
pertussis-like illness. Infect Control Hosp Epidemiol 2001; 22:
Efavirenz Marca comercial: Clase de medicamento: Inhibidor de la transcriptasa inversa no análogo El efavirenz, conocido también como EFV o Sustiva, es un tipo de antirretroviral llamado inhibidor de latranscriptasa inversa no análogo de los nucleósidos (NNRTI). Esta clase de medicamentos bloquea latranscriptasa inversa, una proteína que necesita el VIH para multiplicarse.
284jumonline.qxp:Layout 1 3/17/09 2:30 PM Page 541 Case Report Inspissated Bile Syndrome in a Neonate Treated With Cefotaxime Sonographic Aid to Diagnosis, Management,and Follow-up Tamir Miloh, MD, Henrietta Kotlus Rosenberg, MD,Israel Kochin, MD, Nanda Kerkar, MD holelithiasis in neonates is usually pigmentary and composed of calciumbilirubinate.1,2 In children, cholelithisasis and inspis
CSIRO PUBLISHING