Importance
Calcium-channel blockers are metabolized by the cytochrome P450 3A4 (CYP3A4; EC 1.14.13.97) enzyme. Blood concentrations of these drugs may rise to harmful levels when CYP3A4 activity is inhibited. Clarithromycin is an inhibitor of CYP3A4 and azithromycin is not, which makes comparisons between these 2 macrolide antibiotics useful in assessing clinically important drug interactions.
Objective
To characterize the risk of acute adverse events following coprescription of clarithromycin compared with azithromycin in older adults taking a calcium-channel blocker.
Design, Setting, and Participants
Population-based retrospective cohort study in Ontario, Canada, from 2003 through 2012 of older adults (mean age, 76 years) who were newly coprescribed clarithromycin (n = 96 226) or azithromycin (n = 94 083) while taking a calcium-channel blocker (amlodipine, felodipine, nifedipine, diltiazem, or verapamil).
Main Outcomes and Measures
Hospitalization with acute kidney injury (primary outcome) and hospitalization with hypotension and all-cause mortality (secondary outcomes examined separately). Outcomes were assessed within 30 days of a new coprescription.
Results
There were no differences in measured baseline characteristics between the clarithromycin and azithromycin groups. Amlodipine was the most commonly prescribed calcium-channel blocker (more than 50% of patients). Coprescribing clarithromycin vs azithromycin with a calcium-channel blocker was associated with a higher risk of hospitalization with acute kidney injury (420 patients of 96 226 taking clarithromycin [0.44%] vs 208 patients of 94 083 taking azithromycin [0.22%]; absolute risk increase, 0.22% [95% CI, 0.16%-0.27%]; odds ratio [OR], 1.98 [95% CI, 1.68-2.34]). In a subgroup analysis, the risk was highest with dihydropyridines, particularly nifedipine (OR, 5.33 [95% CI, 3.39-8.38]; absolute risk increase, 0.63% [95% CI, 0.49%-0.78%]). Coprescription with clarithromycin was also associated with a higher risk of hospitalization with hypotension (111 patients of 96 226 taking clarithromycin [0.12%] vs 68 patients of 94 083 taking azithromycin [0.07%]; absolute risk increase, 0.04% [95% CI, 0.02%-0.07%]; OR, 1.60 [95% CI, 1.18-2.16]) and all-cause mortality (984 patients of 96 226 taking clarithromycin [1.02%] vs 555 patients of 94 083 taking azithromycin [0.59%]; absolute risk increase, 0.43% [95% CI, 0.35%-0.51%]; OR, 1.74 [95% CI, 1.57-1.93]).
Conclusions and Relevance
Among older adults taking a calcium-channel blocker, concurrent use of clarithromycin compared with azithromycin was associated with a small but statistically significant greater 30-day risk of hospitalization with acute kidney injury. These findings support current safety warnings regarding concurrent use of CYP3A4 inhibitors and calcium-channel blockers.
The commonly used macrolide antibiotics clarithromycin and erythromycin are clinically important inhibitors of the cytochrome P450 3A4 (CYP3A4; EC 1.14.13.97) enzyme, while azithromycin is much less so.1,2 In older adults, coprescription of clarithromycin or erythromycin with a CYP3A4 metabolized statin (atorvastatin, simvastatin, and lovastatin) has been shown to be associated with a greater risk of hospitalization with rhabdomyolysis, hospitalization with acute kidney injury, and all-cause mortality compared with azithromycin coprescription.3
Calcium-channel blockers are a popular class of antihypertensive drugs that are metabolized by the CYP3A4 enzyme. In pharmacokinetic studies, coadministration of various inhibitors of this enzyme (eg, erythromycin, antifungals, protease inhibitors, and grapefruit juice) raised plasma calcium-channel blocker concentrations by up to 500%.4-6 As a result, there is the possibility of excessive systemic calcium-channel blocker concentration and associated toxicity with concurrent use of a CYP3A4 inhibitor.
Enhanced blood pressure lowering was observed in several studies (up to 12 healthy volunteers) after a CYP3A4 inhibitor was administrated with a calcium-channel blocker.4,7,8 Several case reports described hospitalization with hypotension soon after a CYP3A4 inhibitor was taken with a calcium-channel blocker.9-12 Moreover, a population-based case-crossover study of older adults found a greater risk of hospitalization with hypotension when a calcium-channel blocker was coprescribed with erythromycin or clarithromycin compared with azithromycin.13 Currently, the US Food and Drug Administration warns that “serious adverse reactions have been reported in patients taking clarithromycin concomitantly with CYP3A4 substrates, which includes hypotension with calcium-channel blockers metabolized by CYP3A4 (eg, verapamil, amlodipine, diltiazem).”14 Yet calcium-channel blockers and clarithromycin continue to be frequently coprescribed in routine care.
When hypotension occurs, the kidney is particularly prone to acute ischemic injury from poor perfusion. Acute kidney injury is a clinically important event that impacts morbidity, mortality, and resource use.15 Despite this knowledge, the risk of acute kidney injury following coprescription of clarithromycin with a calcium-channel blocker is unknown. Therefore, we conducted a population-based cohort study of older adults to investigate the interaction between calcium-channel blockers and the antibiotic clarithromycin with a focus on acute kidney injury.
We conducted this study at the Institute for Clinical Evaluative Sciences (ICES) according to a prespecified protocol that was approved by the research ethics board at Sunnybrook Health Sciences Centre (Toronto, Canada). Participant informed consent was not required for this study.
We conducted a population-based, retrospective cohort study of older adults from June 2003 through March 2012 using linked health care databases in Ontario, Canada. Ontario has approximately 13 million residents, 14% of whom are aged 65 years or older.16 Residents have universal access to hospital care and physician services and those aged 65 years or older have universal prescription drug coverage. The reporting of this study followed guidelines for observational studies (eTable 1 in the Supplement).17
We ascertained patient characteristics, drug use, covariate information, and outcome data using records from 5 databases. We obtained vital statistics from the Ontario Registered Persons Database, which contains demographic information on all Ontario residents who have ever been issued a health card. We used the Ontario Drug Benefit Program database to identify prescription drug use. This database contains highly accurate records of all outpatient prescriptions dispensed to patients aged 65 years or older, with an error rate of less than 1%.18 We identified diagnostic and procedural information on all hospitalizations from the Canadian Institute for Health Information’s Discharge Abstract Database. We obtained covariate information from the Ontario Health Insurance Plan database, which includes health claims for inpatient and outpatient physician services. We also used the ICES Physician Database to ascertain antibiotic prescriber information. Previously, we have used these databases to research adverse drug events and health outcomes (including outcomes of acute kidney injury and health services).3,19-22
With the exception of infection type and prescriber (missing in the cohort by approximately 50% for infection type and 14% for prescriber), the databases were complete for all variables used in this study. International Classification of Diseases, 9th revision (ICD-9; pre-2002) and 10th revision (ICD-10; post-2002) codes were used to assess baseline comorbidities in the 5 years prior to receipt of the relevant coprescription (eTable 2 in the Supplement). Codes used to ascertain outcomes are detailed in eTable 3 in the Supplement, which lists only ICD-10 codes because all events would have occurred after the implementation of this coding system. A subpopulation in southwestern Ontario had outpatient serum creatinine measurements available before a new antibiotic coprescription and was in the catchment area of 12 hospitals in which linked inpatient serum creatinine values were also available.23
We established a cohort of older adults in Ontario, Canada, with continuous calcium-channel blocker use who also had evidence of a coprescription for the CYP3A4 inhibitor clarithromycin. For the referent group, we considered older adults coprescribed azithromycin. We previously demonstrated that clarithromycin and azithromycin in Ontario are prescribed for near-identical infections (eg, respiratory tract, sinus, and oropharyngeal infections), prescribed by the same type of physicians (approximately 75% primary care physicians), prescribed to patients with similar comorbidities, and are not statistically different in their risk of hospitalization with acute kidney injury or hospitalization with hypotension in the absence of another interacting drug (clarithromycin vs azithromycin: acute kidney injury odds ratio [OR], 1.06 [95% CI, 0.71-1.58]; hypotension OR, 1.21 [95% CI, 0.61-2.41]).24 Thus, comparison of outcomes among older adults prescribed 1 of these 2 antibiotics serves as a useful model in which to study CYP3A4 drug interactions with calcium-channel blockers in routine clinical practice. For reasons of statistical power and result interpretation, we elected a priori not to study the CYP3A4 inhibitor erythromycin, because in Ontario this drug was prescribed less than once for every 20 clarithromycin prescriptions.3
The date of the clarithromycin or azithromycin prescription served as the index date (referred to as cohort entry date or start time for follow-up). We considered the following CYP3A4 metabolized calcium-channel blockers: amlodipine, felodipine, nifedipine, verapamil, and diltiazem. Continuous use was defined as a second consecutive prescription claim for the same calcium-channel blocker. To ensure that the calcium-channel blocker and macrolide antibiotic were coprescribed, the dates covered by the calcium-channel blocker prescription had to overlap with the dates covered by the antibiotic prescription; in our cohort, the median (interquartile range [IQR]) overlap in each of the 2 macrolide antibiotic groups was 100% (100%-100%).
We excluded the following patients from analysis: (1) those in their first year of eligibility for prescription drug coverage (aged 65 years) to avoid incomplete medication records, (2) those who received a prescription for more than 1 type of antibiotic on the index date to compare mutually exclusive groups, (3) those who received any antibiotic in the 30 days prior to the index date to ensure new antibiotic use and to exclude patients with severe infections that failed to respond to initial antibiotic treatment, (4) those with 1 or more prescriptions for a nonstudy calcium-channel blocker to ensure any observed associations were due to the study drugs, (5) those who were discharged from the hospital in the 2 days prior to their index date to ensure these were new outpatient antibiotic prescriptions (because in Ontario, patients continuing an antibiotic treatment initiated in the hospital would have their oral outpatient antibiotic prescription dispensed on the same day or the day after hospital discharge), (6) those who had potent CYP3A4 inhibitors (such as protease inhibitors or antifungals) dispensed within the 30 days prior to the index date,25 and (7) those with a history of end-stage renal disease receiving chronic dialysis because the assessment of acute kidney injury is no longer relevant in such individuals. A patient could only enter the cohort once.
The primary outcome was hospitalization with acute kidney injury and the 2 secondary outcomes were hospitalization with hypotension and all-cause mortality. We assessed these outcomes within 30 days of the index date (because macrolide antibiotics are prescribed for short durations and adverse events due to drug interactions would occur soon thereafter). The diagnostic codes used to identify the outcomes are presented in eTable 3 in the Supplement. For hospitalization records, up to 25 diagnostic codes can be assigned per hospitalization. As such, patients with codes for multiple study outcomes were accounted for in the assessment of each outcome.
In Ontario, we previously demonstrated that a database code for hospitalization with acute kidney injury identifies a median absolute acute increase in serum creatinine of 1.11 mg/dL (to convert to micromoles per liter, multiply by 88.4) at the time of hospital presentation (IQR, 0.49 to 2.26) above the most recent value prior to hospitalization, and the absence of such a code represents no significant change in serum creatinine (0.07 mg/dL; IQR, −0.04 to 0.23).26 As the absolute increase in a serum creatinine value becomes more extreme (ie, higher levels of acute kidney injury) a code is more likely to be recorded for a particular diagnosis. Although the specificity is greater than 95%, the sensitivity of the hospital diagnosis code is limited particularly for milder forms of the condition. Specifically, the incidence of acute kidney injury, as defined by the diagnosis code, can be underestimated up to 5-fold compared with definitions using serum creatinine measurements. For this reason we examined a subpopulation with linked hospital laboratory values and defined hospitalization with acute kidney injury by evidence of an absolute increase in serum creatinine of 0.3 mg/dL or more from the baseline (preantibiotic) value or a relative increase of 50% or more.27
The code for hospitalization with hypotension has not been validated in our region but is expected to be insensitive. Mortality data are coded accurately in our region with a sensitivity of 97.8% and specificity of 100% for the finding of death.28
We compared baseline characteristics between those coprescribed clarithromycin or azithromycin with calcium-channel blockers using standardized differences.29,30 This metric describes differences between group means relative to the pooled standard deviation and is considered a clinically meaningful difference if greater than 10%. We expressed the risk of developing an outcome in both relative and absolute terms. Absolute risk was also expressed as the number needed to harm (NNH) (1 / absolute risk difference). This measure indicates how many patients need to receive a coprescription with clarithromycin to cause harm to 1 patient who otherwise would not have been harmed if all patients received a coprescription with azithromycin (a lower number indicating greater harm). The NNH was calculated for ease of interpretation, and not to imply causality.
We used PROC LOGISTIC, SAS version 9.2, for multivariable logistic regression analyses to estimate ORs and 95% CIs. We adjusted for 17 potential confounders: age, sex, baseline use of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, nonsteroidal anti-inflammatory agents, oral hypoglycemic agents or insulin, nonpotassium-sparing diuretics or potassium-sparing diuretics, statins, β-blockers, β2-agonists, anticholinergics, and corticosteroids (with all drugs defined by evidence of at least 1 prescription in the preceding 6 months), as well as baseline evidence of chronic kidney disease, coronary artery disease, cerebrovascular disease, peripheral vascular disease, congestive heart failure, and major cancers (defined using hospital diagnosis and physician claim codes in the preceding 5 years; in Ontario the validated algorithm for chronic kidney disease identifies older adults with a median estimated glomerular filtration rate [eGFR] of 38 mL/min per 1.73 m2 [IQR, 27-52], whereas its absence identifies those with a median eGFR of 69 mL/min per 1.73 m2 [IQR, 56-82]).31
We also evaluated the association between coprescription and hospitalization with acute kidney injury in 3 prespecified subgroups. The first subgroup analysis consisted of the type of calcium-channel blocker. The second subgroup analysis consisted of patients with and without chronic kidney disease. It is recommended that the clarithromycin dose be reduced by 50% in chronic kidney disease due to impaired clearance, but in practice this seldom occurs.32 Thus, we thought the relative association between coprescription with clarithromycin and hospitalization with acute kidney injury might be greater in those with chronic kidney disease than in those without. The third subgroup analysis examined effect modification by statin use, as we recently showed that coprescriptions with clarithromycin lead to rhabdomyolysis and acute kidney injury from statin toxicity.3
Odds ratios can be interpreted as relative risks (appropriate given the incidences observed). We conducted all analyses with SAS version 9.2. This includes additional analyses we undertook after knowledge of the primary results (see Results section). In all outcome analyses we interpreted 2-tailed P values lower than .05 as statistically significant.
We identified 190 309 patients taking a calcium-channel blocker who received a coprescription for clarithromycin (n = 96 226) or azithromycin (n = 94 083) (eFigure 1 in the Supplement). Baseline characteristics of the 2 groups were nearly identical, including type and dose of calcium-channel blocker used (Table 1; all standardized differences for 38 characteristics were less than 10%). More than half the patients in each group received amlodipine. The median dosage for coprescribed clarithromycin was 1000 mg daily for 10 days and 300 mg daily for 5 days for azithromycin, which was consistent with drug prescribing references.32 Coprescriptions of calcium-channel blockers and clarithromycin continued to occur in each year of the study period including the years of most recent accrual (Table 1).
Results for the outcome of hospitalization with acute kidney injury are presented in Table 2 (using hospital-based diagnosis codes). Coprescribing clarithromycin with a calcium-channel blocker was associated with a higher risk compared with coprescribing azithromycin (420 patients of 96 226 taking clarithromycin [0.44%] vs 208 patients of 94 083 taking azithromycin [0.22%]; OR, 1.98 [95% CI, 1.68-2.34]). In absolute terms, coprescription with clarithromycin resulted in a 0.22% (95% CI, 0.16%-0.27%) higher incidence of hospitalization with acute kidney injury. The NNH was 464 (95% CI, 374-609).
Results from subgroup analyses are presented in Figure 1 and Figure 2. When examined by type of calcium-channel blocker, the risk of hospitalization with acute kidney injury was highest among patients coprescribed clarithromycin with nifedipine (OR, 5.33 [95% CI, 3.39-8.38]; absolute risk increase, 0.63% [95% CI, 0.49%-0.78%]; P value for interaction, <.001 with amlodipine as the reference group). The corresponding NNH was lowest at 160 (95% CI, 128-205). Median doses of clarithromycin were similar among patients with and without chronic kidney disease (1000 mg/d in each group; standardized difference, <1%). The risk of hospitalization with acute kidney injury following clarithromycin coprescription was not modified by the presence of chronic kidney disease (P value for interaction = .47) or statin use (P value for interaction= .48). However, the NNH was lower in patients with chronic kidney disease compared with those without it (95 [95% CI, 70-145] for patients with chronic kidney disease vs 723 [95% CI, 545-1059] for patients without.
The baseline characteristics for the subpopulation with linked hospital serum creatinine measurements were similar between treatment groups (clarithromycin, n = 3164; azithromycin, n = 2094) (eTable 4 in the Supplement). Approximately 40% of patients had a baseline eGFR lower than 60 mL/min per 1.73 m2. With the outcome of hospitalization with acute kidney injury defined using changes in serum creatinine, coprescribing clarithromycin with a calcium-channel blocker remained associated with a higher risk compared with coprescribing azithromycin (63 patients of 3164 taking clarithromycin [1.99%] vs 26 patients of 2094 taking azithromycin [1.24%]; OR, 1.62 [95% CI, 1.02-2.56]) (eTable 5 in the Supplement). The absolute risk increase was 0.75% (95% CI, 0.03%-1.42%) and the NNH was 133 (95% CI, 70-3004).
Coprescribing clarithromycin with a calcium-channel blocker was associated with a higher risk of hospitalization with hypotension (111 patients of 96 226 taking clarithromycin [0.12%] vs 68 patients of 94 083 taking azithromycin [0.07%]; absolute risk increase, 0.04% [95% CI, 0.02%-0.07%]; OR, 1.60 [95% CI, 1.18-2.16]) and all-cause mortality (984 patients of 96 226 taking clarithromycin [1.02%] vs 555 patients of 94 083 taking azithromycin [0.59%]; absolute risk increase, 0.43% [95% CI, 0.35%-0.51%]; OR, 1.74 [95% CI, 1.57-1.93]), compared with coprescribing azithromycin (Table 2).
The primary associations proved robust in multiple additional analyses. First, we adjusted for 17 relevant confounders and found no meaningful difference with unadjusted results for all 3 outcomes (Table 2). Second, we stratified by physician identification number to account for potential differences in prescribing practices and observed results consistent with the above (eTable 6 in the Supplement). Third, we restricted the cohort only to those coprescribed a calcium-channel blocker with clarithromycin, and compared those coprescribed a higher dose of clarithromycin (1000 mg/d; n = 28 591) with those coprescribed a lower dose of clarithromycin (500 mg/d; n = 65 801) (eTable 7 in the Supplement). Receiving a higher dose of clarithromycin with a calcium-channel blocker was associated with a higher risk of hospitalization with acute kidney injury compared with a lower dose (307 patients of 28 591 taking a high dose [0.47%] vs 95 patients of 65 801 taking a low dose [0.33%]; absolute risk increase, 0.13% [95% CI, 0.05%-0.22%]; OR, 1.42 [95% CI, 1.131.79]). Finally, we compared 30-day outcomes in a restricted cohort at 3 periods: the time of coprescription, 90 days before the coprescription, and 90 days after the coprescription (clarithromycin, n = 53 070; azithromycin, n = 52 244). Full methods and cohort selection are presented in eFigure 2 in the Supplement. Baseline characteristics are presented in eTable 8 in the Supplement and results are in Table 3. As observed in our primary analyses, coprescribing clarithromycin with a calcium-channel blocker was associated with a higher 30-day risk of hospitalization with acute kidney injury (OR, 2.92 [95% CI, 1.93-4.42]) and hypotension (OR, 3.57 [95% CI, 1.63-7.81]) compared with coprescribing azithromycin. Moreover, no significant difference in the 30-day risk of these outcomes was observed when the start of follow-up was 90 days prior and 90 days following the coprescription.
In this population-based study of older adults, we observed that coprescribing clarithromycin with a calcium-channel blocker was common in routine care. This coprescription was associated with a higher risk of hospitalization with acute kidney injury, hypotension, and all-cause mortality compared with coprescription with azithromycin. These findings proved to be robust in multiple additional analyses. Although the absolute increases in the risks were small, these outcomes have important clinical implications. Our results suggest it is possible that hundreds of hospitalizations and deaths in our region may have been associated with this largely preventable drug-drug interaction. This burden on the health care system, given the high costs of managing acute kidney injury, might have been avoided.35
Macrolide antibiotics are frequently prescribed medications and different agents within this class (ie, clarithromycin and azithromycin) are used for similar clinical indications. Since clarithromycin is an inhibitor of CYP3A4 metabolism, concurrent use with a CYP3A4-metabolized calcium-channel blocker may intensify the calcium-channel blocker effect. This would not be expected with azithromycin, which is only a weak inhibitor of CYP3A4 metabolism. The kidney is especially prone to injury from poor perfusion and thus hospitalization with acute kidney injury represents a clinically important consequence of the interaction between macrolide antibiotics and calcium-channel blockers. It is reasonable to assume that the mechanism of kidney injury was hemodynamic in nature as we also noted a higher risk of hospitalization with hypotension in this setting. This finding is consistent with the observation made in a prior case-crossover study where a higher risk of hospitalization with hypotension was detected (OR, 3.7 [95% CI, 2.3-6.1]).13
We also observed that the risk of hospitalization with acute kidney injury was most pronounced when clarithromycin was coprescribed with nifedipine, followed by felodipine and amlodipine. These 3 drugs are dihydropyridines and are selective arterial vasodilators.36 The 2 nondihydropyridines (diltiazem and verapamil) are less potent vasodilators but have the additional properties of direct negative effects on cardiac contractility and conduction. However, these subgroup results must be interpreted cautiously as estimates were also less precise from the small sample sizes (particularly for verapamil).
As the kidney plays an important role in the elimination of clarithromycin, prescribing guidelines state that the dose needs to be reduced in patients with chronic kidney disease.32 However, we demonstrated that this rarely occurs in routine practice. Furthermore, since chronic kidney disease is the most potent risk factor for acute kidney injury, it is possible that the excess toxicity of calcium-channel blockers from clarithromycin coprescription would be most evident in patients with chronic kidney disease. Although we observed no greater relative risk of hospitalization in this group of patients, the absolute NNH from the coprescription was far lower in patients with chronic kidney disease than in those without.
It is concerning that clarithromycin continues to be coprescribed with calcium-channel blockers despite previous studies, as well as warnings in drug prescribing references.14,32 The results of this study reinforce our knowledge about the dangers of this type of drug interaction and have the potential to influence prescribing to prevent adverse events. Furthermore, our study highlights the need for quality improvement initiatives that will mitigate the clinical effects of such drug interactions. Potential strategies may include temporary cessation of the calcium-channel blocker for the duration of clarithromycin therapy or selection of a non–CYP3A4-inhibiting antibiotic when clinically appropriate.
Our study has several strengths. To our knowledge it is the first population-based study to assess hospitalization with acute kidney injury from CYP3A4 inhibition of common antihypertensive medications. The use of Ontario’s health care databases with data on universal prescription drug coverage in older adults provided us with a large representative sample of patients who received coprescriptions. This allowed us to estimate the risks of uncommon but serious adverse events with good precision and excellent external validity. Although the absolute risk increases may have been underestimated due to the limited sensitivity of the diagnostic codes, we captured the more severe forms of the conditions (ie, requiring hospitalization), making these findings of particular interest to clinicians and policy decision makers. We used the antibiotic azithromycin as a comparator group to clarithromycin to reduce concerns about confounding by indication. Furthermore, there was marked similarity of measured baseline characteristics in the 2 groups. Additionally, we previously confirmed that clarithromycin and azithromycin are not different in the 30-day risk of hospitalization with acute kidney injury in the absence of other interacting medications.24 We also verified that the calcium-channel blocker prescriptions alone (when assessed 90 days before and after the antibiotic coprescription) did not impact 30-day outcomes. This information reinforces the primary results of our study.
Our study does have some limitations. Prospective data collection with independent outcome adjudication would be the preferred methodology. However, conduct of such a study might not be possible if physicians were required to intervene after learning about a potential coprescription. In this study, we analyzed retrospective data using administrative diagnosis codes, which we did not expect to be assigned in a different manner in the 2 antibiotic groups. Nonetheless, these codes have their shortcomings, and for this reason we supplemented our primary outcome findings by observing a subpopulation with serum creatinine values and showed a similar signal of hospitalization with acute kidney injury following clarithromycin coprescription. Our findings can only be generalized to older adults, as younger patients are often healthier and may not be as susceptible to drug-drug interactions.37 As with all observational studies, we may have failed to account for important unknown or unmeasured confounding variables. Also, drug-drug interactions are complex and factors beyond CYP3A4 inhibition may have affected the results. For example, we previously observed a small difference in all-cause mortality between clarithromycin and azithromycin alone that we could not explain when causes of death were examined (OR, 1.27 [95% CI, 1.04-1.55]).24 Thus, it is possible that some of deaths observed in the present study (in which the observed OR was 1.74 [95% CI, 1.57-1.93]) could have been attributed to clarithromycin itself or the unique clinical circumstances that led to clarithromycin being chosen over azithromycin. As such, we cannot be entirely certain that the observed associations are causal or attributable to the mechanism we suggest. However, the results are consistent with the known increase in blood calcium-channel blocker concentrations seen after a CYP3A4 inhibitor is used.
Among older adults taking a calcium-channel blocker, concurrent use of clarithromycin compared with azithromycin was associated with a small but statistically significant greater 30-day risk of hospitalization with acute kidney injury. These findings support current safety warnings regarding concurrent use of CYP3A4 inhibitors and calcium-channel blockers.
Corresponding Author: Amit X. Garg, MD, PhD, London Kidney Clinical Research Unit, Room ELL-101, Westminster, London Health Sciences Centre, 800 Commissioners Rd E, London, ON, Canada N6A 4G5 (amit.garg@lhsc.on.ca).
Published Online: November 9, 2013. doi:10.1001/jama.2013.282426.
Author Contributions: Dr Garg had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Gandhi, Fleet, Bailey, Rehman, Garg.
Acquisition of data: Gandhi, McArthur, Garg.
Analysis and interpretation of data: Gandhi, Fleet, Bailey, McArthur, Wald, Garg.
Drafting of the manuscript: Gandhi, Bailey, Garg.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: McArthur, Garg.
Obtained funding: Garg.
Administrative, technical, or material support: Fleet, Wald, Rehman, Garg.
Study supervision: Garg.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Wald reported receiving unrestricted research funding and speaking fees from Alere, and honoraria from Thrasos for serving on a scientific advisory board. Dr Garg reported receiving an investigator-initiated grant from Astellas and Roche to support a Canadian Institutes of Health Research study in living kidney donors, and his institution received unrestricted research funding from Pfizer. No other disclosures were reported.
Funding/Support: This project was conducted at the Institute for Clinical Evaluative Sciences (ICES) Western Site. The ICES is funded by an annual grant from the Ontario Ministry of Health and Long-term Care. ICES Western is funded by an operating grant from the Academic Medical Organization of Southwestern Ontario. This project was conducted with members of the provincial ICES Kidney, Dialysis, and Transplantation Research Program, which receives programmatic grant funding from the Canadian Institutes of Health Research.
Role of the Sponsor: The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The opinions, results, and conclusions are those of the authors, and no endorsement by the Institute for Clinical Evaluative Sciences, Ontario Ministry of Health and Long-Term Care, or Academic Medical Organization of Southwestern Ontario is intended or should be inferred.
Additional Contributions: We thank Brogan Inc, Ottawa, for use of its drug product and therapeutic class database. We thank Gamma Dynacare for their use of the outpatient laboratory database and the team at London Health Sciences Centre, St Joseph’s Health Care, and the Thames Valley Hospitals for providing access to the Cerner laboratory database. We thank Stephanie Dixon, PhD, Lihua Li, MSc, and Salimah Shariff, PhD (all from ICES Western in London, Canada), for administrative support, analytic advice, and graphing contributions to this article. None received compensation for their role in the study.
2.Westphal
JF. Macrolide-induced clinically relevant drug interactions with cytochrome P-450A (CYP) 3A4: an update focused on clarithromycin, azithromycin and dirithromycin.
Br J Clin Pharmacol. 2000;50(4):285-295.
PubMedGoogle ScholarCrossref 3.Patel
AM, Shariff
S, Bailey
DG,
et al. Statin toxicity from macrolide antibiotic coprescription: a population-based cohort study.
Ann Intern Med. 2013;158(12):869-876.
PubMedGoogle ScholarCrossref 4.Jalava
KM, Olkkola
KT, Neuvonen
PJ. Itraconazole greatly increases plasma concentrations and effects of felodipine.
Clin Pharmacol Ther. 1997;61(4):410-415.
PubMedGoogle ScholarCrossref 5.Glesby
MJ, Aberg
JA, Kendall
MA,
et al; Adult AIDS Clinical Trials Group A5159 Protocol Team. Pharmacokinetic interactions between indinavir plus ritonavir and calcium-channel blockers.
Clin Pharmacol Ther. 2005;78(2):143-153.
PubMedGoogle ScholarCrossref 6.Bailey
DG, Bend
JR, Arnold
JM, Tran
LT, Spence
JD. Erythromycin-felodipine interaction: magnitude, mechanism, and comparison with grapefruit juice.
Clin Pharmacol Ther. 1996;60(1):25-33.
PubMedGoogle ScholarCrossref 7.Bailey
DG, Spence
JD, Munoz
C, Arnold
JM. Interaction of citrus juices with felodipine and nifedipine.
Lancet. 1991;337(8736):268-269.
PubMedGoogle ScholarCrossref 8.Josefsson
M, Zackrisson
AL, Ahlner
J. Effect of grapefruit juice on the pharmacokinetics of amlodipine in healthy volunteers.
Eur J Clin Pharmacol. 1996;51(2):189-193.
PubMedGoogle ScholarCrossref 10.Kaeser
YA, Brunner
F, Drewe
J, Haefeli
WE. Severe hypotension and bradycardia associated with verapamil and clarithromycin.
Am J Health Syst Pharm. 1998;55(22):2417-2418.
PubMedGoogle Scholar 11.Gerónimo-Pardo
M, Cuartero-del-Pozo
AB, Jiménez-Vizuete
JM, Cortiñas-Sáez
M, Peyró-García
R. Clarithromycin-nifedipine interaction as possible cause of vasodilatory shock.
Ann Pharmacother. 2005;39(3):538-542.
PubMedGoogle ScholarCrossref 12.Reed
M, Wall
GC, Shah
NP, Heun
JM, Hicklin
GA. Verapamil toxicity resulting from a probable interaction with telithromycin.
Ann Pharmacother. 2005;39(2):357-360.
PubMedGoogle ScholarCrossref 13.Wright
AJ, Gomes
T, Mamdani
MM, Horn
JR, Juurlink
DN. The risk of hypotension following coprescription of macrolide antibiotics and calcium-channel blockers.
CMAJ. 2011;183(3):303-307.
PubMedGoogle ScholarCrossref 15.Coca
SG, Peixoto
AJ, Garg
AX, Krumholz
HM, Parikh
CR. The prognostic importance of a small acute decrement in kidney function in hospitalized patients: a systematic review and meta-analysis.
Am J Kidney Dis. 2007;50(5):712-720.
PubMedGoogle ScholarCrossref 17.von Elm
E, Altman
DG, Egger
M, Pocock
SJ, Gøtzsche
PC, Vandenbroucke
JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.
J Clin Epidemiol. 2008;61(4):344-349.
PubMedGoogle ScholarCrossref 18.Levy
AR, O’Brien
BJ, Sellors
C, Grootendorst
P, Willison
D. Coding accuracy of administrative drug claims in the Ontario Drug Benefit database.
Can J Clin Pharmacol. 2003;10(2):67-71.
PubMedGoogle Scholar 19.Zhao
YY, Weir
MA, Manno
M,
et al. New fibrate use and acute renal outcomes in elderly adults: a population-based study.
Ann Intern Med. 2012;156(8):560-569.
PubMedGoogle ScholarCrossref 20.Shih
AWY, Weir
MA, Clemens
KK,
et al. Oral bisphosphonate use in the elderly is not associated with acute kidney injury.
Kidney Int. 2012;82(8):903-908.
PubMedGoogle ScholarCrossref 21.Molnar
AO, Coca
SG, Devereaux
PJ,
et al. Statin use associates with a lower incidence of acute kidney injury after major elective surgery.
J Am Soc Nephrol. 2011;22(5):939-946.
PubMedGoogle ScholarCrossref 22.Siddiqui
N, Coca
SG, Devereaux
PJ,
et al. Secular trends in acute dialysis after major elective surgery—1995 to 2009.
CMAJ. 2012;184(11):1237–1245.
PubMedGoogle ScholarCrossref 23.Gandhi
S, Shariff
SZ, Beyea
MM,
et al. Identifying geographical regions serviced by hospitals to assess laboratory-based outcomes.
BMJ Open. 2013;3(1).
PubMedGoogle Scholar 24.Fleet
JL, Shariff
SZ, Bailey
DG,
et al. Comparing 2 types of macrolide antibiotics for the purpose of assessing population-based drug interactions.
BMJ Open. 2013;3(7):1-8.
PubMedGoogle ScholarCrossref 26.Hwang
YJ, Shariff
SZ, Gandhi
S,
et al. Validity of the International Classification of Diseases, 10th Revision code for acute kidney injury in elderly patients at presentation to the emergency department and at hospital admission.
BMJ Open. 2012;2(6):1-11.
PubMedGoogle ScholarCrossref 28.Jha
P, Deboer
D, Sykora
K, Naylor
CD. Characteristics and mortality outcomes of thrombolysis trial participants and nonparticipants: a population-based comparison.
J Am Coll Cardiol. 1996;27(6):1335-1342.
PubMedGoogle ScholarCrossref 29.Mamdani
M, Sykora
K, Li
P,
et al. Reader’s guide to critical appraisal of cohort studies: 2. Assessing potential for confounding.
BMJ. 2005;330(7497):960-962.
PubMedGoogle ScholarCrossref 30.Austin
PC. Using the standardized difference to compare the prevalence of a binary variable between 2 groups in observational research.
Comm Stat Simul Comput. 2009;38(6):1228-1234. doi:10.1080/03610910902859574.
Google ScholarCrossref 31.Fleet
JL, Dixon
SN, Shariff
SZ,
et al. Detecting chronic kidney disease in population-based administrative databases using an algorithm of hospital encounter and physician claim codes.
BMC Nephrol. 2013;14(1):81.
PubMedGoogle ScholarCrossref 32.Compendium of Pharmaceuticals and Specialties. CPS 2012: Compendium of Pharmaceuticals and Specialties: The Canadian Drug Reference for Health Professionals.47th ed. Ottawa: Canadian Pharmaceutical Association; 2012.
33.Charlson
ME, Pompei
P, Ales
KL, MacKenzie
CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.
J Chronic Dis. 1987;40(5):373-383.
PubMedGoogle ScholarCrossref 34.Quan
H, Sundararajan
V, Halfon
P,
et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data.
Med Care. 2005;43(11):1130-1139.
PubMedGoogle ScholarCrossref 35.Chertow
GM, Burdick
E, Honour
M, Bonventre
JV, Bates
DW. Acute kidney injury, mortality, length of stay, and costs in hospitalized patients.
J Am Soc Nephrol. 2005;16(11):3365-3370.
PubMedGoogle ScholarCrossref 36.Sica
DA. Calcium-channel blocker class heterogeneity: select aspects of pharmacokinetics and pharmacodynamics.
J Clin Hypertens (Greenwich). 2005;7(4)(suppl 1):21-26.
PubMedGoogle ScholarCrossref 37.Higashi
T, Shekelle
PG, Solomon
DH,
et al. The quality of pharmacologic care for vulnerable older patients.
Ann Intern Med. 2004;140(9):714-720.
PubMedGoogle ScholarCrossref