Abstract
Keywords
Aspirin No. 86
Risk of Age-Related Macular Degeneration
A meta-analysis to evaluate the potential association of developing age-related macular degeneration with aspirin use reviewed 10 eligible studies including 180 834 patients. When the data were analyzed, there was a weak but statistically significant association between the use of aspirin and the risk of age-related macular degeneration (relative risk = 1.137; 95% confidence interval = 1.003-1.289; I2 = 68.4%). Subgroup analyses according to age-related macular degeneration classification showed that in the early and late age-related macular degeneration subgroups, the association was excluded in the early groups (relative risk = 1.19; 95% confidence interval = 0.92-1.53; I2 = 82.6%) and a nonsignificant association was observed in the late age-related macular degeneration group (relative risk = 1.22; 95% confidence interval = 0.87-1.72; I2 = 55.7%). When the type of late age-related macular degeneration was considered, a significant association was observed between aspirin use and the risk of neovascularization (relative risk = 1.95; 95% confidence interval = 1.40-2.72; I2 = 27%) but none in the geographic atrophy group (relative risk = 0.84; 95% confidence interval = 0.62-1.15; I2 = 0%).
The authors concluded that this meta-analysis demonstrated a weak but significant association between the use of aspirin and the risk of age-related macular degeneration. They noted that the results of this study differs from previous work in this area. Proposed mechanisms of action for aspirin in the pathophysiology of age-related macular degeneration include inhibition of cyclooxygenase 1 and prostaglandin-endoperoxide synthetase-2 leading to hypoxia, disturbance of lipid oxidation, and/or anticoagulant effects, which may increase the degree of choroidal neovascularization.
Aspirin [ASA, Acetylsalicylic Acid]
Li L et al (C Chen, Department of Ophthalmology, Renmin Hospital of Wuhan University, 238 Jiefang Road, Wuhan 430060, Hubei Province, China; e-mail:
Turmeric, Fluindione No. 87
Drug Interaction: Increased International Normalization Ratios
A 56-year-old female patient experienced an increased international normalized ratio (6.5) without clinical signs of bleeding approximately 5 days after starting turmeric (one teaspoon; 2.5 g) each evening. Prior to this event, the patient had been treated with fluindione for mitral valvulopathy for several years with international normalized ratios in therapeutic range (international normalized ratio 2-3). No other medications were mentioned in the report, but the patient started the herbal infusion for the treatment of ankylosing spondylarthritis after several prescription treatments had failed. Within 15 days after withdrawal of turmeric, the international normalized ratio returned to target range.
The authors concluded that this patient experienced increased international normalized ratios related to the addition of turmeric therapy to fluindione based on the temporal relationship between the drug administration and the appearance and resolution of effects. The role of turmeric was evaluated as “probable,” according to the Horn drug interaction probability scale, and as “likely,” according to the Begaud method of causality assessment. According to the authors’ discussion of the case, the mechanism of action of this interaction is not clearly established.
Turmeric [Curcuma longa]
Fluindione [“Previscan”]
Daveluy A et al (A Daveluy, Centre régional de pharmacovigilance Hôpital Pellegrin, CHU, 33076 Bordeaux Cedex, France; e-mail:
Hydroxyzine No. 88
QT Prolongation (First Report*)
A 72-year-old male patient was hospitalized for syncopal episodes 4 days after starting hydroxyzine therapy for the management of pruritus. On the first day, he was treated with 100 mg followed by 175 mg for the next 2 days. Concurrent medications on admission included irbesartan, clopidogrel, simvastatin, and insulin. Prior to hydroxyzine therapy, the patient had QT intervals of 450 ms, which increased to 590 and 582 ms on days 6 and 10 after initiation of hydroxyzine, respectively. Additional issues included sinus bradycardia and hypokalemia (3.3 mmol/L). Hydroxyzine blood levels were 96 and 506 ng/mL, respectively. Treatment included the discontinuation of hydroxyzine, which resulted in eventual normalization of QT prolongation and potassium levels. No further recurrences were noted.
The authors concluded that this patient experienced a case of hypokalemia and QT prolongation associated with the administration of hydroxyzine based on the temporal relationship between the administration and discontinuation of the drug. They also noted that this is the first case report of QT prolongation associated with hydroxyzine therapy. They suggested that the patient’s underlying liver impairment may have contributed to the accumulation of hydroxyzine, resulting in high blood concentrations.
Hydroxyzine [“Vistaril”]
Vigne J et al (A Coquerel, Department of Pharmacology and Pharmacovigilance, CHU de Caen, Avenue de la cote de nacre, 14000 Caen, France; e-mail:
Yellow Fever Vaccine No. 89
Vaccine-Associated Viscerotropic Disease
A female patient in her 60s was hospitalized with malaise, dyspnea, vomiting, and diarrhea approximately 6 days after receiving a single dose of yellow fever vaccine and typhoid vaccine. On physical examination in the emergency department, symptoms included tachycardia and weakness. Abnormal laboratory values included a decreased white blood cell count (4400/µL), decreased platelet count (84 000/µL), decreased serum potassium (2.8 mmol/L), and decreased serum calcium (8.0 mg/dL). Within 10 hours of hospitalization, the patient developed acute respiratory failure requiring mechanical ventilation, cardiogenic shock, and acute renal failure. The patient died 3 days after admission. A scan of the chest prior to death revealed a substantial mediastinal mass. At autopsy, the thymus was diffusely enlarged. Analysis of tissue and serum samples revealed yellow fever vaccine–associated viscerotropic disease. A serum sample obtained 1 day before death demonstrated evidence of yellow fever immunoglobulin M, with a yellow fever virus–specific neutralizing antibody titer of 640.
The authors concluded that this patient experienced yellow fever vaccine–associated viscerotropic disease, a serious, rare, and often fatal adverse reaction caused by the uncontrolled replication of vaccine virus, eventually resulting in multisystem organ dysfunction. This conclusion was based on the findings of yellow fever virus–specific antigen in multiple organs. They suggested that the patient’s age and her occult thymic disease may have been predisposing factors for the development for the reaction.
Yellow Fever Vaccine [Yellow Fever Vaccine]
DeSilvia M et al (M DeSilvia, Epidemic Intelligence Service, CDC; Neurocritical Care/Critical Care Medicine, Asante Health System, Medford, OR; e-mail:
Baclofen No. 90
Use of Continuous Venovenous Hemofiltration in Overdose (First Report*)
A 57-year-old male patient was hospitalized in an unconscious state after ingesting approximately 340 mg of baclofen 2 hours prior to admission. Concurrent medication on admission included diazepam, metformin, diclofenac, amlodipine, losartan, pantoprazole, alprazolam, mirtazapine, calcium carbonate/vitamin D3, and rosuvastatin. On admission, he was responsive to painful stimuli and required intubation. Two doses of intravenous naloxone (0.4 mg) produced no response, suggesting that the patient had not also ingested opiates. N-acetylcysteine was also administered empirically for potential acetaminophen co-ingestion but acetaminophen levels were negative. Continuous venovenous hemofiltration was initiated and was continued till the patient became fully conscious on the third day after admission. On the fourth day, the patient was discharged from the intensive care unit to general care and eventually recovered.
The authors concluded that this patient experienced severe central nervous effects as a result of an overdose. They also noted that this was the first report of using continuous venovenous hemofiltration to assist in the clearance of the drug. Baclofen concentrations varied from 1.81 to 0.05 mg/L in the serum and between 0.74 and 0.05 mg/L in hemofiltrate. Total clearance was estimated at 6.6 and 2.4 L/h. The elimination half-life during hemofiltration was estimated at 4.8 hours. The authors concluded that this case suggested that continuous venovenous hemofiltration may be an alternative to hemodialysis in the treatment of baclofen overdose.
Baclofen [“Lioresal”]
Meulendijks D et al (JH Beijnen, Department of Pharmacy & Pharmacology, The Netherlands Cancer Institute, PO Box 90440, 1006 BK Amsterdam, Netherlands; e-mail:
Ipilimumab No. 91
Dermatomyositis (First Report*)
A 50-year-old woman developed an erythematous eruption on the face, upper chest, posterior neck, upper back, and lateral thighs within 2 weeks of initiating ipilimumab (3 mg/kg every 4 weeks) for the treatment of malignant melanoma with metastases of the liver, lungs, and peritoneum. No concurrent medications were reported. Ipilimumab therapy was initially continued, and prednisone (60 mg/day tapered over 4 weeks) was initiated due to worsening cutaneous symptoms and the development of muscle weakness. Restaging computed tomography scans revealed impressive tumor response and further ipilimumab therapy was withheld following the third dose. The cutaneous reaction improved over the subsequent several months. Approximately 14 months later, reinduction therapy with ipilimumab (3 mg/kg every 4 weeks) was initiated due to metastatic melanoma involving the T3 spine and presumed metastatic lesions to the right acetabular fossa, distal right femur, and multiple ribs. Following the first dose, erythematous eruption recurred. Following the third dose, acute proximal muscle weakness of upper and lower extremities and worsening of the cutaneous eruption occurred. Physical examination revealed a photodistributed erythematous to violaceous eruption and was consistent with cutaneous dermatomyositis. Musculoskeletal examination revealed grade 4/5 strength on resisted abduction of shoulders bilaterally and limited flexion of both hips. Serum laboratory analyses revealed elevated creatine kinase (1088 units/L) that increased further the following day (1854 units/L), elevated aldolase activity (23.0 units/L), elevated C-reactive protein (28.6 mg/L), an antinuclear antibody titer of 1:640 with a speckled pattern, and negative Jo-1 antibody. Magnetic resonance imaging findings demonstrated mild short T1 inversion recovery hyperintensity of the bilateral vastus lateralis and rectus femoris muscles. Further doses of ipilimumab were withheld; however, muscle weakness progressed rapidly over the subsequent 3 days. Additional treatment included oral prednisone (80 mg/day) and was changed to intravenous methylprednisolone (80 mg twice daily) due to worsening of symptoms. A biopsy of the vastus lateralis muscle 6 days after initiation of corticosteroids demonstrated atrophy but no evidence of inflammatory myositis. Treatment was continued with oral prednisone (1 mg/kg/day) and was tapered over the following 8 weeks. Serum creatine and aldolase activity normalized within 14 days of initiation of corticosteroid therapy; however, cutaneous symptoms improved only minimally. Ipilimumab was not restarted. Further treatment for melanoma included a combination of dabrafenib and trametinib, which led to rapid and significant tumor shrinkage but only a partial response to this regimen.
The authors concluded that this was the first report of dermatomyositis related to ipilimumab therapy in a patient with metastatic melanoma. The proposed mechanism was an immune-related adverse event associated with cytotoxic T-lymphocyte antigen 4 blockade by ipilimumab.
Ipilimumab [“Yervoy”]
Sheik Ali S et al (RA Vleugels, Department of Dermatology, Brigham and Women’s Hospital, Harvard Medical School, 221 Longwood Ave, Boston, MA 02115; e-mail:
Dietary Supplement No. 92
FDA Public Notification: Hidden Prescription Ingredient
On March 18, 2015, the FDA alerted consumers regarding a hidden prescription ingredient (phenolphthalein) found in a dietary supplement advertised for use in promoting weight loss. Phenolphthalein is not an active ingredient in any FDA-approved drug in the United States, and its use has been associated with cancer. Patients are advised not to purchase the product and, if used, to report any adverse events that occur.
Dietary Supplement [“Ultimate Boost”]
FDA Public Notification: Ultimate Boost contains hidden prescription ingredient. http://www.fda.gov/Drugs/ResourcesForYou/Consumers/BuyingUsingMedicineSafely/MedicationHealthFraud/ucm438679.htm?source=govdelivery&utm_medium=email&utm_source=govdelivery (Mar 18) 2015
Minoxidil No. 93
Anorexia in Infant
An 2-month-old girl, born at 35 weeks gestation, developed anorexia within 3 days of initiating minoxidil (0.15 mg/kg/day titrated to 0.6 mg/kg/day) for the treatment of resistant idiopathic hypertension. Serum laboratory analysis revealed elevated plasma renin activity (270 ng/mL/h), aldosterone level (238 ng/dL), and renal β2-microglobulin clearance (10.2 mg/L). Thyroid studies and urine catecholamine levels revealed no abnormalities. An echocardiogram revealed left ventricular hypertrophy, and a renal sonogram was normal. Initial therapy for blood pressure control included administration of captopril, amlodipine, and clonidine (doses not reported). Therapy led to elevated serum creatinine and oliguria, prompting substitution of propranolol for captopril. Blood pressure control was not established despite escalation of doses, and systolic blood pressure remained at 120 to 130 mm Hg. At 1 month of age, congestive heart failure occurred with a dilated and hypertrophied left ventricle and systolic and diastolic dysfunction. Peak B-type natriuretic peptide level reached 2097 pg/mL. Sustained blood pressure control was achieved with aggressive parenteral labetalol and nitroprusside therapy (doses not reported). Antihypertensive therapy was transitioned to oral amlodipine, clonidine, furosemide, and enalapril. Parenteral labetalol was continued, but enalapril was discontinued due to recurrent deterioration of renal function, increased serum creatinine concentration, and hyperkalemia. At 2 months of age, oral minoxidil was initiated. Systolic blood pressure ranged from 95 to 110 mm Hg. Oral intake decreased from 210 to 63 mL/kg/day and no weight gain occurred for 1 week. Treatment included placement of a feeding tube to supplement oral intake and weaning of minoxidil treatment. Anorexia resolved within 2 days, and no evidence of hypertrichosis was noted. Rebound hypertension occurred (systolic blood pressure160 mm Hg) and was controlled with doxazosin (dose not reported). Antihypertensive therapy was continued with propranolol, amlodipine, and doxazosin. Plasma renin activity, aldosterone, and B-type natriuretic peptide levels all decreased and heart failure had resolved by 5 months of age.
The authors concluded that the development of anorexia in this case was a definite adverse drug reaction related to minoxidil via assessment with the Naranjo adverse drug reaction probability scale. The authors stated that the mechanism was unclear but cited models demonstrating a rapid increase in gene expression of minoxidil sulfotransferase, a cytosolic phenol sulfotransferase that converts minoxidil to the active compound minoxidil sulfate in response to fasting.
Minoxidil [“Loniten”]
Vesoulis ZA et al (ZA Vesoulis, Washington University School of Medicine, 1 Children’s Place, St Louis, MO 63110; e-mail:
Varenicline No. 94
FDA Safety Communication: Increased Alcohol Effects
On March 9, 2015, the FDA alerted patients and health care professionals that new safety labeling for varenicline will reflect recent data regarding a potential interaction with alcohol. Based on reports received by the FDA, interactions between alcohol and varenicline have resulted in increased intoxicating effects of alcohol, sometimes associated with aggressive behavior and/or amnesia. In addition, treatment with the agent has been associated with rare accounts of seizures.
In addition, the FDA encouraged health care professionals to consider these new data and weigh the potential benefit–risk ratio of this new information when prescribing varenicline in patients with a history of seizures or other factors that can lower the seizure threshold. Patients are also advised to stop the medication if they develop agitation, hostility, aggressive behavior, depressed mood, or changes in behavior or thinking that are not typical for them, or if they develop suicidal ideation or behavior. Patients should be counseled regarding the use of alcohol with varenicline.
Varenicline [“Chantix”]
FDA Drug Safety Communication: FDA updates label for stop smoking drug Chantix (varenicline) to include potential alcohol interaction, rare risk of seizures, and studies of side effects on mood, behavior, or thinking. http://www.fda.gov/Drugs/DrugSafety/ucm436494.htm (Mar 9) 2015
Mirtazapine, Venlafaxine No. 95
Increased Platelet Count
A randomized study was conducted to investigate changes in platelet count and mean platelet volume in 62 institutionalized adults treated with mirtazapine (mean = 46 ± 9 mg/day) or venlafaxine (mean = 202 ± 57 mg/day) for major depression. The majority of participants (n = 43) were women, and mean age was 51.7 ± 15.8 years. No participants suffered from concurrent acute or chronic inflammatory disorders. Concurrent medications that were allowed included lorazepam and zolpidem and medications for physical disorders as prescribed by participants’ general practitioners. No significant difference between participants treated with mirtazapine versus venlafaxine were noted at baseline with regard to gender, body mass index, Hamilton Rating Scale for Depression scores (22.1 ± 3.9 vs 23.4 ± 4.6, respectively), age, or platelet count and mean platelet volume. The Hamilton Rating Scale for Depression scores declined after receipt of 28 days of treatment with mirtazapine and venlafaxine (8.3 ± 4.7, P < .0001, and 10.0 ± 7.4, P < .0001, respectively). A positive correlation between platelet count and mean platelet volume was demonstrated (r = .319, P = .0479). A significant increase in platelet count was demonstrated (P = .02), and no changes of mean platelet volume values were demonstrated. There was no significant difference between participants treated with mirtazapine versus venlafaxine; however, the increased platelet count was stronger in the mirtazapine group as comparted to the venlafaxine subgroup. Mean platelet count (per nL) for venlafaxine at baseline and day 28 was 279.15 ± 71.41 and 289.82 ± 84.42, respectively (P not significant). Mean platelet count (per nL) for mirtazapine at baseline and day 28 was 256.13 ± 70.89 and 271.32 ± 68.08, respectively (P = .46). The mean platelet volume (fL) for venlafaxine at baseline and day 28 was 8.16 ± 0.82 and 8.23 ± 0.97, respectively (P not significant). The mean platelet volume (fL) for mirtazapine at baseline and day 28 was 8.16 ± 0.62 and 8.11 ± 0.49, respectively (P not significant).
The authors concluded that results of this study demonstrated that mirtazapine and venlafaxine increased platelet count but had no effect on mean platelet volume in patients suffering from major depression. No mechanism was proposed.
Mirtazapine [“Rameron”]
Venlafaxine [“Effexor”, “Effexor-XR”]
Gronau W et al (M Deuschle, Central Institute of Mental Health, University of Heidelberg, Medical Faculty Mannheim, PO Box 122120, 68072 Mannheim, Germany; e-mail:
Asthma Medications (Nonprescription) No. 96
FDA Safety Communication
On March 19, 2015, the FDA alerted patients and health care professionals regarding potential safety issues with the use nonprescription asthma products labeled as homeopathic preparations The FDA noted that these products do not undergo the typical FDA evaluation process that drugs complete for safety and effectiveness. The communication also noted that many asthma products labeled as homeopathic are widely available for sale via retail stores and the Internet. The safety alert also noted that though many of these products are promoted as “natural,” “safe and effective,” and include indications for the relief or treatment of asthma, they may carry unknown risk associated with use. The FDA encourages patients to consult with their health care provider if you or family member have asthma. In addition, if an adverse event has occurred, it should be reported to the FDA MedWatch system.
Asthma Medications, Nonprescription [Homeopathic Asthma Medications]
FDA Safety Alert: Over-the-counter asthma products labeled as homeopathic: FDA statement—Consumer warning about potential health risks. http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm439014.htm (Mar 19) 2015
Pradigastat, Acetaminophen No. 97
Drug Interaction—Pharmacokinetic Study
In an open-label, 7-treatment period, 5-sequence, and crossover study, 25 healthy adult volunteers (mean age = 35.5 years) were randomized to receive various regimens of pradigastat and acetaminophen. During period 1, subjects received a single oral dose of acetaminophen (1000 mg) with breakfast. During period 2, subjects received a loading dose of pradigastat (100 mg) for 3 days followed by 40 mg/daily for 7 days. During periods 3 to 7 (days 12-16), subjects received pradigastat 1 hour prior to breakfast and also received acetaminophen single 1000-mg doses at 5 different time points relative to breakfast (−2, −1, 0, +1, +3 hours). When pradigastat was administered alone, the agent reached steady state levels (900-1000 ng/mL) by the end of period 2. During all the combined treatment periods, the area-under-the curve and peak plasma concentration (Cmax) of acetaminophen were within 80% to 125% of data alone, suggesting that acetaminophen was not affected by the administration of pradigastat nor by the timing of breakfast. However, when acetaminophen was given 1 hour after a meal, the time to maximum concentrations was delayed by 1.25 hours without affecting the peak concentration or the area-under-the curve.
The authors concluded that this study demonstrated that the presence of steady-state pradigastat did not affect the pharmacokinetics of acetaminophen when administered at various times in relation to a meal.
Pradigastat [“LC908,” “Pradigastat”]
Acetaminophen [“Tylenol”]
Ayalasomayajula S et al (J Chen, Novartis Institutes for Biomedical Research, East Hanover, NJ; e-mail:
Nonsteroidal Anti-Inflammatory Drugs No. 98
Gastrointestinal Adverse Events in At-Risk Patients With Rheumatic Diseases
n an observational study, 4144 adult patients (mean age = 66 years) with at least one gastrointestinal risk factor who also had recently started nonsteroidal anti-inflammatory drug therapy for some type of rheumatic disease were followed for the development of gastrointestinal events. The majority of patients were receiving a nonsteroidal anti-inflammatory drug for osteoarthritis (85%) and rheumatoid arthritis (11%). The median length of time on the nonsteroidal anti-inflammatory drug at entry into the study was 27 days (range = 1-85 days). The most common nonsteroidal anti-inflammatory drugs taken during the study were diclofenac (29%), ibuprofen (19%), naproxen (10%), and meloxicam (8%). The percentage of patients that took nonsteroidal anti-inflammatory drugs continuously during the study was 52.4%. All patients had at least one gastrointestinal risk factor, 25% had 2 risk factors, and 10% were considered high risk (having at least 3 risk factors or history of gastrointestinal bleeding). The incidence rate of overall gastrointestinal events (complicated and uncomplicated) was 19.0/100 person-years. Symptomatic uncomplicated rate was 18.5/100 person-years, and 0.7/100 person-years was the rate for complicated gastrointestinal events. Upper gastrointestinal events were more common than lower gastrointestinal events (12% vs 1%). Reported rates for cardiovascular, anemia, osteoporosis, and pneumonia complications were much lower. Patients taking concurrent proton pump inhibitors had a cardiovascular rate of 0.5/100 person-years compared to 0.3/100 person-years for those that did not take proton pump inhibitors.
The authors concluded that this trial (EVIDENCE trial) was the largest prospective trial monitoring gastrointestinal events in European patients taking nonsteroidal anti-inflammatory drugs for rheumatic diseases who are at risk for gastrointestinal complications. The study showed ongoing risk in patients taking nonsteroidal anti-inflammatory drugs and who at risk for gastrointestinal events.
Nonsteroidal Anti-Inflammatory Drugs [“Voltaren,” “Motrin,” “Naprosyn”]
Lanas A et al (A Lanas, Service of Digestive Diseases, University Hospital Lozano Blesa, C/Domingo Miral s/n, Zaragoza 50009, Spain; e-mail:
Oxaliplatin No. 99
Organizing Pneumonia
A retrospective review of electronic medical records was performed of patients with advanced digestive tumors treated with oxaliplatin-based therapy to identify the incidence of organizing pneumonia. Over the 2.5-year study period, 438 patients were treated with oxaliplatin-based chemotherapy. The incidence of oxaliplatin pulmonary toxicity (eg, interstitial pneumonitis) was 7 in 1000 patients (n = 3) after receiving 7 to 12 courses of therapy. Approximately one third of patients had pulmonary symptoms at diagnosis. The outcome was favorable after the discontinuation of oxaliplatin. Successful treatment in these patients included a short course of corticosteroids.
The authors concluded that these patients experienced interstitial lung disease related to oxaliplatin therapy during treatment for digestive cancer. According to the Naranjo Causality Scale, the cases were rate as possible and probable. Based on this retrospective study, the authors noted that though the incidence rate of oxaliplatin-induced organizing pneumonia is low, clinicians should be aware of this rare adverse event.
Oxaliplatin [“Eloxatin”]
Bellanger C et al (C Bellanger, Gastroenterology and Digestive Oncology Unit, Universite Parise Descartes, Sorbonne Paris Cite, Faculte de medicine, AP-HP, Paris, France) Incidence of organizing pneumonia induced by oxaliplatin chemotherapy for digestive cancer. Ann Pharmacother 49:494–495 (Apr) 2015
Azithromycin, Erythromycin No. 100
Pyloric Stenosis in Early Infancy
In a retrospective cohort study using a military health system database, 1 074 236 children born between June 1, 2001, and April 1, 2012, were studied. Children younger than 90 days who received erythromycin, azithromycin, or cephalexin (control) were followed for the development of infantile hypertrophic pyloric stenosis. Of the children identified, 1902 were prescribed oral erythromycin and 4875 were prescribed oral azithromycin within the first 90 days of life. Over the study period, the incidence of azithromycin prescriptions increased at an annual rate of 15.8% while erythromycin decreased at a rate of 5% per year. Of all the children identified, 2466 developed infantile hypertrophic pyloric stenosis, making an incidence of 2.29/1000. Of those prescribed erythromycin, 17 developed infantile hypertrophic pyloric stenosis (8.94/1000). Of those prescribed azithromycin, 8 developed infantile hypertrophic pyloric stenosis (1.64/1000). The relative risk in the first 90 days of life for developing infantile hypertrophic pyloric stenosis was 3.94 (95% confidence interval = 2.44-6.36; P < .001) and 0.71 (95% confidence interval = 0.36-1.43; P = .34) with erythromycin and azithromycin, respectively. The median age for children at the time of infantile hypertrophic pyloric stenosis onset was 34 days. For those prescribed erythromycin and azithromycin developing infantile hypertrophic pyloric stenosis it was 49 days and 68 days, respectively (P < .001). The median number of days after drug exposure to pyloromyotomy was 13 and 29.5 days, respectively (P = .08). There was no significant association between the administration of cephalexin and infantile hypertrophic pyloric stenosis. Of those that received azithromycin, 3 developed infantile hypertrophic pyloric stenosis within the first 14 days. The number needed to harm in the 0- to 14-day drug exposure group was 56 in the azithromycin group and 35 in the erythromycin group. The adjusted odds ratio for the first 14 days was 8.26 (95% confidence interval = 2.62-26.0) in the azithromycin group and 13.3 (95% confidence interval = 6.80-25.9) in the erythromycin group. The adjusted odds ratio for 15 to 42 days of life was 2.98 (95% confidence interval = 1.24-7.20) and 4.10 (95% confidence interval = 1.69-9.91) for the azithromycin and erythromycin groups, respectively.
The authors concluded that the ingestion of oral erythromycin and azithromycin in infants is associated with an increased risk of developing infantile hypertrophic pyloric stenosis, with the strongest association if the exposure occurs in the first 2 weeks of life. The association persists to the age of 6 weeks, although to a lesser degree. The authors also suggested that if infants are prescribed these agents, they should be monitored for signs and symptoms of pyloric stenosis for 6 weeks posttreatment.
Erythromycin [“E-mycin”]
Azithromycin [“Zithromax”]
Eberly MD et al (MD Eberly, Department of Pediatrics, F. Edward Hebert School of Medicine, Uniformed Services University of Health Sciences, 4301 Jones Bridge Rd, Bethesda, MD 20814; e-mail: