Safety and Efficacy of Oral Azithromycin for Gastrointestinal Dysmotility in Pediatric Patients

Abstract

OBJECTIVE

Pediatric patients with gastrointestinal (GI) dysmotility have limited pharmacologic options due to questionable efficacy and/or safety profiles. The primary objective of this study was to assess the safety of azithromycin in pediatric patients for GI dysmotility. The secondary objective of this study was to evaluate the efficacy of azithromycin in treating GI dysmotility in pediatric patients.

METHODS

A retrospective cohort analysis was conducted from October 1, 2019, to September 30, 2023. Pediatric patients (18 years and younger) who received azithromycin specifically for GI dysmotility were included in the study. Safety endpoints assessed included description of known common and serious adverse events (ADEs) for azithromycin and corrected QT (QTc) interval exceeding 450 milliseconds (ms) using an electrocardiogram. Demographics, indications, dosing regimens, electrocardiographic data, treatment duration, clinical outcomes, and ADEs were assessed.

RESULTS

Forty-three patients met inclusion criteria. The cohort was predominantly composed of infants (0–2 years, 55.8%) and female (53.5%). Baseline electrocardiograms were conducted in 41.9% of patients, one of whom had a QTc interval exceeding 450 milliseconds. Median azithromycin dosing was 3 mg/kg/day with a maximum single dose of 300 mg. Treatment duration ranged from 7 days to over 4 years with a median of 56 days. ADEs were described in six patients and were minor in nature with most occurring within two weeks of therapy initiation. Symptom improvement was documented in 18 patients (41.9%).

CONCLUSIONS

Oral azithromycin may represent a safe and viable option in pediatric patients with GI dysmotility.

Introduction

Gastrointestinal (GI) dysmotility is defined as the abnormal movement of food, fluids, and waste through the digestive tract due to impaired function of the muscles and nerves in the

GI tract, affecting its ability to empty properly.¹ GI dysmotility is prevalent in about 25% of pediatric patients with GI disorders.² These disorders are often associated with neuromuscular dysfunction in the GI tract, viral and Helicobacter pylori infections. Additionally, GI dysmotility can be induced by medications such as opioids, dopamine agonists, and clonidine.^3–5

There are currently no clinical practice guidelines for the treatment of GI dysmotility in pediatric patients. Pharmacological treatment options for GI dysmotility in children are limited by medications with mixed efficacy data and significant, sometimes serious adverse events (ADEs). The two most common agents prescribed in children for GI dysmotility are metoclopramide and erythromycin,⁶ though both agents are advised to be used with caution in children and avoided in neonates.⁷

Presently, there is little evidence supporting the safety and efficacy of metoclopramide in pediatric patients and available data suggest that metoclopramide may not be as effective in pediatric patients, possibly due to the downregulation of dopamine-2 receptors over the first 12 weeks of therapy.^8,9 Metoclopramide use is further limited by extrapyramidal symptoms and tardive dyskinesia, which are known to manifest after 12 weeks of use.¹⁰ Erythromycin is a macrolide antibiotic commonly used to treat GI dysmotility in both adults and pediatrics. Erythromycin works through direct action on motilin receptors in the small intestine, resulting in increased GI motility.¹¹ Erythromycin is associated with corrected QT (QTc) prolongation, resulting in an increased risk for cardiac dysrhythmias as well as significant drug-drug interactions through inhibition of the CYP3A4 isoenzyme.^12,13 In the non-neonatal population, erythromycin use in pediatric GI dysmotility has limited evidence. In the neonatal population, a systematic review of randomized controlled trials revealed conflicting results regarding erythromycin's effectiveness in addressing gastric hypomotility.¹⁴

Azithromycin is also a macrolide antibiotic and has been shown to possess prokinetic characteristics similar to erythromycin. Advantages of azithromycin over erythromycin include a longer half-life allowing for once-daily administration, lower CYP3A4 inhibition, and lower incidence of ADEs, including QTc prolongation.¹⁵ To date, there is no evidence evaluating the safety and efficacy of azithromycin for pediatric GI dysmotility beyond single-dose administration.^16,17

Materials and Methods

All research was conducted at Parkview Regional Medical Center (PRMC) in Fort Wayne, Indiana. The Parkview Women’s & Children’s Hospital is the only children’s hospital in the 14-hospital Parkview system, which has 120 beds including a 7-bed PICU, 45-bed NICU, and 28-bed general pediatrics unit. The Pediatric GI outpatient clinic cares for children and adolescents with digestive, nutritional, and liver problems, while providing diagnostic tests, procedures, and medical management. Physicians additionally provide inpatient consulting services for patients admitted to the pediatric unit with any suspected GI disorder.

This study was a single-center, retrospective chart review of patients who were prescribed azithromycin from October 1, 2019 to September 30, 2023. The primary objective of this study was to assess the safety of azithromycin in pediatric patients for GI dysmotility. This was defined as any known common or serious ADE associated with azithromycin, such as nausea, vomiting, diarrhea, rash, prolonged QTc interval (defined as > 450 ms), increased transaminases (defined as greater than three times the upper limit of normal), and hepatitis. The secondary objective of this study was to evaluate the efficacy of azithromycin in treating GI dysmotility in pediatric patients, which was defined as documented clinical improvement by the provider or described subjective improvement in GI symptoms (such as constipation, vomiting, abdominal discomfort) by the patient or caregiver.

Patients were included in the study if they were 18 years of age or younger and were prescribed azithromycin for GI dysmotility by the Pediatric GI department at PRMC. Concomitant GI medication use, such as histamine₂-receptor antagonists (H2RAs) and proton pump inhibitors (PPIs) did not exclude patients, however concurrent GI motility agents (such as metoclopramide, erythromycin, or cyproheptadine) were exclusionary. GI conditions excluded from this study included inflammatory bowel disease, cyclic vomiting syndrome, or indiscriminate abdominal pain.

An initial patient list of all pediatric patients prescribed azithromycin over the described timeline was obtained by a third party through the Parkview system. Patient charts were then reviewed in the electronic health record to determine eligibility for study inclusion. Any procedure performed by the pediatric gastroenterology team to confirm diagnosis were documented. QTc interval was reported via electrocardiogram (ECG) that was assessed at baseline prior to initiating therapy with interpretation by a cardiologist. Included patient data collection included demographic information, indication, dose, route of administration, frequency, duration of therapy, QTc interval, concomitant GI medication use, clinical improvement in symptoms, and ADEs. If no changes were documented, a negative outcome was assumed. Patients who were lost to follow-up were also considered to have failed azithromycin therapy. Descriptive statistics were performed for collected data. Categorical data were analyzed using Mann-Whitney U and Kruskal-Wallis tests. All statistical analysis was performed using IBM SPSS Statistics, version 30.

Results

The initial patient list contained 84 patients, of which 43 satisfied the inclusion criteria. Forty-one patients were excluded for the following reasons: infection (20 patients), other GI condition (8 patients), not ordered azithromycin (7 patients), on concurrent prokinetic agent (3 patients), and prescribed by a provider outside of pediatric gastroenterology (2 patients). One patient was excluded due to being prescribed azithromycin prior to the study time period. All patients were started on azithromycin in the outpatient setting. Most patients were between the ages of 0 to 2 years of age (55.8%) and female (53.5%). Eighteen patients (41.9%) had a baseline ECG performed prior to initiation of azithromycin, with 1 patient having a documented prolonged QTc interval (460 ms) prior to initiation. The dysmotility diagnosis was confirmed by procedure in 11 patients (25.6%). Complete demographic information can be found in Table 1.

Table 1.

Baseline Characteristics

Demographic	n = 43 (%)
Age (years)
0-2	24 (55.8)
3-5	3 (7.0)
6-9	3 (7.0)
13-15	6 (14.0)
16-17	7 (16.3)
Gender
Male	20 (46.5)
Female	23 (53.5)
Ethnicity
Hispanic/Latino	2 (4.7)
Not Hispanic/Latino	41 (95.3)
Baseline QTc (ms)*
< 400	4 (9.3)
401-420	6 (14.0)
421-450	7 (16.3)
> 450	1 (2.3)
Not collected	25 (58.1)

IBD = inflammatory bowel disease; ms = millisecond

* Data not available for all included patients.

All 43 patients were prescribed azithromycin orally once daily. Dosing of azithromycin ranged from 2 mg/kg/day to 3.3 mg/kg/day with a median dose of 3 mg/kg/day. Dosing was capped at 250 mg for 6 patients with 2 patients capped at 300 mg per day. Duration of therapy ranged from 7 days to more than 4 years, with 37% of patients receiving azithromycin for 15 to 60 days. Six patients (14%) were on azithromycin for more than 1 year. Table 2 provides a breakdown of the dosing regimen, duration of therapy, and rationale for therapy discontinuation.

Table 2.

Therapy outcomes n = 43 (%)

Dosing (mg/kg)
< 2.7	5 (11.6)
2.7-3.3	38 (88.4)
> 3.3	0 (0)
Duration of Therapy (days)
< 15	7 (16.3)
15-30	8 (18.6)
31-60	8 (18.6)
61-90	8 (18.6)
91-365	6 (14.0)
> 365	6 (14.0)
Reason for discontinuation of therapy
No improvement	13 (30.2)
Symptom resolution	6 (14.0)
Lost to follow-up	5 (11.6)
Stopped by caregiver	4 (9.3)
Decreased efficacy	4 (9.3)
Stopped by other provider	3 (7.0)
ADE	3 (7.0)
Not described	2 (4.7)
Poor adherence	1 (2.3)
Disease interaction	1 (2.3)
Death	1 (2.3)
Statistical comparisons	p-value
ADE vs age	0.053
ADE vs dose	0.262
ADE vs duration of therapy	0.076
Improvement vs age	0.486
Improvement vs gender	0.19
Improvement vs duration of therapy	< 0.001
Improvement vs procedure-confirmed diagnosis	0.329

ADEs were described in 6 patients, 2 of whom reported multiple ADEs: 4 patients reported vomiting, 3 patients had nausea, 2 patients reported abdominal pain, and 1 patient reported persistent diarrhea. No serious ADEs were reported. Most ADEs were reported within 2 weeks of starting therapy and resulted in discontinuation in 4 patients. No statistical difference was found when comparing reported ADE occurrences to age (p=0.053), dose (p=0.262), or duration of therapy (p=0.076).

Described improvement was seen in 18 patients (41.9%). There were no statistically significant differences between described improvement and patient age (p=0.486), gender (p=0.19) or procedure-confirmed diagnosis (p=0.329), however there was a statistically significant difference between described improvement and longer duration of therapy (p<0.001). The most common reasons for discontinuing azithromycin therapy were no improvement noted (27.9%) followed by symptom resolution (14%) and lost to follow-up (11.6%). One patient was deceased while taking azithromycin, however their death was attributed to their chronic illness and not any specific medication. All but 1 patient were taking at least one concurrent GI medication with 33 patients on multiple agents. Figure 1 describes the frequency of concurrent GI medications by drug class.

Figure 1.

Frequency of concurrent GI medications.

Discussion

To the authors’ knowledge, this is the first study to describe the use of azithromycin to manage GI dysmotility in the pediatric population as a chronic regimen. Only minor, transient ADEs were reported in this study with no severe ADEs or additional complications described. Results of this study also support previous literature describing azithromycin as safe to use as a prokinetic agent in pediatric patients,^16,17 though the efficacy rate in improving GI dysmotility symptoms was below 50%.

This study aligns with findings from previous studies that have demonstrated azithromycin’s safety profile and tolerability.^18,19 Similar findings have been described in adult populations with fewer ADEs associated with azithromycin compared to erythromycin for gastroparesis^20,21 and other indications.^22,23 A concern with the alternative prokinetic agents is the serious and potentially irreversible ADE profile. Despite their selection and usage as prokinetics in pediatric patients, both metoclopramide and erythromycin are cited on the KIDs list with notable ADEs.⁷ The medications removed from the market for severe ADEs, domperidone and cisapride, are available for refractory GI disorders, but require an investigational new drug application in order to be prescribed and dispensed.^24,25

Metoclopramide may cause dyskinesia and increased risk of respiratory depression in children, and erythromycin and azithromycin should be avoided in neonatal populations due to infantile hypertrophic pyloric stenosis (IHPS). The extrapyramidal symptoms with metoclopramide can progress to tardive dyskinesia and may become irreversible, though the risk is low. In a meta-analysis of metoclopramide involving 2699 patients, 9% of pediatric patients developed extrapyramidal symptoms and few progressed to tardive dyskinesia.¹⁰ For this reason, metoclopramide carries a box warning to not be used beyond 12 weeks.²⁶ In a meta-analysis from 2016, 0.38% (63/16,431) of included infants who received erythromycin were diagnosed with IHPS, more than double the described incidence of infants who did not receive erythromycin (0.15%, 4,632/2,992,022).²⁷ The incidence of IHPS associated with azithromycin is also equally rare. In a study of 10,898 infants who received a single dose of azithromycin, only one case of IHPS was reported.²⁸ A pharmacovigilance study reported ten instances of IHPS out of 2,203 ADE reports, with most occurring in early infancy.²⁹ In this study, IHPS was not described in any case.

Fewer than halfof patients in this study described symptom improvement with azithromycin use, though this result may be influenced by the retrospective design of this study and reliance on provider documentation. Patients who were lost to follow-up were also considered non-responders. Previous studies evaluating both pediatric and adult populations suggest that azithromycin efficacy is comparable to erythromycin for gastroparesis, though they were single-dose studies and evaluated efficacy with antroduodenal manometry or gastric-emptying scintigraphy.^16,17,21,30 While these methods can show normal intestinal physiology in patients with gastroparesis, they are complicated, sometimes intensive procedures requiring direct oversight by trained professionals and are not a practical way to evaluate the effectiveness of azithromycin beyond single-dose administration.^31,32 Dosing of azithromycin in this study was consistent with previous studies.^16,17

Most patients took azithromycin for one to three months, though 14% of patients continued therapy beyond one year. There is no clear duration of therapy with prokinetic usage, though there are some concerns with tachyphylaxis development with macrolide agents used as a prokinetic.³³ Additionally, the use of an antibiotic for a non-infectious indication may lead to increased antimicrobial resistance due to increased antibiotic exposure, though it is unclear to what extent.³⁴

Limitations

There are several limitations of this study that warrant further discussion. First, there are inherent challenges associated with the retrospective nature of this study including reliance on provider documentation and loss of patient follow-up. Changes in GI symptoms from baseline to the discontinuation of azithromycin therapy were elicited from provider documentation, which can be incomplete or inconsistent from provider to provider. To mitigate this, loss to follow-up and unclear or no documentation on patient response were both considered as lack of response to azithromycin, which may have skewed the response rate. Additionally, both authors reviewed all patient charts to ensure accuracy of review and data collection. Second, this study only involved a single center, limiting external application of the results. Third, there was no comparator group, preventing assessment of azithromycin’s effect to placebo or another prokinetic agent. Finally, diagnostic testing of GI dysmotility involving antroduodenal manometry or gastric emptying scintigraphy was not performed post-azithromycin initiation. These processes can be lengthy, invasive and are generally not conducive to study evaluation beyond single dose administration.

Conclusions

Oral azithromycin was shown to be well tolerated in pediatric patients with GI dysmotility with minimal ADEs and no serious events. Described symptom improvement was observed in more than 40% of patients studied. Oral azithromycin may be a viable alternative prokinetic agent for pediatric patients with GI dysmotility.

ABBREVIATIONS

ECG

electrocardiogram;

GI

gastrointestinal;

H2RA

histamine-₂ receptor antagonist;

IHPS

infantile hypertrophic pyloric stenosis;

IV

intravenous;

PPI

proton pump inhibitor;

QTc

corrected QT