The Indian Journal of Pediatrics

, Volume 80, Issue 11, pp 891–895

To Compare the Effect of Dextromethorphan, Promethazine and Placebo on Nocturnal Cough in Children Aged 1–12 y with Upper Respiratory Infections: A Randomized Controlled Trial

Authors

    • Department of PediatricsMaulana Azad Medical College & Lok Nayak Hospital
  • Neha Joshi
    • Department of PediatricsMaulana Azad Medical College & Lok Nayak Hospital
  • Sangita Yadav
    • Department of PediatricsMaulana Azad Medical College & Lok Nayak Hospital
Original Article

DOI: 10.1007/s12098-013-1002-2

Cite this article as:
Bhattacharya, M., Joshi, N. & Yadav, S. Indian J Pediatr (2013) 80: 891. doi:10.1007/s12098-013-1002-2

Abstract

Objectives

To evaluate whether promethazine and dextromethorphan reduce nocturnal cough and improve sleep quality in children aged 1–12 y with upper respiratory tract infection (URI).

Methods

This randomised double-blinded placebo-controlled trial was conducted in Pediatric outpatient department of Lok Nayak Hospital, Delhi. After randomization into promethazine, dextromethorphan and placebo groups, parental assessment of 120 children with URI for nocturnal cough severity (child), post-tussive vomiting (child) and sleep quality (child and parent) on the night before enrolment and after 3 d of assigned medication was measured using an internally validated indigenously prepared ordinal scale.

Results

Entire cohort improved in all the study parameters after 3 d. However, no superior benefit was noted when individual parameters were compared in the promethazine and dextromethorphan groups with the placebo group. Adverse effects were more frequent in the dextromethorphan and promethazine groups although the difference was not statistically significant.

Conclusions

Nocturnal cough in URI is self-resolving and dextromethorphan and promethazine prescribed for the same are not superior to placebo.

Keywords

CoughUpper respiratory infectionChildrenCough formulation

Introduction

Upper respiratory tract infection (URI) is one of the commonest illnesses for which a pediatrician’s consultation is sought. Nocturnal acute cough and consequent sleep disturbance are the most bothersome symptoms of URI and drive parents to seek medical advice or self-administer the widely available over the counter (OTC) cough formulations to their children. Previous studies [14] and a systematic review [5] have reported uncertain efficacy of these medications in children with acute URI. In 2008, a British Thoracic Society guideline stated that, OTC anti-tussives, antihistamines and decongestants are as effective as placebo with the additional risk of adverse effects [6]. Recent evidence has also highlighted that parental reporting of frequency and severity of child’s cough is often an overestimation and hence unreliable [7, 8]. This limits the certainty that can be placed on most published literature that has not used objective cough measurements.

The pediatric outpatient department at the authors’ hospital treats on an average 500 children per day and URI constitutes 35 % of these. An antitussive (such as dextromethorphan) or an antihistamine (such as promethazine) is prescribed to at least 80 % of these children and are distributed free of cost at the hospital pharmacy.

A paucity of Indian studies and guidelines on the use of various cough formulations prompted the authors to conduct this study. The specific objectives of the study were to determine whether either dextromethorphan or promethazine were better than placebo in alleviating nocturnal coughing and sleep disturbances in children with acute URI.

Material and Methods

This randomised double blinded placebo controlled trial was conducted in the pediatric outpatient department of Lok Nayak Hospital, a tertiary care hospital in north India. From June through August, 2011, children aged 1–12 y with cough attributed to upper respiratory tract infections (characterized by the presence of rhinorrhoea and cough for <7 d duration) were enrolled. Patients with signs and symptoms of diseases like asthma, pneumonia, laryngotracheobronchitis, sinusitis and allergic rhinitis, those with history of reactive airway disease, chronic lung disease, asthma and allergic rhinitis and patients who had taken a preparation containing dextromethorphan or an antihistamine in the previous 24 h were excluded.

After stratification for age (age groups 1–5 y and 6–12 y), each child was assigned in a double-blinded manner to receive dextromethorphan (5 mg/dose q6-8 h as per instruction on package), promethazine (0.5 mg/kg/dose q8h) or a placebo (a multivitamin preparation) as per a computer-generated randomisation sequence. Allocation concealment was done by opaque sealed envelopes technique. The study medications were dispensed from the outpatient department pharmacy in identical appearing coded bottles. The randomization sequence and the key to the code on the medication bottles were kept with an investigator who was involved in neither patient enrolment nor administration of study medications or measuring outcome. Thus, investigators were also blinded to the intervention.

Parental assessment of child’s cough frequency and severity, sleep quality and post-tussive vomiting and parent’s own sleep quality were measured and recorded for the night prior to enrolment and after 3 d of medication using indigenously prepared and internally validated objective scale given in Table 1. It took, on an average, 10 min to administer. As per a previous study, normal children can have upto 5 nocturnal coughs and no prolonged nocturnal coughing episode [9]. This was the basis of the cough frequency score. To assess internal validity of the scale, 5 different raters administered the scale independently to 15 patients and free marginal kappa thus determined was found to be 0.884. This indicated strong inter-rater agreement and reproducibility of the scale.
Table 1

Symptom scores used to assess nocturnal cough and sleep quality

A. Cough frequency score:

 (1) None

 (2) Occasional (<10 coughs/night; no prolonged episode)

 (3) Often (10–20 coughs/night; ≤2 prolonged episode)

 (4) Very often (>20 coughs/night; >2 prolonged episodes)

B. Child’s sleep score:

 (1) Slept all night

 (2) Woke up occasionally (≤2/night)

 (3) Woke up frequently (>2/night)

 (4) Did not sleep at all

C. Parents’ sleep score:

 (1) Slept all night

 (2) Woke up occasionally (≤2/night)

 (3) Woke up frequently (>2/night)

 (4) Did not sleep at all

D. Post-tussive vomiting score:

 (1) No

 (2) Yes

E. Composite symptom score:

 Obtained by cumulative score of the individual symptoms

Medicines other than the study formulations, when prescribed, were noted. Any adverse event was also recorded.

Written informed consent of the caregivers of the children and institutional ethical clearance of Institutional Ethical Committee of Maulana Azad Medical College were obtained for the study.

For sample-size estimation the mean cough score of 30 children who fulfilled the inclusion criteria was determined. For detecting a 1 point reduction in the cough score with a 5 % chance of Type I error and a 10 % chance of a Type II error, a sample size of 35 was needed in each group. To allow for 10 % attrition, the total sample size was determined to be 116.

Statistical analyses was done using SPSS version 13. A p value of <0.05 was considered significant in all analyses. The comparison between pre- and post-intervention scores for the entire study population and within the individual treatment groups was done with paired t test. Comparison of treatment groups was done with one-way analysis of variance.

Results

The study enrolled 120 children and all completed the 3 d study period. Their median age was 5.5 y (range: 1–12 y) and 68 (56.7 %) were males. Each intervention group constituted 40 patients. The common presenting complaints other than cough were fever (92 %) and coryza (90 %). The median duration of illness was 2 d (range 1–7 d). The baseline demographic variables and symptom scores at enrollment in the three intervention groups are displayed in Table 2.
Table 2

Baseline demographic and clinical characteristics of the 3 intervention groups

Characteristics

Promethazine

Dextromethorphan

Placebo

(N = 40)

(N = 40)

(N = 40)

Age (mean ± sd)

5.4 ± 3.8 y

5.7 ± 3.6 y

4.8 ± 2.9 y

Males

25 (62.5 %)

16 (40 %)

27 (67.5 %)

Mean duration of illness

4.1 ± 1.7 d

3.7 ± 1.3 d

4.0 ± 1.7 d

Cough score

3.05 ± 0.22

3.0 ± 0.01

3.13 ± 0.33

Patient’s sleep score

2.53 ± 0.51

2.40 ± 0.40

2.33 ± 0.57

Parents’ sleep

2.53 ± 0.51

2.40 ± 0.40

2.33 ± 0.57

Post-tussive vomiting score

1.23 ± 0.42

1.25 ± 0.44

1.23 ± 0.42

Composite symptom score

9.33 ± 1.16

9.05 ± 1.18

9.00 ± 1.60

Other medications prescribed

 • Antibiotics

12

10

9

 • Antipyretics

38

38

40

 • Others

2

0

2

Symptom scores were obtained for the night prior to enrolment and after 3 d of assigned treatment. For the entire cohort all the scores namely, cough , patient’s sleep, parent’s sleep, post-tussive vomiting and composite symptom scores showed dramatic improvement with mean reductions of 1.70, 1.22, 1.28, 0.20 and 4.35 respectively (p < 0.0001).

However, when individual treatment groups were compared on the basis of the above scores, no significant differences were noted. The mean reduction in cough scores in the promethazine, dextromethorphan and placebo groups were 1.68, 1.68 and 1.78 respectively (p = 0.686). The sleep quality of the patients were not significantly better in the promethazine and dextromethorphan groups as compared to the placebo group with mean reduction of scores being 1.35, 1.28 and 1.23 respectively (p = 0.522). The mean reduction in the parents’ sleep score in the dextromethorphan, promethazine and placebo groups were 1.28, 1.37 and 1.25 respectively (p = 0.636). Similarly, the post-tussive vomiting scores also showed significant improvement irrespective of treatment received (p = 0.217). When the scores were combined to yield the composite system score, there was no significant difference between the three groups with mean reductions of 4.6, 4.5 and 4.0 points in the promethazine, dextromethorphan and placebo groups respectively (p = 0.103) (Table 3). Of the 120 patients, 85 (71 %) had symptom duration less than or equal to 2 d. When this group of patients was separately analysed no difference in the three groups with regards to effect of the intervention was found.
Table 3

Symptom scores after 3 d of assigned intervention in the 3 groups

Symptom score

Promethazine

Dextromethorphan

Placebo

P value

(N = 40)

(N = 40)

(N = 40)

Cough score

1.38 ± 0.49

1.33 ± 0.47

1.35 ± 0.62

0.914

Patient’s sleep score

1.18 ± 0.38

1.13 ± 0.33

1.30 ± 0.61

0.269

Parent’s sleep score

1.18 ± 0.38

1.13 ± 0.33

1.28 ± 0.60

0.323

Post-tussive vomiting score

1.00 ± 0.01

1.03 ± 0.16

1.08 ± 0.27

0.142

Composite symptom score

4.73 ± 1.13

4.60 ± 1.08

5.00 ± 1.98

0.434

The authors also attempted to assess whether this lack of effect of the study medications was reflected in different age groups by comparing the day-3 symptom scores of children aged 1–5 y and 6–12 y. No significant difference was found between the two groups.

All parents reported 100 % compliance to the prescribed medicines. This was corroborated by the volume of residual medications brought back by the parents on the follow-up visit. Adverse effects were noted in 29 children. Drowsiness (10), irritability (8), abdominal pain (7), nausea (2) and vomiting (2) were the most frequently reported side effects. The frequency of individual side effects was higher in the two intervention groups as compared to the placebo group although the differences were not statistically significant (Table 4).
Table 4

Adverse events profile in the 3 intervention groups

Adverse event

Promethazine

Dextromethorphan

Placebo

P value

(N = 40)

(N = 40)

(N = 40)

Abdominal pain

3

3

1

0.54

Nausea

1

1

0

0.60

Vomiting

1

1

0

0.60

Drowsiness

6

3

1

0.13

Irritability

3

5

0

0.08

Discussion

This study was undertaken to determine whether commonly prescribed medications for cough due to URI were more efficacious than placebo. Since nocturnal cough and sleep difficulty are the most commonly enumerated complaints among parents who bring their children suffering from URI for medical attention, these were studied as the outcome variables. The study questionnaire was designed in a manner that was easily reproducible as well as quick to administer. Moreover, unlike multiple previous studies [14] the authors used an objective scale for cough assessment to prevent parental subjective overestimation of child’s cough severity.

The most significant observation of the present study was that none of the cough formulations were superior to placebo in reducing cough frequency and severity. The medications also failed to show any benefit in sleep quality of the children as well as their parents when compared with placebo. This is in agreement with other studies done to assess the impact of over-the-counter cough formulations on these parameters in children suffering from URI [14]. The authors also failed to document any beneficial effect of these medications in the younger age group (1–5 y) or when given early in the course of the illness (<2 d). For the whole study population, however, all the symptoms showed significant improvement at the end of the 3 d study period, underlining the self-resolving nature of a URI.

Some of the earlier studies have assessed the effect of a single evening dose or 1 d of the cough formulation on nocturnal cough and sleep quality [2, 4]. These studies have failed to detect the impact of the medications given over a certain duration (e.g., 3–5 d), which is the usual clinical practice. Since the authors recorded parental assessment of cough and sleep quality after 3 consecutive days of medication, it can be concluded that longer duration of these medications also lack a beneficial effect. The authors assessed compliance both by parental recall as well as by measuring residual volume of medications brought on the follow-up visit [2, 4]. This is more accurate than compliance as measured in some studies only by parental recall through telephonic follow-up.

Although, the therapeutic efficacy of cough formulations is uncertain as per studies cited above and the current study, their potential for adverse effects is well documented. This is especially important, since most of these formulations are available over the counter. First generation anti-histamines, especially promethazine, in standard therapeutic doses have been reported to cause serious and life-threatening respiratory depression, oversedation, agitation, hallucinations, seizures, and dystonic reactions when used in children [10, 11]. Dermatologic reactions, gastrointestinal intolerance and neurolept malignant syndrome have also been reported [12]. Promethazine and other antihistamines used in cough formulations such as diphenyhydramine have potential for abuse and chronic ingestion and overdosage is associated with dependence [13], psychosis [11, 14], arrhythmias [15, 16], seizures [11, 17] and rarely death [18, 19]. Adverse reactions described with dextromethorphan include, dystonia [20], anaphylaxis [21], gastrointestinal symptoms and skin reactions [22]. In cases of abuse or accidental overdose, there may be neuropsychiatric manifestations such as euphoria, restlessness, misperception, hallucination and schizophrenic reactions [23, 24]. The present study also reported higher incidence of adverse effects in the dextromethorphan and promethazine groups as compared to the placebo group.

This study was conducted because cough due to URI can be an exceedingly distressing symptom for the affected child, his parents and treating pediatrician. It is also responsible for school absenteeism, prompting parents to seek medical care or self-medicate. Moreover, there is a paucity of studies on the subject from developing countries and none from India.

The limitation of the index study is that each patient served as her/his own control as their responses at enrolment and after 3 d of treatment were compared. Taking separate controls would have been better. Secondly, a no-treatment arm was not kept in the present study protocol. Placebo effect in studies that evaluated antitussives, as documented by Eccles [25], may have been present in the current interventions.

To conclude, there is strong parental need to alleviate the distressing symptoms of URI. This is aided by the easy availability of cough formulations. This study supports the fact that URIs improve with time and symptomatic treatment may not be required. Moreover such formulations are not free of adverse effects. Hence, pediatricians need to make a judicious decision keeping in mind severity of symptoms, parental anxiety, potential adverse effects and cost while prescribing cough formulations.

Conflict of Interest

None.

Role of Funding Source

None.

Copyright information

© Dr. K C Chaudhuri Foundation 2013