There are limited trials emphasizing effectiveness and safety on caffeine versus aminophylline in developing country. Besides, Small for Gestation Age (SGA) growth category or Intrauterine Growth Retarded (IUGR) babies are a significant problem in many developing and underdeveloped countries, and the effect of methylxanthines in them is incompletely understood [3]. Therefore the present study was designed to compare the effectiveness of methylxanthines in preterms, particularly the SGA babies.

This single centered, parallel, open label, randomized controlled trial was conducted from February 2012 to January 2015 in a tertiary level Neonatal Intensive Care Unit (NICU) at Kasturba Hospital, Manipal University. The study was approved by the Institutional Ethics committee, and written informed parental consent was obtained for each participating infant.

Computer generated block randomization was used with block size of 10. Allocation concealment was executed by using sequentially numbered, sealed, opaque envelops. Both random allocation sequencing and concealment was done by research officer who was not concerned with current trial or in management of recruited infants. The treatment assignment was carried out by attending clinician in Neonatology unit.

Infants allocated to Caffeine group received a loading dose of 20 mg/kg of caffeine citrate (10 mg/kg caffeine base) diluted in 5% dextrose given for 30 minutes and were continued on a maintenance dose of 5 mg/kg (2.5 mg caffeine base) 24 hourly (iv or oral preparation of caffeine citrate solution 20 mg/mL). If adequate response was not seen then the dose was optimized up to 7.5 mg/kg. Neonates allocated to aminophylline group received a loading dose of 5 mg/kg of aminophylline, diluted in 5 % dextrose followed with a maintenance dose of 1.5 mg/kg 8 hourly (Inj. aminophylline 250 mg/10 mL). If adequate response was not seen, the dose was titrated up to 2 mg/kg.

Sample size: Based on anticipated SD of 5 apneic episodes in 24 hours and minimal relevant difference at 3 continuing apneic episodes in 24 hours, a power of 90%, 5% as a level of significance, it was estimated that 84 neonates would be needed in each arm. Assuming 30% attrition rate, a sample of 120 in each arm was taken.

Statistical analysis: Mann- Whitney U test was used to compare episodes of apnea, MAR, isolated desaturations, time taken for apnea resolution, NICU stay and hospital stay. Repeated measures ANOVA was used to compare changes in HR at various time periods. Difference in Mean HR on individual day was analyzed by applying Independent sample t test. Adverse effects and co-morbidities were analyzed by Pearson’s chi-square test. Association of occurrence of apnea and isolated desaturation was analyzed by Pearson’s chi-square test. A subgroup analysis for the outcomes was carried out stratified by intrauterine growth status. In each growth strata the difference in apnea spells, MAR, desaturation frequencies and adverse events between the groups were compared using generalized linear model with poisson loglinear scale response. Factorial ANOVA was adopted to compare significant difference in Median days of apnea resolution, hospital stay, NICU stay, and Mean HR between groups in different growth category. P<0.05 was considered to be statistically significant.

A Data Safety Management Board (DSMB) periodically reviewed the study progress.

During the period of 36 months, out of admitted preterm neonates, 240 (23%) newborns were identified experiencing AOP and underwent randomization. The study flow is depicted in Fig. 1. Baseline characteristics were similar in both the groups (Table I).

Fig.1 Study flow.

TABLE I Baseline Demographics Characteristics

TABLE II	Comparison of Apnea Frequency and Secondary Outcomes 

Apneic episodes during 4-7 days of therapy was found to be significantly higher in Caffeine group (P=0.03). Complete resolution of apnea was achieved after similar duration of median 6 days in either group. There was no difference in median (IQR) duration of methylxanthine administration [24 (14, 35.5) d vs 22 (14, 38) d; P=0.97]. No difference was noted in median length of hospital stay and NICU stay between both the groups. Mean HR was significantly higher in aminophylline group at various time intervals (P=0.001) (Table II). Among caffeine-treated neonates, the risk of developing tachycardia was lesser than Aminophylline group (RR 0.30; 95% CI range 0.15 to 0.60, P<0.001). No significant difference found in the risk of developing feed intolerance (17% vs 22.8%, RR 0.74; 95% CI range 0.39 to 1.40, P=0.35). jitteriness (8% vs 9%, RR 0.87; 95% CI range 0.31 to 2.49, P=0.81) and glucose abnormality (3% vs 3%, RR 1.02; 95% CI range 0.21 to 4.92, P=0.97) were also similar in both the groups.

In 42 neonates who were on standard maintenance doses of caffeine, mean (SD) plasma caffeine level was 14.4 (5) mg/L. Approximately 93% (n=39) fell within a recommended wider therapeutic range whereas only 7.1% (n=3) had exceeded therapeutic window. Odds ratio (95% CI) of occurrence of apnea and isolated desaturation in the recommended therapeutic range of caffeine was 0.34 (0.02, 6.04) and 0.09 (0.007, 1.17), respectively as that of supra-therapeutic levels. In 50 neonates on aminophylline, median (IQR) level of the drug was 12.9 (6.3, 19.5) from minimum 0.68 to maximum 50.37 mg/L. Majority of them 52% (n=26) had achieved above the therapeutic range. 24% of neonates had attained therapeutic (n=12) and sub-therapeutic range (n=12). Odds (95% CI) of occurrence of apnea and isolated desaturation in the recommended therapeutic range of aminophylline was 0.30 (0.05, 1.58), and 0.93 (0.08, 10.0) as that of inadequate therapeutic levels, respectively.

One neonate was LGA from each group were excluded for subgroup analysis as the number of LGA babies were too less to analyze. In AGA category continuing apnea episodes during first week of therapy was consistently high in Caffeine group [1-3 days (P=0.01) and 4-7 days (P<0.001)]. During 4-7 days of therapy caffeine group proved to have higher MAR (P=0.001). Persistent desaturation in caffeine group was more than aminophylline during first 3 days of therapy (P=0.006) than in second week of therapy, aminophylline group reported high desaturation episodes (P=<0.001). Among SGA infants, during first 3 days of therapy apnea episodes were higher in caffeine group (P=0.01). Neonates in caffeine group had significant higher isolated desaturations during first week of treatment [1-3 days (P<0.001) and 4-7 days (P<0.001)]. In AGA babies, aminophylline group had higher mean HR on day 2 (P=0.007) and day 3 (P=0.002) (Table III).

TABLE III Comparison of Apnea in AGA and SGA Neonates Treated by Caffeine and Aminophylline

In present study, apneic episodes were found to be similar in both the caffeine- and aminophylline-treated groups during first 3 days and second week of therapy. No differences were also noted in MAR, isolated desaturations and time to resolution of apnea. Risk of developing tachycardia in caffeine treated neonates was less compared to aminophylline group. However, there were significant differences noted in efficacy or adverse events between caffeine and aminophylline in subgroup of AGA and SGA neonates.

Limitations of present study was that it was a single center, open label study. Attending clinicians were not blinded. There are only few trials reported that have compared the effectiveness and safety of caffeine and aminophylline [4-7]. Most of the trials were conducted in developed countries. A recent study [8] compared the incidence of apnea between caffeine and aminophylline treated babies and followed up recruited neonates for 10 days of life. The study reported infants with aminophylline dihydrate had approximately 10% less risk of developing apnea compared to anhydrous caffeine. The results of the present study were comparable to that of Scanlon, et al. [6] that standard doses of aminophylline resulted in less apnea frequency in first week of therapy when compared with standard doses of caffeine. Tachycardia was the major acute adverse event noted in aminophylline-treated neonates and was comparable to observation with other studies [4-8].

In conclusion both caffeine and aminophylline are equally effective in reducing apneas. Perhaps further research should be undertaken to compare the safety and effectiveness of methylxanthines in AGA and SGA separately. Since tachycardia is prominently seen in aminophylline-treated infants, dosage optimization of aminophylline can be tried for preterm population in developing countries. Aminophylline usage can be continued under close supervision at resource poor settings in India.

Contributors: SM, MN, Shashikala: acquisition of data, drafting the article or analysis and interpretation of data; LEL, JP, RBY: substantial contribution to concept and design, revising the work critically for important intellectual content and final approval of the version published; AK: revising the work critically on research methodology and Statistical guidance: SM: co-author who should be approached for the access to raw data.

Funding: Indian Council of Medical Research, New Delhi.

Competing interest: None stated.