Enteral feeding intolerance is a major issue in premature infants, resulting in prolonged hospitalization and a predisposition to serious complications due to prolonged use of parenteral nutrition. A delay in reaching full enteral feedings is also associated with a poorer mental outcome in preterm neonates at 24 months corrected age [1].

As the maturation of motor activity in premature infants lags behind that of digestive and absorptive functions, it is most frequently a disorder of gastrointestinal motility that limits the use of enteral feeding in this population. An adequate establishment of the intestinal flora after birth is strictly related to motility maturation and plays a crucial role in the development of gut barrier function and the innate and adaptive immune system [2].

Enteral supplementation of probiotics prevents severe necrotizing enterocolitis (NEC) and all-cause mortality in preterm infants [3]. Moreover, among the many strategies tried for prevention of feeding intolerance, probiotics are the most promising. However, they are yet to be used as standard of care. The objective of this trial was to determine the efficacy of probiotics on feed tolerance in very low birthweight (VLBW) neonates.

All neonates with a birth weight between 750 g to 1499 g admitted to the NICU in whom enteral feeds were started were eligible for enrolment. Neonates with gastro-intestinal anomalies, severe congenital malformation, and those not started on enteral feeds by day 14 of life were excluded. The study was a double blind randomized controlled trial (RCT) conducted in a tertiary care NICU between August 2012 to November 2013.

We hypothesized that by establishing a normal intestinal flora probiotics could reduce the incidence of feed intolerance. The study protocol was approved by the Institutional Ethics Committee and a written informed consent was taken from the parent/guardian before enrollment.

Outcomes: The primary outcome of the study was the time taken to reach full enteral feeds. Secondary outcomes were episodes of feed intolerance, incidence of NEC stage 2 or more, duration of hospital stay, days on total parenteral nutrition (TPN), weight gain and mortality during hospital stay.

Sample size: The sample size was determined based on data from a pilot study done in the same clinical unit. To achieve a significant mean difference of 3.37 days (with SD 5.05 and 6.6 in the 2 groups) in time to reach full feeds with a Type 1 error of 5% and a Power of 80% (2 sided), the sample size required was 47 (in each group) which was inflated by a further 10% based on the usual percentage of deaths and discharges against medical advice in the VLBW group in our Unit.

The subjects were randomly allocated into two groups using computer generated random numbers by an investigator not directly involved in the study. We followed a parallel group design and block randomization was done with block sizes varying from 8 to 12. Sequentially numbered opaque sealed envelopes were used for allocation concealment. The two groups were coded as A and B and the group code was kept off site in an opaque sealed envelope and opened only after the final analysis was done.

Feeding protocol: Feeding was initiated, advanced, stopped and restarted as per unit protocol derived by consensus for the purpose of the study. The protocol was attached to all the study case files to ensure compliance. Trophic feeds i.e 10 to 20 mL/kg/day at 2 hourly interval of either colostrum (if available) or donor breast milk feeds were initiated in hemodynamically stable infants. Feeds were advanced by 20 mL/kg/day (in babies 750-1249 g and those with abnormal antenatal Doppler) or by 35mL/kg/day (in babies 1250-1499g and well). Feeds were given every 2 hourly and pre-feed aspirates were measured in babies on gavage feeds. Feeds were withheld if there were signs of feed intolerance, hemodynamic instability, suspected NEC or voluminous gastric residuals. Feeds were restarted when all the above mentioned signs were resolved. Parenteral nutrition was continued till 100 mL/kg/day of feeds were reached. Full feeds were defined as 150 mL/kg/day. Oral feeds were initiated in babies more than 30-32 weeks with good suck reflex and otherwise well. 5% Dextrose was used if milk was not available. Human milk fortifiers were used as per NNF India recommendations.

The weights were checked daily on a calibrated digital weighing machine with a sensitivity of ±5 g. A best gestational age was given for each infant based on LMP and corroborated by early first trimester ultrasound when available. NEC was defined and staged as per modified Bell’s staging. The infants were discharged as per unit protocol which is baby maintaining hemodynamic stability without support, on full oral feeds-either by breast milk or paladay, showing consistent weight gain of 15 to 20 g/kg/day for 3 days and 1.4 kg or more. The management protocols, clinical practices, equipment, infrastructure, and key personnel were unchanged during the study period. The baseline illness severity documented by Score for Neonatal Acute Physiology Perinatal extension (SNAPPE-II) was used in both groups [4]. Adequate antenatal steroids were defined as an interval of at least 24 hrs after the initial dose.

The probiotic supplementation did not change the physical appearance of the milk, and the syringes were labeled only with the patient’s name and identification number with no indication of study group assignment. Attending physicians and nurses caring for the infants were blinded to the group assignments. To ensure blinding, the mixing was done after the milk required for each study infant was gathered away from the patient care area in the NICU irrespective of the assigned group. The probiotic was stored as per manufacturer’s guidelines and was prescribed to all infants enrolled in the study to ensure blinding.

Nurses followed strict asepsis during preparation and compliance was monitored regularly by one of the investigators. The infants were clinically monitored daily by the consultants for feed intolerance and sepsis. A septic screen followed by blood culture was done as per clinical suspicion.

Statistical analysis: Continuous variables were compared by using Student’s t test or the Mann-Whitney U test when appropriate; chi square analysis or Fisher’s exact test when appropriate, was used to ascertain significant differences in categorical variables between groups. All tests were 2-tailed. Significance was defined as P<0.05. Intention to treat analysis of data was performed. The final statistical analysis was performed using SPSS 20.0 software (SPSS Inc, Chicago, IL, USA)

Of the 162 VLBW babies admitted to the NICU during the study period, 104 VLBW newborns were enrolled; 52 in each group. Fig. 1 shows the flow of study subjects through the phases of the study.

Fig. 1 Trial flow.

The baseline characteristics are depicted in Table I. The groups were comparable except for slightly more number of Caesarean deliveries in the no probiotic group (Table I). There were 5 extremely preterm babies (4 in control and 1 in probiotic group) and 15 extremely low birth weight babies (10 in control and 5 in probiotic group). Enteral feeding was initiated at a similar postnatal age in the probiotic and no probiotic groups. Oral supplementation with probiotics began in parallel with enteral feeding (<24 h after initiation of feeds). A mean duration of 26.3 (17.6) days of probiotic supplementation was received in the intervention group. 22 babies in the probiotic group and 25 in the control group were exclusively fed with breast milk.

TABLE I Baseline Characteristics of Participants

The primary outcome of time to full enteral feeding was 11.2 (8.3) days in probiotic vs 12.7 (8.9) in no probiotic group and was not significantly different (Table II). The secondary outcomes were also comparable between the groups. Treatments such as duration of ventilation and antibiotic usage, and other co morbidities like intraventricular hemorrhage, patent ductus arteriosus, respiratory distress syndrome and bronchopulmonary dysplasia were not significantly different between the two groups.

TABLE II Outcomes in Preterms in Probiotic and Control Groups

No unexpected adverse events were observed during the course of the study. There was no significant difference in the incidence of nosocomial infection, including fungal sepsis, in the probiotic group.

The present RCT comparing use of probiotic versus no probiotic did not observe any significant reduction in the time to reach full enteral feeding. However, VLBW infants receiving probiotics reached full feeds 1.5 days earlier, which was similar to the result of recently published systematic review [9].

The strength of the study was the blinding used. We did not evaluate the successful colonization of infants’ gut by the organism. The sample size derived from our pilot study was probably over-optimistic on the expected effect size, as the difference observed was much smaller.

In conclusion, in the present double-blind randomized trial, oral supplementation with multicomponent probiotic formulation of L. acidophilus, L. rhamnosus, B.longum and S. boulardii did not improve the gastrointestinal tolerance to enteral feeding in very-low-birth weight infants, but was a safe intervention. We suggest larger multicentric trials of probiotic supplementation to achieve early feed tolerance before accepting/rejecting probiotics for wider clinical use.

Acknowledgements: Nursing staff of the St John’s Medical College NICU for their invaluable help in this research, parents of the infants who took part in this study, Dr.Tinku Sarah for helping with the statistical analysis.

Contributors’: SA: conceptualized and designed the study, drafted the initial manuscript, and approved the final manuscript as submitted; SR, SB: reviewed and revised the manuscript, and approved the final manuscript as submitted; SN, CBS: designed the data collection instruments, and coordinated and supervised data collection.

Funding: None; Competing interest: None stated.