MANUSCRIPT CITATION

Ochoa TJ, Zegarra J, Cam L, et al. Randomized controlled trial of lactoferrin for prevention of sepsis in peruvian neonates less than 2500 g. Pediatr Infect Dis J. 2015;34(6):571-6. PMID:25973934

REVIEWED BY

Dr Janet Berrington
Consultant Neonatal Paediatrician
Newcastle upon Tyne Hospitals NHS Foundation Trust

Dr Otilia Osmulikevici
Paediatric Trainee
Northern Deanery

TYPE OF INVESTIGATION

Prevention. Randomized placebo-controlled trial.

QUESTION

Population: Peruvian infants <2500g birth weight
Intervention: administration of bovine lactoferrin
Comparitor: placebo (maltodextrin)
Outcome: incidence of first episode of late-onset sepsis
Time: within 4 weeks from enrollment

METHODS

• Design: Randomized placebo controlled double blind study.
• Allocation: By assigning pre-allocated numbers : with 1:1 allocation, stratified by weight (500-1000g, 1501-2000g and 2001-2500g)
• Blinding: Physicians and study personnel were blinded to the treatment assignment throughout. In addition data manager and statistician remained blinded to the group assignment until the end of the data analysis.
• Follow up period: For primary outcome – 4 weeks from enrolment. Specific time frames for secondary outcomes are not given, but include those measurable only by hospital discharge. Clinic visits took place at 1 and 3 months of postnatal age.
• Setting: Neonatal Intermediate and Intensive Care Units of 3 Peruvian hospitals: Hospital Nacional Cayetano Heredia (Cayetano), Hospital Nacional Guillermo Almenara Irigoyen (Almenara) and Hospital Nacional Alberto Sabogal Sologuren (Sabogal)
• Patients: Inclusion criteria were: neonates with a birth weight between 500g and 2500g born in or referred in the first 72 hours of life to one of the participating hospitals. Exclusion criteria were neonates with underlying gastrointestinal problems that prevent oral intake, predisposing conditions that affect growth and development (chromosomal abnormalities, structural brain anomalies), family history of cow milk allergy, neonates that lived far from Lima and neonates whose parents declined to participate.
• Intervention: Oral bovine lactoferrin (200mg/kg/day) or placebo (maltodextrin – dose unstated), for 4 weeks starting as soon as infant was receiving any amount of oral feeds.
• Outcomes:
  • Primary outcome: first episode of late onset sepsis (culture proven sepsis or culture negative probable or possible infection) within 28 days from enrollment. Clear definitions of outcomes are given.
  • Secondary outcomes: frequency of culture positive sepsis, pathogen specific late onset sepsis, necrotising enterocolitis (NEC), duration of hospitalization, mortality rate, infection related mortality, frequency of adverse events and treatment intolerance
• Analysis and Sample size: Sample size was based on an expected background rate of first episode of sepsis of 30%, and a reduction in this of 66%. This estimated risk reduction was based on the Manzoni lactoferrin trial (9). A total of 190 infants hus gave a significance level of 0.05 and a power of 80% (not stated whether two or one sided). Analysis was on an intention to treat basis. Unadjusted relative risks for sepsis are presented, and birth weight adjusted risks using generalized linear models. A survival analysis of time to first episode of infection is presented, using a Cox proportional hazards regression model.
• Patient follow up: Of 375 infants that were assessed for eligibility, 185 were excluded: 45 parents declined, 33 did not meet eligibility criteria, and a further 107 for unknown other reasons. Therefore, 190 neonates were included. None were lost to follow up. However, 7 patients in the lactoferrin arm and 3 patients in the placebo group discontinued intervention before the end of the four week period due to a combination of parental requests and transfer to other hospitals.

MAIN RESULTS

There were 33 episodes of first late onset sepsis during the 4 weeks after enrolment. 12 babies were in the lactoferrin group and 21 babies were in the placebo group. Of these 8 were culture positive, 4 in the lactoferrin group and 4 in the placebo group. 6 episodes of sepsis occurred before the randomization, but it is not stated how many were in which intervention arm. The crude risk ratio between groups showed a RR of 0.57 (95%CI 0.3-1.09).

For infants weighing <1500g there were 8 cases of late onset sepsis in the 40 VLBW infants receiving lactoferrin and 15 in those receiving placebo, a 46% reduction in sepsis. After adjustment for birth weight category the RR for late onset sepsis was 0.577 (95% CI 0.3-1.07). In the Cox model, birth weight was the only significant risk factor. No effect of lactoferrin was seen on the hazard ratio of late onset sepsis after adjustment for birth weight (0.507 (95% CI 0.249-1.034)).

The authors noted that many infections occurred early, postulating that lactoferrin receipt would take longer to establish an effect and so a further analysis, using lactoferrin exposure time was done, and this did reach significance (P=0.03) with the hazard ratio 95% CI limits being between 0.0003 and 0.665, but no point estimate is given.

There were no significant differences in secondary outcomes and no established risks related to lactoferrrin are reported.

CONCLUSION

The authors conclude that they did not demonstrate a statistically significant effect of lactoferrin on the rate of first late onset sepsis episodes in infants with a birth weight less than 2500g. The authors also suggest that the hazard ratio implies an effect may be seen with a larger sample size as does the exposure time analysis.

COMMENTARY

Late-onset sepsis is an important problem for neonatologists in all care settings. In the poor resourced countries this is an enormous problem.¹ In the developed world as survival of the most immature infants has improved, the incidence of late onset sepsis has increased,² and remains an important cause of mortality and morbidity in survivors.³ In recent years, the neonatal community has made significant efforts to reduce the burden of LOS. These efforts have included breast feeding initiatives, long line associated care bundles,⁴ use of probiotics,⁵ and feed strategies for at risk infants, including comparing delayed initiation of feeds with early initiation of feeds in neonates with abnormal dopplers as done in the completed ADEPT study.⁶ Furthermore, large feeding trials are in progress to assess LOS as a key outcome measure (the SIFT Trial: comparing two rates of feed increases).⁷

Lactoferrin is a highly conserved protein from an evolutionary perspective, with a wide range of roles related to protection from infection and promotion of nutritional status. Lactoferrin is available as a commercial extract from bovine milk and is worthy of exploration as a modifier of late onset sepsis in neonates, as it has many different potential mechanisms of action against a disease with multiple aetiological factors. Therefore as a prophylactic intervention, lactoferrin may well offer neonates protection through multiple and potentially individually varying mechanisms.⁸ Paediatric studies suggest a role for lactoferrin in prevention and management of diarrhoeal illnesses and inflammatory bowel disease. In preterm infants, establishment of the gut microbiota (on which lactoferrin is likely to impart an effect) is of importance for key morbidities like late onset sepsis (LOS) and necrotising enterocolitis (NEC),⁹ both significant causes of mortality.

To date, 472 infants <1500g have participated in a single randomized controlled trial of lactoferrin, reducing LOS from 17.3% to 5.9% .⁹ This study was performed in Italian units, and the authors of this current paper thus wished to explore whether in a more resource poor setting lactoferrin may reduce LOS similarly. Because of the different nature of late onset infection in more resource poor settings (more infants of >1500g develop and die from late onset sepsis than in developed countries), babies >1500g were enrolled in the current study making this population different from that previously studied. Enrolled infants had a median gestational age of 32 weeks, but infants (<2500g) of up to 38 weeks were also enrolled. It appears that all infants received treatment for the allocated 4 weeks, implying these nearer term infants remained hospitalised for the full month, which would be unusual in other settings.

This difference in health care provision also influences the choice of primary outcome, and the power calculation for this. The Italian study used first culture proven sepsis to discharge as the primary outcome measure. The choice of clinical LOS rather than culture proven LOS for this population is pragmatic and justified, but brings into question the use of Manzoni’s data as pilot data for the power calculation.¹⁰ In addition to this, the actual event rate in the control group was lower than the expected 30% (22%), affecting actual power achieved. The nature of the infections seen in the present study also differed from the Italian group. All culture positive infections in the current study were bacterial, but Manzoni et alreported a 5.4% rate of invasive fungal infection in the control group, the absence of which in the lactoferrin group is responsible for an important proportion of the total reduction seen in that study. In populations like those enrolled in the current study where fungal infections are likely to be rarer due to their more mature gestations, this affects both power calculations and changes the number needed to treat for clear benefit. These factors may be responsible for the absence of effect seen in this study.

The analysis of exposure time is justified by the authors, who do however acknowledge that this was not an a priori analysis. It seems reasonable to address the question of whether lactoferrin can be expected to reduce very early sepsis. The 6 infections that occurred before the intervention clearly would not be influenced by the study. It is a shame that the authors do not identify which groups these were in in order to better interpret the number of infections occurring in each group after the intervention was started.

What then should physicians conclude currently, faced with the discrepancy between the effect in the Italian trial, and the lack of effect in this trial? Most clinicians are currently unable to access lactoferrin for clinical use, so the question remains academic to some degree. The currently underway large, well powered UK lactoferrin trial ELFIN¹¹ will hopefully provide further answers. In the interim, clinicians should continue to promote the source of lactoferrin that is most freely and readily available with minimal regulatory control – mothers own fresh colostrum and breast milk.

REFFERENCES

1. Thaver D, Zaidi AK. Burden of neonatal infections in developing countries: a review of evidence from community-based studies. Pediatr Infect Dis J 2009;28:S3-9.
2. Berrington JE, Hearn RI, Bythell M, Wright C, Embleton ND. Deaths in preterm infants: changing pathology over 2 decades. J Pediatr 2012;160:49-53 e1.
3. Stoll BJ, Hansen NI, Adams-Chapman I, et al. Neurodevelopmental and growth impairment among extremely low-birth-weight infants with neonatal infection. JAMA 2004;292:2357-65.
4. Bion J, Richardson A, Hibbert P, et al. ‘Matching Michigan’: a 2-year stepped interventional programme to minimise central venous catheter-blood stream infections in intensive care units in England. BMJ Qual Saf 2013;22:110-23.
5. Jacobs SE, Tobin JM, Opie GF, et al. Probiotic effects on late-onset sepsis in very preterm infants: a randomized controlled trial. Pediatrics 2013;132:1055-62.
6. Leaf A, Dorling J, Kempley S, et al. Early or delayed enteral feeding for preterm growth-restricted infants: a randomized trial. Pediatrics 2012;129:e1260-8.
7. SIFT Study. www.npeu.ox.ac.uk/sift
8. Embleton ND, Berrington JE, McGuire W, Stewart CJ, Cummings SP. Lactoferrin: Antimicrobial activity and therapeutic potential. Semin Fetal Neonatal Med 2013.
9. Berrington JE, Stewart CJ, Embleton ND, Cummings SP. Gut microbiota in preterm infants: assessment and relevance to health and disease. Arch Dis Child Fetal Neonatal Ed 2013;98:F286-90.
10. Manzoni P, Rinaldi M, Cattani S, et al. Bovine lactoferrin supplementation for prevention of late-onset sepsis in very low-birth-weight neonates: a randomized trial. JAMA 2009;302:1421-8.
11. ELFIN Study. https://www.npeu.ox.ac.uk/elfin