EFFECT OF ADMINISTRATION OF COMBINED ENTERAL LACTOFERRIN AND PROBIOTIC ON INVASIVE FUNGAL INFECTIONS IN PRETERM

Background: Invasive fungal infections (IFI) are responsible for late-onset sepsis in neonates especially preterm, with significant morbidity and mortality. Oral bovine lactoferrin (LF) alone or with probiotics has several potentially significant health benefits and immunomodulation. Objectives: to evaluate efficacy and safety of enterally administered bovine LF alone or in combination with probiotics in prevention of IFI in preterm in comparison to placebo. Patients and Methods: This study is registered in clinicalTrials.gov (ClinicalTrials.gov Identifier: NCT05283278). A prospective double blinded study conducted on 80 preterm, randomly received LF (100 mg/day) (n=20), LF + (LF 100 mg/day plus probiotics 109–1010 colony forming unit) (n=20), or placebo (n=40) for 4 weeks. Blood culture for fungal infection was done at enrollment, days 7, 14, 21 and 28. Results: There was no significant difference between LF and LF+ vs placebo as regards the initial CBC and CRP. Then, placebo showed significantly more clinical signs of sepsis vs LF and vs LF+ (P = 0.003 and 0.001, respectively). Rate of increase


INTRODUCTION
Invasive fungal infections (IFI) are the third most common cause responsible for late-onset sepsis in preterm in neonatal intensive care units (NICUs) after coagulasenegative staphylococci, followed by Gram-negative bacilli.In the past decade, the prevalence of IFIs has increased dramatically.Rates range 1.6-9% in very low birth weight (VLBW)

Sample size:
Sample size determination assuming a rate of IFIs ranging between 0.7% and 2.0% in placebo group and treated group respectively, sample size of 20 patients in each group is enough to detect such difference if true, at 0.05 alpha error and 0.08 power of the test.

Ethical consideration:
Funding Source: None.

Financial
Disclosure: The authors have no financial relationships relevant to this article to disclose.

Conflict of Interest:
The authors have no potential conflicts of interest to disclose.
Parents of the newborns were provided written consent for the protocol which was approved by the Research Ethical Committee of Ain Shams University hospitals, and in accordance with the Helsinki Declaration of 1975.

Exclusion criteria:
• Patients failed to enroll in the 1st 72 hours of age.
• Preterm with conditions necessitating nothing per os such as intestinal obstruction, etc.
• Patients who started prophylaxis with antifungal drugs.
• Patients with clinical congenital anomalies or suspected inborn error of metabolism.Nutritional and feeding policies were stable during the study following our NICU protocol.Administration of expressed maternal milk was encouraged.When needed, feeding supplementations were made with a premature formula.Minimal enteral feeding with small amounts of maternal milk or premature formula (10 mL/kg per day divided every 3 hours) was initiated within the first 2 days of life.

PATIENTS AND METHODS
Cautious volume advancements were performed by adding 10-30 mL/kg per day.Parenteral nutrition was started at admission and continued until enteral feeding reached 150 mL/kg per day.
• Bacterial blood cultures and fungal cultures: For all enrolled neonates blood culture for fungal infection was done at enrollment, days 7, 14, 21 and 28 using Sabouraud agar for detecting Candida species and Hichrome agar to detect other fungal types (primary outcome), we compared length of hospital stay (LOS), use and duration of inotropes, use and duration of mechanical ventilation, rate of increase of enteral feeding, time to reach full enteral intake, signs of feeding intolerance, complete blood count (CBC), C reactive protein (CRP), packed red blood cell transfusion (PRBCs) and mortality in the 3 groups as (secondary outcome).
Under strict aseptic technique we collected blood samples for bacterial blood cultures routinely withdrawn on baby's admission using conventional blood culture technique and done using BD BACTEC PEDS PLUS/F culture vials (Becton, Dickinson and Company Spark, Ireland), and fungal cultures at enrolment, days 7, 14, 21 and 28.

Blood
cultures incubated aerobically at 37°C for 2-3 days for yeast and for 2-3 weeks for fungi.Incubated media is then subcultured on Sabauroud agar (Himedia company, India ® ) for detecting fungal species and Hichrome candida differential agar (Himedia company, India ® ) to detect candida species.Sabouroud and Hichrome agars were incubated at 37°C and at room temperature and weren't discarded until 2 weeks.

Statistical analysis:
Data were collected, revised, coded, and entered Statistical Package for Social Science (IBM SPSS), version 23.Parametric quantitative data were presented as mean, standard deviation and range, while non-parametric data were presented as median and inter-quartile ranges (IQR).Qualitative variables were presented as numbers and percentages.We used chi-square test and/or Fisher exact test for comparison of qualitative data within groups.Independent t test was used to compare parametric quantitative data and Mann-Whitney for non-parametric data.
The confidence interval was set to 95% and the margin of error accepted was set to 5%.So, p value was considered significant if <0.05 and highly significant if <0.01.

RESULTS
Our results will be demonstrated in the following tables and figures:   There was no significant difference between LF and LF+ vs placebo as regards the type of feeding.However, both LF and LF+ showed highly significant rapid rate of increase of milk intake and significantly were faster to reach full enteral intake than placebo (Table 2).The placebo significantly needed more inotropic supports vs either LF or LF+.We also found that both LF and LF+ required mechanical ventilation less than placebo, though it only reached statistically significant in LF+ vs placebo.Both LF and LF+ groups had less necrotizing enterocolitis (NEC) than placebo yet did not reach statistical significance.And though both LF and LF+ groups had less PRBC transfusion than placebo, only LF vs placebo reached statistical significance (Table 4).

Figures (3): Comparison between the 3 studied groups as regards platelets level on days 7, 14, 21 and 28
There was no significant difference between LF and LF+ vs placebo as regards the initial CBC and CRP, however, as shown in Figures (1, 2, and 3) both LF and LF+ had higher hemoglobin and hematocrit levels than placebo, and that LF+ had higher platelets than placebo.In the present study, we demonstrated that there was no significant difference between LF vs placebo nor LF+ vs placebo as regards type of feeding, however, both LF and LF+ showed significantly faster rate of feeding advancement, reached full enteral intake faster, additionally they had less NEC than placebo.In the present study we found that both LF and LF+ had higher hemoglobin and hematocrit levels than placebo and that LF+ had higher platelets than placebo on day 7, 14, 21 and 28 of life.This was comparable to a study by Ke et al., 2015 who declared that infants aged 4 to 6 months in LF group had significantly higher hemoglobin and serum ferritin, than those infants in control group after 3 months of intervention.

DISCUSSION
There were no significant differences between both LF and placebo groups as regards serum ferritin, Hb, hematocrit, mean corpuscular volume, red cell distribution width, platelet count, and total leukocytic count on day 7 (El Barbary et al., 2018), however, they found that there was statistically significant higher serum ferritin, Hb, hematocrit, and mean corpuscular volume, and lower red cell distribution width and total leukocytic count in the LF group than the placebo group on day 30.The placebo had more clinical signs of sepsis, needed more inotropic supports and mechanical ventilation vs LF and vs LF+ groups.Also, we found that there were no reported deaths in both LF and LF+ groups however, in placebo group there was 5 (25%) deaths, also both LF and LF+ needed shorter LOS than placebo.This was comparable to Indrio et al., 2017, as they declared that newborns receiving probiotics had a significant reduction in LOS (p < 0.01), and days of antibiotic treatment (p < 0.01).
In one meta-analysis of 13 randomized controlled trials (RCT) involving 1,969 patients including neonates, pediatrics, adults, and geriatrics, it was found that probiotics significantly decreased risk of ventilator-associated pneumonia in mechanically ventilated patients (Weng et al., 2017).Also, Manzoni et, 2009 found that both LF and LF + probiotic groups had less LOS vs placebo (P=0.002 and P<0.001, respectively), and less reported mortality in the 2 treatment groups vs placebo (P=0.008 and P=0.04, respectively).
However, in contrast to our results, a meta-analysis that included nine RCTs, indicated that enteral LF was not associated with lowering in LOS in all neonates.However, only in subgroup analysis could demonstrate that enteral LF significantly decreased the incidence of LOS in VLBW and ELBW neonates.LF supplementation did not reduce the incidence of NEC stage II or III, bronchopulmonary dysplasia, retinopathy of prematurity, IFI, intraventricular hemorrhage, urinary tract infection, or mortality compared with placebo (Gao et al., 2020).
In the present study, we reported that placebo had more fungal infection vs LF and LF+ groups, and that for Candida albicans, there was no infection reported in both LF and LF+, while 12/40 cases were reported in placebo.As regards Candida Glabrata, tropicalis and krusei, they were more reported in the placebo than in LF and LF+.

Manzoni et al., 2012, in their
study showed that neonates with high intensity of colonization (> 3 different sites concomitantly colonized) was lower in LF or LF+ probiotic, but the difference was not significant (P = 0.10).The overall incidence of IFI was lower in both LF and LF plus probiotic groups compared with placebo (P = 0.002 and 0.02, respectively).The rate of progression from colonization to infection was significantly lower in both LF groups (P= 0.001 and P = 0.02 respectively).In LF groups, IFIs were less frequent in infants < 1500 g.Also, comparable to our study, Manzoni et al., 2012 found that Candida albicans was the most frequent Candida.And due to the small numbers of isolates, only Candida albicans was possible to reveal a significant difference between groups, with 60% and 80% reduction in the treatment groups.No intolerances or adverse effects to bovine LF with/ without probiotics were recorded in the current study.

CONCLUSIONS
We concluded from our study that prophylactic enteral administration of LF either alone or in combination with probiotic has positive impact on reduction of IFI, increased hemoglobin concentration, decrease frequency of blood transfusion, LOS, and mortality rate in preterm.

RECOMMENDATION
Routine supplementation of preterm with LF either alone or in combination with probiotics to reduce IFI.Also, studies on larger patient samples are recommended and should assess whether different types of feeding: breastfeeding, formula or mixed may influence the outcome of preterm patients supplemented by either LF alone or in combination of probiotics vs placebo group.

LIMITATIONS
We did not stratify patients according to the type of feeding, small sample size and difficulty of follow up of studied neonates after discharge for the period of the study.

Figures
Figures (1): Comparison between the 3 studied groups as regardshemoglobin level on days 7, 14, 21 and 28 LF has a specific antifungal action through binding to cell wall's receptor and its disturbance (Lupetti et al., 2003).It could also modify the growth and development of the nascent enterocytes, promoting gut barrier thus blocking translocation of colonizing fungal colonies from the gut to the blood (Buccigrossi et al., 2007) (Manzoni et al., 2012).Probiotics have been shown to be able to prevent enteric fungal colonization, thus a synergistic activity with LF can enhance such effect and further lower the rate of colonization and possibly infection in the LF + group (Romeo et al., 2011).