Accepted Manuscript Probiotics prevent Candida colonization and invasive fungal sepsis in preterm neonates: a systematic review and meta-analysis of randomized controlled trials Hua-Jian Hu, Guo-Qiang Zhang, Qiao Zhang, Shristi Shakya, Zhong-Yue Li PII:
S1875-9572(16)30140-1
DOI:
10.1016/j.pedneo.2016.06.001
Reference:
PEDN 607
To appear in:
Pediatrics & Neonatology
Received Date: 9 October 2015 Revised Date:
3 May 2016
Accepted Date: 27 June 2016
Please cite this article as: Hu H-J, Zhang G-Q, Zhang Q, Shakya S, Li Z-Y, Probiotics prevent Candida colonization and invasive fungal sepsis in preterm neonates: a systematic review and meta-analysis of randomized controlled trials, Pediatrics and Neonatology (2016), doi: 10.1016/j.pedneo.2016.06.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
PEDN-D-15-00268_After Eng Edited_final 1
ACCEPTED MANUSCRIPT Review Article Probiotics Prevent Candida Colonization and Invasive Fungal Sepsis in Preterm Neonates: a Systematic Review and Meta-Analysis of Randomized Controlled
Running Title: Probiotics on fungal infection in preterm
RI PT
Trials
a
SC
Hua-Jian Hu a, Guo-Qiang Zhang a, Qiao Zhang a, Shristi Shakya a, Zhong-Yue Li a,*
Department of Gastroenterology, Children’s Hospital of Chongqing Medical
M AN U
University, Ministry of Education Key Laboratory of Child Development and Disorders, Key Laboratory of Pediatrics in Chongqing, Chongqing International Science and Technology Cooperation Center for Child Development and Disorders, Chongqing, China.
TE D
* Corresponding author. Department of Gastroenterology, Children’s Hospital of Chongqing Medical University, Ministry of Education Key Laboratory of Child Development and Disorders, Key Laboratory of Pediatrics in Chongqing, Chongqing
EP
International Science and Technology Cooperation Center for Child Development and Disorders, No.136, Zhongshan 2nd Road, Yuzhong District, 400014, Chongqing,
AC C
China. E-mail:
[email protected].
2
ACCEPTED MANUSCRIPT ABSTRACT
To investigate whether probiotic supplementation could reduce risk of fungal infection in preterm neonates in neonatal intensive care units (NICUs), PubMed,
RI PT
EMBASE, and CENTRAL were systematically searched for randomized controlled trials (RCTs) regarding the effect of probiotics on fungal infection in preterm neonates. The outcomes of interest were Candida colonization and invasive fungal
SC
sepsis. Seven trials involving 1,371 preterm neonates were included. Meta-analysis
M AN U
(fixed-effects model) showed that probiotic supplementation was significantly associated with a lower risk of Candida colonization (two RCTs, n = 329; RR 0.43, 95% CI 0.27 to 0.67; P = 0.0002; I2 = 0%), and invasive fungal sepsis (seven RCTs, n = 1,371; RR 0.64, 95% CI 0.46 to 0.88; P = 0.006; I2 = 13%). After excluding the
TE D
study with a high baseline incidence (75%) of fungal sepsis, the effect of probiotics on invasive fungal sepsis became statistically nonsignificant (RR 0.88, 95% CI 0.44 to 1.78; P = 0.72; I2 = 15%). When using random-effects model, the favorable effect
EP
of probiotics on Candida colonization remained (RR 0.43, 95% CI 0.27 to 0.68; P =
AC C
0.0002; I2 = 0%), but not that on fungal sepsis (RR 0.64, 95% CI 0.38 to 1.08; P = 0.10; I2 = 13%). Current evidence indicates that probiotics can reduce risk of Candida colonization in preterm neonates in NICUs. There are limited data to support probiotic supplementation preventing invasive fungal sepsis in preterm. High-quality and adequately powered RCTs are warranted.
Key Words: Candida colonization; invasive fungal sepsis; meta-analysis; preterm;
3
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
probiotics
4
ACCEPTED MANUSCRIPT INTRODUCTION
Candida species are the third most common agent responsible for late-onset sepsis in preterm neonates in neonatal intensive care units (NICUs).1-3 The prevalence of
RI PT
invasive fungal infections (IFIs) has dramatically increased in the past decade,4,5 with rates of 1.6% to 9% in very low birth weight neonates (VLBW; < 1500 g) and 15% in extremely low birth weight neonates (ELBW; < 1000 g).6-9 Neonatal IFIs are
SC
associated with increased length of hospital stay, high morbidity and mortality, and
M AN U
neurodevelopmental impairment, which have been questioning the survival of preterm infants.5,8,10
Preterm neonates in NICUs are highly prone to develop IFIs due to the immaturity of skin/mucosal barrier and immune response, invasive diagnostic and
TE D
therapeutic procedures, and use of broad-spectrum antimicrobial drugs, as well as exposure to the hospital milieu, which leads to gastrointestinal colonization with fungi.8 It has been shown that fungal colonization is associated with an increased risk
EP
of developing IFIs, and enteric colonization by Candida is the most important
AC C
predictor of IFIs.11-15 Reducing fungal colonization by using systemic antifungal drugs has been reported to be effective in preventing IFIs.16-21 Despite of these reports on the effective use of antifungal agents for prophylaxis, concerns still exist with respect to cost, tolerability, long-term safety, and emergence of resistant strains. Probiotics, defined as live microorganisms, confer health benefits to the host when administered at adequate doses.22 It has been substantially reported that probiotics could reduce both Candida colonization and IFIs in mice models.23-26
5
ACCEPTED MANUSCRIPT
However, researches in preterm neonates, the most vulnerable patient group cared for in NICUs, are surprisingly scant and controversial. Furthermore, due to small sample sizes, these studies were not adequately powered to detect the effect of probiotics on
RI PT
enteric colonization by Candida and fungal sepsis. Thus, in order to provide the latest and most convincing evidence, we systematically reviewed the current available
M AN U
and IFIs in preterm neonates in NICUs.
SC
literature to investigate whether probiotics reduced the risk of Candida colonization
METHODS
This systematic review and meta-analysis was conducted and reported in adherence to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement
of Interventions.28
TE D
(PRISMA),27 and the guidelines of the Cochrane Handbook for Systematic Reviews
EP
Literature Search and Selection Criteria
AC C
PubMed, EMBASE and the Cochrane Central Register of Controlled Trials (CENTRAL) databases were searched for records that compared enteral probiotics to placebo or no intervention in preterm neonates. We also searched ClinicalTrials.gov (https://clinicaltrials.gov/) and the European Union Clinical Trials Register (https://www.clinicaltrialsregister.eu/). The keywords were as follows: ‘probiotic’, ‘probiotics’, ‘yogurt’, ‘yoghurt’, ‘lactic acid bacteria’, ‘acidophilus’, ‘Lactobacillus’, ‘Lactococcus’, ‘Saccharomyces’, ‘Streptococcus’, ‘Bifidobacterium’, ‘Enterococcus’,
6
ACCEPTED MANUSCRIPT
‘Escherichia coli’, ‘low birth weight’, ‘LBW’, ‘very low birth weight’, ‘VLBW’, ‘extremely low birth weight’, ‘ELBW’, ‘preterm’ and ‘premature’. The last search was conducted on August 20, 2015. No language restriction was imposed. All
RI PT
citations were imported into a bibliographic database (EndNote X7; Thomson Reuters) for assessment of eligibility. Two authors (H-JH and G-QZ) independently conducted the initial search, removed duplicate articles, screened the titles and abstracts for
SC
relevance, and identified as excluded or requiring further assessment. Then the
M AN U
full-text articles were reviewed for inclusion. The references of the retrieved articles and relevant reviews were also manually checked to identify any additional eligible trials.
Studies meeting the following inclusion criteria were included: (i) population:
TE D
gestational age < 37 weeks or birth weight < 2500 g, or both; (ii) intervention: any species/strains/doses of live probiotics administered for more than seven days; (iii) comparators: placebo or no probiotics; (iv) outcome: the primary outcomes were
culture,
and
invasive
fungal
sepsis,
confirmed
by
positive
AC C
specimens
EP
Candida colonization, monitored by oropharyngeal/gastric aspirate/stool/rectal
blood/urine/cerebrospinal fluid culture; and (v) study design: randomized controlled trials (RCTs). We excluded interventions other than live probiotics, administration of probiotics with prebiotics or other agents, and those conducted in full-term infants or children. Discrepancies regarding study inclusion between the two authors (H-JH and G-QZ) were resolved by consensus.
7
ACCEPTED MANUSCRIPT Data Extraction and Quality Assessment
Two of the authors (H-JH and G-QZ) independently extracted relevant data from each included study by using a unified data form. Extracted data were entered into a
RI PT
standardized Word file. The items included in the data form were as follows: source (first author, year of publication, country), number of preterm neonates enrolled, strains/doses/duration of probiotics administered, type of milk (human milk or
SC
formula), and outcomes of interest (Candida colonization and/or fungal sepsis). To
M AN U
minimize the possibility of errors, all data were compared by the same two authors (H-JH and G-QZ), and disagreements were resolved after further checking on the original articles. Article authors were contacted in case of inadequate information to identify relevant data.
TE D
To measure study quality, two authors (H-JH and G-QZ) independently assessed the following criteria according to the Cochrane Risk-of-Bias Tool:29 adequate sequence generation, allocation concealment, blinding of participants and personnel,
EP
binding of outcome assessment, incomplete outcome data, selective reporting, and
AC C
other bias. Judgment of bias pertaining to each item was categorized as low, high, or unclear risk, according to the criteria specified in the Cochrane handbook.29
Statistical Analysis
To evaluate the effect of probiotics, we calculated relative risks (RRs) for the incidence of Candida colonization and invasive fungal sepsis between intervention and control groups. When trials investigated two separate probiotic groups versus
8
ACCEPTED MANUSCRIPT
placebo, data on the two probiotic groups were combined into a single RR, which we included in the meta-analysis. Heterogeneity across studies was tested by using the I2 statistic. Studies with an I2 value greater than 50% were considered to have significant
RI PT
heterogeneity.30 The Mantel-Haenszel method with fixed-effects or random-effects model was used to calculate pooled RRs and 95% confidence intervals (CIs). A sensitivity analysis was conducted to assess the influence of individual studies on the
SC
pooled result, by excluding each study one by one and recalculating the combined
M AN U
RRs on the remaining trials. An assessment of publication bias was performed by using the Begg’s and Egger’s tests.31,32 A P value less than 0.05 was considered as statistically significant, except where otherwise specified. All statistical analyses were performed using the Stata 12.0 (Stata Corporation, College Station, TX, USA) and
RESULTS
TE D
RevMan 5.3 (The Nordic Cochrane Center, Copenhagen, Denmark).
EP
A total of 637 records were identified by the initial electronic literature search. 128
AC C
records were excluded for duplicates, and 478 records were excluded based on the titles and abstracts. The remaining 31 full-text articles were assessed for eligibility, of which 21 were excluded because incidence of fungal sepsis was not reported. Three trials were further excluded due to ineligible population33 and ineligible intervention.34,35 Finally, seven trials were included in our review.36-42 The selection process is shown in Figure 1. Characteristics of the seven trials are summarized in Table 1, and the outcome data of each included study are presented in Table 2. The
9
ACCEPTED MANUSCRIPT
quality of the trials assessed by the Cochrane Risk-of-Bias Tool is summarized in Table 3. The incidence of enteric colonization by Candida was reported in two trials.37,39
RI PT
The trial by Roy et al40 reported stool fungal counts rather than incidence of Candida colonization. Figure 2 shows the results from each trial and overall, using a fixed-effects model, for probiotics in the prevention of Candida colonization and
SC
fungal sepsis. Our analysis indicated that prophylactic probiotics significantly reduced
M AN U
the incidence of Candida colonization (RR 0.43, 95% CI 0.27 to 0.67; P = 0.0002; I2 = 0%), and the risk of developing invasive fungal sepsis (RR 0.64, 95% CI 0.46 to 0.88; P = 0.006; I2 = 13%). When excluding Roy et al40 from the overall analysis, the effect of probiotics on invasive fungal sepsis became statistically nonsignificant (RR
TE D
0.88, 95% CI 0.44 to 1.78; P = 0.72; I2 = 15%). When using random-effects model, the protective effect of probiotics on Candida colonization remained (RR 0.43, 95% CI 0.27 to 0.68; P = 0.0002; I2 = 0%), but not that on fungal sepsis (RR 0.64, 95% CI
EP
0.38 to 1.08; P = 0.10; I2 = 13%). For the meta-analysis of probiotics on fungal sepsis,
AC C
there was no evidence of significant publication bias by formal statistical tests (Egger’s test, P = 0.445; Begg’s test, P = 0.260). None of the included studies reported any systemic infection caused by the supplemented probiotic organisms.
DISCUSSION Seven RCTs with a total of 1,371 preterm neonates were included in our review. With the limited evidence available, our results indicated that the use of probiotics could
10
ACCEPTED MANUSCRIPT
reduce the incidence of Candida colonization in preterm neonates in NICUs. There were limited data to support probiotic supplementation preventing invasive fungal sepsis in preterm. None of the included trials reported any systemic infection caused
RI PT
by the supplemented probiotic organisms. Studies in mice models have shown the efficacy of probiotics in reducing the risk of both enteric fungal colonization and systemic fungal infections.23-26,43,44 Two recent
SC
trials reported that prophylactic supplementation of Saccharomyces boulardii or
M AN U
Lactobacillus reuteri was as effective as nystatin in preventing fungal colonization and IFIs in preterm neonates in NICUs.45,46 The efficacy of a mixture of species of probiotics in the prevention of rectal colonization by Candida was also investigated and confirmed in critically ill children in a PICU.33 Moreover, several trials have
TE D
reported that probiotics are effective in preventing and treating vulvovaginal candidiasis.47-49 In general, current evidence indicates that colonization by probiotics can afford protection against the fungal proliferation in the gastrointestinal tract and
EP
prevent subsequent IFIs. The plausible biological mechanisms by which probiotics
AC C
might prevent IFIs include competitively colonizing the gut, competitive exclusion of fungi,25,26 augmentation of IgA mucosal responses,50,51 modulation of the gut barrier function and permeability,52 production of antimicrobial peptides,53 and upregulation of immune responses.54
Implications for Practice The results of our review provided preliminary evidence that probiotics might be
11
ACCEPTED MANUSCRIPT
useful in reducing enteric Candida colonization and further preventing invasive fungal sepsis in preterm neonates in NICUs. However, in clinical practice, the common strategy of preventing fungal infection in preterm infants is using
RI PT
intravenous fluconazole or oral nystatin. To our knowledge, only two trials compared probiotics with nystatin regarding their effects on fungal infection in preterm infants. Oncel et al45 and Demirel et al46 consistently concluded that supplementation of
SC
Lactobacillus reuteri or Saccharomyces boulardii not only acted as effectively as
M AN U
nystatin with respect to fungal colonization and IFIs, but more effectively reduced the incidence of sepsis, duration of hospitalization or feeding intolerance. Moreover, updated meta-analyses have confirmed the efficacy and safety of probiotics in preventing necrotizing enterocolitis and late-onset sepsis in preterm neonates.55,56
TE D
Given the concerns about medical costs, tolerability, long-term safety, and emergence of resistant strains with the use of antifungal agents, prophylactic probiotics merit further consideration as a potential prevention strategy to prevent fungal infection in
EP
preterm infants. Manzoni et al37 also recommended the use of combinations of
AC C
antifungal drugs and probiotics in all VLBW neonates, with greater reliance on probiotics in larger neonates and lesser reliance in smaller neonates. Hence, this prophylactic option could be discussed with patients or caregivers in clinical practice. However, due to the paucity of data comparing probiotics with fluconazole or nystatin, head-to-head comparative studies are still required to assess the most effective preparations. It is noteworthy that the beneficial effects of probiotics seem to be strain-specific
12
ACCEPTED MANUSCRIPT
and the exact mechanism by which probiotic organisms prevent fungal infection remains unknown. Several cases of systemic infections caused by supplemented probiotics have been reported, which were reviewed elsewhere.55,56 Hence, caution is
RI PT
needed when prescribing probiotics. The uncertain efficacy and safety of probiotics has restricted the clinical use of probiotics in ICUs and most ICU pharmacists would not currently recommend probiotics for the prevention of ventilator-associated
SC
pneumonia.57 Hence, further clinical and experimental studies are strongly needed to
M AN U
accurately determine suitable probiotic organisms, optimal dose, timing of administration, duration of treatment, and safety.
Quality of Evidence
TE D
The strengths of our study included the completeness of the search strategy, which reviewed multiple citation databases and trial registries. By omitting outcome-related search terms, we identified trials that were not primarily focused on fungal infection,
EP
but nevertheless reported relevant outcomes.36,38,41,42 Moreover, low heterogeneity and
AC C
lack of publication bias added robustness to our main findings. Several limitations should be taken into consideration when interpreting the
results. First, although no statistical heterogeneity was found for the primary outcomes, population characteristics, probiotic regimens (various organisms, daily doses, time of initiation, and length of intervention) and type of milk differed across the included studies. We adopted random-effects model to try to account for this variability. Second, only a limited number of trials, most of which included small
13
ACCEPTED MANUSCRIPT
sample sizes, were available for this meta-analysis, which could lead to a spurious result. However, to increase power is one of the reasons why we conducted this review. Third, seven trials were included for the outcome of invasive fungal sepsis, of
RI PT
which only one trial by Roy et al40 was primarily designed to evaluate the effect of probiotics on invasive fungal sepsis. When excluding Roy et al from the overall analysis, the pooled incidence of fungal sepsis was 1.9% and 2.2% in the probiotics
SC
group and placebo group, respectively. Due to the low incidence of fungal sepsis and
M AN U
relatively small sample sizes, the six trials were not adequately powered to detect the effect of probiotics on fungal sepsis, and therefore the effect of probiotics on fungal sepsis became statistically nonsignificant. The trial by Roy et al has a very high incidence of fungal sepsis in their center, 41% in probiotics group and 75% in placebo
TE D
group. We can speculate that the reason why the effect of probiotics on fungal sepsis reached statistical significance in the trial by Roy et al is the high incidence of fungal sepsis, which results in enough power to detect beneficial effects. However, this result
EP
needs to be confirmed by further adequately powered RCTs. Fourth, the
AC C
methodological quality of the included trials varied and was poor overall with an unclear risk of selection bias and performance bias, both of which have been known to significantly increase the likelihood that the intervention would appear effective. Hence, high-quality RCTs are warranted. Fifth, only one of the included RCTs enrolled ELBW infants as the study subjects, and our study was unable to define the effects of probiotics in ELBW infants, who were at greatest risk of developing sepsis and necrotizing enterocolitis. Studies in ELBW infants are greatly needed.
14
ACCEPTED MANUSCRIPT
CONCLUSIONS Current evidence indicates that probiotics can reduce risk of Candida colonization in
RI PT
preterm neonates in NICUs. There are limited data to support probiotic supplementation preventing invasive fungal sepsis in preterm. This finding largely relies on several trials of low methodological quality, i.e., unclear risk of selection
SC
bias and performance bias. High-quality and adequately powered RCTs are warranted,
M AN U
especially in ELBW infants. While it is too soon to recommend its routine use in clinical practice, the prophylactic option could be discussed with patients or
AC C
EP
TE D
caregivers.
15
ACCEPTED MANUSCRIPT
Conflicts of Interest Statement: None of the authors has conflicts of interest relevant to this article to disclose.
AC C
EP
TE D
M AN U
SC
Chongqing City Foundation (2013-1-023).
RI PT
Acknowledgments: This study was supported by the Health Department of
16
ACCEPTED MANUSCRIPT REFERENCES 1.
Manzoni P. Use of Lactobacillus casei subspecies Rhamnosus GG and gastrointestinal
colonization by Candida species in preterm neonates. J Pediatr Gastroenterol Nutr
2.
RI PT
2007;45:S190－4. Dong Y, Speer CP. Late-onset neonatal sepsis: recent developments. Arch Dis Child Fetal
Neonatal Ed 2015;100:F257－63.
Tsai MH, Hsu JF, Chu SM, Lien R, Huang HR, Chiang MC, et al. Incidence, clinical
SC
3.
Infect Dis J 2014;33:e7－13. 4.
M AN U
characteristics and risk factors for adverse outcome in neonates with late-onset sepsis. Pediatr
Shane AL, Stoll BJ. Recent developments and current issues in the epidemiology,
diagnosis, and management of bacterial and fungal neonatal sepsis. Am J Perinatol
5.
TE D
2013;30:131－41.
Filioti J, Spiroglou K, Panteliadis CP, Roilides E. Invasive candidiasis in pediatric
intensive care patients: epidemiology, risk factors, management, and outcome. Intensive Care
Stoll BJ, Hansen N, Fanaroff AA, Wright LL, Carlo WA, Ehrenkranz RA, et al.
AC C
6.
EP
Med 2007;33:1272－83.
Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics 2002;110:285－91. 7.
Tezer H, Canpolat FE, Dilmen U. Invasive fungal infections during the neonatal period:
diagnosis, treatment and prophylaxis. Expert Opin Pharmacother 2012;13:193－205. 8.
Spiliopoulou A, Dimitriou G, Jelastopulu E, Giannakopoulos I, Anastassiou ED,
Christofidou M. Neonatal intensive care unit candidemia: epidemiology, risk factors, outcome,
17
ACCEPTED MANUSCRIPT and critical review of published case series. Mycopathologia 2012;173:219－28. 9.
Ali GY, Algohary EH, Rashed KA, Almoghanum M, Khalifa AA. Prevalence of Candida
colonization in preterm newborns and VLBW in neonatal intensive care unit: role of maternal
RI PT
colonization as a risk factor in transmission of disease. J Matern Fetal Neonatal Med 2012;25:789－95.
10. Adams-Chapman I, Bann CM, Das A, Goldberg RN, Stoll BJ, Walsh MC, et al.
SC
Neurodevelopmental outcome of extremely low birth weight infants with Candida infection. J
M AN U
Pediatr 2013;163:961－7.
11. Cerikcioglu N, Ilki A, Bilgen H, Ozek E, Metin F, Kalaca S. The relationships between candidemia and candidal colonization and virulence factors of the colonizing strains in preterm infants. Turk J Pediatr 2004;46:245－50.
TE D
12. Manzoni P, Farina D, Antonielli d’Oulx E, Leonessa ML, Gomirato G, Arisio R. An association between anatomic site of Candida colonization and risk of invasive candidiasis exists also in preterm neonates in neonatal intensive care unit. Diagn Microbiol Infect Dis
EP
2006;56:459－60.
AC C
13. Vendettuoli V, Tana M, Tirone C, Posteraro B, La Sorda M, Fadda G, et al. The role of Candida surveillance cultures for identification of a preterm subpopulation at highest risk for invasive fungal infection. Pediatr Infect Dis J 2008;27:1114－6. 14. Singhi S, Rao DS, Chakrabarti A. Candida colonization and candidemia in a pediatric intensive care unit. Pediatr Crit Care Med 2008;9:91－5. 15. Saiman L, Ludington E, Dawson JD, Patterson JE, Rangel-Frausto S, Wiblin RT, et al. Risk factors for Candida species colonization of neonatal intensive care unit patients. Pediatr
18
ACCEPTED MANUSCRIPT Infect Dis J 2001;20:1119－24.
16. Manzoni P, Stolfi I, Pugni L, Decembrino L, Magnani C, Vetrano G, et al. A multicenter, randomized trial of prophylactic fluconazole in preterm neonates. N Engl J Med
RI PT
2007;356:2483－95. 17. Manzoni P, Mostert M, Latino MA, Pugni L, Stolfi I, Decembrino L, et al. Clinical characteristics and response to prophylactic fluconazole of preterm VLBW neonates with
SC
baseline and acquired fungal colonisation in NICU: data from a multicentre RCT. Early Hum
M AN U
Dev 2012;88:S60－4.
18. Kaufman D, Boyle R, Hazen KC, Patrie JT, Robinson M, Grossman LB. Twice weekly fluconazole prophylaxis for prevention of invasive Candida infection in high-risk infants of
