Initiation of enteral feeding in preterm infants – how ... - iDOC Africa

MINI-REVIEW
Initiation of enteral feeding in preterm infants – how does evidence inform
current strategies
Authors: Newton Opiyo 1
Eric Ngetich 2
Mike English 1, 3
1. KEMRI/Wellcome Trust Research Programme, Nairobi, Kenya
2. Department of Paediatrics, Moi Teaching and Referral Hospital, Eldoret, Kenya
3. Department of Paediatrics, Oxford University, Oxford, UK
Summary
In initiating enteral feedings in preterm infants, clinicians struggle with questions on (1) when to
initiate (early or late) enteral feeding and how rapidly (slowly or rapidly) to increase feed
volumes. This review summarizes the evidence from systematic reviews and randomised
controlled trials of enteral feeding strategies in preterm neonates, with a particular focus on
defining optimal timing (early versus late) and rates of increase in feed volume (slow versus
rapid) suited to low-income settings where parenteral (intravenous) nutrition is not feasible. The
GRADE (Grading of Recommendations, Assessment, Development and Evaluation) system
was used to appraise the quality of available evidence. The review identified a number of limited
short-term benefits of initiating feeding early and increasing the volume at a relatively rapid rate
using a standardized approach as long as the infant is clinically well. Overall, available very low
to low quality evidence suggests that despite some evidence of benefit from early feeding, it
remains unclear whether preterm infants who receive parenteral nutrition should be given early
or delayed feedings. Similarly, low to moderate quality evidence suggests that there is currently
insufficient evidence to determine whether slowly advancing enteral feed volumes reduces the
risk of mortality or necrotising enterocolitis in preterm neonates. There is need for further clinical
research in this area to develop a scientific basis to current feeding strategies.
1

Background
Very low birth weight (VLBW) and small for gestational age (SGA) infants are at increased risk
of adverse neonatal outcomes. These ‘high risk infants’ frequently demonstrate intolerance of
feeds and high rates of necrotising enterocolitis (NEC) - a syndrome of acute intestinal necrosis
of unknown aetiology [1]. Risk factors for NEC include: extreme prematurity, SGA (due to intrauterine
growth restriction) and enteral feeding [1, 2]. Compared with stable term infants without
NEC, high risk infants who develop NEC have: a greater than 20% mortality, lower levels of
nutrient intake, slower rate of post-natal growth, a higher incidence of nosocomial infections,
longer duration of intensive care and hospital stay, and higher incidence of long-term
neurological disability [1-3].
As preventing prematurity remains an elusive goal feeding regimens therefore try and steer a
course between firm observational evidence suggesting a strong association between enteral
nutrition and NEC (an adverse effect) and the clear relationship between improved nutrition and
better early post-natal growth (a positive effect). Unfortunately, to date, there is no consensus
on how to balance these competing risks and optimise initial feeding strategies in high-risk
infants. More conservative strategies in use include: delaying feeds, slowly increasing feeds,
use of total parenteral nutrition, and prophylactic antibiotics [4, 5]. In this paper we summarize
available best evidence on the safety and effectiveness of methods of establishing enteral
feeding in small, preterm infants. Our aim is to consider how this evidence helps define which
strategies are likely to be best suited to low-income settings in which parenteral nutrition (either
total or partial) is not available to provide additional caloric and nutrient intake. Furthermore the
context in which any guideline might be implemented is one in which the cause of low or very
low birth weight, whether prematurity or SGA, may not be identified by the health workers who
provide the majority of care who have limited training or experience. We also consider the
evidence in the light of an existing Kenyan recommendation, based on expert opinion, that
promotes: (1) relatively ‘early’ introduction of enteral feeds, and; (2) a standardized approach of
relatively ‘rapid’ rates of increasing initial enteral feed volumes (see box 1 outlining current
recommendation).
Methods
Search strategy and study selection criteria
Potential articles for inclusion were identified by direct searches of The Cochrane Library and
MEDLINE (both from inception to March 2009). MEDLINE was searched via PubMed clinical
filters. The searches were performed using combinations of the following Medical Subject
Heading (MeSH) terms: ‘infant’, ‘newborn’, ‘neonate’, ‘premature’, ‘low birth weight’, ‘infant
formula’, ‘formula’, ‘milk’, ‘enteral nutrition’, and ‘enteral feeding’. No time or language
restrictions were applied in these searches. Where high quality, previously conducted
systematic reviews had been performed we included these reviews and did not retrieve or-reevaluate
the individual trials that were included. Trials identified that were not included in
systematic reviews were considered for eligibility. Additional published and unpublished studies
2

were sought by screening through references of identified reviews / study reports and writing to
authors of identified relevant papers / abstracts.
The criteria for considering studies for inclusion in this review are summarised in table 1. Study
selection was done by two independent reviewers (NO and EN). The studies were screened in
two rounds, first on title and abstract, and second on full text, against the pre-defined eligibility
criteria summarised in table 1. Discrepancies between the reviewers were resolved by
discussion.
Quality assessment
Quality of available evidence from the selected studies was assessed using the Grading of
Recommendations, Assessment, Development and Evaluation (GRADE) approach [6]. The
approach classifies the quality of evidence (defined as: ‘the extent to which one can be
confident that an estimate of effect or association can be correct’) into 4 categories: high,
moderate, low, or very low (table 2). The GRADE evidence profiles were prepared by one
reviewer (NO) and verified by a second reviewer (EN). Discrepancies between the reviewers in
the quality ratings were resolved by discussion.
Data extraction and analysis
Data on study characteristics (designs, participants, etc) and outcome measures were extracted
by one reviewer (NO) and checked by a second reviewer (EN). Data abstraction errors were
discussed and corrected. Study results were summarised narratively given important
differences in: study settings, birth weights and gestational ages of studied infants and feeding
regimens.
Results
The study selection process is summarised in figure 1. Overall, 6 papers were included in this
review: four systematic reviews [1, 2, 7, 8], one published randomised controlled trial (RCT) [9],
and one unpublished RCT [10]. Details of the methods and results of the reviews / randomised
controlled trial (RCT) are presented below and summarised in the GRADE evidence profiles
(see tables 2, 4 to 7). Overall, the GRADE quality of evidence varied from low to moderate for
the critical / important outcomes considered.
i) When should enteral feedings be initiated in high-risk infants?
a) Plausibility evidence (derived from RCTs) on ‘maintenance volume’ enteral
feeding on day 1 of life
Evidence for the effectiveness / safety of ‘maintenance volume’ enteral feeding on day 1 of life
was derived from one rigorously conducted World Health Organisation (WHO) technical review
[8]. A common practice in developing countries, where provision of total parenteral nutrition is
very rarely feasible, is to initiate maintenance enteral feeds on day 1 of life and monitor the
3

infants closely. However, some clinicians delay initiation of enteral feeds to day 2, providing
simple glucose / electrolyte maintenance crystalloid fluids intravenously, until the infants have
been assessed to be stable and not at risk of respiratory distress.
To date, only three studies [11-13], conducted in the 1960s in the US and UK, have examined
the effects of such ‘early’ initiation of ‘maintenance’ enteral feeding on day 1 of birth in high risk
infants (table 3). Two of the studies [11, 12] found no significant difference in mortality between
‘early’ (day 1) and ‘delayed’ enterally fed infants. One of the studies [13] however found an
increased risk of death in infants given milk feeds within 2 to 4 hours of birth, compared to those
fed smaller volumes from 12 to 16 hours (mortality rate ratio 2.93, 95% confidence interval (CI):
1.29 to 6.67). The effect of ‘early’ feeding on the time to regain birth weight was varied:
significant improvement was found in one study [12] while no evidence of effect was found in
another [11].
The only other evidence about timing of onset of enteral feeding is from a recent Cochrane
systematic review [7] which identified 3 RCTs (N=115 infants) (table 2). This review found no
significant difference in weight gain, g/kg/week (mean difference -1.00, 95% CI: -127.4 to
125.4), NEC (risk ratio 1.27 to 95% CI: 0.54 to 3.00), mortality (rate ratio 0.87, 95% CI: 0.34 to
2.28) or duration of hospital stay (post-menstrual weeks at discharge) (mean difference, 0.90,
95% CI: -1.21 to 3.01) among infants started on progressive enteral feeds at > 4 days (‘delayed’
feeding) compared to those started on enteral feeds at a mean or median age of ≤ 4 days
(‘early’ feeding) although important beneficial or harmful effects could not be excluded given the
small number of infants studied. The authors concluded that available data are insufficient to
guide clinical practice and recommended further trials to provide robust evidence to inform this
key area of care.
b) Mixed evidence (derived from RCTs / observational studies) on initial enteral
feeding
While early (small) RCTs found no effect of delayed feeding on NEC [14, 15], limited early
(trophic) enteral nutrition has been shown to improve the functional adaptability of the
gastrointestinal tract and eventually feeding tolerance, especially when human milk is used [16,
17] although the practice has been reported to be associated with an increased risk of NEC [18,
19].
Overall, interpretation of the available RCT and observational data on the benefits and risks of
initial enteral feeds is limited by the small number of infants included in studies, considerable
variation and overlap in the definitions of ‘early’ and ‘delayed’ feeding schedules and a number
of inherent methodological flaws (e.g. unclear method of randomization, lack of blinded
assessments of outcomes, and exclusion of infants who developed complications during
hospitalization). Thus, considerable uncertainty persists regarding the optimal timing of initiation
of enteral feeding in high risk infants.
ii) When feedings are advanced, how rapidly should the volume be increased?
4

Controversy persists regarding how fast to advance enteral feeding: limited retrospective data
suggests that rapid rates of feed advancement are associated with increased risk of NEC in
VLBW infants; conversely, slow rates of feed advancement may result in undernutrition,
prolonged exposure to hazards of parenteral nutrition, prolonged hospital stay and delayed
establishment of full enteral nutrition [18, 19].
Evidence on the effects of slow versus rapid rates of advancing feeding volumes in preterm
infants is derived from one updated Cochrane review [2] and one RCT [9] published after the
Cochrane review. The Cochrane review identified three small RCTs (N=396 infants) [20-22].
Entry criteria and rates of advancement of feeding volumes in the included RCTs were varied: in
the Caple trial, infants 1000-2000 g were enrolled and feedings were advanced at 20 mL/kg/day
and 30 mL/kg/day in the slow and rapid groups; in the Salhotra trial, infants with birth weight
less than 1205 g were enrolled, and feedings were advanced at 15 mL/kg/day and 30
mL/kg/day in the slow and rapid groups respectively; in the Rayyis trial, infants 501-1500 g were
enrolled and feedings were advanced at 15 mL/kg/day and 35 mL/kg/day.
Overall, combined results did not detect statistically significant effects on the risk of NEC (risk
ratio 0.96, 95% CI: 0.48 to 1.92) or all-cause mortality (rate ratio 1.40, 95% CI: 0.71 to 2.80).
Infants who had slow rates of feed volume advancement took longer to regain birth weight
(reported mean difference between three and five days). No statistically significant effect on the
total duration of hospital stay was detected. The authors concluded that: (1) available data do
not provide evidence that slow advancement of enteral feed volumes reduces the risk of NEC in
VLBW infants; and (2) increasing the volume of enteral feeds at slow rather than faster rates
results in several days delay in regaining birth weight and establishing full enteral feeds but the
long-term clinical importance of these effects are unclear.
In the RCT [9], a total of 100 stable neonates weighing between 1000 and 1499 g and
gestational age less than 34 weeks were randomly allocated to enteral feeding (expressed
human milk or formula) advancement of 20 mL/kg/day (n=50) or 30 mL/kg/day (n=50).
Neonates in the rapid feeding advancement group achieved full volume feedings before the
slow advancement group (median 7 days versus 9 days, p

feeding protocols: current feeding regimens (CFRs, n=37 babies) and standardised feeding
regimens (SFRs, n=35 babies). The CFR consisted of slow introduction and low volume
increment of enteral feeds. In the SFR group, feeds were introduced within 4 to 24 hours after
birth, and volumes increased rapidly. SFR was associated with (1) a clinically significant
increase in mean weight gain, g/kg/day (SFR 14.1 versus CFR 9.8; p

Conclusions
Available very low to low quality evidence suggests that despite some evidence of benefit from
early feedings, it remains unclear whether low or very low birth weight infants who receive
parenteral nutrition should be given early or delayed feedings. Similarly, low to moderate quality
evidence suggests that there is currently insufficient evidence to determine whether slowly
advancing enteral feed volumes reduces the risk of mortality or necrotising enterocolitis in very
low birth weight infants. Conversely, low quality evidence suggests standardised feeding
regimens in the form of clinical practice guidelines may be important in preventing / minimizing
necrotising enterocolitis in preterm neonates. Overall, further randomised controlled studies that
compare different ages of initiation of feedings, and that assess the effect of varying the rates of
enteral feedings given to very low birth weight infants are needed to strengthen the evidence
base.
7

References
1. Patole S, de Klerk N: Impact of standardised feeding regimens on incidence of neonatal
necrotising enterocolitis: a systematic review and meta-analysis of observational studies.
Arch Dis Child Fetal Neonatal Ed 2005, 90(2):F147-151.
2. McGuire W, Bombell S: Slow advancement of enteral feed volumes to prevent
necrotising enterocolitis in very low birth weight infants. Cochrane Database Syst Rev
2008, 2:CD001241.
3. Bisquera J, Cooper T, Berseth C: Impact of necrotising enterocolitis on length of stay
and hospital charges in very low birth weight infants. Pediatrics 2002, 109:423-428.
4. Churella H, Bachhuber W, MacLean WJ: Survey: methods of feeding low-birth-weight
infants. Pediatrics 1985, 76(2):243-249.
5. Siu Y, Ng P, Fung S, et al: Double blind, randomised, placebo controlled study of oral
vancomycin in prevention of necrotising enterocolitis in preterm, very low birthweight
infants. Arch Dis Child Fetal Neonatal Ed 1998, 79(2):F105-109.
6. Guyatt G, Oxman A, Vist G, et al: GRADE: an emerging consensus on rating quality of
evidence and strength of recommendations. BMJ 2008, 336(7650):924-926.
7. Bombell S, McGuire W: Delayed introduction of progressive enteral feeds to prevent
necrotising enterocolitis in very low birth weight infants. Cochrane Database Syst Rev
2008; 16(2).
8. Edmond K, Bah lR: World Health Organisation. Optimal feeding of low-birth-weight
infants. Technical review. 2006.
9. Krishnamurthy S, Gupta P, Debnath S, Gomber S: Slow versus rapid enteral feeding
advancement in preterm newborn infants 1000-1499 g: a randomized controlled trial.
Acta Paediatr 2010, 99(1):42-46.
10. Ngetich E, Were F, English M, Musoke RN: Effect of a standardised feeding regimen on
early neonatal growth in low and very low birth weight neonates at Kenyatta National
Hospital: a pilot randomised controlled trial. Unpublished MMed thesis, 2009.
11. Cornblath M, Forbes A, Pildes R, et al: A controlled study of early fluid administration on
survival of low birth weight infants. Pediatrics 1966, 38(4):547-554.
12. Smallpeice V, Davies P: Immediate feeding of premature infants with undiluted breastmilk.
Lancet 1964, 2(7374):1349-1352.
13. Wharton B, Bower B: Immediate or later feeding for premature babies? A controlled trial.
Lancet 1965, 2(7420):769-772.
14. LaGamma E, Ostertag S, Birenbaum H: Failure of delayed oral feedings to prevent
necrotizing enterocolitis. Results of study in very-low-birth-weight neonates. Am J Dis
Child 1985, 139(4):385-389.
15. Ostertag S, LaGamma E, Reisen C, Ferrentino F: Early enteral feeding does not affect
the incidence of necrotizing enterocolitis. Pediatrics 1986, 77(3):275-280.
16. Berseth C: Neonatal small intestinal motility: motor responses to feeding in term and
preterm infants. J Pediatr 1990, 117(5):777-782.
17. Lucas A, Bloom S, Aynsley-Green A: Gut hormones and 'minimal enteral feeding'. Acta
Paediatr Scand 1986, 75(5):719-723.
18. Brown E, Sweet A: Preventing necrotizing enterocolitis in neonates. JAMA 1978,
240:2452-2454.
19. Uauy R, Fanaroff A, Korones S, et al: Necrotizing enterocolitis in very low birth weight
infants: biodemographic and clinical correlates. National Institute of Child Health and
Human Development Neonatal Research Network. J Pediatr 1991, 119(4):630-638.
20. Caple J, Armentrout D, Huseby V, et al: Randomized, controlled trial of slow versus
rapid feeding volume advancement in preterm infants. Pediatrics 2004, 114:1597-1600.
8

21. Rayyis S, Ambalavanan N, Wright L, Carlo W: Randomized trial of “slow” versus “fast”
feed advancements on the incidence of necrotizing enterocolitis in very low birth weight
infants. Journal of Pediatrics 1999, 134:293-297.
22. Salhotra A, Ramji S: Slow versus fast enteral feed advancement in very low birth weight
infants: a randomized control trial. Indian Pediatrics 2004, 41:435-441.
9

Tables and Figures
Box 1. Feeding regimens for ‘early’ / ‘delayed’ and ‘slow’ / ‘rapid’ schedules
No respiratory distress / asphyxia
Significant respiratory distress†/ severe asphyxia‡
Weight <
1500g:
‘Early’
Feeding
Weight <
1500g:
‘Delayed’
Feeding
Weight
1500–1999g:
‘Early’
Feeding
Weight
1500 –1999g:
‘Delayed’
Feeding
Start Feeding: 3mls / 3 hrly from birth
Incremental feed volume: Add 3mls to 3 hourly feed
volume every 24 hour period from 24 hours of life until
full feed volume achieved (e.g. at 24 hours feed = 6mls
3hrly, at 48 hours = 9mls / 3hrly)
Start Feeding: 3mls / 3 hrly from 24 hours of age
Incremental feed volume: Add 3mls to 3 hourly feed
volume every 24 hour period from 72 hours of life until
full feed volume achieved (e.g. at 72 hours feed = 6mls
3hrly, at 96 hours = 9mls / 3hrly)
Start Feeding: Initiate full volume feeds appropriate for
weight within 3 hours of birth
Start Feeding: 6mls / 3 hrly from birth
Incremental feed volume: Add 6mls to 3 hourly feed
volume every 24 hour period from 24 hours of life until
full feed volume achieved (e.g. at 24 hours feed = 12mls
3hrly, at 48 hours = 18mls / 3hrly)
Start Feeding: 3mls / 3 hrly from 24 hours of age
Incremental feed volume: Add 3mls to 3 hourly feed volume every 24
hour period from 72 hours of life until full feed volume achieved (e.g.
at 72 hours feed = 6mls 3hrly, at 96 hours = 9mls / 3hrly)
Start Feeding: 3mls / 3 hrly from 72 hours of age
Incremental feed volume: Add 3mls to 3 hourly feed volume every 24
hour period from 120 hours of life until full feed volume achieved
(e.g. at 120 hours feed = 6mls 3hrly, at 144 hours = 9mls / 3hrly)
Start Feeding: 6mls / 3 hrly from birth
Incremental feed volume: Add 6mls to 3 hourly feed volume every 24
hour period from 24 hours of life until full feed volume achieved (e.g.
at 24 hours feed = 12mls 3hrly, at 48 hours = 18mls / 3hrly)
Start Feeding: 6mls / 3 hrly from 24 hours
Incremental feed volume: Add 6mls to 3 hourly feed volume every 24
hour period from 72 hours of life until full feed volume achieved (e.g.
at 72 hours feed = 12mls 3hrly, at 96 hours = 18mls / 3hrly)
†Persistent severe hypoxaemia (SpO 2 remains

Table 1. Study selection criteria
Design
• Randomised or quasi-randomised controlled trials
• Editorials, comments, and case-series were excluded
Participants
• Newborn infants at high risk of feeding intolerance or necrotising enterocolitis on the
basis of low birth weight (

Figure 1. Flow diagram for selection of studies
Identification
Eligibility
Included
478
Articles (after duplicates
removed) identified through
electronic and manual searches
12
Full-text articles assessed for
inclusion
6
Papers included†
466 Articles excluded on
screening titles and
abstracts electronically
6 Articles excluded
4 Narrative reviews
1 Without intervention of
interest
† - 3 Cochrane reviews, 1 non-Cochrane review, 1 randomised
controlled trial (RCT), and 1 unpublished RCT
12

Table 2. GRADE evidence profile
Question: Delayed introduction of progressive enteral feeds to prevent necrotising enterocolitis in very low birth weight infants
Intervention: Early initiation (at a mean or median of ≤4 days) of enteral feedings of human milk or formula
Comparison: Delayed initiation (at a mean or median of >4 days) of enteral feedings of human milk or formula
Settings: Neonatal care units
Bibliography: Bombell S, McGuire W. Delayed introduction of progressive enteral feeds to prevent necrotising enterocolitis in very
low birth weight infants. Cochrane Database Syst Rev. 2008 Apr 16 ;( 2).
Quality assessment
Summary of findings
Effect size
Quality††
Importance
No of
studies
No of
infants
Design Limitations Inconsistency Indirectness Imprecision
(95% CI)
(GRADE)
Mortality prior to discharge
2 101 randomised
controlled
trials
serious†
no serious
inconsistency
no serious
indirectness
serious‡ RR 0.87
(0.34 to
2.28)
⊕⊕
LOW
CRITICAL
Necrotising enterocolitis (NEC)
2 101 randomised
controlled
trials
serious†
no serious
inconsistency
no serious
indirectness
serious‡
RR 1.27
(0.54 to 3.0)
⊕⊕
LOW
CRITICAL
Weight gain (g/kg/week)
1 12 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
very
serious‡
WMD -1.00
(-127.37 to
125.37
⊕
VERY LOW
IMPORTANT
Duration of hospital admission (post-menstrual weeks at discharge)
13

1 60 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious‡ MD 0.90
(-1.21 to
3.01)
⊕⊕
LOW
IMPORTANT
† - small sample sizes, no blinding of intervention and outcome measurement; ‡ – few number of events
††: Quality of evidence – the extent to which we can be confident that an estimate of effect or association is correct. The
judgements are based on the: study design (randomised versus observational studies); likelihood of bias; consistency of the results
across the studies; precision (wide or narrow confidence intervals) of overall estimates and; directness of the evidence with respect
to the populations, interventions and settings where the proposed intervention may be used.
Quality of evidence is categorized as ‘high’, ‘moderate’, ‘low’ or ‘very low’.
• HIGH: Further research is very unlikely to change our confidence in the estimate of effect.
• MODERATE: Further research is likely to have an important impact on our confidence in the estimate of effect and may
change the estimate.
• LOW: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to
change the estimate.
• VERY LOW: We are very uncertain about the estimate.
14

Table 4. GRADE evidence profile
Question: Slow advancement of enteral feed volumes to prevent necrotising enterocolitis in very low birth weight infants
Intervention: Slow rates (up to 24 mls/kg/day) of advancement of enteral feeding
Comparison: Faster rates of advancement of enteral feeding
Settings: Neonatal care units
Bibliography: McGuire W, Bombell S. Slow advancement of enteral feed volumes to prevent necrotising enterocolitis in very low
birth weight infants. Cochrane Database Syst Rev. 2008 Apr 16.
Quality assessment
Summary of findings
Effect size
Quality
Importance
No of
studies
No of
infants
Design Limitations Inconsistency Indirectness Imprecision
(95% CI)
(GRADE)
All cause mortality
2 238 randomised
controlled
trials
serious†
no serious
inconsistency
no serious
indirectness
serious‡ RR 1.40
(0.71 to
2.80)
⊕⊕
LOW
CRITICAL
Necrotising enterocolitis (NEC)
3 396 randomised
controlled
trials
no serious
limitations
no serious
inconsistency
no serious
indirectness
serious‡ RR 0.96
(0.48 to
1.92)
⊕⊕
LOW
CRITICAL
Feeds intolerance (causing interruption of enteral feeding)
1 53 randomised
controlled
trial
serious
no serious
inconsistency
no serious
indirectness
serious‡ RR 1.26
(0.8 to 1.99
⊕⊕⊕
MODERATE
IMPORTANT
Duration of hospital stay (days)
15

2 343 randomised
controlled
trials
serious†
no serious
inconsistency
no serious
indirectness
no serious
imprecision
NSD
⊕⊕⊕
MODERATE
IMPORTANT
Time to regain birth weight
3 396 randomised
controlled
trials
serious†
no serious
inconsistency
no serious
indirectness
no serious
imprecision
Longer in
slow rate
group
(range: 2 to
5 days)
⊕⊕⊕
MODERATE
IMPORTANT
† – parents, care givers, and investigators were aware of allocated interventions; ‡ – few number of events (n=24 deaths, 29 cases
of NEC, 31 cases of feeding intolerance); - small sample size; NSD – no statistically significant difference in the duration of hospital
stay between slow and fast advancement groups
16

Table 5. GRADE evidence profile
Question: Impact of standardised feeding regimens on the incidence of neonatal necrotising enterocolitis: a systematic review and
meta-analysis
Settings: Neonatal care units
Bibliography: Patole SK, de Klerk N. Impact of standardised feeding regimens on incidence of neonatal necrotising enterocolitis: a
systematic review and meta-analysis of observational studies. Arch Dis Child Fetal Neonatal Ed. 2005 Mar; 90(2):F147-51.
Quality assessment
Summary of findings
Risk ratio
Quality
Importance
No of
studies
No of
infants
Design Limitations Inconsistency Indirectness Imprecision
(95% CI)
(GRADE)
Incidence of necrotising enterocolitis (NEC)
6 9,453 observational
studies
no serious
limitations
no serious
inconsistency
no serious
indirectness
no serious
indirectness
0.13 (0.03
to 0.50)
⊕⊕⊕
MODERATE
CRITICAL
17

Table 6. GRADE evidence profile
Question: Slow versus rapid enteral feeding advancement in preterm newborn infants 1000-1499 g: a randomised controlled trial
Intervention: Rapid enteral feeding (expressed human milk or formula) advancement at 30 mL/kg/day
Comparison: Slow enteral feeding (expressed human milk or formula) advancement at 20 mL/kg/day
Settings: Neonatal care units
Bibliography: Krishnamurthy S, Gupta P, Debnath S, Gomber S. Slow versus rapid enteral feeding advancement in preterm
newborn infants 1000-1499 g: a randomized controlled trial. Acta Paediatr. 2010 Jan; 99(1):42-6.
Quality assessment
Summary of findings
Effect size
Quality
Importance
No of
studies
No of
infants
Design Limitations Inconsistency Indirectness Imprecision
(95% CI)
(GRADE)
Mortality
1 100 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious‡
NSD
⊕⊕
LOW
CRITICAL
Necrotising enterocolitis (NEC)
1 100 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious‡
NSD
⊕⊕
LOW
CRITICAL
Days to regain birth weight
1 100 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious‡
Earlier in
rapid
group‡‡
⊕⊕
LOW
IMPORTANT
Length of hospital stay (days)
18

1 100 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious‡
Shorter in
rapid
group††
⊕⊕
LOW
IMPORTANT
† – small sample size, n=100 infants; ‡ – few number of events; NSD – no statistically significant difference in mortality between slow
and fast groups; ‡‡ median 16 days versus 22 days, p

Table 7. GRADE evidence profile
Question: Effect of a standardised feeding regimen on early neonatal growth in low and very low birth weight neonates at Kenyatta
National Hospital: a pilot randomised controlled trial
Settings: Neonatal care units
Bibliography: Ngetich E, Were FN, English M, Musoke RN. Effect of a standardised feeding regimen on early neonatal growth in
low and very low birth weight neonates at Kenyatta National Hospital: a pilot randomised controlled trial. Unpublished MMed thesis,
2009.
Quality assessment
Summary of findings
Effect size
Quality
Importance
No of
studies
No of
infants
Design Limitations Inconsistency Indirectness Imprecision
(GRADE)
Mortality
1 13 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious† CFR 18.9%
versus SFR
17.4%
⊕⊕
LOW
CRITICAL
Necrotising enterocolitis (NEC)
1 6 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious† CFR 8.1%
versus SFR
8.6%
⊕⊕
LOW
CRITICAL
Mean weight gain, g/kg/day
1 55 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious† CFR 9.8
versus
SFR 14.1
p

Time to attain full feeds (mean days)
1 55 randomised
controlled
trial
serious†
no serious
inconsistency
no serious
indirectness
serious† CFR 7.0
versus
SFR 5.6
p