Push versus gravity for intermittent bolus gavage tube feeding of premature and low birth weight infantsSkip sharing on social media links
1Neonatal Services, The Royal Women's Hospital, Parkville, Australia [top]
2Neonatal Intensive Care Unit, University College London Hospital, London, UK [top]
3Grace Centre for Newborn Care, The Children's Hospital at Westmead, Sydney, Australia [top]
4Faculty of Medicine, Central Clinical School - Discipline of Obstetrics, Gynaecology & Neonatology & Faculty of Nursing, University of Sydney, Sydney, Australia [top]
Citation example: Dawson JA, Summan R, Badawi N, Foster JP. Push versus gravity for intermittent bolus gavage tube feeding of premature and low birth weight infants. Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD005249. DOI: 10.1002/14651858.CD005249.pub2.
Jennifer A Dawson
The Royal Women's Hospital
20 Flemington Road
|Assessed as Up-to-date:||11 July 2012|
|Date of Search:||01 May 2012|
|Next Stage Expected:||11 July 2014|
|Protocol First Published:||Issue 2, 2005|
|Review First Published:||Issue 11, 2012|
|Last Citation Issue:||Issue 11, 2012|
|Date / Event||Description|
|07 August 2008
Converted to new review format.
|16 May 2006
New citation: conclusions changed
Many small, sick and premature infants are unable to coordinate sucking, swallowing and breathing, and therefore, require gavage feeding. In gavage feeding, milk feeds are delivered through a tube passed via the nose or mouth into the stomach. Intermittent bolus milk feeds may be administered using a syringe to gently push milk into the infant's stomach (push feed). Alternatively, milk can be poured into a syringe attached to the tube and allowed to drip in by gravity (gravity feed).
To determine whether the use of push compared with gravity gavage feeding results in a more rapid establishment of full gavage feeds without increasing adverse events in preterm or low birth weight, infants who require intermittent bolus gavage feeding.
We searched the following electronic databases to locate randomised controlled or quasi-randomised trials: Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, 2012, Issue 5), MEDLINE (from 1966 to May 2012), EMBASE (from 1980 to May 2012), and CINAHL (from 1982 to May 2012). We used the standard search strategy of the Cochrane Neonatal Review Group.
Randomised or quasi-randomised controlled trials comparing push versus gravity intermittent gavage tube feeding in premature or low birth weight, or both, infants.
Data collection and analysis
We assessed the methodology of trials regarding blinding of randomisation and outcome measurement. We evaluated treatment effect with a fixed-effect model using risk ratio (RR), relative risk reduction, risk difference (RD) and number needed to treat (NNT) for categorical data; and using mean, standard deviation and weighted mean difference (WMD) for continuous data. We analysed outcomes measured as count data, for example frequency of apnoea, bradycardia and episodes of pulse oximeter oxygen (SpO2) desaturation, by comparing rates of events and the rate ratio. We evaluated heterogeneity to help determine the suitability of pooling results.
Only one small cross-over trial met the criteria for inclusion in this review and therefore meta-analysis for any of the treatment outcomes was not performed. Symon 1994 reported a trend towards a higher respiratory rate at 10 to 30 minutes following push gavage feeding and no statistical difference in the time taken to give the feeds regardless of the method used.
There was one small cross-over study that was included in this review. There is insufficient evidence to recommend either method of gavage feeding. A randomised trial is needed to evaluate the benefits and harms of push versus gravity bolus tube feeding in preterm infants. Infants should be stratified by gestational age at birth (above and below 32 weeks) or birth weight (above and below 1500 grams) and respiratory support (ventilated versus non-ventilated) and the sample size should be of sufficient size to evaluate the primary outcomes outlined in this review (time to establish full tube feeds and feeding intolerance).
Plain language summary
Push versus gravity for intermittent bolus gavage tube feeding of premature and low birth weight infants
Infants born prematurely (before 37 weeks) may be unable to coordinate sucking, swallowing and breathing, and require gavage feeding. In gavage feeding, milk is delivered intermittently through a tube passed via the nose or mouth into the stomach. Intermittent bolus milk feeds may be administered using a syringe to gently push milk into the infant's stomach (push feed). Alternatively, milk can be poured into a syringe attached to the tube and allowed to drip in by gravity (gravity feed). There is insufficient evidence from randomised trials comparing push versus gravity intermittent gavage tube feeding in premature or low birth weight infants (less than 2500 grams) to inform practice.
Description of the condition
Many small, sick and premature infants are unable to coordinate suck, swallow and breathing, and require gavage feeding. In gavage feeding, milk feeds are delivered through a tube passed via the nose or mouth into the stomach. Nutritional management influences immediate survival as well as the subsequent growth and development of preterm and low birth weight infants (Lucas 1983).
The physiological effects of gavage feeding include a transient rise in postprandial oxygen consumption (Mukhtar 1982), a decrease in oxygenation (Wilkinson 1974; Yu 1976; Krauss 1978; Blondheim 1993; Hammerman 1995), increase in heart rate (Mukhtar 1982), decreased functional residual capacity (Heldt 1988) and decreased lung volumes (Pitcher-Wilmott 1979). It is postulated that these cardiorespiratory effects are related to the volume displacement caused by feeds introduced into the stomach (Yu 1976). Additionally, intermittent bolus gastric tube feeding (gavage feeding) may decrease cerebral perfusion (Nelle 1997).
Description of the intervention
Gavage milk feeds may be delivered by intermittent bolus, a continuous drip, or a slow infusion over one hour, with an interval of at least one hour until the next feeding (Evans 2001). Interminttent bolus feeds can be given by push or gravity. A 'push feed' is defined as an intermittent bolus milk feed that is administered using a syringe to push milk into the infant's stomach. Alternatively, a 'gravity feed' is defined as an intermittent bolus milk feed when milk is poured into a syringe (using the syringe as a funnel) that is attached to an intra-gastric feeding tube, and the milk is allowed to drip into the infant's stomach by gravity (Stronati 1982). The flow of the milk feed is controlled by altering the height of the syringe. Lowering the syringe slows the milk flow, raising the syringe makes the milk flow faster (Pagano 2010; Sankar 2008). A Cochrane review has shown that small babies weighing less than 1500 grams that were fed by intermittent bolus (push or gravity) compared with continuous infusion found no difference in time to achieve full oral feeds, growth, days to discharge or the incidence of necrotizing enterocolitis (Premji 2011).
How the intervention might work
The authors of two neonatal textbooks recommend that bolus gavage tube feeds should be given slowly, preferably by gravity (Wilkinson 1992; Townsend 1998). Sun 1998 strongly recommends against injecting fluid (milk) under pressure and similarly, Cox 1982 states 'feeding is never injected under pressure'. The reasoning behind these authors' recommendations is not provided. Chinn 1971, argued that a preterm infant does not always have the "advantage of esophageal peristalsis". In addition, "stomach and intestinal peristalsis depend on the presence of fluid in the stomach". She reported that feeding by gravity stimulates the physiological response of peristalsis as seen in the slight fluctuation of the formula level in the syringe. Chinn 1971 asserted that the 'forced-syringe (push method) of gavage feeding is more likely to result in milk being given too rapidly. This in turn, interferes with oesophageal peristalsis and results in oesophageal regurgitation (Barrie 1968; Leidig 1989; Haxhija 1998). In addition, fast gastric filling may also lead to greater vagal stimulation leading to bradycardia and apnoea, especially in low birth weight infants (Kindley 1980). Conversely, Ziemer 1978 argued that some pressure must be exerted as gravity is generally insufficient to administer an entire feed. In an experimental laboratory study Pagano 2010 reported that intermittent gravity feeding may result in an infant being fed a new feeding before the infants has digested the previous feed if the feed is delivered slowly. Nurses who declare a preference for either method of intermittent gavage feeding claim that with their preferred method they can more easily control the speed at which milk is delivered to the infant.
Why it is important to do this review
Since the 1960s there has been disparate recommendations and inconsistency in practice regarding which method of intermittent bolus gavage tube feeding (push or gravity) is the most efficacious in relation to the establishment of full gavage feeding without adverse effects. Choice appears to be primarily based on personal preference, unit policy, or tradition. In a survey of neonatal units in the United Kingdom, the method of bolus feeding varied with some units declaring a preference for push feeds while others preferred the gravity method of feeding (Symon 1994). We noted significant variation in gavage feeding while examining individual unit protocols published online. Those units that This variation in practice is also demonstrated by a number of researchers who have used either the push (Barrie 1968; Heldt 1988; Leidig 1989; Grant 1991; Nelle 1997) or gravity method (Kindley 1980; Parker 1981; Greer 1984; Blondheim 1993; Moscha 1995; Pagano 2010) for bolus gavage feeding in their clinical studies.
Because intermittent bolus gavage feeding is a common practice in neonatal intensive care and special care baby nurseries, it is important to determine the clinical risks and benefits of each method of feeding to enable clinicians to make informed decisions regarding the most appropriate feeding method for an individual infant. Therefore, a systematic review of trials that compares the two methods of milk feeding was performed.
To determine whether the use of push feeding compared with gravity feeding results in a more rapid establishment of full gavage feeds without increasing adverse events in preterm or low birth weight, or both, infants who require intermittent bolus tube feeding.
- infants of very low birth weight (< 1500 grams);
- infants receiving partial or all feeds via gavage tube;
- infants receiving respiratory support during the tube feed;
- positioning of infants during gavage feeding.
Criteria for considering studies for this review
Types of studies
We considered all published and unpublished randomised controlled trials (RCTs) or quasi-randomised trials eligible for inclusion in this review. Studies reported only by abstracts were eligible. We planned to accept cluster and cross-over randomised trials.
Types of participants
Preterm infants (less than 37 weeks gestation) or low birth weight infants (less than 2500 grams) who require partial or complete gavage tube feeding and have no congenital anomalies that might interfere with establishing enteral feeds.
Types of interventions
Push versus gravity gavage feeding.
Types of outcome measures
- Time taken to establish full gavage feeding (days).
- Feeding intolerance (number of episodes per day): defined as significant abdominal distension or discolouration, signs of perforation, obvious blood in stool; gastric residuals 25% to 50% of interval volume for two to three feedings; bilious gastric residual or emesis; significant apnoea or bradycardia; significant cardiopulmonary instability (Kuzma-O'Reilly 2003).
All to be assessed during gavage feeding:
- apnoea during gavage feeding (frequency of episodes), defined as a cessation of breathing for more than 20 seconds or a shorter pause associated with bradycardia or cyanosis (AAP 2003);
- apnoea (frequency of episodes within 30 minutes following gavage feeding), defined as a cessation of breathing for more than 20 seconds or a shorter pause associated with bradycardia or cyanosis (AAP 2003);
- bradycardia during gavage feeding (frequency of episodes), defined as a fall in heart rate of more than 30% below the baseline or less than 100 beats per minute for 10 seconds;
- bradycardia (frequency of episodes within 30 minutes following gavage feeding), defined as a fall in heart rate > 30% below the baseline or less than 100 beats per minute for 10 seconds;
- spontaneous episodes of oxygen desaturation during gavage feeding (frequency of episodes), defined as a spontaneous fall in SpO2 85% for 10 seconds in duration;
- spontaneous episodes of oxygen desaturation (frequency of episodes within 30 minutes following gavage feeding), defined as a spontaneous fall in SpO2 85% for 10 seconds in duration;
- severe apnoea during gavage feeding (frequency of events), defined as a cessation of breathing and a fall in heart rate > 30% below the baseline or less than 100 beats per minute for 10 seconds and a concurrent fall in SpO2 85%;
- incidence of aspiration pneumonia (frequency of episodes);
- incidence of necrotising enterocolitis (Bell's Stage II or greater) (including suspected and confirmed) (Bell 1978);
- days to regain birth weight;
- length of hospital stay (days) from admission to discharge.
Search methods for identification of studies
We used the standard search strategy of the Cochrane Neonatal Review Group as outlined in The Cochrane Library.
We sought to locate randomised controlled or quasi-randomised trials from the electronic databases Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, Issue 5, 2012) MEDLINE (from 1966 to May 2012), EMBASE (from 1980 to May 2012), and CINAHL (from 1982 to May 2012) using the following subject headings (MeSH) and text words: [infant -newborn / OR infan*, or Neonat*, OR prematur* OR preterm OR low birth weight] AND [GORD, GERD, gastro-oesophageal reflux, gastroesophageal reflux, infantile reflux, gastric regurgitation, gastric emptying, enteral feeding, enteral nutrition, feeding behaviour AND infant feeding, gavage feeding, intermittent feeding, bolus feeding, tube feeding, gastric feeding]. We did not apply any language restriction. Two review authors (JD, JF) independently performed the electronic database searches.
Searching other resources
The search strategy also included communication with expert informants, searching bibliographies of reviews and trials for references to other trials, abstracts, conferences and symposia proceedings of the Perinatal Society of Australia and New Zealand and Pediatric Academic Societies (American Pediatric Society, Society for Pediatric Research, and European Society for Pediatric Research) from 1990 to 2011. If we identified any unpublished trial, we intended to contact the corresponding investigator for information. We intended to consider unpublished studies or studies only reported as abstracts as eligible for the review if methods and data could be confirmed by the author. We also intended to contact the corresponding authors of identified RCTs for additional information about their studies, if further data were required.
Data collection and analysis
We used the standard methods of the Cochrane Neonatal Review Group (CNRG).
Selection of studies
The review authors (JD, RS, JF) independently assessed all the potential studies identified as a result of the search strategy for inclusion. We resolved any disagreement through discussion or, if required, we planned to consult a Cochrane review arbiter.
Data extraction and management
Two review authors (JD, RS) separately extracted data. We resolved any differences by discussion with the third review author (NB). We requested additional information from the authors of trials to clarify methodology and seek further data, if necessary.
Assessment of risk of bias in included studies
We used the standardised review methods of the Cochrane Neonatal Review Group to assess the methodological quality of included studies. The review authors independently assessed study quality and risk of bias using the following criteria, documented in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
- Sequence generation: was the allocation sequence adequately generated?
- Allocation concealment: was allocation adequately concealed?
- Blinding of participants, personnel, and outcome assessors for each main outcome or class of outcomes: was knowledge of the allocated intervention adequately prevented during the study?
- Incomplete outcome data for each main outcome or class of outcomes: were incomplete data adequately addressed?
- Selective outcome reporting: are reports of the study free of suggestion of selective outcome reporting?
- Other sources of bias: was the study apparently free of other problems that could put it at a high risk of bias? We planned to give particular attention to completeness of follow-up of all randomised infants and to the length of follow-up in the studies to identify whether any benefits claimed were robust.
We planned to request additional information and clarification of published data from the authors of individual trials. We also planned to assess each trial for risk of bias based on the criteria listed above and mark the trials as:
- 'low' risk of bias;
- 'unclear' risk of bias;
- 'high' risk of bias.
We resolved any discrepancies by mutual discussion and consensus.
Measures of treatment effect
Dichotomous variables were to be analysed using risk ratio and risk difference with 95% confidence intervals. We planned to report the number needed to treat (NNTB) and the number needed to harm (NNTH) if statistically significant results were found. Continuous variables were to be analysed using weighted mean differences and 95% confidence intervals.
Unit of analysis issues
The unit of analysis is the participating infant in individually randomised trials and the neonatal unit (or subunit) for cluster randomised trials; for cross-over trials see below.
Cluster randomised trials
Where the trials involved clustered randomisation, we anticipated that study investigators would have presented their results after appropriately controlling for clustering effects (robust standard errors or hierarchical linear models). If it was unclear whether a cluster randomised trial had used appropriate controls for clustering, the study investigators would have been contacted for further information. Where appropriate controls were not used, individual participant data would have been requested and re-analysed using multi-level models which control for clustering. Following this, effect sizes and standard errors would have been meta-analysed in RevMan using the generic inverse method (Higgins 2011). If appropriate controls were not used and individual participant data were not available, statistical guidance would have been sought from the Cochrane Method Group and external experts as to which method to apply to the published results in attempting to control for clustering. If there was insufficient information to control for clustering, outcome data would have been entered into RevMan using individuals as the unit of analysis, and then sensitivity analysis would have been used to assess the potential bias from inadequately controlled clustered trials (Donner 2001).
We planned to meta-analyse cross-over trials as recommended by Elbourne 2002. Ideally we would have used first period data from cross-over trials and combined these data with data from parallel studies. Elbourne 2002 advises "the results of two or more cross-over trials might be combined, but with this pooled result kept separate from the data from parallel group trials". However, we were unable to obtain such first period data for all the included cross-over studies. In addition, no parallel trials were identified and, therefore, only cross-over design trials were analysed. Elbourne 2002 recommends that use of the cross-over design should be restricted to situations where there is unlikely to be a carry-over of treatment effect across periods. Thus, we planned to assess the possible carryover effect of change in the intervention, that is push and gravity gavage feeding, across the study periods in each of the included trials (see Risk of bias in included studies).
Dealing with missing data
We planned to contact the authors of all published studies if clarifications were required, or to provide additional information. In the case of missing data the number of participants with the missing data were to be described in the results section and in the table 'Characteristics of Included Studies'. The results were to be presented for the available participants. We planned to discuss the implications of the missing data in the discussion of the review.
Assessment of heterogeneity
We planned to use RevMan 5 (RevMan 2011) to assess the heterogeneity of treatment effects between trials. We planned to use the two formal statistics described below.
- The Chi2 test for statistical homogeneity (P < 0.1). Since this test has low power when the number of studies included in the meta-analysis is small, we set the probability at the 10% level of significance (Higgins 2011).
- The I2 statistic, to ensure that pooling of data is valid. The impact of statistical heterogeneity will be quantified using this statistic, which describes the percentage of total variation across studies due to heterogeneity rather than sampling error, in RevMan 5 (RevMan 2011). We planned to grade the degree of heterogeneity as: 0% to 30%, might be important; 31% to 50%, moderate heterogeneity; 51% to 75%, substantial heterogeneity; 76% to 100%, considerable heterogeneity.
Assessment of reporting biases
We planned to assess reporting and publication bias by examining the degree of asymmetry of a funnel plot in RevMan 5 (RevMan 2011).
We planned to use the fixed-effect model in RevMan 5 (RevMan 2011) for meta-analysis.
Subgroup analysis and investigation of heterogeneity
- Infants of very low birth weight (less than 1500 grams).
- Infants receiving partial gavage feeding (more than one gavage tube per 24 hours) or all feeds via gavage tube.
- Infants receiving respiratory support (mechanical ventilation or continuous positive airway pressure) during the gavage feed.
- Positioning (supine, prone, lateral, head elevated).
Description of studies
See Characteristics of Included Studies table.
Results of the search
Our search strategy identified one small cross-over study Symon 1994 (N = 31).
Symon 1994 studied infants that had a birth gestational age ranging from 24 to 32 weeks (mean 31.3 weeks). They had a corrected gestational age ranging from 26 to 37 weeks at the time of feeding. They all had a birthweight of under 1750 grams and were under that weight when given the gravity/plunge feed. The average weight of babies when feeds were given by plunge was 1240 grams and 1216 grams when feeds were given by gravity. Days of age for plunge feeds ranged from 1 to 73 (mean 18.6 days) and for gravity feeds from 1 to 53 (mean 21.55).
The included study (Symon 1994) is a cross-over trial. Thirty one infants received 60 nasogastric feeds (30 plunge and 30 gravity).
No baby received more than three study feeds, and no more than one study feed on any given day. All feeds were given while the infant was in a sleeping or resting state. A size 5 Vygon XRO feeding tube without an extension was used for all of the feeds.
No studies were excluded.
Details of the methodological quality of the included trial (Symon 1994) are given in the table ' Characteristics of Included Studies'. The method of randomisation in the Symon 1994 trial was not identified. Correspondence with the primary author did not enable us to answer this question. There was no blinding of the intervention. Outcome measures were not blinded. All feeds were given by one researcher.
In the one included cross-over study identified in this review (Symon 1994), infants were exposed to more than one feeding method (push and gravity) and this may have confounded the outcome measures. The primary author of Symon 1994 was not able to provide any information regarding the length of time between each treatment and thus possible carryover effect from one treatment to another.
Effects of interventions
Push versus gravity bolus tube feeding (Comparison 1)
One small cross-over study was identified for this comparison (Symon 1994).
The Symon 1994 study did not measure either of the specified primary outcomes: time taken to establish full tube feeding (days); or feeding intolerance.
There were no data available for any of the specified secondary outcomes, however, Symon 1994 compared gravity versus push gavage feeding for the following outcomes.
Effect of feeding method on respiratory rate (Outcomes 1.1 to 1.3)
Symon 1994: There was no significant difference in respiratory rate at the completion of feeds (mean difference (MD) 0.58, 95% CI -5.97 to 7.13 (Outcome 1.1); at 10 minutes (MD 2.43, 95% CI -4.35 to 9.21 (Outcome 1.2); or at 10 to 30 minutes after the feed was completed (MD 3.10, 05% CI -3.43 to 9.63) (Outcome 1.3).
Symon 1994 reported the statistical significance of the effect of feeding method on respiratory rate within each of the groups rather than reporting the statistical difference between the two groups at each time period as described above.
Effect of feeding method on heart rate (Outcomes 1.4 to 1.6)
Symon 1994|: There was no significant difference in heart rate at the completion of feeds (MD -2.60, 95% CI -9.71 to 4.51) (Outcome 1.4); at 10 minutes (MD -2.40, 95% CI -9.59 to 4.79) (Outcome 1.5); or at 10 to 30 minutes after the feed was completed (MD -2.40, 95% CI -9.16 to 4.36) (Outcome 1.6).
Symon 1994 reported the statistical significance of the effect of feeding method on heart rate within each of the groups rather than reporting the statistical significance between the two groups at each time period as described above.
Symon 1994 reported bradycardia (not defined) during and immediately after gavage feeding. There were 4 episodes of bradycardia in each group.
Symon 1994 reported no episodes of apnoea (defined apnoea as a complete cessation of breathing for more than 15 seconds) during or after gavage feeding.
There were no studies that examined time taken to establish full gavage feeding (days), feeding intolerance, apnoea during gavage feeding (defined as a cessation of breathing for more than 20 seconds or a shorter pause associated with bradycardia or cyanosis), apnoea (frequency of episodes within 30 minutes following gavage feeding), bradycardia duringgavage feeding, bradycardia following gavage feeding, spontaneous episodes of oxygen desaturation during gavage feeding, spontaneous episodes of oxygen desaturation following gavage feeding, severe apnoea during gavage feeding, incidence of aspiration pneumonia, incidence of necrotising enterocolitis, days to regain birth weight or length of hospital stay from admission to discharge.
There were no data available for subgroup analyses of:
- Infants of very low birth weight (less than 1500 grams)
- Infants receiving partial gavage feeding (more than one gavage tube per 24 hours) or all feeds via gavage tube
- Infants receiving respiratory support (mechanical ventilation or continuous positive airway pressure) during the gavage feed
- Positioning (supine, prone, lateral, head elevated)
Gavage feeding is commonly used when infants are unable to suck and swallow feeds due to prematurity, low birthweight, or other conditions. Gavage feeds can be given intermittently using push or gravity methods. Intermittent bolus feedings are considered to be more physiologic than continuous feedings because they simulate a more normal feeding pattern and allow for cyclical surges in gut hormones (Aynsley-Green 1990). Continuous gavage feedings are more commonly considered in very premature infants with severe respiratory distress, or in infants who have shown a previous intolerance of intermittent feedings as evidenced by severe reflux and/or persistent pre-feed gastric residuals. However a Cochrane review has shown that small babies weighing less than 1500 grams that were fed by intermittent bolus (push or gravity) compared with continuous infusion found no difference in the time to reach full feeds (Premji 2011). There was no subgroup analysis of the type of intermittent feeding, push versus gravity, in included studies.
One small quasi-randomised controlled trial comparing push and gravity intermittent gavage feeding (Symon 1994) found a trend towards a higher respiratory rate at 10 to 30 minutes following gavage feeds administered using the push method. Conversely Symon 1994 reported a significantly higher heart rate following gravity gavage feeding. It is important to note that Symon reported the statistical significance of the effect of feeding method on heart rate within each of the groups. He did not compare the effect of feeding method on respiratory and heart rate betweengroups.
There are confounding factors that might effect the speed of a feed, these include the diameter of the feeding tube, the type of materials used in the manufacture of the tube, and the type of feed administered. When using the gravity method the height of the syringe above the baby will effect the speed of the feed.
We found little evidence to support the optimal speed for gavage feeding. Small sick infants may be prone to respiratory instability during intermittent gavage feeding (Blondheim 1993). Blondheim showed that gravity feeds given over 15 to 20 minutes by gravity, when compared with continuous tube feeds, reduced adverse effects after feeding. They did not measure the effect of push feeds. In the Symon study, feeds were not timed to last a specific number of minutes and seconds but were given in what the single observer felt to be a 'safe manner', however, there was little difference in the time taken to give the feeds regardless of the method used (1.38 ml/kg/min push feeds and 1.39 ml/kg/min gravity feeds). The University of Manitoba guidelines recommend that the infusion rate of a bolus feed should be no faster than 2 ml per minute for smaller feeding volumes. When a larger volume of bolus feed is given this should take approximately 15 minutes to complete. However, a small and large volume is not defined. Further advice when administering a feed via gravity is to position the syringe so that the feed infuses slowly (University of Manitoba 2007). If the push method is given then it should be pushed slowly. Sankar recommends that feeding should take about 10 to 15 minutes to complete (Sankar 2008). Whilst these recommendations seem sensible they are not evidence based.
The internal diameter, or bore, of the feeding tube will have an effect on the speed of feed and may influence the choice of whether a bolus feed is given by push or gravity. Ziemer favours push feeding when a small diameter feeding tube is used. She states that gravity is generally insufficient to administer an entire feeding and some pressure must be exerted because of the small lumen of the feeding tube (Ziemer 1978). In a laboratory study Pagano showed that using a 6 fg tube without changing any other conditions (height above the infant and type of feed) resulted in a faster feeding time than using a 5 fg tube (Pagano 2010).
The height of the feeding tube above the infant is an important factor in determining the speed of a feed. Increasing the height will increase the force of gravity, as the height of feeding is increased the feed will be given at a faster rate (Pagano 2010). There is little evidence regarding the optimum height of the tube when giving a bolus feed by gravity. Chin recommends that the height should not exceed eight inches above the infant's head (Chinn 1971). Symon controlled for tube size and height above the tube above infants in his cross-over study, all study feeds were given with a standard length size 5 fg feeding tube (Symon 1994).
The type of feed administered can have an effect on the speed of the feed. Pagano compared flow rates for artificial formula against human breast milk and fortified human breast milk. She showed that human breast milk had the fastest flow rate. Milks with increased calorie contents were the slowest. The type of milk administered in the Symon study is not described (Symon 1994). Polyurethane feeding tubes compared to silicone tubes result in feeds given at a faster rate (Pagano 2010).
Despite the paucity of evidence, clinical practice guidelines give clinicians recommendations regarding administering a bolus feeding only by gravity (Sankar 2008 ) or via push or gravity (University of Manitoba 2007). This systematic review has not established which method of intermittent gavage feeding has better outcomes. Future studies should enrol preterm and low birth weight infants who require intermittent gavage feeding. Important outcomes would include those specified in our criteria for considering studies for this review.
Quality of the evidence
One small cross-over trial.
Implications for practice
There is insufficient evidence to determine whether the use of push compared with gravity gavage feeding results in a more rapid establishment of full gavage feeds without increasing adverse events in preterm, low birth weight, or both, infants who require intermittent bolus gavage feeding.
Implications for research
Intermittent bolus gavage feeding is a common practice in neonatal intensive care and special care baby nurseries. A randomised controlled trial is needed to evaluate the benefits and harms of push versus gravity bolus gavage feeding in preterm infants. Infants should be stratified by gestational age at birth (above and below 32 weeks) or birth weight (above and below 1500 g). Subgroup analysis should include size of enteral feeding tube, position of the infant and speed at which the feed is delivered.
The sample should be of sufficient size to be able to evaluate the primary outcomes outlined in this review (time to establish full tube feeds and feeding intolerance).
We would like to thank the Australasian Satellite of the Cochrane Neonatal Review Group.
Contributions of authors
JD and RS wrote the review. NB and JPF provided editorial assistance.
Declarations of interest
- None noted.
Differences between protocol and review
- None noted.
Potential conflict of interest
- None noted.
Characteristics of studies
Characteristics of Included Studies
Randomised cross-over study design
60 nasogastric feeds (20 plunge, 30 by gravity) were given to 31 babies in the intensive care section of the study unit. No baby received more than three study feeds, and no more than one study feed on any given day. All feeds were given by one researcher in order to obviate inter-observer error, and all were given while the baby was in a sleeping or resting state
31 infants 24 to 32 weeks gestation (mean 28.5); at time of feed, corrected gestational age ranged from 26 to 37 weeks (mean 31.3)
Days of age for plunge feeds ranged from 1 to 73 days (mean 18.6), and for gravity feeds from 1 to 53 days (mean 21.6)
All participants had a birth weight of < 1750 grams. The average weight of babies when feeds were given by plunge was 1240 grams, and 1216 grams when feeds were given by gravity
Push versus gravity gavage feeding
Risk of bias table
|Bias||Authors' judgement||Support for judgement|
|Random sequence generation (selection bias)||Unclear risk||
Insufficient information provided. Authors state: 'Each feed was allocated randomly to either the gravity or plunge method'
|Allocation concealment (selection bias)||High risk||
No information provided
|Blinding (performance bias and detection bias)||Unclear risk||
Insufficient information provided
|Blinding of participants and personnel (performance bias)||High risk||
|Blinding of outcome assessment (detection bias)||Unclear risk||
No information provided
|Incomplete outcome data (attrition bias)||Low risk|
|Selective reporting (reporting bias)||Unclear risk||
We were unable to obtain the study's protocol
|Other bias||Unclear risk||
No information provided (such as period between treatment changeover) regarding the possible carryover effect from one treatment to another
References to studies
Published data only (unpublished sought but not used) [Other: ]
Symon A, Cunningham S. Nasogastric feeding methods in neonates. Nursing Times 1994;90(35):56-60.
References to excluded studies
- None noted.
Studies awaiting classification
- None noted.
- None noted.
Committee on Fetus and Newborn. American Academy of Pediatrics. Apnea, sudden infant death syndrome, and home monitoring. Pediatrics 2003;111(4 Pt 1):914-7.
Aynsley-Green A, Lucas A, Lawson GR, Bloom SR. Gut hormones and regulatory peptides in relation to enteral feeding, gastroenteritis, and necrotizing enterocolitis in infancy. Journal of Pediatrics 1990;117(1 Pt 2):S24-32.
Barrie H. Effect of feeding on gastric and oesophageal pressures in the newborn. The Lancet 1968;2(7579):1158-60.
Bell MJ, Ternberg JL, Feigin RD, Keating JP, Marshall R, Barton L, et al. Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Annals of Surgery 1978;187(1):1-7.
Blondheim O, Abbasi S, Fox WW, Bhutani VK. Effect of enteral gavage feeding rate on pulmonary functions of very low birth weight infants. Journal of Pediatrics 1993;122(5 Pt 1):751-5.
Chinn PL. Infant gavage feeding. American Journal of Nursing 1971;71(10):1964-7.
Cox MA, Thrift MC. Nutrition. In: Cloherty JP, Stark AR, editor(s). Manual of Neonatal Care. 6th edition. Boston: Little, Brown & Co, 1982:315-48.
Donner A, Piaggio G, Villar J. Statistical methods for the meta-analysis of cluster randomization trials. Statistical Methods in Medical Research 2001;10(5):325-38.
Elbourne DR, Altman DG, Higgins JP, Curtin F, Worthington HV, Vail A. Meta-analyses involving cross-over trials: methodological issues. International Journal of Epidemiology 2002;31(1):140-9.
Evans RA, Thureen PJ. Early feeding strategies in preterm and critically ill neonates. Neonatal Network 2001;20(7):7-18.
Grant J, Denne SC. Effect of intermittent versus continuous enteral feeding on energy expenditure in premature infants. Journal of Pediatrics 1991;118(6):928-32.
Greer FR, McCormick A, Loker J. Changes in fat concentration of human milk during delivery by intermittent bolus and continuous mechanical pump infusion. The Journal of Pediatrics 1984;105(5):745-9.
Hammerman C, Kaplan M. Oxygen saturation during and after feeding in healthy term infants. Biology of the Neonate 1995;67(2):94-9.
Haxhija EQ, Rosegger H. Effects of bolus tube feeding on cerebral blood flow velocity in neonates. Archives of Disease in Childhood. Fetal and Neonatal Edition 1998;78(1):F78-9.
Heldt GP. The effect of gavage feeding on the mechanics of the lung, chest wall, and diaphragm of preterm infants. Pediatric Research 1988;24(1):55-8.
Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org.
Kindley AD, Harris F. Heart rate changes during gavage feeding of neonates. Early Human Development 1980;4(4):387-92.
Krauss AN, Brown J, Waldman S, Gottlieb G, Auld PA. Pulmonary function following feeding in low-birth-weight infants. American Journal of Diseases of Children 1978;132(2):139-42.
Kuzma-O'Reilly B, Duenas ML, Greecher C, Kimberlin L, Mujsce D, Miller D, Walker DJ. Evaluation, development, and implementation of potentially better practices in neonatal intensive care nutrition. Pediatrics 2003;111(4 Pt 2):e461-70.
Leidig E. Doppler analysis of superior mesenteric artery blood flow in preterm infants. Archives of Diseases in Childhood 1989;64(4 Spec No):476-80.
Lucas A, Bloom SR, Aynsley-Green A. Metabolic and endocrine consequences of depriving preterm infants of enteral nutrition. Acta Paediatrica Scandinavica 1983;72(2):245-9.
Mosca NW. Holding premature infants during gavage feeding: effect on apnoea, bradycardia, oxygenation, gastric residual, gastrin and behavioral state. Doctoral Dissertation, Case Western Reserve University 1995.
Mukhtar A, Stothers JK. Cardiovascular effects of nasogastric tube feeding in the healthy preterm infant. Early Human Development 1982;6(1):25-30.
Nelle M, Hoecker C, Linderkamp O. Effects of bolus tube feeding on cerebral blood flow velocity in neonates. Archives of Disease in Childhood. Fetal and Neonatal Edition 1997;76(1):F54-6.
Pagano T, Stewart D. Flow rates of gavage feedings provided to very preterm infants. The Ohio State University, College of Nursing 2010. [Other: ]
Parker P, Stroop S, Greene H. A controlled comparison of continuous versus intermittent feeding in the treatment of infants with intestinal disease. Journal of Pediatrics 1981;99(3):360-4.
Pitcher-Wilmott R, Shutack JG, Fox WW. Decreased lung volume after nasogastric feeding of neonates recovering from respiratory disease. Journal of Pediatrics 1979;95(1):119-21.
Premji SS, Chessell L. Continuous nasogastric milk feeding versus intermittent bolus milk feeding for premature infants less than 1500 grams. Cochrane Database of Systematic Reviews 2011, Issue 11. Art. No.: CD001819. DOI: 10.1002/14651858.CD001819.pub2.
Review Manager (RevMan) [Computer program]. Version 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011.
Sankar MJ, Agarwal R, Mishra S, Deorari A, Paul V. Feeding of low birth weight infants. http://www.newbornwhocc.org/pdf/Feeding_of_Low_Birth_weight_Infants_050508.pdf 2008.
Stronati M, Lombardi G, Rondini G. A simple modification of the intermittent gavage feeding technique in low-birth-weight infants. Pediatrics 1982;70(3):507-8.
Sun Y, Awnetwant EL, Collier SB, Gallagher LM, Olsen IE. Stewart JE. Nutrition. In: Cloherty JP Stark AR, editor(s). Manual of Neonatal Care. Fourth edition. Philadelphia: Lippincott-Raven, 1998:120.
Townsend SF, Johnson CB, Hay Jr WW. Enteral Nutrition. In: Merenstein GB, Gardner SL, editor(s). Handbook of Neonatal Intensive Care. Fourth edition. St Louis: Mosby, 1998:295.
University of Manitoba 2007
Neonatal Patient Care Team. Neonatal Guidelines - Enteral feeding of infants. http://umanitoba.ca/faculties/medicine/units/pediatrics/sections/neonatology/media/410_Enteral_Feeding_2007.pdf.
Wilkinson A, Yu VY. Immediate effects of feeding on blood-gases and some cardiorespiratory functions in ill newborn infants. Lancet 1974;1(1):1083-5.
Wilkinson A, Calvert S. Procedures in neonatal intensive care. In: Roberton NRC, editor(s). Textbook of Neonatology. Edinburgh: Churchill Livingstone, 1992:1172.
Yu V. Cardio-respiratory response to feeding in newborn infants. Archives of Disease in Childhood 1976;51(4):305-9.
Ziemer M, Carroll JS. Infant gavage reconsidered. American Journal of Nursing 1978;78(9):1543-4.
Other published versions of this review
- None noted.
Classification pending references
- None noted.
Data and analyses
1 Comparison 1. Push versus gravity bolus tube feeding
For graphical representations of the data/results in this table, please use link under "Outcome or Subgroup".
|Outcome or Subgroup||Studies||Participants||Statistical Method||Effect Estimate|
|1.1 Respiratory rate at completion of feed||1||59||Mean Difference (IV, Fixed, 95% CI)||0.58 [-5.97, 7.13]|
|1.2 Respiratory rate 10 minutes after feed completed||1||59||Mean Difference (IV, Fixed, 95% CI)||2.43 [-4.35, 9.21]|
|1.3 Respiratory rate 10-30 minutes after feed completed||1||59||Mean Difference (IV, Fixed, 95% CI)||3.10 [-3.43, 9.63]|
|1.4 Heart rate at completion of feed||1||60||Mean Difference (IV, Fixed, 95% CI)||-2.60 [-9.71, 4.51]|
|1.5 Heart rate 10 minutes after feed completed||1||60||Mean Difference (IV, Fixed, 95% CI)||-2.40 [-9.59, 4.79]|
|1.6 Heart rate 10-30 minutes after feed completed||1||60||Mean Difference (IV, Fixed, 95% CI)||-2.40 [-9.16, 4.36]|
Sources of support
- The Royal Women's Hospital, Australia
- Children's Hospital at Westmead, Australia
- JAD, Australia
- NHMRC Post doctoral fellowship
- Eunice Kennedy Shriver National Institute of Child Health and Human Development National Institutes of Health, Department of Health and Human Services, USA
- Editorial support of the Cochrane Neonatal Review Group has been funded with Federal funds from the Eunice Kennedy Shriver National Institute of Child Health and Human
- Development National Institutes of Health, Department of Health and Human Services, USA, under Contract No. HHSN275201100016C.
This review is published as a Cochrane review in The Cochrane Library, Issue 11, 2012 (see http://www.thecochranelibrary.com for information). Cochrane reviews are regularly updated as new evidence emerges and in response to feedback. The Cochrane Library should be consulted for the most recent version of the review.