Types of studies

Only randomized clinical trials (RCTs) were included. Quazi‐randomized trials were searched for and eligible for inclusion in the 2014 update.

Types of participants

Only studies where participants consisted entirely or mainly of premature infants (infants born before 37 weeks gestation) were included.

Types of interventions

Studies of varying water intake were included. Trials were excluded if the participants received water mainly or entirely as enteral feedings. Because we wished to examine the effects of water per se, rather than feedings, our review was limited to trials that included infants whose water intake was provided mainly or entirely by parenteral means, that is intravascular infusion.

Types of outcome measures

Electronic searches

RCTs identified in previous versions of this review were re‐examined and, in each case, retained. These included trials were identified from multiple sources, including a previous review by one of the authors (Bell 1992) and a MEDLINE search. Additional trials were sought that compared the outcomes of interest in groups of premature infants who were given different levels of water intake according to an experimental protocol. Such trials were sought in a list of trials provided by the Cochrane Neonatal Review Group, in the authors' personal files, and with a PubMed search using the following strategy:

1. infant, low birth weight (18,196 sources identified);

2. infant, premature (40,795 sources identified);

3. 1 or 2 (52,832 sources identified);

4. water intake (14,417 sources identified);

5. fluid intake (5979 sources identified);

6. 4 or 5 (18,998 sources identified);

7. 3 and 6 (176 sources identified).

In April 2010, we updated the search as follows: CENTRAL (The Cochrane Library), MEDLINE (search via PubMed), EMBASE and CINAHL were searched from 2007 to 2010. Search terms: fluid intake OR water intake. Limits: human, newborn infant and clinical trial. No language restrictions were applied.

In October 2014 we updated the search as follows: MEDLINE (search via PubMed), CINAHL, EMBASE and CENTRAL (The Cochrane Library) were searched from 2010 to 2014. Search terms: fluid intake OR water intake. Limits: human, newborn infant and clinical trial. No language restrictions were applied.

Searching other resources

Clinical trials registries were also searched for ongoing or recently completed trials (clinicaltrials.gov; controlled‐trials.com; and who.int/ictrp).

Selection of studies

All randomized and quasi‐randomized controlled trials fulfilling the selection criteria described in the previous section were included. Both review authors reviewed the results of the search and separately selected the studies for inclusion. The review authors resolved any disagreement by discussion.

Data extraction and management

The data were then entered into tables using RevMan software.

Assessment of risk of bias in included studies

The standard methods of the Cochrane Neonatal Review Group were employed. The methodological quality of each trial was reviewed independently by the two review authors. Each identified trial was assessed for methodological quality with respect to: a) masking of allocation, b) masking of intervention, c) completeness of follow‐up, and d) masking of outcome assessment. This information is included in the 'Characteristics of included studies' table.

For the updated review in 2010 and 2014, the 'Risk of bias' table was completed. The two review authors independently assessed the risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Any disagreement was resolved by discussion.

The risk of bias table addressed the following questions.

1. Sequence generation: was the allocation sequence adequately generated?

For each included study, we described the method used to generate the allocation sequence as: adequate (any truly random process for example a random number table, computer random number generator); inadequate (any nonrandom process for example odd or even date of birth, hospital or clinic record number); or unclear.

2. Allocation concealment: was allocation adequately concealed?

For each included study, we described the method used to conceal the allocation sequence as: adequate (for example telephone or central randomization, consecutively numbered sealed opaque envelopes); inadequate (open random allocation, unsealed or non‐opaque envelopes, alternation, date of birth); or unclear.

3. Blinding of participants, personnel and outcome assessors: was knowledge of the allocated intervention adequately prevented during the study? At study entry? At the time of outcome assessment?

For each included study, we described the methods used to blind study participants and personnel from knowledge of which intervention a participant received. We assessed the methods as: adequate, inadequate or unclear for participants; adequate, inadequate or unclear for study personnel; and adequate, inadequate or unclear for outcome assessors and specific outcomes assessed.

4. Incomplete outcome data: were incomplete outcome data adequately addressed?

For each included study, and for each outcome, we described the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total number of randomized participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. We assessed methods as: adequate (< 20% missing data), inadequate (≥ 20% missing data) or unclear.

5. Selective outcome reporting: were reports of the study free of suggestion of selective outcome reporting?

For each included study, we assessed the possibility of selective outcome reporting bias as: adequate (where it was clear that all of the study's pre‐specified outcomes and all expected outcomes of interest to the review were reported), inadequate (where not all the study's pre‐specified outcomes have been reported, one or more reported primary outcomes were not pre‐specified, outcomes of interest were reported incompletely and so could not be used, study failed to include results of a key outcome that would have been expected to have been reported) or unclear. 

6. Other sources of bias: was the study apparently free of other problems that could put it at a high risk of bias?

For each included study, we described any important concerns regarding other possible sources of bias (for example whether there was a potential source of bias related to the specific study design or whether the trial was stopped early due to some data‐dependent process). We assessed whether each study was free of other problems that could put it at risk of bias as: yes, no or unclear.

Measures of treatment effect

The adverse event rates were calculated for the restricted and liberal water intake groups for each dichotomous outcome; the relative risk and risk difference were computed for each outcome. In addition, the maximal weight loss results were recorded and the weighted mean difference was computed. The analyses, including calculation of relative risk, risk difference, weighted mean difference and tests of heterogeneity were accomplished using RevMan software and a fixed‐effect model.

Assessment of heterogeneity

We examined heterogeneity between trials by inspecting the forest plots and quantifying the impact of heterogeneity using the I² statistic.

Data synthesis

Meta‐analysis was carried out using the Review Manager software (RevMan 5, The Cochrane Collaboration). For estimates of typical relative risk and risk difference we used the Mantel‐Haenszel method. For measured quantities we used the inverse variance method. All meta‐analyses were done using the fixed effect model.

Subgroup analysis and investigation of heterogeneity

No subgroup analyses were performed.