Early Versus Delayed Feeding After Placement of a Percutaneous Endoscopic Gastrostomy: A Meta-Analysis (original) (raw)
2008, Gastrointestinal Endoscopy
This review compared early versus delayed feeding after PEG placement. It showed that early feeding was associated with a higher incidence of significant residual gastric volume on the first day. The review contained several methodological limitations and its conclusions do not seem to be reliable. Searching MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL), DARE, HealthSTAR and CINHAL were searched up to 2006. Abstracts from conference proceedings of the Digestive Disease week and the American College of Gastroenterology National Meeting from 1993 to 2007 were handsearched. Search terms were reported. There were no apparent language restrictions. Study selection Randomised controlled trials (RCT) that compared early (≤4h) versus delayed or next-day feeding after PEG placement in adult patients (>18 years) were eligible. Death within 72 hours, complications and the incidence of significant gastric residual volumes during day one represented the outcomes for inclusion. Types of complications were not defined. Significant residual volumes was defined as it appeared in each article. Age of the participating patients ranged from 63 to 76 years. Time of early feeding ranged from less than one hour to four hours. Delayed feeding ranged from from 24 hours to next day. Significant volume definition varied from more than 60ml to 50% of the last volume administered. Length of follow-up was not described. Two authors independently selected the articles after an initial examination. It was unclear how many authors were involved in the initial selection. Assessment of study quality Studies were assessed for description and appropriateness of randomisation, blinding and dropouts using the Jadad scale. It appeared that two reviewers independently evaluated study quality. Data extraction Two authors independently extracted data in a predefined review form. Differences were resolved by consensus. The numbers of patients who experienced complications, died or had significant residual volumes were extracted to calculate odds ratios (OR). Methods of synthesis The pooled odds ratios and their corresponding 95% confidence intervals (CI) of each outcome were calculated using both fixed-effect and random-effects models. The results from the fixed-effect and random-effects models were consistent; only the random-effects model results were reported. Statistical heterogeneity was assessed using the I 2 test. An I 2 of 50% or more was considered significant. Authors stated that publication bias was assessed using a funnel plot for the outcome all complications.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.