H L Cornwall et al, 2019. Coffee to go? The effect of coffee on resolution of ileus following abdominal surgery: a systematic review and meta-analysis of randomised controlled trial, Clinical Nutrition, published online.Print this page
BACKGROUND & AIMS:
Ileus following abdominal surgery is a common postoperative complication. It is a source of considerable morbidity to patients and prolongs hospital stay. There is limited evidence to support the use of coffee to promote resolution of post-operative ileus.
We performed a systematic review, a risk of bias assessment, and meta-analysis of randomised controlled trials that compared the effect of coffee consumption with control intervention on gastrointestinal motility after abdominal surgery. We searched PubMed, EMBASE, CINAHL, ISI Web of Science, clinicaltrials.gov, Cochrane Library, CENTRAL, WHO ICTRP, and Google Scholar, from inception to 24th February 2018.
Data from seven studies were extracted (606 patients). 31% were men. 69% were women. 342 underwent colorectal resection, 114 gynaecological resection and 150 elective caesarean section. Estimates from meta-analysis revealed that coffee consumption reduced time to defecation by 14.8 h (95% CI: -11.9, -17.7) after colorectal resection and 17.8 h (95% CI: -13.6, -22.0) after gynaecological resection, but had no significant effect after caesarean section. Coffee also reduced time to first bowel sound, and time to tolerance of solid food, but not time to first reported flatus. No measures of ileus were increased by coffee consumption. Complications and length of hospital stay were similar for coffee and control groups. Coffee was well-tolerated with no adverse effects. Cost was low.
Risk of bias was moderate or high across studies. Assessed with GRADE criteria, there is low to moderate quality evidence that coffee accelerates postoperative recovery of gastrointestinal function after colorectal and gynaecological surgery.
This information is intended for Healthcare professional audiences.
Please consider the environment before printing.