Web site designed and maintained by Chris Cairns  © SICS EBM Group 2004                                  

Evidence-based feeding

Citation:  Multi-centre, cluster-randomized clinical trial of algorithms for critical-care enteral and parenteral therapy (ACCEPT). Claudio M. Martin et al for the Southwestern Ontario Critical Care Research Network. Canadian Medical Association Journal   2004: 170; 197-204

Lead author's name and email: Claudio M. Martin, drcm@rogers.com

Three-part Clinical Question:

Patients: ICU patients.

Intervention: Algorithms for enteral and parenteral therapy.

Outcomes: (i) Adequate nutritional support, (ii) ICU stay, hospital stay, hospital mortality.

The Study: A prospective, cluster-randomised controlled trial in 14 hospitals without intention to treat.

The Study Patients: ICU patients ≥16 yrs and expected to stay at least 48hrs. Excluded if expected to be receiving sufficient nutrition orally to meet their daily energy requirements within 24hrs after ICU admission, were admitted for palliative care, were moribund and not expected to survive for more than 6 hours, or were suspected to be brain dead.
 

Control group (N = 227; analysed 223): Patients in ICUs in control hospitals (N=7). The ICUs in control hospitals received no new instructions on nutrition and continued with their standard nutritional practices.  Dietitians collected data for the study but no other changes were made.
 

Experimental group (N = 272; analysed 269): Patients in ICUs in intervention hospitals (N=7). The ICUs in intervention hospitals received training in implementing evidence based recommendations (formulated by the CCR-Net in 1996) including algorithms on enteral and parenteral nutrition; pocket cards with the algorithms printed on them were provided to nursing and medical staff and dietitians recorded nutritional support daily, giving immediate feedback as to how to optimize compliance with the recommendations. The guidelines suggested starting either enteral nutrition (EN) or parenteral nutrition (PN) in all patients within 24 hours, and adopting a strategy to maximise delivery of EN.  If EN failed to meet 80% of a patient’s requirements within 72 hours, PN would be started in addition. 

The Evidence:

Other Outcomes:

Comments:
The benefit in reduced mortality and length of hospital stay does not appear to be related simply to the energy intake as there was a non-significant increase in energy intake per patient day with the main effect being an increase in the number of days on which patients were fed.  The authors suggest that enteral nutrition may confer benefits independent of total energy and protein delivery.

There was a non-significant increase in TPN use in the intervention group.

EBM questions:

1) Do the methods allow accurate testing of the hypothesis?  Yes.  A cluster randomised trial enables a complex intervention to be studied without the risk of contamination between groups when one group is receiving a different package of care to the other.  It is also simpler for each individual centre to concentrate on one intervention only.  A run-in period analysing patients in both types of hospital showed that they were well balanced in terms of patient data and outcomes except there were more post-operative patients in the intervention hospitals.  After adjustment for admission source there was still a significant reduction in mortality.  The disadvantage of a cluster design is that the investigators collecting the data cannot be blinded but the authors state that the use of objective outcomes should minimize the effect of this. In this trial there was a risk of  a Hawthorne effect, as the dietitians spent more time in the ICUs in the intervention hospitals, thereby increasing the time spent discussing nutrition with ICU staff, which may have had an additional effect on nutritional support over and above introduction of the protocol.

2) Do the statistical tests correctly test the results to allow differentiation of statistically significant results?  Yes

3) Are conclusions valid in light of the results? The conclusion was that introducing guidelines improved nutritional support and improved outcome and this is a valid conclusion.

4) Did results get omitted, and why?  Yes. In the intervention group, 3 eligible patients were not recruited and 3 enrolled patients were not included in the analysis because of missing data.  In the control group 19 eligible patients were not recruited and 4 enrolled patients had missing data.

5) Did they suggest areas of further research?  No

6) Did they make any recommendations based on the results and were they appropriate? Yes.  They suggest that current evidence-based nutritional guidelines should be adopted with a suitable implementation strategy.

7) Is the study relevant to my clinical practice?  Yes; Scottish ICUs demonstrate varying nutritional practices and patients may benefit from a more unified approach to feeding.

8) What level of evidence does this study represent? 1⁺

9) What grade of recommendation can I make on this result alone? B

10) What grade of recommendation can I make when this study is considered along with other available evidence? There is little evidence available on the effect of introducing guidelines in ICU nutrition on a large scale.  Studies of introduction of limited protocols e.g. for NG feeding have shown an improvement in NG feeding rates.

11) Should I change my practice because of these results?   One should consider examining available guidelines on nutrition in ICU (CCR-Net, ESPEN, ICS guidelines on TPN) and consider introducing  these into practice if they are not already being used.

12) Should I audit my current practice because of these results? Yes: an audit of prescription and intake of nutrition in ICU patients may reveal that patients are not receiving nutrition at the earliest opportunity and may be falling significantly behind their goals which could be having a detrimental effect on their recovery.

Appraised by: Dr. M. McDougall, Consultant, Intensive Care. Queen Margaret Hospital, Fife, UK.
Email: marcia.mcdougall@faht.scot.nhs.uk

Kill or Update By: August 2009

Reviewed & edited by BC & CC.

Citation: EBM Critical Appraisals. Scottish Intensive Care Society EBM Group. 2006. McDougall M. Multi-centre, cluster-randomized clinical trial of algorithms for critical-care enteral and parenteral therapy (ACCEPT). Claudio M. Martin et al for the Southwestern Ontario Critical Care Research Network. Canadian Medical Association Journal   2004: 170; 197-204

Printer friendly view

©SICS EBM 2006