By Dr Au Shek Yin, ICU of QEH
Objectives: To evaluate the impact and safety of implementing an enteralfeeding(EN)protocol,based on updated evidence,in criticallyillpatients in a local mixed medical and surgical intensive care unit (ICU)
Design: This is a prospective observational 3-monthstudy. Patients’outcomes and feeding chartsin the protocol group in these 3 months were analysed, and the data were compared with that of the historical pre protocolgroup 3 months before the intervention
Setting: A 19-bed mixed medical and surgical Intensive Care Unitof Queen Elizabeth Hospital
Patients: From mid March 2012 to mid May 2012, 101 ICU patients who could not start oral feeding or who were on mechanical ventilation for more than 48 hours and who had nocontraindications to enteral feeding by their physical conditions orbeing fasted for procedures were recruited as the protocol group. Before this was a 3-month intervention period from Dec 2011 to Feb 2012.94 similar patients who were admitted from Sept 2011 to Nov 2011 were recruited asthe historicalpre protocolgroup.
Intervention: The interventions comprised 3 components: 1) A departmental nurse-led feeding protocol with higher gastric residual volume threshold, early regular use of prokinetics and early nutritional supplement in the form of total parenteral nutrition ( TPN ) or post pyloric feeding to achieve energy target was set; 2) An education programme on the feeding protocol to both medical and nursing colleagues from December 2011 to February 2012 was conducted; 3) An audit on the nursing compliance to the protocol was conducted.
Measurements and Results: The baseline characteristics were similar between the pre protocol (n=94 ) and the protocol (n=101) groups. The protocol group started EN earlier (47 hr + 25 hr vs 69hr + 87hr; p=0.021 ) than the pre protocol group and achieved the feeding target within a shorter time after ICU admission (61hr + 28 hr vs 100hr + 83 hr; p=0.001 ). The protocol group achieved 90% + 24% of the target feed rate on day 5 while the pre protocol group only achieved 57% + 42% of the target feed rate on day 5 ( p <0.001 ). The protocol group received a higher mean daily calorie ( 822kcal + 400kcal vs 441kcal + 329kcal; p <0.001 ); and a higher mean daily protein per body weight (0.57g/kg + 0.35g/kg vs 0.30g/kg + 0.26g/kg; p <0.001 ) when compared with the pre protocol group. In the protocol group, there was a trend towards shorter ventilator days (9 days + 9 days vs 10 days + 12 days; p =0.452 ) and a shorter ICU length of stay (11 days + 11 days vs 13 days + 13days; p = 0.395 ). There were no significant difference in the number of days on inotropes, hospital length of stay, number of patients that required care in the convalescence hospitals nor the incidence of hospital acquired infection ( HAI ) between the 2 groups. There was also a trend, in the protocol group, towards a lower overall mortality rate ( 26.7% vs 35.1%; p = 0.206 ). In the protocol group, there was a significantly lower percentage of patients with feeding withheld due to intolerance ( 7.92% vs 48.94%; p < 0.001 ) and a lower incidence of constipation ( 41.58% vs 56.38%; p = 0.039 ). There was a trend towards a higher mean blood glucose level at 8 am with higher mean insulin requirement at 8 am in the first few days after starting EN in the protocol group, but this did not reach statistical significance. There was no increase in the incidence of other EN related adverse effects. The overall nursing compliance to the protocol was 74%. Subgroup analyses in patients with different body mass index ( BMI ), medical or surgical patients with different nature of operations still showed similar outcomes in the protocol group except in the reduction in incidence of feeding withheld due to intolerance which could not reach statistical significance in the subgroup with BMI < 18.5 kg/m2 ( 14.29% vs 27.27%; p = 1.000 . Also, the number of days on inotropes in the protocol group were significantly longer in the surgical subgroup ( 2.8 days + 4.1 days vs 1.0 days + 1.5 days; p = 0.010 ), especially the surgical subgroup receiving emergency operation (2.4 days + 3.3 days vs 0.9 days + 1.3 days; p = 0.028 ) than the pre protocol group.
Conclusion: The implementation of an evidence based enteral feeding protocol, together with an education programme and audit in a local mixed medical and surgical ICU, could improve EN delivery. There was a trend towards shorter ventilator days and ICU length of stay in the protocol group. No increase in EN associated adverse effects was noted in patients fed on the feeding protocol.