Greetings,
Spring has finally arrived in Charlottesville and our April trainees were treated to Dogwoods, Redbuds and Azaleas in bloom. Although the early part of the week was filled with April showers, we had prime weather at weeks end. This month our trainees hailed from Highland, CA; Palmdale, CA and Kettering, OH. Our article for journal club was about hypocaloric feedings in critically ill patients.
April Citation: Petros S, Horbach M, Seidel F, Weidhase L. Hypocaloric vs Normocaloric Nutrition in Critically Ill Patients: A Prospective Randomized Pilot Trial. JPEN 2014 Apr 3.(epub ahead of print)
Spring has finally arrived in Charlottesville and our April trainees were treated to Dogwoods, Redbuds and Azaleas in bloom. Although the early part of the week was filled with April showers, we had prime weather at weeks end. This month our trainees hailed from Highland, CA; Palmdale, CA and Kettering, OH. Our article for journal club was about hypocaloric feedings in critically ill patients.
Summary of article:
This was a single-center, randomized, unblinded study (with concealed allocation into groups) of 100 critically ill, medical ICU patients comparing hypocaloric with full calorie nutrition support for the first 7 days of admission. Patients were randomized to receive either full nutrition or 50% of energy expenditure as measured by indirect calorimetry or estimated with Ireton-Jones equation. If the feedings were briefly interrupted, the EN feeding rate could be increased, or if a longer interruption was anticipated, supplemental PN was provided. The EN formula was 1 kcal/mL and provided 38 gm protein/1000 kcals. The PN formulation provided 1.2 kcal/mL and 40 gm protein/1200 kcals. All patients who received < 1500 mLs of EN, or all of those on PN, received supplemental vitamins and trace elements.
The primary outcome of the study was the rate of nosocomial infections while in the ICU. Secondary end points were insulin requirement and ICU, hospital and 28-day all-cause mortality rates. Gastrointestinal intolerance was defined as a gastric aspirate >300 mL/d and/or diarrhea (3 or more watery BM/day). Blood glucose was monitored every 3 hours, with a target blood glucose level of 6–8 mmol/L (108-144 mg/dl).
After the initial 7 day study period, all patients who still required nutrition support were fed a target of 25 kcals/kg ideal weight.
Inclusion and Exclusion Criteria:
Inclusion criteria:
Adult medical ICU patients with a presumed need for nutrition support for at least 3 days who provided informed consent.
Exclusion criteria:
Pre-existing malnutrition (defined as body mass index <18.5 kg/m2), age <18 or >80 years, pregnancy, active malignant disease, current
immunosuppressive therapy, readmission to the ICU, DNR status or refusal to participate by the patient or the guardian or consent provided too late for study inclusion.
Major Results:
There were 46 patients randomized to the hypocaloric group and 54 in the full feeding group. Sepsis, “acute cardiovascular dysfunction” and acute respiratory insufficiency were the most common admission diagnoses. After randomization, there were significantly more patients with a history of DM and COPD in the hypocaloric group, and significantly more patients with a history of neuropsychiatic illness in the full feeding group.
Feedings were started within 24 hours after admission.
Indirect calorimetry was completed in only 37 (37%) patients due to FiO2 >0.6 in some patients, and due to unrepairable equipment failure in the middle of the study. The full feeding group eventually received an average of 19.7 ± 5.7 kcal/kg/d (75.5% of energy expenditure), while the hypocaloric group received an average of 11.3 ± 3.1 kcal/kg/d (42.6% of energy expenditure) during the 7-day study (P = 0.0001). Protein intake was not described in the results section, but table 3 reveals that median protein intake of the full feeding group by day 3 was approx. 0.8gm/kg ideal weight, while the hypocaloric group received a median of approx. 0.4 gm protein/kg ideal weight.
The route of feeding was not significantly different between the groups with 54.3% of all patients receiving only EN, 35.8% receiving only PN and 9.9% of patients receiving a mix of EN and PN. However, the median percentage of calories provided by EN and PN were very different between the groups. The full feeding group received a median of 39.6% of calories from EN and 49.9% from PN (4.8% from other fluids). The hypocaloric group received a median of 69.9% of calories from EN and 8.0% from PN (11.9% from other fluids).
Nosocomial infections were reported in 6 patients (11.1%) of the full feeding group and in 12 patients (26.1%) of the hypocaloric group (P = 0.046).
There was no significant difference in daily blood glucose between the groups, but the full feeding group required more insulin than the hypocaloric group, and there was a significantly greater incidence of hypoglycemic events (16 events in 12 patients) in the hypocaloric group, compared to the full feeding group (9 events in 8 patients) (p= 0.036).
There was no significant difference in the number of patients with a gastric residual > 300 mLs between the groups. There was significantly more patients with diarrhea (3 or more watery BM/day) in the full feeding group compared to the hypocaloric group on day 4 (16 vs 5 patients, respectively;
P = 0.042) and on day 5 (23 vs 9 patients, respectively; P = 0.024) of the study.
The median duration of mechanical ventilation, time on vasopressors, hospital mortality and 28-day mortality were not significantly different between the groups.
Author’s Conclusions:
“…in this prospective randomized pilot study, hypocaloric feeding in the first 7 days was associated with more frequent nosocomial infections, but better glycemic control and less gastrointestinal intolerance, compared with a near normocaloric regimen in critically ill patients. However, there was no significant difference in clinical outcomes between the 2 feeding strategies.”
Evaluation:
The strengths of this study include the fact that patients were randomized and with concealed allocation into groups. However, the weaknesses of this study limit any take-home or practice conclusions.
The authors mention the lack of blinding as a limitation in the discussion section, and note that the lack of data to allow them to develop power calculations means that this was only a pilot study. It is not clear how the decision was made to enroll 100 patients, but if enrollment was stopped once statistical significance was reached, this introduces bias, especially in an unblinded study.
The small size of this study means that there could be an imbalance in the risk of infections between the groups due to chance. We can see that there were significantly more patients with pulmonary disease and diabetes in the hypocaloric group, and significantly more with neuropsychiatric disorders in the full feeding group, suggesting that the groups were not homogenous right from the start (breakdown of randomization). The differences in co-morbid conditions that may predispose to infectious complications could influence the difference in risk of infections between the groups outside of any nutrition interventions.
There are also major nutrition issues with this study. This was not just a study of hypocaloric feeding. This study compared full feeding with hypocaloric, severely protein restricted feeding. The decision to use a low-protein enteral formula without protein supplements determined from the start that patients in the hypocaloric group would receive inadequate protein. The median protein intake of the hypocaloric group was less than ¼ of the protein provided in hypocaloric feeding studies of obese patients. Even the 95th percentile of protein intake for the hypocaloric group was less than 1 gm/kg ideal weight.
This study was in progress before multicenter trials of tight glucose control in the ICU were published. In light of the results of NICE SUGAR and GLUCONTROL, it is not surprising that patients receiving very limited nutrition (hypocaloric group) had significantly more hypoglycemia while receiving intensive insulin therapy. It is a bit surprising that the authors conclusions in the text describe the hypocaloric group as having “better” glycemic control.
We also discussed several other points, including the description of intolerance as a gastric residual > 300 mL or more than 3 liquid stools in a day, or the fact that more than 45% of the patients received PN, which would be very unlike our medical ICU population.
Our Take Home Message (s)
1. Hypocaloric, protein restricted feedings may not be the ideal nutrition support regimen in critically ill adults, but the limited study size of this pilot study with breakdown of the randomization limits any take-home conclusions.
2. Be wary of publications or speakers who would reference this study as evidence supporting full feedings in the early stage of critical illness.
Other News on the UVAHS GI Nutrition Website: (www.ginutrition.virginia.edu):
Upcoming Webinars 2014:
–May 20: Peri-op Nutrition, CHO Loading Prior, and Earlier Post-op Feeding
–June 25: Managing Absorption in the Adult Short Bowel Patient
Check out What’s New:
–“Nutrition Support Blog”
–“ Resources for the Nutrition Support Clinician”
Latest Practical Gastroenterology article:
–Fisher C, Bethany Blalock B. Best Practice for Clearing Clogged Feeding Tubes. Practical Gastroenterology 2014;XXXVIII(3):16.
–Krenitsky J. Nutrition Update in Hepatic Failure. Practical Gastroenterology 2014;XXXVIII(4):47.
Joe Krenitsky MS, RD
Carol Rees Parrish MS, RD