December 2012 e-journal club
We frequently skip our Journal Club in December to accommodate everyone’s busy schedule around the holidays, but we have been waiting for the full text of this study ever since the preliminary results were presented last year at Clinical Nutrition Week.
Heidegger CP, Berger MM, Graf S, et al. Optimisation of energy provision with supplemental parenteral nutrition in critically ill patients: a randomised controlled clinical trial. Lancet. 2012 Dec 3. [Epub ahead of print]
This was a randomized, controlled study conducted in the mixed medical-surgical ICUs of two tertiary care hospitals, designed to determine if supplementing enteral nutrition (EN) with parenteral nutrition (PN) to meet full energy expenditure from ICU days 4 to 8 would improve patient outcome. The study was unblinded, but allocation into groups was concealed with a computer generated sequence stratified by sex, and medical or surgical admission. The primary endpoint of the study was nosocomial infections that occurred day 9 through day 28.
The main secondary endpoints were the number of antibiotic days for nosocomial infection and antibiotic-free days. A number of other secondary endpoints were recorded, including duration of invasive and non-invasive mechanical ventilation, length of stay in the ICU and hospital until day 28, mortality in the ICU, general mortality, duration of renal replacement therapy, glucose control (blood glucose concentration and area under the curve), serum phosphorus, C-reactive protein, hepatic panel, and drug administration (insulin, steroids, and anti-fungal).
Initial calorie goals were 25 kcals/kg of ideal weight for women and 30 kcals/kg of ideal weight for men and “anamnestic bodyweight” weight (not defined, and unclear to us) was used for patients with a BMI ≤ 20. Calorie goals were adjusted on day 3 based on a single indirect calorimetry measurement. Protein needs were calculated based on 1.2 gm/kg of ideal body weight/day. Calories from fluids and medications (propofol) were included in the calculations, and supplemental PN was adjusted 2X/day depending on EN and other calories being provided.
Inclusion and Exclusion Criteria were:
Patients who had received < 60% of their energy target from EN at day 3 after admission to the ICU, were expected to stay for more than 5 days, expected to survive for more than 7 days, and had a functional gastrointestinal tract.
Patients who were already receiving PN, had persistent GI dysfunction or ileus, pregnant, refused to consent, or had been readmitted to the ICU after previous randomization into a study.
Out of 2556 patients assessed, the investigators randomized 305 patients, with 153 assigned to supplemental PN (SPN) and 152 received only EN. Ultimately 20 patients in the SPN group and 10 in the EN group discontinued the intervention due to protocol violation, consent withdrawal or therapeutic withdrawal, but analysis was completed on an intention to treat (everyone randomized) basis.
Mean energy delivery between day 4 and day 8 was 28 kcal/kg per day
in the SPN group and 20 kcal/kg per day in the EN group (103% of energy target in the SPN group vs 77% in the EN group; p<0.0001 (about 600 kcals/day difference).
Mean protein delivery between day 4 and day 8 was 1.2 g/kg ideal weight per day for the SPN group and 0.8 g/kg ideal weight per day for the EN group (100% vs 71%, p<0.0001).
The adjusted probability of nosocomial infection between days 9 and 28 was significantly lower in the SPN group than in the EN group (41 of 153 patients [27%] in the SPN group had a nosocomial infection during follow up compared with 58 of 152 patients [38%] in the EN group p=0.0338. There were also significantly less antibiotic days, and more antibiotic-free days in the group receiving SPN.
We did note that per the data in table 3 there did not appear to be a substantial difference in the actual number of infectious episodes between the two groups across the entire intervention period - 100 episodes of nosocomial infections in the SPN group, and 114 episodes in the EN group from day 4 to day 28.
“Individually optimised energy supplementation with SPN starting 4 days after ICU admission could reduce nosocomial infections and should be considered as a strategy to improve clinical outcome in patients in the ICU for whom EN is insufficient.”
This study appeared to be carefully designed and well conducted with proper concealed allocation into groups. Although it is generally not feasible to double-blind studies with PN/no PN, the investigator responsible for obtaining information on infectious complications was not aware of group allocation. The study was not designed to apply to all ICU patients, only those expected to require ICU care for > 5 days, and ended up as only 12% of ICU patients that were admitted, and ultimately the subjects required about 2 weeks in the ICU and one month in the hospital.
One major concern that we had about this study was the very low protein estimation used for ICU patients. We would normally provide at least 1.5 gm protein/kg of usual or pre-ICU actual (not ideal) weight for non-obese patients, and 2 gm/kg of ideal weight (or more); for those with a BMI >30 which would be 38% or 62% more protein (normal or obese, respectively) than the goals for this study. The amount of protein the EN only group actually received in this study was only 0.8 gm protein/kg of ideal weight (about 0.6 g/kg actual weight) which is only 40-45% of the protein we would provide to critically ill patients. In order to maintain nitrogen equilibrium, patients should receive increased protein when calorie supply is marginal. The EDEN study (see our February e-journal: http://www.healthsystem.virginia.edu/pub/dietitian/inpatient/dh/E-journal%20Club/february-2012.html ) had more than 6X as many patients as the present study, and did not find any significant difference in clinical outcomes with a 900 calorie/day variation in calorie supply between the groups – but the EDEN study did have greater protein goals, and allowed patients in the control group to receive protein supplements. Our clinical experience suggests that the use of scheduled protein modules allows us to provide much closer to estimated protein goals, even in those patients not meeting full calories.
Our group also discussed the relative value of probability of infection that has been adjusted for a few known variables over a specific period of time, compared to all of the infections that occurred during the study. The groups were randomized, which should have balanced all of the known and unknown factors that would affect outcome, so adjusting this data for only a few selected factors may skew the influence of many other unknown or unaccounted-for factors (nutrition status on admission?). Table 3 reported 100 infection events in the SPN group and 114 in the EN only group, so from a real world standpoint, it was not clear to us that the expense of SPN (and the labor cost required to achieve this level of calorie scrutiny/delivery) realized any improvement that mattered.
We noted that there was no difference in glucose control between the two groups, but we were surprised that although the SPN group received significantly more calories (average 600 more kcals/day) from PN containing 51% carbohydrates, the authors reported that there was no significant difference in insulin requirements between the 2 groups.
We also noted that calorie goals were estimated from a single indirect calorimetry measurement, and this in only of 2/3 of the patients. The fact that a single indirect calorimetry measurement is not any more accurate than most prediction equations for determining calorie expenditure over the first weeks in the ICU has been known for years1, and has been revalidated several times recently.2, 3 Also, considering that kcal adjustments after indirect calorimetry were less than 90 calories/day (mean of 66 kcals/day for both groups), it would be a stretch worthy of Pinocchio to suggest that this study is evidence that the use of indirect calorimetry improves patient outcomes in the ICU.
Overall, it is very noteworthy that in selected patients, very short-term, modest calorie PN did not appear to cause notable increase in infectious complications. It is certainly possible that patients with an extended ICU and hospital stay, who are really unable to have enteral feeding advanced to meet calorie and protein goals due to persistent GI tract dysfunction may benefit from supplemental PN. Our experience however, is that the vast majority of patients tolerate adequate EN when educated and dedicated staff are involved in their care.
Please note that there is no way to cover every aspect of this study that deserves mention and keep this readable. We strongly recommend that you read the full text of the study and accompanying editorial.4 This study will be discussed at length and likely be a source of controversy for some time – which is a healthy process in any field of scientific inquiry.
Our Take Home Message (s)
1) Extended periods of hypocaloric feeding with protein restriction should be avoided in critically ill patients.
2) Five days of supplemental PN did not appear to affect the overall number of infections during a patient’s hospital stay to day 28.
3) Five days of supplemental PN to patients with extended ICU and hospital admissions receiving 0.6 gm protein/kg was associated with a significantly decreased adjusted probability of infection from day 9 to day 28.
1. Vermeij CG, et al. Day-to-day variability of energy expenditure in critically ill surgical patients. Crit Care Med. 1989 Jul;17(7):623-6.
2. Frankenfield DC, et al. Longitudinal prediction of metabolic rate in critically ill patients. JPEN J Parenter Enteral Nutr. 2012 Nov;36(6):700-12.
3. Reid CL. Poor agreement between continuous measurements of energy expenditure and routinely used prediction equations in intensive care unit patients. Clin Nutr. 2007;26(5):649-657.
4. Vincent JL, Preiser JC. When should we add parenteral to enteral nutrition? Lancet. 2012 Dec 3.
Other News on the UVAHS GI Nutrition Website:(www.ginutrition.virginia.edu):
Now registering for the March 16-17, 2013 "Weekend Warrior
Mini-Traineeship" Nutrition Support Training Program in
Click Here for Details
Upcoming Webinars for 2013:
--Tuesday, January 15: Outpatient PEGs & the “Stat RD Consult--Carol Rees Parrish, MS, RD
Other Upcoming Topics include: The role of PT in the nutrition plan (exact title TBD), Nutrition Support in Cystic Fibrosis, Fiber in Nutrition Support, and Food Allergies
Check out What’s New:
--“Nutrition Support Blog”
Latest Practical Gastroenterology article:
--Thomas, JR, Nanda R, Shu LH. A FODMAP Diet Update: Craze or Credible? Practical Gastroenterology 2012;XXXVI(12):37.
Joe Krenitsky MS, RD
Carol Rees Parrish MS, RD
PS – Please feel free to forward on to friends and colleagues.