• DOI:10.1007/s00134-016-4248-0
• Corpus ID: 34072936

Comments on Marik and Hooper: Normocaloric versus hypocaloric feeding on the outcomes of ICU patients: a systematic review and meta-analysis

@article{Bitzani2016CommentsOM,
  title={Comments on Marik and Hooper: Normocaloric versus hypocaloric feeding on the outcomes of ICU patients: a systematic review and meta-analysis},
  author={Militsa Bitzani},
  journal={Intensive Care Medicine},
  year={2016},
  volume={42},
  pages={628-629}
}

• M. Bitzani
• Published 15 February 2016
• Medicine
• Intensive Care Medicine

Dear Editor, In a recent review and meta-analysis published online ahead of print on 10 November 2015 [1], Marik and Hooper addressed the impact of hypocaloric feeding on the outcome of critically ill patients. They reported no difference in the outcome parameters between patients receiving hypocaloric feeding and those fed normocalorically. Yet, both methodology and interpretation of the findings raise serious concern. Patients enrolled in the six studies included in the analysis were overweight or obese, with a BMI in the range 28.1–32.9 kg/m2. The caloric goal was set at 25–30 kcal/ kg/day, whereas current recommendations suggest 20–25 kcal/kg/day for normal weight patients and 11–14 kcal/kg/day for obese patients [2]. Therefore, as can be deduced from the results of each separate study, patients on permissive underfeeding were receiving 10–15 kcal/kg/day; according to the aforementioned recommendations, this might have been adequate, at least for the obese patients, while obese patients on the “normocaloric” diet were probably overfed. So, it is not clear whether patients on permissive underfeeding were really underfed, nor can overfeeding in patients on the “normocaloric” diet be ruled out. This could in part explain why no difference was observed in outcome between these two patient groups. Besides, protein administration was very low in all patient groups, with the lowest values (11 g/day) occurring in the trophic feeding patients, while the “normocaloric” group patients were receiving 0.6–0.8 g/kg/ day. Although protein intake in critically ill patients is still a challenging topic, the above amount is, according to available data, considered insufficient and is probably strongly related to the findings. Observational studies have shown decreased mortality when protein administration was consistent with current guidelines (1.2–2 g/ kg/day) [2, 3]. Besides, mortality was reduced in patients reaching both energy and protein goals, instead of the energy goal alone. In a recent study [4], early high protein intake (at least 1.2 g/kg/day) was related to a significantly lower mortality 19.1 vs 36.8 % when protein intake was less than 0.8 g/kg/day. Consequently, definite conclusions regarding the efficacy of the tested methods of nutritional support cannot be made if protein intake is inappropriate. The ICU patient population is widely heterogeneous differing in caloric and protein needs. The issue of optimal nutrition delivery in this patient group is still largely unclear. Emerging new hypotheses like trophic feeding or permissive underfeeding seem very attractive, but they have to be tested also in both normal weight and underweight critically ill patients, in well-designed studies. In the meantime we should be very cautious prior to extrapolating findings in overweight and obese patients to all ICU population. *Correspondence: [e-mail] Department of ICU, G. Papanikolaou Hospital, Exohi, Thessaloniki 57010, Greece