The authors reply

    loading  Checking for direct PDF access through Ovid


We appreciate very much the careful reading of our article (1) by Drs. Fan and Zhu (2). We had hoped that they found our study more useful than useless. Our project describes current practices in ICUs around the world, sadly with their limitations in the provision of goal nutrition therapy. Although we agree with Drs. Fan and Zhu (2) that such levels of intake are far from optimal, similar levels of intake have been reported by other investigators (3–5).
We did not conduct subgroup analyses by diagnostic etiology of ICU admission because the limited sample size per diagnosis would constrain statistical power. We used the combined variables collected in the Sequential Organ Failure Assessment (SOFA) or Acute Physiology and Chronic Health Evaluation (APACHE) II score (temperature, heart rate, oxygenation, mechanical ventilation, bilirubin, hemodynamic stability, Glasgow Coma Score, and renal function) and the NUTrition Risk in the Critically Ill (NUTRIC) score (days in hospital prior to ICU admission, age, number of comorbidities, SOFA, and APACHE II scores) to reflect clinical acuity of patients. Further variables would not have been available consistently across all global ICU settings. We concur that there were undoubtedly differences in ICU practice patterns across sites; thus, our statistical analyses were designed to account for these differences by adjusting models for the specific ICU and for the geographic region.
Our project provides reassurance that the NUTRIC score predicts patients who will respond most positively to greater levels of energy intake, as originally observed (6). However, this current study extends that finding and the replication study by Rahman et al (7) in another database by providing similar findings of greater benefit of “both protein and energy intake” in high NUTRIC risk patients in a more diverse, globally representative sample but expanding the timeframe to 60-day mortality and adding the outcome of time to discharge alive. Furthermore, we confirmed that patients with low NUTRIC score did not fare worse with greater energy and protein intake within the levels observed in our study. This is why we conclude that the best approach may be an attempt to feed all patients optimally with an understanding that low-risk and shorter stay patients are less likely to benefit significantly from near-goal protein or energy intake.
    loading  Loading Related Articles