New England Journal of Medicine AUGUST 11, 2011

Early versus Late Parenteral Nutrition in Critically Ill Adults

Michaël P. Casaer, Greet Hermans, Alexander Wilmer, Djalila Mekhfi, et al.

Bottom Line

In critically ill adults, the initiation of parenteral nutrition to supplement enteral nutrition after one week in the ICU resulted in faster recovery and fewer complications compared to early initiation within 48 hours.

Key Findings

1. Late initiation of parenteral nutrition was associated with a higher likelihood of discharge alive from the ICU earlier than the early initiation group.

2. The late-initiation group experienced a significantly lower rate of new infections (22.8% vs. 26.2%; P=0.008).

3. Patients in the late-initiation group had a lower proportion of patients requiring more than 2 days of mechanical ventilation (36.3% vs. 40.2%; P=0.006).

4. The duration of renal replacement therapy was significantly shorter in the late-initiation group (median 7 days vs. 10 days; P=0.008).

5. There was no significant difference in mortality between the early and late groups (6.2% vs. 5.8%; P=0.62).

6. The mean duration of hospital stay was shorter in the late-initiation group (14 days vs. 16 days; P=0.004).

Study Design

Design

RCT

Open-Label

Sample

4,640

Patients

Duration

Hospital discharge

Median

Setting

Multicenter, Belgium

Population Adult critically ill patients at nutritional risk (NRS score ≥ 3) admitted to the ICU.

Intervention Parenteral nutrition initiated within 48 hours of ICU admission to supplement enteral nutrition to reach calculated caloric targets.

Comparator Late initiation of parenteral nutrition, withheld until day 8 of ICU stay if enteral nutrition was insufficient.

Outcome A composite of clinical outcomes defined as the duration of ICU stay and the incidence of new infections during the ICU stay.

Study Limitations

• The study population had a high proportion of post-cardiac surgery patients, potentially limiting generalizability to more complex, heterogenous ICU populations.

• The overall median ICU stay was short (3 days), which may reduce the impact of nutritional interventions.

• The study was conducted in a setting where intensive insulin therapy for tight glycemic control was standard, which might influence the metabolic effects of parenteral nutrition.

• Nutritional management after ICU discharge was not controlled, which could impact long-term recovery metrics.

Clinical Significance

The EPaNIC trial challenged the previous consensus that early caloric supplementation via parenteral nutrition was beneficial in critically ill patients, providing evidence that delaying supplemental parenteral nutrition until after the first week of ICU stay is safe and potentially superior for clinical recovery.

Historical Context

Prior to this study, European guidelines recommended early initiation of parenteral nutrition (within 48 hours) when enteral nutrition failed to meet caloric targets, while North American guidelines favored early enteral nutrition but allowed for hypocaloric nutrition during the first week; this trial provided landmark evidence supporting the latter, more conservative approach.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

What is the physiological role of autophagy in critical illness, and how does the early administration of parenteral nutrition theoretically interfere with this process?

Key Response

Autophagy is a cellular 'housekeeping' mechanism that breaks down damaged organelles and proteins to provide energy and maintain cellular health during stress. Early parenteral nutrition (PN) increases insulin levels and activates the mTOR pathway, which suppresses autophagy. This inhibition may lead to the accumulation of cellular 'trash' and exacerbate organ dysfunction, explaining why late PN (allowing natural fasting/autophagy) resulted in better outcomes in the EPaNIC trial.

Resident

A patient on ICU day 3 is unable to tolerate more than 10% of their caloric needs via enteral nutrition. Based on the EPaNIC trial results, what is the most appropriate management regarding supplemental nutrition?

Key Response

According to the EPaNIC trial, the most appropriate management is to wait until day 8 before initiating supplemental parenteral nutrition. The study demonstrated that withholding PN for the first week, despite inadequate enteral intake, led to a faster discharge from the ICU and a lower incidence of new infections compared to initiating PN within 48 hours.

Fellow

The EPaNIC trial included a broad population of critically ill patients. How should a fellow interpret the study's findings regarding patients who were severely malnourished at baseline (BMI < 18.5)?

Key Response

In the EPaNIC subgroup analysis, even malnourished patients (those with a high nutritional risk score) did not benefit from early PN and actually shared the same benefits from late PN as well-nourished patients. This challenges the common subspecialty practice of aggressive early feeding in malnourished individuals, though it remains a controversial point because the sample size of severely malnourished patients was relatively small.

Attending

The EPaNIC trial shifted the paradigm from 'aggressive feeding' to 'permissive underfeeding' in the acute phase of critical illness. How does this change our understanding of the catabolic response to sepsis?

Key Response

It suggests that the acute catabolic state is an adaptive, evolutionarily conserved survival mechanism rather than a pathological state requiring immediate correction. Providing exogenous fuel (PN) during this phase may overwhelm mitochondrial capacity or suppress essential stress responses. This teaching point emphasizes that 'feeding the sick' might actually be 'feeding the fire' during the hyperacute phase of critical illness.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

The EPaNIC trial utilized a 'time to discharge' metric as a primary outcome. What are the statistical limitations of using time-to-event outcomes in nutrition trials when death is a competing risk?

Key Response

When using 'time to discharge' as an outcome, mortality acts as a competing risk because a patient who dies will never be discharged. If the intervention increases mortality, the 'time to discharge' for survivors might look deceptively good. Researchers must use specific statistical models, such as Fine-Gray subdistribution hazard models or 'days alive and out of the hospital,' to ensure that the treatment effect on discharge is not confounded by its effect on survival.

Journal Editor

Given that the EPaNIC trial was an open-label study, how might the lack of blinding regarding the nutritional intervention introduce performance bias in the primary endpoint of 'readiness for ICU discharge'?

Key Response

Because clinicians were aware of which patients were receiving PN, their subjective assessment of whether a patient was 'stable enough' for discharge could be subconsciously influenced by their beliefs about nutrition. This is a significant threat to internal validity, as 'readiness for discharge' is more subjective than objective mortality, making the study vulnerable to the 'Hawthorne effect' or investigator bias.

Guideline Committee

How do the EPaNIC results reconcile with ESPEN and ASPEN/SCCM guidelines regarding the timing of supplemental PN in patients for whom enteral nutrition is insufficient?

Key Response

At the time, ESPEN guidelines suggested starting PN within 2 days, whereas ASPEN/SCCM suggested waiting 7 days for previously well-nourished patients. EPaNIC strongly supported the ASPEN approach of waiting, eventually leading to a convergence in international guidelines that emphasize caution with early PN. Current guidelines (e.g., ASPEN 2016/2021) now generally recommend that for patients at low or high nutritional risk, supplemental PN should be delayed until after day 7 if EN is not possible, aligning with the trial's evidence of reduced infection rates.

Clinical Landscape

Noteworthy Related Trials

2012

EDEN Trial

n = 1,000 · NEJM

Tested

Trophic versus full enteral feeding

Population

Patients with acute lung injury

Comparator

Full enteral feeding

Endpoint

60-day mortality

Key result: Trophic feeding did not result in better clinical outcomes compared with full enteral feeding in patients with acute lung injury.

2015

PermiT Trial

n = 894 · JAMA

Tested

Permissive underfeeding

Population

Critically ill patients

Comparator

Standard enteral feeding

Endpoint

90-day mortality

Key result: Permissive underfeeding was not associated with increased mortality compared with standard enteral feeding.

2018

NUTRIREA-2 Trial

n = 2,410 · Lancet

Tested

Early full enteral nutrition

Population

Critically ill patients requiring mechanical ventilation

Comparator

Early parenteral nutrition

Endpoint

Day-28 mortality

Key result: Early full enteral nutrition did not reduce mortality or the risk of secondary infections compared with early parenteral nutrition.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis