The New England Journal of Medicine August 11, 2011

Early versus Late Parenteral Nutrition in Critically Ill Adults

Michael P. Casaer, Dieter Mesotten, Greet Hermans, Pieter J. Wouters, Miet Schetz, Geert Meyfroidt, et al.

Bottom Line

In critically ill adults, withholding supplemental parenteral nutrition until day 8 of the ICU stay, compared to early initiation within 48 hours, accelerates recovery, shortens ICU stay, and significantly reduces the rate of infections and complications.

Key Findings

1. Patients in the late-initiation group had a relative increase of 6.3% in the likelihood of being discharged alive earlier from the ICU (hazard ratio, 1.06; 95% CI, 1.00 to 1.13; P=0.04).

2. Late initiation of parenteral nutrition resulted in fewer ICU infections compared to early initiation (22.8% vs. 26.2%; P=0.008).

3. The late-initiation strategy led to a 9.7% relative reduction in the proportion of patients requiring more than 2 days of mechanical ventilation (P=0.006) and a median reduction of 3 days in the duration of renal-replacement therapy (P=0.008).

4. The incidence of cholestasis was significantly lower in the late-initiation group (P<0.001), contributing to a mean reduction in healthcare costs of €1,110 per patient (P=0.04).

5. There was no significant difference between the two groups in rates of ICU mortality, hospital mortality, or overall survival at 90 days.

Study Design

Design

Randomized Controlled Trial

Open-Label

Sample

4,640

Patients

Duration

90 days

Median

Setting

Multicenter, Belgium

Population Adult patients admitted to the ICU with a Nutritional Risk Screening (NRS) score of at least 3 who were not expected to meet their caloric targets via enteral nutrition alone.

Intervention Early initiation of supplemental parenteral nutrition within 48 hours of ICU admission to reach calculated caloric goals.

Comparator Late initiation of parenteral nutrition, where supplementation was withheld until day 8 of the ICU stay if enteral feeding remained insufficient (with early enteral feeding and intravenous 5% dextrose provided during the first week).

Outcome Time to discharge alive from the intensive care unit (ICU).

Study Limitations

• The trial was open-label, which inherently risks bias, though the investigators attempted to mitigate this by using strict, objective criteria to define readiness for ICU discharge.

• A tight glycemic control protocol (intensive insulin therapy targeting normoglycemia) was uniformly applied, which is no longer standard practice in many modern ICUs, potentially limiting the external validity of the findings.

• A large proportion of the enrolled cohort consisted of elective, uncomplicated cardiac surgery patients with inherently short ICU lengths of stay (median stay was 3-4 days).

• Because many patients had a short ICU stay, the majority of patients in the late-initiation group were discharged before day 8, meaning they never actually received the late parenteral nutrition intervention.

Clinical Significance

The EPaNIC trial demonstrated that 'permissive underfeeding'—tolerating a caloric deficit during the first week of critical illness—is clinically superior to aggressively achieving caloric goals via early supplemental parenteral nutrition. This fundamentally changed clinical practice by showing that early aggressive nutrition in the acute stress phase exacerbates complications like infections, prolonged mechanical ventilation, and liver dysfunction, favoring a more delayed, cautious approach to artificial nutrition.

Historical Context

Prior to the publication of the EPaNIC trial, there was significant international equipoise and conflicting guidelines regarding the optimal timing of parenteral nutrition in the ICU. European (ESPEN) guidelines strongly advocated for early parenteral nutrition (within 2 days) to prevent cumulative energy deficits, while American and Canadian (ASPEN/SCCM) guidelines recommended tolerating hypocaloric feeding and withholding parenteral nutrition for the first week. By conducting a direct head-to-head comparison, EPaNIC provided definitive, high-quality evidence that validated the North American guidelines and led to a global paradigm shift away from early parenteral overfeeding in critically ill patients.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

Why might providing full caloric and protein requirements via parenteral nutrition early in the course of critical illness actually increase infection risk and delay recovery?

Key Response

This touches on the pathophysiology of the acute stress response. Early administration of exogenous macronutrients, particularly amino acids, suppresses autophagy, a crucial cellular mechanism for clearing damaged organelles and intracellular microbes during acute stress. Additionally, early parenteral nutrition can lead to hyperglycemia and metabolic overfeeding, which impair macrophage function and increase infection rates.

Resident

You are managing a patient admitted to the ICU 48 hours ago with severe acute pancreatitis who is not tolerating enteral feeds. Based on the EPaNIC trial, how should you manage their nutritional needs over the next 5 days?

Key Response

According to the EPaNIC trial, you should withhold supplemental parenteral nutrition until day 8 of the ICU stay. You can continue to attempt trophic enteral feeds if appropriate, and provide standard intravenous hydration and micronutrients, but delaying the initiation of full parenteral macronutrients reduces the risk of ICU-acquired infections, shortens the duration of mechanical ventilation, and decreases the overall ICU length of stay.

Fellow

The EPaNIC trial included both well-nourished and malnourished patients. How does baseline nutritional status complicate the decision to delay PN until day 8, and how do we reconcile the physiological risk of severe caloric deficit in cachectic patients with these trial results?

Key Response

Fellows must critically evaluate subgroups. While EPaNIC showed harm with early PN overall, patients with preexisting severe malnutrition might theoretically lack the physiological reserve to endure 8 days of macronutrient deficit. Although EPaNIC's subgroup analyses did not show a benefit to early PN even in malnourished patients, the sample size for severe malnutrition was limited. This creates a nuanced clinical tension, requiring synthesis of this data with physiological principles when managing severely cachectic patients who have absolute contraindications to enteral nutrition.

Attending

In adopting the late parenteral nutrition strategy from EPaNIC, we shift our paradigm from preventing caloric deficit to tolerating acute macronutrient deficit. How do you balance this strategic shift while simultaneously managing the risks of refeeding syndrome and ensuring adequate micronutrient delivery?

Key Response

Attendings need to implement this paradigm shift safely on rounds. While macronutrient deficit is well-tolerated and beneficial acutely, micronutrient deficiencies, such as thiamine and trace elements, can be rapidly fatal. A key teaching point is that withholding parenteral nutrition does not mean withholding standard maintenance vitamins and electrolytes, and vigilant monitoring for refeeding syndrome remains critical when feeds are eventually ramped up on or after day 8.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

The EPaNIC trial utilized a large, unblinded design at a single center where tight glycemic control via intensive insulin therapy was strictly enforced. How might this background practice introduce a significant interaction effect, and what statistical methods could evaluate its impact on the generalizability of the findings?

Key Response

Early parenteral nutrition often causes hyperglycemia. If intensive insulin therapy mitigated the severe hyperglycemic harms of early PN, the true harm of early PN in standard ICU settings might be underestimated. Conversely, hypoglycemia from tight control could interact with the fasting state in the late PN group. A PhD-level researcher would explore interaction terms, mixed-effects models, or meta-regression across multicenter trials to isolate the nutritional intervention's independent effect from the center-specific intensive glycemic protocol.

Journal Editor

A major potential confounder in the EPaNIC trial is the administration of intravenous glucose to the control group during the first week to match the fluid volume of the early PN group. If you were peer-reviewing this manuscript, how would you evaluate the threat this poses to the study's construct validity regarding starvation versus feeding?

Key Response

The control group received substantial IV glucose, meaning they were not in a true fasting state. This exogenous glucose could still partially suppress autophagy or provoke hyperglycemia. An editor would demand a rigorous sensitivity analysis or detailed supplementary data quantifying the daily carbohydrate load in the control arm to ensure the late PN group was not simply functioning as an isolated IV glucose group, which would fundamentally alter the interpretation of the physiological mechanism driving the results.

Guideline Committee

Current ASPEN/SCCM guidelines recommend withholding exclusive parenteral nutrition for the first 7 days in critically ill patients who are at low nutritional risk. How does the EPaNIC trial inform the strength and quality of this recommendation, and does this trial provide sufficient evidence to apply the 7-day delay to patients with high nutritional risk?

Key Response

The EPaNIC trial provides high-quality, Level 1 evidence supporting the ASPEN/SCCM guideline to delay PN for 7 days in low-risk patients. However, the committee must deliberate on the high-risk population. Current ASPEN guidelines suggest considering earlier PN in patients at high nutritional risk or severe malnutrition if enteral nutrition is not feasible. Because EPaNIC's severely malnourished subgroup was relatively small, the evidence is insufficient to confidently mandate late PN in that specific cohort, leading guidelines to rely more on expert consensus for high-risk patients.

Clinical Landscape

Noteworthy Related Trials

2014

CALORIES Trial

n = 2400 · NEJM

Tested

Early parenteral nutrition

Population

Critically ill adults

Comparator

Early enteral nutrition

Endpoint

30-day all-cause mortality

Key result: No significant difference in 30-day mortality was observed between early parenteral and early enteral nutrition.

2016

PEPaNIC Trial

n = 1440 · NEJM

Tested

Late parenteral nutrition

Population

Critically ill children

Comparator

Early parenteral nutrition

Endpoint

New acquired infection and duration of ICU dependency

Key result: Late parenteral nutrition resulted in fewer new infections and a shorter duration of ICU stay compared to early parenteral nutrition.

2017

NUTRIREA-2 Trial

n = 2410 · Lancet

Tested

Early enteral nutrition

Population

Critically ill adults with shock on mechanical ventilation and vasopressors

Comparator

Early parenteral nutrition

Endpoint

28-day mortality

Key result: There was no difference in 28-day mortality, but early enteral nutrition was associated with a higher risk of gastrointestinal complications.

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