COADY CODE

Could a Ketogenic Diet Help in Severe Sepsis? Early Evidence from a 2024 RCT

Written by

tomc

in

Here is a blog post based on the 2024 study promoted in the X post from @metabolicsummit, written in an informative yet balanced tone suitable for a health/science-interested audience.

Sepsis remains one of the leading causes of death in intensive care units worldwide, driving massive inflammation, organ dysfunction, and metabolic chaos. Standard ICU nutrition has long relied on high-carbohydrate enteral or parenteral feeds to meet energy demands — but emerging research is questioning whether that’s always optimal.

A single-center study published in July 2024 in Science Translational Medicine tested something radically different: a ketogenic diet (very low carbohydrate, high fat) in critically ill adults with sepsis.

Study Design – Small but Promising

  • Type: Open-label, randomized controlled trial (RCT)
  • Participants: 40 critically ill adults diagnosed with sepsis (conducted at a single German center)
  • Intervention: Patients were randomized to either:
  • Ketogenic diet (KD) group — designed to induce and maintain nutritional ketosis
  • Standard high-carbohydrate nutrition (control group, reflecting usual ICU care)
  • Primary aim: Could a KD safely achieve stable ketosis (measured via β-hydroxybutyrate levels) without major metabolic harm?
  • Key exclusion criteria typical for ICU nutrition trials (e.g., severe liver failure, contraindications to enteral feeding)

The trial wasn’t powered for hard mortality endpoints — it was an early-phase feasibility + exploratory outcomes study.

Main Safety & Feasibility Findings

The ketogenic approach proved surprisingly feasible in this very sick population:

  • Stable ketosis was achieved in all KD patients, with β-hydroxybutyrate levels significantly higher than controls (mean difference ~1.4 mmol/L, p < 0.001).
  • No major adverse events linked to the diet: no harmful acidosis, dangerous dysglycemia, or dyslipidemia.
  • Striking insulin difference: After day 4, 0% of the KD group required insulin infusions, versus 35–60% of controls on any given day (p = 0.009). This suggests the KD dramatically reduced stress hyperglycemia and exogenous insulin needs.

Exploratory Clinical Outcomes – Eye-Catching but Preliminary

While 30-day survival was similar between groups (as expected in a small trial), several organ-support-free days favored the ketogenic arm over 30 days (reported as incidence rate ratios, all p < 0.001):

  • Ventilator-free days: IRR 1.7 (95% CI 1.5–2.1)
  • Vasopressor-free days: IRR 1.7 (95% CI 1.5–2.0)
  • Dialysis-free days: IRR 1.5 (95% CI 1.3–1.8)
  • ICU-free days: IRR 1.7 (95% CI 1.4–2.1)

In plain terms: patients on the KD group spent more time off mechanical ventilation, off blood pressure support, off dialysis, and out of the ICU compared to those on standard nutrition.

Immune & Inflammatory Signals

The authors dug into mechanisms via next-generation sequencing of CD4⁺/CD8⁺ T cells and cytokine profiling:

  • Reduced markers of excessive T-cell activation and signaling
  • Lower pro-inflammatory cytokine secretion (e.g., trends toward decreased IL-6 and others)

This aligns with preclinical data suggesting ketones have anti-inflammatory and immunomodulatory effects — potentially blunting the cytokine storm that drives sepsis progression.

Important Caveats – This Is Not (Yet) Practice-Changing

  • Open-label design → risk of bias in care decisions
  • Small sample (n=40) → results need replication
  • Single center → generalizability uncertain
  • No difference in mortality (expected at this size)
  • Larger, multicenter, blinded (or at least feeding-protocol-blinded) RCTs are essential before considering KD in routine sepsis care.

Follow-up protocols are already emerging (e.g., a planned Chinese multicenter RCT protocol published in 2025 exploring organ protection), and reviews continue to call for bigger trials.

Bottom Line for Now

This 2024 RCT is one of the first human demonstrations that a carefully supervised ketogenic diet can be safely implemented in septic ICU patients, induces reliable ketosis, virtually eliminates insulin requirements after a few days, and is associated with meaningfully more organ-support-free days — all without obvious harm.

Whether these organ-recovery signals hold up in larger studies — and whether they translate to improved survival or long-term function — remains the big open question.

For clinicians and researchers interested in metabolic interventions in critical illness, this paper (Rahmel et al., Sci Transl Med 2024) is worth reading in full. For everyone else: intriguing first step, but not yet time to rewrite ICU nutrition guidelines.

What do you think — could rethinking carbs in the ICU become a real paradigm shift, or is this another metabolic fad that won’t survive bigger trials?

(References: Rahmel T et al. Science Translational Medicine 2024;16:eadn9285. Additional context from PubMed, Nature Reviews Nephrology commentary, and related 2024–2025 publications.)

More posts