Intravenous vitamin C megadose: the Marik protocol collapse and what replaced it

For about three years between 2017 and 2020, intravenous vitamin C at 6,000 mg/day combined with hydrocortisone and thiamine, dubbed the HAT or Marik protocol, was promoted as the breakthrough adjunctive treatment for sepsis. The original case series reported a 91% relative reduction in hospital mortality. Five subsequent randomized controlled trials and a Cochrane review have not replicated the finding; the largest trial actually reported worse outcomes. The story is a clean illustration of why retrospective single-center series cannot substitute for randomized data.

The 2017 case series that started it

Marik and colleagues published a before-after retrospective cohort study at a single hospital comparing 47 consecutive sepsis patients treated with HAT to 47 historical controls, reporting hospital mortality of 8.5% vs 40.4% (p < 0.001) [1]. The findings were widely covered, replicated rapidly into ICU protocols without trial-level evidence, and discussed extensively on critical-care social media as the "Marik cocktail." The study's authors were transparent that randomized confirmation was required.

The randomized trials

The VITAMINS trial in 216 septic shock patients reported no difference in time alive and vasopressor-free at 7 days with HAT versus hydrocortisone alone (median 122 vs 124 hours) [2]. The CITRIS-ALI trial in 167 patients with sepsis-induced ARDS reported no improvement in modified SOFA score, biomarkers of inflammation, or mortality at 96 hours, despite achieving target plasma vitamin C levels [3]. The ACTS trial in 205 septic shock patients found no difference in 30-day mortality. The 2021 LOVIT trial in 872 adults with sepsis randomized to IV vitamin C 50 mg/kg every 6 hours or placebo found a statistically significant increase in the composite of death or persistent organ dysfunction at 28 days (44.5% vs 38.5%, RR 1.21) — a signal of harm rather than benefit [4]. The 2024 follow-up at 6 months showed persistence of the early adverse signal.

The Cochrane review and the guideline removal

The 2022 Cochrane review of 15 trials with 2,490 sepsis patients concluded high-certainty evidence that vitamin C as an adjunct to standard care does not reduce mortality [5]. The Surviving Sepsis Campaign 2021 update issued a strong recommendation against the routine use of intravenous vitamin C for sepsis [6]. As of 2024, the HAT protocol has been removed from most institutional sepsis order sets.

Why high-dose IV vitamin C is not benign

Vitamin C is filtered and partially reabsorbed by the kidney. Intravenous infusion of doses above 2 g produces an oxalate load that, in patients with renal impairment, can precipitate calcium oxalate in renal tubules. Several case series have reported acute oxalate nephropathy attributable to high-dose IV vitamin C, particularly in patients with chronic kidney disease or pre-existing oxalate burden [7]. Vitamin C also interferes with point-of-care glucose meter readings, producing falsely elevated values that have led to inappropriate insulin dosing in the ICU.

Where IV vitamin C still has legitimate use

Cancer-related quality-of-life adjunctive use remains an active research area, with Phase 2 trials showing some symptom improvement during chemotherapy [8]. Severe scurvy treatment requires IV repletion when oral therapy is not feasible. Burns resuscitation in the early hours after major thermal injury has limited supporting data but is not standard. Outside these specific indications, high-dose IV vitamin C should not be used outside trials.

Bottom line

The Marik HAT protocol for sepsis does not improve outcomes and may worsen them. Intravenous vitamin C at megadoses carries a meaningful renal risk, especially in patients with kidney impairment, and the global guideline position is now against routine use in sepsis. The remaining legitimate indications are limited and specific.