Vitamin C and Intensive Care Unit (ICU) Recovery: The Length-of-Stay Meta-Analysis

Table of Contents

1. Overview
2. The Meta-Analysis at a Glance
3. Key Findings
4. Oral Dosing: 1–3 Grams per Day
5. Why Critically Ill Patients May Need More Vitamin C
6. What Later Trials Showed
7. Limitations
8. What the Evidence Means in Practice
9. Safety
10. Key Research Papers
11. Live PubMed Searches
12. Connections
13. Featured Videos

Overview

The paper at the center of this page is “Vitamin C Can Shorten the Length of Stay in the intensive care unit: A Meta-Analysis” by Harri Hemilä and Elizabeth Chalker, published in the journal Nutrients in March 2019. It is open access, meaning anyone can read the full text for free — no paywall, no library login. The PubMed record is PMID 30934660.

Harri Hemilä is a public-health researcher at the University of Helsinki who has spent decades systematically analyzing vitamin C trials — he is also the lead author of the well-known Cochrane review of vitamin C and the common cold. That background matters: this is not a supplement company's white paper but a career methodologist's careful pooling of other people's trial data.

Why measure intensive care unit stay at all? Earlier trials had suggested vitamin C might influence a scattered list of things in hospitalized patients — blood pressure, infections, bronchoconstriction (airway narrowing), atrial fibrillation after heart surgery, and acute kidney injury. The authors' point was that the practical significance of those scattered effects was unclear. So they chose two outcomes that cut through the noise:

• Length of intensive care unit stay – if vitamin C genuinely helps recovery, patients should leave the intensive care unit sooner, whatever the intermediate mechanism.
• Duration of mechanical ventilation – time spent on a breathing machine is one of the hardest, most objective markers of how sick a patient still is.

These are called pragmatic outcomes: they matter directly to patients, families, and hospitals. Every intensive care unit day is expensive — intensive care is among the costliest services a hospital runs — so an effect expressed as a percentage of stay translates straight into freed-up beds, lower bills, and less time attached to monitors. The authors' own framing was that, given the insignificant cost of vitamin C, even an 8% reduction in intensive care unit stay would be worth exploring further.

Back to Table of Contents

The Meta-Analysis at a Glance

• Design: meta-analysis of controlled trials — studies where one group received vitamin C and a comparison group did not, so the difference between groups can be attributed to the vitamin rather than to ordinary recovery.
• Included: 18 controlled trials with 2,004 patients in total.
• Who the patients were: 13 of the 18 trials studied patients undergoing elective cardiac surgery — planned heart operations such as bypass or valve procedures, where a short intensive care unit stay afterward is routine. This detail becomes very important later on this page.
• Statistical method: an inverse-variance, fixed-effect meta-analysis on the ratio-of-means scale.

That last line is dense, so here is what each term means in everyday language:

• Meta-analysis – instead of running a new trial, researchers mathematically combine the results of existing trials. Pooling many small studies gives a more stable estimate than any single one could.
• Ratio of means – rather than asking “how many days shorter was the stay?”, the analysis asks “by what percentage was the stay shorter?” This is a clever fit for intensive care unit data, because a typical stay after planned heart surgery might be a day or two while a stay for severe illness can run weeks. A percentage travels across those very different baselines; a fixed number of days would not.
• Inverse-variance weighting – trials whose results are more precise (usually the larger ones) count for more in the pooled answer. A 300-patient trial should move the needle more than a 20-patient trial, and this weighting makes sure it does.
• Fixed-effect model – the calculation assumes all the trials are estimating one common underlying effect, and differences between their results are just chance. That assumption keeps the math simple, but it is a real limitation when trials differ in dose, route, and patient type — more on that in the Limitations section.

Back to Table of Contents

Key Findings

Three numbers carry the paper. Each comes with a 95% confidence interval (CI) — the range of values the data are reasonably compatible with — and a p-value, which answers: “if vitamin C truly did nothing, how likely is it we would see a difference at least this large by pure chance?” The smaller the p-value, the harder the result is to explain away as luck.

1. Intensive care unit stay shortened by 7.8%. Across the 12 trials (1,766 patients) that could be pooled for this outcome, vitamin C reduced the length of intensive care unit stay by 7.8% on average (95% confidence interval 4.2% to 11.2%; p = 0.00003). A p-value of 0.00003 means a chance finding of this size would be expected roughly 3 times in 100,000 — this is a statistically firm result, whatever one concludes about its clinical size.
2. Oral vitamin C alone still showed an effect. In the six trials that used ordinary oral vitamin C at 1–3 grams per day, intensive care unit stay was reduced by 8.6% (p = 0.003). This subgroup matters enough to get its own section below.
3. Time on the ventilator shortened by 18.2%. In three trials of patients who needed mechanical ventilation for more than 24 hours — that is, the sicker patients — vitamin C shortened ventilation time by 18.2% (95% confidence interval 7.7% to 27%; p = 0.001). It is an intriguing pattern that the effect looked larger in patients who were sicker, though with only three trials it rests on thinner evidence than the main finding.

What does 7.8% actually feel like? The paper reports percentages, not hours, so treat the following as illustration only — our arithmetic, not the paper's claim. A typical intensive care unit stay after planned cardiac surgery runs on the order of two to four days. Take 7.8% of that:

• 7.8% of a 2-day (48-hour) stay is roughly 3½ to 4 hours.
• 7.8% of a 4-day (96-hour) stay is roughly 7½ hours.

A few hours per patient sounds small until you remember it is a few hours of the most expensive, most resource-constrained bed in the hospital, multiplied across every eligible patient, for the price of a vitamin. That is precisely the authors' argument: the effect does not need to be dramatic to be worth investigating properly, because the intervention costs almost nothing. By the same illustrative arithmetic, an 18.2% reduction for a patient facing two full days on a ventilator would be roughly eight to nine fewer hours on the machine.

Back to Table of Contents

Oral Dosing: 1–3 Grams per Day

Much of the excitement (and later disappointment) around vitamin C in critical care involves high-dose intravenous protocols. That makes this subgroup the quietly practical part of the paper: six of the trials used plain oral vitamin C — swallowed, not infused — at doses of 1 to 3 grams per day, and in those six trials intensive care unit stay was reduced by 8.6% (p = 0.003).

The weighted mean dose was 2.0 grams per day. “Weighted mean” simply means the average dose after giving larger trials proportionally more say — the same weighting logic used in the main analysis. So the representative regimen in the oral trials was about 2 g/day, in the same ballpark as what many people already take from a drugstore bottle during a cold.

Why this subgroup matters:

• It is cheap and available everywhere. Oral vitamin C at these doses costs pennies a day and requires no intravenous (IV) line, no pharmacy compounding, and no special monitoring infrastructure.
• It shows the effect did not depend on mega-dose intravenous infusions. The oral-only estimate (8.6%) was similar to the overall estimate (7.8%), suggesting the signal in these mostly surgical patients was not an artifact of a few extreme-dose studies.
• It is the dose range a care team could actually consider routinely. A hospital does not need a research protocol to give a tablet — though whether to do so remains the care team's call, as the What the Evidence Means in Practice section discusses.

Back to Table of Contents

Why Critically Ill Patients May Need More Vitamin C

Why would a vitamin do anything at all in an intensive care unit? The standard rationale rests on a well-documented observation: plasma vitamin C levels fall sharply during critical illness and after major surgery. Patients who arrive with normal levels can drop to clearly deficient levels within days of severe infection, trauma, burns, or a big operation — even while receiving the vitamin C content of standard hospital nutrition.

The usual explanation runs like this:

• Severe illness is an oxidative storm. Inflammation and tissue injury generate large amounts of reactive oxygen molecules. Vitamin C is one of the body's front-line water-soluble antioxidants, and it is consumed — literally used up — while quenching them.
• Humans cannot make their own. Unlike most animals, we lack the enzyme to synthesize vitamin C, so a sudden surge in demand can only be met from intake or from limited body stores.
• Vitamin C has day jobs that matter in recovery. It is a required cofactor for collagen synthesis (wound and surgical-site healing), for carnitine and certain neurotransmitter syntheses, and it supports the normal function of blood-vessel lining and immune cells — all systems under strain in an intensive care unit patient.
• Requirements plausibly rise just when intake falls. A sedated, ventilated patient is not eating oranges. If demand rises while supply drops, ordinary “sufficient” intakes may not keep pace during the acute phase.

None of this proves supplementation improves outcomes — that is exactly what the trials are for — but it explains why researchers thought the question worth asking. You can browse the primary literature on depleted vitamin C status in the critically ill through this live PubMed search on vitamin C plasma levels in critical illness.

Back to Table of Contents

What Later Trials Showed

The story did not end in 2019, and this page would be misleading if it stopped there. Several randomized trials of intravenous vitamin C in septic intensive care unit patients followed, including the small CITRIS-ALI trial (167 patients, which its authors called a preliminary study) and the large LOVIT trial (872 patients) — and their results push firmly against over-reading the meta-analysis.

• CITRIS-ALI (Fowler and colleagues, JAMA, 2019). This trial tested intravenous vitamin C in patients with sepsis and acute respiratory distress syndrome — among the sickest people in any hospital. Its primary endpoints were negative: vitamin C did not improve the organ-failure scores or inflammation biomarkers the trial was designed around. Over 96 hours the modified organ-failure (SOFA) score fell by 3 points on vitamin C and 3.5 points on placebo (P = .86), with no difference in C-reactive protein or thrombomodulin, the markers of inflammation and blood-vessel injury. A secondary signal suggesting lower mortality in the vitamin C group drew intense debate, but secondary findings in a trial whose primary endpoint failed are hypothesis-generating at best, and the authors and critics alike said so.
• LOVIT (Lamontagne and colleagues, New England Journal of Medicine, 2022). This was the decisive test for high-dose intravenous vitamin C in sepsis: septic intensive care unit patients randomized to intravenous vitamin C at 50 mg per kilogram of body weight every 6 hours, or placebo. The trial enrolled 872 patients who needed drugs to support their blood pressure (vasopressors), with infusions for up to 96 hours. The result went the wrong way: patients receiving vitamin C had a higher risk of the combined outcome of death or persistent organ dysfunction at day 28 — 44.5% versus 38.5% (risk ratio 1.21, 95% confidence interval 1.04 to 1.40); deaths by day 28 were 35.4% versus 31.6%. Not neutral — worse.
• VITAMINS (Fujii and colleagues, JAMA, 2020). In 216 patients with septic shock, intravenous vitamin C plus hydrocortisone and thiamine was compared with hydrocortisone alone. Time alive and free of vasopressors over 7 days did not differ (122.1 versus 124.6 hours); 90-day mortality was 28.6% versus 24.5%.
• VICTAS (Sevransky and colleagues, JAMA, 2021). In 501 patients with sepsis, intravenous vitamin C, thiamine and hydrocortisone did not increase ventilator- and vasopressor-free days compared with placebo (median 25 versus 26 days); 30-day mortality was 22% versus 24%. The trial was stopped early for administrative reasons (funding was withheld) and may have been underpowered.

The 2021 Surviving Sepsis Campaign international guideline (Recommendation 70) states: “For adults with sepsis or septic shock, we suggest against using IV vitamin C” — a weak recommendation based on low-quality evidence.

Read plainly, these trials mean the meta-analysis cannot be stretched to cover high-dose intravenous vitamin C in sepsis. The populations are almost entirely different: 13 of the 18 trials Hemilä and Chalker pooled were elective cardiac surgery patients — people having a planned operation, expected to spend a short, orderly stint in the intensive care unit — not patients in septic shock fighting for their lives. The doses and routes differ too: the oral subgroup averaged 2 g/day by mouth, while LOVIT infused far larger amounts intravenously around the clock into a profoundly different physiology.

A fair reading of all the evidence together: modest-dose vitamin C around planned cardiac surgery is associated with slightly shorter intensive care unit stays and remains a reasonable research question; aggressive intravenous vitamin C in established sepsis has been tested at scale and, on current evidence, does not help and may harm. Those are two different claims about two different situations, and conflating them is exactly the mistake this page exists to prevent.

Back to Table of Contents

Limitations

Hemilä and Chalker were explicit that their finding is a reason to investigate further, not a settled clinical verdict. The honest caveats:

• The fixed-effect model assumes one true effect. With trials spanning different doses, routes, and patient groups, that assumption is generous. A model allowing the true effect to vary between trial settings could give wider, less confident intervals.
• Ratio-of-means gives percentages, not days. A relative measure is elegant, but a 7.8% reduction means very different absolute time in a 1-day stay versus a 10-day stay, and the pooled percentage cannot tell you which patients contributed most of the saved hours.
• The evidence is dominated by elective cardiac surgery. Thirteen of eighteen trials studied one, comparatively predictable, patient type. Generalizing to trauma, sepsis, or general medical intensive care unit patients is exactly the extrapolation the later sepsis trials punished.
• Doses and routes varied widely. Oral and intravenous regimens of different sizes were pooled together; the overall estimate blurs them.
• “intensive care unit discharge” is not a laboratory measurement. Hospitals differ in their discharge criteria, bed pressure, and staffing rhythms; a patient may stay in the intensive care unit an extra night simply because no ward bed is free. That fuzziness adds noise — and in unblinded trials, potentially bias — to the outcome itself.
• Most individual trials were small. The pooled total of 2,004 patients is respectable, but no single large definitive trial anchors the result.

Back to Table of Contents

What the Evidence Means in Practice

What this evidence supports in practice is less than the headline suggests, but more than nothing:

• Intensive care unit decisions belong to the treating team. Anyone in an intensive care unit is under minute-by-minute management of fluids, electrolytes, medications, and machines. Nothing on this page is an instruction to add anything to that care; intensive care unit patients receive everything, including food and supplements, through the care team, which manages fluids, drugs and feeding minute by minute.
• The meta-analysis is published evidence about planned cardiac surgery. It is the kind of finding patients and families raise with surgical and anesthesia teams, for example by asking whether perioperative vitamin C is something the team uses or would consider.
• Arriving replete is the uncontroversial part. Whatever supplementation does, entering a hospital stay with normal vitamin C status is a plain good. That is achievable with ordinary food: citrus, bell peppers, kiwi, strawberries, broccoli, and cabbage-family vegetables are all rich sources.
• Oral 1–3 g/day is inexpensive and generally well tolerated — that is why the authors argue the question deserves bigger trials. But note carefully what this is not: it is not a home intensive care unit protocol, not a substitute for any prescribed treatment, and not a reason to expect a specific outcome for a specific person. An 8% average is a statistical statement about groups.
• The cardiac-surgery result does not carry over to sepsis. In people critically ill with infection, the high-dose intravenous question has been tested in randomized trials with unfavorable results (see What Later Trials Showed), and the 2021 Surviving Sepsis Campaign guideline suggests against intravenous vitamin C in adults with sepsis or septic shock.

Back to Table of Contents

Safety

Vitamin C at ordinary supplemental doses is one of the better-tolerated substances in the supplement aisle, but the specific contexts on this page come with specific cautions:

• Digestive limits come first. As oral doses climb toward each person's “bowel tolerance,” unabsorbed vitamin C draws water into the gut and causes osmotic diarrhea, gas, and cramping. This is the usual ceiling on oral dosing — unpleasant rather than dangerous, and it resolves when the dose drops.
• Kidney-stone risk in predisposed people. Part of a vitamin C load is metabolized to oxalate, the main ingredient of the most common kidney stones. Gram-level doses raise urinary oxalate, and the risk concentrates in people with a history of calcium-oxalate stones and in people with significant kidney disease.
• Glucose-6-phosphate dehydrogenase (G6PD) deficiency and very high intravenous doses. In people with glucose-6-phosphate dehydrogenase deficiency — a common inherited enzyme variant — very high-dose intravenous vitamin C can trigger hemolysis (red blood cells breaking apart). This is why intravenous protocols screen for glucose-6-phosphate dehydrogenase status; it is not a concern at ordinary oral doses.
• Glucose-meter interference. At the high blood levels reached with intravenous dosing, vitamin C can fool some point-of-care glucose meters into reading falsely high or low. In a hospitalized diabetic patient whose insulin is dosed off those readings, that is a genuinely dangerous artifact — another reason high-dose vitamin C belongs under clinical supervision, with the team aware it is on board.

This page is for information and education. It is not medical advice, and it is no substitute for the judgment of the clinicians actually caring for a patient.

Back to Table of Contents

Key Research Papers

Every citation below was verified against the PubMed record before being listed. The link on each entry leads to its PubMed page.

1. Hemilä H, Chalker E. Vitamin C Can Shorten the Length of Stay in the ICU: A Meta-Analysis. Nutrients. 2019;11(4):708. — The paper this page summarizes. PubMed PMID: 30934660; open access (PMCID PMC6521194; doi:10.3390/nu11040708). — PubMed PMID: 30934660
2. Fowler AA 3rd, Truwit JD, Hite RD, et al. Effect of Vitamin C Infusion on Organ Failure and Biomarkers of Inflammation and Vascular Injury in Patients With Sepsis and Severe Acute Respiratory Failure: The CITRIS-ALI Randomized Clinical Trial. JAMA. 2019;322(13):1261-1270. — IV vitamin C in sepsis-ARDS: negative primary endpoints, debated secondary mortality signal. — PubMed PMID: 31573637
3. Lamontagne F, Masse MH, Menard J, et al. Intravenous Vitamin C in Adults with Sepsis in the Intensive Care Unit. New England Journal of Medicine. 2022;386(25):2387-2398. — The LOVIT trial: high-dose IV vitamin C in septic ICU patients increased the risk of death or persistent organ dysfunction. — PubMed PMID: 35704292
4. Fujii T, Luethi N, Young PJ, et al. Effect of Vitamin C, Hydrocortisone, and Thiamine vs Hydrocortisone Alone on Time Alive and Free of Vasopressor Support Among Patients With Septic Shock: The VITAMINS Randomized Clinical Trial. JAMA. 2020;323(5):423-431. — PubMed PMID: 31950979
5. Sevransky JE, Rothman RE, Hager DN, et al. Effect of Vitamin C, Thiamine, and Hydrocortisone on Ventilator- and Vasopressor-Free Days in Patients With Sepsis: The VICTAS Randomized Clinical Trial. JAMA. 2021;325(8):742-750. — PubMed PMID: 33620405
6. Evans L, Rhodes A, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Medicine. 2021;47(11):1181-1247. — PubMed PMID: 34599691
7. Hemilä H, Chalker E. Vitamin C for preventing and treating the common cold. Cochrane Database of Systematic Reviews. 2013;2013(1):CD000980. — The same authors' widely cited Cochrane review, which found modest, consistent reductions in cold duration with regular supplementation; useful context for their methods and long track record on this vitamin. — PubMed PMID: 23440782

Back to Table of Contents

Live PubMed Searches

These searches open the current literature on PubMed, so they stay up to date as new trials are published:

1. Vitamin C and intensive care unit length of stay
2. Vitamin C in cardiac surgery — randomized trials
3. Vitamin C and duration of mechanical ventilation
4. Vitamin C and post-operative atrial fibrillation
5. Ascorbate depletion in critical illness
6. Vitamin C in sepsis — randomized trials

Back to Table of Contents

Connections

• Vitamin C, Collagen & Why Sailors Fell Apart — interactive animation
• Vitamin C
• Vitamin C Benefits
• Intravenous High-Dose Vitamin C and Cancer
• Vitamin C and Immune Function
• Vitamin C and Immune Defense
• Vitamin C and Iron Absorption
• Remedies
• Pulmonology

Back to top

Featured Videos