Hypovitaminosis C and vitamin C deficiency in critically ill patients despite recommended enteral and parenteral intakes

Every one of these patients was being fed. Tubes were running, bags were dripping, and vitamin C was calculated into the formula. By the guidelines, they were adequately nourished. And yet, one-third of them were vitamin C deficient anyway. That gap—between being fed according to the guidelines and being deficient regardless—is exactly what Carr and colleagues set out to understand. Here is why it matters. Vitamin C is an essential, water-soluble nutrient that humans cannot synthesize or store. Every molecule in circulation came from food or a feeding tube. And it does a lot: it's a potent antioxidant that protects proteins, lipids, and DNA from oxidative damage; it serves as an enzyme cofactor for synthesizing the vasopressor hormones noradrenaline and vasopressin; it supports immune responses and collagen production; and it contributes to gene regulation and epigenetic enzyme function. These are not marginal jobs. In a critically ill patient—someone already fighting to maintain blood pressure, fend off infection, and repair damaged tissue—every one of those functions is under strain. Critical illness creates a vitamin C emergency in slow motion. Infection, sepsis, and systemic inflammation generate a surge of reactive oxygen and nitrogen species, and vitamin C scavenges them, burning through the supply. Meanwhile, the body's ability to absorb and retain vitamin C is compromised in multiple ways. The intestinal transporter that pulls vitamin C into the bloodstream—called sodium-dependent vitamin C transporter one, or SVCT1—is saturable, meaning it has a ceiling. Intensive care unit patients often have gastrointestinal dysfunction, gastroparesis, impaired gut perfusion from vasopressors, and high gastric residuals that further blunt uptake. Fluid resuscitation dilutes plasma concentrations, though Carr and colleagues estimated this accounts for only about a quarter to a third of the total deficit. Continuous renal replacement therapy can remove at least half of circulating vitamin C. Add it all up, and the stage is set for depletion even in patients receiving what the guidelines say is enough. To measure exactly how bad it was, Carr and colleagues recruited forty-four critically ill adults from the Christchurch Hospital Intensive Care Unit between December 2015 and August 2016. Twenty-four met predefined criteria for septic shock; seventeen were non-septic. Blood was drawn through arterial lines at enrollment and at twelve, twenty-four, forty-eight, seventy-two, and ninety-six hours—daily snapshots over the first four days in the intensive care unit. Plasma vitamin C was measured by high-performance liquid chromatography with electrochemical detection, and C-reactive protein, the standard inflammation marker, was measured by endpoint nephelometry. Daily vitamin C intake was calculated directly from the volume and composition of whatever enteral or parenteral nutrition each patient received. The team defined hypovitaminosis C as plasma vitamin C below twenty-three micromoles per liter, and outright vitamin C deficiency as below eleven micromoles per liter. The results are striking. Across the full cohort, mean plasma vitamin C was seventeen point eight plus or minus eight point seven micromoles per liter. Two-thirds of patients—sixty-eight percent—had hypovitaminosis C. One-third—thirty-two percent—were frankly deficient. These are not marginal findings. Levels that low impair the antioxidant and cofactor-dependent physiology those patients most needed. Septic shock patients were in considerably worse shape. Their mean plasma vitamin C was fifteen point three plus or minus seven point nine micromoles per liter, compared with twenty point eight plus or minus eight point nine micromoles per liter in non-septic patients—a statistically significant difference, with a p-value of zero point zero three. Eighty-eight percent of septic shock patients had hypovitaminosis C, versus fifty percent of non-septic patients. Nearly forty percent of those with septic shock were deficient, compared with twenty-five percent of non-septic patients. The sicker the patient, the lower the vitamin C. The inflammation data explain why. At baseline, septic shock patients had C-reactive protein concentrations of two hundred fifty-four plus or minus one hundred twenty-one milligrams per liter—two point four times higher than the one hundred five plus or minus eighty-four milligrams per liter seen in non-septic patients. And when the researchers stratified by vitamin C status, patients with hypovitaminosis C had a mean C-reactive protein of two hundred sixteen plus or minus one hundred twenty-two milligrams per liter, compared with one hundred thirty plus or minus one hundred twenty-two milligrams per liter in those with plasma vitamin C above twenty-three micromoles per liter. Higher inflammation, lower vitamin C. The inflammatory process appears to be consuming the supply. Now here is the pharmacokinetic heart of the paper. Carr and colleagues didn't just describe the deficiency—they modeled it against what the patients were actually receiving. They applied a four-parameter log-logistic dose-response model, using published bioavailability data from Levine and colleagues to predict what steady-state plasma vitamin C should look like at a given daily intake. Think of it as a sigmoidal curve: as the dose increases, predicted plasma concentration rises steeply at first, then flattens as the system saturates. From the model, you can estimate what plasma level a given daily intake should produce in a healthy person. More than eighty percent of the cohort received enteral or parenteral nutrition delivering an average of one hundred twenty-five plus or minus eighty-eight milligrams of vitamin C per day. Enteral-only patients averaged one hundred two plus or minus fifty-four milligrams per day; those on total parenteral nutrition averaged two hundred six plus or minus one hundred six milligrams per day. By the model, those intakes should have produced plasma concentrations substantially higher than what was measured. The reality: measured plasma concentrations were only about one-third of the predicted values at every time point—a statistically significant mismatch with a p-value below zero point zero zero zero one. Standard nutrition was delivering vitamin C. The patients just weren't retaining it at anything close to the expected level. The urinary data add another piece. In a subset of eleven patients with urine samples, urinary vitamin C was only forty-one plus or minus thirty micromoles per liter at baseline—far below the one hundred fifty-four micromoles per liter previously observed in healthy individuals. Estimated urinary excretion accounted for roughly fifteen to thirty percent of the administered dose, meaning excess loss in the urine wasn't the main driver of the deficit. The body was using the vitamin C up, not spilling it. Inflammatory metabolic demand, not simple wastage, appears to be the dominant mechanism. So what would it actually take to normalize plasma vitamin C in these patients? The pharmacokinetic modeling and prior clinical data point to a number that is sharply higher than what current nutrition provides. Carr and colleagues conclude that doses of at least two grams per day are likely required to raise plasma levels out of the hypovitaminosis range, and approximately three grams per day to achieve saturating plasma concentrations—around sixty-eight micromoles per liter—with enhanced urinary excretion signaling the system is finally replete. That is ten to thirty times what these patients were getting from standard intensive care unit nutrition. The route matters too. Intravenous delivery bypasses the saturable intestinal sodium-dependent vitamin C transporter one and the gastrointestinal dysfunction that undermines enteral absorption. Patients on total parenteral nutrition—who were already absorbing vitamin C directly into circulation—still had mean plasma vitamin C of only sixteen point six plus or minus seven point five micromoles per liter, still in the hypovitaminosis range. That finding alone tells you that intravenous administration is necessary but not sufficient at current doses; the dose itself has to go up. The authors acknowledge a safety consideration: rare cases of acute oxalate nephropathy have been reported following high-gram intravenous vitamin C in patients with impaired renal function. But for low-gram parenteral regimens in patients with normal renal function, oxalate excretion remained less than zero point two percent with no reported nephropathy—a signal that warrants monitoring, not a reason to abandon higher dosing. They call for clinical trials to define optimal dose, route, and patient-relevant outcomes. The deeper problem this study reveals is structural. Intensive care unit nutrition guidelines were built on bioavailability data from healthy volunteers—people with intact gut function, no systemic inflammation, and no hypermetabolic state driven by sepsis. Transplant those numbers into the intensive care unit, and the math breaks down. Critically ill patients, particularly those with septic shock, face higher demand, worse absorption, some extracorporeal losses, and an inflammatory environment that actively consumes vitamin C faster than standard formulas can replenish it. The result is a cohort that is, on paper, adequately nourished and, in practice, chronically depleted of a nutrient critical to the very processes that might keep them alive. What Carr and colleagues leave open is the question that most urgently follows from their data: if you corrected the deficiency—if you gave these patients two or three grams a day intravenously and actually normalized their plasma vitamin C—what would happen to outcomes? That trial has not yet been conducted at the scale needed to answer it. But this study makes clear that we have been operating with a blind spot in critical care nutrition, and that closing it starts with acknowledging what the numbers already show: recommended is not the same as sufficient. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.