Euthyroid Sick Syndrome: A Low T3 That Might Be Protecting Him
A man three weeks into a critical illness has thyroid labs that look hypothyroid on paper. The disagreement is whether that pattern is his body adapting to survive a crisis, or a real hormone deficiency his body simply hasn't been able to correct on its own.
Vincent D., a 57-year-old former commercial diver, is three weeks into an ICU admission for necrotizing pancreatitis complicated by sepsis, now on his second course of pressors after a brief period off them. He has no history of thyroid disease and had normal labs on a physical eighteen months ago. Today's thyroid panel, sent as part of a broader workup for his persistent hypotension, shows a free T3 well below normal, a free T4 at the low end of normal, and a TSH that is itself low-normal rather than the elevated value a primary hypothyroid process would produce — a pattern his ICU team has seen before in prolonged critical illness but has never been entirely sure whether to treat.
This combination — low T3, low-to-normal T4, TSH that hasn't risen to compensate — is the textbook signature of nonthyroidal illness syndrome, and the weight of the evidence describes it as an adaptive response rather than a true deficiency: peripheral deiodinase activity shifts under severe illness to reduce conversion of T4 to the metabolically active T3, plausibly lowering the body's overall metabolic demand at a moment when energy conservation may aid survival. Randomized trials of thyroid hormone replacement in critically ill patients with this pattern — Brent and Hershman's is the one most often cited — have not shown a survival or outcome benefit, and some have raised concern about accelerated catabolism and added cardiac stress from artificially restoring a metabolic rate the body may have deliberately downregulated. What complicates a clean read on Vincent specifically is the duration: three weeks is longer than the acute-phase adaptation this evidence base most directly describes, and a minority of prolonged critical illness patients do develop a more central component — a genuinely blunted TSH response, not just peripheral conversion changes — which Van den Berghe's work on prolonged critical illness treats as a distinct, later-phase picture rather than a simple extension of the same early adaptive process.
In the ICU, deciding whether the thyroid numbers need correcting
I wouldn't treat this. The randomized evidence for thyroid hormone replacement in nonthyroidal illness syndrome hasn't shown a survival benefit, and some data raises real concern about accelerated catabolism and cardiac stress from correcting a pattern the body may have adaptively created on purpose. His TSH not being elevated is itself informative — a primary hypothyroid process would typically show compensatory TSH rise, and his hasn't.
I agree with not treating based on the acute-phase evidence, but I'd flag that his duration pushes past what most of that evidence base directly describes. Three weeks, with a TSH that hasn't risen to compensate, is longer than the acute adaptive picture the negative trials were largely built around, and some literature on prolonged critical illness describes a more central component developing later that isn't simply the same early peripheral-conversion story extended further.
Treating his TSH's failure to rise as purely reassuring assumes his hypothalamic-pituitary axis is responding normally to a low T3 the same way it would early in an acute illness, but at three weeks that assumption is exactly what's less certain, not more.
I wouldn't start thyroid hormone today, but I also wouldn't treat this as fully settled the way it would be at day three of his illness. If his hemodynamics remain refractory despite otherwise appropriate sepsis management over the next week, and his TSH still hasn't risen, that combination — not today's single snapshot — would be the actual trigger to test the central-axis question directly rather than continue assuming this is ordinary adaptive nonthyroidal illness syndrome indefinitely.
Agreed: no thyroid hormone replacement today, with a specific plan to reassess if his hemodynamics remain refractory and his TSH still hasn't risen over the coming week.
Not agreed: how much weight his three-week duration should carry on its own. The endocrinologist views prolonged duration as a real reason to hold a lower threshold for reassessing the central-axis question; the critical care physician would let his actual hemodynamic trajectory, not the calendar alone, decide when reassessment is warranted, wary of over-weighting duration as a trigger by itself.