Myxedema Coma: Loading Two Hormones Into a Fragile Heart
An 81-year-old man found unresponsive and profoundly hypothermic has myxedema coma, a diagnosis with a real mortality rate and almost no randomized evidence behind its treatment. The disagreement is whether to load T3 alongside T4 in a heart that may not tolerate either drug's full effect.
Harold T., an 81-year-old retired postal supervisor who has lived alone since his wife's death two years ago, was found on his kitchen floor by a neighbor who hadn't seen him collect his mail in three days. He arrived hypothermic at 89°F, bradycardic at 42, and minimally responsive, with a puffy, doughy quality to his face and shins that the admitting resident recognized before the labs came back. His TSH returned markedly elevated at 78 μIU/mL with a free T4 too low to detect on the standard assay — severe, longstanding, untreated hypothyroidism, almost certainly the same Hashimoto thyroiditis his outpatient chart notes he stopped filling levothyroxine for over a year ago, around the time his wife died. His admission ECG shows sinus bradycardia with a QTc of 510 milliseconds and diffusely low-voltage complexes; a random cortisol drawn on arrival, sent because his exact adrenal reserve is unknown, is pending.
Myxedema coma carries a mortality rate that case series have placed as high as thirty to forty percent, and the honest limitation of every treatment recommendation for it is that the condition is too rare, and too immediately dangerous to randomize, for any trial to have settled how aggressively to replace his hormone. What the case-series literature does show is that T4 alone can take days to weeks to meaningfully raise peripheral T3, since it depends on his body's own, likely impaired, conversion capacity — the same impairment probably contributing to how he got here. Combined IV T4 and T3 loading restores active hormone faster, but T3 itself is the more arrhythmogenic and myocardial-oxygen-demanding of the two hormones, and his QTc and bradycardia describe a heart that may not have much room to absorb a rapid metabolic acceleration.
In the ICU, deciding how much hormone to load and how fast
I'd load combined IV T4 and T3, not T4 alone. The case-series literature Wartofsky's reviews summarize consistently describes better survival with combined therapy than T4 monotherapy in myxedema coma, precisely because a patient this sick may not convert T4 to T3 fast enough on his own to matter in the window that counts. And hydrocortisone has to go in first, or at the same time — not after — because pushing thyroid hormone into a patient with unrecognized coexisting adrenal insufficiency can precipitate adrenal crisis by accelerating cortisol clearance faster than an impaired axis can replace it.
I don't disagree about the hydrocortisone timing — that part isn't in dispute. But his QTc is already at 510 and his voltages are low, which in an elderly man can reflect a myxedematous pericardial effusion as much as the hypothyroidism itself, and T3 is the hormone most directly implicated in precipitating arrhythmia and myocardial ischemia when repletion outpaces what a compromised heart can tolerate.
The combined-therapy survival data you're citing comes from case series, not randomized trials, in a condition too rare and too acute to have generated better evidence — that's a real limitation on how confidently we should load his particular heart the same way an average case-series patient was loaded. I'd start T4 alone at a reduced initial dose and reassess his rhythm before adding T3 at all.
I'd split the difference in a way that respects both arguments rather than picking one: hydrocortisone now, combined T4 and T3, but T3 at the lower end of the case-series dosing range with continuous telemetry and a specific plan to hold the next T3 dose if his QTc widens further or he develops any new ectopy. His mortality risk from undertreating a true myxedema coma is real and immediate; so is his arrhythmia risk. Neither gets to be the only consideration tonight.
Agreed: hydrocortisone first, then combined IV T4 and reduced-dose T3, continuous telemetry, with an explicit rule to hold the next T3 dose for QTc widening or new ectopy.
Not agreed: whether to have started T3 at all before seeing at least one set of repeat labs. The cardiologist would have preferred confirming the rhythm tolerated T4 alone first; the endocrinologist views that sequencing as costing hours this patient's survival odds may not afford. The reduced T3 dose was the compromise; the sequencing disagreement itself was not resolved, only overtaken by the need to act.