Who knows? How would anyone know if someone survived without medical intervention? It might take them a long time of maybe 24 hours also. It's highly variable and there are probably all kinds of scenarios.
I looked into it a little further. It is confusing because the primary natural pathway humans use for reducing heme iron has multiple steps and variables. I found the paper I was thinking of.. and just sharing the information.
(copied & pasted)
Hemoglobin Autooxidation/Oxidation Mechanisms and Methemoglobin Prevention or Reduction Processes in the Bloodstream Literature review and outline of autooxidation reaction:
Red blood cells contain enzymatic or chemical reduction systems that restore the molecule to its active form and physiological function, keeping the circulating methemoglobin below 1%. These processes are equally important in methemoglobinemia, but they often appear to be insufficient.
This [The NAD-dependent
pathway or Embden-Meyerof] is normally the only pathway involved in methemoglobin reduction and the reaction is slow. Reduction varies from one species to another; it is fast in the rabbit, and slow in humans, dogs, rats and guinea pigs . It can correct a methemoglobinemia of up to 30% of the total hemoglobin due to poisoning spontaneously within 24-72 hours. One molecule of cytochrome b5 can link to one subunit of methemoglobin and transfer the electron required to reduce the femc iron to ferrous iron. The cytochrome is indispensable and its concentration is what determines the rate of
pathway is different... This pathway has a very low physiological output because it needs an electron acceptor not found in the body between the enzyme and the methemoglobin . This must, therefore, be brought from outside. Clinical treatment generally uses methylene blue as electron acceptor.
(end copy & paste)
I did not look into the non-enzymatic pathways too closely. The non-enzymatic pathways can only account for 30% of the total erythrocyte reduction potential at best.