Isoniazid overdose associated with lactic acidosis

Isoniazid is a mycobacteria-selective anti-tuberculosis drug, which is one of those antibiotics which are metabolised almost entirely by the liver (by acetylation). It has several modes of action, of which the better-studied one is the inhibition of mycolic acid synthesis, which leads to failure of cell wall construction.

In overdose, isoniazid causes most spectacular acidosis; the authors of one case report claim their pH (of 6.49) is “the lowest reported pH level with patient survival” . As little as 1.5g is enough to cause serious problems. The central issue with toxicity from this drug is not the acidosis but rather the relentless seizures which result from pyridoxine (vitamin B6) depletion; isoniazid binds to pyridoxine and prevents it from acting as a cofactor in the synthesis of GABA. No GABA means endless refractory seizures, until pyridoxine is supplemented.

 So, the seizures are enough to produce a lactic acidosis, but in absence of other pathological mechanisms this process will go unnoticed as the lactate is rapidly cleared. Hepatic lactate dehydrogenase will chew through it, converting it all into pyruvate, and returning it into the circulation by gluconeogenesis, or by returning it into the waiting arms of the oxidative phosphorylation enzymes. Right?

isoniazid toxicity

If only it were so. Isoniazid  also seems to inhibit lactate dehydrogenase, preventing its hepatic clearance. This seems to occur by its effect on NAD, a diphosphopyridine nucleotide. Normally isoniazid inhibits mycobacterial pyridoxal (the oxidised form of pyridoxine) reducing its availability to the enzymes which participate in mycolic acid synthesis. In overdose it may affect human NAD in the same way, which then results in decreased lactate dehydrogenase activity (because of decreased NAD availability; remember that pyruvate synthesis requires both lactate and NAD). This is really more of a theory; there don’t seem to be any papers about this effect in humans.

However, the seizures are definitely a major factor in the origin of the extra lactate. Dogs who were fed vast quantities of isoniazid did not develop any lactic acidosis until after they developed seizures; and those dogs who had been given muscle relaxants never developed the acidosis.

In any case, this drug fairly rare. Let us consider for a moment a person who is intent on killing themselves with lactic acidosis, specifically. In such a situation, even the author (who is immersed daily in a critical care environment) would have to work hard to get to a sizeable amount of isoniazid- and while doing so, would trip over a half-dozen other drugs which are more easily available and which also cause lactic acidosis. In short, the exotic


I offer my thanks to Antje R. Weseler, who has kindly corrected this page (it used to incorrectly state that pyruvate decarboxylase was the enzyme affected by biguanides), and who has provided some excellent references to stimulate a biguanide toxicity enthusiast:


Large, Valérie, and Michel Beylot. "Modifications of citric acid cycle activity and gluconeogenesis in streptozotocin-induced diabetes and effects of metformin."Diabetes 48.6 (1999): 1251-1257.


Kwong, Shun C., and Jeffrey Brubacher. "Phenformin and lactic acidosis: a case report and review." The Journal of emergency medicine 16.6 (1998): 881-886.


Jitrapakdee, Sarawut, et al. "Structure, mechanism and regulation of pyruvate carboxylase." Biochem. J 413 (2008): 369-387.




Gjedde S, Christiansen A, Pedersen SB, Rungby J. Survival following a metformin overdose of 63 g: a case report. Pharmacol Toxicol. 2003;93(2):98–9


Peters, Nicolas, et al. "Metformin-associated lactic acidosis in an intensive care unit." Critical Care 12.6 (2008): R149.