M. Meyergessner et al., ANTITHYROID DRUG-INDUCED AGRANULOCYTOSIS - CLINICAL-EXPERIENCE WITH 10 PATIENTS TREATED AT ONE INSTITUTION AND REVIEW OF THE LITERATURE, Journal of endocrinological investigation, 17(1), 1994, pp. 29-36
The frequency, predisposing factors and course of agranulocytosis (gra
nulocytes <250/mu l) secondary to antithyroid drugs were studied in a
cohort of 1256 continously treated outpatients with hyperthyroidism du
ring the 15 year period from 1973 to 1987. Two cases of agranulocytosi
s were detected; the frequency was 0.18% (95%-confidence intervals, 0.
0-0.44%). This prevalence appears to be lower than reported in previou
s studies (up to 1.8%). For other adverse drug reactions, there was a
clear-cut relationship to initial thiyonamide dose and to the body mas
s index; most reactions occurred during the first weeks of treatment.
In addition, eight patients referred for thionamide drug- induced agra
nulocytosis were studied, and the following results obtained: Methimaz
ole dose in patients with agranulocytosis was almost twice as in other
patients (63.3+/-9.7 vs 34.3+/-29.7 mg daily) suggesting that this co
mplication was related to dose. The interval between start of antithyr
oid drug treatment and first symptoms of agranulocytosis was 33 days (
median; range 23-55 days); hence, prolonged treatment beyond this peri
od would appear relatively safe. Withdrawal of the causative agent and
treatment of infection led to recovery of leukocyte counts within 15
days (median; range, 5-31 days). Two fatal outcomes were seen in refer
red patients. In one severely hyperthyroid patient with methimazole-in
duced agranulocytosis, recombinant human granulocyte/macrophage colony
stimulating factor induced clinical and hematologic recovery within a
few days of administration. In conclusion, agranulocytosis is the mos
t severe side effect of antithyroid drugs. According to our results an
d a literature review, it occurs almost exclusively during the first t
en weeks of treatment and is probably related to the drug dose. Physic
ians should be aware of this potentially serious complication particul
arly during this period. A reduction of traditionally prescribed thion
amide doses seems quite possible and might lead to considerable reduct
ion of adverse reactions. In severe cases, a trial of recombinant huma
n granulocyte/macrophage colony stimulating factor should be considere
d.