Fourteen RCTs (n=1,306) were included in the meta-analysis. Details of the non-randomised studies were reported in the authors' discussion and online (Table A).
The quality of the included trials was generally poor, with few reporting adequate methods of randomisation or blinding. However, 97% of the participants completed the follow-up.
Treatment failure.
Adjunctive antithyroid treatment was associated with a higher risk of treatment failure (RR 1.28, 95% CI: 1.07, 1.52, p=0.006) than the control treatment. There was no differential impact between antithyroid drugs. The per-protocol analysis confirmed this finding, as did a sensitivity analysis where the timing of administration was explored. Heterogeneity amongst the study results was low. When 3 small studies were excluded from the analysis, the significance of the risk remained high and publication bias was lowered (Egger's test, p=0.15).
Hypothyroidism.
A reduced risk of hypothyroidism was reported for adjunctive antithyroid treatment compared with control (RR 0.68, 95% CI: 0.53, 0.87, p=0.006). There was no differential impact between antithyroid drugs. The per-protocol analysis revealed a similar trend, as did a sensitivity analysis on the timing of administration. Heterogeneity amongst the treatment effects was considered low to moderate. When the 3 smallest studies were excluded from the analysis, there was a similar significant reduction in risk and publication bias was also lowered (Egger's test p=0.13).
Further sensitivity analysis revealed higher risks of treatment failure associated with fixed radioiodine doses compared with an adaptive regimen when prior antithyroid treatment was given. A high risk of treatment failure and reduced risk of hypothyroidism were reported when high doses compared with lower doses of antithyroid drug were administered after radioiodine. A highly significant correlation was reported between administered radioiodine activity and rates of hypothyroidism (p<0.001) and successful treatment (p<0.001).
Adverse events.
Adverse events associated with the antithyroid drugs were reported to be low (1.8%). These included allergic skin reactions, transient neutropenia, new onset atrial fibrillation and death.