Letters to the Editor marked reduction of muscle permeability to potassium.' Effects of lithium on potassium metabolism have been the topic of in vivo and in vitro studies. Results are contradictory, depending on the study design and the patient's psychiatric state. No noticeable and consistent systematic effect of lithium on body potassium has been reported. Nevertheless, lithium could enhance Na-K pump activity, similar to potassium.2 Lithium therapy has already been proposed in various forms offamilial periodic paralysis with varying results.34 To our knowledge, there is only one other report concerning lithium therapy in a patient with FHPP.' In this case, carbonate lithium was administered to reach serum lithium levels up to 10 mmol/l. No benefit was observed, notably on attack frequency which remained about one per week. Biochemical homogeneity of FHPP may be questioned on the basis of such discrepant results. Some forms could be lithium sensitive and others, lithium resistant. Further studies are clearly needed to elucidate this problem. Lithium, as an oral potassium add-on therapy, is worth trying in FHPP cases resistant to standard therapies. It is safe and can be beneficial on rate of attack. We are grateful to Drs N Daieff and S Sirot from LABCATAL laboratories for their help in the trial design and the provision of the drug, and to Professor Guy Chazot for his valuable help.