Editor—Maternal clomipramine use during pregnancy may cause epilepsy in newborns.1-3 We present the case of a neonate who developed a thus far unrecognised manifestation of clomipramine withdrawal.

A 38 year old woman had been treated for depression throughout pregnancy with clomipramine (100 mg/day). No other medication had been taken. A boy weighing 3370 g was born at term; he had excellent Apgar scores. Two days later, the child was jittery following startling impulses. On the fifth day, simultaneous and symmetrical jerks of arms and legs developed, but the child remained alert. The jerks could be provoked by touch or auditory stimuli, and stopped by briefly stabilising the limb. Neither the face nor the eyes showed abnormal movements. Clomipramine serum concentration was <10 ng/ml. Other laboratory tests, cerebrospinal fluid analysis, and ultrasound brain scans were unremarkable. Suspecting a status epilepticus, the child was treated intravenously with clonazepam and phenobarbital, without success. Subsequently, a 10-channel electroencephalogram, recorded while limb jerks were present, showed no epileptic activity and a normal background pattern.

We diagnosed a generalised, stimulus sensitive, status myoclonus. A single dose of 0.5 mg clomipramine suppressed the myoclonus immediately, but caused somnolence. Over the next four days, the patient had only sporadic myoclonic jerks, while consciousness improved. Examination three weeks later was normal, except for mild jitteriness in response to touch.

This report shows that the clinical spectrum of clomipramine withdrawal in newborns includes myoclonus. Limb jerks appeared several days after birth and were suppressed by clomipramine, which suggests a withdrawal effect1-4 Myoclonus has hitherto not been recognised, but epileptic seizures are commonly reported.1-3 However, EEGs are never performed, except for one neonate with generalised clinical manifestations and a focal epileptogenic focus.3Hence, epilepsy was not unambiguously demonstrated and myoclonus may have been present in some patients.

In this case, several observations argued against epilepsy. Antiepileptic drugs were ineffective, limb jerks were stimulus sensitive, and the child remained responsive during generalised jerks. Importantly, the EEG showed no abnormalities during the generalised jerks, which proved they were not epileptic.

Recognition of myoclonus has therapeutic consequences because status epilepticus requires aggressive treatment, with phenobarbital as the first choice. However, phenobarbital is often ineffective for myoclonus1 3 and may theoretically aggravate withdrawal signs due to liver enzyme induction which facilitates clomipramine clearance.4 Low dose clomipramine, which allows for graduate tapering of clomipramine concentrations, seems more effective for myoclonus.1 3