Editor—We read with interest the article by Ulrike et al.1 They diagnosed hereditary thrombophilia in seven out of 35 (20%) neonates investigated; the cerebral arteries were involved in seven of 79 (9%) of cases in the study. Tests were not available at the time of the study to detect resistance to activated protein C.

In another study of 37 children with venous and arterial thrombosis, 5.1% of a control group, 52% with venous thrombosis, and 38% with arterial thrombosis, had the factor V mutation associated with activated protein C resistance (APCR)).2 In this study cerebral artery thrombosis occurred in 14 neonates, suggesting that APCR and other defects of the protein C pathway may have a major role in neonatal cerebral thrombosis.

Of those with APCR, most cases of venous thrombosis (60%) and 28% of cases with arterial thrombosis were associated with exogenous risk factors, and the authors therefore suggested that an acquired risk of thrombo-embolic disorders masks the coagulation deficiency in most patients with an inherited pre-thrombotic state. However, arterial thrombosis occurred in the absence of risk factors in both studies.

The genetic defect causing APCR is a point mutation of the factor V gene, often referred to as factor V Leiden. The mutation is often present in heterozygous form in 5% of white European and North American populations. It is detected in 20% of cases of deep vein thrombosis and in 50 to 60% of pregnancy associated venous thromboses. The odds ratio for increased risk of venous thrombosis is 7 for heterozygotes and 80 for homozygotes. Although typically associated with venous thrombosis, recent studies of children2 and adults3 have suggested a possible role for APCR in arterial thrombosis.

We recently diagnosed heterozygosity for the Leiden mutation in a 2 year old boy with ACPR (1.21; normal range 2.08–3.62), who has cerebral palsy but not other adverse neonatal risk factors. Investigations for abnormalities of protein C, S, antithrombin III and anticardiolipin antibodies were normal. At 8 weeks he had a divergent squint; at 3 months he developed infantile spasms and hypsarrhythmia which were quickly controlled with Vigabatrin. A computed tomogram indicated a well defined cerebral infarct in the left posterior occipital region. He currently has a right hemianopsia and mild right hemiparesis.

It was thought that the cerebral infarction occurred in utero or perinatally, but the history did not indicate any explicable event or risk factor. His mother had recurrent deep vein thrombosis and pulmonary embolism at an early age. Both she and the father were screened for thrombophilia, as well as the child. Interestingly, the father and child were both heterozygous for factor V Leiden, but the mother was not. The mother may have another, as yet, unidentified cause of thrombophilia, so the child could have several risk factors for thrombosis.

The question is whether the presence of heterozygosity is relevant to the aetiology of cerebral palsy in this case. Although heterozygosity is common, we feel that it would be unwise to preclude a possible connection. We believe that not only are further studies warranted to investigate the prevalence of APCR and other defects of proteins C and S in neonates with thrombosis, but also in children who have cerebral palsy as a result of vascular thromboses that have occurred at an undetermined time, often in utero. How many children with unexplained cerebral palsy or cerebral palsy with relatively minor risk factors, are heterozygous for ACPR? If there is an association, it may afford some comfort to parents and clinicians, that there is, at least in some cases, a potentially inherited risk factor for cerebral palsy.