Intracerebral hemorrhage and COL4A1 mutations, from preterm infants to adult patients

Intracerebral hemorrhage (ICH) accounts for 10% of stroke and is a particularly severe form of stroke. Therapeutic tools are still limited, and prevention remains the most important way to reduce morbidity and mortality. However, for effective prevention, we need to understand the factors underlying the occurrence of hemorrhagic stroke and to identify individuals at greatest risk. Hypertension is a leading factor for ICH in adult patients, and cerebrovascular malformations are the most common causes of ICH in children. However, ICH is a heterogeneous condition, and in a substantial proportion of ICH patients, particularly in the younger ones, no cause can be identified despite extensive investigations. Recently, mutations of COL4A1 have been shown to cause ICH and porencephaly both in mouse and human. 8 COL4A1 encodes the 1 chain of type IV collagen, a major component of basement membranes, including vascular basement membrane. Mutated mice and human patients show an increased fragility of brain vessels that renders them highly sensitive to environmental factors such as birth trauma or antithrombotic agents. Gould and colleagues previously showed that surgical delivery of mouse pups dramatically reduces the risk for cerebral bleeding in the perinatal period and suggested that caesarean delivery may decrease the risk for ICH in human neonates. In this issue, de Vries and coworkers report on two siblings born preterm, at 33 weeks, and at 31 weeks of gestation, with a very low birth weight and in whom perinatal cranial routine ultrasound and magnetic resonance imaging (in the second infant) detected resolving hemorrhages and porencephalic cavities. Both siblings carry a deleterious mutation affecting a glycine residue located in the second C-terminal tandem repeated domain of COL4A1 1 chain. This mutation was inherited from their asymptomatic mother. The presence of these lesions in the first days of life strongly suggests that these hemorrhagic events occurred antenatally and not during vaginal delivery. If these two preterm infants would have been delivered at term, one would assume that they might have been diagnosed later on as having congenital hemiplegia. The first obvious and important implication of this observation is that caesarean delivery would not be sufficient to completely prevent the occurrence of ICH in mutated infants. The second one is that prevention and genetic counseling are important in this condition. This observation raises a number of questions of interest not only for neuropediatricians, obstetricians, and geneticists but also for neurologists taking care of adult patients. One major question is when should we suspect a COL4A1 mutation and conduct a genetic testing in a preterm or at-term infant showing an ICH. This question is not only important for medical care and genetic counseling but has also medicolegal implications. Currently, in cases of a familial porencephaly (more than one affected case), genetic screening of COL4A1 should be considered mandatory and will be of great help for genetic counseling. However, there are many situations in which the answer is so far unclear. Presumed antenatal or perinatal onset of a unilateral intraventricular/ICH leading to a porencephalic cyst is observed in 5 to 8% of very-low-birth-weight infants. Should all these infants be tested for COL4A1 mutations in the absence of any affected relative and any other known causative factor? What should be done when intracranial bleeding is detected in utero by ultrasound screening? Another major question is when should we suspect a COL4A1 mutation in an adult patient, who would then be at risk for ICH and at risk to have an affected child? COL4A1 mutations lead to a number of distinct phenotypes both in mouse and human, depending on the genetic background or environmental factors, or both. Relatives of familial porencephaly children have been shown to have leukoencephalopathy and an increased risk for ICH and/or cerebral microbleeds and/or ischemic lacunar infarcts. Another autosomal dominant condition characterized by the association of arteriolar retinal tortuousities, infantile hemiplegia, and leukoencephalopathy has been associated with a COL4A1 mutation. A distinct phenotype associating anterior eye chamber developmental anomalies including congenital cataract, leukoencephalopathy, ischemic lacunar infarcts, and microbleeds has been reported in a COL4A1-mutated family. At last, a novel phenotype associated with COL4A1 mutations has recently been reported under the acronym HANAC (hereditary angiopathy nephropathy aneurysms and muscle cramps). Screening of COL4A1 is clearly indicated when a patient with hemorrhagic stroke and leukoencephalopathy has suggestive associated symptoms such as retinal arteriolar tortuosity and/or hematuria, or has a history of infantile hemiplegia with porencephaly and/or anterior eye malformation, regardless of whether the proband has affected relatives. However, associated symptoms are often missed, particularly in relatives. COL4A1 mutaEDITORIALS