Septo-optic-pituitary dysplasia (SOD; de Morsier syndrome) is a complex developmental malformation of the brain that is a principal cause of optic nerve hypoplasia with variable visual impairment from normal to total blindness and of pituitary insufficiency with variable neuroendocrine defects ranging from none to isolated growth hormone deficiency to panhypopituitarism. It shares some features with holoprosencephaly, such as agenesis of the septum pellucidum and agenesis or hypoplasia of olfactory bulbs, ^{Reference Sarnat and Flores-Sarnat1} but does not fulfill criteria to classify it as a variant of holoprosencephaly. Other frequently associated brain malformations include rhombencephalosynapsis of the cerebellum, agenesis or dysgenesis of the corpus callosum, schizencephaly, heterotopia and ectopic neurohypophysis. Clinical features of SOD are well described both in the neonate and later childhood; they also include cognitive impairment, autism, ataxia, impaired olfaction and epilepsies, including infantile spasms in some. ^{Reference Alt, Shevell, Poulin, Rosenblatt, Saint-Martin and Srour2} Neuroimaging and neuropathological findings are well described. ^{Reference Alt, Shevell, Poulin, Rosenblatt, Saint-Martin and Srour2–Reference Blackburn, Thomas, Hughes and Pierson5} Prenatal diagnosis of agenesis of the septum pellucidum can be made in the 2^nd or 3^rd trimesters by fetal ultrasonography or MRI, ^{Reference Pilliod, Pettersson and Gibson6} but this isolated finding postnatally without optic nerve hypoplasia or pituitary insufficiency is insufficient to diagnose SOD. Multiple mutated genes have been associated, but even the most frequent, HESX1, involves a small minority of cases, ^{Reference McNay, Turton and Kelberman7} and in many patients with SOD even extensive genetic testing fails to disclose a pathogenic mutation. Early pregnancy vascular disruption in the basal forebrain has also been implicated, especially if there was fetal exposure to teratogenic drugs, neurotoxins or maternal diabetes mellitus, which are risk factors. ^{Reference Dominguez, Vila-Coro, Slopis and Bohan8}

Long-term prognosis of mortality in affected patients is poorly documented, however. This aspect of mortality is the focus of this novel study by Dr Salman and his colleagues at the University of Manitoba, who not only have considerable clinical experience but have also contributed novel research data in the field of developmental brain malformations, including SOD. They have now demonstrated with convincing data from 124 patients with SOD and age-matched controls with mean follow-up of 12 years that the life expectancy of affected children is shortened, with late childhood deaths due to a variety of factors, including seizures, adrenal crises and other co-morbidities. ^{Reference Salman, Ruth, Yogendran and Lix9} Such risk factors had previously been described by these authors. ^{Reference Salman, Ruth, Yogendran, Rozovsky and Lix10} Their present study defines criteria of SOD and was well designed, controlled and statistically valid. Management of co-morbidities and complications can reduce mortality risk. It is a major contribution to the literature on SOD.