To our knowledge, this is one of the main single-center series of patients with neonatal CVT and DMVT. In the entire Paduan cohort as well as in all subgroups, there is a preponderance of the male sex in analogy with what is also reported in the literature. [1, 8]
In the population investigated, no significant family history of thromboembolism emerged. In purely speculative terms, the absence of family history of thromboembolism could be considered an indirect demonstration of the predominantly acquired nature of the neonatal cerebral venous thrombotic phenomenon. [15] Furthermore, the absence of a consistent finding of fetal thrombosis in patients (only one in our population) places the cerebral venous thrombotic phenomenon, especially that of the medullary veins, in a neonatal/perinatal rather than prenatal context, stressing the important role of causal factors attributable to the perinatal period. [16, 17] Analyzing the maternal risk factors, almost a fifth of newborns with cerebral venous thrombosis were conceived using medically assisted procreation (MAP). This percentage is significantly higher than the 2.8% reported in Italy by the ISS (Istituto Superiore di Sanità) in 2020 for the general population,[18] suggesting a potential association between MAP and thrombotic risk in the newborn, which however were not previously reported. The presence of risk factors of placental origin, predominantly infectious, had been identified in approximately 8% of our population, which raised to 12% if we consider exclusively patients with DMVT. Furthermore, approximately one third of patients with DMVT were born through a complicated birth, (i.e. urgent caesarean section or dystocic vaginal delivery). There is a lack of reference data in the literature regarding this group and the expansion of the population could be useful to verify any associations.
In relation to neonatal risk factors, one third of patients with DMVT and two thirds of patients with DMVT associated with thrombosis in other intracerebral venous sites, was born preterm. Similar results were also obtained regarding the need for neonatal resuscitation and low Apgar score. Considerations related to low birth weight are limited by the fact that in the register this data was not associated to gestational age. Unlike neonatal arterial stroke, in which there is a relatively high rate of congenital heart disease, 39.5% in the Italian registry [12], only 10% of newborns with CVT were affected. Analyzing the subgroups, in approximately one fifth of the subjects with CVT without DMVT there was a major congenital heart disease (excluding patent ductus arteriosus and foramen ovale), while this condition was not diagnosed in any of the patients with DMVT. The finding of a neonatal infection was rather common in our cohort, affecting approximately a quarter of cases, similar to other case series previously reported.[19] In particular, infectious events reached a higher percentage in the case of involvement of the medullary veins (40% of newborns).
Onset of symptoms within the first 7–10 days of life was comparable to data previously reported. [1, 8] This timing was very different from other neonatal neurological conditions, such as hypoxic ischemic encephalopathy which manifested itself in the first 12 hours of life or arterial ischemic stroke which presented with epileptic seizures starting from 72 hours of life.[1, 8] We highlighted that 20% of newborn with CVT was asymptomatic and the diagnosis represented an incidental finding to the cranial ultrasound. This data was also confirmed by considering the subgroup of patients with isolated DMVT, unlike previously reported. [8] This discrepancy suggested that this condition could sometimes be underdiagnosed due to the presence of asymptomatic patients. In any case, seizures were the predominant clinical presentation in all subgroups, which, similarly to the literature, occurred in approximately 50% of cases, followed by drowsiness and irritability.[20]
Regarding the radiological presentation, in the present study we analyzed parenchymal involvement secondary to the thrombotic event. The data relating to the occurrence of parenchymal lesions was similar to previously published data about isolated DMVT. [21–24] As previously reported, DMVT was associated to hemorrhagic venous infarctions in the frontal, parietal, and temporal lobes. [25–28]
Therapy of neonatal CVT is a controversial topic: there are no definitive therapeutic guidelines, and the decision of starting treatment is often left to physician’s personal experience. In our case series, a third of total CVT received anticoagulation treatment, of which a quarter of patients with isolated DMVT. This data was comparable to those previously reported. [1, 8] More than a half of the subjects had no neurological impairment at discharge, which is similar to previous published data.[29]
From the comparison between the two groups, no statistically significant differences emerged from the point of view of clinical presentation, and outcomes, even if patients with isolated DMVT. Regarding risk factors, sepsis and thrombophilia seemed to be more related to isolated DMVT and CVT respectively.
Hemorrhagic lesions appeared to be more common in case of isolated DMVT. This could be explained by a greater fragility of small caliber vessels of the deep cerebral venous circulation. Furthermore, the vulnerability of the periventricular regions, sites of the well-known periventricular leukomalacia (PVL) and intraventricular hemorrhage (IVH), whose multifactorial genesis included also prematurity, which presented 33% of DMVT newborns. [30, 31]
Comparing our DMVT case series to recent systematic review conducted by Pin et al, [8] our cohort differed for the presence of asymptomatic patients, for the higher incidence of uneventful pregnancies and the lack of family history of thromboembolism. However, a greater tendency towards neonatal thrombophilia was observed within our population, although this data was limited to a few patients for whom the outcome of thrombophilic tests were known (3 out of 9).
Limitations
Some limitations must be taken into consideration for the correct interpretation of this study. First of all, the small number of the sample did not allow us to identify statistically significant differences for most of the variables. Secondly, the retrospective nature of the study can lead to distortions and missed data. Consulting multiple sources, both paper and electronic data, as well as reviewing brain MRIs with an expert neuroradiologist was attempted in order to improve the quality of the research. Although the Department of Women's and Children's Health of Padua is a Hub third level neonatal stroke unit, our study is monocentric. Finally, we cannot ignore the paucity of data regarding long term follow up both clinical and radiological.