Long‐term follow‐up of neonatal intracranial haemorrhage in children with severe haemophilia

Abstract

Intracranial haemorrhage (ICH) in neonates with severe haemophilia is a potentially life-threatening complication and has been reported to occur in about 2–4% of boys with severe haemophilia; around 44–60 times higher than that expected in a non-haemophilia population.1, 2 ICH requires intensive replacement therapy with factor concentrates, which has been identified as a risk factor for inhibitor development in patients with haemophilia A (HA).3 Death and neurological sequelae (motor function problems, seizures and developmental delay) are the most feared complications and have been reported to affect 50–60% of children with ICH.4 The aim of this paper was to study the incidence and symptoms of ICH in neonates with severe haemophilia and the impact of early diagnosis and treatment of ICH on long-term outcomes. Thirty-three haemophilia centres from the ICH study group, 28 from the PedNet Haemophilia Research Foundation or the International Network of Pediatric Hemophilia and five other haemophilia centres participated in this retrospective review by including all children with severe HA or haemophilia B (FVIII/IX < 1%), born between 2006 and 2015. The study recorded ICH in the neonatal period (first 28 days of life). Data were available for 450 neonates. Ethical approval and informed consent were obtained. The case report form registered patients’ type of haemophilia, mode of delivery, ICH, and in cases of ICH, neurological signs and symptoms at presentation, treatment and sequelae. ICH had to be radiologically diagnosed. All analyses were performed using Statistical Package for the Social Sciences (SPSS Inc., version 26.0, IBM Corp., Armonk, NY, USA) and R (version 3.4.3, R Foundation for Statistical Computing, Vienna, Austria); P values of <0·05 were considered significant. Statistical comparisons between different groups were made using Fisher's exact test and incidences by the Poisson test. Nine out of the 450 (364 severe HA and 86 severe HB) had an ICH in the neonatal period, resulting in an incidence of 2%. Table I summarizes data on the nine patients with ICH during their neonatal period. All nine ICH were diagnosed in the first 11 days of life (median age for imaging-confirmed diagnosis was 4·0 days). The most frequent symptoms (alone or combined) were lethargy in six and vomiting in three patients. Other symptoms reported in one patient each were: non-responsiveness, crying, torticollis, seizures and cephalic haematoma. One child (patient 7, Table I) with no known family history of haemophilia presented with lethargy and feeding problems and was diagnosed with severe HA the day after onset of symptoms (>24 h). The boy died after two days of FVIII treatment and neurosurgical resection. The median follow-up time after neonatal ICH for the surviving eight boys was 7·1 (range 4·1–7·6) years. In four newborns, patients 3, 4, 5 and 9, early diagnosis of haemophilia was made before or on the day of the ICH, and two had a known family history of haemophilia. These four neonates received immediate treatment with factor concentrate within 24 h after onset of symptoms and diagnosis of ICH by imaging the same day and showed no neurological sequelae in long-term follow-up. Five patients, patients 1, 2, 6, 7 and 8, did not receive treatment within 24 h of showing symptoms of an ICH. Of these five patients, one died (patient 7), one showed no long-term neurological symptoms (patient 2) while three showed neurological sequelae: motor and intellectual problems and visual problems (patient 1, 6 and 8). Neonates with severe haemophilia who did not receive immediate treatment within less than 24 h after the onset of symptoms and diagnosis of ICH were significantly more likely to develop neurological impairments or death (4/5, 80%) than those receiving immediate treatment (0/4, 0%; P = 0·047). To be able to compare neonatal ICH incidences to ICH incidences later in life, the 450 neonatal months were added up to 34·52 patient years, resulting in an incidence of 0·26 ICH/patient year. In a direct comparison of the data from the published non-neonatal part of our ICH study,4 the relative risk of ICH in neonates was found being 11·2 times higher than in our 1–12-month-old children (0·023 ICH/patient year), 7·5 times higher than in our 1–2-year-old age group (0·034 ICH/patient year) and 15·4 times higher than in the all children (<18 year) age group (0·017 ICH/patient year). These differences were all statistically significant (P < 0·001). Our results confirm neonatal ICH as a major contributor to morbidity and mortality in children with severe haemophilia. The incidence in the present study of 2% is concordant with earlier studies.5 Mortality in our study was 1/9 (12%), data from a single centre showed a 22% (4/18) mortality rate after neonatal ICH with haemophilia.6 Both the incidence and mortality rate might be underestimated. The neonate might not be diagnosed as having haemophilia if there is no family history of haemophilia, symptoms like lethargy and vomiting are non-specific and not all cases of ICH are detected. In an American study from 1999, four of eight ICH were clinically silent but had haemorrhages seen on MRI in a yearly follow-up.7 Diagnosing haemophilia can be challenging since approximately 50% of the patients are sporadic cases without a family history of haemophilia.8, 9 Despite the low numbers of subjects in our cohort, the study illustrates the need of immediate treatment with factor replacement for good long-term outcomes without neurological impairment. However, minor neurological impairments may have been missed. In studies re-evaluating haemophilia patients with ICH with in-depth testing, more patients than expected had mild cognitive dysfunction or impairments (up to 56%).6, 10 The comparison of the data from the published non-neonatal part of our ICH study to the neonatal part shows how vulnerable the neonatal period is, the relative risk being 11·2 times higher than in children between 1 and 12 months of age and 15·4 times higher than in all children. Diagnosis of haemophilia and treatment should be as early as possible to prevent neurological impairment. We highly appreciate the participation of all subjects in our study. The use of data from the PedNet Registry belonging to the PedNet Haemophilia Research Foundation and all help from its personnel is greatly acknowledged. A list of contributors is given in the Appendix. The study was funded by the Arosenius Foundation Research Grant (founded by the Swedish Haemophilia Patient Association, N.G. Andersson), by the Swedish Research Council (grant number: 2015-02957, R. Ljung) and “Bayer-Award” (R. Ljung). All authors participated in data collection. RL and NGA are responsible for the concept and study design and analysis and of the data. NGA wrote the manuscript. NGA, RW, MC, SC, RK, RL, AM, SR and RL took part in the interpretation of the data. All the authors reviewed the manuscript and gave final approval to the manuscript. All authors declare to have no potential conflicts of interest regarding the present work. G. Auerswald, Bremen, Germany, C. Barnes, Melbourne, Australia, E. Chalmers, Glasgow, UK, H. Chambost, Marseille, France, N. Clausen, Aarhus, Denmark, A.L. Dunn, Columbus, OH, USA, C. Escuriola Ettinghausen, Frankfurt, Germany, K. Fischer, Utrecht, The Netherlands, K. Fijnvandraat, Amsterdam, The Netherlands, C. van Geet, Leuven, Belgium, M. Hoffmann, Copenhagen, Denmark, K. Kavakli, Izmir, Turkey, G. Kenet, Tel Aviv, Israel, C. Königs, Frankfurt, Germany, K. Kurnik, Munich, Germany, M. Manco-Johnson, Denver, CO, USA, M.E. Mancuso, Milan, Italy, C. Molinari, Genua, Italy, W. Muntean, Graz, Switzerland, B. Nolan, Dublin, Ireland, R. Perez Garrido, Valencia, Spain, E. Platokouki, Athens, Greece, A. Rafowicz, Paris, France, E. Santagostino, Milan, Italy, A. Shapiro, Indianapolis, IN, USA, A. Thomas, Edinburgh, UK, M. Williams, Birmingham, UK.

Topics & Concepts