Vitamin K deficiency — Late onset intracranial haemorrhage

European Journal of Paediatric

ORIGINAL

Neurology

1998; 2: 199-203

ARTICLE

Vitamin K deficiency haemorrhage

- Late onset intracranial

NUR AYDINLI,’ ACOP CiTAK,’ MiNE SERPIL BAY~AL,~ MERAL ~ZMEN l

CALI$KAN,3

METiN

KARABbCijOhJ,2

‘Department of Paediatrics, Division of Paediatric Neurology; ‘Division of Emergency Medicine, Institute of Child Health; ‘Division of Paediatric Neurology, institute of Child Health, Istanbul University, Istanbul, Turkey

A retrospective study is presented of the clinical features and outcome of late onset haemorrhagic disease due to vitamin K deficiency in 11 babies who were admitted to the emergency or child neurology unit during a 4-year period Uanuary 1994-December 1997). The disease occurred in infants between 30 and 119 days of age (mean: 56+24 days). None of them received vitamin K after birth and all were breastfed. The presenting complaints were seizures (91%), drowsiness (82%), poor sucking (64%), vomiting (46%), fever (46%), pallor (46%), acute diarrhoea (27%), irritability and high-pitched cry (18%). On examination, tense or bulging fontanelle (73%), anisocoria (36%), weak neonatal reflexes (18%), cyanoses (18%) were the most frequent findings. The localizations of the intracranial haemorrhage were as follows: intracerebral (91%), subarachnoid (46%), subdural (27%), and intraventricular (27%). No fatality was observed. However, after a follow-up period ranging from 6 to 48 months (mean: 21 f13 months), only three (27%) infants remained neurologically normal. Seizure disorders (73%), severe psychomotor retardation (46%), cerebral palsy (46%) and microcephaly (46%) were observed in the remainder. Hydrocephalus developed in three (27%) babies but none of them required shunt replacement. The value is emphasized of vitamin K prophylaxis in the newborn to reduce the incidence of late onset intracranial haemorrhage and handicap in children. Keywords:

Late haemorrhagic

disease

of the newborn.

Intracranial

haemorrhage.

Introduction Three patterns of haemorrhages due to vitamin K deficiency in infancy are identified. Early haemorrhagic disease of the newborn (HDN) occurs at birth or within the first 24 hours of delivery generally due to antenatal drugs (coumarin derivatives and anticonvulsants) and is often life threatening. l Classic HDN accounts for gastrointestinal, nasal, skin and circumcision bleeding between 2nd and 7th days of life. Late HDN is almost exclusively confined to breastfed infants and occurs after the first month of life (or according to some after the first week)2J as an acute intracranial bleeding, sometimes as skin or

Received 4.3.98. Correspondence: Turkiye 1090-3798/98/020199+5

Revised Dr Nur

15.5.98. Aydinli,

$18.00

Accepted 195.98. Istanbul TIP Fakiiltesi,

Cocuk

Sa(jl@

gastrointestinal bleeding. 4-a Analysis of cumulative cases of late HDN from different countries showed that the majority of cases had not received vitamin K prophylaxis at birth.g This retrospective study clinical and laboratory describes presenting features, and outcome of late HDN.

Material

and methods

Between January 1994 and December 1997, 11 patients were admitted to the emergency or paediatric neurology unit with the diagnosis of late onset HDN. The hospital records of these

ve Hastallklan

Anabilim

Dab,

Celi$im

Niirolojisi

0

European Paediatric Neurology Society

1998

Bilim

Dab,

Capa

34390,

istanbul,

200 patients over 1 month of age were retrospectively evaluated. The detailed history included the mother’s general health and medical condition prior to delivery, the place of birth, and whether vitamin K was given at birth. Laboratory studies included complete blood count, partial thromboplastin time (FIT), prothrombin time (PT), and liver function tests. Vitamin K-dependent factor activities were performed when possible. After blood had been drawn for the baseline evaluation, vitamin K2 (l-2mg) was given intravenously. In babies with severe anaemia packed red blood cells and fresh frozen plasma were given. Routine spinal tap was not done because of the possibility of transtentorial hemiation and bleeding from the puncture site. Cranial ultrasonographies and computed tomographies (CT) were performed in all babies for the diagnosis and the evaluation of intracranial bleeding. Follow-up of all infants was carried-out monthly in the child neurology unit for the first 3 months, and every 3 months thereafter. The outcome was assessed by head circumference, neurologic examination, cranial CT and/or magnetic resonance imaging (MRI), electroencephalographic study, and Denver II developmental screening test.10

Original Table 1

Clinical

Seizure Drowsiness Poor sucking vomiting Fever Pallor Acute diarrhoea Irritability/high Examination

pitched

Our study encompassed eight males and three females aged between 30 and 119 days. The mean age of the population was 56&24 days. All babies were born at term of healthy mothers (four by caesarean section, seven by normal spontaneous delivery). Two of the deliveries were at home. The delivery histories were uneventful and the family histories were negative for any bleeding disorder. None of them had received vitamin K at birth. All babies were fully breastfed. The presenting complaints and examination findings are described in Table 1. The laboratory findings included haemoglobin 6.6 f 2.1 g/d1 (range: 4.5-8.1 g/dl), haematocrit 20 f3.7% (range: E-26%), platelets 404 000 + 202 233 x 109/litre (range: 261000-614 000 x log/&e), leukocytes 14 275 f 6 104 x log/hire (range: 6 800-28 000 x 109/litre). The PT and MT values of all babies were longer than control values. In six patients PT was longer than 60 seconds and MT longer than 120 seconds. Fibrinogen levels were 405.7& 128.9 g/litre (range: 295-626 g/litre). Liver function tests were normal. Vitamin Kdependent factor activities (factors II, VII, IX and X) measured in three infants were decreased. Cerebrospinal fluid was xanthochromic in one

N Aydinli

et al.

data at admission

Complaints”

cry

No.

%

10 9 7 5 5 5 3 2

90.9 81.2 63.6 45.5 45.5 45.5 27.3 18.2

8 4 4 2 2 1 of each

72.2 36.6 36.6 18.2 18.8 9.1

findings”

Tense or bulging fontanelle Bleeding from puncture sites Anisocoria Weak neonatal reflexes Cyanoses Strabismus, opisthotonos, dehydration, asymmetric clonus, increased deep tendon reflexes, decreased deep tendon reflexes, bleeding from mouth

“More than one finding could be present in one patient. Table 2

Neuroimaging

Findings”

Results

article:

Intracranial haemorrhage parenchymal” subarachnoid subdural intraventricular Infarction Oedematous change Ventricular compression Ventricular dilation Midline shift

findings No.

%

11 10 5 3 3 4 3 3 3 2

100 90.9 45.5 27.3 27.3 36.4 27.3 27.3 27.3 18.2

“More than one finding could be present in one patient. bFrontotemporoparietal in three, temporoparietal in two, frontal in two, temporal in two, parietal in one.

and bloody in three patients. Lumbar puncture was not done in other cases. The localization of intracranial haemorrhage and other intracranial pathologies are shown in Table 2. Representative CT scans are shown in Figs 1 and 2. In all cases there was more than one site of haemorrhage. No patient had other organ bleeding. Bleeding tendency was treated in all patients with 1-2 mg vitamin K2 intravenously. Fresh frozen plasma was given to seven and packed red blood cells to six patients. One infant needed endotracheal intubation with mechanical ventilation. The PT and MT were corrected 6-12 hours after administration of vitamin K. An anticonvulsive therapy with phenobarbital or diphenylhydantoine

Original

article:

Late haemorrhagic

disease

and intracranial

haemorrhage

201

Fig. 2. Axial CT scan showing intracranial haemorrhage in a J-week-old patient. Right temporal and occipital parenchymal haemorrhage, ventricular compression, left ventricular dilation, midline shift, and large infarction at right parietal and left temporal area.

tension. In two patients intracerebral haematomas of temporoparietal localization were surgically removed. Acetazolamide therapy (30-50 mg/kg/ day) was used in four patients during the acute phase. Three patients developed hydrocephalus but none of them required shunt replacement. All patients were followed-up for a mean period of 21.1+ 13.4 months (range 648 months). Follow-up findings are presented in Table 3. Three children remained normal at the end of 1 year. Eight patients were still prescribed anticonvulsive therapy, and three suffered from West’s syndrome. Five children (46%) had cerebral palsy (two hemiparetic, and three tetraparetic), and five (46%) were microcephalic. Visual disturbances Table 3 Fig.

CT scans demonstrating evolution of 1. Axial intracranial haemorrhage in a T-week-old patient. (a) Left frontal parenchymal haemorrhage surrounded by oedema, left parietal infarction. (b) After 19 days: left frontal haemorrhage resolves, oedema is still present, mild ventricular enlargement is remarked.

was instituted in all patients. In three patients seizures could be controlled with phenobarbital plus diphenylhydantoine. All were prescribed anticonvulsive therapy at discharge. Dexamethasone and mannitol were administered in seven patients because of the suspicion of intracranial hyper-

Follow-up

Findings” Normal Seizure disorders West’s syndrome Severe psychomotor retardation Cerebral palsy spastic tetraparesis spastic hemiparesis Microcephaly Hydrocephalus Strabismus Blindness

findings

No. 3 8 3 5 5 3 2 5 3 3 1

“More than one finding could be present in one patient.

% 27.3 72.8 27.3 45.5 45.5 27.3 18.2 45.5 27.3 27.3 9.1

Original

202 developed observed.

in four (36%) patients.

No mortality

was

Discussion The incidence of vitamin K deficiency bleeding in early infancy was calculated to be decreased in Germany from l/14 000 to l/70 000 with single oral prophylaxis and to l/420 000 with single parenteral vitamin K prophylaxis.g The occurrence of late HDN in Sweden was l/19 570 among newborns who had received vitamin K orally at birth.11 No figure was available for the incidence of late HDN among babies with or without vitamin K prophylaxis in Turkey. The content of vitamin K in human milk is significantly less than in cow’s milk or in formula.2JJ2 The combination of low to absent levels of vitamin K at birth, no supplementation after birth, and the reduced availability from breast-feeding explain the likely pathogenesis of vitamin K-dependent bleeding after 1 month of life. Late onset HDN may also be secondary to warfarin exposure or to an underlying disorder such as biliary atresia, cystic fibrosis, alpha-l antitrypsin deficiency, hepatitis, abetalipoproteinaemia, coeliac disease or prolonged diarrhoea.zn4*6*8J3 Although we were not able to measure vitamin K dependent factors of all patients, diagnosis was confirmed since none of the patients received vitamin K before admission and prolongation of PT and PTT was corrected after administration of vitamin K in all cases. All babies were fully breastfed and no evidence of impaired vitamin K absorption or liver disease were identified during the follow-up period. Three patients had a short episode of diarrhoea which was corrected within 1 week. Therefore prolonged diarrhoea was not a cause among our patients. In the report of Ekelund” eight infants of 17 with late HDN suffered from intracranial haemorrhage. All had received 1-2 mg vitamin K orally at birth and were breastfed. Liver disease or laboratory signs of impaired liver function were demonstrated in six of these patients. Liver disease was not identified among our patients as in Ekelund’s study. This difference may be due to the vitamin K administration rate at birth. If vitamin K prophylaxis can be applied to all newborns, other causes of late HDN will relatively predominate. Intracranial haemorrhage due to vitamin K deficiency is the most common form of late HDN and accounts for approximately 50% of the bleeding episodes at presentation.4gg It is reported that

article:

N Aydinli

et al.

subarachnoid haemorrhage is the major finding in 91% of the babies but it is always combined with other lesions such as subdural haemorrhage in 38%, or parenchymal haemorrhage in 31% of the patients.12 In our study the most common site of haemorrhage was intracerebral (48%) followed by subarachnoid (24%), subdural (14%) and intraventricular (14%). All patients with subarachnoid haemorrhage suffered also from another site of bleeding within the cranium. We have been able to observe our patients for 648 months. Only three (27%) infants remained physically and neurologically normal. One of them had subarachnoid haemorrhage and frontal subdural haematoma, the second had subarachnoid haemorrhage and left parietal small haematoma, and the third had right frontal small haematoma. All the other infants (73%) had microcephaly or some degree of handicap. Hydrocephalus was observed in three children with intraventricular haemorrhage but subsided without shunting. All children with refractory epilepsy have suffered from large intracerebral haematomas and survived with major cerebral damage. Similar results on sequelae after HDN were reported in the literature. In the study of Chaou et a1.12only one infant of 32 with late HDN developed normally. All the other babies have had microcephaly (90%), cerebral palsy (24%), severe psychomotor retardation (17%), or epilepsy (14%). The mortality rate was 9.4%. In the study of von Kries,g the sequelae of central nervous system haemorrhage observed in seven infants were as follows: epilepsy (29%), hemiparesis (14%), hydrocephalus (14%), and death (14%). Motohara et ~1.~described ten infants with late HDN and intracranial haemorrhage. Three (30%) did well while three (30%) died, two (20%) survived with cerebral palsy and one (10%) with hydrocephalus. There are still controversies concerning the best way of providing effective prophylaxis. Single oral or intramuscular administration of 1 mg vitamin Kr postnatally may not offer complete protection against late biochemical vitamin K deficiency.14 According to the study by Comelissen et al.115a daily low oral dose of 25 pg vitamin Ki following an initial oral dose of 1 mg after birth for exclusively breastfed infants may be as effective as parenteral vitamin K prophylaxis. Since intramuscular administration of vitamin K does not provide complete protection from late HDN, it is reasonable to consider administering further doses of vitamin K to infants at high risk of HDN: those who fail to thrive or have liver disease or longterm diarrhoea.4r6J6J7

Original

article:

Late haemorrhagic

disease

and intracranial

Physicians should also consider the possibility of vitamin K deficiency at an early stage in the evaluation of any bleeding which occurs during the first 6 months of life. Although intramuscular administration of vitamin K has been reported to be associated with an increased risk of childhood cancer, subsequent extensive studies have yielded no evidence of any such relationship.*J8 Ln Turkey 1.5 million babies are born each year; 60% of the births take place in hospital. Although the Turkish Ministry of Health recommends parenteral vitamin K prophylaxis for newborns, still some babies born in hospital do not receive vitamin K. On the other hand vitamin K is not administered at home deliveries, half of which are administered by traditional birth attendants. Therefore the families of the babies under 3 months of age, at the first admission to the hospital for any reason, should be asked for details of vitamin K prophylaxis at birth, and in all uncertain cases babies should receive 1 mg vitamin K1 intramuscularly. This study once again emphasizes the value of vitamin K prophylaxis in the newborn to reduce the incidence of late onset intracranial haemorrhage and handicap in children.

6

Soci&! Franqaise de PCdiatrie Con-&! de Nutrition. La vitamine K en Pediatric. Recommendations de prescriptions. Arch Fr PPdiatr 1991; 48: 57-59.

7 Volpe JJ. Intracranial hemorrhage: subdural, primary subarachnoid, intracerebellar, intraventricular (term infant), and miscellaneous. In: Volpe JJ (ed) Neurology ofthe Newborn (3rd edn). Philadelphia: WB Saunders Co 1995: 373402. 8 Urvoas Leblanc rhage disease.

E, Pariente D, Rousset A, de Victor D, A. Ultrasound diagnosis of thymic hemorin an infant with late-onset hemorrhagic Pediatr Radio1 1994; 24: 96-97.

9 Von Kries R, Gabel U. Vitamin K prophylaxis and vitamin K deficiency bleeding in early infancy. Acta Paediatr 1992; 81: 655-657. 10

Anlar B, Yalaz K. Denver II Test Manual: Adaptation and Standardisation for Turkish Children. Ankara: Metaksan, 1996.

11

Ekelund H. Late haemorrhagic disease in Sweden 1987-89. Acta Paediatr Stand 1991; 80: 966-968.

12

Chaou W-T, Chou M-L, Eitzman DV. Intracranial hemorrhage and vitamin K deficiency in early infancy. 1 Pediatr 1984; 105: 880-884.

13

Soylu A, Aslan Y, Sari A, Erduran E. Intracerebral hemorrhage: a rare late manifestation of vitamin-K deficiency in a breastfed infant. A case report. Turk J Pediatr 1974; 39: 265-269.

14

Comelissen EA, KollCe LA, De Abreu RA et al. Effects of oral and intramuscular vitamin K prophylaxis on vitamin Kl, PIVKA-II, and clotting factors in breast fed infants. Arch Dis Child 1992; 67: 1250-1254.

15

Comelissen M, von Kries R, Loughnan P, Schubiger G. Prevention of vitamin K deficiency bleeding: efficacy of different multiple oral dose schedules of vitamin K. Eur J Pediatr 1997; 156: 126-130.

16

McMillan DD. Administration of vitamin newborns: implications and recommendations. Med Assoc J 1996; 154: 347-349.

17

Canadian Paediatric Society, Committee. The use of vitamin CMA] 1988; 139: 127-130.

18

Brousson MA, Klein MC. Controversies surrounding the administration of vitamin K to newborns: review. Can Med Assoc J 1996; 154: 307-315.

References Astedt B. Antenatal drugs affecting vitamin K status of the fetus and the newborn. Senzin Throtnb Hemost 1995; 21: 364-370. Huysman MWA, Sauer PJJ. The vitamin K controversy. Cwr Opin Pediatr 1994; 6: 129-134. Kiegman CM. The fetus and the neonatal infant. In: Nelson WE (ed) Textbook of Pediatrics (15th edn). Philadelphia: WB Saunders Co 1996: 431505. Lane PA, Hafhaway WE. Vitamin K in infancy. 1 Pedintr 1985; 106: 351-359. Motohara K, Matsukura M, Matsuda I et al. Severe vitamin K deficiency in breast-fed infants. J Pedintr 1984; 105: 943-945.

203

haemorrhage

K to Can

Fetus and Newborn K in perinatal period.

a