Advertisement
Original Article| Volume 50, ISSUE 6, P564-568, June 2014

Download started.

Ok

Rise in Late Onset Vitamin K Deficiency Bleeding in Young Infants Because of Omission or Refusal of Prophylaxis at Birth

      Abstract

      Background

      Newborns are at risk for vitamin K deficiency and subsequent bleeding unless supplemented at birth. Vitamin K deficiency bleeding is an acquired coagulopathy in newborn infants because of accumulation of inactive vitamin K–dependent coagulation factors, which leads to an increased bleeding tendency. Supplementation of vitamin K at birth has been recommended in the United States since 1961 and successfully reduced the risk of major bleeding. Refusal or omission of vitamin K prophylaxis is increasing and puts newborn infants at risk for life-threatening bleeding.

      Patients

      Over an eight month period, we encountered seven infants with confirmed vitamin K deficiency; five of these patients developed vitamin K deficiency bleeding.

      Results

      The mean age of the seven infants with vitamin K deficiency was 10.3 weeks (range, 7-20 weeks); manifestations ranged from overt bleeding to vomiting, poor feeding, and lethargy. None of the infants had received vitamin K at birth, and all were found to have profound derangement of coagulation parameters, which corrected rapidly with administration of vitamin K in IV or intramuscular form. Four of the seven infants had intracranial hemorrhage; two of these infants required urgent neurosurgical intervention.

      Conclusion

      Supplementation of vitamin K at birth for all newborns prevents major hemorrhagic complications, such as intracranial bleeding, due to vitamin K deficiency. Parental refusal of vitamin K is increasingly common. It is critical that health care providers and the public be made aware of the varied presentation of this preventable acquired coagulopathy.

      Keywords

      See related editorial on page 545.

      Introduction

      Vitamin K deficiency bleeding (VKDB), formerly known as hemorrhagic disease of the newborn, is an acquired coagulopathy in infants because of an inability to activate the vitamin K–dependent coagulation factors (II, VII, IX, and X) because of a relative deficiency in available vitamin K. VKDB manifestations range from mild “warning bleeds” (epistaxis, umbilical stump, or gastrointestinal bleeding) to severe (intracranial hemorrhage [ICH]). VKDB can be classified by the age of onset: early (<24 hours after birth), classical (2-14 days), and late (2-12 weeks). Newborn infants are at risk for VKDB for the following reasons: (1) reduced bioavailability because of poor placental transfer of vitamin K and the relatively short half-life of the K1 liver stores; (2) reduced vitamin K content in breast milk compared with fortified cow's milk–based formula; and (3) reduced production of vitamin K because of immature or altered gut flora. Because dietary intake is an infant's main source of vitamin K, exclusively breastfed infants have a higher risk for VKDB than formula-fed infants. In infants, the plasma concentrations of all vitamin K–dependent clotting factors are 40-60% of the adult values and slowly rise during infancy but can take up to 90 days to completely normalize even with adequate vitamin K stores.
      • Williams M.D.
      • Chalmers E.A.
      • Gibson B.E.
      The investigation and management of neonatal haemostasis and thrombosis.
      VKDB was first described by Townsend in 1894.
      • Townsend C.W.
      The haemorrhagic disease of the newborn.
      In a definitive study published in 1944, prophylactic vitamin K given at birth was shown to reduce VKDB-associated death by greater than fivefold in the first 2 weeks of life.
      • Lehmann J.
      Vitamin K as a prophylactic in 13,000 infants.
      With evidence mounting over the next two decades, the American Academy of Pediatrics
      American Academy of PediatricsCommittee on Nutrition
      Vitamin K compounds and the water-soluble analogues: use in therapy and prophylaxis in pediatrics.
      issued a statement in 1961 recommending that a single dose of vitamin K be given to all neonates shortly after birth, either 0.5-1 mg intramuscular (IM) or 1-2 mg oral. In the era of prophylaxis, VKDB has become rare, with most reported cases being classical or late onset and occurring in infants who either did not receive adequate vitamin K prophylaxis at birth and are exclusively breastfed or who had an undiagnosed malabsorptive or hepatobiliary disorder. Early VKDB is mainly because of the effect of maternal medications and can be effectively prevented by vitamin K at birth; when no prophylaxis is given, rate of early VKDB is 6-12%.
      • Sutor A.H.
      • von Kries R.
      • Cornelissen E.A.
      • McNinch A.W.
      • Andrew M.
      Vitamin K deficiency bleeding (VKDB) in infancy: on behalf of the ISTH Pediatric/Perinatal Subcommittee.
      Without prophylaxis, the incidence of classical VKDB is as high as 1.7% of live births
      American Academy of PediatricsCommittee on Fetus and Newborn
      Controversies concerning vitamin K and the newborn.
      , whereas the incidence of late VKDB ranges from 4.4 to 7.2 per 100,000 live births.
      Centers for Disease Control and Prevention (CDC)
      Notes from the field: late vitamin K deficiency bleeding in infants whose parents declined vitamin K prophylaxis–Tennessee, 2013.
      When IM vitamin K prophylaxis is given at birth, the rate of late VKDB ranges from 0.24 to 3.2 cases per 100,000 live births.
      • von Kries R.
      • Hachmeister A.
      • Göbel U.
      Oral mixed micellar vitamin K for prevention of late vitamin K deficiency bleeding.
      • Ijland M.M.
      • Pereira R.R.
      • Cornelissen E.A.
      Incidence of late vitamin K deficiency bleeding in newborns in the Netherlands in 2005: evaluation of the current guideline.
      • Takahashi D.
      • Shirahata A.
      • Itoh S.
      • Takahashi Y.
      • Nishiguchi T.
      • Matsuda Y.
      Vitamin K prophylaxis and late vitamin K deficiency bleeding in infants: fifth nationwide survey in Japan.
      • Busfield A.
      • Samuel R.
      • McNinch A.
      • Tripp J.H.
      Vitamin K deficiency bleeding after NICE guidance and withdrawal of Konakion Neonatal: British Paediatric Surveillance Unit study, 2006-2008.
      • Laubscher B.
      • Bänziger O.
      • Schubiger G.
      Prevention of vitamin K deficiency bleeding with three oral mixed micellar phylloquinone doses: results of a 6-year (2005-2011) surveillance in Switzerland.
      ICH occurs frequently in cases of late VKDB and can lead to significant morbidity and mortality. In a pooled analysis of 131 cases, 63% of late VKDB presented with ICH, with 14% mortality and 40% long-term neurological morbidity among surviving infants.
      • Loughnan P.M.
      • McDougall P.N.
      Epidemiology of late onset haemorrhagic disease: a pooled data analysis.
      As part of a Centers for Disease Control investigation that followed a recent rise in the number of infants presenting with VKDB, we determined that 28% (61 of 218) of the parents of children born at local private birthing centers in Tennessee refused vitamin K prophylaxis.
      Centers for Disease Control and Prevention (CDC)
      Notes from the field: late vitamin K deficiency bleeding in infants whose parents declined vitamin K prophylaxis–Tennessee, 2013.
      Over an eight month period after improvement in our education of parents of the risks of declining vitamin K prophylaxis, we demonstrated that 3.4% (104 of 3080) of local parents (Vanderbilt University Medical Center) declined oral or IM vitamin K prophylaxis. There is concern regarding an apparent rise in the rate of parental decline of vitamin K prophylaxis in the middle Tennessee area. However, no state or national mechanism exists to track decline rates at this time. We sought to describe encountered cases of late VKDB in an effort to educate health care providers on the variable presentation of this potentially life-threatening disorder.

      Patients

      We present five cases of late VKDB occurring at a single tertiary care children's hospital between February and September 2013 and two additional infants noted to have laboratories consistent with severe vitamin K deficiency but no bleeding (one asymptomatic infant evaluated as twin had ICH related to VKDB and one infant with acholic stools undergoing liver disorder evaluation). The details of these children are described with approval by the institutional review board at Vanderbilt University. The demographics, presenting signs, treatment, and outcomes of the seven infants are shown in Table 1. Of the children that presented with vitamin K deficiency, the mean age was 10.3 weeks (range, 7-20 weeks). Presenting signs ranged from overt bleeding to vomiting, poor feeding, altered mental status, and hypothermia. All infants were treated with vitamin K on presentation; three also received fresh frozen plasma (FFP). One child was given recombinant factor VII (rFVII) because of severe intracranial bleeding and need for emergent life-saving neurosurgical intervention.
      Table 1Demographics and Clinical Features of Seven Children With Vitamin K Deficiency With or Without Bleeding
      PatientAge at DiagnosisGenderBirth HistoryFeedingPresenting SignsBleeding LocationHead CT FindingsAdditional TreatmentOutcome
      15 moMaleBorn at home and no vitamin K or circumcisionExclusively breastfedEmesis, bruising, somnolence, and poor feedingIntracranialMultiple subdural hematomas and midline shiftNeurosurgical evacuation of hematomaNeurologically intact 3 months post-ICH and undergoing workup for hepatobiliary disease
      28 wkFemaleBorn in hospital and no vitamin KExclusively breastfedVomiting, fever, and lethargyIntracranialLeft frontoparietal and bilateral frontal subarachnoid hemorrhagesAnticonvulsant for recurrent seizuresMild neurological delays 5 months post-ICH and PSOM score 0.5 (mild)
      33 moFemaleBorn at home and no vitamin KExclusively breastfedBloody stoolsGastrointestinalNot doneNoneNo long-term effects to date
      46 wkMaleBorn in hospital and no vitamin KExclusively breastfedFussiness, somnolence, and poor feedingIntracranialParenchymal hemorrhage filling the entire left hemisphere and effacement of left lateral ventricle, midline shiftNeurosurgical evacuation of hematoma and anticonvulsant for recurrent seizuresRight hemiparesis and cognitive delays at 3 mo post-ICH and PSOM score 2.5 (poor, severe deficits)
      57 wkMale (twin)Born in hospital and no vitamin KExclusively breastfedFussiness, emesis, pallor, and poor feedingIntracranialMultifocal acute subdural, subarachnoid, and parenchymal hemorrhagesErythrocyte transfusions × 3 and anticonvulsant for recurrent seizuresModerate neurological delays, 4 mo post-ICH, and PSOM score 1.5 (moderate deficits)
      67 wkMale (twin)Born in hospital and no vitamin KExclusively breastfedAsymptomatic twinNoneNegativeNoneNo long-term effects to date
      712 wkFemaleBorn in birthing center and no vitamin KExclusively breastfedVomiting and acholic stoolsNoneNot doneNoneExcised choledochal cyst
      Abbreviations:
      CT = Computed tomography
      ICH = Intracranial hemorrhage
      PSOM = Pediatric stroke outcome measurement
      The laboratory data for the seven infants are shown in Table 2. Brain imaging of the four children with ICH due to VKDB is shown in the Figure. All of the cases were noted to have severe prolongation of prothrombin time (PT) and partial thromboplastin time (PTT) and corrected as early as 1.5 hours after vitamin K administration. Two of the children required emergent neurosurgical intervention. A description of one of the cases is as follows: A 6-week-old boy presented with a 1-day history of fussiness and decreased alertness without fever. Parents noted pallor, decreased breastfeeding and that his anterior fontanelle was “full and hard.” On presentation to our emergency room, he was bradycardic and had several brief episodes of apnea. A full sepsis evaluation was initiated and because of the neurological changes and presence of blood in the cerebrospinal fluid sample, a head computed tomography (CT) was performed demonstrating a large ICH (Figure panel C) with transfalcine and transtentorial herniation. CT angiography showed no evidence of vascular malformation. After emergent correction of coagulopathy with IV vitamin K and FFP, he was taken to the operating room for emergent hematoma evacuation. Because of an inability to obtain post-vitamin K laboratories (PT and PTT) in a timely manner, low dose rFVII (20mcg/kg/dose) was given to ensure adequate surgical hemostasis. After hematoma evacuation, he had short, repetitive focal seizures requiring two IV anticonvulsants to halt them. He has remained seizure-free on oral anticonvulsants since that time.
      Table 2Laboratory Findings of Cases of Vitamin K Deficiency
      PatientInitial Laboratories: PT (s)/INR/PTT (s)Treatments Given
      Expressed as % of normal values.
      Follow-Up Laboratories: PT (s)/INR/PTT (s)Time Between Vitamin K and Follow-Up LaboratoriesActivity of Vitamin K Dependent Factors
      Expressed as % of normal values.
      IIVIIIXX
      160.1/7.6/77.8Vitamin K 2 mg IV and FFP 20 mL/kg16.2/1.3/37.01.5 hr4158
      2>100/>10/169.7Vitamin K 1 mg SQ and FFP 20 mL/kg16.4/1.4/36.47 hr
      3>100/>10/NAVitamin K 1 mg IV13.3/1.1/28.18 hr3633
      4>100/>10/104.3Vitamin K 2 mg IV; FFP 20 ml/kg; and rFVII14.6/1.2/37.22 hr2<62<3
      5>100/>10/>200Vitamin K 2 mg IV14.6/1.2/30.94.5 hr<3<51<3
      6>100/>10/99.5Vitamin K 2 mg SQ15.5/1.3/31.05 hr
      7>99.9/>10/82.4Vitamin K 2 mg IV41.8 or 14.8/4.6 or 1.2/64.5 or 35.83 hr/8 hr92524
      Abbreviations:
      NA = Not applicable
      SQ = Subcutaneous
      FFP = Fresh frozen plasma
      rFVII = Recombinant factor VII
      Expressed as % of normal values.
      Figure thumbnail gr1
      FigureIntracranial hemorrhages in children with vitamin K dependent bleeding; panels A through D show the key image for each of four children with intracranial hemorrhage. (A) Patient 1: large subdural hematoma and left occipital-parietal region with associated midline shift. (B) Patient 2: parenchymal hematoma in the left parietal region with layering of blood as observed in ICH due to coagulopathy. (C) Patient 4: massive ICH of the left hemisphere with midline shift. (D) Patient 5: subdural hematoma (dashed arrow), right occipital region, and small left frontal parenchymal hematoma (solid arrow).

      Discussion

      These cases represent a substantial increase in the number of infants diagnosed with late onset VKDB in our region. We conducted a review of Tennessee hospital discharge data, searching for International Classification of Diseases-9 codes related to vitamin K deficiency and VKDB, and found six probable cases in 2007 (one with ICH), two cases in 2008, zero case in 2009, three cases in 2010, and one case in 2011; however, on further review, none of these met criteria for classical or late VKDB. The most common additional International Classification of Diseases-9 codes associated with these cases were sepsis, apnea, failure to thrive, biliary atresia or other biliary disorder, seizure, prematurity, kidney anomalies, multiple congenital anomalies, and anemia. The cases presented are mainly from otherwise healthy infants, except two infants with underlying hepatobiliary disease.
      Concerns about unknown risks of IM vitamin K were raised in the early 1990s, when a case-control study suggested a link with childhood cancer; multiple subsequent studies have failed to corroborate this finding.
      American Academy of PediatricsCommittee on Fetus and Newborn
      Controversies concerning vitamin K and the newborn.
      In response to these concerns, oral preparations of vitamin K were developed and studied in many countries worldwide but not in the United States. A single dose of oral vitamin K at birth can effectively prevent early and most cases of classic VKDB, but the failure rate of oral vitamin K for the prevention of late VKDB was unacceptably high. Regimens of repeated oral dosing have slightly higher efficacy but are still significantly less effective than a single dose of IM vitamin K at birth,
      • Busfield A.
      • McNinch A.
      • Tripp J.
      Neonatal vitamin K prophylaxis in Great Britain and Ireland: the impact of perceived risk and product licensing on effectiveness.
      with rates of late VKDB between 1.4 and 6.4 per 100,000 live births when a single oral dose is given at birth.
      American Academy of PediatricsCommittee on Fetus and Newborn
      Controversies concerning vitamin K and the newborn.
      There has been a recent increase in the rate of parental decline of vitamin K prophylaxis at our institution and in the surrounding communities. None of the infants presented here received vitamin K prophylaxis at birth, either because of parental decline or omission at home birth. Some reasons for parental decline include concern over an increased risk for leukemia, concern about the use of a synthetic medication which is perceived as a toxin, and impression that vitamin K administration is not natural and is unnecessary in uncomplicated births, with the last reason being the most common in our series. In spite of these concerns, only one family in this series reported awareness of the potentially life-threatening nature of VKDB.
      The presentation of these described cases of VKDB varied as shown in Table 1. It is important for clinicians to ascertain whether vitamin K was received at birth for any infants presenting in the first 6 months of life with bleeding, bruising, lethargy, or fussiness and consider laboratory testing of coagulation parameters and rapid neuroimaging for infants with unexplained neurological findings and uncertain vitamin K status. Head CT was utilized in three of these infants because of declining neurological status in the emergency department. Infants 5 and 6 (Table 1) were not critically ill at presentation, but head CT was performed at an outside hospital before transfer to our facility. At our tertiary pediatric medical center, we are fortunate to have a short protocol magnetic resonance imaging (MRI) of the brain to rapidly screen for conditions such as ischemic stroke and brain hemorrhage. This 10-minute MRI protocol is preferred at our center because of the increased sensitivity of MRI for cerebral ischemia and concerns about radiation exposure even with pediatric head CT protocols. However, when a child is unstable or MRI is not available, head CT is performed. Of note, minor bleeding often precedes an infant's presentation with ICH from VKDB by days to weeks.
      • Busfield A.
      • Samuel R.
      • McNinch A.
      • Tripp J.H.
      Vitamin K deficiency bleeding after NICE guidance and withdrawal of Konakion Neonatal: British Paediatric Surveillance Unit study, 2006-2008.
      • Loughnan P.M.
      • McDougall P.N.
      Epidemiology of late onset haemorrhagic disease: a pooled data analysis.
      Prolongation of the PT and PTT is always found in vitamin K deficiency in infants; evaluation of the platelet count and fibrinogen, which should be normal in VKDB, helps exclude other causes of bleeding in infancy. The coagulopathy associated with vitamin K deficiency has been shown to rapidly correct with administration of vitamin K, normalizing within 2-3 hours with rapid increase in activation of vitamin K–dependent coagulation factors within 20-30 minutes.
      • Sutor A.H.
      • Kunzer W.
      Time interval between vitamin K administration and effective hemostasis.
      This rapid correction of coagulopathy with vitamin K administration was confirmed by our cases (Table 2).
      Because of the potential consequences of ICH, infants strongly suspected to have VKDB (based on reported history and signs of severe illness) should promptly be given 1 mg of IV vitamin K (phytonadione), even before laboratory results are available.
      • Williams M.D.
      • Chalmers E.A.
      • Gibson B.E.
      The investigation and management of neonatal haemostasis and thrombosis.
      • Sutor A.H.
      • Kunzer W.
      Time interval between vitamin K administration and effective hemostasis.
      In cases where there is life-threatening hemorrhage or surgical intervention is urgently needed, other treatment may be considered, including FFP, whole blood, or rFVII, as clinical situation warrants.
      • Sutor A.H.
      • Kunzer W.
      Time interval between vitamin K administration and effective hemostasis.
      We would suggest that the use of rFVII is not indicated in the vast majority of situations because of the rapid correction of the acquired coagulopathy with vitamin K administration.

      Conclusions

      It is critical for health care providers to continue to advocate for the use of IM vitamin K prophylaxis at birth. There is concern regarding an apparent rise in the rate of parental decline of vitamin K prophylaxis in the middle Tennessee area. However, there is no state or national mechanism to track decline rates. We were able to establish a rise in the rate of VKDB in newborns in the middle Tennessee area in part because of collaboration with the CDC. The authors feel that education of families by health care providers and a concerted governmental effort to track vitamin K refusal and VKBD are needed. Being aware of the varied presentation of VKDB in infancy is also important and primary care providers should inquire whether parents declined it when presented with an ill appearing infant in the first 6 months of life. Regardless of how quickly vitamin K is given at presentation with classic and/or late VKDB and ICH, the long-term neurological sequelae are common.

      References

        • Williams M.D.
        • Chalmers E.A.
        • Gibson B.E.
        The investigation and management of neonatal haemostasis and thrombosis.
        Br J Haematol. 2002; 119: 295-309
        • Townsend C.W.
        The haemorrhagic disease of the newborn.
        Arch Paediatr. 1894; 11: 559-565
        • Lehmann J.
        Vitamin K as a prophylactic in 13,000 infants.
        Lancet. 1944; 243: 493-494
        • American Academy of Pediatrics
        • Committee on Nutrition
        Vitamin K compounds and the water-soluble analogues: use in therapy and prophylaxis in pediatrics.
        Pediatrics. 1961; 28: 501-507
        • Sutor A.H.
        • von Kries R.
        • Cornelissen E.A.
        • McNinch A.W.
        • Andrew M.
        Vitamin K deficiency bleeding (VKDB) in infancy: on behalf of the ISTH Pediatric/Perinatal Subcommittee.
        Thromb Haemost. 1999; 81: 456-461
        • American Academy of Pediatrics
        • Committee on Fetus and Newborn
        Controversies concerning vitamin K and the newborn.
        Pediatrics. 2003; 112: 191-192
        • Centers for Disease Control and Prevention (CDC)
        Notes from the field: late vitamin K deficiency bleeding in infants whose parents declined vitamin K prophylaxis–Tennessee, 2013.
        MMWR Morb Mortal Wkly Rep. 2013; 62: 901-902
        • von Kries R.
        • Hachmeister A.
        • Göbel U.
        Oral mixed micellar vitamin K for prevention of late vitamin K deficiency bleeding.
        Arch Dis Child Fetal Neonatal Ed. 2003; 88: F109-F112
        • Ijland M.M.
        • Pereira R.R.
        • Cornelissen E.A.
        Incidence of late vitamin K deficiency bleeding in newborns in the Netherlands in 2005: evaluation of the current guideline.
        Eur J Pediatr. 2008; 167: 165-169
        • Takahashi D.
        • Shirahata A.
        • Itoh S.
        • Takahashi Y.
        • Nishiguchi T.
        • Matsuda Y.
        Vitamin K prophylaxis and late vitamin K deficiency bleeding in infants: fifth nationwide survey in Japan.
        Pediatr Int. 2011; 53: 897-901
        • Busfield A.
        • Samuel R.
        • McNinch A.
        • Tripp J.H.
        Vitamin K deficiency bleeding after NICE guidance and withdrawal of Konakion Neonatal: British Paediatric Surveillance Unit study, 2006-2008.
        Arch Dis Child. 2013; 98: 41-47
        • Laubscher B.
        • Bänziger O.
        • Schubiger G.
        Prevention of vitamin K deficiency bleeding with three oral mixed micellar phylloquinone doses: results of a 6-year (2005-2011) surveillance in Switzerland.
        Eur J Pediatr. 2013; 172: 357-360
        • Loughnan P.M.
        • McDougall P.N.
        Epidemiology of late onset haemorrhagic disease: a pooled data analysis.
        J Paediatr Child Health. 1993; 29: 177-181
        • Busfield A.
        • McNinch A.
        • Tripp J.
        Neonatal vitamin K prophylaxis in Great Britain and Ireland: the impact of perceived risk and product licensing on effectiveness.
        Arch Dis Child. 2007; 92: 754-758
        • Sutor A.H.
        • Kunzer W.
        Time interval between vitamin K administration and effective hemostasis.
        in: Suzuki S. Hathaway W.E. Bonnar J. Sutor A.H. Perinatal thrombosis and haemostasis. Springer, Berlin, Heidelberg, New York, London, Tokyo, Hong Kong1991