Prevention of Vitamin K defficency

7/23/2019 Prevention of Vitamin K defficency

1/6

P F hosted at the Radboud Repository of the Radboud UniversityNijmegen

The following full text is a publisher's version.

For additional information about this publication click this link.

http://hdl.handle.net/2066/24702

Please be advised that this information was generated on 2015-10-22 and may be subject tochange.
http://hdl.handle.net/2066/24702http://hdl.handle.net/2066/24702


2/6

r J Pediatr (1997) 156: 126 - 130 Springer-Verlag 199

. Cornelissen R. von Kries *P. Loughnan Schubiger

revention of vitamin K deficiency bleeding: fficacy of different multiple oral dose schedules of vitamin K

ceived: 12 April 1996 / Accepted: 21 July 1996

bstract There is consensus that late vitamin K defiency bleeding (VKDB) should be prevented by vitaminprophylaxis, One single dose of 1 mg vitamin Ki isfective if given i.m. or s.c., but not if given orally.

epeated oral doses might be as effective as the parnteral dose but the optimal dose regimen remains to betablished. Different oral dose schedules are presentlyed in different countries. In Australia, Germany, Theetherlands and Switzerland active surveillance data onte VKDB were collected in a similar manner andilure rates compared. Identical case definitions wereed. There were three basic strategies for oral and oner parenteral vitamin K prophylaxis for healthy new

orns in the four countries: (1) daily supplementation ofw dose vitamin K (25 pig) for breast-fed infants (Theetherlands); (2) 3 x 1 mg orally [Australia (January993 - March 1994) and Germany (December 1992 ecember 1994)]; (3) 1 mg vitamin K i.m. (Australiance March 1994); and (4) 2 x 2 mg vitamin K (newixed micellar preparation) (Switzerland). The re

pective failure rates per 100,000 live births (includingases given all recommended doses and those given inomplete prophylaxis) were for strategy: (1) 0.2 (0-1.3)

The Netherlands; (2) 2.3 (95% Cl 1.6-3.4) in Germanynd 2.5 (1.1-4.8) in Australia (oral prophylaxis); (3)ustralia (i.m. prophylaxis) 0 (0-0.9); and (4) 3.6 (0.7-

. Cornelissenepartment of Paediatrics, University Hospital Nijmegen,ijmegen, The Netherlands

v. Kries (EH)epartment of Epidemiology, Kinderzentrum Mncheneiglhofstrasse 63, D-81377 Mnchen, Germanyax: (089) 71009-315

Loughnan

epartment of Neonatology, Royal Childrens Hospital,arkville, Victoria, Australia Schubiger

epartment of Paediatrics, Kinderspital Luzern,uzern, Switzerland

10.6) in Switzerland. The failure rates for comple prophylaxis only were: strategy (1) 0 (0-0.7) in T

Netherlands; (2) 1.8 (1.1-2.8) in Germany and 1.5 (03.6) in Australia; (3) Australia (i.m.) 0 (0-0.9); and (1.2 (0-6.5) in Switzerland.Conclusions The Australian data confirm that threoral doses of 1 mg vitamin K are less effective than i.mvitamin K prophylaxis. A daily low oral dose of 25 jivitamin Kt following an initial oral dose of 1 mg aft bir th for exclusively breast-fed infants may be aeffective as parenteral vitamin K prophylaxis. Theffectiveness of the mixed-micellar preparation ovitamin Ki needs further study.

Key words Vitamin K Infancy Prophylaxis Latevitamin K deficiency Bleeding

Abbreviations C l confidence interval H IV humanimmunodeficiency virus VKBD vitamin K deficiency bleeding

Introduction

Vitamin K prophylaxis is required for the prevention late vitamin K deficiency bleeding (VKDB), which m be fatal or cause serious morbidity. Without prophylaxincidences of 5-7 per 100,000 live births have been ported in Europe [10]. Administration of 1 mg vitamKj i.m. is safe and effective in preventing late VKDand has been common practice in the U.S.A. since 19whereas a single oral dose of vitamin Ki at birth is leeffective [10].

In 1992, an unexpected association between i.m. tamin K prophylaxis in the neonatal period and lachildhood cancer was reported [7]. Although uconfirmed, the report caused considerable conceworldwide and prompted changes in national prophlaxis policies in several countries [5, 6]. In most of thcountries the use of repeated oral vitamin K was


3/6

127

ommended for healthy neonates, reserving i.m. adinistration for ill newborns, in whom oral adminisation is impractical. Different countries, however,dopted different regimens. In Germany and Australiaree oral doses of 1 mg vitamin Kj were recommended1992 [5]. In the Netherlands a small daily dose of 25 \xg prescribed for breast-fed infants in addition to a 1 mgtamin K dose at birth either orally for healthy neo

ates or i.m. in those unwell [22]. In Switzerland twoal doses of a new mixed-micellar vitamin Ki preara ti on have been recommended since 1995 [19].Is repeated oral prophylaxis as effective as i.m. ad

inist ration for the prevention of late onset VKDB? Areere differences between oral regimens regarding theirficacy? Here we report the results of an active sur

eillance trial for late onset VKDB which was carriedut in four different countries with three different oralophy laxis schedules.

ethods

The Netherlands, Germany, Australia and Switzerland activerveillance programmes for VKDB are in operation, usingethods developed by the British Paediatric Association Surveilnce U nit [8], Surveillance is based on monthly report cards sent toediatricians responsible for inpatient care in paed iatric hospitals

y the nat ional paediatric organisation. A nothing to reporttio n was included. A response rate of 85-99% was achieved in alluntries.

The participating countries have agreed on uniform case defiions [21] to allow for meaningful international comparison. Thefinition for the first step (a postcard mailed to all paediatricians)deliberately open, encouraging the reporting of all infants in

hom bleeding may have resulted from VKDB. All notificationse verified by questionnaire, asking for information about thefant, vitamin K administration, clinical presentation and lara to ry data.Case definition of late onset VKDB: any infant between 8 days

d the end of week 12, with spontaneous bruising/bleeding orracranial haemorrhage associated with prolonged prothrombin

say, not due to an inherited coagulopathy or disseminated in-vascular coagulation.

Confirmed VKDB was diagnosed, when the prothrombin assaysults were grossly abnormal compared with standards for ageuick values < 15%; international normalised ratio > 4; pro

rom bin time > 4 x control value) and at least one of the folwing was present:

Platelet count normal or raised and normal fibrinogen andabsent fibrin degradation products.Prothrombin time returned to normal after vitamin K administration.Concentra tion of PIVKA proteins exceeded normal controls,

obable VKDB was assumed when the prothrombin and partialomboplastin assay results were abnormal for age, but not as

ossly as in confirmed cases and if at least one of the aboveentioned items was documented.

Cases were also classified according to the cause of VKDB:

Idio path ic cases were defined as cases in whom no factor predisposing to vitamin K deficiency was identified.Secondary cases were those in whom a predisposing cause, illness or diagnosis was discovered after presentation with bleeding.both idiopathic and secondary cases, vitamin K prophylaxis cansaid to have failed either through inadequacy or omission. In

other cases already known to have a disease predisposing to vitamin K deficiency, vitamin K administration was p art of the norm adisease management. Bleeding in those cases was not a failure o prophylaxis, bu t a failure of management. These are reported s parately.

Oral vitamin K prophylaxis was considered complete if all nationally recommended doses had been given at the time of bleedinOral vitamin K prophylaxis was considered incomplete, if at leaone oral dose had been given but not all recommended doses foage or if an inadequate dose or preparation had been administeredThe number of oral vitamin K doses given was documented in th

respective questionnaires. There were no means to estimate, th proportions of healthy neonates given complete or incomplete orvitamin K prophylaxis at different ages.

Results

The relevant vitamin K recommendations for differentime periods and countries together with the periods owhich active surveillance are shown in Table 1. Thsurveillance data allowed comparisons between thDutch recommendations (25 pig vitamin K/day fo breast-fed infants + 1 mg vitamin K at birth), threoral doses of 1 mg (Australia and Germany) and tworal doses o f 2 mg vitamin K (mixed micellar p re paration), as recommended in Switzerland.

The surveillance data are shown in Table 2. In Th Netherlands, five confirmed cases of late VKDB werreported from October 1992 until December 1994: threcases had known predisposing liver disease not properdealt with, and two were idiopathic cases. In the firsidiopathic case vitamin K prophylaxis was omitted atogether, and in the second case the postnatal dose wa

not given. Both babies had been entirely breast-fedThere were no cases in infants who received the complerecommended prophylaxis. One of the idiopathic casehad fatal intracranial bleeding.

In Germany 32 infants met the case definition (2confirmed and 4 probable cases) in the study periofrom April 1993 to September 1994. One had a predi posing disease requiring specific prophylaxis, not adquately dealt with. Seventeen cases had intracrania bleeding. The reported incidence of VKDB despite com plete prophylaxis according to national reco mmendtions was relatively high, 1.8 (1.1-2.8) per 100,000 liv

births. iIn Australia, a similar prophylaxis protocol to that iGermany was used from January 1993 to March 1994During this period intramuscular vitamin K|, 1.0 mgwas much less commonly used than the oral doses usein healthy neonates (Peter Loughnan, personal communication). There were eight cases (seven confirmeone probable), over this time [15]. Five of these hareceived prophylaxis according to national recommendations. Three cases had intracranial bleedin

t

one died. The failure rates were almost identical to thoobserved in Germany. In March 1994 the national policwas reversed to the previously recommended 1 mg i.mat birth. Until May 1995, no cases of VKDB were observed with this regimen. .


4/6

ble 1 Recommendations fo r vitamin K prophylaxis and surveillance periods for late VKDB in different countries

8

untry

he Netherlands

ermany

ustralia*

witzerland

Recommended prophylaxis Time period Surveillance dataavailable for

i * iveil babies

,funw el l ba bi es ":

"wel l babies :

"unwell bab ies 3:

"w ell bab ies 1:"unwell babies ' :

all babies:

"well babies:

i t

unwell babies ' :

1 mg vitamin K p .o , & 25 ig daily fromweek 1 to week 13 (breast-fed)

1 mg vitamin K i.m. & 25jug daily(breast-fed)

1 mg vitamin K p.o. day 1, 4-10, 28-42

0,1-0,2 mg vitamin K i.m. or s.c. day 1& 1 mg vitamin K p.o. day 4-10, 28-42

1 mg vitamin K p.o. day 1, 3-5, 21-280,1 mg vitamin K i.m. day 1& day 3-5,21-28 0,1 mg i.m. vitamin K (or 1 mg p.o.)

1 mg vitamin Ku t l day 1

2 mg mixed-mi cellar preparation p .o. day1, 4

0,5 mg mixed-mi cellar prep aration i.v.or i.m., 2 mg MM-preparation p.o, week 4-6

since 1990

December 1992 -

December 1994

January 1993 -March 1994

since March 94

since 1995

October 1992 untilDecember 1994

April 1993 until

September 1994

January 1993 -March 1994

March 1994 - May 1995

January 1995 -December 1995

Unless otherwise started the cremopher preparation Konakion was used for single oral or i.m, doses

In Switzerland a new mixed-micellar preparation oftamin Ki has been used from January 1995. All healy neonates should receive 2 mg of vitamin Kj (Ko

akion MM) orally on the 1st and 4th day of life. Fromnuary 1995 to December 1995 four cases of lateKDB were reported (three confirmed and one probablease); two of the confirmed cases were idiopathic andne of them had hepatobiliary disease. One of theseree infants had received no prophylaxis, and two had

een given the old fat soluble drops (Konakion prearation), The probable case had very low vitamin Kependent clotting factors, which returned to normalter vitamin K administration. However, vitamin K

deficiency may not have been the cause of the intracranial bleeding because the baby additionally hadlow platelet count of 14,000/jil, and angiomatosis of thchoroid plexus. Because of the limited number of chdren studied in Switzerland to date, 95% confidenclimits of the failure rate are broad: 1.2 (0-6.5).

Discussion

There is good evidence that parenteral vitamin K pr phylaxis (1 mg i.m. at birth) can prevent almost all cas

ble 2 Late VKDB in countries with different recommendations for vitamin K prophylaxis

Ne therland s Germany Australia Switzerland

rth population 439,000 1,200,000 325,000(vitamin K p.o.)

325,000(vitamin K i.m.)

83,000

u mb er of cases 5 32 8 0 4idiopathic cases 2 13 3 0 3secondary cases 0 18 5 0 1predisposing illness 3 1 0 0 0tal incidence 1.1 2.7 2.5 0 4.7b+c' (95% Cl)2 (0.4-2.7) (1.8-3.8) (1.1-4.8) (0-0.9) (1.3-11.9)ue incidence 0.5 2.6 2.5 0 4.7b (95% Cl)2 (0.1-1.6) (1.8-3.7) (1.1-4.8) (0-0,9) (1.3-11.9)ophylaxis

omitted 1 2 0 0 1incomplete 1 6 3 0 2 3

. complete 0 22 5 0 1V. not documented 0 1 0 0 0

ophylaxis failures 15% Cl)2complete 0 1.8 1.5 0 1.2rophylaxis (0-0.7) (1.1-2.8) (0.5-3.6) (0-0.9) (0-6.5)complete or 0.2 2.3 2.5 0 3.6

ncomplete prophylaxis (0-1.3) (1.6-3.4) (1.1-4.8) (0-0.9) (0.7-10.6)er 100,000 live birthsssuming a Poisson distributionAn older preparation, Konakion oral drops was given instead of 'mixed-micellar5


5/6

129

f late VKDB [10], Data from switch-over studies in oneountry [4], and from the simultaneous use of differentrophylaxis strategies in three other countries [9, 16, 20]how that vitamin K prophylaxis with one oral dose (1-

mg) is less effective than parenteral vitamin K. Reeated oral dose schedules have therefore been reommended. However, a recent publication fromermany suggests that even repeated oral vitamin Krophylaxis may not be as effective as the 1 mg dosedministered i.m. at birth [12].

The Dutch recommendation for daily supplementaon of vitamin K in exclusively breast-fed infants miics formula-feeding, which prevents late VKDB almost

ompletely [11], by providing a daily intake of about0 jig vitamin Kj. Daily supplementation already existsor vitamin D. A diluted vitamin K preparation wasarketed as a food supplement in The Netherlands,

herefore bypassing time-consuming drug licensingrocedures. Gynaecologists, midwives and generalracti tioners instruct the parents. The incidence of late

KDB despite oral prophylaxis (failure incidence) was 095% Cl 0-0.7) per 100,000 live births. Before vitamin Krophylaxis was introduced in 1990, the estimated indence was 7 per 100,000 (1984-1986) [24]. A recom

mendation for daily low oral doses therefore appearsound.

In Germany and Australia three oral doses of 1 mgtamin Kj had been recommended in 1992. The Cre-

mopho r solution (Konakion) was used for oral vitaminprophylaxis in bo th countries. The failure rates were

most identical in both countries and were much higherhan with parenteral vitamin K prophylaxis [9]. Failures

sulted not only from non compliance (forgottenepeat doses) but were also observed in children givenhe recommended vitamin K dose. These observationsrompted a reversal to i.m. vitamin K in Aus tralia, anddose increase to 3 x 2 mg oral vitamin K in Germany.he data following the switch over in Australia confirmrevious observations, that i.m. prophylaxis with 1 mgitamin K can prevent almost all cases of late VKDB [4,

10, 16, 19].The new mixed-micellar preparation is well absorbed

18], even in children with cholestasis [2]. In Switzerland,wo of the three reported failures of oral vitamin Krophylaxis occurred in children given the Konakionreparation. In 1993/94 when two oral doses of 2 mg ofhe old Konakion were recommended in Switzerland,he incidence of late VKDB had been about 4.2 per 00,000 [17], Under the assumption that most childrenceived the mixed-micellar preparation during 1995 inwitzerland it appears possible that the number of failres will be lower with the use of the new mixed-micellarreparation (Konakion MM) although the evidence isot conclusive.

Several potential sources of bias must be consideredor between country comparisons of prophylaxis failureates for the assessment of the relative efficacy of diferent regimes for oral vitamin K prophylaxis.

1. Constant and identical baseline incidences (withouvitamin K prophylaxis) in the respective countries aremandatory. These baseline incidences are influenced by e.g. the breastfeeding habits and the prevalence ocholestatic liver disease in the respective countriesAlthough the baseline incidence was similar in reportsfrom The Netherlands [24] and Germany [9, 10]corresponding data from Switzerland and Australiaare lacking.

2. Completeness of ascertainment is an issue in all formof surveillance systems. Even with active surveillancecomplete ascertainment is unlikely, as shown by attempts to measure the incidence of childhood dia betes [3, 23], vertical HIV infections [1], multiple births [14] and systemic Haemophilus influenzae infections [13]. The under ascertainment, however, ap

pears to be in a similar range for most conditionswithin one national surveillance system [1, 14, 23] andeven between similar active surveillance systems [3,23]. Major differences in the ascertainment between

different active surveillance programmes thereforeappear unlikely and meaningful between countrycomparison seems possible.

3. AH failure rates had to be calculated with all live births in the denom inator because the exact numbersof children eligible for oral vitamin K prophylaxis ineach co unt ry is unknown. The proportions of eligiblchildren, however, are likely to be similar, due to thesimilarities of the eligibility criteria.

4. The proportions of children, who received all recommended doses for oral vitamin K prophylaxis areunknown. It is likely, however, that less than 100% o

the eligible cases received the recommended prophylaxis in all countries. All failure rates therefore arelikely to be underestimated due to a greater denominator than the true numbers exposed. If thesame proportions of children failed to receive therecommended prophylaxis in each country this wouldnot account for bias in the international comparisonBias could be introduced, however, if the proportionsof children given the recommended doses was different between the countries compared.

The contribution of these potential sources of bias fo

the comparison of different oral vitamin K prophylaxischedules is difficult to measure. In terms of properepidemiology these results therefore can only be used togenerate hypotheses to be tested in randomized controlled trials. Such trials would require about 700,000children on standard oral vitamin K prophylaxis (e.g3 x 1 mg: 2 expected cases of VKDB /100,000 live birthsversus abou t 700,000 children on an optimal regime (e.g3 x 2 mg mixed micellar preparation or daily doses o25 j.ig in breast-fed babies plus 1 mg oral vitamin K a birth: 0.01 expected cases of late VKDB per 100,00livebirths). It is evident that such a trial is logisticallyextremely difficult and expensive. The size of the clinica problem (about 7 relevant bleedings due to VKDB pe100,000 livebirths without any vitamin prophylaxis i


6/6

130

western countries) will hardly convince grant donatingbodies even in extremely affluent societies to suppor tsuch a study,

Poor evidence from inte rnational comparisons - aspresented in this paper - will therefore remain the onlybasis for decision making on the optimal form of vitamin K prophylaxis.

Acknowledgements This work would not have been possible without the support of the national surveillance units: APSU in Ausralia, ESPED in Germany, NSCK in The Netherlands, and SPSUn Switzerland.

References

1. Ades AE, Davison CF, Holland FJ, Gibb DM, Hudson CN, etal (1993) Vertically transmitted HIV infection in the BritishIsles. BMJ 306:1296-1299

2. Amade-Manesme O, Lambert WE, Alagille D, De Leenheer AP(1992) Pharmacokinetics and safety of a new solution of vitamin K I{2o) in children with cholestasis. J Pedia tr Gastroenterol Nu tr 14:160-165

3. Bunka M, Cremer HJ, Giani G, Helwig H, Herzig P et al imAuftrag des ESPED-Beirates (1995) ESPED-Jahresbericht1994. Monatsschr Kinderheilkd 143:1035-1046

4. Ekelund Ii (1991) Late haemorrhagic disease in Sweden 1987-89. Acta Paediatr Scand 80:966-968

5. Ernhrungskommission der Deutschen Gesellschaft fr Kinderheilkunde, vorbereitet durch von Kries R, und Gbel U(1993) Vitamin-K-Prophylaxe bei Neugeborenen. Sozialpdiatrie 15:36

6. Expert Committee (1992) Vitamin K prophylaxis in infancy.British Paediatric Association, London

7. Golding J, Greenwood R, Birmingham K, M ott M (1992)Childhood cancer, intramuscular vitamin K, and pethidine given during labour. BMJ 305:341-346

8. Hall SM, Glickman M (1988) The British Paediatric Surveillance Unit. Arch Dis Child 63:344-346

9. Kries R von, Gbel U (1992) Vitamin K prophylaxis and latehemorrhagic disease of newborn (HDN). Acta Pediatr Scand81:655-657

10. Kries R von, Hanawa H (1993) Neonatal vitamin K prophylaxis, report of scientific and standardization subcommittee on

perinatal haemostas is. Throm b Haem os t 69:293-295

lLKries R von, Shearer MJ, Gbel U (1988) Vitamin K in infancy. Eur J Pediatr 147:106-112

12. Kries R von, Hachmeister A, Gbel U (1995) Repeated orvitamin K prophylaxis in West Germany: acceptance and effcacy. BMJ 310:1097-1098

13. Kries R von, Heinrich B, Bhm O, Windfuhr A, Helwig H fHib-Arbeitsgruppe der DGPI (1996) Systemische Haemophiluinfluenzae-Erkrankungen in Deutschland: 1992-1995. Montsschr Kinderheilkd (im Druck)

14. Levene MI, Wild J, Steer P for the British Association o

Perinatal Medicine (1992) Higher multiple births and thmodern management of infertility in Britain. Br J Obstet Gynaecol 99:607-613

15. Loughnan P, Chaot K, Elliott E, Henderson-Smart DMcDougall P (1995) Haemorrhagic disease in Australia, 19931994. Report from the Australian paediatric surveillance uniJ Paediatr Child Health 31:A2

16. McNinch A, Tripp JH (1991) Haemorrhag ic disease of thnewborn in the British Isles: a two year prospective study, BM303:1105-1109

17. Schubiger G, Tnz O (1994) Epidemiological Data on VitamK Prophylaxis in Switzerland. In: Sutor AH, Hathaway W(eds) Vitamin K in infancy. Schattauer Verlag, Stuttgart, p177-178

18. Schubiger G, Tnz O, Grter J, Shearer MJ (1993) Vitamin Kconcentration in breast-fed neonates after oral or intramusculadministration of a single dose of a new mixed-micellar pr

para tion of phylloquinone. J Pediatr Gastroenterol Nu16:435-439

19. Schubiger G, Roulet M, Laubscher B (1994) Vitamin-Ki-Pr phylaxe bei Neugeborenen: Ne ue Empfehlungen. Schweizeische Arztezeitung 75:2036

20. Tnz O, Schubiger G (1988) Neonatale Vitamin K Prophylaxund Vitamin K Mangelblutungen in der Schweiz 1986-1988Schweiz Med Wochenschr 118:1747-52

21. Tripp J, Cornelissen M, Loughnan P, McNinch A, SchubigeG, Kries R von (1994) Suggested protocol for the reporting prospective studies of vitamin K deficiency bleeding (previouscalled hemorrhagic disease of the newborn). In: Sutor AHHathaway WE (eds) Vitamin K in infancy. Schattauer VerlagStuttgart, pp 395-401

22. UitenUiis J (1990) Vitamin K prophylaxis in newborns anduring the first months of life. Ned Tjidschr Geneesk 134:1641645

23. Wadsworth E, Shield J, Hunt L, Baum D (1995) Insulin d pendent diabetes in chi ldren und er 5: incidence and ascertaiment validation for 1992. BMJ 310:700-703

24. Widdershoven JAM (1987) Vitamin K deficiency in infancyMedical Thesis, University of Nijmegen

Prevention of Vitamin K defficency

Documents

Transcript of Prevention of Vitamin K defficency