Acta PBdiatr 81: 658-61. 1992
Vitamin K status in cystic fibrosis E 4 M Cornelissen, A F van Lieburg, K Motohara’ and CG van Oostrom Departments of Paediatrics. University Hospital Nijmegen, Nijmegen, The Netherlands and Amakusa-Sumoto Hospital‘. Kumamoto. Japan
Cornelissen EAM, Lieburg van AF, Motohara K, Oostrom van CG. Vitamin K status in cystic fibrosis. Acta Paediatr 1992;81:658-61. Stockholm. ISSN 0803-5253 Appearance of PIVKA-I1 (protein induced by vitamin K absence-11)in serum is a biochemical sign of insufficient vitamin K-dependent carboxylation of prothrombin. Plasma concentrations of PIVKA-I1 and vitamin K I were determined in 24 children with cystic fibrosis. Eight were supplemented with vitamin KI. The purpose of the study was to determine the occurrence of vitamin K deficiency in cystic fibrosis and to evaluate the effect of vitamin K supplementation. PIVKA-I1 was detectable in only one unsupplemented child. In this patient, the concentration of vitamin K I was below the limit ofdetection of 60 ng/l. Vitamin K I levels in the other unsupplemented children were normal (mean 476 ng/l= 1 mmol/l). The supplemented patients showed extremely high levels of vitamin K1(mean 22 445 ng/l= 50 nmol/l). In conclusion, vitamin K deficiency occurs infrequently in cystic fibrosis. Checking the coagulation system is advised, but routine vitamin K supplementation is not recommended. If additional vitamin K is needed, the starting dose should not exceed 1 mg daily. 0 Cysticfibrosis, haemorrhagic disease. PIVKA-[I, vitamin K , vitamin K dejciency EAM Cornelissen. Department of Paediatrics, University Hospital Nijmegen, P.O. Box 9I01*NLdSOO HB Nijmegen, The Netherlands
In nature, vitamin K is present as vitamin K I (phylloquinone) in green plants and as vitamin K2(menaquinones) produced by bacteria. Vitamin K is a cofactor necessary for hepatic post-translational carboxylation of glutamic acid residues in prothrombin and other vitamin Kdependent proteins. Gamma-carboxylation of glutamic acid residues creates effective calcium and phospholipid binding sites on the clotting factors, necessary for proper coagulation activities. In vitamin K deficiency, incompletely carboxylated proteins that are functionally defective appear in plasma. These abnormal coagulation factors are called “proteins induced by vitamin K absence” (PIVKA). PIVKA-I1 refers to descarboxylated prothrombin. PIVKA-I1 is not detected in plasma unless there is a defect in carboxylation attributable to an impairment of carboxylase enzyme and/or a deficiency or antagonism of vitamin K. On this basis, coagulation abnormalities in patients with cystic fibrosis (CF) can be caused by vitamin K deficiency, liver dysfunction or both. The frequency of coagulation problems in C F due to vitamin K deficiency is unknown. Recently, techniques have been developed for measurement of physiological plasma concentrations of vitamin K I . Despite several case reports of cutaneous, intestinal and intracranial haemorrhages caused by vitamin K deficiency in children with C F (14), vitamin K I concentrations in C F patients have been reported only once ( 5 ) . In that study, vitamin K I levels were found to be similar to controls. The authors concluded that routine vitamin K supplementation is not required.
However, children with C F may require more vitamin K I than controls for effective carboxylation, due to (sub)clinical impairment of liver function. Moreover, diarrhoea and antibiotic therapy could interfere with vitamin KZ production by intestinal flora. Reduced hepatic stores of vitamin KZ could in turn contribute to an increased need for vitamin K I in children with CF. Detection of PIVKA-I1 is a more direct way of demonstrating ineffective carboxylation of clotting factors. Determination of PTVKA-I1 by monoclonal antibody is a very specific and sensitive method (6). In the present study, simultaneous measurements of vitamin KI and PIVKA-I1 were performed in children with CF, who were or were not supplemented with vitamin K I . Our purpose was to establish the frequency of vitamin K deficiency in C F and to evaluate the effect of vitamin K supplementation.
Patients and methods Blood samples were collected from 24 C F patients (nonfasting) who attended our outpatient clinic for routine biannual check-up. Ten were males and 14 females. Their ages ranged from 6 months to 23 years (mean 10 years). Eight patients were supplemented with vitamin K I (Fytomenadion, Konakion, Hoffman-La Roche, Basel, Switzerland) in an oral dose of 4-30 mg daily. Two of the 16 unsupplemented patients were fed with an infant formula enriched with vitamin K I . Blood was sampled by venepuncture: 3 ml of citrated
Vitamin K status in CF
ACTA P E D I A T R X I (1992)
blood (in silicone-coated tubes containing 10% (v/v) of sodium citrate 3.8%) and 8 ml of coagulated blood without additive. After performing a Thrombotest according to Owren (7) using Thrombotest Reagents (Nijegaard & Co, Oslo, Norway), the citrated blood was centrifuged (5000 x g for 10 min) and plasma was stored at - 70" C for PIVKA-I1 analysis. The coagulated blood was kept in the dark immediately after sampling, centrifuged (3000 x g for 5 min) and the serum stored at -20" C until determination of vitamin K I and liver function tests (GGT, ALT, AST, ALP and direct and total bilirubin). PIVKA-I1 concentrations were determined by an enzyme-linked immunosorbent assay, using a monoclonal antibody as described previously by Motohara et al. (8). This antibody reacts with descarboxylated prothrombin (PIVKA-11) quantitatively and does not cross react with native prothrombin. PIVKA-I1 levels are expressed in arbitrary units (AU) per ml, so that 1 AU corresponds to 1 pg of purified prothrombin. In severe vitamin K deficiency, PIVKA-I1 levels can exceed 20 AU/ml. Vitamin K1 was extracted from I-ml serum samples and measured by a two-step high performance liquid chromatographic (HPLC) procedure, according to the method of Lambert et al. (9, 10). A few minor modifications were applied. Recovery of vitamin K I from standard solutions added to normal serum was 85 f5%. The detection limit was 60 pg/ml. Retrospectively, clinical data at the time of blood sampling, including vitamin and enzyme supplementation, antibiotic therapy, other drug intake, diet, bleeding diathesis, length, weight-for-height, presence of steatorrhoea and echographically confirmed liver cirrhosis and/or splenomegaly, were derived from medical records. For statistical calculations the chi-square test, Fisher's exact test, Mann-Whitney U-test and Spearman's rank correlation were used where appropriate.
Results Clinical data of the vitamin K unsupplemented and supplemented patients are presented in Table 1. The groups did not differ significantly (p > 0.05; chi-square and Fisher's exact test). All 24 children took pancreatic enzymes daily. In addition, 2 1 were also supplemented with vitamins A, D and E. Ten patients were on antibiotic treatment at the time of blood sampling. Eight had used these antibiotics (co-trimoxazole, ciprofloxazin, doxycycline, amoxycillin and tobramycin spray) for more than four weeks at the time of blood sampling. No serious bleeding tendencies were reported. The results of laboratory data are shown in Table 2. Individual plasma concentrations of vitamin K I are shown in Fig. 1. One patient (aged 14 years) had detectable PIVKA-I1 concentrations (0.159 AU/ml).
659
Tuble I . Clinical characteristics of the patients studied, grouped by supplementation with vitamin K I .
Unsupplemented Number Mean age (years) (range) Sex (male: female) Antibiotics (see text) Length $PI0 Weight-for-height < PI0 Clinical steatorrhoea Echographic cirrhosis* Echographic splenomegaly
16 9 (0.5-23) 719 5/16(31)t 10/16 (63)t 8/16 (50)t 2/11 (18)t 4/ 15 (2711 3/14 (21)t
Supplemented 8 12 (3-17) 3:5 518 (63)t 218 ( 2 5 ~ 518 (63)t 115 (20)t 417 (57)t 216 (33)t
* As indicated by inhomogeneous, more echo-dense liver pattern on ultrasonographic examination. -f Numbers in parentheses are percentages.
Table 2. Laboratory parameters in cystic fibrosis patients with and without supplementation of vitamin KI.
Unsupplemented ( n = 16) ~
~~
Supplemented (n = 8)
~~
Thrombotest (%)" Median Mean SD Vitamin KI (ng/l) Median Mean fSD PIVKA-I1 >0.10 AU/ml (No.) Elevated liver enzymes (No.)
65 65+ 15 428 476 & 5 I I I 3
56 56 7'
+
25872 22445 f I7340** 0 3
~
Values are expressed as percentage of adult pooled plasma. One value below detection limit excluded. ns. **p
Vitamin K status in cystic fibrosis E 4 M Cornelissen, A F van Lieburg, K Motohara’ and CG van Oostrom Departments of Paediatrics. University Hospital Nijmegen, Nijmegen, The Netherlands and Amakusa-Sumoto Hospital‘. Kumamoto. Japan
Cornelissen EAM, Lieburg van AF, Motohara K, Oostrom van CG. Vitamin K status in cystic fibrosis. Acta Paediatr 1992;81:658-61. Stockholm. ISSN 0803-5253 Appearance of PIVKA-I1 (protein induced by vitamin K absence-11)in serum is a biochemical sign of insufficient vitamin K-dependent carboxylation of prothrombin. Plasma concentrations of PIVKA-I1 and vitamin K I were determined in 24 children with cystic fibrosis. Eight were supplemented with vitamin KI. The purpose of the study was to determine the occurrence of vitamin K deficiency in cystic fibrosis and to evaluate the effect of vitamin K supplementation. PIVKA-I1 was detectable in only one unsupplemented child. In this patient, the concentration of vitamin K I was below the limit ofdetection of 60 ng/l. Vitamin K I levels in the other unsupplemented children were normal (mean 476 ng/l= 1 mmol/l). The supplemented patients showed extremely high levels of vitamin K1(mean 22 445 ng/l= 50 nmol/l). In conclusion, vitamin K deficiency occurs infrequently in cystic fibrosis. Checking the coagulation system is advised, but routine vitamin K supplementation is not recommended. If additional vitamin K is needed, the starting dose should not exceed 1 mg daily. 0 Cysticfibrosis, haemorrhagic disease. PIVKA-[I, vitamin K , vitamin K dejciency EAM Cornelissen. Department of Paediatrics, University Hospital Nijmegen, P.O. Box 9I01*NLdSOO HB Nijmegen, The Netherlands
In nature, vitamin K is present as vitamin K I (phylloquinone) in green plants and as vitamin K2(menaquinones) produced by bacteria. Vitamin K is a cofactor necessary for hepatic post-translational carboxylation of glutamic acid residues in prothrombin and other vitamin Kdependent proteins. Gamma-carboxylation of glutamic acid residues creates effective calcium and phospholipid binding sites on the clotting factors, necessary for proper coagulation activities. In vitamin K deficiency, incompletely carboxylated proteins that are functionally defective appear in plasma. These abnormal coagulation factors are called “proteins induced by vitamin K absence” (PIVKA). PIVKA-I1 refers to descarboxylated prothrombin. PIVKA-I1 is not detected in plasma unless there is a defect in carboxylation attributable to an impairment of carboxylase enzyme and/or a deficiency or antagonism of vitamin K. On this basis, coagulation abnormalities in patients with cystic fibrosis (CF) can be caused by vitamin K deficiency, liver dysfunction or both. The frequency of coagulation problems in C F due to vitamin K deficiency is unknown. Recently, techniques have been developed for measurement of physiological plasma concentrations of vitamin K I . Despite several case reports of cutaneous, intestinal and intracranial haemorrhages caused by vitamin K deficiency in children with C F (14), vitamin K I concentrations in C F patients have been reported only once ( 5 ) . In that study, vitamin K I levels were found to be similar to controls. The authors concluded that routine vitamin K supplementation is not required.
However, children with C F may require more vitamin K I than controls for effective carboxylation, due to (sub)clinical impairment of liver function. Moreover, diarrhoea and antibiotic therapy could interfere with vitamin KZ production by intestinal flora. Reduced hepatic stores of vitamin KZ could in turn contribute to an increased need for vitamin K I in children with CF. Detection of PIVKA-I1 is a more direct way of demonstrating ineffective carboxylation of clotting factors. Determination of PTVKA-I1 by monoclonal antibody is a very specific and sensitive method (6). In the present study, simultaneous measurements of vitamin KI and PIVKA-I1 were performed in children with CF, who were or were not supplemented with vitamin K I . Our purpose was to establish the frequency of vitamin K deficiency in C F and to evaluate the effect of vitamin K supplementation.
Patients and methods Blood samples were collected from 24 C F patients (nonfasting) who attended our outpatient clinic for routine biannual check-up. Ten were males and 14 females. Their ages ranged from 6 months to 23 years (mean 10 years). Eight patients were supplemented with vitamin K I (Fytomenadion, Konakion, Hoffman-La Roche, Basel, Switzerland) in an oral dose of 4-30 mg daily. Two of the 16 unsupplemented patients were fed with an infant formula enriched with vitamin K I . Blood was sampled by venepuncture: 3 ml of citrated
Vitamin K status in CF
ACTA P E D I A T R X I (1992)
blood (in silicone-coated tubes containing 10% (v/v) of sodium citrate 3.8%) and 8 ml of coagulated blood without additive. After performing a Thrombotest according to Owren (7) using Thrombotest Reagents (Nijegaard & Co, Oslo, Norway), the citrated blood was centrifuged (5000 x g for 10 min) and plasma was stored at - 70" C for PIVKA-I1 analysis. The coagulated blood was kept in the dark immediately after sampling, centrifuged (3000 x g for 5 min) and the serum stored at -20" C until determination of vitamin K I and liver function tests (GGT, ALT, AST, ALP and direct and total bilirubin). PIVKA-I1 concentrations were determined by an enzyme-linked immunosorbent assay, using a monoclonal antibody as described previously by Motohara et al. (8). This antibody reacts with descarboxylated prothrombin (PIVKA-11) quantitatively and does not cross react with native prothrombin. PIVKA-I1 levels are expressed in arbitrary units (AU) per ml, so that 1 AU corresponds to 1 pg of purified prothrombin. In severe vitamin K deficiency, PIVKA-I1 levels can exceed 20 AU/ml. Vitamin K1 was extracted from I-ml serum samples and measured by a two-step high performance liquid chromatographic (HPLC) procedure, according to the method of Lambert et al. (9, 10). A few minor modifications were applied. Recovery of vitamin K I from standard solutions added to normal serum was 85 f5%. The detection limit was 60 pg/ml. Retrospectively, clinical data at the time of blood sampling, including vitamin and enzyme supplementation, antibiotic therapy, other drug intake, diet, bleeding diathesis, length, weight-for-height, presence of steatorrhoea and echographically confirmed liver cirrhosis and/or splenomegaly, were derived from medical records. For statistical calculations the chi-square test, Fisher's exact test, Mann-Whitney U-test and Spearman's rank correlation were used where appropriate.
Results Clinical data of the vitamin K unsupplemented and supplemented patients are presented in Table 1. The groups did not differ significantly (p > 0.05; chi-square and Fisher's exact test). All 24 children took pancreatic enzymes daily. In addition, 2 1 were also supplemented with vitamins A, D and E. Ten patients were on antibiotic treatment at the time of blood sampling. Eight had used these antibiotics (co-trimoxazole, ciprofloxazin, doxycycline, amoxycillin and tobramycin spray) for more than four weeks at the time of blood sampling. No serious bleeding tendencies were reported. The results of laboratory data are shown in Table 2. Individual plasma concentrations of vitamin K I are shown in Fig. 1. One patient (aged 14 years) had detectable PIVKA-I1 concentrations (0.159 AU/ml).
659
Tuble I . Clinical characteristics of the patients studied, grouped by supplementation with vitamin K I .
Unsupplemented Number Mean age (years) (range) Sex (male: female) Antibiotics (see text) Length $PI0 Weight-for-height < PI0 Clinical steatorrhoea Echographic cirrhosis* Echographic splenomegaly
16 9 (0.5-23) 719 5/16(31)t 10/16 (63)t 8/16 (50)t 2/11 (18)t 4/ 15 (2711 3/14 (21)t
Supplemented 8 12 (3-17) 3:5 518 (63)t 218 ( 2 5 ~ 518 (63)t 115 (20)t 417 (57)t 216 (33)t
* As indicated by inhomogeneous, more echo-dense liver pattern on ultrasonographic examination. -f Numbers in parentheses are percentages.
Table 2. Laboratory parameters in cystic fibrosis patients with and without supplementation of vitamin KI.
Unsupplemented ( n = 16) ~
~~
Supplemented (n = 8)
~~
Thrombotest (%)" Median Mean SD Vitamin KI (ng/l) Median Mean fSD PIVKA-I1 >0.10 AU/ml (No.) Elevated liver enzymes (No.)
65 65+ 15 428 476 & 5 I I I 3
56 56 7'
+
25872 22445 f I7340** 0 3
~
Values are expressed as percentage of adult pooled plasma. One value below detection limit excluded. ns. **p
