Curves Vitamin Supplements Articles & Resources

METHODS

Patient Population

This study was done at the University of Wisconsin Perinatal Center at Meriter Hospital, the site of the University's teaching programs in neonatology and perinatology. Inclusion criteria for infants and mothers included a singleton delivery at more than or equal to 38 weeks' gestational age, as determined from the last menstrual period and confirmed by prenatal ultrasound dating. All infants were appropriate for gestational age using the curves of Babson et al.[10] Mothers had to be planning to breastfeed exclusively for 6 to 12 weeks. Mothers could not be receiving anticonvulsant therapy, and both mothers and infants were required to be free from gastrointestinal diseases associated with malabsorption. Maternal weights could not exceed 82 kg or be less than 50 kg. Informed consent was obtained from all mothers, and the study was approved by the institutional review boards at both Meriter Hospital and the University of Wisconsin Medical School. All infants received 1 mg of phylloquinone intramuscularly at birth.

For this study, two stages were required with two separate groups of pahents. It was first necessary to determine an appropriate phylloquinone supplementation for breastfeeding mothers to achieve the aims of the study. Thus in stage 1, breastfeeding mothers were randomized to 6 weeks of daily oral supplements of either or mg of phylloquinone within 3 days of delivery. These test levels of supplementation were based on data from our previous study.[6] Stage II was dependent on the results of stage I (see below). In this second phase, mothers were randomized in a double-blind fashion to mg of phylloquinone or a placebo, once a day for 12 weeks. The phylloquinone supplement was prepared from 5-mg tablets of Mephyton [Merck and Co, West Point, PA] (phylloquinone). The placebo was glucose. Both preparations were prepared and dispensed through the hospital pharmacy. Mothers were instructed to take the vitamin K or the placebo with the evening meal. A record was kept of maternal refills, and study measurements of maternal plasma levels of phylloquinone were a second test for compliance for the vitamin K-supplemented group.

In both stages, breast milk samples were collected from mothers at baseline (within 3 days of delivery) and at 2 and 6 weeks of lactation. In stage 11, breast milk samples were also collected at 12 weeks of lactation. Samples of breast milk were collected 18 to 24 hours after maternal ingestion of vitamin K or the placebo. A complete expression of a single breast was collected with an electric pump. Ten milliliters was placed in sterile, polypropylene bottles and frozen immediately at -70 [degrees] C for later analysis under aerobic conditions.

In stage II, 3-day dietary recalls were completed by the mothers at the time of the breast milk sampling to assess maternal vitamin K intake based on current knowledge of the vitamin K content of foods.[11] The volume of breast milk intake for each infant was determined gravimetrically at the time of milk sampling (2, 6, and 12 weeks). This was done by weighing infants before and after each feeding for three consecutive 24-hour periods. Mothers were instructed in the home using an automated electronic balance (Olympic, Model 20, Olympic Medical, Seattle, WA; precision, [+ or -] ), which integrates repetitive weighing every seconds. Mothers were also instructed to change diapers before each feeding and to estimate observed losses of milk by regurgitation. The accuracy of this test weighing procedure has been validated.[12]

Mothers were expected to exclusively breastfeed their infants in both stages of the study. Standard formulas and solid foods were not introduced. Mothers were supplied with a limited quantity of Similac PM 60/40 without added phylloquinone (supplied by Ross Laboratories, Columbus, OH) for emergency use only. If more than one bottle of the formula was used per week, infants were withdrawn from the study.

Plasma samples were collected in stage II from mothers at baseline (within 3 days of delivery) and at 2, 6, and 12 weeks of lactation. Plasma samples in this stage were also collected at 2, 6, and 12 weeks from the infants. Baseline plasma was not collected from the infants, because previous work from our group and others has shown vitamin K to be largely unmeasureable in cord blood and extremely high after the newborn intramuscular phylloquinone injection.[13,14] Blood samples were collected in citrate tubes for infants and in tubes without anticoagulant for mothers and separated, and plasma or serum was frozen at -70 [degrees] C for later analysis.

Plasma or serum phylloquinone concentrations were measured after hexane extraction of the samples and analysis by the method of Haroon et al.[15] Phylloquinone was quantitated by post-high-performance liquid chromatographic column zinc reduction, and detection of the reduced form of the vitamin was by fluorescence spectrophotometry. One milliliter of plasma is required for duplicate determination of phylloquinone concentration with a coefficient of variation of 10%. The adult normal plasma value using this assay is ng/mL in our population. The phylloquinone concentration of human milk was measured as for plasma.[16] One milliliter of milk is required for duplicate determination of phylloquinone concentration with a coefficient of variation of 8%. The normal concentration in pasteunzed cow milk is ng/mL and in skim milk is ng/mL in our laboratory.

The PIVKA-II was measured in stage II on infant plasma samples at 12 weeks using a murine monoclonal antibody available in an enzyme immunoassay kit from Diagnostica Stago (Asnieres Sur Seine, France). The normal value for PIVKA-II in adults is less than 2 ng/mL with this method.[17]

A one-stage prothrombin time (PT) was measured by the coagulation laboratory at Meriter Hospital on infants' plasma samples at 2, 6, and 12 weeks in stage II. The normal value for PT is 11 to 13 seconds, depending on the specific thromboplastin activity index of the thromboplastin reagent used.

Statistics

Sample size was determined by power calculations using the values of phylloquinone concentration in breastfeeding infants in this same general population from our previous study.[6] For stages I and II, at least 10 mothers or 10 infants in each group completing the protocol were required. The power was calculated on the basis of a one-tailed, two-sample t test, with a significance level of [alpha] = .

Data were analyzed using Sigma Stat (Jandel Scientific, San Rafael, CA). Normally distributed data were compared by the unpaired Student's t test, and nonparametric data were compared by the Mann-Whitney rank sum test. Correlation coefficients were determined by standard linear regression.

RESULTS

Stage I

Twenty mothers completed the protocol. Table 1 shows the effect of maternal vitamin K supplementation on human milk vitamin K concentration in stage I. Although vitamin K concentrations were strikingly increased in human milk in both the and supplementation groups, significantly higher concentrations were obtained in the daily supplementation group, as expected. There was considerable variability in the breast milk concentrations, as indicated by the SDs.

TABLE 1. Effect of Vitamin K Supplementation
on Human Concentration,ng/mL

 Time point Vitamin K, mg/d (n=10)(*)

Before supplement [+ or -] 
2 wk [+ or -] 
 ( - )
6 wk [+ or -] 
 ( - )

 Time point Vitamin K, 5 mg/d (n=10)(*)

Before supplement [+ or -] 
2 wk [+ or -] 
 ( - )
6 wk [+ or -] 
 ( - )

 Time point Significant([dagger])

Before supplement NS
2 wk P [is less than] .01

6 wk P [is less than] .01

(*) Values are means [+ or -] SD. Values in parentheses represent ranges.

([dagger]) Vitamin K, mg/d, versus vitamin K, 5, mg/d.

Stage II

Twenty-two mother-infant pairs completed this protocol. Infants did not differ by weight at any time point during the study. Weight at 2 weeks for the vitamin K group was 4135 [+ or -] 570 g (mean [+ or -] SD) and 3911 [+ or -] 625 g for the placebo group. At 12 weeks, weights for the vitamin K and placebo groups were 5836 [+ or -] 537 and 5745 [+ or -] 748 g, respectively.

Table 2 shows the maternal plasma and milk vitamin K concentrations during the first 12 weeks of lactation. The vitamin K supplementation had a very significant effect on maternal plasma concentration at each time point, increasing the serum concentration by 25- to 50-fold. As in stage I, a maternal vitamin K supplement of 5 mg/d had a dramatic effect on breast milk vitamin K concentration, increasing it 70-fold.

TABLE 2. Maternal Plasma and Milk Concentration of Vitamin K During Lactation

 Concentration

 Before Study
Maternal plasma, ng/mL [+ or -] SD
 Vitamin K (n = 11) [+ or -] 
 Placebo (n = 11) [+ or -] 
Human milk, ng,/mL [+ or -] SD
 Vitamin K (n = 11) [+ or -] 
 Placebo (n = 11) [+ or -] 

 Concentration

 2 wk
Maternal plasma, ng/mL [+ or -] SD
 Vitamin K (n = 11) [+ or -] 
 Placebo (n = 11) [+ or -] (*)
Human milk, ng,/mL [+ or -] SD
 Vitamin K (n = 11) [+ or -] 
 Placebo (n = 11) [+ or -] (*)

 Concentration

 6 wk
Maternal plasma, ng/mL [+ or -] SD
 Vitamin K (n = 11) [+ or -] 
 Placebo (n = 11) [+ or -] (*)
Human milk, ng,/mL [+ or -] SD
 Vitamin K (n = 11) [+ or -] 
 Placebo (n = 11) [+ or -] (*)

 Concentration

 12 wk
Maternal plasma, ng/mL [+ or -] SD
 Vitamin K (n = 11) [+ or -] 
 Placebo (n = 11) [+ or -] (*)
Human milk, ng,/mL [+ or -] SD
 Vitamin K (n = 11) [+ or -] 
 Placebo (n = 11) [+ or -] (*)

(*) Placebo compared with vitamin K; P [is less than] .0001.

Table 3 shows the effect of maternal vitamin K supplements on infant plasma concentrations of vitamin K. Again, at each time point, the supplements had a highly significant effect on plasma vitamin K concentration in the infants, with a 6- to 10-fold increase in serum concentration.

TABLE 3. Infant Plasma Vitamin K Concentration

 Concentration, ng/mL + or - SD

 2wk

Vitamin K (n = 11) + or - 
Placebo (n = 11) + or - (*)

 Concentration, ng/mL + or - SD

 6wk

Vitamin K (n = 11) + or - 
Placebo (n = 11) + or - ([dagger])

 Concentration, ng/mL + or - SD

 12wk

Vitamin K (n = 11) + or - 
Placebo (n = 11) + or - ([dagger])

(*) Vitamin K versus palcebo; P = .016.

([dagger]) Vitamin K versus placebo; P < .001.

Table 4 shows the breast milk and phylloquinone intakes in the two groups per kilogram of body weight. Milk intake was significantly greater in the placebo group at 2 weeks. The vitamin K supplement had a very significant effect on infant vitamin K intake at all time points, increasing intakes 50-fold.

TABLE 4. Infant Human Milk (ml/kg/day)
and Vitamin K Intakes ([micro]g/kg/day) [+ or -] SD

Intake 2 wk

Milk, mL/kg/d
 Vitamin K (n = 11) 120 + or - 32
 Placebo (n = 11) 158 + or - 37(*)
Vitamin K, [micro]g/kg/d
 Vitamin K (n = 11) + or - 
Placebo (n = 11) + or - ([dagger])

Intake 6 wk

Milk, mL/kg/d
 Vitamin K (n = 11) 119 + or - 19
 Placebo (n = 11) 142 + or - 33
Vitamin K, [micro]g/kg/d
 Vitamin K (n = 11) + or - 
Placebo (n = 11) + or - ([dagger])

Intake 12 wk

Milk, mL/kg/d
 Vitamin K (n = 11) 113 + or - 22
 Placebo (n = 11) 127 + or - 34
Vitamin K, [micro]g/kg/d
 Vitamin K (n = 11) + or - 
Placebo (n = 11) + or - ([dagger])

(*)Significantly different from vitamin K group; P = .018.

([dagger]) Vitamin K versus placebo; P < .001.

For infants in the placebo group, the correlation of vitamin K intake versus plasma vitamin K concentration was not significant (r = .114; P = .55). However, for the vitamin K-supplemented group, there was significant correlation between infant vitamin K intake and plasma concentration (r = .39; P = .03).

Finally, Table 5 shows the PIVKA-II values and PTs at 12 weeks of age, at a time we have previously shown that the vitamin K plasma concentration in unsupplemented breastfed infants is at its low Although PIVKA-II was significantly increased in the placebo group compared with the supplemented group, only one infant in the placebo group had a PIVKA-II value ( ng/mL) out of the normal adult range (up to 2 ng/mL). This infant did not have any signs or symptoms of hemorrhagic disease. The highest PIVKA-II value in the vitamin K group was ng/mL. PTs did not differ at any time point during the study.

TABLE 5. Des-[gamma]-carboxy-prothrombin (PIVKA-II) and
Prothrombin Time at 12 Weeks(*)

Group PIVKA II, ng/mL([dagger])

Vitamin K (n = 8) + or - 
Placebo (n = 8) + or - [double dagger]

Group Prothrombin Time, s

Vitamin K (n = 8) + or - 
Placebo (n = 8) + or - 

(*) All values are means + SD.

([dagger]) Normal adult value, less than 2 ng/mL.

([double dagger]) Vitamin K versus placebo, P = .018.

The daily dietary intakes (not including supplements) of vitamin K in stage II are not reported. Values for this population were reported in our previous study and did not differ in the present study.[6] Daily averages were less than 500 [micro]g/d and did not differ between groups when the daily supplement was excluded for the vitamin K group. However, as noted previously, there were large individual differences because of the variation in intake of high-vitamin K-containing foods (eg, broccoli and cauliflower). All mothers exceeded the daily recommended intake of 1 [micro]g/kg per day without the vitamin K supplement at every time point during the study.

DISCUSSION

In our previous publication,[6] we demonstrated that infants exclusively breastfed had phylloquinone intakes of less than 1 [micro]g/kg per day (the RDA) persistently. Plasma phylloquinone concentrations were less than the adult normal value ( ng/mL) by 6 weeks of age, an observation that persisted through 26 weeks of breastfeeding. PIVKA-II determinations were not available to us in this first study. Also, in this earlier report,[6] we noted the very low concentrations of vitamin K in human milk despite the vitamin K intakes of mothers, which consistently exceeded the recommended dietary intake of one [micro]g/kg per day. On the other hand, infants receiving standard formula (Similac) containing at least 55 ng/mL added phylloquinone had vitamin K intakes of at least 10 times the RDA. As a result, plasma phylloquinone values in these infants were 10 times the adult normal value of ng/mL.

Our aim in the present study was to increase the phylloquinone content of human milk significantly to levels comparable with that in infant formula. This was accomplished with a maternal supplement of 5 mg/d. The resulting milk concentration ranged from 75 to 82 ng/mL. Although this exceeded the label claim of the phylloquinone content of the formula (55 ng/mL) used in our earlier study, recent analysis of that formula (Similac, 20 calories/oz) determined that the phylloquinone concentration was 120 ng/mL. (F. R. Greer, and J. W. Suttie, unpublished data, 1994). Similarly, our aim was to increase the plasma phylloquinone concentrations in the plasma of breastfeeding infants from much less than that of the adult normal to values closer to those of formula-fed infants found in the previous study. With the breast milk concentrations we achieved, plasma phylloquinone levels in the breastfeeding infants averaged to ng/mL (up from a mean of ng/mL in the placebo group; see Table 3) compared with values of to ng/mL in the formula-fed infants in the previous study.[6]

In this study we achieved the vitamin K intake of the formula-fed infants in the previous study, .41 [micro]g/kg per day compared with to [micro]g/kg per day. However, in our previous study, we calculated the vitamin K intakes based on the product label rather than the measured amount determined since that study's completion. Thus, in point of fact, the vitamin K intakes of the formula fed-infants in the previous study were likely 50% greater than that originally reported.[6] This may account for the fact that in the present study, plasma phylloquinone concentrations in breastfed infants only reached 50% of those of the formula-fed infants in the original study.

PTs at 12 weeks showed no significant differences in the two groups. PIVKA-II values were significantly elevated in the placebo group compared with the vitamin K group, indicating increased levels of partially carboxylated prothrombin. This may be indicative of a relative vitamin K deficiency in this group, although in point of fact only one infant had a PIVKA-II value greater than ng/mL, the upper limit of the adult normal value. This is the first data reported on US infants using this specific monoclonal antibody for PIVKA-II. We regret that limited supplies of the PIVKA-II antibody did not allow for PIVKA-II measurements at baseline and 2 and 6 weeks. None of the infants in this group had any bleeding episodes, and we speculate that much higher values of PIVKA-II would occur in infants with significant hemorrhagic disease of the newborn.

In conclusion, we have demonstrated that in exclusively breastfed infants who received intramuscular phylloquinone at the time of birth, the vitamin K status as measured by plasma phylloquinone and PIVKA-II concentrations is improved by maternal oral supplements of 5 mg/d phylloquinone through the first 12 weeks of life. Because at the present time there are no licensed oral preparations of vitamin K for infants in the United States, this may be an alternative method of supplementation in situations in which parents refuse intramuscular newborn vitamin K prophylaxis at the time of birth. However, even in this group an oral vitamin K supplement at the time of birth would be ideal, given the decreased intake of breast milk during the first few days of life and the risk of the very early onset form of hemorrhagic disease of the newborn.[1]

ACKNOWLEDGMENTS

This work was funded by United States Department of Agriculture grant 82-CSRS-9201-1205, National Institutes of Health grant DK-14881, and a grant from the Medical Surgical Foundation, Meriter Hospital (Madison, WI).

We acknowledge the technical support of Ann Nutt, perinatal pharmacist at Meriter Hospital, for help in preparing the vitamin K supplements and placebo used in this study. We also thank Diagnostica Stago for providing the monoclonal antibody for measuring PIVKA-II.

REFERENCES

[1.] Greer FR, Suttie JW. Vitamin K and the newborn. In: Tsang RC, ed. Nutrition During Infancy. Philadelphia, PA: Hanley & Belfus; 1988: 289-297

[2.] Khayata S, Kindberg C, Greer FR, et al. Vitamin [.1] and vitamin [.2] in infant human liver. J Pediatr Gastroenterol Nutr. 1989;8:304-307

[3.] Shirahata A, Nakamur T, Ariyoshi N. Vitamin [.1] and [.2] contents in blood, stool, and liver tissues of neonates and young infants. In: Suzuki S, Hathaway WE, Bonnar J, Sutor AH, eds. Perinatal Thrombosis and Hemostasis. Tokyo, Japan: Springer-Verlag; 1991:213-223

[4.] Thierry MJ, Hermodson MA, Suttie JW. Vitamin K and warfarin distribution and metabolism in the warfarin-resistant rat. Am I Physiol. 1970; 219:854-859

[5.] Canfield LM, Hopkinson JM, Lima AF, et al. Vitamin K in colostrum and mature human milk over the lactation period--a cross sectional study. Am J Clin Nutr. 1991;53:730-735

[6.] Greer FR, Marshall S, Cherry J, et al. Vitamin K status of lactating mothers, human milk and breast feeding infants. Pediatrics. 1991;88: 7;1-756

[7.] Golding J, Greenwood R, Birmingham K, et al. Childhood cancer, intramuscular vitamin K, and pethidine given during labor. Br Med J. 1992;305:341-346

[8.] Food and Nutrition Board, Commission on Life Sciences, National Research Council. Recommended Dietary Allowances. 10th ed. Washington DC: National Academy Press; 1988:111

[9.] Von Kries R, Greer FR, Suttie JW. The assessment of the vitamin K status of the newborn--a review. J Pediatr Gastroenterol Nutr. 1993;16: 231-238

[10.] Babson SG, Behrman RE, Lessel R. Liveborn birth weights for gestational age of white middle class infants. Pediatrics. 1970;45:937-944

[11.] Pennington JAT, Church HN. Bowes and Church's Food Values of Portions Commonly Used. 13th ed. Philadelphia, PA: JB Lippincott Co; 1980

[12.] Butte NF, Garza C, Smith EO, et al. Human milk intake and growth in exclusively breast-fed infants. I Pediatr. 1984;104:187-195

[13.] Greer FR, Mummah-Schendel LL, Marshall S, et al. Vitamin [.1] (phylloquinone) and vitamin [.2] (menaquinone) status in newborns during the first week of life. Pediatrics. 1988;81:137-140

[14.] Schubiger G, Tonz O, Gruter J, et al. Vitamin [.1] concentration in breast-fed neonates after oral or intramuscular administration of a single dose of a new mixed-micellar preparation of phylloquinone. J Pediatr Gastroenterol Nutr. 1993;16:435-439

[15.] Haroon Y, Bacon DS, Sadowski JA. Liquid-chromatographic determination of vitamin [.1] in plasma with fluorometric detection. Clin Chem. 1986;32:1925-1929

[16.] Haroon Y, Shearer MJ, Rahim S, et al. The content of phylloquinone (vitamin [.1]) in human milk, cows' milk, and infant formula foods determined by high performance liquid chromatography. 1 Nutr. 1982; 112:1105-1117

[17.] Amiral J, Grosley M, Plassart V, Mimilla F, Chambrette B. Development of a monoclonal immunoassay for the direct measurement of decarboxy prothrombin in plasma. Thromb Haemost. 1991;65:64