European Journal of Clinical Nutrition (2002) 56, 387–392 ß 2002 Nature Publishing Group All rights reserved 0954–3007/02 $25.00 www.nature.com/ejcn ORIGINAL COMMUNICATION Vitamin B6 intakes and status of mechanically ventilated critically ill patients in Taiwan

Y-C Huang1*, P-H Lan1, C-H Cheng2, B-J Lee3 and M-N Kan3

1School of Nutrition, Chung Shan Medical University, Taiwan, Republic of China; 2Critical Care and Respiratory Therapy, Taiwan, Republic of China; and 3Intensive Care Unit, Taichung Veterans General Hospital, Taichung, Taiwan, Republic of China

Objective: To assess vitamin B6 intake and status of critically ill patients. The relationship between vitamin B6 status indicators and the severity of illness and outcome in these patients was also examined. Design: Prospective clinical study. Setting: The study was performed at the Taichung Veteran General Hospital, in the central part of Taiwan. Subjects: Ninety-four patients in the intensive care unit (ICU) entered the study and 46 patients successfully completed this study. Interventions: No intervention. Main outcome measures: Vitamin B6 intake was recorded for 14 days. Vitamin B6 status was assessed by direct measures 0 (plasma pyridoxal 5 -phosphate (PLP), pyridoxal (PL), and urinary 4-pyridoxic acid (4-PA)) and indirect measures (erythrocyte alanine (EALT-AC) and aspartate (EAST-AC) aminotransaminase activity coefficient). The severity of illness (APACHE II score), the length of ventilation dependency, and the length of ICU and hospital stay were recorded. Results: Patients had an adequate mean vitamin B6 intake (16.26Æ 19.39 mg) during the 14 day study. Mean vitamin B6 intake was significantly higher on day 14 than on day 1 (P < 0.001). However, plasma PLP and PL concentrations significantly decreased at the 14th day after admission (P < 0.05). Erythrocyte alanine aminotransaminase activity coefficient and EAST-AC did not change significantly. Urinary 4-PA significantly increased at the 14th day (P < 0.001). No significant relationships were found between APACHE II scores and clinical outcomes (the length of ICU and hospital stay, the length of ventilation dependency) of patients, vitamin B6 intake or status indicators. Conclusions: Critically ill patients received nutritional support in the ICU, and had sufficient mean vitamin B6 intake and adequate vitamin B6 status. Therefore, the severity of illness and the results should not be affected by vitamin B6 status. However, we have noted that plasma PLP and PL concentrations significantly decreased while vitamin B6 intake significantly increased on day 14. Critical clinical conditions and complex metabolism in the critically ill may account for the reduction of plasma PLP and PL. Since vitamin B6 deficiency causes profound effects on immune system function, dietary or supplemented vitamin B6 intake is suggested for hospitalized patients. European Journal of Clinical Nutrition (2002) 56, 387 – 392. DOI: 10.1038=sj=ejcn=1601321

Keywords: vitamin B6; pyridoxal 5’-phosphate; pyridoxal; transaminase; 4-pyridoxic acid; critically ill

Introduction Several studies (Potera et al, 1977; Vir & Love, 1978; Louw *Correspondence: Y-C Huang, No. 110 Section 1 Chien-Kuo N Road, School of Nutrition, Chung Shan Medical University, Taiwan, Republic of et al, 1992; Pfitzenmeyer et al, 1997) have reported lower China. vitamin B6 status in hospitalized patients. Potera et al (1977) Guarantor: YC Huang. reported that advanced breast cancer patients had signifi- Contributors: Y-CH was responsible for the development of cantly lower plasma PLP levels than healthy women of intellectual content and the study design, interpretation of the results and manuscript drafting. P-HL was responsible for nutrient data similar age. Vir and Love (1978) investigated vitamin B6 coding and sample analyses. C-HC was responsible for the screening status in 102 hospitalized aged people. There was a high of subjects and interpretation of the results. B-JL was responsible for prevalence of lower vitamin B6 status in both aged male and the screening of subjects. M-NK was responsible for the critical female subjects. They (Vir & Love, 1978) suggested that a revisions of manuscript. Received 25 April 2001; revised 31 July 2001; diet higher in vitamin B6 should be given to aged patients. accepted 1 August 2001 Louw et al (1992) found PLP decreased to values below the Vitamin B6 intake of critically ill patients Y-C Huang et al 388 respective normal range during the acute-phase response. to the patients based on the physician’s decision. Daily However, little is known about the intake and nutritional macronutrients (carbohydrate, lipid and protein) and vita- status of vitamin B6 in critically ill patients who have more min B6 intakes from EN, TPN and intravenous crystalloid complicated clinical conditions. infusions were recorded routinely by the ICU nurses and The physiological active coenzyme form of vitamin B6, dietitians. Medical records including diagnosis, the APACHE PLP, is required for normal nucleic acid and protein synthesis II score, length of ICU stay and of hospital stay, and venti- and for cellular multiplication. Isolated vitamin B6 deficien- latory dependency were obtained. Patient’s age, sex, height, cies cause more profound effects on immune system func- weight and severity of illness on admission were assessed tions than deficiencies of any other B group vitamins within 24 h of admission and again after 2 weeks in the ICU. (Leklem, 1988). Thus, in critically ill patients, having a The body mass index (BMI; kg=m2) was calculated. The lower vitamin B6 status may depress immune responses. severity of illness was evaluated by the ACPACHE II score 0 Pyridoxal 5 -phosphate and PL are both bound to serum (Knaus et al, 1985). Informed consent for participation was albumin while being transported by the blood. The binding obtained. The study was approved by the Committee for of PLP and PL to serum albumin in the circulation serves to Ethics of Chung Shan Medical and Dental College. protect it from hydrolysis and allows for the delivery of PLP and PL to other tissues. Critically ill patients often have a compromised protein status (Bassili & Deitel, 1981; Larca & Greenbaum, 1982; Christman & McCain, 1993; Huang et al, 2000). The question has been raised whether vitamin B6 Biochemical measurements status is affected in these patients. Fasting venous blood samples were collected to estimate The purpose of this study was to assess vitamin B6 intake hematological (ie complete blood count, total lymphocyte and vitamin B6 status of critically ill patients by using direct count (TLC), serum albumin, and hemoglobin) and vitamin measures (plasma PLP and PL, and urinary 4-PA) and indirect B6 status on the 1st and 14th day in the ICU. Plasma and red measures (EALT-AC and EAST-AC). In addition, the relation- blood cells were separated and stored frozen (720C) until ship between vitamin B6 status parameters, the severity of analysis. Plasma and erythrocytes were prepared as pre- illness (Acute Physiology and Chronic Health Evaluation; viously described (Woodring & Storvick, 1970). APACHE II score) and clinical outcomes (the length of ICU Plasma PLP and PL were determined by high performance and hospital stay, the length of ventilation dependency) in liquid chromatography (HPLC) as previously described these patients was also examined. (Edwards et al, 1989). The fluorescence detector excitation and emission wavelengths were 300 and 400 nm, respec- tively. Plasma PLP and PL concentrations were identified and quantified by comparing retention times and peak Methods areas with standards. Erythrocyte alanine aminotransferase Patients and erythrocyte EAST with and without PLP stimulation in This study was conducted at the Taichung Veteran General vitro were measured by the method of Woodring and Storvick Hospital, which is a 1359-bed teaching hospital in the (1970). The interassay variabilities of plasma PLP, PL, EALT- central part of Taiwan. All patients admitted or transferred AC and EAST-AC were 11.7% (n ¼ 20), 11.9% (n ¼ 20), 5.0% to the ICU were screened for potential study participation. (n ¼ 10) and 2.1% (n ¼ 18), respectively. Plasma B6 concen- Only patients completing the study and requiring at least 14 trations and transaminase activity measurements were days of mechanical ventilation were included. Patients were carried out under yellow light to prevent photodestruction. excluded if they were clinically unstable or unconscious Twenty-four-hour urine collections were obtained on the during the study. Ninety-four patients entered the study, 1st and 14th day in the ICU. The pH of the urine samples was however only 46 patients (36 males, 10 females) successfully maintained at  3.0 by the addition of 1 N HCl. Twenty-four- completed this study. All patients received nutritional sup- hour urine volumes were determined and aliquots stored port either enteral nutrition (EN; n ¼ 12), total parenteral frozen (720C) until analysis. Urinary 4-PA was analyzed by nutrition (TPN; n ¼ 16) or combined (EN plus TPN; n ¼ 18) HPLC by the modified method of Gregory and Kirk (1979). based on the physician’s recommendations. Patients Initial addition of trichloroacetic acid (TCA) to the urine was received an enteral feed via a nasogastric tube. The composi- omitted in order to avoid formation of the 4-PA lactone. tion of tube feeding formula was prepared from various Instead the urine samples were centrifuged at 3000 g for commercial formula based on the patient’s nutritional 10 min to remove particulate material and syringe filtered needs. The TPN regimen was given as a continuous infusion (0.45 mm; Nalge Co., Rochester, NY). Urine was then injected via a superior venacaval vein. The parenteral feed contained directly onto the HPLC column. All urine samples thawed for glutamine-free crystalline L-amino acids, carbohydrate in the 4-PA determination were kept on ice to prevent lactonization form of 25% dextrose, 10 or 20% fat emulsion (1.1 kcal=ml or of 4-PA. Creatinine concentrations were determined in all 2 kcal=ml), vitamins and minerals. Additional multivitamin urine samples using a colorimetric kit (Sigma Chemical Co., supplements (containing 15 mg vitamin B6) were also given St Louis, MO).

European Journal of Clinical Nutrition Vitamin B6 intake of critically ill patients Y-C Huang et al 389 Statistical analyses the ICU are listed in Table 3. Patients had significantly higher Data were analyzed by using SigmaStat statistical software energy (P < 0.01) and vitamin B6 (P < 0.001) intakes, and the (version 2.03; Jandel Scientific, San Rafael, CA). Pearson ratio of vitamin B6 to protein (mg=g; P < 0.01) on the 1st correlation coefficients were performed to assess the relation- than on the 14th day after admission to the ICU. The mean ship between dietary variables and vitamin B6 status vari- vitamin B6 intake on the 1st and 14th days was higher than ables. Differences in patients’ anthropometric measure- the US Dietary Reference Intakes (Food and Nutrition Board, ments, nutrient intakes, biochemical values, the length of Institute of Medicine, 1998; 1.7 mg=day for elderly men and ICU stay and hospital stay, and the length of ventilatory 1.5 mg=day for elderly women) and the current Taiwan dependence were recorded. Medical records were compared Recommended Daily Nutrient Allowance (Department of for significant differences on the 1st and the 14th day in the Health, Taiwan, 1993) of 1.6 mg=day for elderly men and ICU using Student’s t-test or the Mann – Whitney rank test. 1.4 mg=day for elderly women. Although many patients have Pearson correlation coefficients were used to assess the adequate vitamin B6 intake, individual variation was also relationship between vitamin B6 status parameters with considered. Individual vitamin B6 intake ranged from 0 to nutrient intakes, the length of ventilator dependency, 87.8 mg on the 1st day and 0 to 86.5 mg on the 14th day. APACHE II score, length of ventilator dependency, and length of time in ICU or hospital stay. Statistical results were considered to be significant at P < 0.05. Values Vitamin B6 status presented in the text and tables are meansÆ standard Biochemical indicators of vitamin B6 status on 1st and 14th deviation (s.d.). day after admission to the ICU are shown in Table 4. Although patients had a mean plasma PLP concentration higher than a cutoff of 20 nmol=l (Food and Nutrition Board Results Institute of Medicine, 1998; Lui et al, 1985), mean plasma Characteristics of all patients are shown in Table 1. Patients’ PLP and PL concentrations significantly (P < 0.05) decreased ages ranged from 23 to 85 y. Mean current weight and BMI on the 14th day when compared with the value on the 1st did not significantly change between the 1st and the 14th day. We did not find that vitamin B6 or protein intake day after admission. There were no significant changes for affected plasma PLP or PL concentrations. APACHE II score comparing day 1 with day 14. The mean The mean EALT-AC and EAST-AC values did not signifi- duration of time on the ventilator during hospital stay was cantly change between the 1st and 14th day in the ICU 56 days with a range of 17 – 193 days. The mean duration of (Table 4). The mean EALT-AC was < 1.25 on the 14th day, stay in the ICU was 39 days (range 17 – 98 days) and in the the suggested value for adequate vitamin B6 status (Leklem, hospital was 75 day (range 21 – 346 days) for all patients. No 1990). The mean EAST-AC value of patients was also lower patients died during the study. The diagnosis of patients at than the suggested value ( < 1.8) for adequate vitamin B6 the time of admission to the ICU is listed in Table 2. The status (Leklem, 1990). Dietary energy, protein or vitamin B6 most common diagnoses were gastrointestinal disorders and intakes were not related to EALT-AC and EAST-AC values malignant neoplasms. during patients’ stay in the ICU. On the 1st and 14th day in the ICU, urinary 4-PA excre- tion was determined to be 6.9Æ 5.2 and 9.4Æ 4.1 mmol=day, m = Dietary intakes respectively. The values were above 3.0 mol day, which is Energy, protein, vitamin B6, and the ratio of vitamin B6 to protein (mg=g) on the 1st and 14th day after admission to Table 2 The diagnosis of patients at the time of admission to the ICU

Diagnosis Number of patients (n ¼ 46) Table 1 Patients’ characteristics, severity of illness (APACHE II score), a the length of ventilatory dependency, and the length of time in the ICU Gastrointestinal disorders 19 b and hospital on the 1st and 14th day of admission to the ICU Malignant neoplasms 12 Infectionc 6 d Day 1 Day 14 Respiratory diseases 4 e Characteristics (n ¼ 46) (n ¼ 46) Trauma 3 Cardiovascular diseases (aneurysm) 1 Age (y) 65.4Æ l4.2a Burn 1 Sex (male=female) 36=10 a Height (cm) 162.3Æ 7.4 Gastrointestinal disorders include peptic ulcer, enteritis, acute pancreatitis, Weight (kg) 62.2Æ 11.6 62.7Æ 10.7 acute cholecystitis, gastric ulcer, duodenal ulcer, peritonitis and intestinal Body mass index (kg=m2) 23.6Æ 3.9 23.8Æ 3.8 obstruction. b APACHE II score 18.2Æ 4.6 17.5Æ 5.8 Malignant neoplasms include esophageal tumor, hypopharyngeal tumor, Length of ventilatory dependence (days) 56.4Æ 34.5 and colon cancer c Length of ICU stay (days) 38.6Æ 20.2 Infection includes deep neck infection, viral meningitis and acute maxillary Length of hospital stay (days) 75.3Æ 55.1 sinusitis. dRespiratory diseases include pneumonia. aValues are meanÆ s.d. eTrauma includes pelvic open fracture and diaphragm rupture.

European Journal of Clinical Nutrition Vitamin B6 intake of critically ill patients Y-C Huang et al 390 Table 3 Daily nutrient intakes of patients on the 1st and 14th day of admission to the ICU

Day 1 Day 14 14 days average Nutrients (n ¼ 46) (n ¼ 46) (n ¼ 46)

Energy (MJ=day) 6.3Æ 2.7 7.8Æ 2.4 7.6Æ 1.9 (0.3 – 13.4) (0.5 – 14.0) (4.2 – 13.4) (kcal=day) 1511.1Æ 656.8*,a 1867.5Æ 568.0 1824.0Æ 443.4 (77.3 – 3194.8) (111.5 – 3350.5) (1004.3 – 3196.4) Protein (g) 58.6Æ 37.9 60.1Æ 26.7 65.97Æ 21.77 (0 – 149.4) (4.0 – 130.0) (26.6 – 117.1) Vitamin B6 (mg) 9.3Æ 16.4* 17.5Æ 19.6 16.3Æ 19.4 (0 – 87.8) (0 – 86.5) (0.4 – 92.9) Vitamin B6=protein ratio 0.326Æ 0.291* 0.414Æ 1.855 0.262Æ 0.299 (0 – 12.5) (0 – 0.955) (0.005 – 6.578)

aValues are meanÆ s.d.with the range in the parentheses. *Values are significantly different from the 14th day and 14 day average of admission to the ICU.

Table 4 Hematological measurements and biochemical indicators of vitamin B6 status indicators (plasma PLP, PL, EALT-AC, EAST- vitamin B6 status on the 1st and 14th days of admission to the ICU AC, and urinary 4-PA; data not shown). Day 1 Day 14 Indicators (n ¼ 46) (n ¼ 46) Discussion Hematological measurements Hemoglobin (g=dl) 9.67Æ 2.05a,b 9.91Æ 1.13b To our knowledge, this is the first study to report vitamin B6 (normal: 11 – 18g=dl) intake and status in critically ill patients. A variety of meth- Albumin (g=dl) (normal: 3.5 – 5.0 g=dl) 2.43Æ 0.45b 2.50Æ 0.47b ods have been utilized to assess vitamin B6 status in this 3 ,b Total lymphocyte count (mm ) 795.47Æ 791.41* 1210.34Æ 880.50 study including direct methods (plasma PLP and PL, urinary (normal: > 500 mm3) Vitamin B6 status indicators 4-PA), indirect methods (EALT, EAST and stimulation with Plasma PLP) and vitamin B6 intake. Pyridoxal 5’-phosphate (nmol=l) 42.74Æ 14.52* 34.57Æ 14.83 Very little data on vitamin B6 consumption in hospita- Pyridoxal (nmol=l) 28.51Æ 9.55* 23.19Æ 14.03 lized patients has been reported. Vir and Love (1978) Erythrocyte EALT-AC 1.32Æ 0.60 1.20Æ 0.52 reported that 18.6% of patients consumed less than EAST-AC 1.41Æ 0.38 1.33Æ 0.34 0.66 mg of vitamin B6 per day and only 21.6% of the patients Urine had a vitamin B6 intake higher or equal to 1 mg. In an other 4-pyridoxic acid (mg=ml) 1.30Æ 0.95* 1.97Æ 1.48 study (Louw et al, 1992), 25 subjects who underwent uncom- 4-pyridoxic acid (mmol=day) 6.89Æ 5.15* 9.41Æ 4.12 4-pyridoxic acid=creatinine (mmol=g) 7.70Æ 6.08* 12.52Æ 6.53 plicated orthopedic surgery had an average of 1.7 and 1.6 mg vitamin B6 intake for men and women, respectively, 3 days a Values are meanÆ s.d. after surgery. Elderly patients in the study of Powers et al bValues are lower than the normal value. (1993) consumed about 0.5 – 4mg=day pyridoxine. Mean *Values are significantly different between the 1st and 14th day of admission to the ICU. vitamin B6 intake of our critically ill patients was certainly higher than the previous studies (Vir & Love, 1978; Louw considered to be associated with adequate vitamin B6 status et al, 1992; Powers et al, 1993). This is probably because all (Leklem, 1990). Unlike the results of plasma PLP and PL our patients received nutritional support (EN, TPN or com- concentrations, urinary 4-PA concentration increased signif- bined) in the ICU; patients often get micronutrient supple- icantly after 14 days. ments to meet their requirements. However, it is worth The results of hematological measurements are also noting that our patients had varied vitamin B6 intake, shown in Table 4. The hemoglobin, albumin and TLC receiving anywhere from 0.4 to more than 92 mg vitamin levels were on average below normal for patients on day 1. B6 during the 14 day study (Table 3). Physicians, nurses Patients showed no significant changes in hemoglobin, total and=or dietitians should carefully identify the amount of protein, and albumin levels by day 14, but TLC increased nutrients patients receive in the ICU. Our study group had significantly by day 14. Only plasma PL concentration cor- sufficient mean vitamin B6 intake. This might be the reason related significantly negatively with TLC (r ¼ 70.365, vitamin B6 intake did not have any effect on the APACHE II P ¼ 0.0013) on day 1. Otherwise, no significant correlation score, the length of ventilation dependency, and the length was found between hematological measurements and vita- of time in the ICU and hospital. min B6 status indicators. It is not surprising to observe an adequate mean plasma The severity of illness (APACHE II score), the length of PLP concentration ( > 20 nmol=l) in the critically ill since our ventilation dependency, the length of time in the ICU and patients had higher vitamin B6 intake than other studies (Vir hospital did not correlate with vitamin B6 intake or with any & Love, 1978; Louw et al, 1992; Powers et al, 1993). An

European Journal of Clinical Nutrition Vitamin B6 intake of critically ill patients Y-C Huang et al 391 interesting finding was that mean plasma PLP concentration portionately to vitamin B6 intake. In agreement with pre- significantly decreased while vitamin B6 intake significantly vious reports (Brown et al, 1975; Leklem, 1990; Huang et al, increased by day 14. There was also no correlation between 1998), urinary 4-PA excretion changed rapidly in response to vitamin B6 intake and plasma PLP levels. In agreement with a change in vitamin B6 intake. Powers et al (1993), the lack of responsiveness of plasma PLP Vitamin B6 has a role in erythrocyte function and meta- to vitamin B6 intake might be due to high doses of vitamin bolism (Bottomley, 1983). Pyridoxine-responsive anemias in B6 supplementation in our critically ill patients during their humans have been reported (Horrigan & Harris, 1968). ICU stay. Patients, therefore, might have sufficient PLP Patients had low hemoglobin concentrations in our study; storage in the liver. However, hepatic PLP reserves might however, vitamin B6 status revealed no effect on hemoglo- decrease as the critical clinical condition continues to exist bin concentration. This finding agrees with that of other in the critically ill. In addition, plasma PLP might redistri- studies (Ribaya-Mercado et al, 1991; Huang et al, 1998). bute from plasma to erythrocytes during the inflammatory Patients had low hemoglobin concentrations which may be state (Driskell, 1994). It has been suggested that erythrocyte caused by other factors. PLP is a better indicator of vitamin B6 status than plasma PLP level (Leklem, 1990; Lumeng et al, 1978). Erythrocyte PLP status should be considered when assessing vitamin B6 status Conclusion in the critically ill. Critically ill patients received appropriate and aggressive Pyridoxal 5’-phosphate and PL are both bound to serum nutritional support in the ICU; they had sufficient mean albumin, with the PLP binding more tightly, against rapid vitamin B6 intake and adequate vitamin B6 status. Therefore, hydrolysis by alkaline phosphatase while being transported the severity of illness and clinical outcomes (the length of by the blood. Lower serum albumin levels in our critically ill ventilation dependency, and the length of time in the ICU patients may lead to the dephosphorylation of plasma PLP and hospital) might not be affected by vitamin B6 intake and into PL (Merrill & Henderson, 1987). Because dietary vitamin status indicators. However, we have noted that plasma PLP B6 and protein intake as well as plasma alkaline phosphatase and PL concentrations significantly decreased while vitamin activity may affect plasma PLP concentration, recent B6 intake significantly increased on day 14. Critical clinical research has suggested that plasma PL may be useful as an conditions and complex metabolism in the critically ill may indicator for vitamin B6 status (Leklem, 1990). We did account for the reduction of plasma PLP and PL. Since observe a significant amount of plasma PL concentration in vitamin B6 is required for normal nucleic acid and protein our critically ill patients. Unfortunately, we could not com- synthesis and for cellular multiplication, vitamin B6 defi- pare plasma PL concentration of our patients with other ciency causes profound effects on immune system function studies because no studies reported plasma PL concentration (Axlerod 1971; Talbott et al, 1987). Dietary or supplemented in hospitalized patients. vitamin B6 intake is suggested for hospitalized patients. Like plasma PLP, our critically ill patients had an adequate EALT-AC ( < 1.25) and EAST-AC ( < 1.8) level (Leklem, 1990) during their ICU stay. Hoorn et al, (1975) indicated that mean EAST-AC level was 1.58Æ 0.14 in 153 geriatric patients, which is considered to be adequate. Te´bi et al (2000) reported that 590 diseased elderly people had an adequate mean References Axelrod AE (1971): Immune processes in vitamin deficiency states. EAST-AC level (1.75Æ 0.38). However, studies (Lui et al, Am. J. Clin. Nutr. 24, 265 – 271. 1985; Ribaya-Mercado et al, 1991) indicated that EAST-AC Bassili HR & Deitel M (1981): Effect of nutritional support on could be normal in the presence of chronic and marginal weaning patients off mechanical ventilators. J. Parent. Enternal vitamin B6 deficiency. Therefore, EALT-AC rather than EAST- Nutr. 5, 161 – 163. Bottomley SS (1983): Iron and vitamin B6 metabolism in the side- AC is a more sensitive indicator of vitamin B6 status (Brown roblastic anemias. In Nutrition in Hematology, ed. J Lindenbaum, et al, 1975; Leklem, 1990; Cochary et al, 1990). On the other pp. 203 – 223. New York: Churchill Livingston. hand, no relationship was found between EALT-AC or EAST- Brown RR, Rose DP, Leklem JE, Linkswiler H & Anand R (1975): AC levels with vitamin B6 intake or with any vitamin B6 Urinary 4-pyridoxic acid, plasma pyridoxal phosphate and ery- throcyte aminotransferase levels in oral contraceptive users receiv- status indicators in the present study. Kretsch et al (1995) ing controlled intakes of vitamin B6. Am. J. Clin. Nutr. 28,10– 19. indicated that the aminotransferases were more subject to Cochary EF, Gershoff SN & Sadowski JA (1990): Aging and vitamin B- individual variability; therefore, caution should be taken in 6 depletion effects on plasma pyridoxal 5’-phosphate and erythro- using EALT or EAST values alone to interpret vitamin B6 cyte aspartate aminotransferase activity coefficients in rats. Am. J. Clin. Nutr. 51, 446 – 452. status. Christman JW & McCain RW (1993): A sensible approach to the The major excretory metabolite of vitamin B6 in humans nutritional support of mechanically ventilated critically ill is urinary 4-PA, which is considered to be a short-term patients. Intensive Care Med. 19, 129 – 136. indicator of vitamin B6 status. Our critically ill patients Driskell JA (1994): Vitamin B6 requirements of humans. Nutr. Res. 14, 293 – 324. not only had an adequate mean 4-PA concentration Department of Health, Taiwan, ROC (1993): Recommended Daily ( > 3.0 mmol=day; Leklem, 1990), but it also increased pro- Nutrient Allowances,56– 70, Department of Health, Taiwan, ROC.

European Journal of Clinical Nutrition Vitamin B6 intake of critically ill patients Y-C Huang et al 392 Edwards P, Liu PKS & Rose A (1989): A simple liquid-chromato- Louw JA, Werbeck A, Louw MEJ, Kotze TJvW, Cooper R, Labadarios D graphic method for measuring vitamin B6 compounds in (1992): Blood vitamin concentrations during the acute-phase plasma. Clin. Chem. 35, 241 – 245. response. Crit. Care Med. 20, 934 – 941. Food and Nutrition Board, Institute of Medicine (1998): Dietary Lui A, Lumeng L, Aronoff GR & Li TK (1985): Relationship Reference Intakes. Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, between body store of vitamin B6 and plasma pyridoxal-P clear- Vitamin B12, Pantothenic Acid, Biotin, and Choline. Prepublication ance: metabolic balance studies in humans. J. Lab. Clin. Med. 106, copy. Washington, DC: National Academy Press. 491 – 497. Gregory III JF & Kirk JR (1979): Determination of urinary 4-pyridoxic Lumeng L, Ryan MP & Li T-K (1978): Validation of the diagnostic acid using high performance liquid chromatography. Am. J. Clin. value of plasma pyridoxal 5’-phosphate measurements in vitamin Nutr. 32, 879 – 883. B-6 nutrition of the rat. J. Nutr. 108, 545 – 553. Hoorn RKJ, Flikweert JP & Westerink D (1975): Vitamin B-1, B-2 and Merrill AH & Henderson JM (1987): Diseases associated with B-6 deficiencies in geriatric patients, measured by coenzymes defects in vitamin B6 metabolism or utilization. A. Rev. Nutr. 7, stimulation of enzyme activities. Clin. Chim. Acta 61,151– 162. 137 – 156. Horrigan DL & Harris JW (1968): Pyridoxine responsive anemia in Pfitzenmeyer P, Guilland JC & d’Athis P (1997): Vitamin B6 and man. Vitam. Horm. 26, 549 – 568. vitamin C status in elderly patients with infections during hospi- Huang YC, Chen W, Evans MA, Mitchell ME & Shultz TD (1998): talization. Ann. Nutr. Metab. 41, 344 – 352. Vitamin B-6 requirement and status assessment of young women Potera C, Rose DP & Brown RR (1977): Vitamin B6 deficiency in fed a high-protein diet with varying levels of vitamin B-6. Am. J. cancer patients. Am. J. Clin. Nutr. 30, 1677 – 1679. Clin. Nutr. 67, 208 – 220. Powers JS, Zimmer J, Meurer K, Manske E, Collins JC & Greene HL Huang YC, Yen CL, Cheng CH, Jih KS & Kan MN (2000): Nutritional (1993): Direct assay of vitamins B1, B2, and B6 in hospitalized status of mechanically ventilated critically ill patients: comparison patients: relationship to level of intake. J. Parent. Enteral Nutr. 17, of different types of nutritional support. Clin. Nutr. 19, 101 – 107. 315 – 316. Knaus WA, Draper EA, Wagner DP & Zimmerman JE (1985): APACHE Ribaya-Mercado JD, Russel RM, Sahyoun N, Morrow FD & Gershoff II: a severity of disease classification system. Crit. Care Med. 13, SN (1991): Vitamin B6 requirement in elderly men and women. 818 – 829. J. Nutr. 121, 1062 – 1074. Kretsch MJ, Sauberlich HE, Skala JH & Johnson HL (1995): Vitamin B- Talbott MC, Miller LT & Kerkvliet NI (1987): Pyridoxine supplemen- 6 requirement and status assessment: young women fed a deple- tation: effect on lymphocyte responses in elderly persons. Am. J. tion diet followed by a plant- or animal-protein diet with graded Clin. Nutr. 46, 659 – 664. amounts of vitamin B-6. Am. J. Clin. Nutr. 61, 1091 – 1101. Te´bi A, Belbraouet S, Chau N & Debry G (2000): Plasma vitamin, b- Larca L & Greenbaum DM (1982): Effectiveness of intensive nutri- carotene, and a-tocopherol status according to age and disease in tional regimes in patients who fail to wean from mechanical hospitalized elderly. Nutr. Res. 20, 1395 – 1408. ventilation. Crit. Care Med. 10, 297 – 300. Vir SC & Love AHG (1978): Vitamin B6 status of the hospitalized Leklem JE (1988): Vitamin B-6 metabolism and function in humans. aged. Am. J. Clin. Nutr. 31, 1383 – 1391. In Clinical and Physiological Applications of Vitamin B6,eds.J Woodring MJ & Storvick CA (1970): Effect of pyridoxine supplemen- Leklem & RD Reynolds, 3 – 433. New York: Liss. tation on glutamic-pyruvic transaminase, and in vitro stimulation Leklem JE (1990): Vitamin B6: a status report. J. Nutr. 120, 1503 – in erythrocyte of normal women. Am. J. Clin. Nutr. 23, 1385 – 1507. 1395.

European Journal of Clinical Nutrition