Annals 0/ Clinical Biochemistry, 1978, 15, 307-312
Detection and incidence of Band C vitamin deficiency in alcohol-related illness M. BAINES From the Department 0/ Pathology, Broadgreen Hospital, Liverpool L14 3LB
The activity of the red blood cell enzymes transketolase, glutathione reductase, and aspartate transaminase, and their activation by the coenzymes thiamine, riboflavin, and pyridoxine, the pyruvate tolerance test, the leucocyte vitamin C concentration, and the activity in serum of yglutamyl transferase were measured in a series of 35 patients with alcohol-related illness. The incidence of thiamine deficiency was 31 % as assessed by the activation of transketolase, and 55 % as assessed by the pyruvate tolerance test. The incidence of riboflavin deficiency was 23 % and of ascorbic acid deficiency 91 %. No cases of pyridoxine deficiency were detected. The pyruvate tolerance test was found to be a more sensitive test of thiamine deficiency than the transketolase activation, and the activation of red blood cell aspartate transaminase was found to be a poor indicator of pyridoxine deficiency. There was a poor correlation of the y-glutamyl transferase activity with the degree of vitamin deficiency, suggesting that alcohol exposure is only partly responsible for the observed vitamin deficiency.
SUMMARY
The possibility of detecting B vitamin deficiency in a aim of the present study was to determine the relative hospital population has recently become a more sensitivity of the full pyruvate tolerance test and the feasible proposition with the introduction of methods red blood cell transketolase activation in detecting involving measurement of red blood cell enzyme vitamin Br deficiency in patients admitted to hospital activity with and without their co-enzyme activators. for alcohol abuse and to test recent suggestions Thus indices of vitamin Bi, B2,and B6 status can be (Baron, 1973; Whitby et al., 1975) that red blood cell obtained by measuring red blood cell transketolase transketolase activity appears to be the method of (TK, E.C. 2.2. 1.1) with and without thiamine choice for investigating possible thiamine deficiency. pyrophosphate, red blood cell glutathione reductase The two indices of vitamin Bi status described above, (GR, E.C. 1.6. 4.2) with and without flavin adenine together with indices of vitamins B2 and B6 and the dinucleotide, and red blood cell aspartate trans- measurement of leucocyte vitamin C, formed the aminase (AST, E.C. 2.6. 1.1) with and without assessment of the vitamin status of the patients under pyridoxal phosphate (Bayoumi and Rosalki, 1976). study in order to determine which of the vitamins A more traditional approach to the identification of are most commonly deficient. Finally, an attempt thiamine deficiency, a possible cause of neurological was made to correlate the degree of deficiency with disorders such as peripheral neuropathy in the the exposure to alcohol, as indicated by the serum alcoholic, has been the response of blood pyruvate activity of y-glutamyl transferase, an enzyme known to a glucose load, the so-called 'pyruvate tolerance to be induced by alcohol intake (Rosalki and Rau, test'. In the absence of its co-enzyme, thiamine pyro- 1972) and to be of significance as a test of hepatic phosphate, transketolase is inactivated, and the meta- injury in the alcohol abuser (Boone et al., 1977). bolism of pyruvate, formed from glucose, is impaired. The pyruvate tolerance test has been used by Materials and methods Morgan (1968) to detect thiamine deficiency in two groups of chronic alcoholics, those experiencing SUBJECTS hallucinations and those with no such symptoms. A group of 35 patients admitted to hospital for the Cambier et al. (1973) have compared a single estima- treatment of problems associated with alcohol abuse tion of pyruvic acid with a colorimetrically deter- was studied. As far as was possible all such hospital mined transketolase level as an index of thiamine admissions over a 15-monthperiodwereincluded. The status in patients with neurological disorders. The reasons for hospitalisation included peripheral 307
Downloaded from acb.sagepub.com by guest on June 4, 2016
M. Baines
308 neuritis and paresthesiae and ethanolic liver damage. Five patients were subsequently shown histologically to have developed cirrhosis. As it was not possible to determine the degree of alcohol dependence the term 'alcoholic', implying alcohol dependence, has been avoided. The group probably included those described as 'heavy drinkers' and 'alcoholics' by Rosalki and Rau (1972). A group of 21 healthy laboratory staff, age range 18-55, formed a reference population for the red blood cell enzyme studies. The control group for the pyruvate studies was composed of 15 subjects of age range 25-60 years. The population included laboratory staff and hospital inpatients who were due for discharge in an adequately fit and nourished condition. No subject in the control group had, as far as was known, any illness likely to affect the metabolism of glucose or any known vitamin deficiency. EXPERIMENTAL METHODS
Red blood cell enzymes These were measured in an haemolysate prepared as follows. A 5-10 mI blood sample was collected into a tube containing lithium heparin, centrifuged at 3000 rpm for 10 minutes and the plasma and buffy coat removed. The cells were then washed twice with an equal volume of saline (NaCI, 9 g/l) and finally mixed with a further equal volume of saline before being frozen at - 20°C. For assay the following day, or within two days of being taken, the cells were haemolysed by thawing and centrifuged at 3000 rpm for 10 minutes to remove cell debris, and O'5 mI of the haemolysate was mixed with 1·0 mI of saline. This dilution provided a final haemoglobin concentration of approximately 50 g/I for the measurement of transketolase activity. A further dilution of 0·5 mlof the above mixture with 2·0 mI of saline was prepared for the measurement of glutathione reductase and aspartate transaminase. The haemoglobin content of the haemolysate was measured by the cyanrnethaemoglobin method described by Henry et al. (1974). The materials and methods used for the estimation of the blood cell enzyme activities were essentially those described by Bayoumi and Rosalki (1976), except that all measurements were made at 37°C. A Beckman Model 25 Spectrophotometer with attached Recorder/Controller Unit (Beckman-RIIC Ltd, Gienrothes, Fife, Scotland) was used for the measurements. Pyruvate measurement Blood samples for pyruvate estimation were obtained from an antecubital vein of the supine patient, using minimum stasis and without fist-clenching (Braybooke et al., 1975). The first sample was
obtained after an overnight fast and further specimens at 30, 60, and 120 minutes after an oral dose of 50 g glucose. Blood (4 ml) was immediately mixed with 4 mI of ice-cold perchloric acid (1M) to prevent further metabolism of the pyruvate. After centrifugation, the pyruvate in the supernatant fluid was measured enzymatically at 340 nm by the method of Czok and Lamprecht (1970). Leucocyte vitamin C This was measured on blood (8 ml), collected into a tube containing lithium sequestrene as anticoagulant, by the method of Denson and Bowers (1961). The reference range for an adult population in this laboratory is 114-342 nmol/lO B WBC. y-Glutamyl transferase The activity in serum of y-glutamyl transferase (E.C. 2.3. 2.2) was measured at 37°C by the method of Rosalki and Tarlow (1974) except that the final concentrationofthe substrate (y-glutamyl-p-nitroanilide) was reduced to 3·2 mM as suggested by Tarbit (1975) and recommended by the Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology (1976). The measurements were made on a Gilford 3400 Enzyme Analyser (Gilford Instruments Ltd, Teddington, Middlesex). The reference ranges for an adult population in this laboratory are up to 45 U/I for female subjects and up to 48 U/I for male subjects. Results
Table 1 shows the red blood cell enzyme activity and percentage activation by the co-enzymes in a reference population of 21 laboratory workers (16 men and five women not taking oral contraceptives, age range 18-55 years). The results are in broad agreement with those obtained by Bayoumi and Rosalki (1976) but differ for the enzymes glutathione reductase and aspartate transaminase from those described by Williams (1976).
Table 1 Red blood cell enzyme activities and co-enzyme activation in a reference population (n = 21) Range Transketolase (U/gHb) Thiamine activation %
0·66- 1'16 0 - 21·2
Mean O·BS 10'7
SD 0'13 HO
Glutathione reductase (U/gHb) Flavin activation %
S'19- 13·6 16·0- 64·S
8'94 37'6
12-8
Aspartate transaminase CU/IlHb) Pyridoxine activation %
1·8S- 3-81 32·6 -101·8
2·78 64·8
0·49 19·0
Downloaded from acb.sagepub.com by guest on June 4, 2016
1.80
309
Detection and incidence of Band C vitamin deficiency in alcohol-related illness Table 2
Blood pyruvate levels in response to a glucose load in reference populations Landon et aI. (1961)
Thls sett.. (II = 15)
Whitby et al. (1975)
(II = 15)
Fasting pyruvate (I'mol/I) 30 min post glucose (I'mol/ll 60 min post glucose (I'mol/ll 120 min post glucose (I'mol/ll
Range
Mean
SD
Mean+2SD
Mean
56- 96 59-118 61-124
76'5 84'3 90·9 78·3
13'8 18'2 18·6 10'5
104 121 128 99
58 98 137 105
es-ioo
Table 2 shows the fasting pyruvate level and the response to a 50 g glucose load in a reference population of 15 subjects, age range 25-60 years. The values obtained from this population are compared with those described by Landon et al. (1962), who used a glucose dose of 1 g/kg body weight, and by Whitby et al. (1975), whose figures were based on a 100 g glucose dose. From the data presented in Table 2 it was decided that the criteria of abnormality of a pyruvate tolerance test would be a fasting pyruvate level of greater than 104 fLmolfl together with a 60minute level greater than 128 fLmol/l, or a single 60minute level of greater than 128 fLmol/l. The latter criterion would accommodate those patients who,
+
1 SD
Upper limit o/llormal 80 Not given 110 Not given
although having a normal fasting pyruvate level, were unable to respond in a normal manner to a glucose load. The levels of y-glutamyl transferase activity, the leucocyte vitamin C level, and the incidence of pyruvate tolerance test and red blood cell enzyme abnormality in the population under study are shown in Table 3, and Table 4 summarises the relationship between the two indices of thiamine deficiency. It can be seen from Table 3 that the pyruvate tolerance test was abnormal in 18 out of 33 cases (an incidence of 55 %), the transketolase activation was abnormal in 11 of 35 cases (31 %), the glutathione reductase activation was abnormal in
Table 3 y-Glutamyl transferase activity, pyruvate tolerance test and red blood cell enzyme abnormality, and leucocyte vitamin C levels in a population with alcohol-related illness Patient
Sex/Age
"I-GT (U/I)
I RB 2MM 3 MK 4WH 5 AE 6JM 7 HT 8 RC 9EM 10EP 11 FM 12 VS I3MM 14EE 15 FJ 16 LT 17WW 18CC 19 CA 20PW 21 MC 22 FS 23 JH 24TS 25 FD 26JC 27MH 28 JF 29CW 30MA 31 LR 32 LS 33 PM 34 JK 35 TH
M57 F47 F69 M55 M49 M39 M49 M56 M55 M56 M67 F38 M51 FSJ M51 F67 M58 M39 M48 F46 F52 M54 M45 M31 M67 M29 F57 MOO M65 F50 M41 M43 F37 M57 F54
270 714 41 113 801 50 495 115 32 154 327 1665 504 55 225 167 64 151 63 181 194 106 648 2565 57 27 237 244 448 266 79 73 3llO 61 61
Pyruvate abnormal
V V V V V V V V V
TKabllormal
V V V
GR abllormal
V V
V
V V V V
V V
•
V V V
V
V
V
V
V V
V V V V V V
• Not measured.
Downloaded from acb.sagepub.com by guest on June 4, 2016
AST abllormal
uucocyte vit C (IImol/lO' WBC 89 33 112 86 38 72 41 4 222 47 26 61 52 169 37 42 37 46 91 8 87 17 82 33 76 77 67 44 64 84 30 56 116 91 54
M. Baines
310 Table 4 Relationship between pyruvate tolerance test and red blood cell transketolase activation in population under study Both tests normal Pyruvate abnormal, TK normal Pyruvate normal, TK abnormal Both tests abnormal
13/33 (39%) 10/33 (30%) 2/33 (6~;';) 8/33 (24%)
8 of 35 cases (23 %), and the leucocyte vitamin C was abnormal in 32 of 35 cases (91 %). No case of abnormal aspartate transaminase activation was found in this series. From a total of 35 patients studied, 13 (37 %) were deficient in one B vitamin, and a further three (9 %) were deficient in two B vitamins, as measured by the red blood cell enzyme activation method. Discussion
The alcohol abuser is at risk of becoming thiaminedeficient for a number of reasons, including neglect of eating habits, impaired absorption of thiamine (Baker et al., 1975), and impaired utilisation of thiamine due to decreased phosphorylation by a damaged liver. The signs of deficiency may be subclinical, yet treatment is readily available and the response is usually good in the absence of severe cirrhosis. Estimates of thiamine in blood or urine, using chemical (Johnson et al., 1945) or microbiological methods (Baker and Frank, 1968), are tedious and unsuitable for the routine laboratory and may not give a true indication of tissue stores. The potential advantages of obtaining indices of three B vitamins from a single, small sample of erythrocytes from a non-fasted patient are obvious and look attractive when compared with the need in the pyruvate tolerance test to fast the patient overnight, administer glucose, and obtain four venous samples over two hours. Yet it can be seen from Table 3 that the incidence of pyruvate test abnormality (55 %) is considerably greater than that of transketolase activation abnormality (31 %), and therefore if the pyruvate response to a glucose load is a true indication of thiamine status then that test would appear to be a more sensitive indicator of thiamine deficiency than the transketolase activation. Both tests are, however, only indirect estimates of the thiamine level and may be influenced by conditions other than that of pure thiamine deficiency. The pyruvate level is known to be affected by diseases such as diabetes mellitus, congestive heart failure, and some digestive disorders, but, as far as was known, none of the patients under study was suffering from these disorders. Liver disease will, of course, affect both the pyruvate level and the transketolase activation by
reducing the phosphorylation of thiamine and the production of the apoenzyme of transketolase, but both tests should be affected in a similar manner, depending upon the degree of liver damage. Indeed, of the group of five patients with cirrhosis in this series, two patients had abnormality in both pyruvate tolerance test and transketolase activation, two patients had abnormality in neither, and one patient had abnormality of pyruvate tolerance test only. Further evidence that both the pyruvate tolerance test and transketolase activation may be regarded as indicative of the thiamine status is that both tests responded to thiamine replacement therapy. In two patients who were given large intramuscular doses of thiamine hydrochloride over 10 days (Parentrovite IMRP, Bencard, Brentford, Middlesex TW8 9BE) the 6O-minute pyruvate level fell from 168 p.mol/l to 104 p.mol/l and the transketolase activation from 25'6% to 0% in one (patient LT, no. 16) and from 183 p.mol/l to 119 p.mol/land 29·8 % to 0% in the second (patient MK, no. 3). It may be concluded, therefore, that both the pyruvate tolerance test and the activation of transketolase are sensitive to the thiamine status of the body, but that, even though pyruvate is dependent for its further metabolism upon the pentose phosphate cycle, and thus the enzyme transketolase, the pyruvate test is a more sensitive indicator of thiamine deficiency than the activation of red blood cell transketolase. The incidence of vitamin B2 (riboflavin) deficiency, as assessed by the flavin activation of red blood cell glutathione reductase, was 23 % (8 of 35 cases), a slightly lower incidence than that of thiamine deficiency as detected by transketolase activation. In a smaller study, Bayoumi and Rosalki (1976) found an incidence of 15 % (2 of 13 cases), and in a population of geriatric patients Room et al. (1975) found an incidence of B2 deficiency of 11·7 % and of Bl deficiency of 22·9 %. Thus it appears that riboflavin deficiency is less prevalent than thiamine deficiency, which may be explained by the fact that riboflavin, unlike thiamine, is readily absorbed from food sources by the alcoholic (Baker et al., 1975). No cases of vitamin Be (pyridoxine) deficiency were found in the present series, in agreement with the small series of Bayourni and Rosalki (1976). Indeed, from a total of 85 estimations done in this laboratory on a mixed hospital population (mainly geriatric and alcoholic) only one case of Bo deficiency has been discovered. Room et al. (1975) found an incidence of 19 % in geriatric patients. Vitamin Be deficiency has been noted in uraemic patients by measurement of the plasma pyridoxine level (Stone et al., 1975), in women using oral contraceptives by measurement of blood pyridoxal phosphate (Shane and Contractor,1975). in epileptics on anticonvulsant
Downloaded from acb.sagepub.com by guest on June 4, 2016
Detection and incidence of Band C vitamin deficiency in alcohol-related illness treatment by a method of red blood cell aspartate transaminase activation (Reinken et al., 1972), and in 25 of 50 alcoholics by a microbiological measurement of pyridoxine (Davis and Smith, 1974). A similar incidence (35 of 66 cases) of low plasma pyridoxal phosphate was noted in a series of alcoholic patients without abnormal liver findings by Lumeng and Li (1974). Thus a biochemical deficiency of pyridoxine has been clearly established in the alcoholic population and is supported by the work of Baker et al. (1975), who found a significantly lowered absorption of vitamin Be from a natural food source in a group of alcoholic patients with liver disease. The present findings, therefore, appear to agree with those of Shane and Contractor (1975), who, in a study of pregnant women and oral contraceptive users, concluded that the red blood cell AST activation test appeared to be a poor indicator of vitamin Be status except in pronounced deficiency, as it is less responsive to vitamin depletion than blood pyridoxal phosphate levels. A similar lack of sensitivity of the red blood cell AST activation test in detecting vitamin Be deficiency seems to apply to the present population with alcohol-related illness. One of the most consistent findings in the present series is that of deficiency of the other water-soluble vitamin, vitamin C, the incidence being 91 % (32 of 35 cases). This finding is in agreement with the results of O'Keane et al. (1972), who showed a significantly lower (p < 0'()()()5) mean leucocyte ascorbic acid level in 50 chronic alcoholics as compared with a matched control group. The above authors were also able to link the observed deficiency with diet, demonstrating a strong correlation between dietary intake of vitamin C and the leucocyte ascorbic acid level in the alcoholic and the control groups. As vitamin C is concerned with many metabolic processes, including collagen synthesis, corticosteroid and cholesterol metabolism, and electron transport processes, and requirement may be increased in liver disease to compensate for diminished storage and to allow tissue repair, it seems a prudent step to measure and, if necessary, correct the vitamin C status of patients admitted to care for problems associated with alcohol consumption. This may be easily achieved with intramuscular vitamin supplementation, a 10-day course of such a preparation (Parentrovite IMHP) restoring the leucocyte vitamin C level to 209 nmol/IO" WBC from an initial 42 nmol/10 s WBC in patient LT, no. 16. The range of values for y-glutamyl transferase (y-GT) activity of the patients when seen was 27 VII to 2565 VII (Table 3), with only three patients (9 %) below the upper limit of normal for their sex. Rosalki and Rau (1972), in a study of 76 anicteric patients who were classified as either 'alcoholics' or
311
'heavy drinkers', found an incidence of elevation of the y-GT activity of 75 %. No enzyme other than y-GT was elevated in the group classified as heavy drinkers, which leads Rosalki and Rau to consider the measurement of y-GT to be a sensitive and highly specific test for the detection of liver cell injury in the suspected alcoholic. This link between alcohol exposure and elevation of y-GT has been confirmed by many other authors, for example, Rollason et al. (1972) were able to show an increasing incidence of elevation of y-GT with increased alcohol consumption in five groups of individuals imbibing differing amounts of alcohol. The purpose of the measurement of y-GT in the present work was to provide an indication of the degree of exposure to alcohol, and to determine if any correlation could be made between the elevation of y-GT and the degree of vitamin deficiency, possibly allowing a predictive statement of need for vitamin supplementation for a given y-GT activity. The results presented in Table 3 do not appear to suggest any constant correlation between y-GT activity and the vitamin deficiency. Thus patient TS, no. 24, who shows the highest y-GT activity (2565 Ujl), does indeed have deficiency of vitamins Bi, Bs, and C, but patient VS, no. 12, with a y-GT activity of 1665 VII, has deficiency of vitamin C only. Further, patients MK, no. 3, and LT, no. 16, with y-GT levels of 41 VII and 167 VII respectively, have deficiency in vitamins Br, Bs, and C, but patients JH, no. 23, and MA, no. 30, with jrGT levels of 648 VII and 266 VII, have deficiency of vitamin C only. The conclusion must be made, therefore, that the observed vitamin deficiency is not simply a factor of degree of exposure to alcohol but is probably dependent to a large extent on the dietary habits of the individual patient. It seems that a continued food intake, even in the face of sustained alcohol abuse, is able to maintain vitamin stores in some patients. It can be concluded, therefore, that the patient admitted to hospital for problems associated with alcohol abuse is at risk of having developed a vitamin deficiency, that risk being a function of factors such as a degree of exposure to alcohol and dietary adequacy. The vitamin most commonly deficient is vitamin C, followed by vitamins Bi and Bs, but the actual incidence of deficiency is dependent upon the method of detection. The pyruvate response to glucose load, although more time-consuming than the red blood cell transketolase activation test, does appear to be more sensitive in detecting vitamin Bl deficiency. in the population under study, and its attractiveness, both to the investigator and to the patient, might be increased by reducing the number of samples to two, a fasting specimen and a 60minute sample following a glucose dose. The simple
Downloaded from acb.sagepub.com by guest on June 4, 2016
M. Baines
312
sample requirement of the red blood cell enzyme activation tests make them a useful adjunct to the pyruvate test and the leucocyte vitamin C estimation, but the limitations of the transketolase and aspartate transaminase estimations must be borne in mind. Indeed, with such a low incidence of vitamin B8 deficiency seen in this laboratory with the present method, it is doubtful whether the continued use of this method is justified. Finally, while the level of activity of y-GT gives a useful guide to alcohol exposure in the patient, it is not able to predict vitamin deficiency. Only an accurate dietary history, together with the biochemical tests described above, will identify those patients who would benefit from vitamin supplementation. I thank Mr G. Davies, principal biochemist, for help and advice with this work, the staff of the biochemistry laboratory for performing some of the estimations, the physicians of Broadgreen Hospital for access to patients in their care, and H. J. Crispe for typing the manuscript. References Baker, H., and Frank, O. (1968). Vitamin status in metabolic upsets. World Review ofNutrition and Dietetics, 9,124-160. Baker, H., Frank, 0., Zetterman, R. K., Rajan, K. S., ten Hove, W., and Leevy, C. M. (1975). Inability of chronic alcoholics with liver disease to use food as a source of folates, thiamin and vitamin B6. American Journal of Clinical Nutrition, 28, 1377-1380. Baron, D. N. (1973). A Short Textbook ofChemical Pathology, 3rd edition, p, 55. The English Universities Press, London. Bayoumi, R. A., and Rosalki, S. B. (1976). Evaluation of methods of coenzyme activation of erythrocyte enzymes for detection of deficiency of vitamins B-1, B-2, and B-6. Clinical Chemistry, 22, 327-335. Boone, D. J., Tietz, N. W., and Weinstock, A. (1977). Significance of y-glutamyl transferase (GGn activity measurements in alcohol-induced hepatic injury. Annals of Clinical and Laboratory Science, 7, 25-28. Braybrooke, J., Lloyd, B., Nattrass, M., and Alberti, K. G. M. M.(l975). Blood sampling techniques for lactate and pyruvate: a reappraisal. Annals of Clinical Biochemistry, 12, 252-254. Cambier, J., Masson, M., Dairou, R., Delacoux, E., and Bournique, J. (1973). Evaluation biochimique de la carence en vitamine BI. Etude comparee de la pyruvicemie et de la transcetolase. Annales de Medeclne Interne, 124, 189-192. Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology (1976). Recommendedmethod for the determination of y-glutamyl transferase in blood. Scandinavian Journal of Clinical and Laboratory Investigation, 36, 119-125. Czok, R., and Lamprecht, W. (1970). In Methoden der enzymatischen Analyse, volume 2, edited by H. V. Bergmeyer, p, 1407. Verlag Chernie, Weinheim.
Davis, R. E., and Smith, B. K. (1974). Pyridoxal and folate deficiency in alcoholics. Medical Journal of Australia, 2, 357-360. Denson, K. W., and Bowers, E. F. (1961). The determination of ascorbic acid in white blood cells. A comparison of white blood cell ascorbic acid and phenolic acid excretion in elderly patients. Clinical Science, 21, 157-162. Henry, R. J., Cannon, D. c., and Winkelman, J. W. (Eds.). (1974). Clinical Chemistry, Principles and Technics, 2nd edition, p. 1131. Harper and Row, Maryland. Hoorn, R. K. J., Flikweert, J. P., and Westerink, D. (1975). Vitamin B-1, B-2 and B-6 deficiencies in geriatric patients, measured by coenzyme stimulation of enzyme activities. Clinica Chimica Acta, 61, 151-162. Johnson, R. E., Sargent, F., Robinson, P. F., and Consalazio, F. C. (1945). Estimation of riboflavin thiamine, and N'methylnicotinamide. Industrial and Engineering Chemistry, Analytical Edition. 17, 384-386. Landon, J., Fawcett, J. K., and Wynn, V. (1962). Blood pyruvate concentration measured by a specific method in control subjects. Journal of Clinical Pathology, 15, 579-584. Lumeng, L., and Li, T-K. (1974). Vitamin B-6metabolism in chronic alcohol abuse. Pyridoxal phosphate levels in plasma and the effects of acetaldehyde on pyridoxal phosphate synthesis and degradation in human erythrocytes. Journal of Clinical Investigation, 53, 693-704. Morgan, H. G. (1968). Acute neuropsychiatriccornplications in chronic alcoholism. British Journal of Psychiatry, 114, 85-92. O'Keane, Maureen, Russell, R. I., and Goldberg, A. (1972). Ascorbic acid status in alcoholics. Journal ofAlcoholism, 7, 6-11. Reinken, L., Hohenauer, L., and Ziegler, E. E. (1972). Activity of red blood cell glutamic oxalacetlc transaminase in epileptic children under anti-epileptic treatment. Clinica Chimica Acta, 36, 270-271. Rollason, J. G., Pincherle, G .• and Robinson, D. (1972). Serum gamma glutamyl transpeptidase in relation to alcohol consumption. Clinica Chimica Acta, 39, 75-80. Rosalki, S. B., and Rau, D. (1972). Serum y-glutamyl transpeptidase activity in alcoholism. Clinica Chimica Acta, 39, 41-47. Rosalki, S. B., and Tarlow, D. (1974). Optimised determination of y-glutamyl transferase by reaction-rate analysis. Clinical Chemistry, 20, 1121-1124. Shane, B., and Contractor, S. F. (1975). Assessment of vitamin B6 status. Studies on pregnant women and oral contraceptive users. American Journal of Clinical Nutrition, 28,739-747. Stone, W. J., Warnock, L. G., and Wagner, C. (1975). Vitamin B6 deficiency in uraemia. American Journal 0/ Clinical Nutrition, 28, 950-957. Tarbit, I. F. (1975). More economical y-glutamyl-transferase assay. Clinical Chemistry, 21, 10, 1548. Whitby, L. G., Percy-Robb, I. W., and Smith, A. F. (Eds.), (1975). Lecture Notes on Clinical Chemistry, p, 226. Blackwell Scientific Publications, Oxford. Williams, D. G. (1976). Methods for the estimation of three vitamin dependent red cell enzymes. Clinical Biochemistry, 9,252-255.
Accepted for publication 18 August 1978
Downloaded from acb.sagepub.com by guest on June 4, 2016
Detection and incidence of Band C vitamin deficiency in alcohol-related illness M. BAINES From the Department 0/ Pathology, Broadgreen Hospital, Liverpool L14 3LB
The activity of the red blood cell enzymes transketolase, glutathione reductase, and aspartate transaminase, and their activation by the coenzymes thiamine, riboflavin, and pyridoxine, the pyruvate tolerance test, the leucocyte vitamin C concentration, and the activity in serum of yglutamyl transferase were measured in a series of 35 patients with alcohol-related illness. The incidence of thiamine deficiency was 31 % as assessed by the activation of transketolase, and 55 % as assessed by the pyruvate tolerance test. The incidence of riboflavin deficiency was 23 % and of ascorbic acid deficiency 91 %. No cases of pyridoxine deficiency were detected. The pyruvate tolerance test was found to be a more sensitive test of thiamine deficiency than the transketolase activation, and the activation of red blood cell aspartate transaminase was found to be a poor indicator of pyridoxine deficiency. There was a poor correlation of the y-glutamyl transferase activity with the degree of vitamin deficiency, suggesting that alcohol exposure is only partly responsible for the observed vitamin deficiency.
SUMMARY
The possibility of detecting B vitamin deficiency in a aim of the present study was to determine the relative hospital population has recently become a more sensitivity of the full pyruvate tolerance test and the feasible proposition with the introduction of methods red blood cell transketolase activation in detecting involving measurement of red blood cell enzyme vitamin Br deficiency in patients admitted to hospital activity with and without their co-enzyme activators. for alcohol abuse and to test recent suggestions Thus indices of vitamin Bi, B2,and B6 status can be (Baron, 1973; Whitby et al., 1975) that red blood cell obtained by measuring red blood cell transketolase transketolase activity appears to be the method of (TK, E.C. 2.2. 1.1) with and without thiamine choice for investigating possible thiamine deficiency. pyrophosphate, red blood cell glutathione reductase The two indices of vitamin Bi status described above, (GR, E.C. 1.6. 4.2) with and without flavin adenine together with indices of vitamins B2 and B6 and the dinucleotide, and red blood cell aspartate trans- measurement of leucocyte vitamin C, formed the aminase (AST, E.C. 2.6. 1.1) with and without assessment of the vitamin status of the patients under pyridoxal phosphate (Bayoumi and Rosalki, 1976). study in order to determine which of the vitamins A more traditional approach to the identification of are most commonly deficient. Finally, an attempt thiamine deficiency, a possible cause of neurological was made to correlate the degree of deficiency with disorders such as peripheral neuropathy in the the exposure to alcohol, as indicated by the serum alcoholic, has been the response of blood pyruvate activity of y-glutamyl transferase, an enzyme known to a glucose load, the so-called 'pyruvate tolerance to be induced by alcohol intake (Rosalki and Rau, test'. In the absence of its co-enzyme, thiamine pyro- 1972) and to be of significance as a test of hepatic phosphate, transketolase is inactivated, and the meta- injury in the alcohol abuser (Boone et al., 1977). bolism of pyruvate, formed from glucose, is impaired. The pyruvate tolerance test has been used by Materials and methods Morgan (1968) to detect thiamine deficiency in two groups of chronic alcoholics, those experiencing SUBJECTS hallucinations and those with no such symptoms. A group of 35 patients admitted to hospital for the Cambier et al. (1973) have compared a single estima- treatment of problems associated with alcohol abuse tion of pyruvic acid with a colorimetrically deter- was studied. As far as was possible all such hospital mined transketolase level as an index of thiamine admissions over a 15-monthperiodwereincluded. The status in patients with neurological disorders. The reasons for hospitalisation included peripheral 307
Downloaded from acb.sagepub.com by guest on June 4, 2016
M. Baines
308 neuritis and paresthesiae and ethanolic liver damage. Five patients were subsequently shown histologically to have developed cirrhosis. As it was not possible to determine the degree of alcohol dependence the term 'alcoholic', implying alcohol dependence, has been avoided. The group probably included those described as 'heavy drinkers' and 'alcoholics' by Rosalki and Rau (1972). A group of 21 healthy laboratory staff, age range 18-55, formed a reference population for the red blood cell enzyme studies. The control group for the pyruvate studies was composed of 15 subjects of age range 25-60 years. The population included laboratory staff and hospital inpatients who were due for discharge in an adequately fit and nourished condition. No subject in the control group had, as far as was known, any illness likely to affect the metabolism of glucose or any known vitamin deficiency. EXPERIMENTAL METHODS
Red blood cell enzymes These were measured in an haemolysate prepared as follows. A 5-10 mI blood sample was collected into a tube containing lithium heparin, centrifuged at 3000 rpm for 10 minutes and the plasma and buffy coat removed. The cells were then washed twice with an equal volume of saline (NaCI, 9 g/l) and finally mixed with a further equal volume of saline before being frozen at - 20°C. For assay the following day, or within two days of being taken, the cells were haemolysed by thawing and centrifuged at 3000 rpm for 10 minutes to remove cell debris, and O'5 mI of the haemolysate was mixed with 1·0 mI of saline. This dilution provided a final haemoglobin concentration of approximately 50 g/I for the measurement of transketolase activity. A further dilution of 0·5 mlof the above mixture with 2·0 mI of saline was prepared for the measurement of glutathione reductase and aspartate transaminase. The haemoglobin content of the haemolysate was measured by the cyanrnethaemoglobin method described by Henry et al. (1974). The materials and methods used for the estimation of the blood cell enzyme activities were essentially those described by Bayoumi and Rosalki (1976), except that all measurements were made at 37°C. A Beckman Model 25 Spectrophotometer with attached Recorder/Controller Unit (Beckman-RIIC Ltd, Gienrothes, Fife, Scotland) was used for the measurements. Pyruvate measurement Blood samples for pyruvate estimation were obtained from an antecubital vein of the supine patient, using minimum stasis and without fist-clenching (Braybooke et al., 1975). The first sample was
obtained after an overnight fast and further specimens at 30, 60, and 120 minutes after an oral dose of 50 g glucose. Blood (4 ml) was immediately mixed with 4 mI of ice-cold perchloric acid (1M) to prevent further metabolism of the pyruvate. After centrifugation, the pyruvate in the supernatant fluid was measured enzymatically at 340 nm by the method of Czok and Lamprecht (1970). Leucocyte vitamin C This was measured on blood (8 ml), collected into a tube containing lithium sequestrene as anticoagulant, by the method of Denson and Bowers (1961). The reference range for an adult population in this laboratory is 114-342 nmol/lO B WBC. y-Glutamyl transferase The activity in serum of y-glutamyl transferase (E.C. 2.3. 2.2) was measured at 37°C by the method of Rosalki and Tarlow (1974) except that the final concentrationofthe substrate (y-glutamyl-p-nitroanilide) was reduced to 3·2 mM as suggested by Tarbit (1975) and recommended by the Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology (1976). The measurements were made on a Gilford 3400 Enzyme Analyser (Gilford Instruments Ltd, Teddington, Middlesex). The reference ranges for an adult population in this laboratory are up to 45 U/I for female subjects and up to 48 U/I for male subjects. Results
Table 1 shows the red blood cell enzyme activity and percentage activation by the co-enzymes in a reference population of 21 laboratory workers (16 men and five women not taking oral contraceptives, age range 18-55 years). The results are in broad agreement with those obtained by Bayoumi and Rosalki (1976) but differ for the enzymes glutathione reductase and aspartate transaminase from those described by Williams (1976).
Table 1 Red blood cell enzyme activities and co-enzyme activation in a reference population (n = 21) Range Transketolase (U/gHb) Thiamine activation %
0·66- 1'16 0 - 21·2
Mean O·BS 10'7
SD 0'13 HO
Glutathione reductase (U/gHb) Flavin activation %
S'19- 13·6 16·0- 64·S
8'94 37'6
12-8
Aspartate transaminase CU/IlHb) Pyridoxine activation %
1·8S- 3-81 32·6 -101·8
2·78 64·8
0·49 19·0
Downloaded from acb.sagepub.com by guest on June 4, 2016
1.80
309
Detection and incidence of Band C vitamin deficiency in alcohol-related illness Table 2
Blood pyruvate levels in response to a glucose load in reference populations Landon et aI. (1961)
Thls sett.. (II = 15)
Whitby et al. (1975)
(II = 15)
Fasting pyruvate (I'mol/I) 30 min post glucose (I'mol/ll 60 min post glucose (I'mol/ll 120 min post glucose (I'mol/ll
Range
Mean
SD
Mean+2SD
Mean
56- 96 59-118 61-124
76'5 84'3 90·9 78·3
13'8 18'2 18·6 10'5
104 121 128 99
58 98 137 105
es-ioo
Table 2 shows the fasting pyruvate level and the response to a 50 g glucose load in a reference population of 15 subjects, age range 25-60 years. The values obtained from this population are compared with those described by Landon et al. (1962), who used a glucose dose of 1 g/kg body weight, and by Whitby et al. (1975), whose figures were based on a 100 g glucose dose. From the data presented in Table 2 it was decided that the criteria of abnormality of a pyruvate tolerance test would be a fasting pyruvate level of greater than 104 fLmolfl together with a 60minute level greater than 128 fLmol/l, or a single 60minute level of greater than 128 fLmol/l. The latter criterion would accommodate those patients who,
+
1 SD
Upper limit o/llormal 80 Not given 110 Not given
although having a normal fasting pyruvate level, were unable to respond in a normal manner to a glucose load. The levels of y-glutamyl transferase activity, the leucocyte vitamin C level, and the incidence of pyruvate tolerance test and red blood cell enzyme abnormality in the population under study are shown in Table 3, and Table 4 summarises the relationship between the two indices of thiamine deficiency. It can be seen from Table 3 that the pyruvate tolerance test was abnormal in 18 out of 33 cases (an incidence of 55 %), the transketolase activation was abnormal in 11 of 35 cases (31 %), the glutathione reductase activation was abnormal in
Table 3 y-Glutamyl transferase activity, pyruvate tolerance test and red blood cell enzyme abnormality, and leucocyte vitamin C levels in a population with alcohol-related illness Patient
Sex/Age
"I-GT (U/I)
I RB 2MM 3 MK 4WH 5 AE 6JM 7 HT 8 RC 9EM 10EP 11 FM 12 VS I3MM 14EE 15 FJ 16 LT 17WW 18CC 19 CA 20PW 21 MC 22 FS 23 JH 24TS 25 FD 26JC 27MH 28 JF 29CW 30MA 31 LR 32 LS 33 PM 34 JK 35 TH
M57 F47 F69 M55 M49 M39 M49 M56 M55 M56 M67 F38 M51 FSJ M51 F67 M58 M39 M48 F46 F52 M54 M45 M31 M67 M29 F57 MOO M65 F50 M41 M43 F37 M57 F54
270 714 41 113 801 50 495 115 32 154 327 1665 504 55 225 167 64 151 63 181 194 106 648 2565 57 27 237 244 448 266 79 73 3llO 61 61
Pyruvate abnormal
V V V V V V V V V
TKabllormal
V V V
GR abllormal
V V
V
V V V V
V V
•
V V V
V
V
V
V
V V
V V V V V V
• Not measured.
Downloaded from acb.sagepub.com by guest on June 4, 2016
AST abllormal
uucocyte vit C (IImol/lO' WBC 89 33 112 86 38 72 41 4 222 47 26 61 52 169 37 42 37 46 91 8 87 17 82 33 76 77 67 44 64 84 30 56 116 91 54
M. Baines
310 Table 4 Relationship between pyruvate tolerance test and red blood cell transketolase activation in population under study Both tests normal Pyruvate abnormal, TK normal Pyruvate normal, TK abnormal Both tests abnormal
13/33 (39%) 10/33 (30%) 2/33 (6~;';) 8/33 (24%)
8 of 35 cases (23 %), and the leucocyte vitamin C was abnormal in 32 of 35 cases (91 %). No case of abnormal aspartate transaminase activation was found in this series. From a total of 35 patients studied, 13 (37 %) were deficient in one B vitamin, and a further three (9 %) were deficient in two B vitamins, as measured by the red blood cell enzyme activation method. Discussion
The alcohol abuser is at risk of becoming thiaminedeficient for a number of reasons, including neglect of eating habits, impaired absorption of thiamine (Baker et al., 1975), and impaired utilisation of thiamine due to decreased phosphorylation by a damaged liver. The signs of deficiency may be subclinical, yet treatment is readily available and the response is usually good in the absence of severe cirrhosis. Estimates of thiamine in blood or urine, using chemical (Johnson et al., 1945) or microbiological methods (Baker and Frank, 1968), are tedious and unsuitable for the routine laboratory and may not give a true indication of tissue stores. The potential advantages of obtaining indices of three B vitamins from a single, small sample of erythrocytes from a non-fasted patient are obvious and look attractive when compared with the need in the pyruvate tolerance test to fast the patient overnight, administer glucose, and obtain four venous samples over two hours. Yet it can be seen from Table 3 that the incidence of pyruvate test abnormality (55 %) is considerably greater than that of transketolase activation abnormality (31 %), and therefore if the pyruvate response to a glucose load is a true indication of thiamine status then that test would appear to be a more sensitive indicator of thiamine deficiency than the transketolase activation. Both tests are, however, only indirect estimates of the thiamine level and may be influenced by conditions other than that of pure thiamine deficiency. The pyruvate level is known to be affected by diseases such as diabetes mellitus, congestive heart failure, and some digestive disorders, but, as far as was known, none of the patients under study was suffering from these disorders. Liver disease will, of course, affect both the pyruvate level and the transketolase activation by
reducing the phosphorylation of thiamine and the production of the apoenzyme of transketolase, but both tests should be affected in a similar manner, depending upon the degree of liver damage. Indeed, of the group of five patients with cirrhosis in this series, two patients had abnormality in both pyruvate tolerance test and transketolase activation, two patients had abnormality in neither, and one patient had abnormality of pyruvate tolerance test only. Further evidence that both the pyruvate tolerance test and transketolase activation may be regarded as indicative of the thiamine status is that both tests responded to thiamine replacement therapy. In two patients who were given large intramuscular doses of thiamine hydrochloride over 10 days (Parentrovite IMRP, Bencard, Brentford, Middlesex TW8 9BE) the 6O-minute pyruvate level fell from 168 p.mol/l to 104 p.mol/l and the transketolase activation from 25'6% to 0% in one (patient LT, no. 16) and from 183 p.mol/l to 119 p.mol/land 29·8 % to 0% in the second (patient MK, no. 3). It may be concluded, therefore, that both the pyruvate tolerance test and the activation of transketolase are sensitive to the thiamine status of the body, but that, even though pyruvate is dependent for its further metabolism upon the pentose phosphate cycle, and thus the enzyme transketolase, the pyruvate test is a more sensitive indicator of thiamine deficiency than the activation of red blood cell transketolase. The incidence of vitamin B2 (riboflavin) deficiency, as assessed by the flavin activation of red blood cell glutathione reductase, was 23 % (8 of 35 cases), a slightly lower incidence than that of thiamine deficiency as detected by transketolase activation. In a smaller study, Bayoumi and Rosalki (1976) found an incidence of 15 % (2 of 13 cases), and in a population of geriatric patients Room et al. (1975) found an incidence of B2 deficiency of 11·7 % and of Bl deficiency of 22·9 %. Thus it appears that riboflavin deficiency is less prevalent than thiamine deficiency, which may be explained by the fact that riboflavin, unlike thiamine, is readily absorbed from food sources by the alcoholic (Baker et al., 1975). No cases of vitamin Be (pyridoxine) deficiency were found in the present series, in agreement with the small series of Bayourni and Rosalki (1976). Indeed, from a total of 85 estimations done in this laboratory on a mixed hospital population (mainly geriatric and alcoholic) only one case of Bo deficiency has been discovered. Room et al. (1975) found an incidence of 19 % in geriatric patients. Vitamin Be deficiency has been noted in uraemic patients by measurement of the plasma pyridoxine level (Stone et al., 1975), in women using oral contraceptives by measurement of blood pyridoxal phosphate (Shane and Contractor,1975). in epileptics on anticonvulsant
Downloaded from acb.sagepub.com by guest on June 4, 2016
Detection and incidence of Band C vitamin deficiency in alcohol-related illness treatment by a method of red blood cell aspartate transaminase activation (Reinken et al., 1972), and in 25 of 50 alcoholics by a microbiological measurement of pyridoxine (Davis and Smith, 1974). A similar incidence (35 of 66 cases) of low plasma pyridoxal phosphate was noted in a series of alcoholic patients without abnormal liver findings by Lumeng and Li (1974). Thus a biochemical deficiency of pyridoxine has been clearly established in the alcoholic population and is supported by the work of Baker et al. (1975), who found a significantly lowered absorption of vitamin Be from a natural food source in a group of alcoholic patients with liver disease. The present findings, therefore, appear to agree with those of Shane and Contractor (1975), who, in a study of pregnant women and oral contraceptive users, concluded that the red blood cell AST activation test appeared to be a poor indicator of vitamin Be status except in pronounced deficiency, as it is less responsive to vitamin depletion than blood pyridoxal phosphate levels. A similar lack of sensitivity of the red blood cell AST activation test in detecting vitamin Be deficiency seems to apply to the present population with alcohol-related illness. One of the most consistent findings in the present series is that of deficiency of the other water-soluble vitamin, vitamin C, the incidence being 91 % (32 of 35 cases). This finding is in agreement with the results of O'Keane et al. (1972), who showed a significantly lower (p < 0'()()()5) mean leucocyte ascorbic acid level in 50 chronic alcoholics as compared with a matched control group. The above authors were also able to link the observed deficiency with diet, demonstrating a strong correlation between dietary intake of vitamin C and the leucocyte ascorbic acid level in the alcoholic and the control groups. As vitamin C is concerned with many metabolic processes, including collagen synthesis, corticosteroid and cholesterol metabolism, and electron transport processes, and requirement may be increased in liver disease to compensate for diminished storage and to allow tissue repair, it seems a prudent step to measure and, if necessary, correct the vitamin C status of patients admitted to care for problems associated with alcohol consumption. This may be easily achieved with intramuscular vitamin supplementation, a 10-day course of such a preparation (Parentrovite IMHP) restoring the leucocyte vitamin C level to 209 nmol/IO" WBC from an initial 42 nmol/10 s WBC in patient LT, no. 16. The range of values for y-glutamyl transferase (y-GT) activity of the patients when seen was 27 VII to 2565 VII (Table 3), with only three patients (9 %) below the upper limit of normal for their sex. Rosalki and Rau (1972), in a study of 76 anicteric patients who were classified as either 'alcoholics' or
311
'heavy drinkers', found an incidence of elevation of the y-GT activity of 75 %. No enzyme other than y-GT was elevated in the group classified as heavy drinkers, which leads Rosalki and Rau to consider the measurement of y-GT to be a sensitive and highly specific test for the detection of liver cell injury in the suspected alcoholic. This link between alcohol exposure and elevation of y-GT has been confirmed by many other authors, for example, Rollason et al. (1972) were able to show an increasing incidence of elevation of y-GT with increased alcohol consumption in five groups of individuals imbibing differing amounts of alcohol. The purpose of the measurement of y-GT in the present work was to provide an indication of the degree of exposure to alcohol, and to determine if any correlation could be made between the elevation of y-GT and the degree of vitamin deficiency, possibly allowing a predictive statement of need for vitamin supplementation for a given y-GT activity. The results presented in Table 3 do not appear to suggest any constant correlation between y-GT activity and the vitamin deficiency. Thus patient TS, no. 24, who shows the highest y-GT activity (2565 Ujl), does indeed have deficiency of vitamins Bi, Bs, and C, but patient VS, no. 12, with a y-GT activity of 1665 VII, has deficiency of vitamin C only. Further, patients MK, no. 3, and LT, no. 16, with y-GT levels of 41 VII and 167 VII respectively, have deficiency in vitamins Br, Bs, and C, but patients JH, no. 23, and MA, no. 30, with jrGT levels of 648 VII and 266 VII, have deficiency of vitamin C only. The conclusion must be made, therefore, that the observed vitamin deficiency is not simply a factor of degree of exposure to alcohol but is probably dependent to a large extent on the dietary habits of the individual patient. It seems that a continued food intake, even in the face of sustained alcohol abuse, is able to maintain vitamin stores in some patients. It can be concluded, therefore, that the patient admitted to hospital for problems associated with alcohol abuse is at risk of having developed a vitamin deficiency, that risk being a function of factors such as a degree of exposure to alcohol and dietary adequacy. The vitamin most commonly deficient is vitamin C, followed by vitamins Bi and Bs, but the actual incidence of deficiency is dependent upon the method of detection. The pyruvate response to glucose load, although more time-consuming than the red blood cell transketolase activation test, does appear to be more sensitive in detecting vitamin Bl deficiency. in the population under study, and its attractiveness, both to the investigator and to the patient, might be increased by reducing the number of samples to two, a fasting specimen and a 60minute sample following a glucose dose. The simple
Downloaded from acb.sagepub.com by guest on June 4, 2016
M. Baines
312
sample requirement of the red blood cell enzyme activation tests make them a useful adjunct to the pyruvate test and the leucocyte vitamin C estimation, but the limitations of the transketolase and aspartate transaminase estimations must be borne in mind. Indeed, with such a low incidence of vitamin B8 deficiency seen in this laboratory with the present method, it is doubtful whether the continued use of this method is justified. Finally, while the level of activity of y-GT gives a useful guide to alcohol exposure in the patient, it is not able to predict vitamin deficiency. Only an accurate dietary history, together with the biochemical tests described above, will identify those patients who would benefit from vitamin supplementation. I thank Mr G. Davies, principal biochemist, for help and advice with this work, the staff of the biochemistry laboratory for performing some of the estimations, the physicians of Broadgreen Hospital for access to patients in their care, and H. J. Crispe for typing the manuscript. References Baker, H., and Frank, O. (1968). Vitamin status in metabolic upsets. World Review ofNutrition and Dietetics, 9,124-160. Baker, H., Frank, 0., Zetterman, R. K., Rajan, K. S., ten Hove, W., and Leevy, C. M. (1975). Inability of chronic alcoholics with liver disease to use food as a source of folates, thiamin and vitamin B6. American Journal of Clinical Nutrition, 28, 1377-1380. Baron, D. N. (1973). A Short Textbook ofChemical Pathology, 3rd edition, p, 55. The English Universities Press, London. Bayoumi, R. A., and Rosalki, S. B. (1976). Evaluation of methods of coenzyme activation of erythrocyte enzymes for detection of deficiency of vitamins B-1, B-2, and B-6. Clinical Chemistry, 22, 327-335. Boone, D. J., Tietz, N. W., and Weinstock, A. (1977). Significance of y-glutamyl transferase (GGn activity measurements in alcohol-induced hepatic injury. Annals of Clinical and Laboratory Science, 7, 25-28. Braybrooke, J., Lloyd, B., Nattrass, M., and Alberti, K. G. M. M.(l975). Blood sampling techniques for lactate and pyruvate: a reappraisal. Annals of Clinical Biochemistry, 12, 252-254. Cambier, J., Masson, M., Dairou, R., Delacoux, E., and Bournique, J. (1973). Evaluation biochimique de la carence en vitamine BI. Etude comparee de la pyruvicemie et de la transcetolase. Annales de Medeclne Interne, 124, 189-192. Committee on Enzymes of the Scandinavian Society for Clinical Chemistry and Clinical Physiology (1976). Recommendedmethod for the determination of y-glutamyl transferase in blood. Scandinavian Journal of Clinical and Laboratory Investigation, 36, 119-125. Czok, R., and Lamprecht, W. (1970). In Methoden der enzymatischen Analyse, volume 2, edited by H. V. Bergmeyer, p, 1407. Verlag Chernie, Weinheim.
Davis, R. E., and Smith, B. K. (1974). Pyridoxal and folate deficiency in alcoholics. Medical Journal of Australia, 2, 357-360. Denson, K. W., and Bowers, E. F. (1961). The determination of ascorbic acid in white blood cells. A comparison of white blood cell ascorbic acid and phenolic acid excretion in elderly patients. Clinical Science, 21, 157-162. Henry, R. J., Cannon, D. c., and Winkelman, J. W. (Eds.). (1974). Clinical Chemistry, Principles and Technics, 2nd edition, p. 1131. Harper and Row, Maryland. Hoorn, R. K. J., Flikweert, J. P., and Westerink, D. (1975). Vitamin B-1, B-2 and B-6 deficiencies in geriatric patients, measured by coenzyme stimulation of enzyme activities. Clinica Chimica Acta, 61, 151-162. Johnson, R. E., Sargent, F., Robinson, P. F., and Consalazio, F. C. (1945). Estimation of riboflavin thiamine, and N'methylnicotinamide. Industrial and Engineering Chemistry, Analytical Edition. 17, 384-386. Landon, J., Fawcett, J. K., and Wynn, V. (1962). Blood pyruvate concentration measured by a specific method in control subjects. Journal of Clinical Pathology, 15, 579-584. Lumeng, L., and Li, T-K. (1974). Vitamin B-6metabolism in chronic alcohol abuse. Pyridoxal phosphate levels in plasma and the effects of acetaldehyde on pyridoxal phosphate synthesis and degradation in human erythrocytes. Journal of Clinical Investigation, 53, 693-704. Morgan, H. G. (1968). Acute neuropsychiatriccornplications in chronic alcoholism. British Journal of Psychiatry, 114, 85-92. O'Keane, Maureen, Russell, R. I., and Goldberg, A. (1972). Ascorbic acid status in alcoholics. Journal ofAlcoholism, 7, 6-11. Reinken, L., Hohenauer, L., and Ziegler, E. E. (1972). Activity of red blood cell glutamic oxalacetlc transaminase in epileptic children under anti-epileptic treatment. Clinica Chimica Acta, 36, 270-271. Rollason, J. G., Pincherle, G .• and Robinson, D. (1972). Serum gamma glutamyl transpeptidase in relation to alcohol consumption. Clinica Chimica Acta, 39, 75-80. Rosalki, S. B., and Rau, D. (1972). Serum y-glutamyl transpeptidase activity in alcoholism. Clinica Chimica Acta, 39, 41-47. Rosalki, S. B., and Tarlow, D. (1974). Optimised determination of y-glutamyl transferase by reaction-rate analysis. Clinical Chemistry, 20, 1121-1124. Shane, B., and Contractor, S. F. (1975). Assessment of vitamin B6 status. Studies on pregnant women and oral contraceptive users. American Journal of Clinical Nutrition, 28,739-747. Stone, W. J., Warnock, L. G., and Wagner, C. (1975). Vitamin B6 deficiency in uraemia. American Journal 0/ Clinical Nutrition, 28, 950-957. Tarbit, I. F. (1975). More economical y-glutamyl-transferase assay. Clinical Chemistry, 21, 10, 1548. Whitby, L. G., Percy-Robb, I. W., and Smith, A. F. (Eds.), (1975). Lecture Notes on Clinical Chemistry, p, 226. Blackwell Scientific Publications, Oxford. Williams, D. G. (1976). Methods for the estimation of three vitamin dependent red cell enzymes. Clinical Biochemistry, 9,252-255.
Accepted for publication 18 August 1978
Downloaded from acb.sagepub.com by guest on June 4, 2016
