Zinc status and vitamin A transport in cystic fibrosis1

Transcription

1 Zinc status and vitamin A transport in cystic fibrosis1 Robert A. Jacob, Harold H. Sandstead, Noel W. Solomons,2 Christian Rieger, and Richard Rothberg ABSTRACT Zinc status and the retinol transport system were examined in 18 retinol supplemented cystic fibrosis (CF) patients and 40 age-matched controls. Plasma vitamin A was significantly lower in the CF group as compared to the controls and correlated positively with plasma retinol-binding protein (RBP) in both the CF and control groups. Plasma zinc of the CF group was not significantly lower than controls whereas hair zinc was. Plasma zinc was positively correlated with plasma RBP, vitamin A, and albumin in the CF group but not in the controls. Plasma concentrations of vitamin A, RBP, albumin, and zinc decreased with age in the CF group but not in the controls. The data support previous suggestions that low plasma vitamin A levels in CF are due to defects in the retinol transport system. The zinc status of the CF group as a whole was judged to be low-normal however a subgroup of CF patients were in the marginal to deficient category. This subgroup also had lower levels of plasma vitamin A and RBP. The data suggest that zinc may be a contributing factor in the low plasma vitamin A/RBP levels of CF patients with marginal or deficient zinc status. Am. J. Clin. Nutr. 31: , An increased incidence of vitamin A deficiency in cystic fibrosis (CF) patients has long been known (1). Standard therapy for the disease includes generous oral supplementation with vitamin A but in spite of this plasma levels of vitamin A in CF patients are often subnormal. Vitamin A is mobilized from liver stores and circulated in the plasma primarily as retinol bound to a specific protein-protein complex, retinol binding protein (RBP)- prealbumin (2). A direct correlation between plasma levels of vitamin A and RBP in normal subjects and patients with hepatitis has been observed (3). Previous studies have suggested that low plasma vitamin A levels in CF children are due to a defect in the retinol transport system. Underwood and Denning (4), in CF children supplemented with vitamin A, found subnormal plasma levels of vitamin A coincident with normal to above normal liver levels. When CF patients were supplemented with enough vitamin A to maintain normal liver stores they displayed lower plasma vitamin A and RBP levels than did age matched controls and their plasma vitamin A levels correlated directly with RBP (5). Results of several studies suggest that zinc may play a role in vitamin A transport. Plasma vitamin A and zinc were positively correlated in a group of 61 children in Baltimore, Md. (6). In rats zinc was necessary for normal mobilization of vitamin A from the liver (6). Failure of zinc-deficient rats to mobilize vitamin A from adequate hepatic stores was attributed to impaired hepatic RBP synthesis, as levels of RBP in zinc deficient rats were low in both plasma and liver relative to zinc sufficient ad libitum or pair-fed controls (7). Since low levels of zinc have been reported in growth retarded CF patients (8), and zinc status has been observed to influence retinol transport in the rat (7), we have evaluated the relationship between zinc status and the retinol transport system in a group of CF children.3 Whereas the role of 1 From the United States Department of Agriculture, Agricultural Research Service, Human Nutrition Laboratory, Grand Forks, North Dakota and Departments of Medicine and Pediatrics, Pritzker School of Medicine, University of Chicago, Chicago, Illinois Recipient of a Josiah Macy Jr. Foundation Faculty Fellowship Award and a Nutrition Foundation Future Leaders Award. A detailed report of the overall nutriture of this group will be reported elsewhere (SOLOMONS, N. W., J. B. WAGONFELD, C. RIEGER, R. JACOB, M. BOLT, J. 638 The American Journal of Clinical Nutrition 31: APRIL 1978, pp Printed in U.S.A.

2 ZINC STATUS AND VITAMIN A TRANSPORT IN CF 639 zinc in vitamin A metabolism has been investigated in many animal studies (6, 7, 9-1 2) the present study is one of the first in which a possible interrelationship between zinc and vitamin A metabolism has been investigated in a human disease state. Methods Eighteen CF patients, ages 6 to 1 7 years, were selected from among registrants at the Cystic Fibrosis Clinic of the Wyler Childrens Hospital, University of Chicago Hospitals and Clinics, Chicago, ill. Height-for-age percentiles were calculated from the anthropometric scales of the Iowa Child Welfare Research Station and the Stuart curves of the Harvard School of Public Health. Nine of the patients had heights at or below the 10th percentile with three at or below the 3rd percentile; twelve had weights at or below the 10th percentile and ten at or below the 3rd percentile. The CF patients were receiving daily multi-vitamin supplements (Poly-vi-sol) containing 2500 IU of water miscible vitamin A per tablet. All but two were receiving replacement therapy with pancreatic enzymes and all had some degree of pulmonary disease but none had clinically significant infectious disease at the time that blood and hair samples were taken. Informed consent was obtained from the patients and their parents after the nature, purpose, and risks of the study were fully explained. For comparison purposes, samples were taken from 40 adolescent controls, ages 1 1 to 17, at a private day school after written consent had been obtained from the students parents. The study was carried out in accordance with the principles of the Declaration of Helsinki. Postprandial blood samples were drawn between 10 AM and noon in all subjects. Blood was drawn with plastic syringes with stainless steel needles, placed in trace metal free plastic tubes containing zinc free oxalate anticoagulant, and promptly centrifuged to separate red cells from plasma. Plasma zinc was determined by aspiration of a 5- fold water diluted plasma into the atomic absorption flame and calibration against aqueous standards (13). Values of plasma zinc were not obtained on hemolyzed samples. Vitamin A was determined by the macro procedure of Neeld and Pearson (14). Plasma RBP was determined by radial immunodiffusion ( M-Partigen Retinol Binding Protein Accupak Kit, Behring Diagnostics, American Hoechst Corp., Somerville, N.J ). Values for RBP on some of the samples were not obtained because ofinsufficient sample quantity. Total protein was measured by the Biuret reaction and albumin by electrophoresis on cellulose acetate membranes. Hair samples were taken from the occipital scalp with stainless steel scissors. The proximal 2 to 3 cm were cut into pieces less than 1 mm in length and washed with acetone, ether, and detergent (15). Samples of 15 to 20 mg were wet ashed with nitric and sulfuric acids, the digestate diluted to 5 ml, and zinc measured via flame atomic absorption with 5 volume percent H2SO4 matrix of both samples and calibration standards. The Student s t test for unequal variances was used for statistical comparison of mean values between groups. Linear regression analysis was also performed and included R2 calculations. The product moment correlation coefficients (r) reported represent a subset of the total of 40 correlations calculated. Results Mean concentrations and standard deviations of the various nutrients measured for the CF and adolescent control groups are shown in Table 1. Plasma vitamin A, RBP, and zinc were lower in the CF group as compared to the controls, however, the differences in RBP and zinc were not statistically significant. Hair zinc and plasma albumin were significantly lower in the CF group, whereas plasma total protein was significantly higher. Positive correlations of plasma vitamin A with RBP were found for both the adolescent control (r = , n = 15, P < 0.03) and CF (r = , n = 12, P < 0.001) groups. Plasma zinc correlated positively with both RBP and vitamin A in the CF group (Figs. 1 and 2) but not in the control group. Plasma albumin correlated positively with both zinc (r = , n = 15, P = 0.01) and RBP (r = , n = 1 2, P < 0.001) in the CF group but not in the control group. Correlations of age with plasma constituents vitamin A, RBP, zinc, and albumin are shown in Table 2. All of these constituents decreased with age in the CF population. Only albumin showed a significant correlation (positive) with age in the control group; however, attainment of statistically significant correlations in this group was hampered because of lack of a wide range of age variables. The mean values of plasma vitamin A, RBP, zinc, albumin, and hair zinc for the subgroup of CF patients with plasma zinc <70 j.g/dl are shown in Table 3. All of the values are lower than those found in the adolescent controls, although the difference VANDER HoRsr, R. ROTHBERG AND H. H. SANDSTEAD. Biochemical indices of nutrition in treated cystic fibrosis. (submitted for publication). Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture, and does not imply its approval to the exclusion of other products that may also be suitable.

3 640 JACOB ET AL. TABLE 1 Concentrations of Vitamin A, RBP, zinc, and albumin in CF patients and normal adolescent controls Mean ± SD E a 3l 25 -J 9 0. $3 CONTROLS ± ISO #{149} CONTRO,S.X± ISO p PLASMA ZINC (g/i00mi) FIG. 1. Correlation between plasma RBP and plasma zinc in CF patients. 50 CF Adolescent controls Plasma vitamin A, ±g/dl 30.7 ± ± 6.0 N=33 Plasma RBP, ±g/ml 28.3 ± ± 10.1 N=12 N=21 Plasma zinc, ±Wdl 72.1 ± ± 17.7 N=15 N=38 Hair zinc, pg/g 117 ± ± 23 N=39 Plasma albumin, g/dl 3.87 ± ± 0.25 N=40 Plasma total protein, g/dl 7.49 ± ± 0.30 N=40 b Calculated using Student s t test. Not significant statistically, P> #{176} NSb 1.48 NS 4.76 < < <0.001 E 8 a 40 CONTROLS #{149} n:15 #{149} r:0.5l1 p z 30 ± SD. > 4 (I) 4 -J 0 20 #{149} controls.±is.d. I C I I I I I I I PLASMA ZINC (jg/i00mi) FIG. 2. Correlation between plasma vitamin A and plasma zinc in CF patients.

4 ZINC STATUS AND VITAMIN A TRANSPORT IN CF 641 TABLE 2 Correlations of age with plasma concentrations of Vitamin A, RBP, zinc, and albumin in CF patients and adolescent controls Age vs Vitamin A RBP Zinc Albumin Cystic r Fibrosis P N Adolescent r Controls P N NS 33 NS 21 NS Not significant st atistically, P> TABLE 3 Concentrations of Vitamin A, RBP, zinc, albumin, and hair zinc in adolescent controls and subgroup of CF patients with plasma zinc <70 ±Wdl#{176} Mean ± SD CF plasma Zn <70 g/dl5 Plasma vitamin A, &g/dl 25.7 ± 6.6 N=7 PlasmaRBP,.Lg/ml 23.5±9.5 N=4 Plasma albumin, g/dl 3.70 ± 0.17 N=7 Hairzinc,g/g 117± 40 N=7 Adolescent controls 36.9 ± 6.0 N= ± 10.1 N= ± 0.25 N=40 162± 23 N=39 :( P NSd < Plasma zinc values of less than 70 ±g Zn/dl arbitrarily chosen as indicative of low zinc nutriture. Plasma zinc determinations on 63 male and female normal Human Nutrition Lab employees, mean age 32 years, gave an average value of 88.tg/dl and a minimum value of 69 ±Wdl. b Plasma zinc of the CF subgroup (mean ± SD) was 61.1 ± 6.2 ±g/dl compared with 78.6 ± 17.7 ±g/dl for the adolescent controls, P < Calculated using Student s t test. d Not significant statistically, P > in plasma RBP was not statistically significant.5 Discussion The lower plasma vitamin A levels and the strong correlation between vitamin A and RBP in the retinol supplemented CF children are consistent with previous studies of CF populations (4, 5, 16, 1 7). The lower mean plasma level of RBP found in the cystic fibrosis patients, although not statistically significant at P < 0.05, is also consistent with the previous studies and with the suggestions by Underwood and Denning (4) and Smith et al. (5) that low plasma levels of vitamin A in CF are due to a defect in the retinol transport system. The magnitude of this apparent defect is such that the retinol levels of the CF patients taking daily vitamin A supplements represent low-normal rather than deficient vitamin A nutriture. The positive correlations of plasma zinc with RBP and vitamin A (Figs. 1 and 2) have heretofore not been documented in CF patients. The correlations raise the question of a possible role of zinc in the retinol transport system of CF patients. Similar correlations have been observed in other human populations; between plasma zinc and vitamin A in 5-year-old children (6), and between serum zinc and RBP in patients with viral hepatitis (18). The role, if any, of zinc in vitamin A transport in CF cannot be defined from the presently available data. In general, most CF children appear to have an adequate zinc status, however some are in the marginal to deficient category. Haisted and Smith (8) reported normal plasma zinc The lower number of values for plasma RBP in Tables 1, 2, and 3 was due to insufficient sample quantities and hampered attainment of statistical significance particularly in Table 3 where N = 4 for the CF group.

5 642 JACOB ET AL. levels in CF children without growth retardation and low levels in CF children with moderate to severe growth retardation. Palm et al. (16) found normal levels in 36 CF patients. The plasma zinc values of the cystic fibrosis patients in this study were not significantly lower than the controls whereas the hair zinc values were. The data suggest a low-normal zinc status for the group as a whole while a subgroup of seven patients with plasma zinc <70 g/dl (Table 3) are in the marginal to deficient category. The lower mean values of vitamin A/RBP for this subgroup reflect the positive correlations between plasma zinc and vitamin Al RBP in the CF group as a whole. It cannot be clearly determined whether the low plasma vitamin A/RBP levels observed in this CF subgroup are a direct result of, or merely associated with, their lower zinc status. Work with rats has shown that either zinc deficiency or food restriction (pair-feeding) results in depressed plasma vitamin A/RBP levels (9, 10). To test the possibility that the correlations of vitamin A/RBP with zinc in the CF patients reflected relative growth status and tissue need for the nutrients we correlated both height and weight percentiles of the CF patients with plasma levels of each of the three nutrients. Of the six correlation coefficients calculated, only one approached statistical significance, a positive correlation of weight percentile with plasma zinc, P = This indicates that the correlations of vitamin A/RBP with zinc in the CF patients are not due to the effects of growth status on plasma vitamin A and zinc levels. Although zinc deficiency and food restriction in rats resulted in depressed plasma vitamin A/RBP levels, only zinc deficiency resulted in significantly decreased liver RBP (9). Zinc deficiency in rats has been shown to adversely affect synthesis of some proteins (19). It appears from studies in rats that RBP is one of these proteins (7, 10). Based on these findings in experimental zinc deficiency, it seems possible that poor zinc nutnture in CF patients might adversely affect hepatic RBP and/or prealbumin synthesis and hence impair the retinol transport mechanism. A second abnormality induced by impaired zinc nutriture which might have contributed to our findings is an impairment of the retinol dehydrogenase system. Huber and Gershoff (1 1) have shown that in rats, zinc deficiency significantly decreases the activity of the zinc metalloenzyme, alcohol dehydrogenase, responsible for the retinolretinal conversion. They have suggested that zinc deficiency thus impairs vitamin A utilization. Morrison et al. recently showed that oral zinc sulfate therapy was effective for correcting night blindness in zinc deficient cirrhotics even when vitamin A supplementation alone was ineffective (20). Palm et al. (16), however, found no significant change in the low plasma vitamin A/RBP levels of a group of 36 CF patients upon supplementation with oral zinc sulphate. The initial plasma zinc values of the CF group were normal, implying that the patients were not zinc deficient. It seems highly unlikely that zinc deficiency was a factor in the occurrence of the low vitamin A/RBP levels in their patients. The zinc status of the CF patients in their group however do not appear to have been strictly comparable to ours in that a subgroup of our CF patients were in the marginal to deficient category. Other factors besides zinc deficiency which might contribute to the low vitamin A/RBP profile in CF can be suggested. Intestinal malabsorption may contribute to the low levels of plasma albumin and vitamin A. This was probably not the cause of low plasma vitamin A in our patients however, as they were given water miscible retinol supplements. In addition, the patients in the Underwood and Denning (4) study mentioned previously displayed hepatic retinol stores in the normal or supranormal range. Bacterial infection has been shown to depress plasma zinc, cause an increase in hepatic zinc, increase zinc losses in urine, and cause an alteration in plasma protein patterns (21-24). It is not possible to rule out an influence by chronic pulmonary infection on the plasma or hair zinc levels in our patients, however none had severe active infection at the time of the study. Another possible explanation for low vi-

6 ZINC STATUS AND VITAMIN A TRANSPORT IN CF 643 tamin A/RBP levels in CF patients is the presence of liver disease. Abnormalities of vitamin A metabolism have been associated with various kinds of liver disease (25). In rats apo-rbp is synthesized in the liver and secreted in the plasma primarily as holo- RBP (retinol-bound) (2, 3, 26). Patients with protein-calorie malnutrition have been found to have low plasma vitamin A levels due to impaired hepatic production of RBP (27, 28). Patients with cirrhosis, chronic active hepatitis, and acute viral hepatitis displayed markedly decreased levels of vitamin A, RBP, and prealbumin (29). All three of these plasma constituents were highly correlated. In patients with acute hepatitis, the levels increased as the disease improved and plasma RBP concentrations correlated negatively with plasma bilirubin, GOT, and alkaline phosphatase. It was suggested that hepatic synthesis of RBP and prealbumin was comparably disturbed in the presence of liver disease. Significant liver disease is not unusual in patients with CF (30-32). Smith et al. (5) found hepatic abnormalities in almost half of their patients. Moreover, plasma levels of vitamin A, RBP, and prealbumin were correlated negatively with the degree of hepatic involvement. The mean values of plasma vitamin A and RBP of the half of the CF group in which hepatic abnormalities were not detected were, however, also significantly lower than mean control values. This suggests that liver disease per se was not the cause of the low vitamin A/RBP levels observed. Another consequence of systemic CF disease is hypoalbuminemia. According to Pittman et al. (33) and Strober et al. (34) the hypoalbuminemia is associated with an increase in plasma volume rather than abnormalities in hepatic albumin synthesis or albumin turnover. Increased plasma volume in our patients might account for some of the decrease in plasma vitamin A/RBP zinc/ albumin levels and the observed correlations. However others have concluded that the low vitamin A/RBP profile observed in their CF patients was not a result of hemodilution (5). The strong correlation between plasma albumin and RBP in our patients and in Smith s patients (5) suggests that the depressed levels of these two proteins were due to a common factor(s). Possible common factors have been mentioned above. Giroux (35) found in 156 hospitalized patients and 28 controls, that total serum zincalbumin bound zinc were highly correlated as were albumin bound zinc-albumin. He concluded that total serum zinc concentration reflects serum albumin concentration. This is seen in our CF group by the plasma albumin-zinc correlation. The correlations of zinc with RBP and vitamin A may therefore reflect the albumin-rbp correlation. The unusual inverse correlations of vitamin A/RBP/zinc/albumin with age (Table 2) have been previously reported for vitamin A (5, 1 7) but not for zinc and RBP. The correlations may reflect an increase in the magnitude of pertinent CF disease factors with age. It is clear from the above that it is impossible to define the role of zinc nutriture in the observed findings with certainty. Observations in rats support the concept that zinc was a contributive factor to the low plasma vitamin A/RBP levels in the zinc deficient CF patients. However, the observed correlations of vitamin A/RBP with plasma zinc do not define cause and effect, and may therefore be secondary to CF disease factors such as liver disease and/or hemodilution. Proof of a relationship between zinc status and vitamin A transport in humans will require a therapeutic intervention with zinc in patients known to be zinc deficient.!! The authors are indebted to Mr. George Logan and Ms. LuAnn Johnson for their help with the statistical treatment of the data. We would like to thank Dr. Robert Rosenfleld and Ms. Anne Eggleton for their assistance with the adolescent control subjects. References 1. ANDERSEN, D. H. Cystic fibrosis of the pancreas. Vitamin A deficiency and bronchiectasis. J. Pcdiat. 15: 763, KANAI, M., A. RAZ AND DEW. GooDMAN. Retinol-binding protein: The transport protein for vitamin A in human plasma. J. Clin. Invest. 47: 2025, SMITh, F. R., A. RAZ AND DEW. GooDMAN. Radioimmunoassay of human plasma retinol-binding protein. J. Clin. Invest. 49: 1754, UNDERWOOD, B. A., AND C. R. DENNING. Blood and liver concentrations of vitamins A and E in

7 644 JACOB ET AL. children with cystic fibrosis. Pediat. Res. 6: 26, SMrru, F. R., B. A. UNDERWOOD, C. R. Dr- NING, A. Vas AND D. S. GOODMAN. Dcpressed plasma retinol-binding protein levels in cystic fibrosis. J. Lab. Clin. Med. 80: 423, Mm, J. C., JR., E. G. MCDANIEL, F. F. FAN AND J. A. HALSTED. Zinc: a trace element essential in vitamin A metabolism. Science 181 : 954, SMrru, J. E., E. D. BROWN AND J. C. Sismi, JR. The effect of zinc deficiency on the metabolism of retinol-binding protein in the rat. J. Lab. Chin. Med. 84: 692, FLusmr, J. A., AND J. C. SMrru, JR. Plasma-zinc in health and disease. Lancet 1 : 322, BROWN, E. D., W. Ciwt AND J. C. SMrru, JR. Vitamin A metabolism during the repletion of zinc deficient rats. J. Nutr. 106: 563, St,srru, J. C., JR., E. D. BROWN, E. G. MC- DANIEL AND W. Cu.N. Alterations in vitamin A metabolism during zinc deficiency and food and growth restriction. J. Nutr. 106: 569, HUBER, A. M., AND S. N. GERSHOFF. Effects of zinc deficiency on the oxidation of retinol and ethanol in rats. J. Nutr. 105: 1486, CARNEY, S. M., B. A. UNDERWOOD AND J. D. LOERCH. Effects of zinc and vitamin A deficient diets on the hepatic mobilization and urinary excretion of vitamin A in rats. J. Nutr. 106: 1773, SINMA, S. N., AND E. R. GsamLI. Serum copper and zinc levels in various pathologic conditions. Am. J. Chin. Pathol. 54: 570, NEELD, J. B., JR., AND W. N. PEARSON. Macroand micromethods for the determination of serum vitamin A using trifuoroacetic acid. J. Nutr. 79: 454, KLEVAY, L. M. Hair as a biopsy material. I. Assessment of zinc nutriture. Am. J. Chin. Nutr. 23: 284, PALIN, H. D., B. A. UNDERWOOD AND C. R. DENNING. The effect of oral zinc supplementation on plasma levels of vitamin A and retinol-binding protein. Pediat. Res. 10: 358, KNOEPFLE, G., H. W. ROTrHAUWE AND A. ODENThAL. Serum concentrations of vitamin A, carotene, retinol-binding protein and prealbumin in patients with cystic fibrosis. Z. Kinderheilkd. 119: 279, HENKIN, R. I., AND F. R. SMITh. Zinc and copper metabolism in acute viral hepatitis. Am. J. Med. Sci. 264: 401, KIRCHGESSNER, M., H. P. ROTh AND E. Wm- GAND. Biochemical changes in zinc deficiency. In: Trace Elements in Human Health and Disease, Vol. I. Edited by A. S. Prasad. New York: Academic Press, 1976, pp MORRISON, S. A., R. M. RUSSELL, E. V. OAKS AND E. A. CARNEY. Reversal of night blindness in cirrhotics by zinc sulfate. Am. J. Chin. Nutr. 30: 612, BEISEL, W. R., W. D. SAWYER, E. D. RYLL AND D. CROZIER. Metabolic effects of intra-cellular infections in man. Ann. Internal Med. 67: 744, PEKAREK, R. S., AND W. R. BEISEL. Characterization of the endogenous mediator(s) of serum zinc and iron depression during infection and other stress. Proc. Soc. Expth. Biol. Med. 138: 728, SANDsmD, H. H.,K. P. Vo-KMACnJANDN. W. Soi.oMoNs. Conditioned zinc deficiencies. In: Trace Elements in Human Health and Disease. Vol. I. Edited by A. S. Prasad. New York: Academic Press, 1976, pp BEISEL, W. R., R. S. PEKAREK AND R. W. WAN- NEMACHER. Homeostatic mechanism affecting plasma zinc levels in acute stress. In: Trace Elements in Human Health and Disease, Vol. I. Edited by A. S. Prasad. New York: Academic Press, 1976, pp H.*ismD, J. A., AND J. C. Smith, JR. Night blindness and chronic liver disease. Gastroenterology 67: 193, MUTO, Y., J. E. SMITH, P. 0. MILCH AND D. S. GOODMAN. Regulation of retinol-binding protein metabolism by vitamin A status in the rat. J. Biol. Chem. 247: 2542, SMmi, F. R., DEW. GOODMAN, M. S. ZAKLAMA, M. K. GABR, S. E. MARAGHY AND V. N. PATWAR- DHAN. Serum vitamin A, retinol-binding protein and prealbumin concentrations in protein-calorie malnutrition. I. A functional defect in hepatic retinol release. Am. J. Chin. Nutr. 26: 973, SMrru, F. R., DEW. GOODMAN, G. ARROYAVE AND F. VritR. Serum vitamin A, retinol-binding protein, and prealbumin concentrations in proteincalorie malnutrition. II. Treatment including supplemental vitamin A. Am. J. Chin. Nutr. 26: 982, SMrm, F. R., AND DEW. GOODMAN. The effects of diseases of the liver, thyroid and kidneys on the transport of vitamin A in human plasma. J. Clin. Invest. 50: 2426, DI SANT AGNESE, P. A., AND W. A. BLANC. Distinctive type of biiary cirrhosis of liver associated with cystic fibrosis of pancreas: recognition through signs of portal hypertension. Pediatrics 18: 387, CEAIG, J. M., H. HADDAD AND H. SHWACHMAN. Pathological changes in liver in cystic fibrosis of pancreas. Am. Med. Assoc. J. Diseases Children 93: 357, STERLING, K. Serum albumin turnover in Laennec s cirrhosis as measured by I 31-tagged albumin. J. CIin. Invest. 30: 1238, PITrMAN, F. E., C. R. DENNING AND H. G. BARKER. Albumin metabolism in cystic fibrosis. Am. J. Diseases Children 108: 360, STROBER, W. P., G. PETER AND R. H. SCHWARTZ. Albumin metabolism in cystic fibrosis. Pediatrics 43: 416, Gmoux, E. L., M. DURIEUX AND P. J. SCHECH- TER. A study of zinc distribution in human serum. Bioinorg. Chem. 5: 211, 1976.