Zinc Homeostasis and Kinetics in Children With Cystic Fibrosis (CF)

Overt zinc deficiency presents with features of acrodermatitis enteropathica (AE) including diarrhea, dermatitis, growth failure, and poor immune function. AE has been described as a complication of cystic fibrosis, and can be the initial presenting feature. As AE has almost never been described in otherwise healthy subjects consuming self-selected diets, this suggests a profound disturbance in zinc metabolism, either in absorption or excretion, can occur in CF.

Few studies have measured zinc absorption in children with CF. Those that have had demonstrated significant reductions in zinc absorption in CF subjects with pancreatic insufficiency with a significant increase in zinc absorption with pancreatic enzyme replacement. However, zinc metabolism is maintained by changes in fractional zinc absorption and zinc losses in to the gastro-intestinal tract (endogenous fecal zinc excretion). Of these two mechanisms, changes in endogenous fecal zinc excretion appear to be the more important. To date, only one study has measured endogenous fecal zinc excretion in subjects with CF. Krebs et al. studied 15 infants, less than 4 months of age, newly diagnosed with CF. They found that all infants were in negative zinc balance, and that endogenous fecal zinc excretion was significantly correlated with the degree of fat malabsorption. These results are difficult to interpret as there was no control group studied, and there is little published data on zinc absorption in healthy infants of this age, and no data on endogenous fecal zinc excretion.

Clearly, a better measure of zinc status is required to fully understand the importance of zinc deficiency in CF, and the relationship between fat malabsorption and zinc balance. In the past, we have used novel stable isotope-based multicompartmental models to study zinc status in healthy and disease states. These have proved able to detect changes in zinc metabolism in two experimental paradigms of zinc deficiency, even before the plasma zinc becomes abnormal.

We hypothesize that,

1. Children with CF will have worse zinc status, lower fractional zinc absorption, increased endogenous fecal zinc excretion, and poorer zinc balance, than age-matched controls.

2. In children with CF, endogenous fecal zinc excretion will be significantly positively correlated, and zinc absorption and zinc balance significantly negatively correlated with fecal fat excretion.

3. After two months of zinc supplementation with 20 mg/d zinc as zinc acetate, children with CF will have similar zinc status, fractional zinc absorption, endogenous fecal zinc excretion, and zinc balance as age-matched controls.

4. Two months of zinc supplementation will not affect iron status (Hemoglobin, serum ferritin, serum transferrin receptors) or copper status (serum copper, ceruloplasmin, copper/ zinc superoxide dismutase).

Twenty-four children with CF aged 8-14 years old will be recruited from the Cystic Fibrosis Clinic at Texas Children's Hospital. Children will be considered eligible if they have a diagnosis of CF, weigh more than 25 kg, have a clinical diagnosis of pancreatic insufficiency requiring pancreatic enzyme replacement, and have a history of good compliance with medical therapies. Subjects will be considered ineligible if they have had gastrointestinal (GI) tract surgery, have pancreatic endocrine dysfunction, or are taking more than 10 mg/d zinc supplements. The study will be explained to the subjects and their parents. Informed written consent will be obtained from the parents, and written consent or assent (as age appropriate) from the subjects.

After consent is obtained, a registered dietician will use a food frequency questionnaire to assess the child's usual zinc intake. Advice will be given on how to eat a healthy balanced diet that should meet the recommended daily allowance (RDA) for zinc and other essential minerals. Except for this dietary advice, no other dietary intervention will be carried out.

Twelve healthy children aged 8-14 years old will be recruited by public advertisement, local contacts and by word of mouth. They will be eligible for the study if they are of similar age, gender and ethnicity to the CF subjects. Informed consent and assent will be obtained in the same manner as for the CF children, and the same dietary advice and guidance given.

The children with CF will be randomized to receive a zinc supplement of 20 mg/d as zinc acetate or an identical appearing placebo. Randomization will be stratified by gender and by ethnicity. Randomization tables will be maintained by the Investigation Pharmacy Service at Texas Children's Hospital. Once consent is obtained the Investigation Pharmacy Service will be contacted and will randomize the children to receive one of two identical appearing solutions - one containing 10 mg/mL elemental zinc as zinc acetate and one being the placebo. The subjects, their parents and the investigators will be blinded as to which solution the child has been randomized to receive.

Children will be admitted to the General Clinical Research Center after an overnight fast. A local anesthetic cream will be applied over veins in the antecubital fossa. Blood, 10-15 mL, will be drawn for baseline laboratory investigations including a complete blood count, liver function test, plasma zinc, serum ferritin, serum transferrin receptor concentration, serum copper, serum ceruloplasmin and erythrocyte cU/Zn- superoxide dismutase. After the blood has been taken, subjects will be provided with their first bottle of zinc solution, or placebo. They will be instructed to take 1 mL of the solution twice daily for two months. New supplies of the zinc or placebo solution will be provided as required during the study. All used or partially used bottles will be returned to the Investigation Pharmacy Service and the amount taken calculated from the change in weight of the bottle.

After consuming the placebo or zinc solution for two months, children with CF will be admitted to the General Clinical Research Center following an overnight fast. Subjects will be asked to empty their bladder. An intravenous heparin-lock catheter will be inserted into a vein in the antecubital fossa or the dorsum of the hand and a baseline blood sample taken for serum chemistries.

Subjects will be given a breakfast containing approximately 3 mg of zinc, including 120 mL of orange juice to which 2 mg of zinc-67 had been added 18-24 hours earlier. Immediately after breakfast, 0.5 mg zinc-70 will be infused by venipuncture of the other arm over 30-60 seconds, and the butterfly infusion set flushed with 5 mL of normal saline.

Blood samples will be drawn from the heparin lock catheter 2, 4, 6, 8, 10, 15, 20, 30, 60, 120 and 180 minutes after the intravenous zinc isotope is given. Further blood samples will be drawn by venipuncture 24, 72, and 120 hours after the intravenous isotope was given. These blood samples will be anticoagulated in certified zinc-free ammonium heparin tubes (Monvette 9ml AH, Sarstedt Inc., Newton, NC). Plasma will be separated by centrifugation and stored at -80°C pending analysis. A complete urine and fecal collection will be made during the metabolic study. Urine samples will be collected in 8-h aliquots and all stools passed collected individually. Urine and fecal samples will be stored at -20°C pending analysis.

Control children will be studied in exactly the same manner, approximately 2 months after baseline bloods were drawn.

The primary hypothesis is that children with CF will have lower fractional zinc absorption, increased endogenous fecal zinc excretion, poorer zinc balance and worse zinc status, than age-matched controls. This will be assessed using ANOVA. If this suggests a significant "group" effect, post-hoc testing will be carried out using the Tukey-Kramer test. The kinetic parameter used to assess zinc status will be k(2,1) as this showed the largest change in response to zinc deficiency, in our previous studies. This analysis will test both hypotheses 1 and 3 (see above). Should any significant differences exist in age, weight, etc. between the groups, this will be statistically corrected for using ANCOVA. Data will be analyzed for the entire study population, and for the CF children with pancreatic insufficiency alone. The study is designed to have an 80% power of detecting changes in k(2,1) between the two groups of pancreatic insufficient CF children alone.

The relationship between fecal fat and endogenous fecal zinc excretion, and zinc balance (hypothesis 2), will be assessed using simple and multiple regression analysis. The effect of zinc supplementation on iron and copper status will be examined using a 2 sample t-test, comparing CF children with zinc supplementation, to those who did not receive zinc supplementation.