Case Report. Trace element concentrations in patients on home enteral feeding: two cases of severe copper deficiency. Introduction

Transcription

1 Case Report Trace element concentrations in patients on home enteral feeding: two cases of severe copper deficiency A Oliver 1, KR Allen 1 and J Taylor 2 Abstract Addresses 1 Department of Clinical Biochemistry, Leeds Teaching Hospitals, Britannia House, Britannia Road, Morley, Leeds LS27 0DQ, UK 2 Department of Nutrition and Dietetics, South Leeds Primary Care Trust, St Mary s Hospital, Leeds LS12 3QE, UK Correspondence Mr KR Allen keith.allen@leedsth.nhs.uk Background: Enteral feeding is the fastest growing area of artificial nutrition, with the annual rate of growth being estimated at 20 25% a year. Previous studies have demonstrated trace element deficiencies in patients on long-term home enteral nutrition (HEN). Methods: The trace elements zinc, selenium, copper and manganese were measured in blood samples from 37 patients on HEN using atomic absorption spectroscopy. Results: Plasma zinc concentrations (range mmol/l) were below the reference range ( mmol/l) in 30 patients, plasma selenium concentrations (range mmol/l) were below the reference range ( mmol/l) in only one patient. Whole blood manganese (range nmol/l) and plasma manganese (range nmol/l) were above both respective reference ranges ( nmol/l and 9 24 nmol/l) in four patients. Two patients showed severely low plasma copper concentrations of 2.4 mmol/l and 2.5 mmol/l, and responded to treatment with extra copper supplementation. Conclusion: Although enteral feeds contain adequate concentrations of trace elements, problems with bioavailability may occur and patients receiving long-term enteral feeding should be monitored with regard to plasma trace element concentrations. Introduction Enteral feeding is becoming increasingly used for nutritional support of patients in the community. Conditions that necessitate this type of arti cial feeding include a decreased appetite due to cancer or chronic illness, an inability to feed as a result of neurological disease or cerebrovascular accident and oesaphageal obstruction due to stricture or tumour. Many of the patients on long-term home enteral feeding will receive their nutrition via a gastrostomy tube inserted directly into the stomach (percutaneous endoscopic gastrostomy [PEG]). In some cases, jejunostomy feeding may be indicated and the feed tube is surgically placed directly into the jejunum (jejunostomy feeding). Patients on home enteral nutrition (HEN) have previously been seen to be de cient in a variety of micronutrients, including copper, due to inadequate amounts of trace elements in the feed solutions. 1,2 The trace element status with regard to plasma selenium, copper, zinc and manganese and whole blood manganese was evaluated in 37 patients receiving long-term HEN. Patients and methods Patients Thirty-seven patients (17 men, 20 women) aged between 38 and 95 years (median age 75 years) were included in this study. Ethical permission was obtained from the Research Ethics Committee, Leeds Teaching Hospitals. The patients had been receiving HEN for four months to 6.5 years (median two years). Sixteen of the patients were receiving HEN as a result of a cerebrovascular accident and the remaining patients had a range of miscellaneous conditions, including neurological problems (dementia, multiple sclerosis, Huntington s 136 r 2005 The Association of Clinical Biochemists

2 Trace element concentrations on home enteral feeding 137 chorea) and post-surgical conditions a ecting their ability to eat and swallow. Various brands of commercially available enteral feeds were used to support these patients. Twelve patients required a standard bre-enriched feed (Fresubin Original Fibre), 13 patients a high-energy bre-enriched feed (Fresubin Energy Fibre) and six patients a standard feed (Fresubin Original). These feeds were obtained from Fresenius Cabi Homecare Ltd,Warrington, UK. Four patients were receiving the bre-enriched feed Jevity s and two patients were receiving a specialized peptide-based feed Perative s (Abbott Nutrition, Abbott House, Maidenhead, UK). The majority of patients had their feed administered via a gastrostomy tube, with only two patients receiving their feed via a jejunostomy tube. All received enteral feed via prolonged infusion over a time span of several hours, and where possible overnight. Methods Heparinized blood was collected from all patients for the measurement of plasma zinc, selenium, copper and manganese and whole blood manganese. Copper and zinc were measured by ame atomic absorption spectroscopy and selenium and manganese by graphite furnace atomic absorption spectroscopy. Plasma C-reactive protein (CRP) was measured by an automated immunoassay method as a marker of the in ammatory response. Routine investigations including renal, liver, bone and haematology pro les were also carried out on all patients in this study. Reference ranges for trace elements measured in this study were established by the Supra-Regional Assay Laboratories for Trace Elements in the UK using atomic absorption spectroscopy. 3 Results Table 1. Trace element concentrations in 37 patients receiving long-term home enteral feeding Trace element Median (range) Reference range Zinc (mmol/l) 10.6 ( ) Copper (mmol/l) 18.9 ( ) Selenium (mmol/l) 1.2 ( ) Whole blood manganese 164 (74 309) (nmol/l) Plasma manganese (nmol/l) 28 (13 51) 9 24 The median and range for the trace element concentrations in the patients are shown in Table 1. Plasma zinc concentration was less than the lower limit of the reference range in 30 patients, whereas there was only one patient with a plasma selenium below the lower limit of the reference range. Both whole blood and plasma manganese were slightly elevated in four of the patients compared with their respective reference ranges. Plasma CRP concentrations ranged from o5 to 53 mg/ L, with a median concentration of 12 mg/l. In all, 21 of the patients had a CRP concentration below 20 mg/l. Patients with trace element concentrations outside their respective reference ranges were further investigated by reference to their medical notes and routine biochemistry and haematology results. No obvious clinical explanations for any of the abnormal trace elements could be found (e.g. no cholestasis was evident in patients with raised whole blood and plasma manganese). Four patients had a plasma copper concentration below the reference range. Two of these patients had marginally low copper concentrations (10.6 mmol/l and 10.5 mmol/l); however, the remaining two patients had severely low plasma copper concentrations (2.4 mmol/l and 2.5mmol/L). These two patients underwent further investigation to establish the aetiology of their low plasma copper. Case reports This case involved a 76-year woman, who had been receiving HEN via a gastrostomy for three years and eight months following a cerebral vascular accident. Her past medical history included idiopathic liver disease resulting in cirrhosis and oesophageal varices. Medical records also indicated diabetes mellitus type 2 and a history of renal failure following blood loss due to rupture of oesophageal varices. Renal function tests at the time of investigation of trace element status showed no evidence of renal impairment (plasma creatinine 79 mmol/l). Liver function tests showed no increase in plasma bilirubin or ALT, but alkaline phosphatase was slightly elevated at 515 IU/L (reference range IU/L). Previous investigations for her liver cirrhosis showed no evidence for haemochromatosis, but there was no evidence to rule out the possibility of Wilson s disease. The present medication consisted of frusemide, digoxin, lansoprazole, temazepan, paroxetine and senna. The patient was receiving and tolerating 1 L of bre-enriched enteral feed, Jevity s, and L of water daily. A repeat blood sample from this patient con rmed low plasma copper of 2.7 mmol/l and low plasma caeruloplasmin of 0.07 g/l. The only other trace element abnormality demonstrated in this patient was a slightly low plasma zinc concentration (8.5 mmol/l), possibly

3 138 Oliver et al. related to the low plasma albumin of 30 g/l. Haematological investigations showed a slightly low haemoglobin of 11.2 g/dl ( ), with a normal white blood cell count of /L and a neutrophil count of /L. This patient was a 75-year-old man who had been receiving the same brand of bre-enriched feed as (Jevity s ) for the relatively short time of nine months via a gastrostomy. There was no past medical history of note until the patient had a CVA. Renal and liver function test showed no abnormalities and plasma albumin was 38 g/l (reference range g/l). Medication at the time of investigation included baclofen, frusemide, amiloride, aspirin, omeprazole, paroxetine and simvastatin. Plasma concentrations of other trace elements zinc, selenium and manganese were all within the reference range (10.6 mmol, 1.4 mmol/l and 31nmol/L, respectively). Haematological investigation showed a slightly low haemoglobin of 12.4 g/dl and normal white blood cell and neutrophil counts of /L and / L, respectively. A repeat blood test con rmed the low plasma copper concentration of 2.5 mmol/l and the low plasma caeruloplasmin concentration of 0.05 g/l. A literature search established no known interferences with medication for both patients with regard to copper absorption. The patients were investigated further by measurement of urine copper excretion and eye examination for the presence of Kayser--Fleisher rings, to exclude the likelihood of Wilson s disease. Patient 1 had a urine copper excretion of 0.1 mmol/mmol creatinine, and in Patient 2 there was no detectable copper in the urine (reference range for urine copper o0.1 mmol/mmol creatinine). There was no evidence of copper deposits as Kayser--Fleischer rings in the eye in either of the patients. Copper de ciency in the enteral feed itself was excluded by the measurement of copper using ame atomic absorption spectroscopy. Both patients were receiving at least 33 mmol of copper per day, which is within the recommended daily allowance of mg ( mmol) of copper. 4 Three other patients in this study who were supported by the same brand of feed had plasma copper concentrations within the reference range. Although both patients were receiving the recommended daily intake of copper in their enteral feed, further copper supplementation was sought to see if this would increase their plasma copper concentrations. Copper supplementation in the form of copper histidinate is only available for the treatment of the rare copper-de cient condition Menkes disease, on a named patient basis. The only forms of copper available for supplementation in these enterally fed patients were those containing other trace elements and vitamins. The multivitamin and trace element preparation Minadex (Seven Seas Health Care Ltd, Hull, UK) was chosen because it was available in a low-viscosity liquid form that could be easily administered via the PEG route. Minadex contains 2 mmol of copper per 5 ml of solution and both patients were supplemented with 10 ml of Minadex (4 mmol of copper) a day. Blood samples were taken for the measurement of copper and caeruloplasmin during the course of this treatment and the results are shown in Figures1and 2, respectively. Plasma copper (μmol / L) Plasma caeruloplasmin (g / L) Days during copper supplementation Figure 1. Plasma copper concentrations during copper supplementation in two cases of copper deficiency Days during copper supplementation Figure 2. Plasma caeruloplasmin concentrations during copper supplementation in two cases of copper deficiency.

4 Trace element concentrations on home enteral feeding 139 responded well to the Minadex and after day 45 the dose was increased to 30 ml a day, with the result that plasma copper concentration reached levels within the reference range. After day 100, Minadex supplementation was reduced to 10 ml a day due to the possibility that it was causing bowel disturbances. Plasma caeruloplasmin concentration followed the trend of the increased plasma copper concentration. initially showed no response to supplementation. However, after 45 days the Minadex was administered separately from other medication. This resulted in a slight increase in plasma copper concentration. On day 69, supplementation was ceased due to increased bowel movements and the patient s plasma copper concentration decreased. Discussion Previous cases of severe copper de ciency have been reported in patients receiving long-term enteral feeding and were associated with signs of severe anaemia and neutropenia. 5,6 These cases occurred prior to improvements in the composition of enteral feeds and the cause of copper de ciency was related to an insu cient amount of the element in the feed. The estimated safe and adequate dietary intake for copper in adults has been estimated to be mg/day, 4 and studies carried out by Higuchi et al. 5 have estimated the minimum dietary copper to maintain a normal serum copper concentration to be 1.5 mg/kg/day. Both copper-de cient patients in this study were receiving at least 33 mmol (2 mg) of copper daily, and therefore inadequate intake was not responsible for the hypocupraemia. Other possible explanations for the low plasma copper concentrations in these patients include reduced bioavailability or increased excretion. Radioisotopic studies in experimental animals have shown copper to be absorbed primarily in the duodenum and to some extent in the stomach and distal part of the small bowel. 7 The absorbed copper is transported to the liver where it is incorporated into caeruloplasmin, which is released into the blood stream and carried to the tissues. Most endogenous copper is lost via the bile, with small amounts being lost in pancreatic and intestinal uids and intestinal cells, and eliminated from the body.very little copper is lost in the urine and relatively little copper is stored in the body compared with other elements such as zinc and iron. 8 Excessive losses of copper in the urine have been described in patients with nephrotic syndrome due to increased permeability of the glomerulus to caeruloplasmin. Increased urine loss of copper has also been associated with TPN due to urinary loss of aminoacid-bound copper. 9 However, none of these factors account for the hypocupraemia observed in the two patients in this study. Factors a ecting the bioavailability of copper have been extensively reviewed by Lonnerdal. 10 Much of the research in this area comes from studies in experimental animals and cannot be directly related to humans. One study using stable copper isotopic absorption studies in young men 11 showed that dietary bre had little e ect on copper bioavailability. However, the same study demonstrated gross intra-individual variation in copper absorption. Increased dietary intake of zinc does decrease absorption of copper from the gut and this phenomenon, is used in the treatment of Wilson s disease. 12 Copper absorption may also be a ected by high levels of dietary intake of iron. However, both copperde cient patients in this study were receiving concentrations of zinc and iron in their enteral feed within the recommended daily allowances. The interaction between copper absorption and therapeutic drugs is an area that has not been extensively studied, and this needs to be taken into account in these copper-de cient patients. in this study only showed an increase in plasma copper when further copper supplementation was given at a time di erent from that of the enteralfeedanddrugtreatment. This study demonstrates the need to monitor trace elements in patients on long-term enteral feeding. Copper, in particular, due to its relatively low concentration in enteral feeds compared with elements such as zinc and iron, may be particularly susceptible to problems of bioavailability. References 1 Yagi M, Tani T, Hashimoto T, et al. Four cases of selenium deficiency in postoperative long-term enteral nutrition. Nutrition 1996; 12: Masugi J, Masahiko A, Fukuda T. Copper deficiency anaemia and prolonged enteral feeding. Ann Intern Med 1994; 121: Personal communication from Dr A Taylor, Chairman of the SAS Trace Element Laboratories for the UK 4 National Research Council; Havel RJ, Calloway DH, Gussow JD, Mertz W, Nesheim MC, eds. Recommended Dietary Allowances. 10th edn. Washington, DC: National Academy Press, 1989; Higuchi S, Higashi A, Nakamura T, Matsuda I. Nutritional copper deficiency in severely handicapped patients on a low copper enteral diet for a prolonged period: estimation of the required dose of dietary copper. J Paed Gastroenterol Nutr 1988; 7: Williams DM. Copper deficiency in humans. Sem Haematol 1983; 29: Van Campen DR, Mitchell EA. Absorption of Cu 64,Zn 65,Mo 99, and Fe 59 from ligated segments of the rat gastrointestinal tract. J Nutr 1965; 86: Turnlund JR. Human whole-body copper metabolism. Am J Clin Nutr 1998; 67(Suppl): 960S 4S 9 Beshgetoor D, Hambidge M. Clinical conditions altering copper metabolism in humans. Am J Clin Nutr 1998; 67(Suppl): 1017S 21S

5 140 Oliver et al. 10 Lonnerdal B. Bioavailabilty of copper. Am J Clin Nutr 1996; 63(Suppl): 821S 9S 11 Turnland JR, King JC, Gong B, Keyes WR, Michel MC. A stable isotope study of copper absorption in young men effect of phytate and a cellulose. Am J Clin Nutr 1985; 42: Hoogenraad TU, Van Hattum J, Van den Hamer CJA. Management of Wilson s disease with zinc sulphate. Experience in a series of 27 patients. J Neurol Sci 1987; 77: Accepted for publication 22 November 2004