Aims In high doses zinc may cause copper deficiency, a diagnosis that is often missed resulting in anaemia, neutropenia and irreversible neurological symptoms. The aim of this study was to assess if zinc deficiency is erroneously diagnosed by misinterpretation of plasma zinc concentrations and whether copper deficiency is induced in patients prescribed zinc.
Methods Casenotes of 70 patients prescribed zinc were scrutinised. Plasma concentrations of zinc, copper, C reactive protein and albumin were recorded from the laboratory database.
Results 62% of patients were prescribed zinc at doses sufficient to cause copper deficiency. In 48% of the patients, plasma zinc concentrations were low as a probable result of hypoalbuminaemia or the systemic inflammatory response rather than deficiency. Awareness of copper deficiency was lacking; it was only documented as a possible side effect in one patient and plasma copper was measured in only two patients prescribed zinc. 9% of patients developed unexplained anaemia and 7% developed neurological symptoms typical of copper deficiency.
Conclusions Zinc deficiency is frequently misdiagnosed on the basis of low plasma zinc concentrations. The potential risk of copper deficiency developing in patients prescribed high doses of zinc is apparently infrequently considered. It is probable that a significant minority of patients prescribed with high doses of zinc develop iatrogenic copper deficiency.
- LABORATORY TESTS
Statistics from Altmetric.com
A diagnosis of zinc deficiency is normally made on the basis of a low concentration of plasma zinc. However, there are no guidelines that assign a concentration below which it is appropriate to prescribe zinc or what dose is applicable. Although plasma zinc is a good marker of zinc status,1 zinc concentrations may also fall independently of status in patients with a systemic inflammatory response (SIR) and/or low concentrations of its carrier protein albumin.2 ,3 Consequently, there is potential for erroneously diagnosing zinc deficiency on the basis of low plasma concentrations.
Zinc is essential in trace amounts of 5.5–9·5 mg/day for men and 4–7 mg/day for women.4 However, zinc supplements are usually only available in minimum formulations of 45 or 50 mg of elemental zinc. There is little to suggest that such doses are unsafe over short time periods. However, larger doses ingested as supplements5 ,6 or unwittingly consumed in excess as zinc-containing dental fixatives7 ,8 ,9 may cause copper deficiency. Ingestion of zinc in excess of requirements causes enterocytes to produce metallothionein which binds zinc until these cells are sloughed off into the gut lumen.10 However, copper binds more avidly to metallothionein11 and so is also lost potentially resulting in deficiency. The clinical consequences may be anaemia and neutropenia. However, if the diagnosis is missed or delayed, disabling and frequently irreversible neurological symptoms may ultimately develop.9 ,12
We investigated if cases of zinc deficiency were misdiagnosed through erroneous interpretation of low plasma zinc secondary to an SIR or low plasma albumin and whether undiagnosed copper deficiency may be present in patients prescribed zinc but who are not followed up with this possibility in mind.
Records of prescriptions made in Glasgow Hospitals are stored in a central database and the names and dates of birth of 88 patients who were given zinc supplements during admission were obtained from two hospitals in Glasgow from 2000 to 2010.
Of these 88 patients, 18 were excluded (2 who were prescribed topically applied zinc and 16 whose casenotes contained no relevant information). Details of the 70 remaining patients were compiled for this audit.
Where available, plasma concentrations of zinc, copper, albumin and C reactive protein (CRP), as a marker of the SIR, were obtained from the laboratory database. A CRP concentration >20 mg/L was considered to be sufficiently raised to account for a low plasma zinc concentration being due to an SIR.2 Similarly, an albumin concentration less than 25 g/L was considered sufficiently low to account for low plasma zinc concentrations.
Their casenotes were scrutinised in order to obtain the following information: the reason for prescribing zinc; whether advice was given to general practitioners on the management of zinc therapy and the possibility of copper deficiency developing; the length of time zinc was prescribed; and the subsequent development of haematological or neurological symptoms consistent with copper deficiency.
If patients developed symptoms consistent with copper deficiency and zinc was prescribed in doses of over 100 mg/day, the general practitioner was contacted initially by phone and then by letter. The background to the audit was explained and advice was given on stopping or reducing the dose of zinc.
The local hospital ethics group considered this study to be an audit and so concluded that ethical approval was not required.
The doses and formulations of zinc prescribed were only available for 52 patients, shown in table 1. With the possible exception of one patient, all were prescribed zinc as the sulfate.
In 29 patients, the reason for prescribing zinc was not recorded. In the remainder, zinc was prescribed because of zinc deficiency in 21 (43%), dermatological reasons such as pressures sores, poor skin or lower limb ulcers in 19 (38%), poor nutrition in 4 (8%), alcohol withdrawal support in 1 and alopecia in 1.
Plasma zinc was measured prior to prescription in 43/70 patients (61%) and was found to be low in 37. In 28 of these patients (76% of this subgroup and 40% of all patients) the reason for a low plasma zinc concentration was considered more likely to be secondary to an SIR or hypoalbuminaemia rather than zinc deficiency. In 22 of these 37 patients (59%) there was a significant SIR (mean CRP=84 mg/L, range 22–287 mg/L), in 5 patients (14%) hypoalbuminaemia (≤25 g/L) was present, and 1 patient had both hypoalbuminaemia and SIR. There was no SIR or hypoalbuminaemia in nine patients (24%).
Copper concentration was only measured in two patients and was low in both: 3.1 µmol/L and 7.0 µmol/L (lower reference limit: 10 µmol/L in male patients and 10.7 in female patients).
In 66 casenotes (94%), no further reference to zinc was made following its prescription. Zinc treatment was not usually recorded in the drug information file and consequently it was difficult to assess how long zinc was prescribed. No information on compliance was available. The casenotes were examined to find any record of a warning to the General Practitioner (GP) of the potential risk of copper deficiency when prescribing high dose zinc supplements over an extended time period. Only one such warning was recorded.
Thirteen patients had anaemia, neutropenia and/or neurological symptoms, which are typically associated with copper deficiency. Of 12 patients with anaemia and/or neutropenia, in six the cause was unknown and in six the finding either pre-dated zinc supplementation or was explicable by other diagnoses. The doses prescribed were 90 mg/day in one patient, 135 mg/day in two and unknown in three. Five patients, two of whom also had unexplained anaemia, had neurological symptoms: peripheral neuropathy (two), finger paraesthesia (one), ataxia (one) and lower limb neuropathy (one). In the nine patients who developed unexplained signs and/or symptoms typical of copper deficiency, no mention was made of these in casenotes prior to initiation of zinc. Haemoglobin measurements and neurological investigations, however, had not previously been carried out and so their presence before starting zinc could not be definitively excluded. A putative diagnosis of myelodysplastic syndrome was made in an 82-year-old lady who developed severe macrocytic anaemia and neutropenia (haemoglobin=43 g/L, neutrophil count=0.1×109/L) 9 months after being prescribed 135 mg/day for bed sores. This patient decided to stop zinc supplementation and her haematology subsequently improved. The casenotes of the eight remaining patients contained no information as to whether zinc treatment was stopped.
Copper deficiency usually presents as anaemia and neutropenia and, if left untreated, irreversible neurological sequelae may develop.9 ,12 However, copper deficiency is often not considered in patients with anaemia and neutropenia and so the diagnosis may be missed.9 ,12 It is more common for the diagnosis to be made after the development of more severe and frequently irreversible neurological manifestations.
Of the 70 patients in this study, nine developed unexplained haematological signs and/or neurological symptoms characteristic of copper deficiency (four with anaemia and/or neutropenia alone, three with neurological symptoms alone and two with both). One female patient gave a history that strongly supported a diagnosis of zinc-induced copper deficiency and whose haematological abnormalities normalised after she decided to stop taking zinc. The remaining eight patients had clinical signs and symptoms in keeping with copper deficiency at a time when they were prescribed zinc. However, a definitive conclusion that these patients were suffering from zinc-induced copper deficiency cannot be made since the time of onset of symptoms and response to withdrawal of zinc were not known.
The recommended daily allowance of zinc is 5.5–9.5 mg/day for men and 4–7 mg/day for women. However, the minimum dose of solvazinc, the most commonly used treatment, contains 45 mg of elemental zinc, which exceeds the US tolerable limit of 40 mg/day.13 Ingestion of zinc in amounts exceeding physiological requirements stimulates the body's homeostatic mechanism for zinc thereby limiting its intestinal absorption.14 This is controlled by the enterocyte synthesis of metallothionein, which binds zinc and allows its subsequent excretion as enterocytes are sloughed off into the gut lumen. If zinc is ingested in high enough amounts, copper malabsorption may ensue since copper binds avidly to metallothionein.11 Over an extended time period of months or years, this may lead to copper deficiency.9 ,12 Zinc-induced copper deficiency has been frequently described in the literature as a result of long-term ingestion of supplements, either prescribed or purchased over the counter;, overuse of dental fixatives, some preparations of which contain large amounts of zinc; chronic feeding with some enteral feeds; and swallowing of zinc-containing coins.
Published cases of copper deficiency following long-term zinc supplementation are often associated with ingestion of several hundred milligrams of zinc per day.15 ,16 ,17 However, several cases caused by lower doses of 100–150 mg have been reported.9 ,18 ,19 In the present study, over 60% of patients were prescribed 135 mg/day or more and so were likely to be at potential risk of developing zinc-induced copper deficiency. It is unlikely that this clinical possibility was considered as only two patients (3%) had plasma copper measured and in only one case was copper deficiency recorded in the casenotes as a potential risk. These findings underline the lack of awareness of zinc-induced copper deficiency.
In this study, the principal reason for prescribing zinc in 43% of patients was for correction of zinc deficiency, based on a low plasma zinc concentration. A low plasma zinc concentration, however, does not necessarily indicate a diagnosis of zinc deficiency but instead may be secondary to the SIR or low concentrations of albumin to which zinc is predominantly bound. During the SIR, albumin and bound zinc is redistributed into the extravascular space as well as being sequestered into the liver and kidney. After elective surgery, plasma zinc concentrations fall quickly but return to normal without treatment within about a week as the SIR resolves.20 Similarly, conditions associated with low concentrations of albumin, which carries around 70% of circulating zinc,21 also result in low plasma zinc concentrations. This survey showed that in 76% of patients, low plasma zinc concentrations secondary to hypoalbuminaemia or the SIR were likely to be misinterpreted as indicative of zinc deficiency.
Zinc is an essential trace element and so clinicians may consider it a safe nutrient rather than a drug carrying potential risk. This study offers persuasive evidence of a potential risk of iatrogenic copper deficiency being unwittingly caused by prescribing high doses of zinc. Further work is needed to investigate the likelihood of this scenario. In a significant number of patients, zinc was prescribed after zinc deficiency had been erroneously diagnosed following misinterpretation of a low plasma zinc concentration. In adults the daily zinc requirement is less than 10 mg/day. There is no evidence to support the prescription of zinc at 135 mg/day, the dose most commonly prescribed. This study highlights the potential risk of developing zinc-induced copper deficiency as a result of such prescribing.
Take home messages
C reactive protein concentrations should be checked when interpreting plasma zinc concentrations to determine whether hypoalbuminaemia or systemic inflammatory response may be the cause rather than the deficiency.
When treating zinc deficiency it is important that doses below the upper tolerable intake limit of 45 mg/day are given to avoid copper deficiency.
It is wise to monitor plasma copper concentrations in patients who are prescribed zinc over extended time periods of around three months.
Handling editor Tahir Pillay
Contributors AD designed the study, carried out literature search, analysed and interpreted the data and wrote the initial manuscript. IM advised and modified the final manuscript. CY and NW retrieved clinical information from the patients’ casenotes and reviewed the final manuscript.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.