Biological monitoring of aluminium in renal patients

183

Clinica Chimicu Acra, 160 (1986) 183-188 Elsevier

CCA 03605

Biological monitoring of aluminium in renal patients Frederik Toxicology

Laborutory,

(Received Key words: Aluminium;

A. de Wolff Universiry Hospitul,

and Gijsbert

B. van der Voet

P.O. Box 9600. 2300 RC Leiden (The Netherlunds)

14 May 1986; revision received 1 July 1986: accepted Biological monitoring; Renulputients;

Huemodiu&is;

1 July 1986) ETthroc;vies; Normul o&e

Summary During the past decade, aluminium (Al) has been shown to possess a potential for systemic toxicity. Renal patients form a high-risk group for Al poisoning, and biological monitoring is indicated to diagnose and prevent toxicity. Electrothermal atomic absorption spectrometry is the analytical method of choice for measuring Al levels. Special precautions have to be taken to prevent contamination with environmental Al or adsorption of Al to glassware during analysis. Since this metal is almost evenly distributed between plasma and erythrocytes, either plasma or whole blood may be used to estimate exposure to Al. Measurement in dialysis fluid is suitable to assess parenteral uptake. Hair analysis is of no value. The normal value are mainly in plasma is 7.3 L- 2.0 pg/l (mean f SD, n = 10); toxic symptoms associated with levels above 100 pg/l. A dialysate level not exceeding 5 pg/l may be considered safe; intestinal absorption is then the only remaining, highly variable source of Al uptake.

Introduction Toxic effects of metals constitute an important field of interest in medicine for a long period. From the Stone Age onwards, man has developed industrial methods for the processing of metal ores. It became, therefore, inevitable to be exposed to metal compounds, an exposition which could lead to toxic symptoms when a certain dose was exceeded. In the past, the study of metal toxicity has mainly been limited to arsenic, lead, mercury and cadmium. In the last decade, however, many more elements have been demonstrated to cause disease in man. Aluminium (Al) is a metal in the latter group which was considered relatively non-toxic until its discovery in 1976 as the source of encephalopathy in patients on chronic intermittent haemodialysis (CIHD) [1,2]. Until that date, only very few observations pointed towards systemic toxicity of Al ]3,41. 0009-8981/86/$03.50

8 1986 Elsevier Science Publishers

B.V. (Biomedical

Division)

184

Although Al poisoning deals with the biological toxicity.

is most probably not limited monitoring in this population

to CIHD patients, this paper as a high-risk group for Al

Source of Al poisoning in CIHD patients Initially it was found that in the case of epidemic poisoning in the 1970s contamination of the dialysate was the main source of Al toxicity [1,2]. This problem could be overcome by using water previously purified by reversed osmosis. However, even in centres with a constantly good water quality there are patients who develop increased Al levels. This is due to intestinal absorption of Al from Al hydroxide which is administered to renal patients to bind phosphate in the diet to prevent hyperphosphataemia [5,6]. The capacity to absorb Al seems to be subject to major interindividual variation, as the wide practice of Al hydroxide administration leads to increased levels in only a part of the patient population. Therefore, monitoring of Al in CIHD and other renal patients is necessary for the diagnosis, prevention and treatment control of Al poisoning. In the following paragraphs, the methodology, choice of sample, frequence and interpretation of Al monitoring will be discussed. Methodology There are very few analytical methods available for measurement of Al conNeutron activation analysis is probably an centrations in biological matrices. acceptable method but inaccessible for hospital laboratories. Electrothermal (flameless) atomic absorption spectrometry (AAS) meets perfectly with the requirements of a clinical laboratory. The analysis of Al, however, should only be performed in a laboratory with considerable experience in trace element assays. The major difficulty in the analysis is that Al is a ubiquitous element, and since concentrations in human material are low even in poisoning, it is extremely difficult to avoid interference from contamination. For collection of blood and for serum separation Al-free tubes and pipettes should be used throughout, made from a material that does not absorb Al, which could lead to falsely decreased levels. In general, polypropylene tubes meet with these requirements, but every brand and batch should be checked regularly. Several methods for Al assay in biological material using flameless AAS have been described, the two most recent being d’Haese et al [7] and Van der Voet et al [S]. A synopsis of the latter procedure is presented in Table I. With this method, a linear calibration curve is obtained up to at least 100 pg/l. The detection limit is 1.3 pg/l for aqueous solutions, and 1.9, 1.8, and 2.3 pg/l for serum, plasma, and blood respectively. Within-run precision ranges between 2.8% for high levels in water and 4.2% for low levels in blood. Even when using a reliable procedure, each clinical laboratory performing Al assays should participate in an external quality control scheme, e.g. the Trace

185

TABLE

I

Synopsis 1. 2 3 4

of Al assay in blood, plasma

and serum

Dilute sample (10 yl or more) tenfold with 0.2% Triton X-100 in polypropylene Prepare calibration curve (O-100 pg Al per I) in T&on X-100 (0.2%) Inject 20 ~1 in a pyrolytically coated graphite tube Furnace program:

Step

a b c d e 5 Measure

tubes; vortex

Temperature

Ramp time

Hold time

(“C)

(se@

tsec)

Internal N, flow (mt/min)

110 150 1400 1400 2600

10 50 40 1 0

5 13 0 20 8

100 100 100 0 10

absorption

at 309.3 nm; spectral

bandwidth

0.7 nm

Element Quality Assessment Scheme organised by the Robens Institute at the University of Surrey in Guildford, UK. In cases of chronic exposition to potentially noxious substances, e.g. in occupational toxicology, it is advisable and customary to monitor not only parameters of the body burden (~iu~~g~c~i~oni~~~i~g}, but also the nature and concentration of the substance to which the subject is exposed (enuironmental monitoring). This is especially important in metal toxicology, because the toxicity of two different compounds of the same element can differ greatly. For biological monitoring, using AAS, the metal is measured in its elemental form only; for environmental monitoring it is usually necessary to identify the compound to which subjects are actually exposed. In the case of CIHD patients, environmental monitoring should be done in two ways. To measure intestinal exposure, the dose of the administered phosphate binder is registered; to determine parenteral exposure, Al concentrations in dialysate are measured. Ideally, the chemicat form of the metal in this medium should also be determined, but this is too complicated for clinical routine and adds little, if anything, to diagnosis and treatment. The primary purpose of Al determination in dialysate is to check the level of contamination, which should be kept as low as possible. The maximum acceptable Al concentration in dialysate is discussed below. In the bioiogical monitoring of toxic trace elements it is usual to measure amounts in whole blood. In the case of Al, concentrations in plasma (Alp) or serum (AlS) are determined rather than Al in blood (AIB). To evaluate this custom an investigation was made into the relationship between AlP and AlB; also studied was the distribution within and binding to erythrocytes [9]. In CIHD patients there is a linear relations~p between AlP and AIB, both before and after haemodialysis. This is demonstrated in Fig. 1. The slope amounts to 0.86,

186 r-m~p~ I_

AIP lug/L1 x102

-

7. PATIENTS

O--L) before dlalyslsin=131 .-. afterdtalys,sln.10)

(~10~)AIBlug/LI

Fig. 1. Relationship between ahrminium in blood (AlB) and in plasma (Alp) in renal patients intermittent haemodialysis. (From Ref. 8, with permission of the publisher.)

on chronic

which indicates that Al is distributed between plasma and blood cells with only small quantitative differences. The binding of Al to blood cells was studied in rat blood both in vitro and in vivo. Blood was sampled from 3 rats, one of which was injected 1 h earlier with 10 mg of Al/kg intraperitoneally. Blood from one untreated rat was incubated with Al (addition of approx. 500 pg Al/l of blood), and the blood of the third rat was analysed without pretreatment. Cells were spun down and washed repetitively in saline (9 g NaCl/l). The amount of Al in supernatant and cells was determined after

w

a

Number

of woshmgs

Fig. 2. Effect of the repeated washings in saline on the concentration of Al in blood cells (AlE) from a non-treated rat (0), a rat treated with 10 mg of Al/kg (0) and in rat blood cells that were incubated with 0.5 mg of Al/l for 1 h at 37’C (0). (From Ref. 9, with permission of the publisher.)

187

each centrifuge run. Three to 4 washings of the cells in saline resulted in 100% reduction of Al in the erythrocytes of the 3 blood samples (Fig. 2). The above results show that the concentrations of Al in plasma and blood are almost equal, and that the binding to erythrocytes is very weak, unlike other toxic metals such as lead, mercury and cadmium. This means that plasma and blood have similar value for biological monitoring of Al exposure. As has been well demonstrated elsewhere [&lo], Al assay in hair has no value for the diagnosis of Al intoxication in the individual patient. It can, therefore, be concluded that dialysate is the sample of choice for environmental monitoring of CIHD patients, and plasma or blood for biological monitoring of Al poisoning. Frequency and inte~re~tion

of the analysis

Every renal patient treated with either CIHD or oral Al hydroxide or both is potentially exposed to toxic amounts of Al and should, therefore, be monitored regularly. The proposed Guidelines of the Commission of the European Community [l&12] advise the measurements of Al in plasma once every three months. In most patients this frequency will be sufficient when the oral dose of Al hydroxide is constant, when there are no pathological symptoms attributable to Al toxicity, and when the concentration of Al is well below 100 pg/l. When Al hydroxide therapy is prescribed for the first time, when the dose is increased, or when toxic symptoms are noticed, Al should be monitored weekly. During chelation therapy with desferrioxamine, the frequency may be increased to study the Al balance. A plasma (serum, blood) concentration of 100 pg/l is generally considered to be the maximum level to be safe [11,12]. The normal value of Al in man has been a matter of controversy in the past. In Casarett and Doull’s handbook on toxicology [13], 170 pg/l is given as the normal value. This very high value is most probably due to analytical inconsistencies. In an earlier study of 91 psychiatric, somatically healthy inpatients [5] we found an Al value of 20.5 + 7.6 pg,/l (mean f SD); we scrupulously avoided Al contamination. Recently, in a group of 10 healthy volunteers we have established the normal value in our laboratory as 7.3 * 2.0 pg/l, which is in better agreement with others [14]. The discrepancy in normal values between psychiatric patients and healthy people is a subject of further research. The 100 ~g/l-border~me for toxicity seems to be arbitrarily chosen. In a group of 295 CIHD patients from IO different renal units we found that 72 of them (24.4%) had levels above 100 pg/l. One-third of these (8% of the total patient population) showed signs of Al toxicity [5]. We did not look for toxic symptoms in the low Al group. Although many patients with serum aluminium levels greater than 100 pg,/l show signs of toxicity, it is possible that toxicity may develop when levels are lower than this and that the generally accepted ‘safe concentration’ of 100 pg/l is too high. At the present stage of research it is, therefore, indicated to keep the Al level as close to the normal value as possible. To control ‘environmental’ exposure, Al in dialysate should be checked regularly in each centre. The frequency of the assay should be determined by local cir-

188

cumstances, but in any case Al should be determined after introduction of any change in the preparation of the dialysate (water treatment. chemicals). For the maximum allowable concentration of Al in dialysate, 30 pg/l has been suggested previously [ll]. However, this level is far too high. About 75% of Al is bound to plasma proteins [15]; this means that only 25% of the determined ‘total plasma AI’ is available for transport through the dialysis membrane [5]. From these data it can be derived that 30 pg/l Al in dialysate is in equilibrium with a toxic plasma level of 120 pg/l. If we assume that a plasma level of 20 pg/l is safe though slightly exceeding the normal range, then a corresponding safe dialysate level of 5 pg/l may be considered safe. The proposed Guidelines of the European Community advise that starting in 1988, the Al level in dialysis media should be kept below 10 pg/l

WI. References Flendrig JA, Kruis H, Das HA. Aluminium and dialysis dementia. Lance. 1976; i: 1235. Alfrey AC, LeGendre GR, Kaehny WB. The dialysis encephalopathy syndrome. Possible aluminium intoxication. N Engl J Med 1976; 294: 184-188. Spofforth J. Case of aluminium poisoning. Lancet 1921; i: 1301. MacLaughlin AIG, Kazantzis G, King E, Tedre D, Porter RJ, Owen R. Pulmonary fibrosis and encephalopathy associated with the inhalation of aluminium dust. Br J industr Med 1962; 19: 253-263. De Wolff FA. Toxicological aspects of aluminium poisoning in clinical nephrology. Clin Nephrol 1985; 24 (suppl) S9-S14. Cannata JB, Briggs JD, Junor BJR, Fell GS. Aluminium hydroxide intake: real risk of aluminium toxicity. Brit Med J 1983; 286: 1937-1938. d’Haese PC, Van de Vijver FL, De Wolff FA, De Broe ME. Measurement of aluminium in serum, blood, urine, and tissues of chronic hemodialyzed patients by use of electrothermal atomic absorption spectrometry. Clin Chem 1985; 31: 24-29. Van der Voet GB, De Haas EJM, De Wolff FA. Monitoring of aluminium in whole blood, plasma, serum, and water by a single procedure using flameless atomic absorption spectrophotometry. J Anal Tox 1985; 9: 97-100. Van der Voet GB, De Wolff FA. Distribution of aluminium between plasma and erythrocytes. Human Toxic01 1985; 4: 643-648. De Groot HJ, De Haas EJM, d’Haese P, Heyndrickx A, De Wolff FA. Determination of aluminium in serum and hair of patients on chronic intermittent haemodialysis. Pharm Weekbl Sci Ed 1984; 6: 11-15. Commissie der Europese Gemeenschappen. Voorstel voor een richtlijn van de Raad betreffende de bescherming van dialysepatienten door blootstelling aan aluminium zoveel mogelijk te beperken. Publicatieblad van de Europese Gemeenschappen 29-5-1983 Nr. C 202, 5-X. Commissie der Europese Gemeenschappen. Gewijzigd voorstel voor een richtlijn van de Raad betreffende de bescherming van dialysepatiEnten door de blootstelling an aluminium zoveel mogelijk te beperken. Publicatieblad van de Europese Gemeenschappen 20-6-1985 Nr. C 150, 6-11. Doull J, Klaassen CD, Amdur MO. Cassaret and Doull’s Toxicology: The basic science of poisons. Macmillan, New York 1980: 435. 14 Savory J, Brown S, Bertholf RL, Wills MR Quantitative determination of aluminium. In: Taylor A, ed.. Aluminium and other trace elements in renal disease. London: BailliPre Tyndall, 1986: 246-259. 15 King SW, Wills MR, Savory J. Serum binding of aluminium. Research Communications in Chemical Pathology and Pharmacology 1979; 26: 161-169.