Historical Perspective
There is little doubt that in the experimental setting aluminium is toxic to hundreds of cellular processes both in man and animals. It is interesting, nevertheless, how much controversy surrounds its role in human disease. What is not controversial is that under certain conditions aluminium poisoning may occur and that its most alarming effects are as a result of potent neurotoxicity.
Aluminium is the third most abundant element in the earth's crust, of which it comprises up to 13%, existing naturally in complex compounds such as silicates and oxides. In the late 19th century the manufacture of metallic aluminium became a practical possibility, and since then it has been widely used in industry. In a paper by von Döllken in 1897, reference was made to the earliest publication (by Orfila in 1814) discussing the potential toxic effects of aluminium salts on animals (von Döllken, 1897). Behavioural changes and irritability were noted following parenteral administration, von Dölken also commented on work performed in 1887 by Siem in which detailed animal studies found quite profound neurological changes following the administration of aluminium. In the same article von Döllken confirmed these findings and one of his concluding comments was that the changes induced were clinically similar to human dementia. The earliest reference to aluminium in an English language medical journal was in the Lancet of 1913 where its successful use in the kitchen was discussed (Anonymous, 1913b; Anonymous, 1913a). The authors did not investigate potential toxicity.
Aluminium Toxicity in Subjects with Normal Kidneys Industrial exposure
The first description of the toxic effects of aluminium on man was not until 1921. Spofforth reported the case of a 46 year old metal worker who had been dipping red-hot metal articles into concentrated nitric acid using an aluminium holder and developed âloss of memory, tremor, jerking movements and impaired coordinationâ (Spofforth, 1921). Large amounts of aluminium were found in his urine, although no details were given of the methodology used. The report did not discuss the subsequent course of his illness. In 1957 Campbell and colleagues reviewed 503 references to âaluminium in the environment of manâ (Campbell et al., 1957). Except for cases of industrial poisoning due to inhalation of aluminium dust, they dismissed the potential toxicity of this element to man. In 1974 the same group reviewed the field once more, but again concluded that there was âstill no need for concern by the public or producers of aluminium or its products concerning hazards to human healthâ (Sorenson et al., 1974).
Since the case of encephalopathy reported by Spofforth in 1921 several other cases of industrial poisoning with aluminium have been reported, but only many years later. Pulmonary fibrosis and granulomata were found in aluminium workers but encephalopathy associated with the inhalation of aluminium dust was not reported again until 1962 (McLaughlin et al., 1962). A 49 year old man developed a rapidly progressive encephalopathy associated with epileptiform seizures. The earliest symptoms started three years before his death with an intermittent deficit of short-term memory and difficulty with speech. There followed increasing muscle twitching and speech difficulty. The electroencephalogram was abnormal with generalised slowing and spike activity. Other investigations were all normal except for mild anaemia; renal function had been normal and he died after the onset of bronchopneumonia following further deterioration of his cerebral state. No measurements of serum aluminium concentration were made during life or at post-mortem, but very high levels of aluminium were found in all the tissues examined, including the brain (480 ÎŒg/g dry weight compared with less than 0.6 ÎŒg/g in normals). The lungs showed severe fibrosis associated with areas of stainable aluminium. No comment was made on the bone histology. In 1985 came a further report of neurological disorders (incoordination, tremor, cognitive defects) in three patients who had worked for over 12 years in the same aluminium smelting plant (Longstreth et al., 1985), although no conclusive evidence for the role of aluminium was documented. A type of pneumonia has also been described (Herbert et al., 1982) and an increased incidence of carcinoma of the lung in Norway (Andersen et al., 1982). Accidents involving single subjects exposed to aluminium in medical applications have also occurred (Hantson et al., 1994; Renard et al., 1994). Rifat and colleagues reported a large study in which miners exposed to aluminium for therapeutic purposes (in the form of Mclntyre powder to prevent silicotic lung disease) developed impaired cognitive function which was worse the longer the exposure to the agent (Rifat et al., 1990). More recently White and colleagues described neuropsychological changes in aluminium smelting plant workers (White et al., 1992) confirming earlier worries about this type of exposure. In addition, one study reported an incident where source water had been contaminated by effluent from an aluminium die-casting plant (Kilburn & Warshaw, 1993). Contaminants included a variety of organic chemicals, and although it is quite possible that large amounts of aluminium may have been present also this was not discussed. Residents adjoining the plant did develop quite significant psychomotor difficulties including loss of short-term memory.
Aluminium accumulation in parenteral nutrition
Patients on long-term parenteral nutrition may develop bone pain and osteomalacia. Contamination of parenteral solutions by aluminium may be an important factor in the genesis of this condition, as, in spite of normal renal function, these patients may accumulate large amounts of aluminium in various tissues (Klein et al., 1982a; Klein et al., 1982b; Vargas et al., 1986). High levels of serum, urine and bone aluminium were reported, with deposition of aluminium at the mineralisation front of bone and osteomalacic changes. None had any evidence of encephalopathy. Children may be particularly susceptible, although the reason for this is not clear (Ott et al., 1983; Sedman et al., 1985; McGraw et al., 1986; Bishop et al., 1989).
Aluminium and Chronic Renal Failure
After the discovery that aluminium caused severe bone disease and often fatal encephalopathy in dialysis patients, the exposure of such patients to aluminium was greatly reduced and these conditions rapidly disappeared, except for the occasional sporadic case. Ackrill and colleagues had shown that desferrioxamine was an effective chelating agent which could be useful in the treatment of both aluminium induced brain (Ackrill et al., 1980) and bone disease (Ackrill et al., 1982; Ackrill & Day, 1985), although since then the significant toxicity of this chelating agent has meant that much smaller doses are now used and it may take many months to see resolution of disease (De Broe et al., 1993). Subsequently I became interested in the potential for persistent toxic effects from the relatively low levels of aluminium exposure, as experienced by the great majority of dialysis patients world-wide. Could such levels be regarded as 'safe' or were they still associated with insidious low grade toxic effects? Below I have reviewed the effects of aluminium on cerebral metabolism and function in such patients followed by a brief review of the known effects on bone and parathyroid metabolism, and haemopoiesis in renal failure patients.
Aluminium encephalopathy
The potential of aluminium as a major cause of neurological disease was not realised until approximately 15 years after the emergence, in the early 1960's, of haemodialysis as a successful treatment for patients with end stage renal disease. Neurological complications of renal failure such as uraemic encephalopathy, convulsions, neuromuscular irritability and peripheral nerve damage improve with effective dialysis treatment, or in the case of 'dialysis disequilibrium', occur during dialysis (Peterson & Swanson, 1964; Kiley & Hines, 1965; Jebsen et al., 1967; Tyler, 1968).
Alfrey and colleagues (1972) described a fatal encephalopathy of chronic haemodialysis patients quite distinct from previous reports, arising after 3â7 years' treatment. Five patients were reported whose first symptom was a speech abnormality. The speech became slow and deliberate with a stuttering quality, associated with difficulty in naming objects. There followed increasing tremor, muscle twitching, disturbance of movement, memory loss, concentration defects, personality changes and intermittent psychosis. These symptoms were often worse during or immediately after dialysis. The severe encephalopathy proceeded, in spite of a number of therapeutic manoeuvres, including intensive dialysis, to coma and death after 6â7 months. Only one patient survived, following successful renal transplantation. Laboratory findings were unhelpful: lumbar puncture was normal; electroencephalography was non-specifically abnormal; post mortem examination of the brain was normal or in a few cases focal oedema was noted. It was concluded that, in the absence of other causes of dementia, the syndrome was a...