Dr David J A Goldsmith Consultant Nephrologist and Horonary Senior Lecturer Renal Unit, Guy's Hospital London, United Kingdom

A 47-year old man was referred to the renal outpatients in January 1999 for investigation of abnormal urea and electrolytes.
He had been taking lithium carbonate for bipolar affective disorder since a presentation with hypomania aged 13. The daily lithium dose had ranged from 800 mg to 2.4 g/day, giving a cumulative lithium dosage of over 16 kg. During this time he had married, had children and maintained an unbroken employment record. His only documented psychiatric relapse was 10 years previously after the daily dose had been reduced to 800 mg. Though lithium blood levels were monitored carefully at least twice a year from 1970 to 1998. There was no clinical episode of toxicity, and blood lithium levels were never > 1.2 mmol/l (> 50% of the levels were < 0.6 mmol/l).
He clearly described polyuria, polydipsia and nocturia, which had first become apparent in 1977, some 8 years into lithium therapy where he was already drinking 20 pints of fluid per day.
In 1999 his regular medication was lithium 1.2 g od and atenolol 50 mg od. There was no family history of renal disease. Physical examination was unremarkable showing only that the patient was mildly overweight, the blood pressure was 150/85 mmHg and fundoscopy was normal.
The creatinine level in 1999 was 145 mmol/l, having been 128 mmol/l in 1988, and 126 mmol/l in 1995. Thyroid stimulating hormone was 4.6 mU/l (0.3-5.5).
Renal immunological investigations were negative, normal sized kidneys and no renal tract abnormalities were seen on ultrasound scan, 2 g of protein in a 24 hour urine collection (with a volume of around 6 litres) and an isotope Glomerular Filtration Rate (GFR) of 79.5 ml/min/ 1.73 m2.
The patient refused to countenance a switch to another psychotropic medication citing his concerns about relapse and subsequent socio-economic problems. His Lithium dose remained 1.2 g giving blood levels of 0.9 mmol/L. There was no excess in eosinophil numbers on the white blood count differential nor on the blood film.
When seen in January 2001 his creatinine was 175 mmol/l and the GFR had fallen to 58.4 mls/min/1.73m2. By August 2001 the creatinine had reached 230 mmol/l and the lithium level was 1.7 mmol/L. To take the matter further it was agreed to perform a renal biopsy (Oct 2001).

Questions

Figure 1:
The biopsy showed severe chronic tubulointerstitial damage, interstitial scars and tubular atrophy consistent with lithium toxicity plus active tubulointerstitial nephritis with a significant number of eosinophils in the interstitium. Focal and segmental glomerulosclerosis was also seen.

Figure 1