As the population of Western developed countries becomes older, the well-being of elderly citizens is an emerging challenge. Hormonal replacement therapy is being prescribed ever more frequently for the elderly, but the clinical consequences of these treatments are not fully understood.
A 73-year-old man was hospitalized for evaluation of an 8-kg weight loss, sweats and redness of facial skin. His past medical history included asthma and hypertension for which he received fluticasone and losartan respectively. The symptoms developed 10 months after beginning daily transdermal testosterone therapy with 10 mg of androstanolone. This drug was prescribed within the context of a protocol to slow the natural ageing process, even though the patient had low–normal serum testosterone levels. One year before admission, the haemogram was normal and his pulse was 100 beats/min and regular. The patient had no fever and no hepatosplenomegaly, but he was hypertensive (170/95 mmHg). Results of blood analyses were white blood cells (WBCs) 6 × 109/l (62% granulocytes and 22% lymphocytes), red blood cells (RBCs) 5·870 × 1012/l, haemoglobin (Hb) 17·8 g/dl, haematocrit 56%, mean corpuscular volume 95 fl (normal range 85–95 fl), platelet count 176 × 109/l, serum B12 0·45 ng/l (normal range 250–750 ng/l), serum erythropoietin 1·07 U/l (normal range 1·9–13·7 U/l) and red cell volume 52 ml/kg. Histological examination of bone marrow aspirate and biopsy samples demonstrated hyperplasia of the three cell lines without dystrophy and fibrosis; bone marrow cells were cytogenetically normal. Abdominal computerized tomography scan detected no enlarged spleen or tumoral mass. The diagnosis was consistent with secondary polycythaemia and transdermal testosterone was stopped. Three months later, the patient had gained 4 kg, Hb was 14·6 g/dl, haematocrit was 45%, serum erythropoietin was 7·5 U/l and red cell volume had decreased (40 ml/kg); the last was completely normalized 9 months after testosterone withdrawal.
Serum androgen levels decline with age in normal men, such that most men over 60 years of age have mean serum total testosterone levels near the low end of the normal adult range. It is not known whether lower testosterone in older men has an effect on androgen-responsive organ systems, such as muscle, bone and prostate. Consequently, in Western countries, current studies are trying to evaluate the relative benefits and risks of testosterone supplementation in older men. Testosterone has been touted by some members of the alternative medicine community as a sexual potency enhancer, energy stimulator and all-round fountain of youth. However, haematocrits and RBC volumes have been shown to rise significantly during testosterone replacement therapy and the stimulatory effects of androgens on erythropoiesis are well documented ( Shahidi, 1973; Krauss et al, 1991 ; Tenover, 1992). Drinka et al (1995) recently reported two cases of polycythaemia in two men, one aged 66 and one aged 73 years old who had, respectively, received intramuscular testosterone (150 mg/70 kg every 2 weeks) for 14 and 17 weeks. The risk of haemoconcentration during testosterone therapy might be greater if the patient also has a condition considered as a risk factor for polycythaemia, such as sleep apnoea syndrome ( Drinka et al, 1995 ). Although elevated haematocrits have been reported for parenteral testosterone supplementation ( Tenover, 1992; Drinka et al, 1995 ), we are unaware of reports linking topical testosterone with the development of polycythaemia. We describe the first case in which the transdermal delivery of testosterone across non-scrotal skin of a male patient probably induced supraphysiological concentrations of testosterone. Androstanolone withdrawal led to the normalization of the red cell volume, suggesting that the drug was responsible for the appearance of polycythaemia. Testosterone is able to increase erythropoiesis by stimulating the kidney to produce erythropoietin but in our patient the serum erythropoietin level was low, suggesting an erythropoietin-independent mechanism of polycythaemia induction ( Shahidi, 1973). Thus, during testosterone replacement therapy, haematocrit monitoring should be standard practice in medical centres and older men should be tested periodically throughout therapy.