Whole blood, plasma, and serum viscosity together with red cell deformability were measured before and during treatment of renal anaemia with recombinant human erythropoietin (EPO). Whole blood viscosity (WBV) progressively increased during the first 4 months of treatment in association with the rise in haemoglobin concentration. When the WBV was corrected to a standard haemoglobin concentration no change in blood viscosity was observed, neither was there any alteration in a derived index of red cell deformability, or in the plasma and serum viscosities. In addition, a direct measurement of red cell deformability using a filtration technique before EPO therapy was similar to that obtained in 30 healthy volunteers. There was no significant change in this parameter over the first 9 months of treatment. The rheological changes which occur with correction of anaemia with EPO can be explained solely by the increase in circulating haemoglobin mass rather than to any change in the properties of the plasma or red cells themselves.

An increased incidence of fistula thrombosis has been reported in haemodialysis patients treated with recombinant human erythropoietin (rHuEpo). The present study sought to investigate this problem by measuring fistula blood flow, blood viscosity, and a variety of tests of coagulation and haemostasis in a group of ten haemodialysis patients treated with rHuEpo. Fistula blood flow did not alter during the first 12 months of rHuEpo despite a significant increase in whole-blood viscosity. Bleeding time improved in all ten patients after 4 months of therapy, and this improvement was maintained at 12 months. There were no significant changes in one-stage prothrombin time, kaolin cephalin clotting time, whole-blood clotting time, prothrombin consumption index, plasma fibrinogen factor VII, factor VIII, antithrombin III, or platelet aggregability to ADP over the first 4 months of rHuEpo. In contrast, protein C decreased from 84.3 to 66.4% (P < 0.01) and protein S from 124.1 to 68.3% (P < 0.001) over the first 4 months. By 8 and 12 months, the concentrations of these substances had returned to pretreatment values. The levels of protein C and S attained at 4 months are known to predispose to thrombosis, and it is possible that this effect may contribute to the increased incidence of fistula thrombosis observed in haemodialysis patients treated with rHuEpo.

Summary The treatment of renal anaemia by recombinant human erythropoietin (EPO) is now well established. Several studies have examined the pharmacokinetics and efficacy of the drug given intravenously, intraperitoneally and subcutaneously and there is increasing evidence that the subcutaneous route has several advantages including the requirement for a lower dose. It is also important to stress the need for careful determination of baseline iron status of all patients before commencing EPO therapy. In the long term the extremely high iron stores of transfusion dependent patients will disappear. In the short term, however, the majority of the patients whose serum ferritin is less than 100 micrograms/l will require iron supplementation to allow an appropriate haemoglobin response. Alternatively, a fall in transferrin saturation to less than 20% is certainly an indication for iron supplementation and if oral iron therapy is not adequate then intravenous preparations may have to be considered. Although the anaemia of renal failure can be fully corrected by EPO, partial correction may be sufficient to reverse the problems of reduced exercise capacity, myocardial ischaemia and cardiomegaly which are frequently associated with end-stage renal disease. Partial correction will also result in a lesser rise in whole blood viscosity and, in turn, possibly reduce hypertension, thrombosis and increased peripheral resistance and thus lessen the side effects of EPO therapy.

Treatment with erythropoietin is highly effective and beneficial if given with care. In view of its cost, however, it is essential to exclude and treat other causes of anaemia before considering using this hormone. After treatment is started the important points for success are regular review of iron availability state combined with a slow correction of the anaemia. Failure of response requires a thorough search for a possible cause, which should be corrected before considering an increased dose of the hormone. Regular monitoring for potential complications, particularly a rise in blood pressure, is required.

The long-term cardiorespiratory effects of recombinant human erythropoietin treatment were investigated in ten haemodialysis patients by means of maximum exercise testing, lung function tests, echocardiography, chest X-ray, and rheological assessment over 12 months. There were significant rises in exercise time (mean [SD] 13.2 [5.5] to 20.0 [6.2] min), maximum oxygen consumption (19.1 [7.0] to 25.0 [6.7] ml.min-1.kg-1), and anaerobic threshold (11.7 [3.6] to 15.4 [4.8] ml.min-1.kg-1) after 2 months of erythropoietin treatment. The improvements were maintained but not augmented on repeat testing after 4, 8, and 12 months of therapy. Carbon monoxide transfer [corrected] rose from 15.5 (2.9) to 18.6 (3.7) ml.min-1.mm Hg-1. There was a substantial reduction in exercise-induced cardiac ischaemia (eight patients had significant ST segment depression before erythropoietin, only one after 2 months' treatment, and none after 12 months' treatment), despite a significant rise in whole blood viscosity. Left ventricular mass, as estimated by echocardiography, progressively decreased from 354 (169) g to 251 (95) g after 12 months' treatment, and four patients showed a reduction in cardiothoracic ratio on chest X-ray.

Data from 342 men who are participants in the Caerphilly Collaborative Heart Disease Study were used to replicate a previous report of a significant relationship between measures of whole-blood viscosity and hearing levels in persons with sensorineural hearing impairment. In the unselected data, there were significant relationships between measures of whole-blood viscosity at high shear rates and hearing threshold levels at 2000 and 4000 Hz, even after accounting for the effects of age and socioeconomic group. In a subset of the data containing 124 persons selected on the basis of likely sensorineural hearing impairment, there were significant relationships between whole-blood viscosity and hearing level at all frequencies, with stronger effects at the higher frequencies. The data support the contention of a potentially important relationship between whole-blood viscosity and sensorineural hearing impairment.

human immunodeficiency virus infection in haemophiliacs. Lancet 1986;ii: 1233-6. 12 Centers for Disease Control. Classification system for human T-lymphotropic virus type Illymphadenopathy-associated virus infections. Ann Intern Med 1986;105:234-7. 13 Centers for Disease Control. Revision of the case definition of acquired immunodeficiency syndrome for national reporting. MMWR 1985;34: 374-5. 14 Peto K, Pike MC, Armitage DR, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. II. Analysis and examples. BrJ Cancer 1977;35:1-39. 15 Haseltine WA. Replication and pathogenesis of the AIDS virus. Journal of AIDS 1988;1:217-40. 16 Pedersen C, Nielsen JO, Dickmeiss E, Jordal R. Early progression to the acquired immunodeficiency syndrome following primary human immunodeficiency virus infection. AIDS (in press). 17 Scott GB, Buck BE, Leterman JG, Bloom FL, Parks WP. Acquired immunodeficiency syndrome in infants. N EnglJ Med 1984:310:76-8 1. 18 Anderson KC, Gorgone BC, Marlink RG, et al. Transfusion-acquired human immunodeficiency virus infection among immunocompromised persons. Ann Intern Med 1986;105:519-27. 19 Cooper DA, Gold J, Maclean P, etal. Acute AIDS retrovirus infection. Lancet 1985;i:537-40. 20 Ho DD, Sarngadharan MG, Resnick L, Dimaszo-Veronese R, Rota TR, Hirsch MS. Primary human T-lymphotropic virus type III infection. Ann Intern Med 1985;103:880-3. 21 Tindall B, Barker S, Donovan B, et al. Characterization of the acute clinical illness associated with human immunodeficiency virus infection. Arch Intern Med 1988;148:945-9. 22 Weber JN, Rogers LA, Scott K, et al. Three-year prospective study ofHTLV III infection in homosexual men. Lancet 1986;i: 1179-82. 23 Boyko WJ, Schenchter MT, Craib KJP, et al. The Vancouver lymphadenopathy-AIDS study: 7. Clinical and laboratory features of 87 cases of primary HIV infection. CMAJ 1987;137:109-13. 24 Lange JMA, Parry JV, de Wolf F, Mortimer PP, Goudsmit J. Diagnostic value of specific IgM antibodies in primary HIV infection. AIDS 1988;2:31-5. 25 Pedersen C, Gerstoft J, Lindhardt B0, Sindrup J. Candida esophagitis associated with acute human immunodeficiency virus infection. J Infect Dis 1987;156:529-30. 26 Rouse BT, Horohov DW. Immunosuppression in viral infections. Rev Infect Dis 1986;8:850-73.

Fifteen severely anaemic patients receiving CAPD were treated with subcutaneous recombinant human erythropoietin (Epo). Ten subjects had a good response with the haemoglobin concentration increasing from less than 8 g/dl to greater than 10 g/dl within 16 weeks. Four patients had a poor response, which was due to infection in two, myelofibrosis in one and unknown cause in another. Epo was ineffective in the remaining individual, probably due to the presence of occult metastatic carcinoma. Iron supplementation in the form of intravenous iron dextran was given to 12 patients when transferrin saturation decreased below 20%. There was a significant increase in red cell volume (P less than 0.005), with a decrease in plasma volume (P less than 0.005). Red cell iron turnover increased (P less than 0.05) but there was no change in red cell lifespan or deformability. Biochemical parameters remained unaltered, as did peritoneal function. Exercise duration improved (P less than 0.001), as did maximal oxygen consumption (P less than 0.01). Four patients required an increased dose of hypotensive drugs. Epo is a safe effective therapy for the anaemia of CAPD subjects.

Figures collected over 2 1/2 years of screening of women attending the antenatal clinics of South Glamorgan, Wales, showed that the percentage of women identified as a risk for haemoglobinopathy trait (7.4%) was almost twice that estimated from the 1981 Census data (4.1%) and that the incidence of a thalassaemia trait in these women (0.38% beta; 0.68% alpha) was similar to that of an area of supposedly greater risk in London (0.46% beta; 0.42% alpha). Figures from one laboratory showed that without a scrutiny of all antenatal clinic blood counts for thalassaemic indices some thalassaemia traits will be missed. The apparent incidence of the HbS trait was 0.13%, and reasons for this being an underestimate are given.

Iron deficiency leads to a significant increase in whole blood viscosity in polycythaemic patients. This can be corrected by simultaneous treatment with iron and venesection. In patients with polycythaemia secondary to hypoxia the haemoglobin concentration may be increased by treating coexisting iron deficiency, without altering the whole blood viscosity, thus increasing oxygen availability to the tissues. Values for whole blood viscosity can be derived from the peripheral blood Hb and MCH, allowing therapeutic decisions to be made on the basis of a simple haematological examination.