HAeMODYNAMIC STUDIES IN HIGH OUTPUT BERI-BERI

Despite the general belief that the cardiac output is high in heart failure due to thiamine defi- ciency, the evidence for this is chiefly clinical and the number of measurements reported is remarkably few.Using the ethyl iodide method of Henderson and Haggard (1925), Hayasaka and Inawashiro (1928) found that the cardiac output averaged 5 9 litres per square metre per minute at the "acme" and 4-5 litres on recovery.This confirmed the earlier work of Odaira (1925).Hayasaka and Inawashiro also reported that during treadmill exercise, sufficient to treble the oxygen consumption the cardiac output appeared to rise more in patients with beri-beri than it did in normal subjects and in the patients themselves on recovery.At rest the oxygen unsaturation of antecubital venous blood averaged 5 59 volumes per cent at the "acme" and 8-51 volumes per cent on recovery.During treadmill exercise the antecubital arteriovenous oxygen difference did not alter in the "acme," but increased to 130 volumes per cent on recovery.Weiss and Wilkins (1937) also found a narrowed femoral arteriovenous oxygen difference at rest in thirteen alcoholic cases.They showed that the decholin circulation time was accelerated, and from the clinical evidence of Wencke- bach (1934) inferred that rice beri-beri and alcoholic beri-beri were functionally identical and characterized by generalized arteriolar dilatation.Porter and Downs (1942) used Grollman's acetylene method (1932) in two alcoholic patients given thiamine two and four days previously and the respective cardiac outputs were 4-71 and 9-80 I/min.Burwell and Dexter (1947) were the first to determine the cardiac output at catheterization by collection of mixed venous blood and the use of the direct Fick principle.Their patients had a cardiac output of 11 -7 litres per minute and manometric evidence of failure of both ventricles: the overall peripheral resistance was low.The findings a month later were normal.Later Lahey et al. (1953) reported one case and Iseri et al. (1954) three cases in which cardiac catheterization had been done.Lahey's patient had an output of 16 -litres with normal filling pressures, while Iseri's three patients had outputs of 14 4, 8'3 and 114 litres per min.respectively, high venous pressures, and low peripheral resistances: some also had high pulmonary vascular pressures.The total number of accurate measurements of cardiac output, therefore, amounts to five or possibly seven, if we include the two cases of Porter and Downs.Clearly the cardiac output in these cases was quite variable.In 1951 we realized that alcoholic beri-beri was not a rare disease in the principal cities of Australia.We have taken advantage of this situation to study the clinical features and the hemodynamics in 19 cases with cardiovascular manifestations.The clinical features are being reported elsewhere; this paper concerns itself with the haemodynamics and pathogenesis of the disease.All the patients were severe alcoholics; indeed, in this country cardio- vascular beri-beri does not occur with alcoholic intakes below 1500 calories a day kept up for at least three months.Many of the present patients drank much more, but none of them had cirrhosis of the liver.Strangely enough it is not the rule to find cirrhosis in patients with classical cardiovascular beri-beri.The average age of the patients was 37 years, range 24 to 54.This paper is concerned with classical high output beri-beri described so eloquently by Wencke- bach (1934) and his pupil Aalsmeer (Aalsmeer and Wenckebach, 1929).This disease responds