Decreased dextrose tolerance or delayed removal of sugar from the blood in patients with arthritis was first noted by Pemberton and Foster in 1920.1 This finding has since been corroborated by some workers and denied by others, resulting in confusion of opinion. With the development of a satisfactory method for the determination of sugar requiring only 0.025 cc. of blood and easily applicable to arterial blood by Jeghers and Myers,2 it seemed desirable to study the problem anew and to see if the relation of the sugar levels of the arterial and venous blood would shed new light on the subject.
No attempt will be made to review the literature exhaustively. In Pemberton and Foster's1 series of 400 cases in the Army, the average level in 18 patients with severe arthritis while fasting was 0.113 per cent, which during the dextrose tolerance test for the same series