Hypomagnesemia has been reported to occur in diabetes mellitus in the course of recovery from ketoacidoses (Martin and Wertman 1947), as well as during insulin maintenance therapy (Jackson and Meier 1968). Moreover, the hypomagnesemia in patients with diabetic retinopathy has also been reported in the recent study (McNair, Christensen, Madsbad, Lauritzen, Faber, Binder and Transbol 1978). However, it is well known that the serum magnesium level does not accurately reflect the true state of total body magnesium (Warren and Parisi 1968;Mather and Levin 1979). In the present study, magnesium (Mg) levels in plasma, erythrocyte and urine were determined in patients with diabetes mellitus. These were then investigated in relation to the degree of diabetic retinopathy. Patients and Methods109 diabetics aged 17 to 77 (mean 48.5 years) were selected and were divided into three groups on the basis of retinal finding: (1) diabetics with normal fundi (Group I, n = 63), (2) those with background diabetic retinopathy (Group II, n = 33), and (3) those with proliferative diabetic retinopathy (Group III, n = 13). None of the patients had marked renal damage or had taken magnesium drugs or hypotensive diuretics. The control group consisted of 33 healthy subjects ranging from 15 to 80 years of age (mean 51.8 years). Venous blood samples were taken into heparinized tubes. Lysed blood samples for erythrocyte measurement were prepared by adding 3 ml of deionized water to 2 ml of blood. 5 ml of 0.75 % EDTA was then added to 0.2 ml of this mixture. After standing for 30 minutes, the solution was centrifuged and a supernatant was used for the assay (Lint, Jacob, Dong and Khoo 1969). Mg concentration were measured by atomic absorption spectrophotometry, using the Hitachi 170-50A. Erythrocyte Mg concentration was calculated from plasma and whole blood concentration, using the following formula: Erythrocyte Mg (mEq/1 packed cells) = A -B x (l-Ht/100) x 100/Ht. [A = Mg in whole blood (mEq/1); B = Mg in plasma (mEq/1)]. Statistical analysis was carried out applying the Student's t-test.

No abstract

Thyroid hormone abnormalities in serum were investigated in 47 patients with diabetes mellitus. Although no significant differences in T4 were found between normal subjects and diabetics, a group of diabetics whose fasting blood sugar levels were over 250 mg/dl showed significantly higher reverse T3 (rT3) (p less than 0.01) and lower T3 levels (p less than 0.05) than healthy controls. A significant correlation was observed between rT3 and FFA levels in diabetics. Moreover, patients in diabetic ketoacidosis showed markedly high rT3 with low T3 levels. With insulin treatment, these levels returned to normal in several days. These findings suggest that the reduction of T3 and the increase of rT3 may indicate an adaptation to limit catabolism in diabetics.

In order to ascertain the platelet sensitivity to prostacyclin (PGI2) in patients with diabetes mellitus, we determined the percentage inhibition of platelet aggregation and platelet ATP secretion following PGI2 addition in an in vitro system. The percentage inhibition of platelet aggregation caused by PGI2 in final concentration of 1.25, 2.5, or 5.0 ng/ml was significantly lower in diabetics than in healthy controls. That of platelet ATP secretion by 1.25 or 2.5 ng/ml of PGI2 was also significantly lower in diabetics. These data suggested that in patients with diabetes mellitus, the decreased sensitivity of platelets to PGI2 will bring about hypercoagulability and may become one of the risk factors of diabetic microangiopathy in cooperation with lowered vascular PGI2 generation.

To investigate the feasibility of using salivary urea nitrogen as an index of renal glomerular filtration rate, we developed and applied a new analytical system consisting of a urease-containing test strip and an automatic reflectance spectrometer. The concentrations of urea nitrogen so determined correlate well (r = 0.93) with concentrations in serum. These preliminary data suggest that our method can be used routinely as a simple and reliable means of detecting abnormalities of renal function.