It is well established that erythrocyte 2, 3-diphosphoglycerate (2, 3-DPG) decreases markedly in diabetic ketoacidosis and returns to a normal level following insulin treatment. However, the cause of the marked decrease remains unknown. Alterations in glycolysis in erythrocytes of patients with diabetic ketoacidosis were investigated by enzymatic analysis of glycolytic intermediates and adenine nucleotides, and the following results were obtained.
1. In erythrocytes from untreated patients with diabetic ketoacidosis, the cellular levels of fructose-6-phosphate were significantly increased (P<0.05) and those of fructose-1, 6-bisphosphate, glyceraldehyde-3-phosphate plus dihydroxyacetone phosphate, 2, 3-DPG and 3-phosphoglycerate were significantly decreased (P<0.05, 0.05, 0.001 and 0.01, respectively). The crossover point was shown to be at thephosphofructokinase (PFK) step.
2. This glycolytic pattern in erythrocytes was similar to that of a patient with hereditary partial deficiency of erythrocyte PFK, strongly suggesting that the fall in 2, 3-DPG level in diabetic ketoacidosis is due to inhibition of PFK activity.
3. Normal human erythrocytes were incubated in the presence of β-hydroxybutyrate, acetoacetate, palmitate or insulin or at various pH levels in the absence of these substances. Glycolytic flow and intracellular glycolytic intermediates were not affected by ketone bodies, palmitate or insulin. Marked decreases in glycolytic intermediates below the PFK step and increases in intermediates above the PFK step associated with a significant reduction in glycolytic flow, were observed only at low pH levels.
4. It was also demonstrated in experiments using partially purified enzyme preparations that the PFK activity was dependent on pH and not affected by ketone bodies.
It is concluded that the fall in 2, 3-DPG level in diabetic ketoacidosis is caused by inhibition of PFK activity, and that this inhibition is due maily to an increase in hydrogen ion concentration and not to changes in the β-hydroxybutyrate, acetoacetate, insulin or palmitate concentration in the blood.

View full abstract

We have demonstrated the usefulness of studying anti-a-component antibody for differentiating the sera of patients with insulin autoimmunity from those of insulin-treated patients. That is to say, anti-a-component antibody can be found in the sera of insulin-treated patients but not in the sera of patients with insulin autoimmunity. It is possible, however, that there may be no anti-a-component antibody in the sera of insulin-treated patients as well, since highly purified insulins allow the possibility of not developing anti-a-component antibody. We therefore tested the presence of anti-acomponent antibody in the sera of 36 patients treated with single peak insulin, 12 patients with insulin autoimmunity and one patient who had been treated only with Monocomponent Insulin (MC-Insulin) and developed anti-insulin antibody in his serum.
The percentage of ¹²⁵I-a-component bound with sera preincubated with cold MC-Insulin was 28.0±3.08 (mean±SD) in patients treated with single peak insulin and 20.5±2.74 in patients with insulin autoimmunity. The latter level was as high as in normal subjects, but the former level was higher than in normal subjects. In insulin-treated patients, the bound ¹²5I-a-component in sera preincubated with both cold MC-Insulin and cold a-component was significantly lower than in sera preincubated with MC Insulin alone. This difference was considered to be due to specific (not crossreacting with insulin) anti-a-component antibody. The 125I-a-component binding in serum from a patient who had been treated only with MC-Insulin and showed insulin antibody in his serum, was within the normal range during the first 4 yr. However, it increased somewhat above the upper limit of the normal range after 4 yr had past.
Even at the present time when single peak insulin is widely used, it appears that examination for anti-a-component antibody represents a useful technique for differentiating the sera of patients with insulin autoimmunity from those of insulin-treated patients.

View full abstract

In order to elucidate the role of glucagon in abnormal glucose tolerance after gastrectomy, we performed total gastrectomy in male Wistar rats using Billroth's 1st method, and the plasma glucagon levels after oral glucose loading in 15 gastrectomized rats were determined. As controls, 14 normal rats and 9 rats with a duodenal pouch (a catheter was inserted into the intraduodenum) were also investigated.
In the present experiments, after overnight fasting, the bilateral jugular veins were exposed under anesthesia with chloral hydrate (300mg/kg, intraperitoneally), and blood sampling was carried out with a heparinized syringe at intervals up to 90 min. All subjects were given 2g/kg of glucose (50% solution) orally through a polyethyelene catheter which was inserted into the stomach.
Blood glucose was determined by the orthotoluidine method and plasma insulin by polyethyleneglycol radioimmunoassay using rat insulin as standard. Plasma glucagon (IRG) was measured by the talcum method using antiserum 30 K, and plasma total glucagon (GLI) was determined by the same method using antiserum K 4023.
The blood glucose levels in the gastrectomized group after glucose loading were significantly elevated compared to those in noraml rats and the rats with a duodenal pouch.
The fasting plasma GLI level in the gastrectomized rats was 705±138.5 pg/ml, which was significantly higher than those in normal rats and the rats with a duodenal pouch (P<0.02 and P<0.05, respectively). The plasma GLI response to oral glucose loading in the gastrectomized rats was significantly higher than those in the other tow gourps (P<0.02). The fasting IRG level in the gastrectomized rats was also elevated compared to those in the other two groups. The plasma IRG levels following oral glucose administration rose abruptly to 785±233.0 pg/ml at 15 min after glucose loading. This level was also significantly higher than those in the other two groups (P<0.01).
No significant difference in plasma insulin response to glucose was observed among the three groups.
These results suggest that the IRG and GLI levels in gastrectomized rats may be elevated by a certain mechanism other than rapid passage through the intestine of glucose, and the elevated IRG may contribute to the glucose intolerance observed in gastrectomized rats.

View full abstract

The clinical findings in 27 patients with juvenile-onset diabetes and the effect of pancreatic β-cell function on diabetic stability, including its changes after initiation of insulin therapy, were studied.
All the patients were keto-acidotic at the onset and still require insulin. Their ages were 13.0±0.8 (mean±SEM) years. The degree of stability was quantified by the standard deviation (SD) of eight values of fasting blood sugar (FBS) and by the presence of acetonuria. The SD value was 50.7 mg/dl on the average. The stable group (A) consisted of patients with a SD value of less than 50.7 mg/dl and no acetonuria, and the unstable group (B) consisted of the other patients.
Clinical findings in both groups are as follows:(1) mean age was significantly greater in group A (15.0±1.1 years) than in group B (11.6±1.1 years, P<0.05);(2) age at the onset was significantly greater in group A (13.1±1.0 years) than in group B (6.6±1.1 years, P<0.001);(3) duration of insulin therapy was significantly shorter in group A (1.3±0.5 years) than in group B (5.0±0.8 years, P<0.001);(4) doses of insulin administered were significantlyless in group A (16.4±3.5 units/day) than in group B (35.5±7.2 units/day, P<0.05);(5) CPRvalue was significantly higher in group A (1.62±0.47 ng/ml) than in group B (0.49±0.05 ng/ml, P<0.05). The onset of diabetes peaked at two to three, seven to eight and 11 to 12 years of age.
Age at the onset of diabetes and the CPR value were inversely correlated with the SD value of FBS (P<0.05, respectively). The earlier the age at onset, the lower the CPR value was. Blood glucose was stable in patients with a CPR value of more than 1.16 ng/ml and unstable in those with a CPR value of less than 0.45 ng/ml, indicating that pancreatic β-cell function was important for diabetic stability.
Patients with a short duration of insulin therapy had a high amount of CPR, and those with more than two years of insulin therapy had a CPR value of less than 0.50 ng/ml. Thus, pancreatic β-cell function was thought to be retained for about two years after initiation of insulin therapy.

View full abstract

The lipid peroxide in the plasma of 103 diabetic patients and 331 control subjects was measured by a fluorometric method using thiobarbituric acid.
1) The plasma lipid peroxide values of diabetics (5.48 ±0.76 nmole/ml) were significantly (P<0.001) higher than those of controls (3.74 ±0.13 nmole/ml).
2) The plasma lipid peroxide levels were higher (P<0.001) in poorly controlled diabetics (6.39±1. 21 nmole/ml) than those in well controlled diabetics (4.08 ±0.38 nmole/ml). However, no significant correlation between plasma lipid peroxide values and glucose tolerance was found.
3) The plasma lipid peroxide levels had no relation with hypertriglyceridemia and hypercholesterolemia.
4) The plasma lipid peroxide values were significantly (P<0.001) higher in diabetics with angiopathy (7.69 ± 1.65 nmole/ml) than those without angiopathy (3.89 ±0.55 nmole/ml).
The above results suggest that lipid peroxide plays an important role in the occurrence and progression of angiopathy in diabetics.

View full abstract

This study attempts to clarify distribution of ¹⁴C-tolbutamide in rats after intravenous injection, using whole body autoradiographic procedure. The results are discussed and compared with those obtained by oral administration of ¹⁴C-tolbutamide. Results of oral administration disclosed the following: ¹⁴C radioactivity was mainly excreted in the urine and partially excreted into the gall bladder; the majority of radioactivity in bile juice was reabsorbed in the feces; concentration of ¹⁴C in bone marrow and lymph vessels was highest in the whole body during eight hours of observation; radioactivity in the lungs, liver, kidney, and thyroid gland was high; radioactivity in the heart, pancreas, adrenal glands, and spleen was also high but not as high as in the former group.
Results of intravenous injection revealed the following:
1) ¹⁴C radioactivity in the bile duct was high throughout the observation period. Although a large part of this radioactivity was reabsorbed, part of it was apparently excreted in to the feces. Immediately after injection of the drug, ¹⁴C radioactivity in the gastrointestinal tract, especially in the stomach and duodenum, began to increase, which suggested a high concentration of the drug in the blood.
2) Concentration of ¹⁴C in bone marrow and lymph vessels began to reach the highest level throughout the whole body ten minutes after injection. This level was maintained throughout the experimental period, and in the early stages it showed a higher concentration of ¹⁴C than did the orally administerd drug.
3) There was a high concentration of ¹⁴C in the lungs, liver, and kidneys. Radioactivity in the thyroid gland was not so high in the case of intravenous injection.
4) Radioactivity in the heart muscles and adrenal glands was markedly higher compared with that of the orally administered drug. However, ¹⁴C radioactivity in blood in the heart cavity was very low, as in the case of the orally administered drug.
It is apparent that distribution of the drug was markedly different in some aspects, depending upon the methods of administration. Extra-pancreatic effects and side effects of the drug are discussed.

View full abstract

Kidney sizes were studied in 36 normal and 71 diabetic subjects. The kidney size, “renal ratio”, was determined from measurements in the frontal plane using intravenous pyelograms according to the method of Simon. The ratio varied between 2.68 and 3.49 in normal subjects. The mean ± SD was 3.11 ± 0.22. The ratio was not correlated with age, height, body weight, glomerular filtration rate or renal plasma flow.
In diabetics, the mean renal ratio ± SD was 3.40 ± 0.29. This represented a significant increase compared with normal subjects (p<0.001). The increment in the ratio in diabetics was independent of age, duration of diabetes and degree of proteinuria. The ratio did not correlate with any renal function in the diabetics. However, the ratio in diabetics was positively correlated with the fasting blood sugar on admission: y (renal ratio) =O.001 x (blood glucose level in mg/dl) +3.19, r=0.312 (p<0.01).
The present study confirms that the renal ratio in diabetics was increased. It is considered that increment in the renal ratio provides a clue to the renal involvement in diabetes and that metabolic disturbance in diabetes may be responsible for such renal enlargement.

View full abstract

Since a new type of hypoglycemia associated with insulin auto-immune antibodies was reported by Hirata in 1970, 20 cases of this syndrome have been reported. in patients between 21 and 77 years of age, with the exception of one baby whose hypoglycemia was reportedly due to it's mother's insulin auto-immune antibodies conveyed through the placenta.
An 8-year-old girl, the youngest patient with insulin auto-immune syndrome, received a traffic injury on July 20, 1976, She complained of a headache, and diphenylhydantoin and other drugs were administered. From the beginning of September 1976, she showed hypoglycemic reactions, and in the early morning of September 24th, 1976, she lost consciousness. At that time, her fasting blood glucose was below 30 mg/dl, and she regained consciousness after an intravenous glucouse infusion. Diuranal blood glucose levels, revealed hypoglycemia but an oral glucose tolerance test showed a diabetic pattern. Insulin antibodies were found in the patient's serum although she had never received exogenous insulin. Factitious hypoglycemia was ruled out in this patient because she was so young, and she had never been diagnosed as being diabetic. No one in her family was diabetic so she was not familiar with insulin.
Using a specific precipitation method, insulin-binding immunoglobulins were found only in IgG, and light chains of the immunoglobulins binding with insulin were classified mainly as a kappa type. The patient's HLA antigens, especially B-locus, were determined as B 15 and BW 54.

View full abstract

Two cases of diabetes with gas gangrene caused by non-clostridial infection are reported, and the clinical findings are discussed.
Case 1. A 42-year-old woman was admitted to the hospital because of a poor genehadral condition due to high fever and severe local inflammation of the left leg and foot. She had had diabetes mellitus for ten years but had never been properly treated. Two months before admission she suffered a minor burn on the back of the left foot, which gradually worsened, and gangrene with severe inflammation developed. On admission, large, ulcerous, foulsmelling gangrene was noted on the back and sole of the foot. The lesions communicated each other subcutaneously, the surrounding areas were markedly swollen, and subcutaneous fluctuation and crepitation were observed. Results of laboratory tests revealed severe anemia, hyperglycemia, leukocytosis, and 6+ CRP. Despite intravenous administration of regular insulin and antibiotics, the phlegmonous infection spread rapidly through the left thigh and the abdominal wall to the chest wall. Staphylococcus aureus, Proteus vulgaris, and Corynebacterium were identified from a pus culture. An x-ray fi lm revealed subcutaneous gas accumulation. The patient died of severe bacterial infection.
Case 2. A 56-year-old man was admitted to the hospital. He had gangrene in the fourth toe of the left foot, and the surrounding area was inflamed. He had been diabetic for 20 years and had been taking oral hypoglycemic agents. On admission, he had fever and systemic signs of infection. An x-ray film revealed gas formation around the infected toe and the sole. Resection of the toes and necrotomy failed to improve the local infection, and cellulitis extended over the whole foot. Amputation below the knee joint was performed. Recovery was slow, but finally the patient was discharged. Staphylococcus aureus and Escherichia coli were detected from a pus culture, but an anaerobic culture was negative.
Twenty-one cases of diabetes with non-clostridial gas gangrene have been reported since 1893. Clinical findings reveal a gradual onset and slow progression. The lower extremities, already complicated with severe neuropathy, are affected. Delay in diagnosis is often fatal (mortality 76%). Early detection of subcutaneous gas accumulation appears to be important. In addition to the appropriate antibiotic treatment, prompt amputation of the infected area seems to be necessary.

View full abstract