Yoshikuni Fujita

An Autopsy Case of Troglitazone-Induced Fulminant Hepatitis

OBJECTIVE To study an autopsy case of troglitazone-induced fulminant hepatitis. CASE REPORT A 58-year-old man contracted severe hepatitis 2 months after administration of troglitazone for the treatment of type 2 diabetes. Liver damage progressed gradually, even after withdrawal of the agent. Despite intensive therapy, the hepatitis became fulminant and the patient died 8 weeks after the onset of liver damage. Autopsy revealed massive hepatic necrosis and cholestasis with inflammatory cell infiltration. The pathological findings and the positive result of the drug-induced lymphocyte stimulation test for troglitazone indicated that the liver damage was mediated by hypersensitivity to troglitazone. CONCLUSIONS One report has attributed troglitazone-induced liver damage in less severe cases to idiosyncratic reactions. However, the present case indicates that troglitazone can induce hypersensitivity resulting in fulminant hepatitis. Careful monitoring of serum liver enzymes during troglitazone therapy is therefore essential.

Intestinal Behçet's disease--pathognomonic changes in intramucosal lymphoid tissues and effect of a "rest cure" on intestinal lesions.

To clarify the pathognomonic changes of intestinal lesions of Behçet’s disease and to determine effective therapeutic measures, we recruited 13 patients with the intestinal form of this disease for study. We performed pathology studies on the resected specimens of 7 patients and treated 5 of the other 6 patients with a low-residue diet. Pathology examination revealed that 6 of the 7 had inflammatory ulcerations in the ileocecal region. The ileal ulcers were mainly on the antimesenteric side. We observed remnants of Peyer’s patches at the margins of the major ulcerative lesions in 2 of 2 patients examined. There were aggregations of lymphocytes resembling destroyed lymph follicles in the superficial layer at the mouths of small fissuring lesions, and ulcer scars were also noted in Peyer’s patches in 4 of 5 other patients. X-ray and endoscopic examinations revealed the disappearance of intestinal lesions in 5 patients within 1 month during, or following the low-residue diet treatment. We found the intestinal lesions of Behçet’s disease at sites coinciding with intramucosal lymphoid tissue. The “rest cure” for the affected bowel was effective, i.e., there was significant alleviation of gastrointestinal symptomas and the intestinal lesions disappeared. We speculated that acute exudative inflammation, abscess formation, and consequent ulceration may occur in these tissues by the same mechanisms as those that operate in the positive needle-prick reactions seen in patients with Behçet’s disease.

Usefulness of anaerobic threshold in estimating intensity of exercise for diabetics

We examined the utility of the anaerobic threshold (AT) for quantifying the intensity of exercise that a diabetic patient is capable of handling. Thirteen diabetic patients treated with buformin exercised on a bicycle ergometer, and comparison was made with 20 healthy subjects matched for age and sex. The AT was determined from VO2 and VE with a personal computer. The intensity of exercise at the AT was 93 +/- 6 W in diabetic men and 80 +/- 10 W in diabetic women, values that were less than those of healthy subjects (P less than 0.05). There was a negative correlation between the intensity of exercise at the AT and the plasma concentration of buformin (P less than 0.01). There were no significant differences in either plasma lactic acid or pyruvic acid concentration at the AT between healthy subjects and diabetics. The plasma glucose at the AT or after exercise was lower than the baseline values in all subjects (P less than 0.01). The plasma insulin at the AT was lower than the baseline values in healthy subjects (P less than 0.01), but not in diabetics. There were no changes in plasma glucagon in any group. We concluded that determination of the AT is a simple, non-invasive procedure useful for ascertaining the optimal intensity of exercise for diabetics.

Relationship between age-adjusted heart rate and anaerobic threshold in estimating exercise intensity in diabetics

We compared heart rates at the anaerobic threshold (AT) with age-adjusted heart rates in two groups of diabetics (DMY, mean age 31; and DMO, mean age 48) to ascertain whether the AT is useful for evaluating the intensity of exercise therapy for diabetics. In both the DMY and DMO groups the AT was reached at lower heart rates than in control groups of healthy subjects, indicating that the metabolism of diabetics may become anaerobic if their age-adjusted heart rate is higher than their heart rate at the AT. In this study, 86% of the DMY and 50% of the DMO group had age-adjusted heart rates, at 70% of maximum, that were higher than the heart rates at the AT. Thus, exercise intensity with an age-adjusted heart rate at 70% of maximum may induce anaerobic metabolism in some diabetics. We also studied the relationship between exercise intensity and the glycosylated hemoglobin (HbA1c) level at the AT in order to understand why the heart rate at the AT in diabetics was less than in healthy subjects. Exercise intensity at the AT in both diabetic groups was less than in healthy groups, and so did not increase the heart rates. Both exercise intensity and the heart rate at the AT were inversely related to the HbA1c level, suggesting that HbA1c may be important for keeping low the exercise intensity and the heart rate at the AT. As far as possible, we excluded cardiovascular autonomic neuropathy, so that it could not explain the mechanism.(ABSTRACT TRUNCATED AT 250 WORDS)

Insulin Resistance in Mitochondrial Gene Mutation

From the Saiseikai Central Hospital (Y.S., K.H., Y.A., K.M.); Institute for Diabetes Care and Research, Asahi Life Foundation (H. Kadowaki); Third Department of Internal Medicine (H. Katagiri, T.K., Y.O., Y.Y.), Faculty of Medicine, University of Tokyo; and the Department of Biochemistry (M.S.), School of Medicine, Keio University, Tokyo, Japan. Address correspondence to Yoshihiko Suzuki, MD, at Saiseikai Central Hospital, 1-417, Mita, Minato-ku, Tokyo 108, Japan.

Renal structure as an indicator for development of albuminuria in normo- and microalbuminuric type 2 diabetic patients

Baseline glomerular structure in microalbuminuric (MA) and proteinuric Caucasian type 2 diabetic patients predicted progressive glomerular filtration rate decline while baseline urinary albumin excretion (UAE) did not. Little is known about whether or not renal structure at the early stages of diabetic nephropathy (DN) in type 2 diabetic patients can predict further functional development of DN. Baseline renal structure and function and follow-up data of renal function were examined in 17 type 2 diabetic patients (11 men, 45+/-7 (mean+/-S.D.) years old) with known diabetes duration 11+/-8 years without definable renal disease other than DN. Six patients showed normoalbuminuria (NA), 11 microalbuminuria (MA), and were followed up for 6.4+/-1.8 years after the baseline renal biopsy. Light and electron microscopic morphometric analyses provided quantitative glomerular and tubulointerstitial structural changes. No statistically significant difference was observed in hemoglobin A1c (HbA1c) values or mean blood pressure (MBP) between baseline and follow-up, even though the number of patients placed on antihypertensive drugs increased from 3 to 7. Follow-up UAE was not significantly different from the baseline UAE although 13 of 17 cases showed an increase. Baseline UAE did not correlate with the follow-up UAE or morphometric measures. Glomerular basement membrane width and volume fraction of the mesangium and mesangial matrix positively correlated with follow-up UAE. In NA and MA Japanese type 2 diabetic patients, baseline renal structural measures are more reliable indicators for the development of UAE than baseline UAE.

Anaerobic threshold can provoke microalbuminuria in non-insulin-dependent diabetics

We examined two groups of non-insulin-dependent diabetic men (group A, 13 patients without microalbuminuria; group B, 9 patients with intermittent microalbuminuria) to ascertain whether the anaerobic threshold (AT) can provoke microalbuminuria, comparing them with 12 healthy subjects matched for age and sex (group C). All subjects exercised on a bicycle ergometer until the AT was reached. In intermittent microalbuminuria, the albumin:creatinine ratio (ACR) was over 0.25 mg/mmol.Cr 1-3-fold in 5 measurements. The ACR after exercise was increased to over 0.25 mg/mmol.Cr in 4/9 cases in group B (P < 0.05), in 2/13 cases in group A, but not at all in group C. We also studied the mechanism of exercise-induced microalbuminuria. In group B, ACR before exercise correlated positively with the baseline plasma glucose. Furthermore, positive correlation was found between ACR after exercise and HbA1c in group B. The AT did not affect the urinary beta 2-microglobulin in any groups. The plasma atrial natriuretic factor (ANF) after exercise was elevated most prominently in group B (P < 0.05). Positive correlation was found between increments of ACR and increments of plasma ANF after exercise in group B. We conclude that the AT can provoke microalbuminuria in some non-insulin-dependent diabetics. The plasma ANF and metabolic control may play an important role in the pathophysiology of exercise-induced microalbuminuria.

Levels of albuminuria and risk of developing macroalbuminuria in type 2 diabetes: historical cohort study

Although increased urinary albumin excretion may increase the risk of adverse renal outcomes in patients with diabetes, it remains unclear whether microalbuminuria is associated with a higher incidence of macroalbuminuria in the absence of non-diabetic kidney events that frequently develop during the long-term course of type 2 diabetes. This historical cohort study included patients with type 2 diabetes, spot urine albumin:creatinine ratio (ACR) <300 mg/gCr and normal serum creatinine concentrations treated between August 1988 and April 2015. Patients with any evidence suggesting non-diabetic kidney diseases at baseline were excluded. Over a median follow-up of 50 months, 70 of the 1760 included patients developed macroalbuminuria. Twenty-one of these patients were diagnosed with non-diabetic renal events. The five-year cumulative incidence of macroalbuminuria in patients with ACRs of 0–7.5 mg/gCr, 7.5–30 mg/gCr, 30–150 mg/gCr, and 150–300 mg/gCr were 0%, 0.53%, 3.5%, and 36.0%, respectively, with significant differences between each pair of ACR categories. In type 2 diabetes, higher urinary ACR, even within a level of normoalbuminuria, was associated with a greater incidence of macroalbuminuria when non-diabetic renal events were excluded. These results conflict with findings suggesting that microalbuminuria is a poor indicator for the progression of diabetic nephropathy.

An ATP-sensitive potassium channel blocker suppresses sodium-induced hypertension through increased secretion of urinary kallikrein

It is suggested that an ATP-sensitive potassium channel blocker suppresses sodium-induced hypertension through increased secretion of urinary kallikrein. We reported that glibenclamide, an ATP-sensitive potassium channel blocker, accelerated dose-dependent secretion of renal kallikrein in sliced kidney cortex and in vivo in rats. In vehicle-treated normal Brown- Norway-Kitasato (nBN-Ki) rats, the administration of glibenclamide increased urinary kallikrein secretion, but changed neither the systolic blood pressure nor the urinary sodium on low (0.3%) NaCl diets. Although on high (8%) NaCl diets, the systolic blood pressure of the nBN-Ki rats administrated glibenclamide was significantly lower (P<0.05). The urinary levels of kallikrein and sodium of the nBN-Ki rats administrated glibenclamide were significantly increased (P<0.05, glibenclamide vs. vehicle). A similar result was obtained with a kidney-selective ATP-sensitive potassium blocker, N,N′-dicyclohexyl-4-morpholinecarboxamidine (U18177), in SD rats. Mutant kininogen-deficient Brown-Norway Katholiek (muBN-Ka) rats fed high (8%) NaCl diets showed an increase in urinary kallikrein levels, but showed neither hypotensive nor natriuretic actions by glibenclamide. A bradykinin B2 receptor antagonist, 8-[3-[N-(E)-3-(6-acetamidopyridin-3-yl) acryloylglyycyl]-N-methylamino]-2,6-dichlorobenzyloxy]-2-methylquinoline (FR173657), which was administrated to SD rats, together with glibenclamide, abolished the hypotensive and natriuretic effects of glibenclamide in high-sodium (8%NaCl) hypertension, despite an accelerated secretion of urinary kallikrein. Therefore, these results indicate that glibenclamide, an ATP-sensitive potassium channel blocker suppressed sodium-induced hypertension through sodium excretion from the kidney resulting from accelerated secretion of urinary kallikrein.

Detailed glomerular ultrastructure in Japanese type 2 diabetic patients by the quick-freezing and deep-etching method

Mesangial expansion and glomerular basement membrane (GBM) thickening did not correlate with urinary albumin excretion (UAE) in type 2 diabetic patients in our previous studies; therefore, it was necessary to elucidate more detailed ultrastructural changes in the early stages of diabetic nephropathy (DN) in type 2 diabetic patients. The quick-freezing and deep-etching (QF-DE) method allows us to examine three-dimensional ultrastructures of human renal glomeruli in vivo at high resolution. The QF-DE method was applied to six type 2 diabetic patients without definable renal diseases other than DN. Four patients were normoalbuminuric (NA) and the other two were microalbuminuria (MA). Three control specimens were the normal parts from nephrectomies due to renal cell carcinomas. Electron microscopic morphometric analyses provided quantitative glomerular structural changes. Replica membranes were prepared by the QF-DE method, and diameters of mesh structures at the GBM and mesangial matrix (MM) were measured on electron micrographs as previously described. By the QF-DE method, both the GBM middle layer and MM were composed of polygonal meshwork structures. The mesh pores of the GBM and MM were more enlarged and irregular in shape in NA diabetic patients than those of the controls, and these ultrastructural changes became more obvious in MA patients. The mesh diameters of the GBM and MM in the diabetic patients were also larger than those of the controls. Such a mesh diameter of the GBM was well correlated with the amount of UAE, while the mesh diameter of MM showed a slight correlation with UAE. Although there were small number of subjects in the present study, the detailed ultrastructural changes in NA and MA type 2 diabetic patients, which had not been disclosed by conventional electron microscopy, were revealed by the QF-DE method. Increased mesh diameters of GBM might be related with the increase of UAE.