ETIOLOGICAL FACTORS IN THIAZIDE-INDUCED OR AGGRAVATED DIABETES MELLITUS

Selective ß1-Blockers Are Not Associated With New-onset Diabetes Mellitus in Hypertensive Patients

BACKGROUND

Although ß-blockers are known to increase new-onset diabetes mellitus (DM), previous evidence have been controversial. It has been suggested that newer vasodilatory ß-blockers yield better glycemic control than older nonselective agents. The aim of this study was to evaluate the diabetogenicity of currently used newer ß-blockers based on ß1 receptor selectivity in a series of Asian population.

METHODS

We investigated a total of 65,686 hypertensive patients without DM from 2004 to 2014. Patients with hemoglobin (Hb) A1c ≤6.0%, fasting blood glucose ≤110 mg/dL, and no history of diabetes or diabetic treatment were enrolled for analysis. Patients were divided into the ß-blockers group and non-ß-blockers group. Propensity score matching (PSM) analysis using a logistic regression model was performed to adjust for potential confounders. The primary end point was the cumulative incidence of new-onset DM, defined as a fasting blood glucose ≥126 mg/dL or HbA1c ≥6.5%, and major adverse cardiac and cerebral events (MACCE), defined as a composite of total death, nonfatal myocardial infarction, and cerebrovascular accidents. We investigated predictors of new-onset DM and MACCE based on 2 models, including clinical risk factors and co-medications, respectively.

RESULTS

Mean follow-up duration was 30.91 ± 23.14 months in the entire group before adjustment. The ß-blockers group had a significantly higher incidence of new-onset DM and MACCE than the non-ß-blockers group. After PSM, analysis of a total of 2284 patients (1142 pairs, C-statistic = 0.752) showed no difference between the 2 groups in new-onset DM or MACCE. In multivariate analysis after PSM, baseline HbA1c, stroke, heart failure, nonselective ß-blockers, and age were independent predictors of new-onset DM. Selective ß1-blockers did not increase new-onset DM after adjustment for other antihypertensive medication and statins.

CONCLUSIONS

In the era of newer ß-blockers, selective ß1-blockers were not associated with new-onset DM. More evidence is needed to verify this relationship and the underlying mechanisms.

Is diabetes always diabetes?

Diabetes is always taken to be a life-long diagnosis. In order to re-examine this question, 75 g glucose tolerance tests (OGTT) were performed twice on 37 previously confirmed diabetic patients (mean duration of diabetes of 4.6 years; range 1-15 years) with normal glycosylated haemoglobin levels on regular review. Weight loss since institution of a healthy diet was 7.6 +/- 4.8 kg (+/- SE). Normal glucose tolerance was found in 27% of patients and impaired glucose tolerance in 21% with no significant change on rechallenge. HbA1 was 6.3 +/- 1.5% (+/- SD) (normal < 7.5%) in patients with normal glucose tolerance compared to 7.0 +/- 0.9% (+/- SD) in those with impaired glucose tolerance, P < 0.05. The response of the OGTT in these patients varied with dietary intake and weight. Such individuals could be regarded as having perfectly controlled diabetes or alternatively to have been cured. The definition of diabetes should be reviewed to allow people to escape the diagnosis where permanent change in dietary habits is established.

Diuretics and other antihypertensive drugs and glycemic control in non-insulin-dependent diabetics with hypertension. A survey by the Fukuoka Diabetes Clinic Group

We compared the levels of fasting plasma glucose (FPG) and glycosylated hemoglobin (HbA1) in non-insulin-dependent diabetics with hypertension who were treated with monotherapy of diuretics or other antihypertensive drugs (AHD) and in those who were not given any AHD. Total 162 patients were divided into the four groups; the control group without AHD (n = 45), the diuretics group (n = 42), the beta-blocker group (n = 30) and the Ca-antagonist group (n = 45). FPG and HbA1 levels were 153 +/- 44 mg/dl and 10.0 +/- 2.3% for the control group, 145 +/- 55 mg/dl and 9.8 +/- 2.2% for the diuretics, 165 +/- 63 mg/dl and 10.2 +/- 2.1% for the beta-blocker and 158 +/- 42 mg/dl and 10.4 +/- 2.0% for the Ca-antagonist, respectively. There were no significant differences in the levels of FPG and HbA1 among the four groups. Multivariate analysis also revealed no difference in glycemic control even when anti-diabetic treatment (diet alone, oral hypoglycemic agents or insulin), body mass index, serum potassium, systolic and diastolic blood pressures and age were taken into account. Blood pressure levels did not differ among the groups except control and they were well controlled at the low doses of AHD. Our results suggest that the choice of low dose diuretics for the treatment of hypertension in non-insulin-dependent diabetics might not be necessarily excluded by the only reason of the possible deleterious influence on glycemic control.

Glucose tolerance during antihypertensive therapy in patients with diabetes mellitus

Many antihypertensive drugs have adverse effects on glycemic control when they are used in diabetic patients. This was noted for thiazide diuretics in 1960, and the mechanism of the effects remains uncertain. Indirect evidence suggests that changes in the serum potassium are at least contributory, although the principal mechanism of thiazide-induced hyperglycemia is probably a reduction in the insulin response to glucose. Beta blockers also adversely affect blood sugar control but only by a small margin. The main cause for concern with beta blockers, however, is their effect during hypoglycemia in which nonselective agents delay blood sugar recovery. In diabetic patients, the institution of antihypertensive therapy should be followed by a reassessment to note any changes in sugar, potassium, and lipids, or side effects.

Effects of amiloride on arterial pressure and renal function

Amiloride was administered to 13 male patients with mild essential hypertension and normal renal function. It effectively reduced supine and standing arterial pressures. The antihypertensive response was associated with a significant decrease in renal inulin clearance and a rise in plasma creatinine concentration, although both values at the end of seven weeks of therapy remained within normal limits. Serum potassium concentrations rose but did not reach hyperkalemic levels. Amiloride did not increase blood levels of glucose, uric acid, lipids, calcium, or urea nitrogen. We conclude that amiloride is an effective antihypertensive and antikaliuretic agents for patients with mild hypertension and normal renal function.

Changes in the fasting blood sugar after hydrochlorothiazide and potassium supplementation

A 20% or greater rise in the fasting blood sugar occurred in 11 of 24 patients treated with 50 mg hydrochlorothiazide twice daily for six to ten weeks. The blood sugar decreased when potassium was given in conjunction with the diuretic, but the mean fasting blood sugar remained higher than pretreatment levels. There was no correlation between the percent change in the blood sugar and the per cent change in the serum K or exchangeable 42K.

Insulin, glucagon, aminoacid imbalance, and hepatic encephalopathy

Hepatic encephalopathy (H.E.) is associated with and perhaps caused by changes in plasma-aminoacid patterns--decreased branched-chain aminoacids (B.C.A.A.) and increased aromatic aminoacids (A.A.A.). The decreased B.C.A.A. may be in part secondary to hyperinsulinaemia, but the B.C.A.A. are catabolised by both fat and muscle. The increase in A.A.A. may reflect a "catabolic stimulus" reflected in hyperglucagonaemia, particularly in severe hepatic failure and H.E., and a decreased insulin/glucagon ratio. Endogenous protein, lean body-mass, or liver then releases large amounts of A.A. and the A.A.A. cannot be catabolised by the failing liver, and thus accumulate in the circulation. With decreased plasma-B.C.A.A., the molar ratio of B.C.A.A. and A.A.A. decreases allowing the toxic A.A.A. to penetrate the blood-brain barrier in increased amounts and encephalopathy develops. Appropriate therapy for H.E. must include reversal of the "catabolic state" by providing sufficient B.C.A.A. and calories to decrease the flux of A.A.A. from muscle and liver, and the restoration of the normal molar ratio of B.C.A.A. and A.A.A.

Effect of equivalent antihypertensive doses of mefruside and cyclopenthiazide on serum electrolytes, uric acid and glucose tolerance in hypertensive patients

1. A comparative cross-over trial of mefruside and cyclopenthiazide, each drug being given for 6 weeks, was conducted on thirty hypertensive patients with diabetes mellitus or impaired glucose tolerance. Other antihypertensive therapy, any antidiabetic therapy and potassium supplementation were kept constant throughout the trial. Dosages of mefruside and cyclopenthiazide were adjusted to give approximately equal blood pressure levels in the two drug periods. 2. There was no significant difference in the following parameters studied during the sixth week of each of the two periods of drug therapy: serum electrolytes, total CO2, chloride, urea, amylase, haemoglobin, erythrocyte sedimentation rate, platelet and white blood cell counts, lying and standing blood pressure and pulse rate, weight, fasting and 2-h glucose and insulin levels and five-value glucose and insulin curve areas, fasting calcium and phosphate. 3. Serum creatinine and uric acid showed a small but significant fall during mefruside therapy.