D.A. Pyke

ISLET-CELL ANTIBODIES IN DIABETES MELLITUS

Islet-cell antibodies (I.C.A.) were found in 38% (319/829) of insulin-dependent diabetic patients, in 5% (6/112) of insulin-independent diabetics, and in 1.7% (3/177) of non-diabetic subjects. In the insulin-dependent group I.C.A. were found in 85% of patients immediately after the onset of symptoms and they became less common as the duration of disease increased I.C.A. were equally common in both sexes and the decline in their prevalence was independent of age. The antibodies were directed against cytoplasmic components of islet cells but not against insulin itself. The appearance of I.C.A. probably follows cell damage occurring before the onset of symptoms. By contrast, thyroid and gastric autoantibodies were more common in older patients and females. There was no correlation between the presence of these antibodies and I.C.A. in patients with either diabetes of recent onset or longstanding disease.

HÆMOGLOBIN A1 IN DIABETIC PREGNANCY

Abstract Serum-haemoglobin A 1 (HbA 1 ) did not alter in normal pregnancy but in pregnant diabetic patients HbA 1 was raised in the first trimester, falling as the pregnancy progressed, presumably because of better diabetic control. Post partum, when the control was relaxed, HbA 1 increased. In diabetics at 36 weeks gestation HbA 1 correlated with the fasting blood-glucose 4 weeks previously. 3 out of 5 pregnant diabetics whose HbA 1 was greater on presentation than that of well-controlled diabetics gave birth to children with fatal congenital abnormalities. No abnormalities were detected in children born of mothers with an HbA 1 level within the range for well-controlled diabetes.

SENSITIVITY TO ENKEPHALIN AS A CAUSE OF NON-INSULIN DEPENDENT DIABETES

Non-insulin-dependent diabetes is associated with facial flushing after alcohol in patients on chlorpropamide (chlorpropamide alcohol flushing, C.P.A.F.) especially when there is a family history of diabetes. C.P.A.F. in three subjects (two diabetics, one non-diabetic) was blocked by the specific opiate antagonist naloxone. In nine subjects (six diabetics) C.P.A.F. was reproduced by the enkephalin analogue with opiate-like activity [D-Ala2, MePhe4, Met (O)-ol] enkephalin (DAMME). C.P.A.F. thus may be due to increased sensitivity to endogenous opiates. DAMME and other substances with opiate-like activity, such as morphine and beta-endorphin, affect carbohydrate metabolism and insulin secretion. Increased sensitivity to endogenous opiates such as enkephalin may thus give rise to non-insulin-dependent diabetes associated with C.P.A.F.

Growth in diabetic children. Studies in identical twins.

Abstract An inquiry to determine whether early-onset diabetes leads to retardation of growth and development is described. Healthy monozygotic twins are normally very similar in height and therefore provide ideal comparisons for the effect of disease on height. The adult height and age of menarche in 35 young-onset identicaltwin pairs, one or both of whom has diabetes, have been studied. In all but one of 12 pairs in which one twin developed diabetes before puberty and the other did not, the affected twin was shorter; the mean difference was 21/4 in., which is much more than would be expected by chance. In the pairs in which diabetes appeared after puberty in one or both twins there was no significant difference in height. When both twins developed diabetes before puberty, again there was no significant difference in height. In 2 pairs of girl twins in whom diabetes appeared before puberty in one but not in the other, menarche was 4 and 5 years later in the affected than in the unaffected twin, whereas in the other pairs the difference was only about 1 month. These differences were seen despite apparently satisfactory diabetic control. It is concluded that diabetes of early onset leads to retardation of growth and development, even when its treatment is adequate as judged by conventional criteria, that these criteria are inadequate, and that the control of insulin-dependent diabetes is almost always poor.

Increased proinsulin levels as an early indicator of B-cell dysfunction in non-diabetic twins of type 1 (insulin-dependent) diabetic patients.

SummaryGlucose tolerance and insulin secretion were studied in two groups of non-diabetic identical twins of recently-diagnosed Type 1 (insulin-dependent) diabetic patients: (1) a group of 5 twins with islet cell antibodies, and (2) a group of 6 twins without. Despite similar fasting glucose, insulin and C-peptide concentrations both groups of twins had significantly higher fasting proinsulin concentrations than the control group (p<0.05). The twins with complement-fixing islet cell antibodies had reduced glucose tolerance and clearance, whilst the twins without islet cell antibodies did not. Neither group of twins showed any abnormality in insulin, C-peptide or proinsulin response to oral or intravenous glucose. We conclude that increased fasting proinsulin levels precede abnormalities of insulin secretion, and are an early indication of minor B-cell damage in these twins irrespective of their risk of developing diabetes.

PATHOGENESIS OF INSULIN-DEPENDENT DIABETES: A ROLE FOR ACTIVATED T LYMPHOCYTES

Expression of HLA DR antigens on T lymphocytes indicates that these cells are actively involved in an immune response. Raised levels of activated T lymphocytes were found in 14 of 15 recently diagnosed but in only 7 of 28 long-standing insulin-dependent diabetics. 9 of the recently diagnosed patients retested 6 months later still had high levels of activated T lymphocytes. Even long-standing insulin-dependent diabetics had significantly higher levels of activated T lymphocytes than non-insulin-dependent diabetics and healthy controls. 5 of 7 unaffected co-twins of recently diagnosed insulin-dependent diabetics had high levels of activated T cells; this increase persisted in 2 retested 6 months later, when mildly impaired glucose tolerance had developed. These results confirm that there is an active cellular immune reaction in newly diagnosed insulin-dependent diabetics which may precede the disease and persist for at least 6 months after its appearance.

HISTOCOMPATIBILITY ANTIGENS IN DIABETIC IDENTICAL TWINS

84 pairs of diabetic identical twins were HL-A typed. 22 were maturity-onset diabetics (diagnosed after age 45); all were concordant and they showed no disturbance of HL-A frequencies. Of the remaining 62 pairs of juvenile-onset diabetics, 31 were discordant (only one twin diabetic) and 31 concordant (both twins diabetic). The frequency of the W15 antigen was equally increased in both the concordant and discordant pairs; HL-A8 was increased only in the concordant pairs. The fact that W15 is increased in the discordant pairs shows that there is genetic predisposition to diabetes even in these twins; the fact that HL-A8 is not increased suggests that different alleles may underlie susceptibility in the two groups of twins.

Evidence that the reduced number of natural killer cells in Type 1 (insulin-dependent) diabetes may be genetically determined

SummaryViruses may cause Type 1 (insulin-dependent) diabetes. We wondered whether the number and function of natural killer cells, which are important in anti-viral defense, are disturbed in diabetic patients. We studied 16 recently diagnosed Type 1 diabetic patients, 18 Type 1 diabetic patients diagnosed more than 15 years previously, 18 Type 2 (non-insulin-dependent) diabetic patients and 23 control subjects. We determined the number of natural killer cells (expressed as log10%) using anti-Leu 11 monoclonal antibody and the function (in log10 lytic units) concurrently using a 51Cr release assay with K 562 as target cells. We found that the number of natural killer cells was reduced in Type 1 diabetes (1.01±0.04) as compared with Type 2 diabetic patients (1.16±0.04, p=0.004) and normal control subjects (1.16±0.04, p=0.006). To establish whether the reduced natural killer cell number is genetically determined we studied 19 identical twin pairs discordant for Type 1 diabetes; we found that even the non-diabetic co-twins had a reduced natural killer cell number (0.93±0.05, p= 0.0006) as compared with normal control subjects. Natural killer cell function was similar in all groups while natural killer activity per cell was significantly increased in the recently diagnosed diabetic patients (1.63±0.07) as compared with long-standing diabetic patients (1.26±0.26, p= 0.03) and controls subjects (1.36±0.07, p= 0.006). In conclusion the reduced number of natural killer cells in Type 1 diabetes appears to be genetically determined while their activity at diagnosis is increased.

MANAGEMENT OF THE PREGNANT DIABETIC: HOME OR HOSPITAL, WITH OR WITHOUT GLUCOSE METERS?

Thirteen pregnant insulin-dependent diabetic patients were allocated to blood-glucose self-monitoring (meter) or conventionally treated (non-meter) groups at 30--31 weeks. A day-profile of blood-glucose and intermediary metabolite levels was obtained 2 weeks later at home and after routine admission to hospital at 35--36 weeks. Metabolic profiles were also obtained in eight normal pregnant women of equivalent gestational ages admitted to hospital for study. The mean blood-glucose and metabolite concentrations in both the meter and non-meter groups were similar to those obtained in the non-diabetic pregnant women (mean diurnal blood-glucose 4.6 +/- 1.1 mmol/l, meter group; 5.3 +/- 1.5 non-meter group; 4.8 +/- 0.8 controls). The diabetic control achieved at home was not improved by the use of a meter or admission to hospital (mean glucose at home 5.0 +/- 1.3 and 6.3 +/- 1.6 mmol/l in hospital). Both the use of a meter and admission to hospital may be valuable in some patients. Neither, however, is essential for the good control of blood-glucose in all pregnant diabetics.

BLOCKADE OF CHLORPROPAMIDE-ALCOHOL FLUSHING BY INDOMETHACIN SUGGESTS AN ASSOCIATION BETWEEN PROSTAGLANDINS AND DIABETIC VASCULAR COMPLICATIONS

Chlorpropamide/alcohol flushing (CPAF), found in many patients with non-insulin-dependent diabetes (NIDD), can be blocked by indomethacin in most patients who are free of vascular complications but not in those with such complications. Since indomethacin is a prostaglandin inhibitor this finding suggests that prostaglandins may be involved in the aetiology of vascular diseases in NIDD. All 6 pairs of identical twins with CPAF, of whom 2 pairs were disocrdant for diabetes, were concordant for indomethacin blocking, which suggests that the block has a genetic basis. The difference in the response of CPAF to indomethacin in diabetic patients with and without vascular complications is probably the first indication of a metabolic difference between these two types of patient.