John B. O'Sullivan

Gestational diabetes and perinatal mortality rate

Significantly higher perinatal mortality rates were found in a prospective study of 187 gestational diabetic patients and 259 randomly selected negative control patients. Since the gestational diabetic patients were older and obese, further analyses were made to determine the role of these variables. Age was found to have a disproportionately adverse effect, enhanced slightly by obesity, on the pregnancies of gestational diabetic patients when compared with those of negative control patients. Two classes of gestational diabetic patients are outlined on the basis of age. The first includes those below 25 years of age who show no increased fetal wastage in pregnancy but may be more susceptible to this problem in unfavorable obstetric circumstances. The second class includes gestational diabetic patients 25 years of age or older who have high-risk pregnancies even in centers administering good prenatal care.

The Sorbitol Pathway: Enzyme Localization and Content in Normal and Diabetic Nerve and Cord

The enzymes of the sorbitol pathway, aldose reductase and sorbitol dehydrogenase, were investigated in sciatic nerve and spinal cord. The substrate specificities of spinal cord aldose reductase indicates that it is a variant of TPN L-hexonate dehydrogenase which possesses poor polyol forming ability. In contrast, the cauda equina and sciatic nerve aldose reductase have a true aldose reductase with considerable polyol forming ability. The distribution of the latter enzyme is associated with the presence of Schwann cells. Wallerian degeneration experiments are compatible with the localization of true aldose reductase in the Schwann cell and sorbitol dehydrogenase in the axon. There was no change in the levels of sorbitol dehydrogenase in diabetic nerves, a 30 per cent decrease occurred in the aldose reductase content. This decrease suggests a metabolic abnormality of the Sehwann cell, possibly resulting from alrered cellular integrity, which may have important implication for the otiology of diabetic neuropathy.

Establishing criteria for gestational diabetes.

The general requirements for establishing diagnostic criteria with tests of glucose tolerance are discussed in relation to standards for gestational diabetes that are in current use. Data are presented to show that even when remissions of gestational diabetes occur with the progression of pregnancy, the risk of subsequent diabetes remains unaltered. The prognostic potential that fasting blood glucose levels has for the outcome of pregnancy is also considered. Finally the dependence of prevalence rates for gestational diabetes on the criteria selected and their substantial effect on assessment of screening methods is demonstrated.

Rheumatoid arthritis in a New England town. A prevalence study in Sudbury, Massachusetts.

Abstract Examinations for rheumatoid arthritis were made on 4626 persons comprising 77 per cent of the adult population of Sudbury, Massachusetts. With the use of American Rheumatism Association cr...

Aspects of birth weight and its influencing variables

Abstract 1. 1. Data from 5,883 registering prenatal patients were used to examine variables influencing birth weight. Statistical handling by multiple linear regression techniques was confirmed by simple three-way cross-classification tables. 2. 2. Initial results indicated that maternal age, parity, weight, past history of having borne a baby in excess of 9 pounds, length of gestation, and sex of the infant were all of possible significance. 3. 3. Further analyses removing the overlapping effects of intercorrelated factors were made. Apparent influence of age and parity was then seen to result from parity only. Extension of these analyses indicated that the effect of parity was entirely due to the maternal prepregnant usual weight. Evidence of increasing gestational blood glucose levels seen with increases in maternal weight formed the basis for postulating an environmental cause for the relationship between maternal weight and infant birth weight. 4. 4. The significant effect of having previously borne a large baby, while in keeping with this postulated environmental cause since such persons have higher than average weights as well as blood glucose levels in pregnancy, in no way negates possible genetic explanations. 5. 5. Length of gestation has the greatest single effect on infant birth weight. Its use as an index of maturity is discussed in relation to the difficulties in applying results from group data with large variances to an individual.

Gestational Blood Glucose Levels in Normal and Potentially Diabetic Women Related to the Birth Weight of Their Infants

Data from 5,534 prenatal registrants who had blood glucose determinations following the ingestion of 50 gm. of glucose indicated that age is the only maternal factor of significance to resulting blood glucose levels when other variables are held fixed. The lack of a blood glucose—birth weight correlation in the general population was further explored with the demonstration that such a relationship exists among groups of potentially diabetic women. Overweight women giving birth to a large baby were found to have significantly higher mean blood glucose values. The overweight women with an infant of average birth weight or the normal weight women with a large baby had no such blood glucose elevation. These data imply that more accurate separation of either the overweight women or women giving birth to large babies into groups of special significance to future development of diabetes mellitus is possible.

Insulin Treatment and High Risk Groups

A 16-yr prospective study of 615 gestational diabetic subjects, half of whom were randomly assigned to insulin therapy during pregnancy, is evaluated to report on the potential reduction in subsequent diabetes due to insulin therapy. No such difference in incidence rates was observed between the two groups. In the subsets of women managed with insulin who bore a baby of large birthweight or who had a family history of diabetes, subsequent decompensated diabetes was found to be significantly reduced. Both life table and multivariate analyses to adjust for the effects of age, weight, level of blood glucose at selection, duration of follow-up, and other potentially confounding covariables confirmed this conclusion. The finding suggests the possibility of long-term preventive benefits from insulin treatment in high risk subsets of women with gestational diabetes.

The prevalence of diabetes mellitus and related variables—A population study in Sudbury, Massachusetts

Abstract 1. 1. Seventy-seven per cent of the town of Sudbury was examined for diabetes. Diagnoses required both elevated postprandial blood sugars and an abnormal glucose tolerance test. 2. 2. The prevalence of known diabetes was found to be 1.1 per cent, while newly discovered cases amounted to 0.8 per cent. Evaluation of prevalence includes a discussion of the problems encountered in acceptance of known diabetics as well as the effect of varying diagnostic criteria on the acceptance of new diabetics. 3. 3. Analyses of the sex prevalence of diabetes indicated a preponderance of males. No increase was found in the prevalence of diabetes among childbearing women. The significance of these findings is discussed. 4. 4. No relationship was found between a close family history of diabetes and blood sugar levels. 5. 5. Postprandial blood sugar levels were found to be unrelated to body weight. Values obtained following the ingestion of 100 g of glucose, on the other hand, were related to body weight indicating their greater discriminatory ability.

Age Gradient in Blood Glucose Levels: Magnitude and Clinical Implications

Population studies confirm the finding that average blood glucose levels in the fasting state increase with age. This blood glucose gradient is statistically significant even when confounding factors, such as obesity, are considered. The order of magnitude of the increase, however, may be considered of questionable clinical significance since it amounts to no more than 2 mg. per 100 ml. per decade through the adult years. And yet, it reflects a problem that continues to puzzle both investigators and clinicians alike. The small change in blood glucose levels with advancing years is magnified to a considerable extent in conditions other than fasting. Postprandial values change at the rate of 4 mg. per 100 ml. per decade, and those following a glucose challenge at 8 to 13 mg. per 100 ml. per decade. In circumstances which affect carbohydrate metabolism, such as steroid administration, the age gradient following a glucose challenge is even more impressive—up to 18 mg. per 100 ml. per decade during cortisone-glucose tolerance tests. The magnitude of the change in blood glucose levels attributable to age, then, appears to be related to the metabolic environment, size, character and proximity of the carbohydrate challenge. These changes pose formidable problems for the clinician in situations of diagnostic uncertainty. For example, failure to adjust critical screening or diagnostic blood glucose levels may give rise to an uncomfortably high proportion of older persons being diagnosed diabetic. In this regard, data from the National Center for Health Statistics can be used to calculate the difference in numbers of persons rating positive at

Evaluation of Age-adjusted Criteria for Potential Diabetes

Data from Oxford, Massachusetts, indicate that one to two-hour postprandial blood sugar levels rise 4 mg. per 100 ml. per decade. In order to test whether postprandial data used for screening or diagnostic standards should be correspondingly adjusted for age, two criteria, comparable except that one resulted in a sliding scale age-adjustment, were applied to blood sugar levels obtained in Oxford in 1946–1947. The development of diabetes over the next twenty-two years provided no justification for raising these standards for increases in age. These data are considered to provide a pragmatic solution to the rise in glucose levels that occur with age, but are insufficiently precise to resolve its pathophysiologic meaning. The discussion emphasizes that the conclusion reached does not necessarily apply to data obtained following glucose ingestion.