A number of changes in therapy of uncontrolled diabetes have occurred in recent years. These include low-dose insulin regimens, often routine phosphate repletion, more cautious bicarbonate replacement, infusion of larger fluid volumes, the use of hypotonic solutions in hyperosmolar states, and recently magnesium repletion. These modalities (with the exception of routine magnesium repletion) have been employed at North Central Bronx Hospital since its opening in 1976. Through this retrospective analysis of 275 cases of uncontrolled diabetes we have tried to answer the following questions: What is the outcome of all episodes of uncontrolled diabetes in a municipal hospital population with a uniform treatment protocol? What are the results of treatment with new modalities in various age groups? Are the causes of death different from those tabulated in previous reports? Our results indicate a good outcome in those under the age of 50 yr regardless of the diagnosis of hyperosmolar nonketotic coma (HNC) or diabetic ketoacidosis (DKA). Mortality from DKA was 2% in those under age 50 yr and 26% in the older age group. Surprising was the low mortality in the hyperosmolar group with 0% mortality under age 50 yr and 14% in patients over this age. The major categories of causes of death in the series included sepsis, adult respiratory distress syndrome (ARDS), metabolic, cardiovascular, and shock. With the exception of ARDS, these categories were not different from other reported series. There were few thromboembolic events in this series.(ABSTRACT TRUNCATED AT 250 WORDS)

No abstract

We report the occurrence of the adult respiratory distress syndrome (ARDS) in association with uncontrolled diabetes in nine patients. In reviewing the literature we found nine similar cases reported in little over a decade. In most cases no condition known to precipitate ARDS was discovered. The evidence suggests that the severely uncontrolled diabetic state in some way may initiate pathologic events leading to the capillary leak of ARDS. This description of the association of these two entities not commonly recognized as occurring simultaneously has important clinical implications: the entity should be anticipated in uncontrolled diabetic patients who present with acidosis, hypotension, hypothermia, and/or coma. The clinical or radiologic diagnosis of pneumonia or fluid overload should not be made in the uncontrolled diabetic patient in the absence of unequivocal evidence of infection or congestive heart failure. The development of dyspnea, hypoxemia, rales, or infiltrates in the otherwise routine resuscitation of these patients should lead the clinician to suspect the development of ARDS. Prompt invasive monitoring in these cases is indicated to aid in their management and may help to improve survival. We found calculation of the A-a gradient to be useful in patients with uncontrolled diabetes. Although not necessarily predictive, widened gradients were the earliest detectable abnormality found in all patients who developed ARDS.

A young diabetic patient with ketoacidosis responding to standard treatment abruptly developed pulmonary edema, apnea, coma, cardiac arrythmias and diabetes insipidus. At autopsy, cerebral edema and neuronal degeneration were present. In the hypothalamus many neurones were absent. Similar cases have been reported, but the pathogenesis of the syndrome is not understood. The present patient received large amounts of sodium bicarbonate, and significant potassium deficiency may have been present.