Issue 117 Item 13 Impaired Lungs Seen in Diabetes, Another Complication

Findings in two large trials show— Impaired lung function may be yet another complication of diabetes. Findings could delay the introduction of inhaled insulin. Rates of impaired lung function appear to be higher in patients with elevated hemoglobin A1c levels. Moreover, impaired lung function also appears to predict the onset of type 2 diabetes and to be independently associated with increased mortality rates, researchers reported at the annual scientific sessions of the American Diabetes Association. The cause of the lung impairment remains unclear.

Diabetic patients should be specifically asked about wheezing, dyspnea, and shortness of breath, Dr. Timothy M.E. Davis told this newspaper. Lung function testing is a consideration for those who are symptomatic.

Such symptoms in diabetic patients are frequently presumed to be the result of heart disease rather than a lung problem, added Dr. Davis, professor of medicine at the University of Western Australia, Perth.

The development of inhaled insulin is one reason for the surge in interest in lung function in patients with diabetes. Inhaled insulin is now under study by several pharmaceutical companies. The new lung findings “put a little more burden on the industry to do long-term studies,” Dr. Frederick L. Brancati of Johns Hopkins University, Baltimore, said in an interview.

In a poster, Dr. Brancati’s associate at Hopkins, Dr. Hsin-Chieh Yeh, presented an analysis of spirometry data from the Atherosclerosis Risk in Communities (ARIC) study, a population-based trial of 11,365 adults aged 45-64 years.

The 1,187 ARIC subjects with diabetes had significantly lower mean forced vital capacity (FVC) at baseline (3.5 L) than did those without diabetes (3.7 L) after adjustment for age, sex, race, and height. Forced expiratory volume in 1 second (FEV1) also was lower among the diabetic patients (2.6 L vs. 2.9 L).

These differences remained significant after adjustments for adiposity, smoking, physical activity, education, and coexisting cardiovascular diseases.

The FVC was inversely correlated with hyperglycemia. Compared with subjects whose fasting blood glucose level was 97 mg/dL or less, FVC was 19 mL lower among those with glucose levels of 98-109 mg/dL and steadily declined to 187 mL lower if levels exceeded 169 mg/dL.

A similar dose-response relationship was seen between FEV1 and fasting glucose levels, with an average FVC decline of an additional 6 mL/year among diabetic compared to nondiabetic patients. Diabetes was not associated with a lower FEV1/FVC ratio at baseline or at the 3-year follow-up. The reduced FVC—associated with restrictive, rather than obstructive, lung damage—is driving the decline in lung function, Dr. Yeh said.

A similar association between hyperglycemia and diminished lung function also was seen among 125 participants in the Fremantle Diabetes Study. During 7 years of follow-up, several lung function measures—FVC, FEV1, vital capacity, and peak expiratory flow (PEF), all normalized for height—declined by 15%-31% per year in absolute terms, and by 8%-25% of predicted value, said Dr. Wendy A. Davis, a biostatistician at the University of Western Australia, Perth.

After controlling for age, gender, smoking status, and baseline lung function, higher values of glycosylated hemoglobin and fasting plasma glucose were associated with reductions in FVC, FEV1, and PEF (all normalized for height) as well as percent-predicted lung function.

The link between glycemic exposure and deteriorating lung function was independent of exercise level, diabetes duration, other diabetes complications, and other comorbidities, she said. the.wolfs@verizon.net
Dr. Brancati presented data from a separate prospective analysis of 13,602 ARIC subjects who did not have diabetes at baseline. During 9 years’ follow-up, 1,304 developed the disease.

The risk of developing diabetes was significantly greater when baseline FVC was reduced. Those in the highest FVC quartile (112%-136% of predicted value for age, sex, and height), had a risk of 7%, compared with 18% for the lowest quartile (68%-92% of predicted value).

In a separate analysis of the Fremantle data, Dr. Timothy Davis reported on 102 deaths in a 7-year follow-up of 494 diabetic patients with no respiratory disease at baseline spirometry. In a multivariate analysis, independent predictors of death included age of 65 or older (relative risk 3.4), neuropathy (2.5), coronary heart disease (2), male gender (1.8), and albumin/creatinine ratio of at least 3 mg albumin/mmol creatinine (1.62).

After adjusting for these variables, reduced vital capacity normalized for height still predicted death. “We were expecting cardiovascular disease and nephropathy to swamp it out,” Dr. Davis said.