Eighty type 2 DM cases were randomly selected from patients attending Diabetic clinic Out Patient Department (OPD), R.G.Kar Medical College. Among them 60 patients were included following inclusion and exclusion criteria. Age and sex matched 60 healthy controls were included according to inclusion and exclusion criteria. After proper ethical clearance, a comparative cross-sectional study was conducted at the Department of Physiology, R.G.Kar Medical College & Hospital in collaboration with Department of Medicine, Department of Biochemistry from January 2010 to November 2012.

Diagnosis of type 2 DM were done according to the National Diabetes Data Group and World Health Organization diagnostic criteria for DM based on the following premises [1].

Following patients were excluded from the study: Smokers; history of respiratory diseases such as asthma, COPD, tuberculosis, ILD; H/O occupational exposure; H/O URTI & LRTI; Hypertension, H/O angina; CVA; Obesity (BMI>30 kg/m²); known thyroid disorders, autoimmune disease like SLE, RA; known kidney diseases, Hereditary peripheral neuropathy; individuals with unacceptable spirometric technique, due to various causes like obstruction of teeth or tongue, sub-maximal effort, air escape, effort sustained for less than 6 s duration, failure to attain a plateau on volume time curve, recent surgery.

Detailed history and clinical examinations were done and blood sample after overnight fasting was taken for the Fasting plasma glucose and post prandial plasma glucose (by Glucose Oxidase Peroxidase method [10,11], glycated hemoglobin (HbA1C) level as an index of glycemic control (by Ion Exchange Resin method) [12].

Height and weight of each subject was measured. Self-reported smoking history was considered. Pulmonary functions were measured by the electronic spirometer, model-RMS Helios-702 in accordance with the standards of lung function testing of the American Thoracic Society/European Respiratory Society (ATS/ERS) [13]. Pulmonary function report included patient’s gender, height, weight, age and smoking status. Standard spirometric measures included, forced vital capacity (FVC), forced expiratory volume in one second FEV1, the ratio of forced expiratory volume in one second to forced vital capacity (FEV1/FVC), forced expiratory flow rates (FEF25%, FEF50%, FEF75%, and FEF25%-75%) and peak expiratory flow rate (PEFR). Pulmonary function variables were recorded as a percentage of the normal value predicted on reported height and age [14]. Post bronchodilator (reversibility test) testing was performed 10 min after administration of the bronchodilator.

Measurement of DLCO: The DLCO of the subjects of this study were measured by single breath (DLCOsb) method using computerized DLCO measuring machine, - INPIRE- HD–PFT [15]. Best of three satisfactory readings was taken for analysis. The technique was validated in our laboratory and the prediction equations for normal Indian subjects had been derived and reported previously [15,16]. Normal values are based upon age, height, ethnicity, and sex. A value is usually considered abnormal if it is less than 80% of predicted value [15,16].

Data were analysed in SPSS software- version 17 (IBM, Chicago, Illinois, 2008). p-value of <0.05 was taken as significant with 95% confidence interval.

[Table/Fig-1] demonstrate that cases and control are age matched. The mean BMI, FPG, PPPG, HBA1c were significantly (p<0.05) increased in T2DM patients than the control.

Our study shows that there is a significant (p<0.05) decrease of PFT parameters (FVC, FEV1, PEFR, FEF 25-75 ) and diffusion capacity (DLCO and DL/VA) whereas FEV1/FVC is significantly increased in cases compared to control [Table/Fig-2].

Applying ANOVA test between three groups according to HBA1c level [Table/Fig-3] shows that there is significant decrease in FVC, DLCO and DL/VA and significant increase in FEV1/FVC in that groups having HBA1c level >7%.

This study shows that [Table/Fig-4] there is negative association between HbA1c level and PFT parameters (FVC, FEV1) but FEV1/FVC is positively correlated with HbA1c level. Regarding diffusion capacity both DLCO and DL/VA are negatively correlated with glycemic status.

Comparison of pulmonary function observed in the present study with the corresponding predicted values published from various regions of India is shown in [Table/Fig-5,6]. When compared to the selected studies our result shows almost similar changes like the most of the studies.

Renal, retinal and other manifestations of diabetic microangiopathy have frequently been studied but pulmonary complications of Type 2 Diabetes have been poorly characterized specially in the eastern Indian regions. In this study fasting and post meal blood glucose levels and HbA1c% were found to be significantly more in type 2 diabetics than the controls pointing to the fact that there was poor glycemic control. This may be because of irregular drug intake, inappropriate drugs, sub-dosing, overeating, lack of diabetic life style discipline, etc practiced by the patients [18]. HbA1c% is an indicator of diabetes control. Higher the level of HbA1c%, poor is the diabetic control i.e. higher level of circulating glucose. If circulating glucose is constantly at a higher level for 3 month (as measured by HbA1c%), it can lead to more and more nonenzymatic glycosylation of tissue proteins. If respiratory system is a target, this will be reflected in the PFT parameters analysed.

Our study found that FVC, FEV1 were significantly reduced in type 2 diabetics whereas FEV1/FVC was significantly increased in cases compared to control. This signifies that restrictive lung pathology occurs in diabetes. Similar findings were observed by other authors [6,19].

In type 2 diabetics FEF 25- 75% is significantly reduced compared to controls. FEF 25-75% is an indicator of force of expiration of gases during middle 50% of forced expiration. Forced expiration is supported by muscular and recoil forces of the respiratory system. Thus decrease in muscular and recoiling forces of the respiratory system because of increased glycosylation is responsible for significant decrease in FEF 25-75%. Similar findings were observed in other studies [18,20]. The flow can also be decreased due to obstruction.

PEFR is the gas exhaled in 1/10^th of a second during forced expiratory manoeuvre when recoiling forces of the lungs and contractile forces of respiratory muscles are functioning maximally and supporting the expiration to the maximal. Due to glycosylation of the connective tissues of the respiratory apparatus, the recoiling forces of the lungs and the contractile forces of the respiratory muscles might be decreased, leading to a significant reduction in PEFR [21,22]. Our study also shows the similar finding and extended the observations of previous researchers.

Diffusion capacity of the type-2 diabetic patients also deteriorated in our study. Asanuma et al., [23] reported a significantly lower forced vital capacity in diabetics along with a decreased diffusing capacity. Sinha et al., [6] also concluded that impairment of DLCO was common in type2DM Asian Indian having pulmonary capillary endothelial dysfunction and it could be related to insulin resistance and dyslipidaemia. In contrast to the above findings, Bulbou et al.,[24] did not find any correlation between reduced diffusion capacity in diabetics with diabetic complications and others [25] have reported no significant difference in diffusing capacity between healthy subjects and diabetics.

Study of Agarwal et al.,[7] showed the relation of lung function parameters with glycemic control(HBA1c). There was a decrease in FVC%, FEV1%, PEFR%, FEF25-75%, DLCO%, DL/VA% and an increase in FEV1/FVC% with increase in HBA1c level. Another Study by Sinha et al., [6] also showed that lung function parameters were decreasing with deterioration of glycemic control and there was a negative correlation between DLCO and HBA1c level(r=0.62, p<0.05). They found no difference in other pulmonary function only correlations were observed between DLCO and HbA1C (r=0.62, p<0.05). Other studies [6,7] have observed only correlation between diffusing capacity but spirometric values did not differ in type-2 diabetes patients.

A small sample size and non-measurement of TLC, cross-sectional study with no follow-up are the limitations of the present study. Further, histological studies on pulmonary microvasculature and compliance measurements of the lung would be useful to investigate reasons for reduced DLCO values.

Our study shows significant changes of FVC%, FEV1/FVC%, DLCO% and DL/VA% in Type-2 diabetes patients and it has been correlated with poor glycemic control. The above pattern of changes are possibly due to hyperglycemia induced non enzymatic glycosylation of tissue proteins and chronic diabetic microangiopathy causing basement membrane thickening (capillaries and endothelium) leading to reduction in strength and elasticity of connective tissues and reduced pulmonary blood volume with V/Q mismatch impairing the diffusion capacity.