IMPAIRED LUNG FUNCTION IN A HISPANIC DIABETES POPULATION: THE RELATIONSHIP BETWEEN METABOLIC RISK FACTORS AND PULMONARY FUNCTION By KATHERINE A. DE LA TORRE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2016
2016 Katherine A. De la Torre
To my m o ther
4 ACKNOWLEDGMENTS This thesis wo uld not have been possible with out the constant support and encouragement I received from my academic mentor, Dr. David Sheps. I sincerely thank him for his consistent efforts and true desire to keep me on track. I would also like to thank my committee members, Dr. Catherine Striley and Dr. Mattia Prosperi for serving in my defense committee despite their overwhelmingly bus y schedule. Finally, I would like to express my deepest gratitude to my parents and friends. Their support and unwavering confidence in my ability helped me achieve my academic dreams.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURES ................................ ................................ ................................ .......... 7 LIST OF ABBREV IATIONS ................................ ................................ ............................. 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 Diabetes Mellitus and Lung Impairment ................................ ................................ .. 11 Lung Function ................................ ................................ ................................ ......... 12 2 METHODS ................................ ................................ ................................ .............. 15 Study Population ................................ ................................ ................................ ..... 15 Study Design ................................ ................................ ................................ .......... 15 Statistical Analysis ................................ ................................ ................................ .. 17 Ethical Statement ................................ ................................ ................................ .... 17 3 RESULTS ................................ ................................ ................................ ............... 19 4 DISCUSSION ................................ ................................ ................................ ......... 23 5 CONCLUSIONS ................................ ................................ ................................ ..... 29 APPENDIX: LETTER OF DATA USE PERMISSION ................................ .................... 30 LIST OF REFERENCES ................................ ................................ ............................... 32 BIOGRAPHIC AL SKETCH ................................ ................................ ............................ 39
6 LIST OF TABLES Table page 3 1 Baseline characteristic of Hispanic study population as a whole and by gender ................................ ................................ ................................ ................ 19 3 2 Pulmonary Function Tests of study population and by gender. .......................... 20 3 3 Difference in Means and Odds Ratio of metabolic risk factors by Pulmonary Function Status (n=61) ................................ ................................ ....................... 21 3 4 Categorized age by Pulmonary Function stratified by Gender ........................... 21
7 LIST OF FIGURES Figure page 3 1 ROC curve for glycated hemoglobin by pulmonary impairment status. .............. 22
8 LIST OF ABBREVIATIONS BMI Body mass index DM Diabetes mellitus HbA1c Glycated Hemoglobin HDL c High density lipoprotein cholesterol FEV1 F orced expiratory volume in one second the volume of air exhaled in the first second under force after a maximal inhalation FEV1/FVC ratio P ercentage of the FVC expired in one second FVC F orced vital c apacity the total volume of air that can be exhaled during a m aximal forced expiration ef fort LDL c Low density lipoprotein cholesterol PFT Pulmonary function tests TC Total cholesterol
9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science IMPAIRED LUNG FUNCTION IN A HISPANIC DIABETES POPULATION: THE RELATIONSHIP BETWEEN METABOLIC RISK FACTORS AND PULMONARY FUNCTION By Katherine A. De la Torre August 2016 Chair: Catherine W. Striley Major: Epidemiology Type 2 diabetes mellitus is a chronic disease whose prevalence increas es with age. Inflammatory changes in lungs and glycosylation of connective tissues are o bserved in patients suffering type 2 diabetes mellitus. Additionally, these patients have higher risk than non diabetic o f developing infectious and non infectious pulmonary diseases as well as intrinsic cardiovascular and neuropathic diabetic complications. Lung function tests are painless, low cost and effective tools that produce important information to help with prognosis and reduce f uture complications ; hence lung function tests should be performed in this high risk population. This study aim s to measure the prevalence of abnormal lung function in a cohort of Hispanic p atients with diabetes mellitus and investigate associations between metabolic risk factors and impaired lung function measured by spirometry showing either obstructive or restrictive pattern. A cross sectional study was performed including 386 type 2 diabe tes mellitus (DM2) patients from a Public Health care unit in Quito, Ecuador. Prevalence of lung
10 disease characteristics of sam ple population were described. Wilcoxon test and t test were used for mean differences calculations ; univariate and multivariate logistic regression models were used to identify the correlates of pulmonary function and metabolic variables including glycated hemoglobin ( HbA1c ) total cholesterol, high and low density lipid cholesterol, triglycerides, body mass index and time of disea se. Analyses were adjusted by sex, age and smoking status. A total of 63 subjects had some kind of pulmonary functional impairment (17.8%). Among of them 40 (63.5%) had only restrictive pattern, 2 (3.2%) had only obstructive pattern and 21 (33.3%) subjects had a mixed spirometry pattern. In overall, 61 (15.8%) of the whole population exhibited res trictive type pulmonary changes Glycated hemoglobin did not work as predictive biomarker for lung restrictive impairment Potential impairment of lung function s ho uld be assessed in DM2 patients, p ulmonary function should be considered as important as renal or cardiac function evaluation in diabetic patients assessing pulmonary function tests reference values for this specific population regarding ethnicity and lo ng time high altitude residency.
11 CHAPTER 1 INTRODUCTION Diabetes Mellitus and Lung Impairment Type 2 Diabetes mellitus (DM ) is one of the most prevalent chronic diseases worldwide. In Ecuador (INEC, 2011) and in the rest of the world (Alwan, Armstrong, Cowan, & Riley, 2011) DM is among the top five direct causes of death The trend in adult diabetes prevalence worldwid e has increased, or at best remained unchanged since 1980 (NCD RisC, 2016) In fact, the number of individuals with diabetes mellitus around the world is projected to rise from 135 million in 1995 to 300 million in 2025 (IDF, 2013; King, Hilary. Aubert, Ronald. Herman, 1998) Pulmonary impairment in DM patients has not been investigated as much as cardiovascular, renal, and cerebrovascular complications. However, the prevalence of restrictive lung impairment related to metabolic diseases varies from 5 % to 12% among Americans (Guerra et al., 2010; Kurth & Hnizdo, 2015) Reduced lung volumes and DM the severity of which relates to glycemic exposure (Davis, Knuiman, Kendall, Grange, & Davis, 2004; Mannino et al., 2012) While s ome studies have found that DM patients tend to have poorer lung function than non diabetics 2003) oth ers have found that poor lung function can be treated as a predict ive marker of developing diabetes (Yamane et al., 2013) The proposed mechanisms of pulmonary impairment in DM patients include microangiopathy of lung capillaries and chronic low grade systemic inflammation due to hyperglycemia both of which are related to alterations of lung matrix proteins and basal lamina thickening (Weynand, B. et al 1 999; Rodolfo, D. et al, 2010). Additionally, insulin
12 resistance with or without hyperglycemia is also associated to lung function impairment in the elderly (Fimognari et al., 2007) F ibrosis of lung parenchyma and resp iratory muscle impair ment due to myopathy or autonomic neuropathy are also proposed mechanisms to explain the relationship between DM and lung function impairment (Scarlata, Costanzo, Giua, Pedone, & Incalzi, 2012) An increase in mean glycated hemoglobin ( HbA1c ) was associated with a decrease in the pul monary function tests : forced vital capacity ( FVC ) and forced expiratory volume in one second ( FEV1 ) in diabetic (McKeever, Weston, Hubbard, & Fogarty, 2005) more than non diabetic population s (Oh, Park, Lee, & Park, 2015) and may be a reliable predictor of poor lung function, especially the restr ictive pattern (Godfrey & Jankowich, 2016) There is a predominant reduction in all the pulmonary function tests of diabetic patients toward the restrictive patter n (I., Hamdy, Amin, & Rashad, 2013) ; with significantly lower FEV1 and FVC values in DM patients than in non diabetics, even after adjust ing for age, sex, BMI, smoking status, diabetes duration and HbA1c levels (Yeh et al., 2008) Moreover, i n the first National Health and Nutrition Examination Survey (NHANES I), restrictive and obstructive pulmonary impairment have been shown as significant predictive factor s for ear ly mortality (Mannino, Buist, Petty, Enright, & Redd, 2003) independent ly of smoking body mass index and adiposity (Hickson et al., 2011; Leone et al., 2009) Lung Function Spirometry is a basic low cost tool for the exploration of lung function, and could be performed both in the specialist and primary care consultation. Understand ing its principles, limitations, indications and results is essential to assess the degree and type of respiratory dysfunction, monitor pulmonary diseases and optimize treatments. The
13 American Th or acic Society (ATS) published recommendations for an adequate use of spirometry equipment and procedures in 1987 (Miller, Hankinson, et al., 2005) T hose recommendations have been followed in many studies for a quality assessment during test performance Spirometry is a test designed to quantify functional pulmonary volumes, where a patient needs to expire as hard and fas t in the spirometer as possible (Miller, Crapo, et al., 2005) The results of the test are compared to the predicted values that are calculated from their age, height, gender and ethnic group (Pellegrino et al., 2005) The output are flow volumes curves and pulmonary function test (PFT) as: f orced vital capacity (FVC the total volume of air that can be exhaled during a maximal forced expirat ion effort ), f orced expiratory volume in one second (FEV1, the volume of air exhaled in th e first second under force after a max imal inhalation ), FEV1/FVC ratio and the percentage of the FVC expired in one second Then, on the basis of these P FT percent predict ive values are calculated to discriminate a normal or abnormal spirometry pattern (Barreiro & Perillo, 2004) Pulmonary ventilatory function impairment is define d as the presence of restrictive, obstructive or mixed abnormalities (Pellegrino et al., 2005) People with an obstructive spirometry pattern have difficult y exhaling all the air from the lungs, whereas, people with re strictive pattern have difficult y fully inhaling air (Pollard et al., 1997) The most common causes of an obstructive pattern are chronic obstructive pulmonary disease (COPD) which includes emphysema and chronic bron chitis; asthma, and cystic fibrosis (Athanazio, 2012) Other diseases as interstitial lung disease, autoimmune diseases, obesity and neuromuscular disease are the most
14 common causes of restrictive pattern. These diseases could be established independently of diabetes diagnosis However diabetes is a commonly comorbid ity of these pulmonary diseases (Cavailles et al., 2013) and depend ing on the popul ation there is a 2 % to 16% prevalence of diabetes in people with COPD (Chatila, Thomashow, Minai, Criner, & Make, 2008) Additionally, there is some evidence of a relationship between airway inflammation and insensitivity to insulin in patients with asthma (Gulcan, Bulut, Toker, & Gulcan, 2009; Mansi, Joshi, Pandloskar, & Dhar, 2007) I n a retrospective longitudinal study diabetic subjects were at increased risk of developing asthma, COPD, pulmonary fibrosis and pneumonia, but not lung cancer and that might be a consequence of d iminished pulmonary function in diabetic patients (Ehrlich, Quesenberry, Eeden, Shan, & Ferrara, 2010) The association between DM and reduced pulmonary function has been described for many years (I. et al., 2013; Litonjua, Lazarus, Sparrow, DeMolles, & Weiss, 2005) but the clinical significance of this association is un known Nevertheless, i t is remarkable to understand how lungs might be affected by hyperglycemia and factors related to DM. The relationship between impaired lung function and risk factors related to DM is important to further understanding of pulmonary function in diabetic population and mig ht imply s trategies to impact the burden of significant conditions related to both poor lung function and diabetes However, to the best of my knowledge, t here are no published studies measuring pulmonary function in DM patients in Ecuador The aim of this study is to determine prevalence of lung impairment and the relationship among metabolic risk factors and pulmonary function in Ecuadorian DM patients
15 CHAPTER 2 METHODS Study Population The study population w ere DM patients who visited the Chimbacalle Non Communicable Diseases Club located in Quito, Ecuador for periodic health examination. This is a public health care unit which provides basic and specialized ambulatory care for pati ents with chronic diseases. The dataset has been collected as part of a trend descriptive observational study started in 2009 from medical records for measuring mortality and cardiovascular risk factors over time. W e were granted access to an un identified dataset containing medical records and baseline spirometry data from p atients in 2012. The sample population of this study was composed of Hispanic adults (Ecuadorians) who live in Quito at 2820 meters above sea level (masl), exercise for an hour at least 3 times per week, and attend either two or one one hour lectures abo ut type 2 diabetes or how to improve their health and nutrition per week. Additionally al l subjects ha d been taking daily doses of aspirin (100 mg) and cholesterol lowering drug ( simvastatin 20 mg) as cardioprotection, a s well as their diabetes medication Study Design A cross sectional study design was conducted This study measure d the prevalence of lung function in a cohort of Hispanic DM patients and investigate d associations between metabolic risk factors and the outcome of interest. The sample included 386 patients Inclusion criteria were patients previously diagnosed with type 2 diabetes under A merican Diabetes Association criteria of either sex, who had at least one spirometry in 2012. Patients with previous diagnosis of acute or c hronic respiratory
16 disease pregnancy history of occupational exposure, collagen, neurological or neuromuscular diseases deformities or physical disability that m ight affect lung function or do not allow a reliable spirometry procedure were excluded. S pirometry tests were performed with a Based s pirometer ( Welch Allyn CardioPerfect v1.6, NY, USA ) using the acceptance and repeatability criteria set forth by the American Thoracic Society (Miller, Crapo, et al., 2005) adjusted for age, sex and height based on NHANES III reference values P ulmonary function tests included : FVC, FEV1 and the ratio FEV1/FVC. Blood test s were performed by the Chimbacalle Non Communicable Diseases Club and recorded in the medical histories. Blood samples were obtained after 12 hours overnight fast for the estimation of HbA1c, to tal cholesterol (TC), high density lipid (HDL), triglycerides (TG) and creatinine; LDL was calculated based on a mathematical formula using other lipid profile variables ( LDL = TC HDL TG/5.0 (mg/dL))(Friedewald, W. 1972). Anthropometric measurements as height and weight were recorded and body mass index (BMI) were calculated by the formula of weight/height 2 Additionally, previous diagnosis of hypertension, age and duration of disease in years were recorded. Laboratory and spirometry variables were num e rical. Glycated hemoglobin was used as a biomarker of optimal metabolic control in diabetes patients. Case definition Restrictive pulmonary pattern w as defined as cases in which FVC values were l ower than 80% and FEV1/ FVC values were equal to or higher than 70% Obstructive pulmonary pattern was defined as cases in which FEV1 values were l ower than 80% and FEV1/ FVC ratio s were l ower than 70% Mixed pattern was
17 defined as cases in which both restrictive and obstructive definitions were satisfied (Barreiro & Perillo, 2004) Statistical Analysis Statistical analysis were performed using SAS 9.4 software package ( SAS Institute Inc., Ca ry, NC, USA) in data corresponding to 386 patients. Metabolic variables w ere analyzed with descriptive statistics (mean and standard deviation ) of the sample as a whole, as well as according to pulmonary function status. T o compare differences in means bet ween pulmonary function for the most relevant demographic and clinical characteristics s t test (for parametric variables) and Wilcoxon test ( for non parametric variables ) were used Age was categorized based on NHANES III references values (CDC, 2015) and associated with pulmonary function stratified by gender using Cochran Mantel Haenszel Statistics (CMH) Univariate and multivariate logistic regression were used adjust ed by sex and smoking status. Receiver operating characteristic (ROC) curve were graphically calculated to illustrate the accuracy of glycated hemoglobin to predict restrictive spirometry pattern ; where AUC value s of 0.5 corresponds to random chance and 1.0 for perfect accuracy to identify a disease and no disease subjects. Ethical S tatement The data of Chimbacalle Non Communicable Diseases Club (CENCT Spanish acronym ) in Quito, Ecuador has been gather ed as part of a n ongoing trend descriptive observational study s ince 2009. This study has been authorized by the Ecuadorian Ministry of Public Health with an aim to describe the overall health trends with in this i nstitution. The un identified data from 2012 as long as with the spirometry results were
18 provided to this author by the authorities of Chimbacalle Non Communicable Diseases Clinic for research purposes This secondary analysis study, along with the waiver of informed consent, w ere appr oved b y the Institutional Review B oard of University of Florida Additionally, a letter of data use permission w as provided by the Chimbacalle Non Communicable Diseases Clinic (CECNT). (See Appendix A)
19 CHAPTER 3 RESULTS As shown in Table 3 1 311 (80.57%) subjects were females The average age of the subjects was 60.41 years The median age of the population was 60, and 64.51% of the participants were younger than 65 years old A ge was normally distributed The majority ( 301 7 7.98 %) of subjects had high blood pressure as a comorbidity. Most ( 357 92.49 %) subjects were non smokers at the time of t he study. Table 3 1. Baseline characteristic of Hispanic study population as a whole and by gender All subjects N=386 Women N=311 Men N= 75 Mean/n (SD/%) Mean/n (SD/%) Mean/n (SD/%) Age (years) b 60.41 (11.70) 59.82(11.50) 62.84 (12.26) Duration of Disease (years) 7.54 (7.06) 7.58(7.08) 7.37(7.03) Smoking Status a Current 29(7.51%) 9(2.89%) 20(26.67%) Former 53(13.73%) 24(7.72%) 29(38.67%) Never 304(78.76%) 278(89.39%) 26(26.67%) Hypertension a 301(77.98%) 245(78.78%) 56(74.67%) HbA1c (%) 7.70(1.80) 7.72(1.79) 7.61 (1.81) Body Mass Index (kg/m 2 ) b 29.84(4.79) 30.14(4.99) 28.5 (3.77) Height (cm) 151.6(7.74) 149.32(5.92) 161.12(7.14) Total Cholesterol (mg/dl) b 180.19(39.17) 182.41(39.14) 171.04(38.22) HDL c (mg/dl) b 78.51(17.10) 79.9(16.94) 72.58(16.56) LDL c (mg/dl) 66.73(30.32) 66.81(31.28) 66.36(25.98) Triglycerides (mg/dl) 181.13(102.54) 181.58(93.75) 179.24(133.69) Creatinine (mg/dl) b 1.09(0.40) 1.05(0.33) 1.23(0.59) a Categorical Variables: Number and Percentage. b Statistical significant differences in means (p<0.05) by gender SD, standard deviation HbA1c, glycated hemoglobin
20 The number of years since diagnosis with type 2 DM was positively skewed with a median of 5 and a range of 0 37 years. HbA1c and HDL c showed positive skewness with a median of 7.30 (3.3 14.0) and 76 mg/dl (range of 4 5 166), respectively. Other covariates showed a positive skewness distribution, as well (data not shown). The difference in age, BMI, total cholesterol, HDL c and creatinine between women and men Out of the 63 subjects with impaired spirometry pattern, 40 (63.5%) had only a restrict ive pattern, 2 (3.2%) had only an obstructive pattern and 21 (33.3%) subjects had a mixed spirometry pattern. (Table 3 2) Mild restrictive and mild obstructive pattern were found in 88.52% and 91.30% of the lung impairment group. Therefore, a total of 61 s ubjects had some kind of restrictive spirometry pattern (15.8%). Restricted spirometry was found among 8 (10.7%) men and 53 (17.0%) women. Table 3 2. Pulmonary Function Tests of study population and by gender All subjects n=386 Women n=311 Men n= 75 Mean/n (SD/%) Median (Range) Mean/n (SD/%) Mean/n (SD/%) FVC total volume (L) 2.66 (0.73) 2.60 (0.80 6.70) 2.46 (0.57) 3.51 (0.70) FVC percent predicted (%) 96.47 (17.97) 95.00 (46 208) 95.55 (17.97) 100.31 (17.59) FEV1 total volume (L) 2.29 (0.65) 2.20 (0.80 6.00) 2.12 (0.51) 2.99 (0.67) FEV1 percent predicted (%) 107.4 (22.10) 105.00 (49 252) 105.96 (21.58) 113.3 (23.39) Absolute FEV1/ FVC ratio (%) 86.0 (7.29) 85.7 (56.4 100) 86.18 (7.18) 85.25 (7.75) Restrictive Spirometry Pattern ab 40 (63.5%) 36 (90.0%) 4 (10.00%) Obstructive Spirometry Pattern 2 (3.2%) 1 (50.0%) 1 (50.0%) Mixed Spirometry Pattern ab 21 (33.3%) 17 (80.9%) 4 (19.1%)
21 Subjects with restrictive pattern were found to have significantly higher mean HDL c value and lower mean age than those with normal spirometry pattern ( P < 0. 05). O ther risk factors either unadjusted or adjusted, did not show association with pulmonary function (Table 3 3 ) Table 3 3. Difference in Means and Odds Ratio of metabolic risk factors by Pulmonary Function Status (n=61) Pulmonary Function Status Restrictive (n=61) Normal (n=325) p value a Crude OR (95%CI) Adjusted OR (95%CI) b Mean (SD) Age (years) 57.46 (12.51) 60.96(11.48) 0.046 0.97 (0.95 0.99) 0.98 (0.95 1.00) Duration of Disease (years) 7.23(6.35) 7.60(7.20) 0.584 0.99 (0.95 1.03) 0.99 (0.95 1.03) HbA1c (%) 7.92(1.59) 7.66(1.83) 0.079 1.08 (0.93 1.25) 1.08(0.93 1.25) Body Mass Index 29.66(6.15) 29.90(4.50) 0.247 0.98 (0.93 1.05) 0.98 (0.93 1.04) Total Cholesterol (mg/dl) 188.10(43.76) 178.71(38.14) 0.177 1.01 (0.99 1.02) 1.01(0.99 1.02) HDL c (mg/dl) 84.64(30.38) 77.37(16.20) 0.023 1.02 (1.01 1.04) 1.02(1.01 1.04) Triglycerides (mg/dl) 180.93(117.38) 181.17(99.72) 0.518 1.00 (0.99 1.01) 1.00(0.99 1.01) Creatinine (mg/dl) 1.07(0.28) 1.09(0.42) 0.624 1.00 (0.99 1.01) 1.00(0.99 1.01) a Wilcoxon test, age t student b Adjusted by gender and smoking status The association between age and pulmonary function was significant but weak for females (p = 0.0430). In contrast, the results for males show a non significant association (p < 0.05). The overall tests of association controlling for gender showed non significance based on CMH s tatistics (p = 0.4190) (Table 3 4) \ Table 3 4. Categorized age by Pulmonary Function stratified by Gender Female a n (%) Male n (%) Age (years) Restrictive Normal Restrictive Normal 20 40 1(9.09) 10(90.91) 1(33.33) 2(66.67) 41 60 34(22.08) 120(77.92) 3(11.54) 23(88.46) 61 up 17(88.36) 129(88.36) 5(10.87) 41(89.13) a Female: p value=0.0430
22 T he t est of homogeneity of age and pulmonary function across gender was non significant (p value=0.3506). Thus the association can be considered to be the same for men and women. To evaluate the accuracy of glycated hemoglobin as biomarker which might predict the developing of pulmonary restrictive pattern, we used ROC analysis expec ting values of the area under the curve (AUC) close to 1.00 (Kelly Zhou) However, AUC was 0.57 show ing that glycated hemoglobin as a metabolic biomarker has a weak predictive abi lity to discriminate patients with restrictive spirometr y pattern from normal subjects (Fig. 3 1) Figure 3 1. ROC curve for glycated hemoglobin by pulmonary impairment status
23 CHAPTER 4 DISCUSSION This study investigated the prevalence of pulmonary function impairment among adult Hispanic type 2 DM patients living at high altitudes and determine d metabolic risk factors for pulmonary impairment The key findings of the present study are that 1) high age is a protective factor among diabetes, 2 ) high HDL c values were related with restrictive lung impairment and 3) m etabolic risk factors were not associated wit h pulmonary restrictive pattern. Nevertheless, the prevalence of pulmonary impairment in present study was 1 7 .80 % of the screened pat ients and the prevalence of a restrictive pulmonary impairment was 15.8% which is higher as other population s where the prevalence of pulmonary restrictive impairment varies from 6.5 to 12.3% ( NHANES III 2 007 2010; Guerra et al., 2010, Kurth & Hnizdo, 2015 0tata 1999 ) However, this comparison is not possible since our study population is Hispanic diabetics living at high altitude (9,000FASL), whereas, other studies have used healthy participants, living at sea level or not more than 2,000 FASL and Hispani cs are not completely represented. Out of 63 patients with pulmonary impairment, 33.33% showed a mixed pattern; c ombined obstruction and restriction rarely occurs and is more common caused by a combination of pulmonary parenchymal and non pulmonary disorde rs (Diaz Guzman, McCarthy, Siu, & Stoller, 2010) Nevertheless, pulmonary impairment severity was measured as one of the principal clinical significance of spirometry, mild restrictive pattern was found in 88.5% and severe restrictive pattern just in less th an 2% of the impaired subjects. Finally, in the present population and in others (Mannino et al., 2012) the prevalence of
24 restrictive pa ttern was higher in women than in men ; it might due to the low prevalence of men in this sample population ( 19.43 %). FVC values for pulmonary restrictive impairment diagnosis has a positive predict ive value of 43%, sensitivity of 60 86 % and specificity of 83 90 % (Shawn, Dales, & Cardinal, 1999) a nd patients with this diagnosis need further analy sis to confirm the restrictive impairment. Although, t his study found some evidence of high values of FVC and FEV1 in highland Ecuadorian population might be due to long term residence at high altitudes is known to affect pulmonary capacity and oxygen inta ke (Pollard et al., 1997). However, the evidence of increased values of FVC among highlands populations is mixed. Increases in FVC and FEV1 were obser ved at increasing altitudes in H ispanic Peruvian ( 4,105 masl (Valenzuela & Ramos, 2004) Tibetan and I ndian ( 3300 masl (Wood, Norboo, Lilly, Yoneda, & Eldridge, 2003) populations. In contrast Hispanic Colombian population did not show differences in PFT altitudes over 2,600 masl (Szeinuk, 2016) and in Nepalese Himalaya population showed a decreasing FVC and FEV1 unchanged values with increasing altitude (Mason et al., 2000). Therefore, s ince PFT are influenced by body height, gender, and ethnicity (Nepal, Das, & Bhaila, 2014), a prediction equation for PFT spiro metry parameters should be established for the Ecuadorian population. Hence, future studies about the clinical significance of PFT in DM patients living at high altitudes should be performed. Metabolic risk factors did not show a significant association wi th pulmonary restrictive impairment, probably because diabetes is not the only cause of developing pulmonary restrictive pattern. Increased BMI values are clearly associated with increased prevalence of DM (Bays, Chapman, & Grandy, 2007) Persons with a h igher
25 BMI are thus e xpected to ha ve restrictive pulmonary pattern (McClean, Kee, Young, & Elborn, 2008; Wannamethee, Shaper, & Whincup, 2005) due to a thicker and less flexible body wall (Zammit, Liddicoat, Moonsie, & Makker, 2010) In light of these findings, was expected that diabetics with a higher BMI would have a restrictiv e pulmonary pattern. Nevertheless, our results do n o t support th at theory. Differences in chest dimensions due to ethnicity in pulmonary function (Whittaker, Sutton, & Beardsmore, 2005) intrauterine factors (Lawlor, Ebrahim, & Davey Smith, 2005) or other body weight components (Mohamed et al., 2002; Rossi et al., 2008) may be the cause s for the observed no relationship between BMI and restrictive pulmonary pattern in the present study In this study, glycated hemoglobin used as a biomarker of optimal glucose control showed a low predictive capacity to distinguish patients with lung impairment from normal sub jects. Previous studies have shown that higher values of HbA1c are associated with a restrictive pattern in type 2 DM patients (Lange et al., 1989; Pitocco et al., 2012) Th i s r esult may be explained by the fact that the high blood glucose in diabetes patients is not the only cause of developing a pulmonary restrictive pattern Additionally, m etabolic risk factors such as total cholesterol, triglycerides, serum creatinine or duration of disease unadjusted and adjusted by gender and smoking status, did not show a significant association with lung restrictive impairment. The low percentage of subjects who smoke might be attributable to cultural factors (women generally do not smoke) and economic factors (very low disposable income, tobacco cost). Th ese factors also explain the low prevalence of the obstructive pattern due to the intrinsic relationship between the major causes of obstructive ventilatory diseases and
26 smoking (Laniado Laborin, 2009; Tamimi, Serdarevic, & Hanania, 2012) Furthermore, y ounger patients showed a reduced risk of having restrictive pulmonary im pairment, and those patients with higher values of HDL c showed increased risk. These results are opposite to the cited literature, wh ich lung impairment risk increased with age (Mirabelli et al., 2016; Pitocco et al., 2012) and higher values of HDL c we re consider ed a protective factor for cardio metabolic diseases (Assmann & Gotto, 2004; Goldbourt, Yaari, & Medalie, 1997; Verdier et al., 2013) These findings might be explained by r is k factors for chronic respiratory diseases (smoking, indoor and outdoor pollutants, allergens or occupational agents) are more likely to be related to developing pulmonary restrictive impairment than metabolic risk factors. Considering that this study was carried out at one time point and it is unknown whether lung impairment occurred before, after, or during the onset of diabetes c ausality cannot be inferred Additionally, the dataset include d only hypertension data, without other major co morbidit ies bec ause patients with acute or chronic lung diseas es were excluded from the study and LDL was calculat ed using a mathematical formula H igher HDL c values and triglycerides outlier values c ould have made the LDL c values unreliable to associate with pulmonary restrictive pattern. Even though t his fixed population has its own ethnic particularities their exercise habits at high altitude and daily medication intake ( aspirin and cholesterol lowering drug ) cannot completely explain the higher values of HDL c and the higher values of HDL c among those with the pulmonary restrictive pattern found in this cohort of patients In this study population we foun d high mean HDL c value and several patients exhibited HDL c values over 80 mg/dl. It is challenging to compare HDL c values
27 between this and other populations because most of the available literat ure reports HDL c values above 4 0 mg/dl for men and 5 since the se are the standard HDL c cut offs for diagnoses o f dyslipidemia and cardiovascular risk factor s (Rodriguez et al., 2014) Nevertheless, HDL c genetic mut ations have been associated with elevated HDL c levels. (Bromley et al., 2005; Yamashita et al., 2000) Thus, it is possible that the extreme values of HDL c can be genetically inherited in this population of Hispanic type 2 DM patients Additionally, patients who reside at high altitudes have been found to have high HDL c values in some cases (Riyami et al., 2014; Vats et al., 2013) but not in others (Gonzales & Tapia, 2013; Mlaga, Zevallos Palacios, De Los, Lazo, & Huayanay, 2010) Hence, future studies will be necessary to assess whether genetic, physiological, or a combination of factors are related to high HDL c values. While there were good reasons to support the use of NHANES III spirometry reference values, this decision may have an FVC and FEV1 been faulty. NHANES III spirometry data described normal pulmonary function for three major ethnic groups: Caucasians, African Americans and Mexican Americans Even though there were no longitudinal large sample st udies in Hispanic no n Mexican population s we thought the similar body types observed in Mexicans and Ecuadorians might allow us to use NHANES III as Hispanic references values for spirometry test s H owever, the percent predicted values of PFT in this Ecua dorian population might be higher than the reference values due to geographical location or ethnic factors f urther studies will be necessary to disc ern these differences.
28 While the results of this study cannot be completely generalized to other Hispanic populations because t his population resides in a location found over 2 820 masl compar ison to other highland populations is warranted Most important this study addressed an o verall insight of the prevalence of pulmonary impairment and r elated characteristics which are important in public health for assessing the burden of disease in a diabetes highland population, as long as with a d escriptive analyses for diabetic population pulmonary function tests characteristics. Further studies nee d to be performed to ascertain the factors that contribute to pulmonary impairment amongst people in highlands. First, prediction equation for PFT spirometry parameters should be established for Ecuadorian population. Second, PFT spirometry parameters shou ld be compared comparison between highland and sea level Ecuadorian populations. Finally, studies should be conducted to investigate differences in pulmonary function among healthy and diabetic patients.
29 CHAPTER 5 CONCLUSIONS Pulmonary function impairmen t should be considered as important as renal or cardiac function evaluation in diabetic patients. This study has shown a prevalence of 1 5.8% of pulmonary restrictive impairment among Hispanic diabetic population Glycated hemoglobin was found to be a weak predict association with pulmonary impairment. Metabolic risk factors as lipid profile, BMI, and duration of di sease were not statistical significant associated w ith lung restrictive impairment; however, s ubjects with restrictive lung impairment had lower mean age and higher mean HDL c values than those with no lung impairm ent. This study provided a basic information for further analysis of lung impairment in a diabetic highland Hispanic population and high lighted the importance of having s pirometry reference values for this specific population regarding ethnicity and long time high altitude residency.
30 APPENDIX LETTER OF DATA USE PERMISSION L etter of data use permission p rovided by the Chimbacalle Non Communicable Diseases Cl ub (CECNT Spanish acronyms ) Spanish and English translated versions.
32 L IST OF REFERENCES Alwan, A., Armstrong, T., Cowan, M., & Riley, L. (2011). Noncommunicable Diseases Co Who 1 207. http://doi.org/10.2471/BLT.07.045138 Assmann, G., & Gotto, A. (2004). HDL Cholesterol and Protective Factors in Atherosclerosis. Journal of the American Heart Association 109 8 14. http://doi.org/10.1161/01.CIR.0000131512.50667.46 Athanazio, R. (2012). Airway disease: similarities and differences between asthma, COPD and bronchiectasis. 67 (11), 1335 43. http://doi.org/10.6061/clinics/2012(11)19 Barreir o, T. J., & Perillo, I. (2004). An Approach to Interpreting Spirometry. American Family Physician 69 (5), 1107 1114. Bays, H. E., Chapman, R. H., & Grandy, S. (2007). The relationship of body mass index to diabetes mellitus, hypertension and dyslipidaemia : Comparison of data from two national surveys. International Journal of Clinical Practice 61 (5), 737 747. http://doi.org/10.1111/j.1742 1241.2007.01336.x Bromley, C., Moushumi, Chaudhury Craig, R., Deverill, C., Erens, B., Fuller, E., Gray, Prescott, 2 Kevin Pickering, 2 Paola Primatesta, 3 Shaun Scholes, 2 Nicola Shelton, 3 Svetlana Speight, 2 Emmanuel Stamatakis, 3 Heather Wardle, 2 Paola Zaninotto. (2005). The Scottish Health Survey 2003. In The Scottish Health Survey 2003 (Vol. 1 Cardioa v). Retrieved from http://www.scotland.gov.uk/Publications/2011/09/27084018/58 Cavailles, A., Brinchault Rabin, G., Dixmier, A., Goupil, F., Gut Gobert, C., Marchand European Respiratory Review 22 (130), 454 475. http://doi.org/10.1183/09059180.00008612 CDC. (2015). Spirometry Training Program. NHANES III references values. Retrieved from http://www.cdc.gov/niosh/topics/spirometry/nhanes.html Chatila, W. M., Thomashow, B. M., Minai, O. a., Criner, G. J., & Make, B. J. (2008). Comorbidities in chronic obstructive pulmonary disease. Proceedings of the American Thoracic Society 5 (4), 549 555. http://doi.org/10.1513/pats.200709 148ET Davis, W., Knuiman, M., Kendall, P., Grange, V., & Davis, T. (2004). Glycemic Exposure Is Associated With Reduced Pulmonary Function in Type 2 Diabetes. The Fremante Diabetes Study. Diabetes Care 27 752:757.
33 Diaz Guzman, E., McCarthy, K., Siu, A., & Stoller, J. K. (2010). Frequency and causes of combined obstruction and restriction identified in pulmonary function tests in adults. Respiratory Care 55 (3), 310 316. Retrieved from http://ezproxy.fiu.edu/login?url=http://search.ebscohost.com/login.aspx?d irect=true &db=rzh&AN=2010586535&site=ehost live&scope=site Ehrlich, S., Quesenberry, C., Eeden, S., Shan, J., & Ferrara, A. (2010). Patients diagnosed with diabetes are at increased risk for asthma, chronic obstructive pulmonary disease, pulmonary fibrosi s, and pneumonia but not lung cancer. Diabetes Care 33 (1), 55 60. http://doi.org/10.2337/dc09 0880. Incalzi, R. A. (2007). The association between metabolic syndro me and restrictive ventilatory dysfunction in older persons. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 62 (7), 760 5. http://doi.org/10.1093/gerona/62.7.760 Godfrey, M. S., & Jankowich, M. D. (2016). The vital capacit y is vital epidemiology and clinical significance of the restrictive spirometry pattern. Chest 149 (1), 238 251. http://doi.org/10.1378/chest.15 1045 Goldbourt, U., Yaari, S., & Medalie, J. H. (1997). Isolated low HDL cholesterol as a risk factor for coro nary heart disease mortality. A 21 year follow up of 8000 men. Arteriosclerosis, Thrombosis, and Vascular Biology 17 (1), 107 113. http://doi.org/10.1161/01.ATV.17.1.107 Gonzales, G. F., & Tapia, V. (2013). Association of high altitude induced hypoxemia t o lipid profile and glycemia in men and women living at 4100m in the Peruvian Central Andes. Endocrinologa Y Nutricin (English Edition) 60 (2), 79 86. http://doi.org/10.1016/j.endoen.2012.06.010 Guerra, S., Sherrill, D. L., Venker, C., Ceccato, C. M., Halonen, M., & Martinez, F. D. (2010). Morbidity and mortality associated with the restrictive spirometric pattern: a longitudinal study. Thorax 65 (6), 499 504. http://doi.org/10.1136/thx.2009.12605 2 Gulcan, E., Bulut, I., Toker, A., & Gulcan, A. (2009). Evaluation of glucose tolerance status in patients with asthma bronchiale. of the Association for the Care of Asthma 46 (2), 207 209. http://doi.org/10.1080/0 2770900802627302 Hickson, D., Burchfiel, C., Lui, J., Petrini, M., Harrison, K., White, W., & Sarpong, D. (2011). Diabetes, Impaired Glucose Tolerance, and Metabolic Biomarkers in Individuals with Normal Glucose Tolerance are Inversely Associated with Lun g Function: The Jackson Heart Study. Lung 189 (4), 520 529. http://doi.org/doi:10.1007/s00408 011 9296 1
34 I., A. A. E. A., Hamdy, G., Amin, M., & Rashad, A. (2013). Pulmonary function changes in diabetic lung. Egyptian Journal of Chest Diseases and Tuberculosis 62 (3), 513 517. http://doi.org/10.1016/j.ejcdt.2013.07.006 IDF. (2013). Diabetes Atlas International Diabetes Federation King, Hilary. Aubert, Ronald. Herman, W. (1998). Global Burden of Diabetes, 1995 2025. Diabetes Care 21 (9), 1414 143 1. http://doi.org/10.2337/diacare.21.9.1414 Kurth, L., & Hnizdo, E. (2015). Change in prevalence of restrictive lung impairment in the U.S. population and associated risk factors: the National Health and Nutrition Examination Survey (NHANES) 1988 1994 and 2007 2010. Multidisciplinary Respiratory Medicine 10 (1), 7. http://doi.org/10.1186/s40248 015 0003 6 P. (1989). Diabetes mellitus, plasma glucose and lung function in a cross sectional population study. European Respiratory Journal 2 (1), 14 19. Laniado Laborin, R. (2009). Smoking and chronic obstructive pulmonary disease (COPD). Parallel epidemics of the 21st century. International Journal of Environmental Research a nd Public Health 6 (1), 209 224. http://doi.org/10.3390/ijerph6010209 Lawlor, D. A., Ebrahim, S., & Davey Smith, G. (2005). Association of birth weight with a meta analysis. Thorax 60 (10), 851 8. http://doi.org/10.1136/thx.2005.042408 (2009). Lung function impairment and metabolic syndrome the critical role of abdominal obesity. American Jour nal of Respiratory and Critical Care Medicine 179 (6), 509 516. http://doi.org/10.1164/rccm.200807 1195OC Litonjua, A. A., Lazarus, R., Sparrow, D., DeMolles, D., & Weiss, S. T. (2005). Lung function in type 2 diabetes: The Normative Aging Study. Respiratory Medicine 99 (12), 1583 1590. http://doi.org/10.1016/j.rmed.2005.03.023 Mlaga, G., Zevallos Palacios, C., De Los, M., Lazo, ., & Huayanay, C. (2010). Elevada Frecuencia De Dislipidemia Y Glucemia Basal Alterada En Una Poblacin Peruana De Alt ura* High Frequency of Dyslipidemia and Impaired Fasting Glycemia in a High Altitude Peruvian Population. Rev Peru Med Exp Salud Publica 27 (4), 557 61. Mannino, D., Buist, A., Petty, T., Enright, P., & Redd, S. (2003). Lung function and mortality in the United States: data from the First National Health and Nutrition Examination Survey follow up study. Thorax 58 388 393.
35 Mannino, D., McBurnie, M., Tan, W., Kocabas, A., Anto, J., Vollmer, W., & Buist, A. (2012). Restricted spirometry in the burden of lu ng disease study. International Journal of Tuberculosis and Lung Disease 16 (10), 1405 1411. http://doi.org/10.5588/ijtld.12.0054 Mansi, R., Joshi, S. V., Pandloskar, S. R., & Dhar, H. L. (2007). Correlation between blood sugar, cholesterol and asthma sta tus. Indian Journal of Allergy, Asthma and Immunology 21 (1), 31 34. McClean, K. M., Kee, F., Young, I. S., & Elborn, J. S. (2008). Obesity and the lung: 1. Epidemiology 1. Thorax 63 (1468 3296 (Electronic)), 649 654. http://doi.org/http://dx.doi.org/10.1 136/thx.2007.086801 McKeever, T. M., Weston, P. J., Hubbard, R., & Fogarty, A. (2005). Lung function and glucose metabolism: An analysis of data from the Third National Health and Nutrition Examination Survey. American Journal of Epidemiology 161 (6), 546 556. http://doi.org/10.1093/aje/kwi076 ATS/ERS Task Force. (2005). General considerations for lung function testing. The of the European Society for Clinical Respiratory Physiology 26 (1), 153 161. http://doi.org/10.1183/09031936.05.00034505 Wagner, J. (2005). Standardisation of spirometry. European Respiratory Journal 26 (2), 319 338. http://doi.org/10.1183/09031936.05.00034805 Mirabelli, M. C., Preisser, J. S., Loehr, L. R., Agarwal, S. K., Barr, R. G., Couper, D. J., u p among black and white adults in the ARIC study cohort. Respiratory Medicine 113 57 64. http://doi.org/10.1016/j.rmed.2016.02.003 Mohamed, E. I., Maiolo, C., Iacopino, L., Pepe, M., Di Daniele, N., & De Lorenzo, A. (2002). The impact of body weight components on forced spirometry in healthy Italians. Lung 180 (3), 149 159. http://doi.org/10.1007/s004080000089 NCD RisC. (2016). Worldwide trends in diabetes since 1980: a pooled analysis of 751 population based studies with 4 4 million participants. The Lancet 387 (10027), 1513 1530. http://doi.org/10.1016/S0140 6736(16)00618 8 Oh, H., Park, J. H., Lee, C. H., & Park, J. S. (2015). The association of normal range glycated hemoglobin with restrictive lung pattern in the general population. PLoS ONE 10 (2), 1 10. http://doi.org/10.1371/journal.pone.0117725
36 Wanger, J. (2005). Interpretative strategies for lung function tests. European Respiratory Journal 26 (5), 948 968. http://doi.org/10.1183/09031936.05.00035205 Ghirlanda, G. (2012). R EVIEW The Diabetic Lung 35. http://doi.org/10.1900/RDS .2011.9.23 Milledge, J. S. (1997). Hypoxia, hypocapnia and spirometry at altitude. Clinical 92 (6), 593 598. Riyami, N. B. Al, Banerjee, Y., Al Waili, K., Rizvi, S. G., Al Yahyaee, S., Hassan, M. O., Bayoumi, R. a. (2014). The Effect of Residing Altitude on Levels of High Density Lipoprotein Cholesterol: A Pilot Study From the Omani Arab Population. Angiology 66 (6), 568 73. http://doi.org/10.1177/0003319714544355 Rodriguez, C. J., Daviglus, M. L., Swett, K., Gonz??lez, H. M., Gallo, L. C., Wassertheil Hispanics/Latinos of diverse background in the United States American Journal of Medicine 127 (12), 1186 1194.e1. http://doi.org/10.1016/j.amjmed.2014.07.026 Zamboni, M. (2008). Body composition and pulmonary function in the elde rly: a 7 year longitudinal study. International Journal of Obesity 32 (9), 1423 1430. http://doi.org/10.1038/ijo.2008.103 Scarlata, S., Costanzo, L., Giua, R., Pedone, C., & Incalzi, R. A. (2012). Diagnosis and prognostic value of restrictive ventilatory disorders in the elderly: A systematic review of the literature. Experimental Gerontology 47 (4), 281 289. http://doi.org/10.1016/j.exger.2012.02.001 Shawn, A., Dales, R., & Cardinal, P. (1999). How accurate is spirometry at predicting restrictive pulmonary imapirment. Chest 115 (i3), 869. Tamimi, A., Serdarevic, D., & Hanania, N. A. (2012). The effects of cigarette smoke on airway inflammation in asthma and COPD: Therapeuti c implications. Respiratory Medicine 106 (3), 319 328. http://doi.org/10.1016/j.rmed.2011.11.003 Valenzuela, M., & Ramos, E. (2004). Medicion de la Capacidad Vital Forzada por Espirometria en habitantes adultos naturales de Junin. (4105m.s.n.m). Revista d e La Sociedad Peruana de Neumologia 48 (2).
37 (2013). Changes in cardiovascular functions, lipid profile, and body composition at high altitude in two different ethnic gr oups. High Altitude Medicine Biology 14 (1), 45 52. http://doi.org/10.1089/ham.2012.1071 Verdier, C., Martinez, L. O., Ferrires, J., Elbaz, M., Genoux, A., & Perret, B. (2013). Targeting high density lipoproteins: Update on a promising therapy. Archives of Cardiovascular Diseases 106 (11), 601 611. http://doi.org/10.1016/j.acvd.2013.06.052 Association between glycemic state and lung function: The Framingham He art Study. American Journal of Respiratory and Critical Care Medicine 167 (6), 911 916. http://doi.org/10.1164/rccm.2203022 Wannamethee, S., Shaper, A., & Whincup, P. (2005). Body fat distribution, body composition, and respiratory function in elderly men The American Journal of Clinical Nutrition 82 996 1003. Retrieved from http://discovery.ucl.ac.uk/162357/ Whittaker, a L., Sutton, a J., & Beardsmore, C. S. (2005). Are ethnic differences in lung function explained by chest size? Archives of Disease in Childhood. Fetal and Neonatal Edition 90 (5), F423 8. http://doi.org/10.1136/adc.2004.062497 Wood, S., Norboo, T., Lilly, M., Yoneda, K., & Eldridge, M. (2003). Cardiopulmonary function in high altitude residents of Ladakh. High Altitude Medicine and Biology 4 (4), 445 454. Retrieved from http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L3 8072352 \ nhttp://dx.doi.org/10.1089/152702903322616191 \ nhttp://sfx.umd.edu/hs?s id=EMBASE&issn=15270297&id=doi:10.1089/152702903322616191&atitle =Cardio pulmonary+function+in+high+alti Kohno, N. (2013). Cross sectional and prospective study of the association between lung function and prediabetes. BMJ Open 3 (2), 1 7. http://doi.org/10.1136/bmjopen 2012 002179 Yamashita, S., Maruyama, T., Hirano, K. I., Sakai, N., Nakajima, N., & Matsuzawa, Y. (2000). Molecular mechanisms, lipoprotein abnormalities and atherogenicity of hyperalphalipoproteinemia. Atherosclerosis 152 (2 ), 271 285. http://doi.org/10.1016/S0021 9150(00)00574 8 Yeh, H. C., Punjabi, N. M., Wang, N. Brancati, F. L. (2008). Cross Sectional and Prospective Study of Lung Function in Adults With Type 2. Diabetes Care 31 (4). http://doi.org/10.2337/dc07 1464.Abbreviations
38 Zammit, C., Liddicoat, H., Moonsie, I., & Makker, H. (2010). Obesity and respiratory diseases. International Journal of General Medicine 3 335 343. http://doi.org/10.2147/IJGM.S11926
39 BIOGR APHICAL SKETCH Katherine A. De la Torre is a native of Quito, Ecuador. She received a Medical Degree from the Pontifical Catholic University of Ecuador in 2012 Additionally she graduated from the University of F lorida s Department of Epidemiology master s program in summer 20 16 Her area of interested is in epidemiology of non communicable and endocrinology diseases.