Introduction

Chronic obstructive pulmonary disease (COPD) is a common chronic and debilitating inflammatory condition of the airways clinically characterized by respiratory symptoms and irreversible airflow limitation. Epidemiologic studies suggest that COPD is associated with significant morbidity and mortality, with a total of 3.3 million deaths reported in 2019, which is expected to increase to 4.5 million in 2030.1 Although the exact pathogenesis of COPD remains largely unknown, mounting evidence suggests that unrelenting inflammation, both from environmental exposures and smoking, plays an indispensable role in its pathogenesis.2 For example, C-reactive protein, a frequently used biomarker of inflammation, has previously been shown to be correlated with COPD severity and to predict mortality risk.3

Platelets have been identified as crucial modulators of inflammatory responses and contribute to the development of COPD through multiple mechanisms, including modulation of the hypoxia signaling pathway and breakdown of lung elastin by platelet factor 4.4 In addition, platelets have also been shown to contribute to COPD by mediating immune responses and thrombotic pathways.5 However, the high-density lipoprotein (HDL) cholesterol has been shown to possess antiplatelet, anti-thrombotic, and anti-inflammatory properties.6 Previous studies have yielded conflicting results regarding HDL levels in patients with COPD.7 The platelet/HDL ratio (PHR), a novel biomarker that may reflect the balancing role of platelets and HDL for inflammation and hypercoagulability, was first proposed by Jialal’s group as a valid biomarker for nascent metabolic syndrome.8 Subsequent studies have also indicated that PHR is associated with several conditions, such as nephrolithiasis and non-alcoholic fatty liver disease.9,10 The study by Zhang et al showed that a PHR greater than 223.684 is positively associated with the prevalence of stroke, and PHR is linearly indicative of cardiovascular mortality in patients with stroke.11 However, there is a paucity of data regarding the association between PHR and COPD prevalence.

Therefore, the aim of the present study was to examine the association between PHR and COPD prevalence using cross-sectional data from the US National Health and Nutrition Examination Survey (NHANES) and death certificate data linked to the National Death Index.

MethodsData Source and Subjects

The NHANES is a nationwide survey that was conducted every 2 years to collect information about the health and nutritional status of the non-institutionalized US population. The data were collected through face-to-face interviews, physical examinations, and laboratory tests. The detailed survey protocol is recorded and readily available on the website: https://wwwn.cdc.gov/nchs/nhanes/Default.aspx. The NHANES study was approved by the Institutional Review Board of the National Center for Health Statistics, and informed consent was obtained from all adult participants. According to the national legislation, ethical approval for medical researches involving human subjects can be waived if the research uses publicly available data with an observational design or uses anonymized data (https://www.gov.cn/zhengce/zhengceku/2023-02/28/content_5743658.htm). This study used publicly available NHANES data and all the identifying information in NHANES has been de-identified, so ethical approval was exempted by the Ethics Committee of the Shanghai Tenth Hospital.

In this study, we used the NHANES data across 10 cycles of the survey from 1999–2018. Subjects aged between 40 and 85 years were included. Exclusion criteria were pregnancy (n = 32), missing data on COPD (n = 77), missing data on complete blood count or HDL level (n = 4016), and incomplete data on other covariates (n = 6376). The final analytic sample included 25751 participants with complete data on the primary exposure and the outcome. The detailed process of participant selection and exclusion is shown in Figure 1.

Figure 1 Study flowchart showing the process of participants inclusion and exclusion.

Abbreviation: PSM, propensity score matching.

Criteria for COPD

Participants were considered to have COPD if responded yes to the question “Has a doctor or other health professional ever told you that you had COPD?”. This definition was in accordance with previous studies in terms of COPD.12,13

Blood Lipids and Cell Count Determination

Venous blood samples were collected after an overnight fast for at least 9 hours. The HDL concentrations were measured using the direct precipitation or immunoassays. The PHR was calculated as the ratio of the platelet count (103 cells/μL) to the HDL concentration (mg/dL).9–11

Covariates

Covariates indicate factors that are relevant or potentially confounding according to existing literature and clinical judgment. In this study, the following covariates information were collected. Participants’ socioeconomic and demographics, including age (continuous), sex (male/female), race (non-Hispanic white, non-Hispanic black, Mexican American, other Hispanic, and other races), marital status (single, nonsingle), education level (< high school, high school, and > high school), and poverty–income ratio were collected at the time of the questionnaire. In addition, physical measures and lifestyle factors, including smoking status (never, former, current), use of alcohol (never, former, mild, moderate, and heavy), and body mass index (BMI), were also obtained. The detailed criteria for categorizing smoking and alcohol consumption were consistent with those previously reported.14 Information on concurrent cardiovascular disease, including angina, myocardial infarction, coronary heart disease, congestive heart failure, and stroke, was also acquired via questionnaire. Kidney function assessment with estimated glomerular filtration rate was calculated based on serum creatinine.15

Statistical Analysis

All the statistical analyses were weighted to account for the complex, multistage, probability sampling design of NHANES and to obtain nationally representative estimates. Variables were expressed as means ± standard errors or counts (weighted percentages) and compared using the independent Student’s t-test or Chi-Squared test, as appropriate. To minimize the impact of confounding variables between the 2 groups, a 1:2 propensity score-matching (PSM) for participants’ age, sex, and race was performed. Binary logistic regression was used to estimate the association between PHR and COPD prevalence. During this process, the odds ratio (OR) and 95% confidence interval (CI) were calculated. In addition, this association was also analyzed based on PHR quartiles, with the lowest Q1 as the reference group. Three models in total were constructed: the crude model was unadjusted; model 1 was adjusted for age, sex, race, marital status, education level, and poverty–income ratio; and the model 2 was further adjusted for body mass index, drinking, smoking, and comorbidities of hypertension, diabetes, and cardiovascular diseases. Restricted cubic spline analysis with fully adjusted logistic regression was used to determine the dose–response associations between PHR and COPD risk. Stratified analysis was performed to test whether the relationship between PHR and COPD risk differed by age, sex, race, hypertension, and diabetes. Receiver-operating characteristic (ROC) curves were generated to determine and compare the capabilities of platelet counts, HDL, and PHR for distinguishing COPD from non-COPD. Two-tailed P<0.05 denoted statistical significance.

ResultsBaseline Participant Characteristics

A total of 25751 participants, including 753 with COPD and 24998 without COPD, with a mean age of 57.19 years and 47.83% men, were included. As shown in Table 1, individuals with COPD exhibited significantly higher age, a higher prevalence of non-Hispanic white and single status, less educational attainment, a lower poverty–income ratio, and a higher prevalence of smoking and medical comorbidities of hypertension, diabetes, and cardiovascular disease. In terms of laboratory tests, participants with COPD had significantly lower estimated glomerular filtration rate and higher PHR than participants without COPD. After PSM, the age difference between the COPD group and the non-COPD group was drastically reduced, and there were no longer any significant differences in race/ethnicity composition, diabetes prevalence, and estimated glomerular filtration rate between the two groups.

Table 1 Characteristics of Study Participants Before and After Propensity Score Matching

Associations Between PHR and COPD Prevalence

Weighted logistic regression (Table 2) identified a significant association between PHR and the prevalence of COPD, even in the fully adjusted model 2 (OR = 1.002, 95% CI 1.001–1.003). Compared with the reference Q1 group, the ORs, and 95% CIs for the Q2, Q3, and Q4 PHR quartile were 1.162 (0.874–1.546), 1.225 (0.924–1.625), and 1.510 (1.102–2.069), respectively (P for trend = 0.012). Notably, a significant association between PHR and COPD was still observed in the logistic regression analysis after matching on propensity scores, as shown in Table 3.

Table 2 Associations Between Platelet/High-Density Lipoprotein Cholesterol Ratio (PHR) and Risk of Chronic Obstructive Pulmonary Disease Before Propensity Score Matching

Table 3 Associations Between Platelet/High-Density Lipoprotein Cholesterol Ratio (PHR) and Risk of Chronic Obstructive Pulmonary Disease After Propensity Score Matching

Restricted cubic spline analysis (Figure 2) was performed to evaluate the dose–response relationship between PHR and COPD prevalence. The results showed that the PHR was positively associated with COPD both before (P for nonlinearity = 0.080) and after (P for nonlinearity = 0.135) PSM.

Figure 2 Restricted cubic spline analysis of the multivariable-adjusted association between PHR values and the prevalence of chronic obstructive pulmonary disease before (A) and after (B) propensity score matching.

Subgroup Analysis

Figure 3 shows the results of subgroup analyses stratified by age, sex, race, hypertension, and diabetes status before and after PSM. The associations between PHR and COPD did not differ among participants of different age, sex, race, and diabetes status. Notably, the significant association between PHR and COPD prevalence was observed only in participants without hypertension both before and after PSM.

Figure 3 Forest plot showing the results of subgroup analysis for the association between PHR and chronic obstructive pulmonary disease before and after propensity score matching.

We then compared data from participants with and without hypertension. The results indicated that compared with those without hypertension, participants with hypertension had significantly lower HDL (1.43 ± 0.01 mmol/L vs 1.37 ± 0.01 mmol/L, P < 0.001), higher COPD prevalence (3.53% vs 1.73%, P < 0.001), and statistically insignificant platelet count (251.29 ± 1.00 per 1000 cells/μL vs 249.40 ± 0.99 per 1000 cells/μL, P = 0.109).

ROC Curve Analysis

We compared the capabilities of platelets, HDL, and PHR for discriminating COPD from non-COPD both before and after PSM. As shown in Figure 4 and Table 4, platelets and HDL could not distinguish COPD from non-COPD participants both before and after PSM. In comparison, the discriminative ability of PHR, as indicated by the area under the curve, was significantly higher than that of the platelets and HDL both before and after PSM.

Table 4 Receiver-Operating Characteristic Curve Results for the Discrimination of COPD from Non-COPD

Figure 4 Receiver-operating characteristic curves for the discrimination of COPD from non-COPD by platelet count, HDL, and PHR before (A) and after (B) propensity score matching.

Abbreviations: COPD, chronic obstructive pulmonary disease; HDL, high-density lipoprotein cholesterol; PHR, platelet count to HDL ratio.

Discussion

The present cross-sectional study examined the association between PHR and prevalence of COPD using a nationally representative, population-based cohort from the NHANES. We found that a higher PHR was linearly associated with increased prevalence of COPD, which was further supported by the restricted cubic spline analysis. In addition, hypertension status significantly modified this relationship, as significant association between PHR and COPD prevalence was observed only in participants without hypertension. The ROC results indicated superior performance of PHR over platelet counts and HDL for distinguishing COPD from non-COPD. Subsequent PSM further confirmed the robustness and validity of the study results.

In the current study, COPD patients were associated with certain socio-demographic characteristics even after PSM, such as lower educational level, lower poverty–income ratio, and higher prevalence of being single. This observation is consistent with the fact that smoking, which is an important risk factor for the development of COPD, is more prevalent in socioeconomically disadvantaged individuals. This finding also echoed a previous study showing that higher educational level, higher annual household income, and access to medical services were associated with a lower risk of COPD.16

Comparison of the COPD and non-COPD groups showed no statistical difference in HDL levels, while significantly higher platelet levels were observed in the COPD group. Burkart et al demonstrated that a higher HDL is associated with worse lung function indicated by lower forced expiratory volume at 1 second/forced vital capacity ratio, and greater percent emphysema, highlighting that an elevated HDL may instead increase the risk of COPD.17 A multicenter study indicated that a lower serum HDL level may reduce forced expiratory volume at 1 second in patients with COPD.18 Congruently, Markelić et al showed an elevated level of HDL in COPD patients, especially in those with advanced stages.19 However, the meta-analysis of 615 COPD and 471 controls showed no significant differences in HDL levels between COPD and controls.7 The work by Wen et al indicated a non-linear association between HDL and lung function in COPD, with HDL elevation before 66 mg/dl was associated with emphysema and impaired lung function.20 HDL has been found to be implicated in the inflammatory activation of alveolar macrophages and the induction of neutrophil proliferation,21 both of which significantly contribute to airway inflammation and thus increased risk for the development of chronic lung conditions like COPD. Another theory holds that HDL could modulate type II alveolar cells and thus alters the surface characteristics of pulmonary surfactant.22 In fact, confounding factors of steroid use and the pro-inflammatory nature of HDL in chronic conditions have also been employed to explain the relationship between HDL and COPD.23

Available reports have consistently shown an elevated platelet count in patients with COPD. For example, Zinellu’s group performed a meta-analysis and found that COPD patients had significantly higher platelet counts than non-COPD patients, which is consistent with our findings.5 It has also been observed that platelet count is positively associated with COPD severity.24 Platelets were implicated in the development and exacerbation of COPD through multiple mechanisms, including destruction of lung elasticity by secreting platelet factor 4, and induction of a pro-thrombotic state and pulmonary vascular remodeling.25 Consequently, a significantly elevated platelet count has also been associated with an increased risk of all-cause mortality, and antiplatelet therapy with aspirin may improve COPD symptoms and quality of life.26

Logistic regression analysis showed that PHR, both as a continuous and categorical variable, was significantly associated with COPD prevalence even after adjustment for potential confounders and after PSM. In addition, the ROC curve analysis indicated superior performance of PHR over platelet counts and HDL for distinguishing COPD from non-COPD. The PHR, which combines the predictive value of circulating platelet count and HDL into a single ratio, has previously been reported to be associated with cardiovascular mortality in stroke and all-cause mortality in patients with coronary artery disease in longitudinal studies.11,27 More importantly, the PHR has demonstrated superior utility in predicting cardiovascular mortality risk than platelet and HDL alone, which is compatible with our results.11 In addition, the elevated PHR has also been cross-sectionally associated with an increased prevalence of nephrolithiasis, non-alcoholic liver disease, and acute mania.9,10,28 HDL has been shown to counteract the pro-coagulation effect of platelets by stimulating clot fibrinolysis.29 Thus, an elevated PHR may indicate pro-coagulation status, which has been implicated in the development of airflow obstruction and emphysema. Therefore, the PHR may serve as a convenient, readily available and informative biomarker for COPD.

An interesting finding of this study is that the association between PHR and COPD was only present in those without hypertension. Comparison of participants with and without hypertension showed similar platelet counts, and the higher PHR level in participants with hypertension was primarily contributed by the significantly lower HDL levels in this group. Our main findings suggest that platelet count, rather than HDL, may be the main contributor to the association between PHR and COPD prevalence. Notably, these results indicate that PHR may be discordantly associated with COPD. Consequently, clinical attention should be tailored to specific subgroups to provide more guidance to participants without hypertension. It is likely that there are additional operating mechanisms underlying this difference in COPD patients with and without hypertension that are worthy of further investigation in the future.

Strengths of this study include its national representativeness and large sample size, coupled with PSM controlling for age, sex, and race. Previous studies have shown that people of different sexes and races not only have a completely different risk of COPD but also a disproportionate mortality rate, especially in African Americans.30,31 However, this study also suffers from several limitations that should be acknowledged. First, the observational cross-sectional design renders causal interpretation between PHR and COPD difficult, for which longitudinal cohort studies are needed. Second, the diagnosis of COPD in the current study was self-reported, which may introduce recall bias and defy classification of COPD severity. Documentation of forced expiratory volume in 1 second/forced vital capacity < 70% after bronchodilation would probably be more objective. Finally, it appears that we failed to account for COPD severity and acute exacerbation status in this study, as both have been reported to affect platelet count and HDL levels.

Conclusion

In conclusion, this cross-sectional study revealed a higher PHR is associated with increased prevalence of COPD in participants without hypertension. Thus, PHR serves as a potential biomarker for COPD that deserves additional investigation. Longitudinal cohort studies are needed to establish this relationship.

Data Sharing Statement

The data that support the findings of this study are available on request from the corresponding author upon reasonable request.

Ethics Statement

This study uses data from a publicly available database, in which all study participants gave informed consent in accordance with the Institutional Review Board and study ethic guidelines at the Centers for Disease Control and Prevention. Ethical approval for the current study was exempted from the Ethics Committee of Shanghai Tenth People’s Hospital.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising, or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This study was funded by the 2020 Shanghai Association of Integrated Traditional Chinese and Western Medicine Community Medicine and Health Management Research Project (No. 2020-11) granted to Dr. Min Tan.

Disclosure

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

1. Safiri S, Carson-Chahhoud K, Noori M, et al. Burden of chronic obstructive pulmonary disease and its attributable risk factors in 204 countries and territories, 1990-2019: results from the global burden of disease study 2019. BMJ. 2022;378:e069679. doi:10.1136/bmj-2021-069679

2. King PT. Inflammation in chronic obstructive pulmonary disease and its role in cardiovascular disease and lung cancer. Clin Transl Med. 2015;4(1):68. doi:10.1186/s40169-015-0068-z

3. Leuzzi G, Galeone C, Taverna F, Suatoni P, Morelli D, Pastorino U. C-reactive protein level predicts mortality in COPD: a systematic review and meta-analysis. Eur Respir Rev. 2017;26(143):160070. doi:10.1183/16000617.0070-2016

4. Mallah H, Ball S, Sekhon J, Parmar K, Nugent K. Platelets in chronic obstructive pulmonary disease: an update on pathophysiology and implications for antiplatelet therapy. Respir Med. 2020;171:106098. doi:10.1016/j.rmed.2020.106098

5. Zinellu A, Paliogiannis P, Sotgiu E, et al. Platelet count and platelet indices in patients with stable and acute exacerbation of chronic obstructive pulmonary disease: a systematic review and meta-analysis. COPD. 2021;18(2):231–245. doi:10.1080/15412555.2021.1898578

6. Namiri-Kalantari R, Gao F, Chattopadhyay A, et al. The dual nature of HDL: anti-Inflammatory and pro-Inflammatory. Biofactors. 2015;41(3):153–159. doi:10.1002/biof.1205

7. Xuan L, Han F, Gong L, et al. Association between chronic obstructive pulmonary disease and serum lipid levels: a meta-analysis. Lipids Health Dis. 2018;17(1):263. doi:10.1186/s12944-018-0904-4

8. Jialal I, Jialal G, Adams-Huet B. The platelet to high density lipoprotein -cholesterol ratio is a valid biomarker of nascent metabolic syndrome. Diabetes Metab Res Rev. 2021;37(6):e3403. doi:10.1002/dmrr.3403

9. Ni J, Lv L, Wu P, Xu C. Associations between the platelet/high-density lipoprotein cholesterol ratio and likelihood of nephrolithiasis: a cross-sectional analysis in United States adults. Front Endocrinol. 2024;15:1289553. doi:10.3389/fendo.2024.1289553

10. Lu CF, Cang XM, Liu WS, et al. Association between the platelet/high-density lipoprotein cholesterol ratio and nonalcoholic fatty liver disease: results from NHANES 2017-2020. Lipids Health Dis. 2023;22(1):130. doi:10.1186/s12944-023-01861-9

11. Zhang H, Xu Y, Xu Y. The association of the platelet/high-density lipoprotein cholesterol ratio with self-reported stroke and cardiovascular mortality: a population-based observational study. Lipids Health Dis. 2024;23(1):121. doi:10.1186/s12944-024-02115-y

12. Zheng L, Yu X, Xia Z, Guo Y, Dai Y. The associations between serum vitamins and carotenoids with chronic obstructive pulmonary disease: results from the NHANES. Int J Chron Obstruct Pulmon Dis. 2023;18:2985–2997. doi:10.2147/COPD.S432995

13. Shi Y, Xu Z, Pu S, Xu K, Wang Y, Zhang C. Association between serum klotho and chronic obstructive pulmonary disease in us middle-aged and older individuals: a cross-sectional study from NHANES 2013-2016. Int J Chron Obstruct Pulmon Dis. 2024;19:543–553. doi:10.2147/COPD.S451859

14. Zhang Q, Xiao S, Jiao X, Shen Y. The triglyceride-glucose index is a predictor for cardiovascular and all-cause mortality in CVD patients with diabetes or pre-diabetes: evidence from NHANES 2001-2018. Cardiovasc Diabetol. 2023;22(1):279. doi:10.1186/s12933-023-02030-z

15. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–612. doi:10.7326/0003-4819-150-9-200905050-00006

16. Cai L, Wang XM, Fan LM, Shen JR, Liu YN, Golden AR. Socioeconomic variations in chronic obstructive pulmonary disease prevalence, diagnosis, and treatment in rural Southwest China. BMC Public Health. 2020;20(1):536. doi:10.1186/s12889-020-08687-5

17. Burkart KM, Manichaikul A, Wilk JB, et al. APOM and high-density lipoprotein cholesterol are associated with lung function and per cent emphysema. Eur Respir J. 2014;43(4):1003–1017. doi:10.1183/09031936.00147612

18. Qaisar R, Karim A, Muhammad T. Circulating biomarkers of handgrip strength and lung function in chronic obstructive pulmonary disease. Int J Chron Obstruct Pulmon Dis. 2020;15:311–321. doi:10.2147/COPD.S225765

19. Markelić I, Hlapčić I, Rogić D, et al. Lipid profile and atherogenic indices in patients with stable chronic obstructive pulmonary disease. Nutr Metab Cardiovasc Dis. 2021;31(1):153–161. doi:10.1016/j.numecd.2020.07.039

20. Wen X, Wu X, Deng Z, et al. The nonlinear relationship between high-density lipoprotein and changes in pulmonary structure function and pulmonary function in COPD patients in China. Int J Chron Obstruct Pulmon Dis. 2024;19:1801–1812. doi:10.2147/COPD.S467976

21. van der Vorst EPC, Theodorou K, Wu Y, et al. High-density lipoproteins exert pro-inflammatory effects on macrophages via passive cholesterol depletion and PKC-NF-κB/STAT1-IRF1 signaling. Cell Metab. 2017;25(1):197–207. doi:10.1016/j.cmet.2016.10.013

22. Yu Z, Jin J, Wang Y, Sun J. High density lipoprotein promoting proliferation and migration of type II alveolar epithelial cells during inflammation. State Lipids Health Dis. 2017;16(1):91. doi:10.1186/s12944-017-0482-x

23. Reed RM, Iacono A, DeFilippis A, Eberlein M, Girgis RE, Jones S. Advanced chronic obstructive pulmonary disease is associated with high levels of high-density lipoprotein cholesterol. J Heart Lung Transplant. 2011;30(6):674–678. doi:10.1016/j.healun.2010.12.010

24. Moniruzzaman M, Karim MR, Ahamed F, et al. Platelet count as a severity of chronic obstructive pulmonary disease. Mymensingh Med J. 2020;29(2):241–247.

25. Chebbo M, Duez C, Alessi MC, Chanez P, Gras D. Platelets: a potential role in chronic respiratory diseases? Eur Respir Rev. 2021;30(161):210062. doi:10.1183/16000617.0062-2021

26. Fawzy A, Putcha N, Aaron CP, et al. Aspirin use and respiratory morbidity in COPD: a propensity score-matched analysis in subpopulations and intermediate outcome measures in COPD study. Chest. 2019;155(3):519–527. doi:10.1016/j.chest.2018.11.028

27. Wu WY, Jia CZ, Xu XY, et al. Impact of platelet-to-HDL-cholesterol ratio on long-term mortality in coronary artery disease patients with or without type 2 diabetes: insights from a Chinese multicenter cohort. J Inflamm Res. 2024;17:2731–2744. doi:10.2147/JIR.S458950

28. Korkmaz ŞA, Kızgın S. Neutrophil/high-density lipoprotein cholesterol (HDL), monocyte/HDL and platelet/HDL ratios are increased in acute mania as markers of inflammation, even after controlling for confounding factors. Curr Med Res Opin. 2023;39(10):1383–1390. doi:10.1080/03007995.2023.2260302

29. van der Stoep M, Korporaal SJ, Van Eck M. High-density lipoprotein as a modulator of platelet and coagulation responses. Cardiovasc Res. 2014;103(3):362–371. doi:10.1093/cvr/cvu137

30. Ntritsos G, Franek J, Belbasis L, et al. Gender-specific estimates of COPD prevalence: a systematic review and meta-analysis. Int J Chron Obstruct Pulmon Dis. 2018;13:1507–1514. doi:10.2147/COPD.S146390

31. Mamary AJ, Stewart JI, Kinney GL, et al. Race and gender disparities are evident in COPD underdiagnoses across all severities of measured airflow obstruction. Chronic Obstr Pulm Dis. 2018;5(3):177–184. doi:10.15326/jcopdf.5.3.2017.0145