Clinical Characteristics and Medical Utilization of Smokers with Preserved Ratio Impaired Spirometry

Introduction

Preserved ratio impaired spirometry (PRISm) is a spirometric impairment being defined as decreased forced expiratory volume in one second (FEV₁) <80% predicted with FEV₁/forced vital capacity (FVC) ≥0.7.^1–5 The prevalence of PRISm is 9–20%; however, little is known about PRISm because of the exclusion of individuals with this spirometric impairment from previous obstructive lung disease studies.^6–8

The natural history and prognosis of PRISm is highly variable.^6,7,9 In the US COPDGene cohort, 22.2% of subjects with PRISm were at risk of chronic obstructive pulmonary disease (COPD) with Global Initiative for Obstructive Lung Disease (GOLD) stage 0, and 25.1% of them progressed to GOLD stages 1–4 COPD during the 5-year follow-up period.⁶ In a UK Biobank cohort, 12.2% of subjects with PRISm developed COPD.⁹ In addition, individuals with PRISm could exhibit increased respiratory symptoms, poor health-related quality of life, multiple comorbidities, and increased mortality compared with those with normal spirometry.^6,8,10–12 However, few studies have investigated the disease burden and incidence of pulmonary complications in subjects with PRISm.

The aim of the present study was to investigate the clinical characteristics, real-world treatment pattern, medical cost, and medical utilization of smokers with PRISm using data obtained from the Korean National Health and Nutrition Examination Survey (KNHANES) linked to the Health Insurance Review and Assessment Service (HIRA). In addition, we compared differences in the clinical characteristics of smokers with PRISm according to medical utilization.

Materials and Methods

Study Population

This observational, retrospective cohort study investigated the data obtained from the KNHANES between 2007 and 2012, linked to the HIRA. The HIRA system includes medical reimbursement records for the entire Korean population. The KNHANES is a national cross-sectional surveillance system containing de-identified data on demographics, comorbidities, smoking history, and spirometry results.¹³ Sampling was conducted with a stratified multistage cluster sampling design.

We screened ex- or current smokers aged 40 year or older who underwent spirometry. Subjects who had ever smoked more than 100 cigarettes was defined as smokers. Ex-smokers were defined as smokers who quit smoking more than 3 months ago. Smokers who currently smoke or quit smoking within 3 months were defined as current smokers. Socioeconomic status was stratified into quartiles based on income level. The 2nd and 3rd quartiles were considered middle. Spirometry was performed using a rolling dry-seal spirometer (Vmax series 2130; SensorMedics Corp., Yorba Linda, CA, USA), according to the American Thoracic Society performance criteria.¹⁴ We divided the smokers into three groups based on spirometry results: normal spirometry, PRISm, and COPD. A FEV₁/FVC ratio of <0.7 was used to define COPD.¹ PRISm was defined as described above. Normal spirometry was defined as an FEV₁/FVC ≥0.7 and an FEV₁ ≥80% predicted.

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). All KNHANES participants signed an informed consent form. Open data are available at the KNHANES website (https://knhanes.cdc.go.kr). All data were anonymously managed in all stages. Ethical approval for this study was waived by the Institutional Review Board of Ewha Womans University Mokdong Hospital because this study used anonymously managed open data.

Study Design and Data Collection

We collected data on demographics, comorbidities, smoking history, and spirometry results from the KNHANES database. Based on the HIRA data between 2007 and 2012, the smoker’s hospitalization, emergency department (ED) visit, intensive care unit (ICU) admission, and prescription records were analyzed. We investigated differences in demographic characteristics, clinical characteristics, and prescribed medications related to obstructive lung disease among the three groups. Disease burden and medical utilization were evaluated using claims data on outpatient clinic visit, ED visit, hospitalization, ICU admission, and medical cost. In addition, we compared the demographic and clinical characteristics of the PRISm group divided into smokers with and without medical utilization. Smokers who visited an outpatient clinic with claim codes for the International Classification of Disease 10th edition (ICD-10) codes J43-44 except for J43.0 were included. We also included smokers with acute respiratory events who required an ED visit and hospitalization with the ICD-10 codes J43-J44 except for J43.0, J12-17, I26, I26.0, I26.9, R06.0, and J80. As we used the data available before the release of the dual long-acting muscarinic antagonist/ long-acting beta-2 agonist in South Korea, there were no data regarding on dual therapy.

Statistical Analysis

All statistical analyses were performed using SAS ver. 9.4 (SAS Institute, Cary, NC, USA). Data are expressed as means ± standard deviations or numbers (%). Continuous variables were analyzed by Student’s t-test or one-way analysis of variance. Categorical variables were analyzed by Pearson’s chi-square test. All tests for significance were two-sided, and all variables with p <0.05 were considered significant. Bonferroni correction (alpha = 0.05/3= 0.0167) was performed for multiple comparisons between the groups in post-hoc test.¹⁵

Results

Baseline Characteristics

A total of 7115 smokers were included in the present study, consisting of 4743 smokers (66.7%) in the normal group, 689 smokers (9.7%) in the PRISm group, and 1683 smokers (23.7%) in the COPD group. The mean age was 57.0 years, and 6344 smokers (89.2%) were men (Table 1). The PRISm group had the highest proportion of women and medical aid and showed the highest body mass index (BMI) and waist circumference. The number of smoking pack-years (PY) of the PRISm group (26.0 ± 20.2 PY) was significantly higher than that of the normal group (21.2 ± 18.0 PY) but lower than that of the COPD group (30.7 ± 22.9 PY, p <0.001). The proportion of smokers who reported intermittent wheezing was the second highest in the PRISm group; however, spirometry results revealed that the PRISm group had significantly reduced FEV₁ and FVC % predicted.

Table 1 Baseline Characteristics of Smokers with Normal Spirometry, Preserved Ratio Impaired Spirometry, and Obstructive Spirometry

Treatment and Medical Utilization

During the study period, 92 smokers (13.4%) in the PRISm group and 436 smokers (25.9%) in the COPD group visited hospital for the treatment of respiratory diseases. The various types of respiratory medicines taken by the PRISm group were at a level between that of the normal and COPD groups (Table 2). The outpatient clinic visit rate and cost of the PRISm group were significantly lower than those of the COPD group (p <0.001) but were increased tendency compared with those of the normal group (Table 3). On the other hand, the rates of hospitalization, ICU admission, and ED visit were not significantly different between the PRISm and COPD groups; however, these rates were significantly higher in the PRISm group than in the normal group.

Table 2 Treatment Regimens of Smokers with Normal Spirometry, Preserved Ratio Impaired Spirometry, and Obstructive Spirometry

Table 3 Medical Utilization of Smokers with Normal Spirometry, Preserved Ratio Impaired Spirometry, and Obstructive Spirometry

Medical Utilization of the PRISm Group

Of 689 smokers in the PRISm group, 92 smokers (13.4%) visited the hospital. The mean FEV₁% predicted (70.7 ± 9.1 vs 74.0 ± 5.6) and FVC % predicted (69.1 ± 10.7 vs 74.6 ± 8.4) were significantly lower in the PRISm group with medical utilization than in the PRISm group without medical utilization (Table 4). Hospital visit was associated with old age, women, low BMI, low socioeconomic status, low schooling level, higher amount of tobacco exposure, intermittent wheezing, osteoarthritis, and decreased FEV₁ and FVC % predicted. Interestingly, a higher portion of smokers were hospitalized in the PRISm group (54/92, 58.7%) than in the COPD group (201/436, 46.1%) among PRISm and COPD smokers with medical utilization (p = 0.028).

Table 4 Comparison of Clinical Characteristics of Smokers with Preserved Ratio Impaired Spirometry According to Medical Utilization

Discussion

To the best of our knowledge, this is the first study to investigate the medical utilization of smokers with PRISm and their clinical characteristics. The present study showed that the frequency of clinically significant exacerbation requiring ED visit or hospitalization was comparable between the PRISm and COPD groups.

PRISm remains poorly understood.¹ PRISm is commonly referred to as restrictive lung disease; however, 11.8% of smokers in the PRISm group reported wheezing, which suggests narrowed airways. In comparison with other ethnicities, East Asians seem to have a lower incidence of PRISm.^16,17 In the present study, the PRISm group had the highest BMI, whereas the COPD group had the lowest BMI. Given the relationship between a low obesity rate and a low incidence of PRISm among East Asians as well as the high BMI in the PRISm group, obesity may play an important role in the pathogenesis of PRISm. Leptin, secreted in direct proportion to the adipose tissue mass, exerts proinflammatory effects.^18,19 Although the functional effects of leptin on the respiratory system are less clear, elevated leptin levels may modulate the immune reaction in the airways by inciting a robust proinflammatory response or skewing the cellular response towards a type 1 helper phenotype.^20,21 Indeed, Wan et al reported that a reduction in BMI was associated with transitioning from PRISm to normal spirometry.⁶

Previous studies reported that 15–40% of subjects with PRISm progressed to COPD, and 33–50% of subjects had persistent PRISm.^2,3,6,17 We could not determine the number of smokers with PRISm who developed COPD in a few years; however, we found that PRISm also requires careful observation because of frequent exacerbation and concomitant medical utilization, comparable to those in COPD. The present study showed that the amount of tobacco exposure was associated with medical utilization. Smoking is a major factor in COPD exacerbation. Cumulative cigarette smoke exposure causes airflow limitation, accelerating the rate of decline in FEV_1.²² Parekh et al reported that current smoking and frequent exacerbation were factors contributing to decreased quality of life, which may be associated with a high burden of respiratory symptoms in PRISm.²³ Subjects with PRISm were relatively young and current smokers compared with those with COPD. Therefore, further evaluation is needed to determine whether smoking cessation can reduce the incidence of acute exacerbation or achieve normal lung function.

In this study, low FEV₁ and FVC were associated with medical utilization in PRISm. No study has evaluated the relationship between lung function and the severity of symptoms or prognosis of PRISm. However, Guerra et al reported that individuals with PRISm with low FVC had persistent PRISm over time.² A study of Peruvian adults showed that PRISm was associated with accelerated decline in FEV₁; however, those with PRISm could be classified into high-, and low-risk groups that progressed to COPD or even achieved normal spirometry.^6,24 Although Kim et al reported that lung function was not a predictive factor of COPD, decreased FEV₁ and FVC might imply severe airflow limitation and respiratory symptoms, which would require close observation and early management.⁸

The present study has several limitations. First, spirometry was conducted through the pre-bronchodilator spirometry test even though the reliability of portable spirometry has been proven.²⁵ The lack of post-bronchodilator spirometry may have resulted in an overestimate of the prevalence of both PRISm and COPD. However, although bronchodilator administration has been shown to reduce the prevalence of obstructive lung disease, its effect on the prevalence of PRISm is less well-established. Comorbidities such as heart failure, combined interstitial lung disease, and intermediate tuberculosis prevalence might contribute to the prevalence of PRISm and associated adverse respiratory outcomes; however, we could not examine these comorbidities comprehensively. Second, the European Respiratory Society/ American Thoracic Society technical standard on interpretative strategies for lung function tests recommend to use of z-score of FEV₁/FVC to determine impaired lung function.²⁶ We did not use the z-score, however, a previous study showed that both the fixed ratio and the z-score of FEV₁/FVC showed comparable prediction performance for the 10-year respiratory and COPD mortalities in elderly patients.²⁷ Third, we excluded never smokers; thus, the effects of genetic factors as well as exposure to second-hand smoke and other non-tobacco environmental exposures were not evaluated. Considering that PRISm subjects are known to be young and that a study reported an association between history of childhood asthma and PRISm, lung growth might be an important pathogenic factor underlying one of various phenotypes influencing symptoms, severity, and longitudinal trajectories.²⁸ Fourth, other risk factors of exacerbation, such as the exacerbation history before the study period, comorbidities, infection, and compliance with medications, were not fully considered. Fifth, analyses were limited to the examination of the relationship between cross-sectional PRISm and clinical outcomes; thus, we could not determine whether distinct longitudinal trajectories within PRISm may be differentially associated with clinical outcomes.

Conclusion

The clinical characteristics and medical utilization of the PRISm group were comparable to those of the COPD group. Smokers with PRISm who were older, women, or heavy smokers with low BMI, low socioeconomic status and schooling level, wheezing, or low FEV₁ and FVC might need close observation and early treatment. Further studies on appropriate management and medical treatment for individuals with PRISm are needed to improve their respiratory symptoms and prognosis.