Assessment of comorbidities and prognosis in patients with COPD diagnosed with the fixed ratio and the lower limit of normal: a systematic review and meta-analysis

Chronic obstructive pulmonary disease (COPD) is a non-curable disease with high morbidity and mortality [1]. Globally, it is the 3rd leading cause of death [2]. In America in the 2000s, the prevalence of COPD was 24 million, and half of these were undiagnosed [3]. In 2015, 99.9 million Chinese were diagnosed with COPD, which accounted for 8.6% of the total population [4]. Patients with COPD always have respiratory symptoms and often have comorbidities [5]. Moreover, several co-morbidities of COPD, such as cardiovascular disease and diabetes mellitus have been reported to exacerbate the mortality of COPD [6, 7].

COPD guidelines recommend the use of post-bronchodilator forced expiratory volume in 1 s (FEV₁)/ forced vital capacity (FVC) ratio to diagnose patients with chronic respiratory symptoms or those at risk [1, 8, 9]. However, the American Thoracic Society (ATS) and European Respiratory Society (ERS) recommend the use of a threshold below the lower limit of normal (LLN) adjusted for age criteria to establish a diagnosis of COPD [10]. Given these complex diagnostic methods, the current data about lung function and prognosis of COPD are difficult to compare [11]. Several studies have demonstrated that a decline of lung function is a marker for premature death, especially in cardiovascular disease [12, 13]. Additionally, the criteria of FEV1/FVC<0.70 underestimates the prevalence of COPD, especially among people under 45 years of age [14, 15]. Colak Y et al. reported that underdiagnosis of COPD is closely related to poor prognosis, even among asymptomatic individuals [16]. A study demonstrated that the LLN criterion does not identify important pulmonary pathologies and respiratory-related complications in a large number of patients with COPD compared to a fixed ratio (FR) [17].

Identifying alternate prognoses and outcomes for COPD patients (that meet a different set of diagnostic criteria) can lay the groundwork for determining which diagnostic criterion is better. A meta-analysis comparing the risk of comorbidities and mortality in patients with different diagnostic criteria has not been performed to date. Additionally, high-quality meta-analysis is increasingly recognized as a key factor in obtaining a high level of evidence. Therefore, we conducted a meta-analysis to determine if there is a difference between the 2 diagnostic criteria for morbidity, mortality, and exacerbation.

The methods used in this study followed guidelines from the Cochrane Handbook for Systematic Reviews of Interventions. The meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines [18] and presented based on Assessing the Methodological Quality of Systematic Reviews (AMSTAR) guidelines.

This systematic review and meta-analysis compared different studies for the risk of death, exacerbation of COPD, and comorbidities. There were several key findings: patients who met any 1 of the criterion for COPD were more likely to die than those met neither, individuals with airflow limitation according to FR were more likely to be worse according to symptoms. Irrespective of whether the participants met any criterion or not, the risk of developing heart disease, heart failure, stroke, and diabetes was not significantly different between the participants.

According to LLN or FR criteria, patients with airflow limitation had a higher risk of all-cause mortality; and according to FR, individuals with airflow limitation were more likely to be worse, which concurs with previous studies [24‐26, 31]. Mannino DM et al. demonstrated a 4-fold risk of all-cause mortality in patients who met the criteria for LLN over an 18-year follow-up [25]; and in the same cohort, Akkermans RP et al. demonstrated that patients could suffer exacerbation at the first follow-up year according to the FR criterion [31]. Researchers also reported that patients with airflow limitation according to FR may account for poorer prognosis, including a higher risk of hospitalization and premature mortality [32]. One possible reason for such a phenomenon is the existence of airway hyper-responsiveness. Previous studies have shown that airway hyper-responsiveness may be an independent risk factor for mortality and an indicator of lung function decline [31, 33, 34]. Another possible explanation is that, as gas retention may occur in the early stages of COPD, the patient is worsened, and total lung capacity increases.

The results of mortality and exacerbation had high heterogeneity. There are several possible reasons for this: First, the severity of the participants included in the original studies was different. There was a close relationship between the level of severity and the risk of mortality and exacerbation. This meta-analysis demonstrated that participants with moderate airflow limitation may be more frequently hospitalized [35]. The risk of exacerbation was more frequent in the severe and very severe COPD patients [36]. Second, the proportion of ever-smoker and current-smoker varied; with the highest proportion at 84.5%. Previous studies have reported that the prevalence of COPD increased in smokers [37]; and that they have higher rates of lung function abnormalities and mortality [38]. Evidence has shown that smoking is one of the risk factors for COPD and that it can increase the lung’s total burden of inhaled particles and gases which in turn worsens a patient’s condition [36, 39‐42].. Third, the process of data extraction might also be a source of heterogeneity. In this meta-analysis, some data were not available in the original studies and needed to be extracted by software. This might have caused some errors. Finally, the use of different spirometers and the different experience level of the clinicians could also have been sources for heterogeneity.

According to either the LLN or FR criteria, individuals with airflow limitations seem to have no significant difference for the risk of developing co-morbidities such as heart disease, stroke, heart failure, and diabetes mellitus, which is discordant with previous studies. Studies have demonstrated that the morbidities of COPD are affected by cardiovascular disease, diabetes mellitus, etc. [37, 43]. These chronic conditions are very closely related to age and sex. Although it is still unclear whether healthy aging will lead to COPD, or whether age reflects the sum of cumulative exposures throughout life, it is clear that age is a risk factor for COPD [44]. For a long time heart failure was considered to be closely related to airflow limitation and it was a challenge for doctors to diagnose patients with respiratory disease [45‐47]. Yunus Çolak et al. demonstrated that heart failure could mimic airflow limitation in patients with COPD [21]. Therefore, it is doubtful whether clinicians should suspect cardiovascular disease in patients with airflow limitation according to FR only. A previous study also reported a poorer cardiopulmonary function in the LLN+/FR- group and no difference in the LLN−/FR+ group as compared to the control group [48]. Comorbidities can independently influence mortality and hospitalization in patients with COPD [49], so they deserve specific attention.

Heterogeneity was also high for the comorbidities. One possible reason for this is that the proportion of males in the original studies was unbalanced; the proportion of males varied from 23.1 to 100%. Another is the uneven age distribution of the population included in the original studies. The ages of the included participants in the original studies are different. The youngest and the oldest participants were 20 years and 100 years of age respectively. Several studies have demonstrated that the prevalence of COPD increased steeply with increasing age, especially among those > 60 years of age [37, 43, 50, 51]. Therefore, the studies which included participants > 60 years of age may have caused the heterogeneity.

This meta-analysis has several limitations. On one hand, because of the limited data in the original studies, we could not avoid the information bias. Several data for OR were calculated by the Revman software, and some mortality data for HR were extracted from the Kaplan Meier curve (K-M curve), which might have made our results less accurate. On the other hand, although only a few studies have described the prevalence of COPD for patients < 40 years of age, some studies have shown that patients < 40 years of age with COPD still account for a large portion of the total [4, 52]. Recognizing the characteristics and prognosis in COPD patients < 40 years of age is necessary. Therefore, a large number of studies are needed in the future to explore this group of patients. This is the only way to understand the overall disease characteristics of COPD and corresponding, take measures to reduce the burden of COPD and its effects on our economy and society.

Patients who meet any one of the 2 diagnostic criteria are more likely to have higher mortality as compared to those who meet neither criterion; those who meet only FR are more likely to exacerbate. Considering the influence of developing co-morbidities, we found that in patients who met only the LLN criteria, the risk for developing heart disease, stroke, heart failure, and diabetes mellitus were not different from those meet neither criterion.