The Association between Periodontal Disease and Chronic Obstructive Pulmonary Disease: A Case Control Study

Abstract

Introduction: Although there are studies evaluating the effects of periodontal health on chronic obstructive pulmonary disease (COPD), the effects of COPD – a systemic disease, on periodontal tissue is unknown. The aim of this study is to evaluate the effects of COPD on periodontal tissues by comparing COPD patients and controls. Methods: Fifty-two COPD patients and 38 non-COPD controls were included in this case-control study. Number of teeth, plaque index (PI), gingival index (GI), bleeding on probing, clinical attachment level and probing depth were included in the periodontal examination. In addition to clinical evaluations, gingival crevicular fluid (GCF) levels of high-sensitive C-reactive protein (hs-CRP), interleukin-1 beta (IL-lb) and prostaglandin-E₂ (PGE₂), and serum hs-CRP levels were measured in COPD patients and the controls. Results: The number of teeth was significantly lower while PI and GI were significantly higher in COPD patients when compared to the controls. As well as serum hs-CRP levels, the GCF levels of hs-CRP, IL-1b and PGE₂ were significantly higher in COPD patients than the controls. Conclusion: Our results demonstrated that COPD may be associated with periodontal disease as manifested by lower number of teeth and higher levels of inflammatory mediators especially CRP in GCF. This finding may be a reflection of systemic effects of COPD on periodontal tissues. Poor oral health behavior of COPD patients have to be considered in larger size group studies in the future.

Keywords :

• pulmonary disease
• chronic obstructive
• periodontal disease
• periodontitis
• C-reactive protein
• interleukin-1beta

Introduction

Chronic obstructive pulmonary disease (COPD) is a chronic progressive disease characterized by airflow limitation that is not fully reversible, and is associated with an abnormal inflammatory response of the lungs to inhaled noxious particles or gases. It is a major cause of morbidity and mortality in the World (1). COPD is associated with inflammation, both in the stable phase of the disease and during exacerbations. Inflammation in the lung is also associated with a certain degree of systemic inflammation (2).

For this reason, COPD has been considered as a systemic disease in recent years and some inflammatory mediators are responsible for COPD-related systemic manifestations (3, 4). A meta-analysis done by Gan et al. has shown that even patients with stable COPD have increased white blood cell count and elevated C-reactive protein (CRP), fibrinogen and cytokines (e.g. interleukin-1 beta, interleukin-6,tumor necrosis factor-alpha) levels (5). High levels of Prostaglandin E₂ (PGE₂), which is a pro-inflammatory mediator, correlates with the severity of airflow limitation in stable COPD (6). Therefore, PGE2 may be one of the major contributors to the pathogenesis of inflammation in COPD.

Periodontitis is a chronic inflammatory disease which serves as a reservoir of Gram-negative anaerobic organisms, lipopolysaccharides and some pro-inflammatory mediators (7). Among the pro-inflammatory mediators, PGE₂ and interleukin-1 beta (IL-lβ) play critical roles in inflammatory processes leading to alveolar bone and connective tissue loss in periodontal disease (8–10).

In periodontitis, the local destruction of the periodontal tissues causes a large surface area of ulcerated pocket epithelium which allows exchange between bacterial and host products (11). It has been suggested that while periodontitis may have systemic effects, some systemic diseases such as cardiovascular diseases, diabetes, respiratory diseases, adverse pregnancy outcomes and osteoporosis have negative effects on periodontal tissue (12–14). The relation between periodontal and systemic diseases is believed to be mediated through systemic inflammatory reactants such as acute-phase proteins and immune effectors, but the underlying biological mechanisms are yet unclear, and speculative (2, 13).

The association between periodontal diseases and respiratory diseases has been discussed for years. Large-scale studies have been performed in patients with different respiratory diseases including COPD. These studies were designed to observe the effects of periodontal health on COPD (15–18). There is little information known about the effect of COPD and systemic inflammation in COPD in particular, on periodontal health.

Our hypothesis was that COPD patients might have higher inflammatory response and worse periodontal health due to the subsequent pro-inflammatory events that occur in COPD when compared to non-COPD patients. The aim of this case-control study was to evaluate the effects of COPD on periodontal tissues by measuring clinical parameters and gingival crevicular fluid (GCF) IL-1β, PGE₂ in addition to serum and GCF high-sensitive C-reactive protein (hs-CRP) levels.

Material and Methods

Subjects and study design

We screened the medical records of COPD patients (560 cases) who were followed on an outpatient basis by three chest clinics in Istanbul (Istanbul Medical Faculty Department of Respiratory Diseases, Yedikule Chest Diseases and Thoracic Surgery Training and Research Hospital and Sureyyapasa Chest Diseases and Thoracic Surgery Training and Research Hospital) between August 2008 and October 2011. Of 560 COPD patients, 440 who fulfilled the GOLD diagnostic criteria (1) were evaluated with regard to exclusion criteria. Patients with less than 8 teeth (n = 55) or who were edentulous (n = 288), as well as those who had significant respiratory diseases in addition to COPD (n = 6), a history of significant cardiovascular disorders or diabetes (n = 37), or a history of smoking in the last 2 years (n = 2) were excluded. As a result, 52 stable COPD patients fulfilling the inclusion criteria participated in the study. All patients were stable when included in the study (no infective exacerbation within one month prior to being included in the study).

The COPD group (n = 52) was divided into two subgroups with regard to the disease severity; patients with mild and moderate COPD were included in group I and the patients with severe and very severe COPD were included in group II.

Thirty-eight non-COPD controls who attended to Istanbul University, Faculty of Dentistry Department of Periodontology with matching age and gender constituted the control group. The controls with less than 8 teeth or who were edentulous or who had any respiratory diseases, significant cardiovascular disorders or diabetes were excluded. All controls were former smokers at least 2 years post-cessation of smoking. Local ethics committee of Istanbul University, Medical Faculty approved the study protocol (Project no. 2008/954), and informed written consent was obtained from all individuals prior to participation.

Assessment of COPD severity and lung function

Forced expiratory volume in 1 second (FEV₁), forced vital capacity (FVC) and FEV₁/FVC ratio of the patients were measured via spirometry by trained and certified technicians and the results noted as percentages of the predicted values (FEV₁% predicted). Later, the patients were divided into two subgroups, group-I (patients with mild-moderate COPD) and group-II (patients with severe-very severe COPD) on the basis of FEV₁% predicted values with regard to GOLD classification (1) (mild COPD: characterized by airflow limitation with FEV₁ ≥80% predicted; moderate COPD: airflow limitation with ≤50% FEV₁ < 80% predicted; severe COPD: airflow limitation with ≤30% FEV₁ < 50% predicted; very severe COPD: airflow limitation with FEV₁ < 30% predicted).

Periodontal examination

The periodontal measurements including plaque index (PI) (19), gingival index (GI) (20),% of sites that had bleeding on probing (BoP), probing depth (PD), space between the cemento-enamel junction and gingival margin (SCG) and clinical attachment (CA) level were performed by the same blinded researcher (GO). An intraclass correlation coefficient of 0.85 for PD measurements indicated that the examiner reliability was high. PD and SCG were measured by William's periodontal probe (Hu-Friedy, Chicago, IL) at six sites of all teeth (excluding third molars) and were recorded in millimeters. SCG was recorded as a positive value where the free gingival margin occurred apical to the cemento-enamel junction. CA level was calculated by CA level = PD + SCG formula.

Gingival crevicular fluid and blood sampling

The inflamed non-adjacent pocket sites of patients and controls were selected for gingival crevicular fluid (GCF) collections among the incisors and premolars with at least 4-mm depth. A total of 3 samples were collected to evaluate GCF level of IL-1β, PGE₂ and hs-CRP from each patient. One sample was collected for each biomarker and analyzed separately. Sample collections were performed one week after the clinical measurements.

Sample collections were performed 1 week after the clinical measurements. Briefly, after isolating the tooth with a cotton roll, the supragingival plaque was removed without touching the marginal gingival site and the crevicular site was gently air-dried. In order to avoid the contamination of saliva optimal effort was made. GCF was collected by paper strips (Periopaper, Oraflow Inc., NY, USA) positioned at the orifice of sulcus for 30 seconds. Strips with blood marks were discarded. The volume for each strip was measured with a calibrated meter (Periotron 6000, Oraflow Inc., NY, USA). The paper strips were immediately transferred to a plastic tube and stored at −80ºC until analysis. For each assay, separate samples were collected and analyzed. The strips were not pooled.

Fasting blood samples were collected in the anticoagulant-free vacutainer tubes by trained assistants and samples were immediately directed to the central biochemistry laboratory of Istanbul Medical Faculty for hs-CRP measurement.

Biochemical analysis

GCF was retrieved from the filter strips by eluting in 100-μl phosphate buffered saline solution-Tween buffer for 30 minutes and incubation on a shaking platform overnight (minimum 18 hours) for each parameter. GCF samples were analyzed for IL-1β, PGE₂ (Invitrogen Co., Camarillo, CA, USA) and hs-CRP (R&D systems, Minneapolis, MN, USA), using commercially available sandwich enzyme linked immunosorbent assays according to the manufacturer's instructions. The results of crevicular IL-1β and hs-CRP were expressed as ng/site and PGE₂ as pg/site.

Serum hs-CRP levels were measured by latex-enhanced immuneturbidimetric assay using the Roche Modular P800 analyzer (Roche Diagnostics GmbH Mannheim, Germany) and results were expressed as mg/L.

Statistical analysis

The ideal sample size to assure adequate power for this cross-sectional study was calculated after a pilot experiment, and a 20% difference was obtained. With a power of 80% and α = 0.05, the minimum number of individuals required for the comparisons was 36 for each group. The data were evaluated with statistical software (SPSS Inc. Chicago, IL) and the results were presented as mean ± SD. Each patient was considered as an observation unit for periodontal clinical measurements and for each clinical parameter the average of the whole mouth measurements (excluding the third molars) were used for statistical calculations. Demographic data analyses were done by using Pearson's chi-square and independent sample t-test. The correlation between biochemical and clinical parameters were performed between the site that was sampled and the site's clinical parameters. The comparison of demographical and clinical and biochemical parameters were performed by Mann–Whitney U-test. The relation between variables was evaluated by the Pearson correlation test.

Results

Table 1 shows the demographical and clinical characteristics of COPD patients and the controls. The age, sex, time since the cessation of smoking and education status of the controls and COPD patients were similar. The number of teeth in COPD patients was significantly lower than the controls (p < 0.001). PI and GI were significantly higher in COPD patients (p < 0.001, p = 0.023 respectively). There was not a significant difference between the BoP and CAL scores of COPD patients and the controls. The number of sites with PD ≥ 4 mm was higher in the controls (p < 0.04) but percentage of PD ≥ 4 mm was similar in both controls and COPD patients (p = 0.14).

Table 1.  Demographic and clinical variables of patients with COPD vs. controls

Variable	Control (n = 38)	COPD Patients (n = 52)	p value
Age in years (mean ± SD)	53.5 ± 9.2	57.5 ± 9.7	0.08
Gender
(n; male/female)	32/6	49/3	0.1
(%; male)	85%	92%
Time since cessation of smoking in years (mean ± SD)	6.8 ± 6.3	4.9 ± 4.3	0.07
Education (n)
Primary School	18	28	0.3
High School	9	11
Bachelor's degree	11	13
Number of teeth (mean ± SD)	22.2 ± 4.2	17.9 ± 5.1	<0.001
PI (mean ± SD)	1.3 ± 0.2	1.5 ± 0.4	<0.1001
GI (mean ± SD)	1.2 ± 0.2	1.4 ± 0.3	0.023
BoP (%, mean ± SD)	38.0 ± 20	49.0 ± 29.4	0.14
PD (mm, mean ± SD)	2.8 ± 0.4	2.7 ± 0.6	0.12
CAL (mm, mean ± SD)	3.4 ± 0.8	3.5 ± 1.2	0.7
PD ≥ 4 mm (%, mean ± SD)	15.7 ± 10.5	14.3 ± 15.9	0.14

Values in bold indicate a statistically significant difference (p < 0.05).

GCF IL-1β and PGE₂ levels, serum and GCF hs-CRP levels are presented in Table 2. The result of IL-1β levels and PGE₂ levels were higher in COPD patients than the controls (p < 0.001). The hs-CRP levels were significantly higher, both in GCF and serum of COPD patients when compared to the controls (p = 0.01 and p = 0.035 respectively).

Table 2.  Biochemical  parameters of patients with COPD and controls (mean ± SD)

	Control (n = 38)	COPD (n = 52)	p value
GCF IL-1β (pg/site)	0.28 ± 0.44	0.29 ± 0.33	0.000
GCF PGE2 (ng/site)	0.79 ± 0.26	0.89 ± 0.21	0.000
GCF hs-CRP (ng/site)	0.24 ± 0.30	0.76 ± 0.95	0.001
Serum hs-CRP (mg/L)	3.20 ± 2.50	7.20 ± 8.60	0.035

Values in bold indicate a statistically significant difference (p < 0.05).

Of 52 COPD patients, 29 were in group-I and 23 in group-II. None of the clinical periodontal parameters showed any significant difference between COPD subgroups (Table 3). Serum hs-CRP levels were significantly higher (p = 0.02) and GCF IL-1β levels were lower in group-II patients when compared to group-I (p = 0.03) (Table 4).

Table 3.  Demographical and clinical parameters of COPD subgroups

Variable	Group-I (n = 29)	Group-II (n = 23)	p value
Age in years (mean ± SD)	57.8 ± 8.7	57.2 ± 11.2	0.5
Time since cessation of smoking in years (mean ± SD)	5.2 ± 5.6	9.0 ± 6.6	0.04
FEV1	1.7 ± 0.6	1.1 ± 0.6	0.001
FEV1 predicted%	61.9 ± 12.1	37.5 ± 16.1	<0.000
Number of teeth (mean ± SD)	18.3 ± 4.9	17.3 ± 5.5	0.5
PI (mean ± SD)	1.5 ± 0.4	1.6 ± 0.5	0.6
GI (mean ± SD)	1.4 ± 0.3	1.4 ± 0.4	0.3
BoP (%, mean ± SD)	45.6 ± 28.3	54.9 ± 30.9	0.2
PD (mm, mean ± SD)	2.8 ± 0.7	2.5 ± 0.6	0.2
CAL (mm, mean ± SD)	3.6 ± 1.2	3.3 ± 1.1	0.3
PD ^3 4 mm (%, mean ± SD)	16.3 ± 17.8	11.4 ± 11.9	0.3

Group-I patients vs group-II patients.

Values in bold indicate a statistically significant difference (p < 0.05).

Table 4.   Biochemical parameters of COPD subgroups (mean ± SD)

	Group-I(n = 29)	Group-II (n = 23)	p value
GCF IL-1β (pg/site)	0.38 ± 0.40	0.17 ± 0.15	0.03
GCF PGE2 (ng/site)	0.93 ± 0.20	0.84 ± 0.22	0.2
GCF hs-CRP (ng/site)	0.67 ± 0.84	0.89 ± 1.08	0.6
Serum hs-CRP (mg/L)	5.09 ± 7.20	10.05 ± 9.54	0.02

Group-I patients vs group-II patients.

Values in bold indicate a statistically significant difference (p < 0.05).

Correlation analysis showed that serum CRP level was positively correlated with GCF hs-CRP levels and PD in COPD patients (r = 0.28, p = 0.044 and r = 0.32, p = 0.018, respectively). In controls, the only correlation was between GCF hs-CRP levels and PD measurements (r = 0.39, p = 0.014).

Discussion

Most of the studies evaluating the relation between COPD and periodontitis were focused on the effect of periodontal health on COPD (4, 17, 21, 22). Leuckfeld et al. (4) demonstrated a high prevalence of chronic marginal periodontitis in patients with severe COPD that appears to be independent of possible risk factors for periodontitis such as age, body mass index, use of corticosteroid, bone mineral density and pack years smoked. In a retrospective analysis of cross-sectional data, Scannapieco and Ho (17) showed a moderate association between periodontal attachment loss and COPD prevalence. In that retrospective analysis, a distinct trend was noted whereby the more severe the mean attachment loss, the greater association with COPD.

Zhou et al. (21) evaluated periodontal health and the quality of life in patients with COPD and showed that poor periodontal health as reflected by more missing teeth and high PI was significantly associated with lower quality of life in COPD patients. In a similar study (22), poor periodontal health, dental care, and oral health knowledge were significantly associated with an increased risk of COPD. Two recent studies evaluated the relationship between periodontal health and COPD exacerbations. Liu et al. (23) reported that poor periodontal health, insufficient dental care and oral health knowledge are associated with increased risk of COPD exacerbations. Kucukcoskun et al. (24) demonstrated that initial periodontal treatment prevents or decreases COPD exacerbations.

Periodontitis is a bacteria-host response related chronic inflammatory disease characterized with tissue destruction. It is known that in the presence of a systemic disease such as diabetes mellitus, the severity of the periodontitis is high and success in the treatment of periodontitis is insufficient unless the systemic disease is controlled (12, 25, 26). Since COPD is a systemic inflammatory disease, this study was designed to find an answer to the question ìDoes COPD have negative effect on periodontal health?î Our results revealed that COPD may be associated with periodontal disease as manifested by higher levels of hs-CRP, PGE₂ and IL-1β in GCF as well as higher serum hs-CRP levels compared to controls.

It is well known that smoking is an important risk factor for both COPD and periodontitis (27). Smoking could be either a confounder or an effect modifier in the relationship between periodontitis and COPD (27). Considering that all COPD patients have a smoking history, only former smokers that quit smoking at least 2 years ago were included both in COPD and the control groups in order to have similar groups in terms of smoking situation.

The most striking finding in our study was COPD patients having a significantly lower number of remaining teeth when compared to those in the control group. Similar results were also highlighted in previous studies (22, 23). Several reasons may be proposed to explain this result; the reflection of poor quality of life due to COPD related dyspnea, malnutrition due to COPD, COPD related depression causing poor oral hygiene applications (21, 22, 28, 29), effect of systemic inflammation of COPD on periodontal tissues and preference of radical treatments such as tooth extraction. COPD patients are usually unable to tolerate prolonged dental treatment due to the restrictions on their daily life related to dyspnea and therefore may opt for radical treatment. To minimize the effect of low teeth number on our results, we did not include edentulous patients or patients with less than 8 teeth.

Our results revealed that BoP, CAL, PD and PD ≥ 4 mm ratio were similar between the COPD patients and controls. Additionally COPD patients had higher PI and GI scores than the controls (p = 0.001, p = 0.023 respectively). There are few studies comparing similar periodontal parameters in COPD patients with systemically healthy controls. The first study is the case-control study published by Wang et al. in 2009 (22). In this study, 328 patients with normal lung function are compared with 306 COPD patients (22). PI, BoP, PD, CAL and number of teeth were evaluated and while the PI was higher in COPD patients, number of teeth was found to be lower (p ≤ 0.001 and p ≤ 0.001, respectively).

These results are similar with our study. The second study is a study published by Liu et al. (23) in 2012. In this study, 581 COPD patients were compared with 437 healthy individuals. PI, BI, PD, CAL and number of teeth were evaluated. PI, Bleeding Index and CAL were found to be higher in the COPD group, while number of teeth was found to be lower (p < 0.001 for each) (23). Katancik et al. (30) separated 860 cases over the age of 70 in to those with and without obstruction according to the FEV₁/FVC ratio, and evaluated their periodontal (GI, PI, PD and CAL) status.

In this study, only GI and CAL were found to be higher in patients with airway obstruction according to their FEV₁/FVC ratio. A recent study published in 2013 also evaluated the relationship between COPD and periodontal health (31). The study compared 36 mild COPD and 20 non-COPD patients by clinical periodontal measurements and salivary and serum levels of MMP-8, MMP-13, and TIMP-1. They have concluded that even with mild COPD some systemic inflammatory markers such as MMP-8 tend to increase. However they reported that their results did not support the previously proposed interaction between COPD and periodontal disease (31).

In our study, we evaluated the CRP levels in serum and GCF of both the COPD and control groups. In stable COPD cases, as the disease severity increases (severe-very severe COPD cases), hs-CRP increases due to systemic inflammation (32). For this reason, in our study we used serum hs-CRP levels as a marker of systemic inflammation in COPD. GCF samples were taken from areas where the disease was clinically the most severe (PD ≥ 4 mm and areas where bleeding is seen upon probing) from both groups. Our results showed that COPD patients have higher hs-CRP than the controls (p < 0.001). As the severity of COPD increased, hs-CRP also increased. In the severe-very severe COPD group, hs-CRP was significantly higher than those in the mild-to-moderate COPD group.

Periodontal infections may contribute to elevated systemic C-reactive protein levels (33). Recent studies revealed that serum CRP of patients with periodontal diseases is elevated with deep periodontal pockets, severe attachments loss, subgingival microflora, and alveolar bone loss (34). Treatment of periodontal infection, whether by intensive mechanical therapy, drug therapy, or extraction significantly lowers the serum CRP levels (35). According to a review the majority of the studies showed that CRP levels are higher in chronic periodontitis patients than in controls.

There is strong evidence from cross-sectional studies that plasma CRP in periodontitis is elevated compared with controls (36). However, it has been shown that the origin of CRP in GCF is not from periodontal tissues (37). It can be concluded that localized inflammation during periodontal disease may result in a generalized systemic inflammatory response in distant organs, including the liver (38) or the presence of CRP in GCF may be indicative of systemic inflammation other than periodontal infection (37).

In our study, GCF hs-CRP levels of COPD patients were significantly higher than the controls. Even though the difference was not significant, GCF hs-CRP levels were also higher in group-II patients than in group-I patients, in COPD subgroups. Our results partially support previous studies reporting that CRP in the GCF has systemic origin (37). In our study, the controls had detectable hs-CRP levels in GCF. Since the controls of the present study were systematically healthy; other than chronic periodontitis, CRP detected in the GCF of these patients is probably associated with the mild inflammation caused by periodontitis.

Since CRP is a non-specific marker and the role of CRP in GCF is not yet clear, we also measured IL-1β and PGE₂ in GCF. Recent studies reported that pro-inflammatory mediators such as IL-1β and PGE₂ have a role in the progression of periodontal disease and alveolar bone resorption (39). By a large size population study, Offenbacher et al. (9) showed that higher levels of GCF IL-1β and PGE₂ were significantly associated with clinical signs of periodontal disease. It has been concluded that the levels of these mediators were higher in GCF of patients with periodontitis, and their levels may vary with regard to the progression of periodontal diseases (9). Chronic inflammation and tissue damage are common findings of COPD and chronic periodontitis, and many pro-inflammatory mediators such as PGE₂ play an important role in both of these diseases (6, 40).

No studies evaluating these mediators in GCF have been found when researching the studies considering the relationship between COPD and periodontal disease. Our results showed that COPD patients have significantly higher GCF IL-1β and PGE₂ levels than the controls. Higher level of IL-1β in GCF might be a result of greater number of immunocompetent cells ìin situ.î Although the difference between the groups seems very low especially for PGE₂, the results were statistically significant due to the non-parametric distribution of the data. Higher levels of these mediators in GCF of COPD patients with periodontitis may be due to effect of systemic inflammation on periodontal tissues.

Limitations and strengths of the study

Periodontitis is a multifactorial disease that may be strongly influenced by socio-economic factors and it is hard to manifest obscure differences between individuals especially with respect to health behaviors. Although our results revealed a difference between periodontal health status of COPD patients and the controls, it is not clear whether this difference is a result of systemic inflammation or it is related to oral health behaviors. As this is a cross- sectional study, it can not be definitively concluded that COPD caused the increases in measured GCF inflammatory mediators; rather the results suggest that COPD may have unfavorable effects on the periodontium. The sample size of the study comparatively limited the power of the study to eliminate the effect of differences between the oral health behaviors of the participants. The case-control design, focusing on COPD and non-COPD participants and inclusion of only former smokers permitted better delineation of oral health in COPD and constitute strengths of our study.

The relationship between periodontal disease and COPD has been the subject of numerous studies in recent times. However, most of these studies examined the effects of periodontitis on COPD, while our study evaluated the effects of COPD on periodontal tissues according to periodontal clinical parameters as well as inflammatory markers, as is important in this respect. CRP GCF levels and inflammatory mediators in serum were examined in both controls and COPD patients.

Conclusion

We may conclude that the clinical periodontal findings were not different between the groups except the number of teeth. However, presence of COPD is related to higher levels of especially hs-CRP, and PGE₂ and IL-1β in GCF as well as higher serum hs-CRP levels compared to the controls. Increased levels of these inflammatory markers in COPD patients with periodontitis may be a reflection of systemic inflammation in COPD. Their poor oral health behavior must also be taken into consideration regarding this outcome. Further studies are needed to understand if either the host response or oral health behaviors of COPD patients is responsible for the inflammatory change of periodontal tissues.

Declaration of Interest Statement

This study was supported by Scientific Research Projects Coordination Unit of Istanbul University (Project No: T4100).

The authors declare that there is no conflict of interest. The authors alone are responsible for the content and writing of the paper.

Acknowledgments

The authors would like to thank Zuhal Karakurt (Sureyyapasa Chest Diseases and Thoracic Surgery Training and Research Hospital, Istanbul) for her assistance in the evaluation of some patients.

Local ethics committee of Istanbul University, Medical Faculty approved the study protocol (Project no: 2008/954), and informed written consent prior to participation was obtained from all individuals.