0
Original Research: Diffuse Lung Disease |

Validation of the GAP Score in Korean Patients With Idiopathic Pulmonary FibrosisGAP Score in Idiopathic Pulmonary Fibrosis FREE TO VIEW

Eun Sun Kim, MD; Sun Mi Choi, MD; Jinwoo Lee, MD; Young Sik Park, MD; Chang-Hoon Lee, MD; Jae-Joon Yim, MD; Chul-Gyu Yoo, MD; Young Whan Kim, MD; Sung Koo Han, MD; Sang-Min Lee, MD
Author and Funding Information

From the Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Korea.

CORRESPONDENCE TO: Sang-Min Lee, MD, Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Seoul National University College of Medicine, 101, Daehang-no, Jongno-gu, Seoul 110-744, Korea; e-mail: sangmin2@snu.ac.kr


FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2015;147(2):430-437. doi:10.1378/chest.14-0453
Text Size: A A A
Published online

BACKGROUND:  No study has determined whether the risk of mortality predicted by the GAP (gender, age, and physiologic variables) model matches the observed mortality from idiopathic pulmonary fibrosis (IPF) in non-Western populations. We evaluated the clinical course of IPF and validated the GAP model in Korean patients with IPF.

METHODS:  We included 268 patients who received a diagnosis of IPF at Seoul National University Hospital between 2005 and 2009. For each patient, demographics and clinical data, such as lung physiologic parameters at IPF diagnosis, were evaluated. We validated the GAP model using discrimination and calibration to predict the risk of death in Korean patients with IPF.

RESULTS:  The study population comprised 181 men and 87 women (mean age, 65.9 years). The mean baseline % predicted FVC was 77, and % predicted diffusing capacity of lung for carbon monoxide was 65.9. A total of 157 deaths (58.6%) occurred during follow-up, and the median time to death was 4.64 years. The observed cumulative mortality at 1, 2, and 3 years was 10.4%, 20.9%, and 31.0%, respectively. The GAP model produced estimates of 1-year mortality risk consistent with the observed data (C statistic: GAP calculator, 0.74; GAP index and staging system, 0.72; P < .29). However, calibration of the GAP model at 3 years was not satisfactory.

CONCLUSIONS:  The GAP model showed similar discrimination power compared with the original cohort but did not predict the 3-year risk of death accurately. Further multinational validation studies are needed.

Figures in this Article

Idiopathic pulmonary fibrosis (IPF) is the most common of the idiopathic interstitial pneumonias and carries the worst prognosis, with median survival ranging from 2.5 to 3.5 years. Although IPF has a poor overall prognosis, the clinical course of individual patients varies from slow progression to acute decompensation and death.

Several clinical prediction models have been developed for patients with IPF.13 However, these have not been adopted widely in clinical practice because they lack formal external validation and use some variables that are not measured routinely in current clinical practice. The GAP model was developed using gender (G), age (A), and the two lung physiology variables (P) of FVC and diffusing capacity of the lung for carbon monoxide (Dlco). To our knowledge, it is the first prediction model of IPF based on a competing risks analysis and the only one of its kind that has been externally validated in a distinct cohort of patients with IPF.4 However, one limitation is that both the derivation and validation cohorts were drawn from only two centers in the United States.

The number of incidental and prevalent IPF cases varies greatly in published studies (range, 0.5-27.9 cases per 100,000).57 Few data on IPF incidence or prevalence are available in geographic regions other than the United States and Europe. Some studies reported differences in the epidemiology of IPF between Asian and Western countries. For example, a large population-based study from Taiwan found lower IPF incidence and prevalence (0.5-6.4 per 100,000 and 0.5-1.4 per 100,000, respectively) in Asian than in Western countries.8 Another study from Japan showed that the estimate of overall IPF prevalence was 2.95 per 100,000, which is also lower than that reported in Western counties.9 From these data, we were interested in whether the prognosis of patients with IPF in Korea differs from that in Western countries. We hypothesized that the GAP model would not predict the risk of death accurately in Korean patients with IPF.

Study Design and Patients

Patients with IPF diagnosed between 2005 and 2009 at Seoul National University Hospital (SNUH), a university-affiliated tertiary care hospital in Korea, were included. The diagnosis of IPF was made by ward pulmonologists on the basis of medical history, available pulmonary function test results, high-resolution CT imaging, and surgical lung biopsy specimens following established criteria.10 Patients with no available pulmonary function test results at diagnosis or with clinical evidence of connective tissue disease, lung cancer, or lung metastasis from another malignancy; an occupational or environmental exposure that may cause interstitial lung disease; or a history of ingestion of a drug or an agent known to cause pulmonary fibrosis (Fig 1) were excluded. The study was approved by the Institutional Review Board and Ethics Committee of SNUH (IRB No. H-1304-018-477) and was conducted in compliance with the Declaration of Helsinki.

Figure Jump LinkFigure 1 –  Flowchart of patient enrollment into the study. GAP = gender, age, and physiologic variables; IPF = idiopathic pulmonary fibrosis; PFT = pulmonary function test.Grahic Jump Location
Clinical Assessment and Outcome

We assessed the patients’ demographic characteristics, including smoking status and clinical characteristics. Information on hospitalization for respiratory conditions, acute exacerbation of IPF, mechanical ventilation, and death was also evaluated by medical chart review and interview. Vital status was ascertained through a record linkage with the Korea mortality registry between January 2005 and July 2013. The cause of death was obtained by review of available hospital discharge information. Both the GAP calculator and the GAP index and staging system were applied to each patient to obtain the GAP index, stage, and predicted 1-, 2-, and 3-year mortality. Finally, we compared the observed risk of all-cause mortality with the mortality risk predicted by the GAP model.

Statistical Analysis

Descriptive data are expressed as mean ± SD unless otherwise specified. Student t test was used to compare continuous variables, and χ2 or Fisher exact test was used to compare categorical variables. Survival curves were estimated using the Kaplan-Meier method, and differences in survival time among the three GAP stage groups were calculated by log-rank test. On the basis of the reported Cox proportional hazard, we calculated the 1-, 2-, and 3-year risk for all-cause mortality for all patients and compared the risk of death predicted by the GAP model with the observed mortality using calibration plots and goodness-of-fit statistics (Hosmer-Lemeshow test). Finally, we calculated the C statistic for the GAP model as a measure of discrimination. Unless otherwise noted, all tests were two-sided and performed at the 0.05 significance level. Analyses were performed using SPSS, version 20.0 (IBM Corporation), and the Medical Research Collaborating Center of SNUH reviewed the statistical analyses.

Patient Characteristics

The characteristics of the 268 patients with IPF included in the study are summarized in Table 1. The mean age was 65.9 ± 9.6 years; 181 patients (67.5%) were men, and 151 patients (56.3%) had a smoking history. A surgical lung biopsy was performed to diagnose IPF in 54 patients (20.1%). Two patients had a family history of IPF; they had at least two affected first- or second-degree relatives. The mean baseline % predicted FVC was 77.8 ± 18.8, and the % predicted FEV1 was 89.8 ± 21.5. Dlco measurements were performed in 220 patients (82.1%). The % predicted Dlco (corrected for hemoglobin level if available) was 65.9 ± 21.7. The GAP system showed 157 patients with GAP stage I (58.6%), 73 (27.2%) with GAP stage II, and 38 (14.2%) with GAP stage III.

Table Graphic Jump Location
TABLE 1 ]  Demographic Characteristics of Study Patients Stratified by GAP Stage

Data are presented as mean ± SD or No. (%). Significant differences among the GAP stages were tested with analysis of variance, χ2, or Fisher exact test. Dlco = diffusing capacity of the lung for carbon monoxide; GAP = gender, age, and physiologic variables (FVC and Dlco).

Clinical Assessment

The mean number of admissions and acute exacerbations was 0.57 ± 1.2 and 0.49 ± 1.0 per patient per year, respectively. The frequencies of admission and acute exacerbation tended to increase as the GAP stage increased, but the differences were not significant (P = .192 and .162, respectively). Twenty-nine patients (10.8%) received mechanical ventilation, the rates of which differed significantly according to GAP stage (P < .001). Ten patients (3.7%) were given a diagnosis of lung cancer. The mean time to diagnosis of lung cancer was 37.5 months, and it was significantly associated with GAP stages; the higher GAP stages, the shorter time to diagnosis of lung cancer (P < .020) (Table 2).

Table Graphic Jump Location
TABLE 2 ]  Follow-up Outcomes and Mortality of Study Patients Stratified by GAP Stage

Data are presented as mean ± SD, No. (%), or median (range). Significant differences among the GAP stages were tested using analysis of variance, χ2, or Fisher exact test. IPF = idiopathic pulmonary fibrosis. See Table 1 legend for expansion of other abbreviation.

Survival Analyses and Validation of the GAP Model

The median duration of follow-up was 4.64 years (range, 0.03-20.6 years). Of the 268 patients, 157 (58.6%) died. The median time to death was 3.64 years (range, 0.04-10.4 years). Forty-one of 49 patients (83.7%) with data on the cause of death died of progression of lung fibrosis rather than of a comorbid condition. Eighty-three patients (31.0%) died within 3 years, and the observed cumulative mortality rates at 1, 2, and 3 years were 10.4%, 20.9%, and 31.0%, respectively. The observed mortality rate differed significantly according to GAP stage (P < .001), and we found no apparent differences in the observed and predicted risk of death (Table 3).

Table Graphic Jump Location
TABLE 3 ]  Comparison of Predicted and Observed Cumulative Mortality

Data are presented as median. See Table 1 and 2 legends for expansion of abbreviations.

Figure 2 shows the overall survival of the study population according to GAP stage. The survival rate was significantly higher in patients with GAP stage I than in those with GAP stages II or III. The C statistic values for the GAP calculator at 1, 2, and 3 years were 0.74 (95% CI, 0.35-1.00), 0.71 (95% CI, 0.44-0.92), and 0.68 (95% CI, 0.46-0.87), respectively. The GAP index and staging system produced relatively lower C statistic values than did the GAP calculator at 0.72 (95% CI, 0.34-1.00), 0.69 (95% CI, 0.42-0.91), and 0.66 (95% CI, 0.44-0.85), respectively.

Figure Jump LinkFigure 2 –  Kaplan-Meier plot of survival probability from the time of initial diagnosis in patients with IPF. The 3-y survival for GAP stage I, II, and III groups was 16.6%, 50.7%, and 52.6%, respectively. A statistically significant difference was found between the GAP stage I group and GAP stage II and III groups (log-rank P < .001). See Figure 1 legend for expansion of abbreviations.Grahic Jump Location

Finally, we compared the risk of death predicted by the GAP model with the observed mortality using calibration plots and goodness-of-fit statistics (Hosmer-Lemeshow test) (Figs 3, 4). The GAP model predicted 1-year mortality well, and the differences between the predicted and observed risks were not significant. However, we found that the GAP model did not predict the 3-year mortality accurately; that is, there was a significant difference between the predicted and observed risks of 3-year mortality. The median predicted 3-year risk of mortality was 27.7% (interquartile range, 2.3-91.9) by the GAP calculator and 16.3% (interquartile range, 16.3-76.8) by the GAP index and staging system. Compared with the 31.0% observed 3-year mortality, these corresponded to a relative underprediction of 12.9% and 47.4%, respectively (Table 3).

Figure Jump LinkFigure 3 –  Calibration plots of the GAP calculator in patients with idiopathic pulmonary fibrosis. A-C, The x-axis shows the 1-y (A), 2-y (B), and 3-y (C) risk of mortality as predicted by the GAP model, and the y-axis shows the observed risk. Every data point represents a risk class with a corresponding predicted and observed risk. The solid line represents perfect agreement between predicted and observed risks, and the dashed line represents ± 10% differences between them. The Hosmer-Lemeshow statistic tests whether predicted and observed risks differ significantly across all risk classes. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location
Figure Jump LinkFigure 4 –  Calibration plots of the GAP index and staging system in patients with idiopathic pulmonary fibrosis. A-C, The x-axis shows the 1-y (A), 2-y (B), and 3-y (C) risk of mortality as predicted by the GAP model, and the y-axis shows the observed risk. Every data point represents a risk class with a corresponding predicted and observed risk. The solid line represents perfect agreement between predicted and observed risks, and the dashed line represents ± 10% differences between them. The Hosmer-Lemeshow statistic tests whether predicted and observed risks differ significantly across all risk classes. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location

The GAP model is one of the simplest clinical prediction models for mortality in patients with IPF and has been validated by Ley et al4 in two US centers. However, no validation study has been performed in other Western or Asian countries to determine whether regional or genetic differences exist in the accuracy of this model.

In the present study, we retrospectively evaluated 268 patients who met either the histologic or the clinical criteria for IPF in Korea. In the univariate analysis, mortality was associated with sex (P = .008), age (P < .001), lower FVC (P < .001), lower FEV1 (P < .037), and lower Dlco (P = .015). In the multivariate analysis, mortality correlated independently with sex (P = .013), age (P = .001), lower FVC (P = .003), and lower Dlco (P = .015), which were exactly the same variables included in the GAP model. However, the performance of the GAP model was not fully satisfactory. The discrimination ability of the GAP model was good in the first year (C statistic range, 0.72-0.74). Even though the performance of the GAP model tended to worsen over 3 years in the present study, the C statistic values were similar to those of the original study.4 Because discrimination is not the only property relevant to prognostic indices, we checked whether the GAP score has good calibration using the Hosmer-Lemeshow χ2 statistic. To our disappointment, the GAP model did not accurately predict the 2-year and 3-year mortality in Korean patients with IPF (Figs 3, 4). Actually, Ley et al4 did not show the objective values to prove calibration using the Hosmer-Lemeshow χ2 statistic. Therefore, it is difficult to determine whether this poor calibration of GAP score for 3-year mortality is a limitation of the GAP model in general or a unique characteristic of Korean patients. A prospective multinational validation study of the GAP model is needed to solve this problem.

There are several possible reasons why the GAP model did not do well in the present study. First, lung function in this study population differed from that of the original GAP cohorts.4 The mean % predicted FVC, FEV1, and Dlco were higher in the present population than in the original GAP cohort (FVC, 77.8 vs 68.8; FEV1, 89.8 vs 77.0; Dlco, 65.9 vs 44.2). When the subset of patients with usual interstitial pneumonia as diagnosed from surgical lung biopsy specimen was analyzed separately, the lung function parameters were still better than those of the original GAP cohorts (FVC, 70.4; FEV1, 80.2; Dlco, 59.8). In previous studies, Korean patients with IPF tended to show less impairment in lung function compared with patients in Western countries.1116 Second, the patients in the present study were younger than those in the original GAP cohorts. The mean age at diagnosis of IPF was 65.9 ± 9.6 years in the present study, and that of the original GAP derivation and validation cohorts was 69.7 ± 8.7 years and 66.3 ± 8.7 years, respectively. Therefore, it might be assumed that lung physiology parameters achieve less weight in risk scores than strong predictors such as age in the present study, even though decreased lung volume and gas exchange abnormalities were generally recognized as important prognostic factors in previous studies.1,4,17

Compared with other studies, the overall prognosis of IPF was better in the present patients. Generally, about two-thirds of patients with IPF die within 5 years according to previous studies from Western countries.1824 However, in the present study, only 83 (31.0%) and 119 (44.4%) patients died within 3 and 5 years, respectively, and more than one-half of the patients remained stable over the study period. When we analyzed the patients with pathologically proven IPF separately, the 3- and 5-year mortality rates were 16.7% and 31.5%, respectively, even though they had worse lung function than the other patients. Our hospital is a tertiary referral hospital, and asymptomatic patients with IPF can be found easily. Diagnosis was based on abnormalities found on routine chest radiography and lung biopsy specimens showing usual interstitial pneumonia. Therefore, an increased rate of detection of asymptomatic IPF is another plausible explanation for this difference.

Like Ley et al,4 we evaluated the prognosis of patients with IPF retrospectively and tried to validate the GAP model in terms of both discrimination and calibration. The survival rate was significantly different according to the GAP stages similar to the original study, but the GAP model did not predict the 2-year and 3-year mortality accurately in the present study. This may be a drawback of using the GAP model to predict mortality from IPF in real clinical practice. Because the original study was conducted in only two US centers, the GAP model should be validated in other Western populations for both discrimination and calibration. Additionally, these prognostic differences suggest that environmental and genetic differences might exist between the two populations, making further study to validate the GAP model in various races and countries important.

We also reviewed detailed clinical, radiologic, and histologic data from patients with IPF, which allowed us to compare these characteristics according to GAP stage. The GAP model was developed only to predict mortality in patients with IPF, and prediction of other outcomes was not available in the previous study. We found that the need for mechanical ventilation was significantly related to GAP stage (P < .0001) and that patients with a higher GAP stage were given a diagnosis of lung cancer earlier than those with a lower GAP stage (P < .020). Respiratory admission and acute exacerbation of IPF seemed to be related to GAP stage as well, but the relationships were not significant (Table 2). Therefore, the GAP model could predict other clinical outcomes in addition to mortality in patients with IPF.

This retrospective study has certain limitations and biases. First, the Dlco test was omitted in 48 patients (17.9%) because of respiratory limitations. This may have affected the classification of patients into the three GAP stages and the performance of the GAP model. Second, we enrolled patients who received a diagnosis of IPF between 2005 and 2009 to obtain sufficient information to calculate the 3-year mortality rate. Lung transplantation, which has been shown to improve lung function and survival in patients with IPF, was not used frequently during this time in Korea. In the original GAP study, 15 (6.6%) and 20 (6.1%) lung transplantations occurred in the derivation and validation cohorts, respectively. However, only one patient in the present study was referred to another hospital for lung transplantation. Finally, this validation of the GAP model was conducted in only one tertiary referral hospital; thus, the results may have limited generalizability to other populations. A prospective multicenter validation study of the GAP model is needed to confirm these data in Korea.

The GAP model may be a valuable tool for determining the prognosis and guiding the management of IPF. The GAP model in Korean patients with IPF showed similar discrimination power compared with the original cohort. However, the GAP model did not accurately predict the 2-year and 3-year risk of death in individual patients with IPF. Additional multinational study is needed to confirm these findings and to validate the applicability and accuracy of this risk assessment system.

Author contributions: S.-M. L. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. E. S. K. and S.-M. L. contributed to the study concept and design, data analysis and interpretation, drafting of the manuscript, critical revision of the manuscript for important intellectual content, and final approval of the manuscript; S. M. C., J. L., Y. S. P., and J.-J. Y. contributed to the data analysis and interpretation and final approval of the manuscript; and C.-H. L., C.-G. Y., Y. W. K., and S. K. H. contributed to the data analysis and interpretation, critical revision of the manuscript for important intellectual content, and final approval of the manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Other contributions: The authors thank the study patients who allowed us to conduct clinical research studies in an effort to improve the lives of patients with IPF.

Dlco

diffusing capacity of the lung for carbon monoxide

GAP

gender, age, and physiologic variables

IPF

idiopathic pulmonary fibrosis

SNUH

Seoul National University Hospital

King TE Jr, Tooze JA, Schwarz MI, Brown KR, Cherniack RM. Predicting survival in idiopathic pulmonary fibrosis: scoring system and survival model. Am J Respir Crit Care Med. 2001;164(7):1171-1181. [CrossRef] [PubMed]
 
Wells AU, Desai SR, Rubens MB, et al. Idiopathic pulmonary fibrosis: a composite physiologic index derived from disease extent observed by computed tomography. Am J Respir Crit Care Med. 2003;167(7):962-969. [CrossRef] [PubMed]
 
du Bois RM, Weycker D, Albera C, et al. Ascertainment of individual risk of mortality for patients with idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2011;184(4):459-466. [CrossRef] [PubMed]
 
Ley B, Ryerson CJ, Vittinghoff E, et al. A multidimensional index and staging system for idiopathic pulmonary fibrosis. Ann Intern Med. 2012;156(10):684-691. [CrossRef] [PubMed]
 
Fernández Pérez ER, Daniels CE, Schroeder DR, et al. Incidence, prevalence, and clinical course of idiopathic pulmonary fibrosis: a population-based study. Chest. 2010;137(1):129-137. [CrossRef] [PubMed]
 
Kaunisto J, Salomaa ER, Hodgson U, Kaarteenaho R, Myllärniemi M. Idiopathic pulmonary fibrosis—a systematic review on methodology for the collection of epidemiological data. BMC Pulm Med. 2013;13(1):53. [CrossRef] [PubMed]
 
Nalysnyk L, Cid-Ruzafa J, Rotella P, Esser D. Incidence and prevalence of idiopathic pulmonary fibrosis: review of the literature. Eur Respir Rev. 2012;21(126):355-361. [CrossRef] [PubMed]
 
Lai CC, Wang CY, Lu HM, et al. Idiopathic pulmonary fibrosis in Taiwan - a population-based study. Respir Med. 2012;106(11):1566-1574. [CrossRef] [PubMed]
 
Ohno S, Nakaya T, Bando M, Sugiyama Y. Idiopathic pulmonary fibrosis—results from a Japanese nationwide epidemiological survey using individual clinical records. Respirology. 2008;13(6):926-928. [CrossRef] [PubMed]
 
Raghu G, Collard HR, Egan JJ, et al; ATS/ERS/JRS/ALAT Committee on Idiopathic Pulmonary Fibrosis. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183(6):788-824. [CrossRef] [PubMed]
 
Jeon K, Chung MP, Lee KS, et al. Prognostic factors and causes of death in Korean patients with idiopathic pulmonary fibrosis. Respir Med. 2006;100(3):451-457. [CrossRef] [PubMed]
 
Kim YJ, Park JW, Kyung SY, et al. Clinical characteristics of idiopathic pulmonary fibrosis patients with diabetes mellitus: the national survey in Korea from 2003 to 2007. J Korean Med Sci. 2012;27(7):756-760. [CrossRef] [PubMed]
 
Kim JH, Lee JH, Ryu YJ, Chang JH. Clinical predictors of survival in idiopathic pulmonary fibrosis. Tuberc Respir Dis (Seoul). 2012;73(3):162-168. [CrossRef] [PubMed]
 
Lee JS, Ryu JH, Elicker BM, et al. Gastroesophageal reflux therapy is associated with longer survival in patients with idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2011;184(12):1390-1394. [CrossRef] [PubMed]
 
Lee JS, Song JW, Wolters PJ, et al. Bronchoalveolar lavage pepsin in acute exacerbation of idiopathic pulmonary fibrosis. Eur Respir J. 2012;39(2):352-358. [CrossRef] [PubMed]
 
Kim DS, Park JH, Park BK, Lee JS, Nicholson AG, Colby T. Acute exacerbation of idiopathic pulmonary fibrosis: frequency and clinical features. Eur Respir J. 2006;27(1):143-150. [CrossRef] [PubMed]
 
Collard HR, King TE Jr, Bartelson BB, Vourlekis JS, Schwarz MI, Brown KK. Changes in clinical and physiologic variables predict survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2003;168(5):538-542. [CrossRef] [PubMed]
 
Bjoraker JA, Ryu JH, Edwin MK, et al. Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 1998;157(1):199-203. [CrossRef] [PubMed]
 
Kim DS. Acute exacerbations in patients with idiopathic pulmonary fibrosis. Respir Res. 2013;14(1):86. [CrossRef] [PubMed]
 
Schwartz DA, Helmers RA, Galvin JR, et al. Determinants of survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 1994;149(2):450-454. [CrossRef] [PubMed]
 
Schwartz DA, Van Fossen DS, Davis CS, et al. Determinants of progression in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 1994;149(2):444-449. [CrossRef] [PubMed]
 
Carrington CB, Gaensler EA, Coutu RE, FitzGerald MX, Gupta RG. Natural history and treated course of usual and desquamative interstitial pneumonia. N Engl J Med. 1978;298(15):801-809. [CrossRef] [PubMed]
 
Costabel U, Matthys H. Different therapies and factors influencing response to therapy in idiopathic diffuse fibrosing alveolitis. Respiration. 1981;42(3):141-149. [CrossRef] [PubMed]
 
Dayton CS, Schwartz DA, Helmers RA, et al. Outcome of subjects with idiopathic pulmonary fibrosis who fail corticosteroid therapy. Implications for further studies. Chest. 1993;103(1):69-73. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1 –  Flowchart of patient enrollment into the study. GAP = gender, age, and physiologic variables; IPF = idiopathic pulmonary fibrosis; PFT = pulmonary function test.Grahic Jump Location
Figure Jump LinkFigure 2 –  Kaplan-Meier plot of survival probability from the time of initial diagnosis in patients with IPF. The 3-y survival for GAP stage I, II, and III groups was 16.6%, 50.7%, and 52.6%, respectively. A statistically significant difference was found between the GAP stage I group and GAP stage II and III groups (log-rank P < .001). See Figure 1 legend for expansion of abbreviations.Grahic Jump Location
Figure Jump LinkFigure 3 –  Calibration plots of the GAP calculator in patients with idiopathic pulmonary fibrosis. A-C, The x-axis shows the 1-y (A), 2-y (B), and 3-y (C) risk of mortality as predicted by the GAP model, and the y-axis shows the observed risk. Every data point represents a risk class with a corresponding predicted and observed risk. The solid line represents perfect agreement between predicted and observed risks, and the dashed line represents ± 10% differences between them. The Hosmer-Lemeshow statistic tests whether predicted and observed risks differ significantly across all risk classes. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location
Figure Jump LinkFigure 4 –  Calibration plots of the GAP index and staging system in patients with idiopathic pulmonary fibrosis. A-C, The x-axis shows the 1-y (A), 2-y (B), and 3-y (C) risk of mortality as predicted by the GAP model, and the y-axis shows the observed risk. Every data point represents a risk class with a corresponding predicted and observed risk. The solid line represents perfect agreement between predicted and observed risks, and the dashed line represents ± 10% differences between them. The Hosmer-Lemeshow statistic tests whether predicted and observed risks differ significantly across all risk classes. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location

Tables

Table Graphic Jump Location
TABLE 1 ]  Demographic Characteristics of Study Patients Stratified by GAP Stage

Data are presented as mean ± SD or No. (%). Significant differences among the GAP stages were tested with analysis of variance, χ2, or Fisher exact test. Dlco = diffusing capacity of the lung for carbon monoxide; GAP = gender, age, and physiologic variables (FVC and Dlco).

Table Graphic Jump Location
TABLE 2 ]  Follow-up Outcomes and Mortality of Study Patients Stratified by GAP Stage

Data are presented as mean ± SD, No. (%), or median (range). Significant differences among the GAP stages were tested using analysis of variance, χ2, or Fisher exact test. IPF = idiopathic pulmonary fibrosis. See Table 1 legend for expansion of other abbreviation.

Table Graphic Jump Location
TABLE 3 ]  Comparison of Predicted and Observed Cumulative Mortality

Data are presented as median. See Table 1 and 2 legends for expansion of abbreviations.

References

King TE Jr, Tooze JA, Schwarz MI, Brown KR, Cherniack RM. Predicting survival in idiopathic pulmonary fibrosis: scoring system and survival model. Am J Respir Crit Care Med. 2001;164(7):1171-1181. [CrossRef] [PubMed]
 
Wells AU, Desai SR, Rubens MB, et al. Idiopathic pulmonary fibrosis: a composite physiologic index derived from disease extent observed by computed tomography. Am J Respir Crit Care Med. 2003;167(7):962-969. [CrossRef] [PubMed]
 
du Bois RM, Weycker D, Albera C, et al. Ascertainment of individual risk of mortality for patients with idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2011;184(4):459-466. [CrossRef] [PubMed]
 
Ley B, Ryerson CJ, Vittinghoff E, et al. A multidimensional index and staging system for idiopathic pulmonary fibrosis. Ann Intern Med. 2012;156(10):684-691. [CrossRef] [PubMed]
 
Fernández Pérez ER, Daniels CE, Schroeder DR, et al. Incidence, prevalence, and clinical course of idiopathic pulmonary fibrosis: a population-based study. Chest. 2010;137(1):129-137. [CrossRef] [PubMed]
 
Kaunisto J, Salomaa ER, Hodgson U, Kaarteenaho R, Myllärniemi M. Idiopathic pulmonary fibrosis—a systematic review on methodology for the collection of epidemiological data. BMC Pulm Med. 2013;13(1):53. [CrossRef] [PubMed]
 
Nalysnyk L, Cid-Ruzafa J, Rotella P, Esser D. Incidence and prevalence of idiopathic pulmonary fibrosis: review of the literature. Eur Respir Rev. 2012;21(126):355-361. [CrossRef] [PubMed]
 
Lai CC, Wang CY, Lu HM, et al. Idiopathic pulmonary fibrosis in Taiwan - a population-based study. Respir Med. 2012;106(11):1566-1574. [CrossRef] [PubMed]
 
Ohno S, Nakaya T, Bando M, Sugiyama Y. Idiopathic pulmonary fibrosis—results from a Japanese nationwide epidemiological survey using individual clinical records. Respirology. 2008;13(6):926-928. [CrossRef] [PubMed]
 
Raghu G, Collard HR, Egan JJ, et al; ATS/ERS/JRS/ALAT Committee on Idiopathic Pulmonary Fibrosis. An official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med. 2011;183(6):788-824. [CrossRef] [PubMed]
 
Jeon K, Chung MP, Lee KS, et al. Prognostic factors and causes of death in Korean patients with idiopathic pulmonary fibrosis. Respir Med. 2006;100(3):451-457. [CrossRef] [PubMed]
 
Kim YJ, Park JW, Kyung SY, et al. Clinical characteristics of idiopathic pulmonary fibrosis patients with diabetes mellitus: the national survey in Korea from 2003 to 2007. J Korean Med Sci. 2012;27(7):756-760. [CrossRef] [PubMed]
 
Kim JH, Lee JH, Ryu YJ, Chang JH. Clinical predictors of survival in idiopathic pulmonary fibrosis. Tuberc Respir Dis (Seoul). 2012;73(3):162-168. [CrossRef] [PubMed]
 
Lee JS, Ryu JH, Elicker BM, et al. Gastroesophageal reflux therapy is associated with longer survival in patients with idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2011;184(12):1390-1394. [CrossRef] [PubMed]
 
Lee JS, Song JW, Wolters PJ, et al. Bronchoalveolar lavage pepsin in acute exacerbation of idiopathic pulmonary fibrosis. Eur Respir J. 2012;39(2):352-358. [CrossRef] [PubMed]
 
Kim DS, Park JH, Park BK, Lee JS, Nicholson AG, Colby T. Acute exacerbation of idiopathic pulmonary fibrosis: frequency and clinical features. Eur Respir J. 2006;27(1):143-150. [CrossRef] [PubMed]
 
Collard HR, King TE Jr, Bartelson BB, Vourlekis JS, Schwarz MI, Brown KK. Changes in clinical and physiologic variables predict survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2003;168(5):538-542. [CrossRef] [PubMed]
 
Bjoraker JA, Ryu JH, Edwin MK, et al. Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 1998;157(1):199-203. [CrossRef] [PubMed]
 
Kim DS. Acute exacerbations in patients with idiopathic pulmonary fibrosis. Respir Res. 2013;14(1):86. [CrossRef] [PubMed]
 
Schwartz DA, Helmers RA, Galvin JR, et al. Determinants of survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 1994;149(2):450-454. [CrossRef] [PubMed]
 
Schwartz DA, Van Fossen DS, Davis CS, et al. Determinants of progression in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 1994;149(2):444-449. [CrossRef] [PubMed]
 
Carrington CB, Gaensler EA, Coutu RE, FitzGerald MX, Gupta RG. Natural history and treated course of usual and desquamative interstitial pneumonia. N Engl J Med. 1978;298(15):801-809. [CrossRef] [PubMed]
 
Costabel U, Matthys H. Different therapies and factors influencing response to therapy in idiopathic diffuse fibrosing alveolitis. Respiration. 1981;42(3):141-149. [CrossRef] [PubMed]
 
Dayton CS, Schwartz DA, Helmers RA, et al. Outcome of subjects with idiopathic pulmonary fibrosis who fail corticosteroid therapy. Implications for further studies. Chest. 1993;103(1):69-73. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543