Survival in idiopathic pulmonary
fibrosis
—
cytotoxic agents compared to
corticosteroids
Carlos A.C. Pereira
, Tatiana Malheiros, Ester M. Coletta,
Rimarcs G. Ferreira, Adalberto S. Rubin,
Jaqueline S. Otta, Naile
ˆ S. Rocha
Department of Pulmonology, Paulista School of Medicine, Federal University of Sao Paulo, Av Iraı´ 393 conj 34, Moema 04082 001, Sao Paulo, SP, Brazil
Received 18 June 2004; accepted 2 May 2005
Summary
Study objective: To compare the survival of patients with IPF treated retro-spectively with corticosteroids alone, to survival of patients treated with immunosuppressive and corticosteroids combined.
Design: Non-randomized retrospective cohort study. Setting: Three tertiary centers in Brazil.
Patients: Eighty-two IPF patients were included. The diagnosis was confirmed by open lung biopsy in 48. Patients received either corticosteroids alone (group I) or cytotoxic agents in addition to corticosteroids (group II).
Measurements and results: The primary end-point was mortality. Secondary outcome included longitudinal changes in FVC. Mean age was 6678 years. FVC was 71717% of predicted. There were 48 deaths during the study period (59%), 44 secondary to respiratory causes. From preliminary univariate analysis, for the group as a whole, worse survival was found to be associated with FVC%o70% of predicted
(P¼0:004), evidence of disease progression by follow-up FVC measurements (P¼0:01), and pharmacologic treatment (P¼0:014). Median survival was 25 months for the group I, and 45 months for the group II (Log-Rank¼6.45,
P¼0:01). After adjusting for FVCX70% ando70% of predicted, there was evidence to indicate that survival was associated with recommended pharmacologic treatment only in patients with FVCX70% (Log Rank¼6.84,P¼0:009).
KEYWORDS Fibrosis; Pulmonary; Fibrosing alveolitis; Cyclophosphamide; Prednisone
0954-6111/$ - see front matter&2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.rmed.2005.05.008
Conclusions: The combination of immunosuppressive agents and prednisone results in better survival when compared to prednisone alone in patients with IPF. The benefit seems to occur only in patients with less severe disease, as reflected by FVCX70%.
&2005 Elsevier Ltd. All rights reserved.
Introduction
Idiopathic pulmonary fibrosis (IPF) is a lung disease of unknown cause with a median survival less than 5 years.1–3 IPF is defined pathologically as usual interstitial pneumonia (UIP) by open lung biopsy,4 but in many cases the diagnosis can be made by typical findings on high-resolution computed tomo-graphy (HRCT) associated with appropriate clinical and functional findings.4,5
Corticosteroids (CSs) have been the mainstay of therapy for IPF, but randomized, placebo-con-trolled trials are lacking, and their value is debated.2,6,7 Two recent consensus statements4,8 recommend combining prednisone or prednisolone with either azathioprine or cyclophosphamide as initial therapy for IPF. Two small prospective studies support this recommendation.9,10 A Cochrane re-view on non-CS immunosuppressive treatment of IPF concluded that the older agents have generally not been well evaluated.11 A recent study sug-gested that combined CSs and cyclophosphamide therapy has no impact on survival in patients with IPF.12Expensive anti-fibrotic treatments have been recently become available, but a large placebo-controlled trial of interferon gamma-1b in IPF was disappointing.13
The aim of present study was to compare the survival of patients with IPF treated retrospectively with CS alone, to survival of patients treated with immunosuppressive and CS combined, both cheap treatments for IPF in underdeveloped countries.
Methods
Patients treated between January 1990 and De-cember 2000 were identified from two centers in Sa˜o Paulo and from one center in Porto Alegre, Brazil. At these centers, patients with interstitial lung diseases are evaluated by a systematic protocol.
Inclusion criteria
All patients included (n¼82) in the study fulfilled the following criteria:
With surgical lung biopsy(n¼48): Pattern of UIP, defined according of Katzenstein and Myers.14 Surgical lung biopsies were reviewed by two pathologists experienced in interstitial lung dis-eases (EMC, RGF).
Without surgical lung biopsy (n¼34)—all
criter-ia should be present:
1. Clinical evidence (dyspnea or cough) of UIP present for at least 3 months with progression of symptoms, pulmonary function impairment, or radiographic abnormality. Abnormal pulmon-ary function studies included evidence of re-striction or impaired gas exchange (low DLCO, increased P(A-a)O2, or fall in SpO2 during exercise). In this report only the changes in FVC were analyzed in relation to treatment. 2. Age450.
3. Typical HRCT findings of UIP.5,15
Typical bibasilar end-inspiratory crackles (often described as ‘‘velcro’’ in character) were present on auscultation of the lungs in 94% of cases, but were not essential for inclusion in study.
Exclusion criteria
The exclusion criteria (applied to patients with surgical biopsy or not) were as follows:
1. Evidence of a possible cause of UIP, like hypersensitivity pneumonitis.
2. Stable disease for 41 year, as evidenced by dyspnea absent or unchanged or non-significant changes in FVC (o10% and 0.2 L).
3. Significant associated COPD, characterized by FEV1/FVC ratioo65% of predicted.
4. Presence of significant centrilobular nodules, air-trapping, and extensive ground-glass opaci-ties (X30%) or ground-glass pattern dissociated from fibrosis areas, middle and upper lobe predominance, and absence of honeycombing at HRCT.4
Patients were categorized as non-smokers or as smokers (current or former smokers).
Patients included in the study were treated at the discretion of the assistant physicians with CS alone (group I,n¼26), or with the combination of CS and immunosuppressive drugs (group II,n¼56). Patients were not randomized, and medication use was not blinded to the outcome evaluators (i.e., an ‘‘open-label’’ study), but these were objective (death and FVC decline).
The dose of prednisone or equivalent was at least 0.5 mg/kg/day for 3 consecutive months, followed by tapering to 10 mg given for 1 year or more, associated or not with immunosuppressive agents intended to be given for at least 6 months, a time required for evaluation of response to these agents.16 The CS used were prednisone (median dose¼30 mg) by mouth, or weekly pulses of methylprednisolone (median dose¼1 g). The im-munosuppressive agents used were cyclophospha-mide (1–2 mg/kg/day by oral route, median dose¼100 mg/day or IV each month, median dose¼750 mg) or oral azathioprine (2–3 mg/Kg/ day). Median time of treatment was 14 months for CSs and 12 months for cyclophosphamide.
Study endpoints
The primary outcome measure was mortality. Survival was assessed through September 1, 2003. Deaths were identified from follow-up, with the patient’s family or by search of the state death registry. The cause of death was obtained by review of the hospital information, death certificates and, when available, the autopsy reports.
Causes for death were classified into: respirator-y—when resulting from respiratory failure due to IPF or complications of IPF or its treatment, like lung cancer or respiratory infections; others—those secondary to diseases not related to IPF or its treatment. Longitudinal physiological assessment was done by spirometry. The normal values were those derived for Brazilian population.17 Response to therapy was defined by changes in FVC over a 3-to 6-month period. A failure 3-to respond 3-to therapy was defined asX10% decrease in FVC (orX200-ml change).4 In the remaining patients the response was defined as favorable.
Statistical analysis
Continuous data are expressed as mean7SD, med-ian and range. Group comparisons were made using unpaired t-tests (for normally distributed, contin-uous variables), Mann–Whitney test (for non-nor-mally distributed variables), and chi-square statistics or Fisher exact tests as appropriate (for
comparisons of proportions). Correlations were calculated by Pearson’s coefficient.
Survival was estimated as the time since specific therapy to death or the last clinic visit.
Cumulative survival probabilities were estimated using the Kaplan–Meier method.
18Cox proportional
hazards regression was used to identify variables associated with survival.19 Variables that were associated with mortality at Po0:10 in univariate analysis were included in multivariate models (SPSS-10). The mortality was examined using proportional hazards regression, with the inclusion of percent predicted FVC as a covariate (to control for the global severity of disease).
A P value o0.05 (two-sided) was considered statistically significant.
The study was approved by the ethics committee of all hospitals involved.
Results
Enrollment
Eighty-two patients formed the study group, 26 of whom were treated with CS alone (group I), and 56 with cytotoxic agents, azathioprine (n¼3) or cyclophosphamide (n¼53) (group II). Thirteen patients from group I were treated with pulses of methylprednisolone and 22 patients from group II were treated with pulses of cyclophosphamide. The main reason for prescribing cyclophosphamide was its lower cost.
Forty-eight subjects in this study had open lung biopsy and fulfilled the clinicopathologic criteria; the other 34 fulfilled the clinical and HRCT clinicoradiologic criteria for the diagnosis of IPF.
Clinical variables
Baseline patient characteristics are summarized in
Table 1 according to recommended treatment. Imbalances were apparent between the two groups for clubbing and number of smokers or former smokers (both more frequent in group I). Age, sex, duration of illness, and FVC% were similar between the two study groups.
Survival analysis
From preliminary univariate analysis, for the group as a whole, shorter survival was found to be associated with FVC% of predicted (P¼0:004), evidence of disease progression by follow-up FVC measurements (P¼0:01), and treatment (P¼0:014). Median survival was 25 months for the group treated with CS alone, and 45 months for the group treated with immunosuppressive agents (Log-Rank¼ 6.45,P¼0:01,Fig. 1).
Age, gender, year of treatment, clubbing, smok-ing status, duration of symptoms, diagnosis by open lung biopsy and previous use of CS did not correlate with survival (P40:10). Due to imbalances between the two groups regarding smoking and clubbing, a multivariate analysis (Cox, forward stepwise) was done in 61 patients where all variables were available, including age, gender, FVC%, smoking status and treatment to see their influence on survival. Only treatment (P¼0:05) and FVC% (P¼0:02) remained significantly associated to survival.
Initial spirometry was done in 71 patients. After adjusting for FVCX70% ando70% predicted, there was evidence to indicate that survival was asso-ciated with recommended pharmacologic treat-ment only in patients with FVCX70% (Log Rank¼6.84, P¼0:009): median survival in pa-tients with FVCX70% was 31 months for the 10 patients in group I and 88 months for the 27 patients in group II; median survival in patients with FVCo70% was 24 months for the 10 patients in group I and 33 months for the 24 patients in group II. These data are shown inFig. 2.
Changes in FVC
Eighteen patients from group I and 44 from group II had at least one measurement of FVC after 3–6 months since the beginning of therapy. FVC declined 25170.31 ml in group I compared with
1057230 ml in group II (t¼2.05, P¼0:045). Fail-ure to respond, as measFail-ured byX200-ml decrease in FVC, was seen in 12 (67%) patients in group I, compared with 11 (25%) of patients in group II (x2
¼9:50,P¼0:002).
Adverse effects
Respiratory or urinary infections were observed in 16 (29%) cases treated with cytotoxic agents compared to two (8%) cases treated with CSs (x2¼4:63,
Po0:05). Hemorrhagic cystitis, fatigue and leuco-penia resulted in treatment stop in six patients treated with cyclophosphamide. Diabetes mellitus, complications due to osteoporosis, and CS myopathy were seen in 11 patients from both groups.
Discussion
In this retrospective study, objective benefits were seen in a group of patients with IPF treated with treatment
immunossupressor
-censored
corticosteroid
-censored
Log-Rank= 6.45, p=0.01
1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100
Survival time (months)
Cumulative survival
Figure 1 Survival of patients with IPF treated with immunossupressive agents (n¼56) or corticosteroids
(n¼26).
Table 1 IPF-patient characteristics according to recommended treatment.
Group I CS (n¼26) Group II immunosupressors (n¼56) Pvalue
Sex (M/F) 15/11 31/25 1.00
Age, yr 6678 6578 0.79
Smokers or ex-smokers,n(%) 19/24 (79%) 26/51 (51%) 0.02
Clinical duration of illness, months 23717 (n¼23) 20717 (n¼48) 0.50
FVC, % predicted 68717 (n¼20) 72718 (n¼51) 0.39
Open surgical biopsy 16 (62%) 32 (67%) 0.71
Clubbing,n(%) 15/25 (60%) 19/51 (37%) 0.06
Plus
immunosuppressive agents and CSs in comparison to a group treated with CSs alone.
Diagnosis
Studies were lung biopsy showing UIP pattern as the gold standard for diagnosis of IPF, in the absence of possible secondary causes, have suggested that HRCT lung scans are highly specific for establishing the morphologic pattern of UIP,2,5,20 but honey-combing should be present. In our study all patients without open lung biopsy had honeycombing in a typical distribution of UIP.
ATS/ERS had put forward major and minor criteria for diagnosis of IPF in the absence of a surgical lung biopsy.4 Age450 yr is described as minor criteria but in our study patients less than 50 years old without an open lung biopsy were excluded.
Transbronchial biopsy, a major criteria according ATS4to exclude other diseases than IPF, was done in only 25% of patients of our patients not submitted to open lung biopsy. However, in a large series,21 transbronchial biopsy showed diagnosis other than UIP in only 2% of patients with possible IPF.
Treatment
The recent ATS/ERS joint statement4acknowledged that there is no supportive evidence from well-conducted, randomized, placebo-controlled clini-cal trials that combined treatment with CS and immunosuppressive agents improves survival or
quality of life of IPF patients, and serious treat-ment-related side effects may outweigh any po-tential benefits of therapy. A Cochrane review on non-CS immunosuppressive treatment of IPF con-cludes that the older agents have generally not been well evaluated.11Nevertheless, due to para-digm changes in basic mechanisms of UIP,22it seems likely that these older agents will never be evaluated by large prospective multicenter studies. Several early studies of patients with IPF cited responses of 10 to 30% with CS, but these series failed to classify patients according to histologic entities or well-defined radiologic criteria.7,23In a study from Michigan,229 patients with well-defined IPF were treated with CS only. The survival was less than 3 years. A recent study12 found similar mortality rates among untreated patients with IPF, compared with those treated with a combina-tion of CS and immunosupressor agents, but it is difficult to accept the first group as an adequate control group, since treatment could be avoided due to stable disease, a finding observed in many cases, with resultant better prognosis. In our study, 10 patients not treated were excluded; the median survival exceeded 60 months in these cases. In Collard’s study,12 lung biopsy was obtained in all treated patients, but in only 15% of the untreated patients. The median survival in untreated patients with lung biopsy showing UIP was around 23 months only, compared to 56 months of untreated patients without lung biopsy, suggesting that other diseases were included in the last group. Cyclophospha-mide, either given orally or by pulse therapy, has been used in a number of studies in patients with
Cumulative survival
immunossupressor -censored
corticosteroid -censored
CVF <70% CVF >=70%
Survival time (months) Survival time (months)
Cumulative survival
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
0 10 20 30 40 50 60 0 20 40 60 80 100
Figure 2 Survival of patients with IPF, separated according initial FVCX70% of predicted oro70% and treatment with
possible IPF/UIP.24–27 In the majority of cases, previous treatment with CS had failed. Interpreta-tion of these studies is difficult. Diagnosis of IPF/ UIP was secure only in one trial.25 Many patients had a long time of symptoms or FVCo70% pre-dicted.24,25Duration of treatment varied but ifp6 months was associated with failure.25,27Long-term mortality was not evaluated. Functional responses ranged from 5%25to 60%.26.
A few randomized therapeutic trials compared CS with immunosuppressive or cytotoxic agents.9,10,28 Johnson and colleagues10 compared the effect of prednisolone and cyclophosphamide vs. predniso-lone apredniso-lone in pulmonary fibrosis. Life table analysis suggested better survival in patients in the cyclo-phosphamide-prednisolone series but this was not significant. At 3 years, 10 of 22 patients in the prednisolone only series had died compared with three of 21 patients in the cyclophosphamide-prednisolone series. With death or failure of first treatment regimen as outcome, there was a significant advantage to the patients having cyclo-phosphamide-prednisolone. This advantage was explained in part by the better lung volumes in this group on admission. This study is further limited by the inclusion of patients with collagen vascular disease.
Raghu and colleagues,10evaluated the efficacy of azathioprine and prednisone vs. prednisone and placebo in 27 patients, 23 of whom had IPF diagnosed by open lung biopsy specimens. Patients with associated collagen vascular diseases were excluded. When adjusted for age, the survival advantage of azathioprine/prednisone became marginally significant.
Winterbauer28related a prospective randomized double-blind placebo-controlled trial of treatment of IPF in 27 participants, 13 randomized to receive prednisone and double-blinded placebo and 14 to receive prednisone and azathioprine. Other clinical details and diagnostic criteria were not described in the published study. Mean survival time was 43 months for treatment with prednisone and azathioprine, with eight survivors at last follow-up (57%). For prednisone alone it was 34 months with four survivors (31%).
These figures are similar to our findings.
High-dose intravenous (IV) methylprednisolone has been used therapeutically in a number of medical fields to avoid the complications and side effects of long-term glucocorticoid therapy and because of the perception that high-dose IV methylprednisolone may have ‘‘special’’ therapeu-tic effects. Intermittent high dose ‘‘pulse’’ CS therapy could inhibit neutrophil accumulation in IPF29compared to usual doses. Based on this belief,
13 patients in our study were treated with this regimen, but no difference in survival or FVC decline was observed in comparison to the other 13 patients treated with oral steroids. Similarly, no differences in FVC changes were seen between patients treated with oral and IV cyclophospha-mide.
Functional changes
In our study, patients with FVCp70% predicted had a shorter survival in comparison to those with FVC470%. Three recent papers30–32 reported pre-dictors of survival in a total of 220 well-defined IPF patients. Latsi and colleagues30 in a retrospective study of 61 IPF patients indicate that patients with poorer initial lung function had shorter survivals. All three studies 30–32 indicated that stabilization and/or improvement of the FVC% predicted at 6 and again at 12 months was associated with improved survival. Declines in the FVC% predicted (410%) predicted shorter survivals.
In our study, in patients with FVCp70% pre-dicted, the results of treatment with cytotoxic agents were similar to those seen with CS alone. Our study also shows that treatment of IPF with cytotoxic agents and CS result in a significantly lesser decrease in FVC in comparison to patients treated with CS alone, with greater survival.
Time of treatment and side effects
While the median follow-up was 28.0 months, the median time of treatment was 14 months for CSs and 12 months for cyclophosphamide. The physi-cians in charge stopped therapy during the course of the disease for several reasons: side effects (n¼26), cumulative doses of cyclophosphamide exceeding 50 g (n¼6), clinical and functional stability (n¼4) and others (n¼5). In patients treated with cytotoxic agents the duration of treatment correlated with survival time (Cox analysis, P¼0:05). In the present study 44 deaths were secondary to respiratory causes. Some deaths were ascribed to respiratory infections, but a clear relation with treatment could not be established.
Limitations
trials are considered the gold standard for evalua-tion of drug therapy; our study is retrospective.
Selection bias is always a possibility in patients studied in tertiary centers. Many more patients with possible IPF were seen in our centers during the period of study, but were not included due to stringent inclusion criteria. The percentage of patients submitted to surgical lung biopsy in the study group was greater than all patients with possible IPF evaluated in the same time period (around 25%). Temporal changes occurred in diag-nostic criteria during the period of study, but we retrospectively applied uniform criteria for diag-nosis of IPF. All histologic and HRCT data were reviewed. There was no evidence of temporal changes in severity of disease, as evidenced by lack of correlation between year of diagnosis and duration of symptoms, FVC and age at diagnosis.
Imbalance between the two groups was present in some clinical findings. A study33 showed that survival was extended in patients with IPF who were cigarette smokers at the time of their initial evaluation and clubbing was associated to worse survival. In our study neither smoking status nor clubbing influenced survival.
We tried eliminating all confounding factors by: (1) limiting the characteristics of subjects included and (2) using Cox regression, where the only factors related to survival were FVC and treatment group.
Digital clubbing is associated with many unre-lated serious diseases but its pathogenesis remains a clinical enigma. It has been hypothesized that platelet clusters impacting in the distal vasculature mediate the morphological changes of clubbing. Since the multifunctional cytokines vascular en-dothelial growth factor and platelet-derived growth factor are released on platelet aggregation and are hypoxically regulated,34 patients with clubbing could be more hypoxemic and have a worse prognosis. However, patients with clubbing had similar exercise saturation (8477%) in
compar-ison to those without clubbing (8576%).
Conclusions
From the results of the present study it can be concluded that the combination of cytotoxic agents and CSs results in a longer survival time when compared to CSs alone in patients with IPF. The benefit seems to occur only in patients with less severe disease, as reflected by a FVCX70%.
The median survival found in the present study and in another large series of patients12 treated with CSs and immunosuppressive agents are very
similar (45 and 47 months, respectively), so we suggest that new treatments involving expensive drugs should be planned to exceed these results in a significant way.
References
1. King Jr TE, Schwarz MI, Brown K, et al. Idiopathic pulmonary fibrosis: relationship between histopathologic features and mortality.Am J Respir Crit Care Med2001;164:1025–32. 2. Flaherty KR, Toews GB, Travis WD, et al. Clinical significance
of histological classification of idiopathic interstitial pneu-monia.Eur Respir J2002;19:275–83.
3. Bjoraker JA, Ryu JH, Edwin MK, et al. Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis.
Am J Respir Crit Care Med1998;157(1):199–203.
4. American Thoracic Society. Idiopathic pulmonary fibrosis: diagnosis and treatment. International consensus state-ment. American Thoracic Society (ATS), and the European Respiratory Society (ERS). Am J Respir Crit Care Med
2000;161:646–64.
5. Hunninghake GW, Lynch DA, Galvin JR, et al. Radiologic findings are strongly associated with a pathologic diagnosis of usual interstitial pneumonia.Chest2003;124:1215–23. 6. Lynch JP, Keaner M. Current approaches to the treatment of
parenchymal lung diseases. In: Spina D, Page CP, Metzger WJ, O’Connor BJ, editors.Drugs for treatment of respira-tory diseases; 2003.
7. Richeldi L, Davies HR, Ferrara G, Franco F. CS for idiopathic pulmonary fibrosis. Cochrane Database Syst Rev 2003(3): CD002880.
8. British Thoracic Society, and Standards of Care Committee. The diagnosis, assessment and treatment of diffuse par-enchymal lung disease in adults.Thorax 1999;54(Supl.1): S1–S30.
9. Raghu G, Depaso WJ, Cain K, et al. Azathioprine combined with prednisone in the treatment of idiopathic pulmonary fibrosis: a prospective double-blind, randomized, placebo-controlled clinical trial.Am Rev Respir Dis1991;144:291–6. 10. Johnson MA, Kwan S, Snell NJ, Nunn AJ, Darbyshire JH, Turner-Warwick M. Randomised controlled trial comparing prednisolone alone with cyclophosphamide and low dose prednisolone in combination in cryptogenic fibrosing alveo-litis.Thorax1989;44:280–8.
11. Davies HR, Richeldi L, Walters EH. Immunomodulatory agents for idiopathic pulmonary fibrosis.Cochrane Database Syst Rev2003(3):CD003134.
12. Collard HR, Ryu JH, Douglas WW, et al. Combined corticosteroid and cyclophosphamide therapy does not alter survival in idiopathic pulmonary fibrosis.Chest2004;125: 2169–74.
13. Raghu G, Brown K, Bradford WZ, et al. A placebo-controlled trial of interferon gamma-1b in patients with idiopathic pulmonary fibrosis.N Engl J Med2004;350:125–33. 14. Katzenstein AL, Myers JL. Idiopathic pulmonary fibrosis:
clinical relevance of pathologic classification.Am J Respir Crit Care Med1998;157:1301–15.
15. Douglas WW, Ryu JH, Swensen SJ, et al. Colchicine vs. prednisone in the treatment of idiopathic pulmonary fibrosis.Am J Respir Crit Care Med1998;158:220–5. 16. Lynch 3rd JP, McCune WJ. Immunosuppressive and cytotoxic
17. Pereira CAC, Barreto SP, Simo˜es JG, Pereira FWL, Gerstler JG, Nakatani J. Valores de refereˆncia para espirometria em uma amostra da populac-a˜o brasileira adulta. J Pneumol
1992;18:10–22.
18. Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958;53: 457–81.
19. Cox DR. Regression models and life-tables (with discussion).
J R Stat Soc Series B1972;34:187–220.
20. Raghu G, Mageto YN, Lockhart D, et al. The accuracy of the clinical diagnosis of new-onset idiopathic pulmonary fibrosis and other interstitial lung disease: a prospective study.
Chest1999;116:1168–74.
21. Hunninghake GW, Zimmerman MB, Schwartz DA, et al. Utility of a lung biopsy for the diagnosis of idiopathic pulmonary fibrosis.Am J Respir Crit Care Med2001;164: 193–6.
22. Selman M, King TE, Pardo A. Idiopathic pulmonary fibrosis: prevailing and evolving hypotheses about its pathogenesis and implications for therapy. Ann Intern Med 2001;134: 136–51.
23. Lynch 3rd JP, White E, Flaherty K. CS in idiopathic pulmonary fibrosis.Curr Opin Pulm Med2001;7:298–308.
24. Dayton CS, Schwartz DA, Helmers RA, et al. Outcome of subjects with idiopathic pulmonary fibrosis who fail corti-costeroid therapy. Implications for further studies. Chest
1993;103:69–73.
25. Zisman DA, Lynch 3rd JP, Toews GB, Kazerooni EA, Flint A, Martinez FJ. Cyclophosphamide in the treatment of idio-pathic pulmonary fibrosis: a prospective study in patients who failed to respond to corticosteroids.Chest2000;117: 1619–26.
26. Kolb M, Kirschner J, Riedel W, Wirtz H, Schmidt M. Cyclophosphamide pulse therapy in idiopathic pulmonary fibrosis.Eur Respir J1998;12:1409–14.
27. Baughman RP, Lower EE. Use of intermittent, intravenous cyclophosphamide for idiopathic pulmonary fibrosis. Chest
1992;102:1090–4.
28. Winterbauer RH. The treatment of idiopathic pulmonary fibrosis.Chest1991;100:233–5.
29. Keogh BA, Bernardo J, Hunninghake GW, Line BR, Price DL, Crystal RG. Effect of intermittent high dose parenteral corticosteroids on the alveolitis of idiopathic pulmonary fibrosis.Am Rev Respir Dis1983;127:18–22.
30. Latsi PI, du Bois RM, Nicholson AG, et al. Fibrotic idiopathic interstitial pneumonia: the prognostic value of longitudinal functional trends. Am J Respir Crit Care Med 2003;168: 531–7.
31. Collard HR, King Jr TE, 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 Med2003;168:538–42.
32. Flaherty KR, Mumford JA, Murray S, et al. Prognostic implications of physiologic and radiographic changes in idiopathic interstitial pneumonia.Am J Respir Crit Care Med
2003;168:543–8.
33. King Jr TE, 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;164:1171–81.
34. Atkinson S, Fox SB. Vascular endothelial growth factor (VEGF)-A and platelet-derived growth factor (PDGF) play a central role in the pathogenesis of digital clubbing.J Pathol
