- Email: [email protected]
Prevalence and Outcomes of Pulmonary Arterial Hypertension in Advanced Idiopathic Pulmonary Fibrosis
CHEST
Original Research INTERSTITIAL LUNG DISEASE
Prevalence and Outcomes of Pulmonary Arterial Hypertension in Advanced Idiopathic Pulmonary Fibrosis* Christopher J. Lettieri, MD; Steven D. Nathan, MD, FCCP; Scott D. Barnett, PhD; Shahzad Ahmad, MD; and Andrew F. Shorr, MD, MPH, FCCP
Study objectives: The development of pulmonary arterial hypertension (PAH) can complicate many interstitial lung diseases, including idiopathic pulmonary fibrosis (IPF). We sought to characterize the prevalence of PAH and its impact on survival in patients with advanced IPF. Design: Retrospective analysis of consecutive IPF patients undergoing pretransplantation right heart catheterization. Setting: Lung transplant and IPF referral center. Methods: PAH was defined as a mean pulmonary artery pressure (mPAP) of > 25 mm Hg. We compared demographic, spirometric, 6-min walk test (6MWT) results, and survival outcomes between those with PAH and those without PAH. Measurements and results: Seventy-nine patients were included in the study. PAH was present in 31.6% of patients (mean [ⴞ SD] mPAP, 29.5 ⴞ 3.3 vs 19.1 ⴞ 3.7 mm Hg, respectively). Those patients with PAH had a lower mean diffusing capacity of the lung for carbon monoxide (DLCO) (37.6 ⴞ 11.3% vs 31.1 ⴞ 10.1%, respectively; p ⴝ 0.04) and were more likely to require supplemental oxygen (66.7% vs 17.6%, respectively; p < 0.0001). Mean distance walked (143.5 ⴞ 65.5 vs 365.9 ⴞ 81.8 m, respectively; p < 0.001) and mean pulse oximetric saturation nadir (80.1 ⴞ 3.7% vs 88.0 ⴞ 3.5%, respectively; p < 0.001) during the 6MWT were also lower among those with PAH. PAH was associated with a greater risk of death during the study period (mortality rate, 60.0% vs 29.9%, respectively; odds ratio, 2.6; 95% confidence interval [CI], 2.3 to 3.1; p ⴝ 0.001). One-year mortality rates were higher in those with PAH (28.0% vs 5.5%, respectively; p ⴝ 0.002). As a predictor of mortality, PAH had a sensitivity, specificity, and accuracy of 57.1%, 79.3%, and 73.4%, respectively. There was a linear correlation between mPAP and outcomes with higher pressures associated with a greater risk of mortality (hazard ratio, 1.09; 95% CI, 1.02 to 1.16). FVC and DLCO did not predict outcomes. Conclusions: PAH is common in advanced cases of IPF and significantly impacts survival. A reduced DLCO, supplemental oxygen requirement, or poor 6-min walk performance should raise suspicion of the presence of underlying PAH. Identifying PAH might be an important adjunct in monitoring disease progression, triaging for transplantation, and guiding therapy. (CHEST 2006; 129:746 –752) Key words: catheterization; exercise test; mortality; pulmonary artery; pulmonary fibrosis; pulmonary function tests; pulmonary hypertension Abbreviations: CI ⫽ confidence interval; Dlco ⫽ diffusing capacity of the lung for carbon monoxide; HR ⫽ hazard ratio; ILD ⫽ interstitial lung disease; IPF ⫽ idiopathic pulmonary fibrosis; LV ⫽ left ventricular; mPAP ⫽ mean pulmonary arterial pressure; PA ⫽ pulmonary artery; PAH ⫽ pulmonary arterial hypertension; PAWP ⫽ pulmonary artery wedge pressure; PFT ⫽ pulmonary function test; 6MWT ⫽ 6-min walk test; Spo2 ⫽ pulse oximetric saturation; TLC ⫽ total lung capacity
forms of diffuse parenchymal lung disease M any can result in the development of pulmonary arterial hypertension (PAH).1–11 For example, Arcasoy and coworkers12 found PAH in one quarter of 746
patients with various forms of advanced lung disease who were referred for transplantation. Disease-specific PAH has been described as occurring in 5 to 38% of patients with scleroderma, 4.3 to 43% of Original Research
patients with systemic lupus erythematosus, and 21% of patients with rheumatoid arthritis.4 – 8 Sarcoidosis is also associated with PAH in 1 to 28% of cases and is more prevalent in patients with more advanced disease.9 PAH associated with interstitial lung disease (ILD) has been found to be a significant predictor of mortality.13,14 PAH has been reported in patients with idiopathic pulmonary fibrosis (IPF), but the prevalence has not been well-defined. In an analysis of the United Network of Organ Sharing registry, Shorr and colleagues10,11 found that approximately one quarter of ⬎ 2,000 patients with IPF who were listed for lung transplantation had PAH. While this study was not designed to look specifically at patients with IPF, it did suggest that the incidence of PAH is significant in this population. A better understanding of the prevalence of PAH in patients with IPF is important. Routine measurements from pulmonary function tests (PFTs) may not accurately correlate with survival in patients with IPF.13–22 PAH, on the other hand, is associated with worse outcomes in patients with other advanced lung diseases and may have similar consequences in patients with IPF. A possible explanation is that underlying PAH is not reflected by these standard measures. Patients with IPF often have functional limitations and a decreased exercise capacity, as measured by the 6-min walk test (6MWT), which has been shown to correlate with outcomes better than standard PFTs and may reflect coexistent pulmonary hypertension.23–26 Underlying PAH may explain why survival correlates better with functional limitations than measurements of restrictive lung physiology. Understanding and characterizing the relationship between PAH and IPF may provide important prognostic information, may assist with triaging for transplantation, and may potentially provide a target for therapy. *From the Department of Pulmonary & Critical Care Medicine (Dr. Lettieri), Walter Reed Army Medical Center, Washington DC; the Lung Transplant Program (Drs. Nathan, Barnett, and Ahmad), Inova Fairfax Hospital, Falls Church, VA; and the Department of Pulmonary and Critical Care Medicine (Dr. Shorr), Washington Hospital Center, Washington, DC. The authors have no potential financial conflicts of interests to disclose. The opinions expressed herein are not to be construed as official or as reflecting the policies of either the Department of the Army or the Department of Defense. Manuscript received August 2, 2005; revision accepted September 14, 2005. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Christopher J. Lettieri, MD, Walter Reed Army Medical Center, Department of Pulmonary & Critical Care, 6900 Georgia Ave, NW, Washington, DC 20307; e-mail: [email protected] www.chestjournal.org
We hypothesized that PAH is common in patients with more advanced IPF and may be an independent risk factor for mortality. We attempted to define this association using a cohort of patients with IPF who underwent right heart catheterization as part of an evaluation for lung transplantation.
Materials and Methods Subjects We retrospectively reviewed all patients with IPF who were seen at our clinic between June 1998 and December 2004. Our clinic serves as both an IPF and lung transplant referral center. We included patients who had undergone right-sided cardiac catheterization as part of their initial evaluation prior to being listed for lung transplantation. Spirometry obtained within 1 month of the catheterization was required for inclusion in the study. When available, we included data from the 6MWT performed within 1 month of the catheterization. The 6MWTs were performed while the patient breathed room air and were conducted in accordance with the recommendations of the American Thoracic Society.27 All subjects met accepted diagnostic criteria for IPF per the American Thoracic Society/European Respiratory Society guidelines and had histopathologic evidence of usual interstitial pneumonia.28 Although these guidelines were not published until 2000, we retrospectively verified that all included subjects fulfilled the criteria outlined in these guidelines. Measurements We recorded demographic data and PFT results, and estimated the left ventricular (LV) ejection fraction from echocardiography and cardiac catheterization data. PFT measurements included FVC, total lung capacity (TLC), and diffusing capacity of the lung for carbon monoxide (Dlco). The prediction equations of Knudson et al,29 with appropriate demographic adjustments, were employed. For the 6MWT analysis, only the studies in which patients ambulated while breathing room air were included, and therefore patients with a pulse oximetric saturation (Spo2) of ⬍ 88% were excluded from the study. This was done in order to homogenize the data obtained from the 6MWTs. The 6MWT data included the total distance walked (in meters) and the lowest Spo2 recorded by noninvasive pulse oximetry. The test was terminated if the Spo2 fell to ⬍ 80%. Right-sided cardiac catheterization data included measurements of right atrial pressure, pulmonary arterial pressures, pulmonary artery wedge pressure (PAWP), and cardiac index. Measurements obtained from cardiac catheterization were obtained from the patient during a resting state. We defined PAH as a resting mean pulmonary arterial pressure (mPAP) of ⬎ 25 mm Hg. Those with a mPAP values ⱕ 25 mm Hg were considered to be normal. We defined LV dysfunction as an LV ejection fraction of ⬍ 50% or a PAWP of ⬎ 14 mm Hg. Further evaluation for other secondary causes of PAH (ie, polysomnography or ventilation-perfusion scanning) was conducted when clinically indicated. End Points The primary end point of this study was the prevalence of PAH. Survival to lung transplantation or the end of the observation period represented the secondary end point, and patients CHEST / 129 / 3 / MARCH, 2006
747
were censored at the timing of either of these events. Mortality status was determined in December 2004 and was verified using the US Social Security Death Index. We explored whether demographics, cardiac parameters, PFT measurements, or 6MWT parameters correlated with the incidence of PAH. Statistical Analysis We initially compared continuous variables with the Student t test and analyzed categoric variables with the Fisher exact test. The survival analysis was completed according to the methods of Kaplan and Meier, and the log rank test was used to compare survival curves. Cox proportional hazards regression models were used to calculate hazard ratios (HRs) for mortality, including adjustments for FVC and Dlco. The data are presented as the mean ⫾ SD. All tests were two-tailed, and p values of ⬍ 0.05 were assumed to represent statistical significance. All analyses were completed using a statistical software package (SPSS, version 12.0; SPSS, Inc; Chicago, IL).
Results During the study period, 79 patients with IPF who were evaluated in our clinic were referred for lung transplant evaluation and underwent right-heart catheterization. These individuals comprised the final cohort included in this analysis. No subjects were lost to follow-up, and none were excluded. The mean mPAP was 23.4 ⫾ 5.1 mm Hg (range, 8 to 46 mm Hg) among the cohort (Fig 1). The criteria for PAH were present in 25 subjects (31.6%), with a mean mPAP of 29.5 ⫾ 3.3 mm Hg, compared with 19.1 ⫾ 3.7 mm Hg among those who did not meet the definition. Table 1 displays the characteristics of patients with and without PAH. Age, gender, FVC, TLC, and cardiac index did not differ between those with or without PAH. The PFT values used for this analysis were obtained within a mean period of 24.2 ⫾ 6.7 days11–34 after right heart catheterization. Dlco was significantly lower in those with PAH (37.6 ⫾ 11.3% vs 31.1 ⫾ 10.1% predicted, respectively; p ⫽ 0.04). Among patients in the cohort, 34.6% of subjects with IPF required supplemental oxygen, which was more common in those with PAH (66.7% vs 17.6%, respectively; p ⬍ 0.0001). The need for supplemental oxygen was determined by either a resting Spo2 of ⬍ 88% or a Pao2 of ⬍ 55 mm Hg. Among patients in the cohort, 15.2% had both a supplemental oxygen requirement and Dlco of ⬍40% predicted. These individuals were 10.2 times more likely to have underlying PAH than those with a Dlco of ⱖ 40% predicted who did not require supplemental oxygen (p ⬍ 0.001). The need for supplemental oxygen together with a Dlco of ⬍ 40% predicted identified the presence of PAH with a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 65.0%, 94.1%, 86.7%, 82.1%, and 748
Figure 1. mPAP values. Histogram displaying the distribution of mPAPs among the cohort (values expressed in mm Hg). mPAP ⫽ mean pulmonary arterial pressure
83.3%, respectively. The 1-year mortality rate for individuals with both of these factors was 50.0%, compared with 21.1% for those patients with neither (p ⫽ 0.02). PAH was present in 52.4% of nonsurvivors compared to only 24.1% of survivors (p ⫽ 0.008). During the observation period, 60.0% of those patients with PAH died compared to only 29.9% of those without PAH (odds ratio, 2.6; 95% confidence interval [CI], 2.3 to 3.1; p ⫽ 0.001). One-year mortality rates were similarly higher among those with PAH (28.0% vs 5.5%, respectively; p ⫽ 0.002). Figure 2 shows the Kaplan-Meier curve, which revealed an early and significant difference in survival between those with and without PAH. The presence of PAH predicted mortality with a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 57.1%, 79.3%, 50.0%, 83.6%, and 73.4%, respectively. There was a linear correlation between mPAP and outcomes, with higher pressures associated with a greater risk of mortality (HR, 1.09; 95% CI, 1.02 to 1.16). In other words, the higher the mPAP, the greater the subsequent risk of mortality. After adjustment for both FVC (HR, 1.10; 95% CI, 1.02 to 1.19) and Dlco (HR, 1.07; 95% CI, 1.01 to 1.18), increasing mPAP remained a significant predictor of death. Original Research
Table 1—Patient Characteristics Based on the Presence or Absence of PAH* Characteristics
MAP ⱕ 25 mm Hg (n ⫽ 54)
MAP ⬎ 25 mm Hg (n ⫽ 25)
p Value
Age, yr Male gender, % Supplemental oxygen, % FVC, % predicted TLC, % predicted Dlco, % predicted mPAP, mm Hg RAP, mm Hg Cardiac index, L/min/m2 PAWP, mm Hg Ejection fraction, % Mortality rate, % 1-yr mortality rate, %
56.2 ⫾ 1.0 72.2 17.6 52.5 ⫾ 11.9 55.5 ⫾ 10.7 37.6 ⫾ 11.3 19.1 ⫾ 3.7 4.3 ⫾ 2.8 2.8 ⫾ 0.4 8.4 ⫾ 3.5 61.5 ⫾ 5.9 29.9 5.5
54.7 ⫾ 3.8 64.0 66.7 49.3 ⫾ 11.0 57.8 ⫾ 12.1 31.1 ⫾ 10.1 29.5 ⫾ 3.3 5.3 ⫾ 3.5 3.2 ⫾ 1.3 9.3 ⫾ 3.5 59.4 ⫾ 5.0 60.0 28.8
0.27 0.23 ⬍ 0.001 0.13 0.27 0.04 n/a 0.14 0.20 0.22 0.12 0.001 0.002
*Values are given as the mean ⫾ SD, unless otherwise indicated. RAP ⫽ right atrial pressure; MAP ⫽ mean arterial pressure.
In addition to mPAP, the measured spirometric and cardiac parameters were also analyzed as potential predictors of mortality within 1, 2, or 3 years from evaluation. FVC, TLC, and cardiac index did not differ between survivors and nonsurvivors. At each time period, only the presence of PAH differentiated survivors from nonsurvivors (1 year, p ⫽ 0.027; 2 years, p ⫽ 0.0005; 3 years, p ⫽ 0.018). While Dlco and the need for supplemental oxygen predicted the presence of PAH, they did not determine outcomes over the ensuing 1, 2, or 3 years.
To account for any potential bias, we reanalyzed the measured variables as functions of survival after eliminating 22 subjects who underwent lung transplantation. Of the remaining 57 patients, PAH was present in 33.3%. Although Dlco was significantly lower in those with PAH, only PAH independently predicted survival (p ⬍ 0.0001). Among patients in this subpopulation, the 1-year mortality rate was significantly greater in those with PAH compared to those without PAH (36.8% vs 7.9%, respectively; p ⫽ 0.003). Similarly, we reexplored survival after
Figure 2. PAH as a predictor of survival in patients with IPF. www.chestjournal.org
CHEST / 129 / 3 / MARCH, 2006
749
eliminating 19 subjects with LV dysfunction to ensure that congestive heart failure was not a potential confounding factor. In this instance, Dlco predicted PAH, but only PAH independently predicted survival (p ⬍ 0.0001). Again, 1-year mortality rates were significantly higher among those with PAH (38.9% vs 9.5%, respectively; p ⫽ 0.003). In both instances, PAH was more common in nonsurvivors and neither LV dysfunction nor transplant altered outcomes. 6MWT data contemporaneously obtained with both spirometry and right-heart catheterization were available in 34 subjects (43.0%) [Table 2]. Among these subjects, PAH was present in 29.4% having an mPAP of 29.8 ⫾ 3.0 mm Hg, compared to 18.2 ⫾ 3.6 mm Hg for those subjects without PAH (p ⬍ 0.001). Both the distance walked during the 6MWT and the lowest Spo2 were significantly lower in those with PAH. The mean distance walked during the 6MWT was 143.5 ⫾ 65.5 m in those subjects with PAH, compared to 365.9 ⫾ 81.8 m in those subjects without PAH (p ⬍ 0.001). The Spo2 nadir during the 6MWT was also significantly lower in those with PAH (80.1 ⫾ 3.7% vs 88.0 ⫾ 3.5%, respectively; p ⬍ 0.001). Among these patients, the presence of PAH was associated with considerably higher mortality, occurring in 70.0% of those with PAH compared to only 37.5% of those without PAH (p ⫽ 0.003). Discussion We have reported on the prevalence of PAH and its impact on survival among a large cohort of patients with advanced IPF, all of whom had pulmonary artery (PA) pressure measurements determined by right heart catheterization. Although it has been previously reported that PAH affects the prognosis of patients with IPF, most of the prior studies12,13,30,31 were smaller, did not report outcomes, relied on estimates of PA pressures, and/or included a broader mix of patients. In our cohort, PAH is present in nearly one third of patients with advanced IPF. Its presence has a significant impact on outcomes, with an almost threefold increased risk of death.
While the prevalence has not been previously defined, the association between IPF and PAH has been reported and has been found to correlate with worse outcomes. King and colleagues13 found that radiographic findings suggesting pulmonary hypertension were associated with higher mortality in patients with IPF. In patients with other ILDs, PAH has been identified1,2 as a major contributor to morbidity and mortality. In systemic sclerosis, the presence of PAH represents the greatest risk of death. Similarly, Shorr and colleagues10,11 found that PAH was associated with a worse prognosis in patients with sarcoidosis who were awaiting lung transplantation. In patients with idiopathic (primary) PAH, the degree of hemodynamic derangement correlates strongly with mortality.14 PAH is also associated with a significant increase in mortality in patients with COPD.34 LV dysfunction is the most common etiology for PAH, and many patients with IPF may also have risk factors for coexisting cardiac disease. However, the majority of patients in this study had normal pulmonary capillary wedge pressures and cardiac index values, suggesting that LV dysfunction is not a contributing factor in most individuals. We attempted to demonstrate this further by eliminating those with LV dysfunction. In this subgroup analysis, PAH was still common in patients with IPF and was strongly associated with mortality. This suggests that the mechanism of pulmonary hypertension in these patients with end-stage lung disease is related to the lung disease itself. Whether this is due to progressive fibrosis, uncorrected hypoxemia, endothelial dysfunction, cytokine derangements, genetic factors, or vascular remodeling remains uncertain. In our study, only PAH correlated with mortality. Baseline spirometric measurements did not correlate with the presence of PAH or accurately predict mortality. This differs from the results of previous studies13,17–22 that have found that age, spirometric values, and a reduced Dlco are associated with higher mortality rates. While both FVC and Dlco were lower in nonsurvivors, they did not independently predict outcomes. We found that a reduced Dlco predicted PAH, which subsequently pre-
Table 2—6MWT Measurements Between Those With and Without PAH* Variables
MAP ⱕ 25 mm Hg (n ⫽ 10)
MAP ⬎ 25 mm Hg (n ⫽ 24)
p Value
MPAP, mm Hg 6MWT distance, m Spo2 nadir on 6MWT, % Mortality rate, %
18.2 ⫾ 3.6 365.9 ⫾ 81.8 88.0 ⫾ 3.5 37.5
29.8 ⫾ 5.1 143.5 ⫾ 65.5 80.1 ⫾ 3.7 70.0
NA ⬍ 0.001 ⬍ 0.001 0.003
*Values are given as the mean ⫾ SD, unless otherwise indicated. NA ⫽ not applicable. 750
Original Research
dicted survival. The combination of a Dlco of ⬍ 40% predicted and the need for supplemental oxygen was more accurate in predicting the presence of PAH. Performance on the 6MWT also was predictive of underlying PAH, the presence of which resulted in greater desaturation and shorter distances walked. These parameters may be useful as surrogate markers for PAH when deciding which patients to refer for confirmatory right heart catheterization. Others factors that were not measured in our study have also been shown to predict outcomes in patients with IPF, including b-type (brain) natriuretic peptide and the degree of parenchymal fibrosis.31,35,36 Significant fibrosis, as seen radiographically or by histopathologic examination, may be the etiology for the subsequent development of PAH in patients with end-stage lung disease. Our findings differed from some previous reports evaluating PAH in IPF. In studies by both Timmer et al37 and Harari and colleagues,30 PA pressures were not significantly different between patients who died and those who survived to lung transplantation. Timmer et al37 found that the resting arterial oxygen content discriminated survivors from nonsurvivors.37 Similarly, we found that desaturation during the 6MWT and the need for supplemental oxygen predicted the presence of PAH. However, these were not independently associated with an increased risk for mortality. These prior studies differed with our study in terms of design, and they included a mixed population of ILD patients with limited numbers of IPF patients. These somewhat discordant findings underline the lack of understanding regarding the role of PAH in IPF patients and underscore the need for further prospective studies with serial measures of PA pressures. Understanding the prevalence and related outcomes of PAH patients may provide a new target in the evaluation, monitoring, and treatment of patients with IPF. IPF is associated with poor outcomes, with a median survival time of only 2.5 to 3.5 years following diagnosis.15,16,18,32 When PAH develops as a consequence of IPF, the risk of death appears to be even greater. Advances in the treatment of PAH due to other etiologies, particularly idiopathic PAH, have led to improved exercise tolerance, improved survival, and a decreased need for lung transplantation.33 It is therefore possible that the treatment of the underlying PAH in patients with IPF-induced pulmonary hypertension may improve both functional capacity and survival. Our study had several limitations. This was a retrospective review of patients who were evaluated at a single center. We included patients who were both referred to an IPF center and had undergone www.chestjournal.org
evaluation for transplantation, possibly reflecting the presence of those patients with more advanced IPF. The nature of our cohort may have resulted in an overestimation of the prevalence of PAH and worse outcomes through the inherent selection of patients with more advanced disease. Similarly, since only those patients undergoing evaluation for transplantation were included in the study, the average age of our population may be lower and the results might not reflect the course of older individuals. These factors may limit the applicability of the study to patients seen in non-IPF or transplant referral centers. Comorbid conditions that may affect survival were not included in this analysis. However, we included only those patients who underwent evaluation for lung transplantation, which infers a population without significant or decompensated comorbidities. Additionally, we attempted to control for the most common etiology of PAH by performing a subgroup analysis of those with concomitant LV dysfunction. We also did not correct for potential therapies used in each of the patients that may have influenced outcomes. Length of survival was measured from the time of right-heart catheterization during the initial evaluation for lung transplantation and not from the onset of symptoms or confirmation of the diagnosis. This explains the shorter survival time in our cohort compared with what has previously been described for IPF patients.15,16,18,32 Also, with time zero being the day of the cardiac catheterization, the influence of PAH on survival would not be affected by any lead-time bias. Finally, echocardiography was not obtained systematically; thus, we lacked sufficient temporally related data to perform an adequate comparison with our right heart catheterization data. In summary, the prevalence of PAH in our study population was high, affecting nearly one third of patients. The prevalence of PAH among other patient groups with IPF may be different and likely depends on multiple factors, including patient age and the duration of the disease. Coexisting PAH in patients with IPF is associated with worse outcomes and was a better prognostic marker than standard measures of lung function. In patients with IPF, especially those with more advanced disease, evaluating for the presence of PAH may be useful in determining prognosis and may have a role in monitoring the disease course, triaging for lung transplantation, and deciding on potential therapies. Surrogate markers, such as a reduced Dlco, the need for supplemental oxygen, or a poor performance on the 6MWT, should raise suspicion for the presence of PAH and the need for confirmatory right heart catheterization. The timing of this and the role of serial catheterizations remain to be determined. CHEST / 129 / 3 / MARCH, 2006
751
Future studies, preferably prospective and multicentered, are needed to validate our findings and to further assess for markers of PAH. References 1 Olschewski H, Ghofrani HA, Walmrath D, et al. Inhaled prostacyclin and iloprost in severe pulmonary hypertension secondary to lung fibrosis. Pneumologie 2000; 54:133–142 2 Lee P, Langevitz P, Alderdice CA, et al. Mortality in systemic sclerosis (scleroderma). Q J Med 1992; 82:139 –148 3 Fartoukh M, Humbert M, Capron F, et al. Severe pulmonary hypertension in histiocytosis X. Am J Respir Crit Care Med 2000; 161:216 –223 4 Koh ET, Lee P, Gladman DD, et al. Pulmonary hypertension in systemic sclerosis: an analysis of 17 patients. Br J Rheumatol 1996; 35:989 –993 5 Rolla G, Colagrande P, Scappaticci E, et al. Exhaled nitric oxide in systemic sclerosis: relationships with lung involvement and pulmonary hypertension. J Rheumatol 2000; 27: 1693–1698 6 Hinderliter AL, Willis PW 4th, Barst RJ, et al. Effects of long-term infusion of prostacyclin (epoprostenol) on echocardiographic measures of right ventricular structure and function in primary pulmonary hypertension: Primary Pulmonary Hypertension Study Group. Circulation 1997; 95:1479 –1486 7 Winslow TM, Ossipov MA, Fazio GP, et al. The left ventricle in systemic lupus erythematosus: initial observations and a five-year follow-up in a university medical center population. Am Heart J 1993; 125:1117–1122 8 Dawson JK, Goodson NG, Graham DR, et al. Raised pulmonary artery pressures measured with doppler echocardiography in rheumatoid arthritis patients. Rheumatology (Oxford) 2000; 39:1320 –1325 9 Preston IR, Klinger JR, Landzberg MJ, et al. Vasoresponsiveness of sarcoidosis-associated pulmonary hypertension. Chest 2001; 120:866 – 872 10 Shorr AF, Davies DB, Nathan SD. Outcomes for patients with sarcoidosis awaiting lung transplantation. Chest 2002; 122:233–238 11 Shorr AF, Davies DB, Nathan SD. Predicting mortality in patients with sarcoidosis awaiting lung transplantation. Chest 2003; 124:922–928 12 Arcasoy SM, Christie JD, Ferrari VA, et al. Echocardiographic assessment of pulmonary hypertension in patients with advanced lung disease. Am J Respir Crit Care Med 2003; 167:735–740 13 King TE, Tooze JA, Schwarz MI, et al. Predicting survival in idiopathic pulmonary fibrosis: scoring system and survival model. Am J Respir Crit Care Med 2001; 164:1171–1181 14 D’Alonzo GE, Barst RJ, Ayres SM, et al. Survival in patients with primary pulmonary hypertension: results from a national prospective registry. Ann Intern Med 1991; 115:343–349 15 Schwartz DA, Helmers RA, Galvin JR, et al. Determinants of survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1994; 149:450 – 454 16 Flaherty KR, King TE, Raghu G, et al. Idiopathic Interstitial Pneumonia: What is the effect of a multidisciplinary approach to diagnosis? AJRCCM 2004; 170:904 –910 17 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– 537 18 Martinez FJ, Safrin S, Weycker D, et al. The clinical course of patients with idiopathic pulmonary fibrosis. Ann Intern Med 2005; 142:963–967 752
19 Perez A, Rogers RM, Dauber JH. The prognosis of idiopathic pulmonary fibrosis. Am J Respir Cell Mol Biol 2003; 29: S19 –26 20 Mogulkoc N, Brutsche MH, Bishop PW, et al. Pulmonary function in idiopathic pulmonary fibrosis and referral for lung transplantation. Am J Respir Crit Care Med 2001; 164:103– 108 21 Thomeer MJ, Vansteenkiste J, Verbeken EK, et al. Interstitial lung diseases: characteristics at diagnosis and mortality risk assessment. Respir Med 2004; 98:567–573 22 Erbes R, Schaberg T, Loddenkemper R. Lung function tests in patients with idiopathic pulmonary fibrosis: are they helpful for predicting outcome? Chest 1997; 11:51–57 23 Lama VN, Flaherty KR, Toews GB, et al. Prognostic value of desaturation during a 6-minute walk test in idiopathic interstitial pneumonia. Am J Respir Crit Care Med 2003; 168: 1084 –1090 24 Eaton T, Young P, Milne D, et al. Six-minute walk, maximal exercise tests: reproducibility in fibrotic interstitial pneumonia. Am J Respir Crit Care Med 2005; 171:1150 –1157 25 Miyamoto S, Nagaya N, Satoh T, et al. Clinical correlates and prognostic significance of six-minute walk test in patients with primary pulmonary hypertension. Comparison with cardiopulmonary exercise testing. Am J Respir Crit Care Med 2000; 161:487– 492 26 Hallstrand TS, Boitano LJ, Johnson WC, et al. The timed walk test as a measure of severity and survival in idiopathic pulmonary fibrosis. Eur Respir J 2005; 25:96 –103 27 American Thoracic Society. Guidelines for the six-minute walk test: consensus statement. Am J Respir Crit Care Med 2002; 166:111–117 28 American Thoracic Society. Idiopathic pulmonary fibrosis: diagnosis and treatment: international consensus statement; American Thoracic Society (ATS), and the European Respiratory Society (ERS). Am J Respir Crit Care Med 2000 Feb; 161:646 – 664 29 Knudson RJ, Lebowitz MD, Holberg CJ, et al. Changes in the normal maximal expiratory flow-volume curve with growth and aging. Am Rev Respir Dis 1983; 127:725–734 30 Harari S, Simonneau G, De Juli E, et al. Prognostic value of pulmonary hypertension in patients with chronic interstitial lung disease referred for lung or heart-lung transplantation. J Heart Lung Transplant 1997; 16:460 – 463 31 Leuchte HH, Neurohr C, Baumgartner R, et al. Brain natriuretic peptide and exercise capacity in lung fibrosi and pulmonary hypertension. Am J Respir Crit Care Med 2004; 170:360 –365 32 Bjoraker JA, Ryu JH, Edwin MK, et al. Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1998; 157:199 –203 33 Hosenpud JD, Bennett LE, Keck BM, et al. Effects of diagnosis on survival benefit of lung transplantation for end-stage lung disease. Lancet 1998; 351:24 –27 34 Oswald-Mammosser M, Weitzenblum E, Quoix E, et al. Prognostic factors in COPD patients receiving long-term oxygen therapy: importance of pulmonary artery pressure. Chest 1985; 107:1193–1198 35 King TE, Schwartz MI, Brown K, et al. Idiopathic pulmonary fibrosis: relationship between histopathologic features and mortality. Am J Respir Crit Care Med 2001; 164:1025–1032 36 Gay SE, Kazerooni EA, Toews GB, et al. Idiopathic pulmonary fibrosis. predicting response to therapy and survival. Am J Respir Crit Care Med 1998; 157:1063–1072 37 Timmer SJ, Karamzadeh AM, Yung GL, et al. Predicting survival of lung transplantation candidates with idiopathic interstitial pneumonia: does Pao2 predict survival? Chest 2002; 122:779 –784 Original Research
Original Research INTERSTITIAL LUNG DISEASE
Prevalence and Outcomes of Pulmonary Arterial Hypertension in Advanced Idiopathic Pulmonary Fibrosis* Christopher J. Lettieri, MD; Steven D. Nathan, MD, FCCP; Scott D. Barnett, PhD; Shahzad Ahmad, MD; and Andrew F. Shorr, MD, MPH, FCCP
Study objectives: The development of pulmonary arterial hypertension (PAH) can complicate many interstitial lung diseases, including idiopathic pulmonary fibrosis (IPF). We sought to characterize the prevalence of PAH and its impact on survival in patients with advanced IPF. Design: Retrospective analysis of consecutive IPF patients undergoing pretransplantation right heart catheterization. Setting: Lung transplant and IPF referral center. Methods: PAH was defined as a mean pulmonary artery pressure (mPAP) of > 25 mm Hg. We compared demographic, spirometric, 6-min walk test (6MWT) results, and survival outcomes between those with PAH and those without PAH. Measurements and results: Seventy-nine patients were included in the study. PAH was present in 31.6% of patients (mean [ⴞ SD] mPAP, 29.5 ⴞ 3.3 vs 19.1 ⴞ 3.7 mm Hg, respectively). Those patients with PAH had a lower mean diffusing capacity of the lung for carbon monoxide (DLCO) (37.6 ⴞ 11.3% vs 31.1 ⴞ 10.1%, respectively; p ⴝ 0.04) and were more likely to require supplemental oxygen (66.7% vs 17.6%, respectively; p < 0.0001). Mean distance walked (143.5 ⴞ 65.5 vs 365.9 ⴞ 81.8 m, respectively; p < 0.001) and mean pulse oximetric saturation nadir (80.1 ⴞ 3.7% vs 88.0 ⴞ 3.5%, respectively; p < 0.001) during the 6MWT were also lower among those with PAH. PAH was associated with a greater risk of death during the study period (mortality rate, 60.0% vs 29.9%, respectively; odds ratio, 2.6; 95% confidence interval [CI], 2.3 to 3.1; p ⴝ 0.001). One-year mortality rates were higher in those with PAH (28.0% vs 5.5%, respectively; p ⴝ 0.002). As a predictor of mortality, PAH had a sensitivity, specificity, and accuracy of 57.1%, 79.3%, and 73.4%, respectively. There was a linear correlation between mPAP and outcomes with higher pressures associated with a greater risk of mortality (hazard ratio, 1.09; 95% CI, 1.02 to 1.16). FVC and DLCO did not predict outcomes. Conclusions: PAH is common in advanced cases of IPF and significantly impacts survival. A reduced DLCO, supplemental oxygen requirement, or poor 6-min walk performance should raise suspicion of the presence of underlying PAH. Identifying PAH might be an important adjunct in monitoring disease progression, triaging for transplantation, and guiding therapy. (CHEST 2006; 129:746 –752) Key words: catheterization; exercise test; mortality; pulmonary artery; pulmonary fibrosis; pulmonary function tests; pulmonary hypertension Abbreviations: CI ⫽ confidence interval; Dlco ⫽ diffusing capacity of the lung for carbon monoxide; HR ⫽ hazard ratio; ILD ⫽ interstitial lung disease; IPF ⫽ idiopathic pulmonary fibrosis; LV ⫽ left ventricular; mPAP ⫽ mean pulmonary arterial pressure; PA ⫽ pulmonary artery; PAH ⫽ pulmonary arterial hypertension; PAWP ⫽ pulmonary artery wedge pressure; PFT ⫽ pulmonary function test; 6MWT ⫽ 6-min walk test; Spo2 ⫽ pulse oximetric saturation; TLC ⫽ total lung capacity
forms of diffuse parenchymal lung disease M any can result in the development of pulmonary arterial hypertension (PAH).1–11 For example, Arcasoy and coworkers12 found PAH in one quarter of 746
patients with various forms of advanced lung disease who were referred for transplantation. Disease-specific PAH has been described as occurring in 5 to 38% of patients with scleroderma, 4.3 to 43% of Original Research
patients with systemic lupus erythematosus, and 21% of patients with rheumatoid arthritis.4 – 8 Sarcoidosis is also associated with PAH in 1 to 28% of cases and is more prevalent in patients with more advanced disease.9 PAH associated with interstitial lung disease (ILD) has been found to be a significant predictor of mortality.13,14 PAH has been reported in patients with idiopathic pulmonary fibrosis (IPF), but the prevalence has not been well-defined. In an analysis of the United Network of Organ Sharing registry, Shorr and colleagues10,11 found that approximately one quarter of ⬎ 2,000 patients with IPF who were listed for lung transplantation had PAH. While this study was not designed to look specifically at patients with IPF, it did suggest that the incidence of PAH is significant in this population. A better understanding of the prevalence of PAH in patients with IPF is important. Routine measurements from pulmonary function tests (PFTs) may not accurately correlate with survival in patients with IPF.13–22 PAH, on the other hand, is associated with worse outcomes in patients with other advanced lung diseases and may have similar consequences in patients with IPF. A possible explanation is that underlying PAH is not reflected by these standard measures. Patients with IPF often have functional limitations and a decreased exercise capacity, as measured by the 6-min walk test (6MWT), which has been shown to correlate with outcomes better than standard PFTs and may reflect coexistent pulmonary hypertension.23–26 Underlying PAH may explain why survival correlates better with functional limitations than measurements of restrictive lung physiology. Understanding and characterizing the relationship between PAH and IPF may provide important prognostic information, may assist with triaging for transplantation, and may potentially provide a target for therapy. *From the Department of Pulmonary & Critical Care Medicine (Dr. Lettieri), Walter Reed Army Medical Center, Washington DC; the Lung Transplant Program (Drs. Nathan, Barnett, and Ahmad), Inova Fairfax Hospital, Falls Church, VA; and the Department of Pulmonary and Critical Care Medicine (Dr. Shorr), Washington Hospital Center, Washington, DC. The authors have no potential financial conflicts of interests to disclose. The opinions expressed herein are not to be construed as official or as reflecting the policies of either the Department of the Army or the Department of Defense. Manuscript received August 2, 2005; revision accepted September 14, 2005. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Christopher J. Lettieri, MD, Walter Reed Army Medical Center, Department of Pulmonary & Critical Care, 6900 Georgia Ave, NW, Washington, DC 20307; e-mail: [email protected] www.chestjournal.org
We hypothesized that PAH is common in patients with more advanced IPF and may be an independent risk factor for mortality. We attempted to define this association using a cohort of patients with IPF who underwent right heart catheterization as part of an evaluation for lung transplantation.
Materials and Methods Subjects We retrospectively reviewed all patients with IPF who were seen at our clinic between June 1998 and December 2004. Our clinic serves as both an IPF and lung transplant referral center. We included patients who had undergone right-sided cardiac catheterization as part of their initial evaluation prior to being listed for lung transplantation. Spirometry obtained within 1 month of the catheterization was required for inclusion in the study. When available, we included data from the 6MWT performed within 1 month of the catheterization. The 6MWTs were performed while the patient breathed room air and were conducted in accordance with the recommendations of the American Thoracic Society.27 All subjects met accepted diagnostic criteria for IPF per the American Thoracic Society/European Respiratory Society guidelines and had histopathologic evidence of usual interstitial pneumonia.28 Although these guidelines were not published until 2000, we retrospectively verified that all included subjects fulfilled the criteria outlined in these guidelines. Measurements We recorded demographic data and PFT results, and estimated the left ventricular (LV) ejection fraction from echocardiography and cardiac catheterization data. PFT measurements included FVC, total lung capacity (TLC), and diffusing capacity of the lung for carbon monoxide (Dlco). The prediction equations of Knudson et al,29 with appropriate demographic adjustments, were employed. For the 6MWT analysis, only the studies in which patients ambulated while breathing room air were included, and therefore patients with a pulse oximetric saturation (Spo2) of ⬍ 88% were excluded from the study. This was done in order to homogenize the data obtained from the 6MWTs. The 6MWT data included the total distance walked (in meters) and the lowest Spo2 recorded by noninvasive pulse oximetry. The test was terminated if the Spo2 fell to ⬍ 80%. Right-sided cardiac catheterization data included measurements of right atrial pressure, pulmonary arterial pressures, pulmonary artery wedge pressure (PAWP), and cardiac index. Measurements obtained from cardiac catheterization were obtained from the patient during a resting state. We defined PAH as a resting mean pulmonary arterial pressure (mPAP) of ⬎ 25 mm Hg. Those with a mPAP values ⱕ 25 mm Hg were considered to be normal. We defined LV dysfunction as an LV ejection fraction of ⬍ 50% or a PAWP of ⬎ 14 mm Hg. Further evaluation for other secondary causes of PAH (ie, polysomnography or ventilation-perfusion scanning) was conducted when clinically indicated. End Points The primary end point of this study was the prevalence of PAH. Survival to lung transplantation or the end of the observation period represented the secondary end point, and patients CHEST / 129 / 3 / MARCH, 2006
747
were censored at the timing of either of these events. Mortality status was determined in December 2004 and was verified using the US Social Security Death Index. We explored whether demographics, cardiac parameters, PFT measurements, or 6MWT parameters correlated with the incidence of PAH. Statistical Analysis We initially compared continuous variables with the Student t test and analyzed categoric variables with the Fisher exact test. The survival analysis was completed according to the methods of Kaplan and Meier, and the log rank test was used to compare survival curves. Cox proportional hazards regression models were used to calculate hazard ratios (HRs) for mortality, including adjustments for FVC and Dlco. The data are presented as the mean ⫾ SD. All tests were two-tailed, and p values of ⬍ 0.05 were assumed to represent statistical significance. All analyses were completed using a statistical software package (SPSS, version 12.0; SPSS, Inc; Chicago, IL).
Results During the study period, 79 patients with IPF who were evaluated in our clinic were referred for lung transplant evaluation and underwent right-heart catheterization. These individuals comprised the final cohort included in this analysis. No subjects were lost to follow-up, and none were excluded. The mean mPAP was 23.4 ⫾ 5.1 mm Hg (range, 8 to 46 mm Hg) among the cohort (Fig 1). The criteria for PAH were present in 25 subjects (31.6%), with a mean mPAP of 29.5 ⫾ 3.3 mm Hg, compared with 19.1 ⫾ 3.7 mm Hg among those who did not meet the definition. Table 1 displays the characteristics of patients with and without PAH. Age, gender, FVC, TLC, and cardiac index did not differ between those with or without PAH. The PFT values used for this analysis were obtained within a mean period of 24.2 ⫾ 6.7 days11–34 after right heart catheterization. Dlco was significantly lower in those with PAH (37.6 ⫾ 11.3% vs 31.1 ⫾ 10.1% predicted, respectively; p ⫽ 0.04). Among patients in the cohort, 34.6% of subjects with IPF required supplemental oxygen, which was more common in those with PAH (66.7% vs 17.6%, respectively; p ⬍ 0.0001). The need for supplemental oxygen was determined by either a resting Spo2 of ⬍ 88% or a Pao2 of ⬍ 55 mm Hg. Among patients in the cohort, 15.2% had both a supplemental oxygen requirement and Dlco of ⬍40% predicted. These individuals were 10.2 times more likely to have underlying PAH than those with a Dlco of ⱖ 40% predicted who did not require supplemental oxygen (p ⬍ 0.001). The need for supplemental oxygen together with a Dlco of ⬍ 40% predicted identified the presence of PAH with a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 65.0%, 94.1%, 86.7%, 82.1%, and 748
Figure 1. mPAP values. Histogram displaying the distribution of mPAPs among the cohort (values expressed in mm Hg). mPAP ⫽ mean pulmonary arterial pressure
83.3%, respectively. The 1-year mortality rate for individuals with both of these factors was 50.0%, compared with 21.1% for those patients with neither (p ⫽ 0.02). PAH was present in 52.4% of nonsurvivors compared to only 24.1% of survivors (p ⫽ 0.008). During the observation period, 60.0% of those patients with PAH died compared to only 29.9% of those without PAH (odds ratio, 2.6; 95% confidence interval [CI], 2.3 to 3.1; p ⫽ 0.001). One-year mortality rates were similarly higher among those with PAH (28.0% vs 5.5%, respectively; p ⫽ 0.002). Figure 2 shows the Kaplan-Meier curve, which revealed an early and significant difference in survival between those with and without PAH. The presence of PAH predicted mortality with a sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 57.1%, 79.3%, 50.0%, 83.6%, and 73.4%, respectively. There was a linear correlation between mPAP and outcomes, with higher pressures associated with a greater risk of mortality (HR, 1.09; 95% CI, 1.02 to 1.16). In other words, the higher the mPAP, the greater the subsequent risk of mortality. After adjustment for both FVC (HR, 1.10; 95% CI, 1.02 to 1.19) and Dlco (HR, 1.07; 95% CI, 1.01 to 1.18), increasing mPAP remained a significant predictor of death. Original Research
Table 1—Patient Characteristics Based on the Presence or Absence of PAH* Characteristics
MAP ⱕ 25 mm Hg (n ⫽ 54)
MAP ⬎ 25 mm Hg (n ⫽ 25)
p Value
Age, yr Male gender, % Supplemental oxygen, % FVC, % predicted TLC, % predicted Dlco, % predicted mPAP, mm Hg RAP, mm Hg Cardiac index, L/min/m2 PAWP, mm Hg Ejection fraction, % Mortality rate, % 1-yr mortality rate, %
56.2 ⫾ 1.0 72.2 17.6 52.5 ⫾ 11.9 55.5 ⫾ 10.7 37.6 ⫾ 11.3 19.1 ⫾ 3.7 4.3 ⫾ 2.8 2.8 ⫾ 0.4 8.4 ⫾ 3.5 61.5 ⫾ 5.9 29.9 5.5
54.7 ⫾ 3.8 64.0 66.7 49.3 ⫾ 11.0 57.8 ⫾ 12.1 31.1 ⫾ 10.1 29.5 ⫾ 3.3 5.3 ⫾ 3.5 3.2 ⫾ 1.3 9.3 ⫾ 3.5 59.4 ⫾ 5.0 60.0 28.8
0.27 0.23 ⬍ 0.001 0.13 0.27 0.04 n/a 0.14 0.20 0.22 0.12 0.001 0.002
*Values are given as the mean ⫾ SD, unless otherwise indicated. RAP ⫽ right atrial pressure; MAP ⫽ mean arterial pressure.
In addition to mPAP, the measured spirometric and cardiac parameters were also analyzed as potential predictors of mortality within 1, 2, or 3 years from evaluation. FVC, TLC, and cardiac index did not differ between survivors and nonsurvivors. At each time period, only the presence of PAH differentiated survivors from nonsurvivors (1 year, p ⫽ 0.027; 2 years, p ⫽ 0.0005; 3 years, p ⫽ 0.018). While Dlco and the need for supplemental oxygen predicted the presence of PAH, they did not determine outcomes over the ensuing 1, 2, or 3 years.
To account for any potential bias, we reanalyzed the measured variables as functions of survival after eliminating 22 subjects who underwent lung transplantation. Of the remaining 57 patients, PAH was present in 33.3%. Although Dlco was significantly lower in those with PAH, only PAH independently predicted survival (p ⬍ 0.0001). Among patients in this subpopulation, the 1-year mortality rate was significantly greater in those with PAH compared to those without PAH (36.8% vs 7.9%, respectively; p ⫽ 0.003). Similarly, we reexplored survival after
Figure 2. PAH as a predictor of survival in patients with IPF. www.chestjournal.org
CHEST / 129 / 3 / MARCH, 2006
749
eliminating 19 subjects with LV dysfunction to ensure that congestive heart failure was not a potential confounding factor. In this instance, Dlco predicted PAH, but only PAH independently predicted survival (p ⬍ 0.0001). Again, 1-year mortality rates were significantly higher among those with PAH (38.9% vs 9.5%, respectively; p ⫽ 0.003). In both instances, PAH was more common in nonsurvivors and neither LV dysfunction nor transplant altered outcomes. 6MWT data contemporaneously obtained with both spirometry and right-heart catheterization were available in 34 subjects (43.0%) [Table 2]. Among these subjects, PAH was present in 29.4% having an mPAP of 29.8 ⫾ 3.0 mm Hg, compared to 18.2 ⫾ 3.6 mm Hg for those subjects without PAH (p ⬍ 0.001). Both the distance walked during the 6MWT and the lowest Spo2 were significantly lower in those with PAH. The mean distance walked during the 6MWT was 143.5 ⫾ 65.5 m in those subjects with PAH, compared to 365.9 ⫾ 81.8 m in those subjects without PAH (p ⬍ 0.001). The Spo2 nadir during the 6MWT was also significantly lower in those with PAH (80.1 ⫾ 3.7% vs 88.0 ⫾ 3.5%, respectively; p ⬍ 0.001). Among these patients, the presence of PAH was associated with considerably higher mortality, occurring in 70.0% of those with PAH compared to only 37.5% of those without PAH (p ⫽ 0.003). Discussion We have reported on the prevalence of PAH and its impact on survival among a large cohort of patients with advanced IPF, all of whom had pulmonary artery (PA) pressure measurements determined by right heart catheterization. Although it has been previously reported that PAH affects the prognosis of patients with IPF, most of the prior studies12,13,30,31 were smaller, did not report outcomes, relied on estimates of PA pressures, and/or included a broader mix of patients. In our cohort, PAH is present in nearly one third of patients with advanced IPF. Its presence has a significant impact on outcomes, with an almost threefold increased risk of death.
While the prevalence has not been previously defined, the association between IPF and PAH has been reported and has been found to correlate with worse outcomes. King and colleagues13 found that radiographic findings suggesting pulmonary hypertension were associated with higher mortality in patients with IPF. In patients with other ILDs, PAH has been identified1,2 as a major contributor to morbidity and mortality. In systemic sclerosis, the presence of PAH represents the greatest risk of death. Similarly, Shorr and colleagues10,11 found that PAH was associated with a worse prognosis in patients with sarcoidosis who were awaiting lung transplantation. In patients with idiopathic (primary) PAH, the degree of hemodynamic derangement correlates strongly with mortality.14 PAH is also associated with a significant increase in mortality in patients with COPD.34 LV dysfunction is the most common etiology for PAH, and many patients with IPF may also have risk factors for coexisting cardiac disease. However, the majority of patients in this study had normal pulmonary capillary wedge pressures and cardiac index values, suggesting that LV dysfunction is not a contributing factor in most individuals. We attempted to demonstrate this further by eliminating those with LV dysfunction. In this subgroup analysis, PAH was still common in patients with IPF and was strongly associated with mortality. This suggests that the mechanism of pulmonary hypertension in these patients with end-stage lung disease is related to the lung disease itself. Whether this is due to progressive fibrosis, uncorrected hypoxemia, endothelial dysfunction, cytokine derangements, genetic factors, or vascular remodeling remains uncertain. In our study, only PAH correlated with mortality. Baseline spirometric measurements did not correlate with the presence of PAH or accurately predict mortality. This differs from the results of previous studies13,17–22 that have found that age, spirometric values, and a reduced Dlco are associated with higher mortality rates. While both FVC and Dlco were lower in nonsurvivors, they did not independently predict outcomes. We found that a reduced Dlco predicted PAH, which subsequently pre-
Table 2—6MWT Measurements Between Those With and Without PAH* Variables
MAP ⱕ 25 mm Hg (n ⫽ 10)
MAP ⬎ 25 mm Hg (n ⫽ 24)
p Value
MPAP, mm Hg 6MWT distance, m Spo2 nadir on 6MWT, % Mortality rate, %
18.2 ⫾ 3.6 365.9 ⫾ 81.8 88.0 ⫾ 3.5 37.5
29.8 ⫾ 5.1 143.5 ⫾ 65.5 80.1 ⫾ 3.7 70.0
NA ⬍ 0.001 ⬍ 0.001 0.003
*Values are given as the mean ⫾ SD, unless otherwise indicated. NA ⫽ not applicable. 750
Original Research
dicted survival. The combination of a Dlco of ⬍ 40% predicted and the need for supplemental oxygen was more accurate in predicting the presence of PAH. Performance on the 6MWT also was predictive of underlying PAH, the presence of which resulted in greater desaturation and shorter distances walked. These parameters may be useful as surrogate markers for PAH when deciding which patients to refer for confirmatory right heart catheterization. Others factors that were not measured in our study have also been shown to predict outcomes in patients with IPF, including b-type (brain) natriuretic peptide and the degree of parenchymal fibrosis.31,35,36 Significant fibrosis, as seen radiographically or by histopathologic examination, may be the etiology for the subsequent development of PAH in patients with end-stage lung disease. Our findings differed from some previous reports evaluating PAH in IPF. In studies by both Timmer et al37 and Harari and colleagues,30 PA pressures were not significantly different between patients who died and those who survived to lung transplantation. Timmer et al37 found that the resting arterial oxygen content discriminated survivors from nonsurvivors.37 Similarly, we found that desaturation during the 6MWT and the need for supplemental oxygen predicted the presence of PAH. However, these were not independently associated with an increased risk for mortality. These prior studies differed with our study in terms of design, and they included a mixed population of ILD patients with limited numbers of IPF patients. These somewhat discordant findings underline the lack of understanding regarding the role of PAH in IPF patients and underscore the need for further prospective studies with serial measures of PA pressures. Understanding the prevalence and related outcomes of PAH patients may provide a new target in the evaluation, monitoring, and treatment of patients with IPF. IPF is associated with poor outcomes, with a median survival time of only 2.5 to 3.5 years following diagnosis.15,16,18,32 When PAH develops as a consequence of IPF, the risk of death appears to be even greater. Advances in the treatment of PAH due to other etiologies, particularly idiopathic PAH, have led to improved exercise tolerance, improved survival, and a decreased need for lung transplantation.33 It is therefore possible that the treatment of the underlying PAH in patients with IPF-induced pulmonary hypertension may improve both functional capacity and survival. Our study had several limitations. This was a retrospective review of patients who were evaluated at a single center. We included patients who were both referred to an IPF center and had undergone www.chestjournal.org
evaluation for transplantation, possibly reflecting the presence of those patients with more advanced IPF. The nature of our cohort may have resulted in an overestimation of the prevalence of PAH and worse outcomes through the inherent selection of patients with more advanced disease. Similarly, since only those patients undergoing evaluation for transplantation were included in the study, the average age of our population may be lower and the results might not reflect the course of older individuals. These factors may limit the applicability of the study to patients seen in non-IPF or transplant referral centers. Comorbid conditions that may affect survival were not included in this analysis. However, we included only those patients who underwent evaluation for lung transplantation, which infers a population without significant or decompensated comorbidities. Additionally, we attempted to control for the most common etiology of PAH by performing a subgroup analysis of those with concomitant LV dysfunction. We also did not correct for potential therapies used in each of the patients that may have influenced outcomes. Length of survival was measured from the time of right-heart catheterization during the initial evaluation for lung transplantation and not from the onset of symptoms or confirmation of the diagnosis. This explains the shorter survival time in our cohort compared with what has previously been described for IPF patients.15,16,18,32 Also, with time zero being the day of the cardiac catheterization, the influence of PAH on survival would not be affected by any lead-time bias. Finally, echocardiography was not obtained systematically; thus, we lacked sufficient temporally related data to perform an adequate comparison with our right heart catheterization data. In summary, the prevalence of PAH in our study population was high, affecting nearly one third of patients. The prevalence of PAH among other patient groups with IPF may be different and likely depends on multiple factors, including patient age and the duration of the disease. Coexisting PAH in patients with IPF is associated with worse outcomes and was a better prognostic marker than standard measures of lung function. In patients with IPF, especially those with more advanced disease, evaluating for the presence of PAH may be useful in determining prognosis and may have a role in monitoring the disease course, triaging for lung transplantation, and deciding on potential therapies. Surrogate markers, such as a reduced Dlco, the need for supplemental oxygen, or a poor performance on the 6MWT, should raise suspicion for the presence of PAH and the need for confirmatory right heart catheterization. The timing of this and the role of serial catheterizations remain to be determined. CHEST / 129 / 3 / MARCH, 2006
751
Future studies, preferably prospective and multicentered, are needed to validate our findings and to further assess for markers of PAH. References 1 Olschewski H, Ghofrani HA, Walmrath D, et al. Inhaled prostacyclin and iloprost in severe pulmonary hypertension secondary to lung fibrosis. Pneumologie 2000; 54:133–142 2 Lee P, Langevitz P, Alderdice CA, et al. Mortality in systemic sclerosis (scleroderma). Q J Med 1992; 82:139 –148 3 Fartoukh M, Humbert M, Capron F, et al. Severe pulmonary hypertension in histiocytosis X. Am J Respir Crit Care Med 2000; 161:216 –223 4 Koh ET, Lee P, Gladman DD, et al. Pulmonary hypertension in systemic sclerosis: an analysis of 17 patients. Br J Rheumatol 1996; 35:989 –993 5 Rolla G, Colagrande P, Scappaticci E, et al. Exhaled nitric oxide in systemic sclerosis: relationships with lung involvement and pulmonary hypertension. J Rheumatol 2000; 27: 1693–1698 6 Hinderliter AL, Willis PW 4th, Barst RJ, et al. Effects of long-term infusion of prostacyclin (epoprostenol) on echocardiographic measures of right ventricular structure and function in primary pulmonary hypertension: Primary Pulmonary Hypertension Study Group. Circulation 1997; 95:1479 –1486 7 Winslow TM, Ossipov MA, Fazio GP, et al. The left ventricle in systemic lupus erythematosus: initial observations and a five-year follow-up in a university medical center population. Am Heart J 1993; 125:1117–1122 8 Dawson JK, Goodson NG, Graham DR, et al. Raised pulmonary artery pressures measured with doppler echocardiography in rheumatoid arthritis patients. Rheumatology (Oxford) 2000; 39:1320 –1325 9 Preston IR, Klinger JR, Landzberg MJ, et al. Vasoresponsiveness of sarcoidosis-associated pulmonary hypertension. Chest 2001; 120:866 – 872 10 Shorr AF, Davies DB, Nathan SD. Outcomes for patients with sarcoidosis awaiting lung transplantation. Chest 2002; 122:233–238 11 Shorr AF, Davies DB, Nathan SD. Predicting mortality in patients with sarcoidosis awaiting lung transplantation. Chest 2003; 124:922–928 12 Arcasoy SM, Christie JD, Ferrari VA, et al. Echocardiographic assessment of pulmonary hypertension in patients with advanced lung disease. Am J Respir Crit Care Med 2003; 167:735–740 13 King TE, Tooze JA, Schwarz MI, et al. Predicting survival in idiopathic pulmonary fibrosis: scoring system and survival model. Am J Respir Crit Care Med 2001; 164:1171–1181 14 D’Alonzo GE, Barst RJ, Ayres SM, et al. Survival in patients with primary pulmonary hypertension: results from a national prospective registry. Ann Intern Med 1991; 115:343–349 15 Schwartz DA, Helmers RA, Galvin JR, et al. Determinants of survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1994; 149:450 – 454 16 Flaherty KR, King TE, Raghu G, et al. Idiopathic Interstitial Pneumonia: What is the effect of a multidisciplinary approach to diagnosis? AJRCCM 2004; 170:904 –910 17 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– 537 18 Martinez FJ, Safrin S, Weycker D, et al. The clinical course of patients with idiopathic pulmonary fibrosis. Ann Intern Med 2005; 142:963–967 752
19 Perez A, Rogers RM, Dauber JH. The prognosis of idiopathic pulmonary fibrosis. Am J Respir Cell Mol Biol 2003; 29: S19 –26 20 Mogulkoc N, Brutsche MH, Bishop PW, et al. Pulmonary function in idiopathic pulmonary fibrosis and referral for lung transplantation. Am J Respir Crit Care Med 2001; 164:103– 108 21 Thomeer MJ, Vansteenkiste J, Verbeken EK, et al. Interstitial lung diseases: characteristics at diagnosis and mortality risk assessment. Respir Med 2004; 98:567–573 22 Erbes R, Schaberg T, Loddenkemper R. Lung function tests in patients with idiopathic pulmonary fibrosis: are they helpful for predicting outcome? Chest 1997; 11:51–57 23 Lama VN, Flaherty KR, Toews GB, et al. Prognostic value of desaturation during a 6-minute walk test in idiopathic interstitial pneumonia. Am J Respir Crit Care Med 2003; 168: 1084 –1090 24 Eaton T, Young P, Milne D, et al. Six-minute walk, maximal exercise tests: reproducibility in fibrotic interstitial pneumonia. Am J Respir Crit Care Med 2005; 171:1150 –1157 25 Miyamoto S, Nagaya N, Satoh T, et al. Clinical correlates and prognostic significance of six-minute walk test in patients with primary pulmonary hypertension. Comparison with cardiopulmonary exercise testing. Am J Respir Crit Care Med 2000; 161:487– 492 26 Hallstrand TS, Boitano LJ, Johnson WC, et al. The timed walk test as a measure of severity and survival in idiopathic pulmonary fibrosis. Eur Respir J 2005; 25:96 –103 27 American Thoracic Society. Guidelines for the six-minute walk test: consensus statement. Am J Respir Crit Care Med 2002; 166:111–117 28 American Thoracic Society. Idiopathic pulmonary fibrosis: diagnosis and treatment: international consensus statement; American Thoracic Society (ATS), and the European Respiratory Society (ERS). Am J Respir Crit Care Med 2000 Feb; 161:646 – 664 29 Knudson RJ, Lebowitz MD, Holberg CJ, et al. Changes in the normal maximal expiratory flow-volume curve with growth and aging. Am Rev Respir Dis 1983; 127:725–734 30 Harari S, Simonneau G, De Juli E, et al. Prognostic value of pulmonary hypertension in patients with chronic interstitial lung disease referred for lung or heart-lung transplantation. J Heart Lung Transplant 1997; 16:460 – 463 31 Leuchte HH, Neurohr C, Baumgartner R, et al. Brain natriuretic peptide and exercise capacity in lung fibrosi and pulmonary hypertension. Am J Respir Crit Care Med 2004; 170:360 –365 32 Bjoraker JA, Ryu JH, Edwin MK, et al. Prognostic significance of histopathologic subsets in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1998; 157:199 –203 33 Hosenpud JD, Bennett LE, Keck BM, et al. Effects of diagnosis on survival benefit of lung transplantation for end-stage lung disease. Lancet 1998; 351:24 –27 34 Oswald-Mammosser M, Weitzenblum E, Quoix E, et al. Prognostic factors in COPD patients receiving long-term oxygen therapy: importance of pulmonary artery pressure. Chest 1985; 107:1193–1198 35 King TE, Schwartz MI, Brown K, et al. Idiopathic pulmonary fibrosis: relationship between histopathologic features and mortality. Am J Respir Crit Care Med 2001; 164:1025–1032 36 Gay SE, Kazerooni EA, Toews GB, et al. Idiopathic pulmonary fibrosis. predicting response to therapy and survival. Am J Respir Crit Care Med 1998; 157:1063–1072 37 Timmer SJ, Karamzadeh AM, Yung GL, et al. Predicting survival of lung transplantation candidates with idiopathic interstitial pneumonia: does Pao2 predict survival? Chest 2002; 122:779 –784 Original Research