0
Correspondence |

Utility of CT for Predicting Pulmonary Hypertension in Patients With Parenchymal Lung Disease Response: Similar Results, Different Conclusion? FREE TO VIEW

Hubert Chen, MD, MPH; Teresa De Marco, MD; Jeffrey A. Golden, MD, FCCP; Michael K. Gould, MD, FCCP
Author and Funding Information

Affiliations: University of California, San Francisco, San Francisco, CA,  Stanford University School of Medicine, Palo Alto, CA,  David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA

Correspondence to: Hubert Chen, MD, MPH, 350 Parnassus Ave, Suite 609, San Francisco, CA 94143; e-mail: hubert.chen@ucsf.ed



Chest. 2008;133(4):1053-1054. doi:10.1378/chest.07-2761
Text Size: A A A
Published online

In the September issue of CHEST, Zisman et al1studied the ability of high-resolution CT to predict pulmonary hypertension (PH) in patients with idiopathic pulmonary fibrosis (IPF). They found no correlation between main pulmonary artery diameter (MPAD) and PH by right-heart catheterization, and concluded that “MPAD cannot be used to screen for PH in advanced IPF patients.” In contrast, Tan et al2 studied a similar research question, finding that a MPAD ≥ 29 mm on CT had a sensitivity of 84% and specificity of 75%. They concluded that “MPAD has excellent diagnostic value for detection of PH in patients with advanced lung disease.” How are clinicians to reconcile these apparently conflicting findings?

In fact, Zisman et al1and Tan et al2 reported no significant correlation between MPAD and PH (r = 0.148 and r = 0.124, respectively). Tan et al,,2 however, assessed the diagnostic value of using a MPAD ≥ 29 mm to predict PH. We applied a similar approach to the data recently published by Zisman et al.1 Below is a reproduction of Figure 2 from their article (Fig 1 ), with thresholds for MPAD ≥ 29 mm and mean pulmonary artery pressure > 25 mm Hg added, and each quadrant labeled.

Using this approach, we estimate a sensitivity of 86 to 92% and specificity of 29 to 31%. Such results are comparable to those reported by Tan et al,2 to the extent that both studies demonstrate reasonably good sensitivity. Which conclusion is then correct?

Ultimately, such results must be interpreted in the context of the clinical question. In the case of advanced IPF, the estimated positive predictive value of using MPAD on CT to predict PH appears to be poor (50 to 55%). The estimated negative predictive value, however, is not entirely without merit (67 to 83%). Therefore, in combination with other noninvasive studies, CT may still have some utility in helping clinicians identify those patients in whom right-heart catheterization is either unnecessary or low yield.

The authors have no conflicts of interest to disclose.

Figure Jump LinkFigure 1. Reproduced with permission from Zisman et al.1 MPAP = mean pulmonary artery pressure; NS = not significant.Grahic Jump Location

The authors have no conflicts of interest to disclose.

Zisman, DA, Karlamangla, AS, Ross, DJ, et al (2007) High-resolution chest CT findings do not predict the presence of pulmonary hypertension in advanced idiopathic pulmonary fibrosis.Chest132,773-779. [PubMed] [CrossRef]
 
Tan, RT, Kuzo, R, Goodman, LR, et al Utility of CT scan evaluation for predicting pulmonary hypertension in patients with parenchymal lung disease: medical College of Wisconsin Lung Transplant Group.Chest1998;113,1250-1256. [PubMed]
 
To the Editor:

Chen and colleagues1examined the diagnostic value of mean pulmonary artery diameter (MPAD) ≥ 29 mm to predict pulmonary hypertension using Figure 2 from our study.2Given the relatively high sensitivity and negative predictive value obtained by this approach, they astutely point out that CT may have a role in screening for pulmonary hypertension in patients with idiopathic pulmonary fibrosis. This assertion has merit; however, any test can be made sensitive by selecting a low-enough cutoff but, as in this case, at the expense of specificity. Figure 2 only shows 56 of the 65 observations (9 are superimposed). With the entire data set, a MPAD cutoff ≥ 29 mm had sensitivity, specificity, positive predictive value, and negative predictive value for pulmonary hypertension (defined as mean pulmonary artery pressure [MPAP] from right-heart catheterization > 25 mm Hg) of 85 to 100%, 10 to 38%, 34 to 61%, and 60 to 100%, respectively. Only 8 of 65 patients (12%) in the data set had MPADs < 29 mm, and this is reflected by a wide confidence interval for the negative predictive value estimate. Accordingly, the negative predictive value could be as low as 60%; in our opinion, this figure is too small to accept MPAD as a screening tool for a condition with potentially devastating consequences in idiopathic pulmonary fibrosis. Furthermore, as reflected by an unacceptably low specificity, of those without pulmonary hypertension, only 8 of 38 patients (21%) had MPADs < 29 mm. Simpler and less expensive clinical measurements (oxygen saturation using pulse oximetry, percentage of predicted FVC, and percentage of predicted diffusion capacity of the lung for carbon monoxide) can be used to screen for pulmonary hypertension in patients with idiopathic pulmonary fibrosis.34 Recently, we validated a formula to predict MPAP from standard lung function measurements.4 A formula-predicted MPAP > 21 mm Hg was associated with a sensitivity, specificity, positive predictive value, and negative predictive value of 74 to 99%, 41 to 73%, 35 to 70%, and 78 to 100%, respectively, for pulmonary hypertension defined as MPAP from right-heart catheterization > 25 mm Hg. We recommend this method over MPAD from CT to screen for pulmonary hypertension in patients with advanced idiopathic pulmonary fibrosis.

References
Chen, H, Gould, MK, De Marco, T, et al Utility of CT for predicting pulmonary hypertension in patients with parenchymal lung disease: similar results, different conclusion?Chest2008;133,1053-1054. [PubMed] [CrossRef]
 
Zisman, DA, Karlamangla, A, Ross, DJ, et al High-resolution chest computed tomography findings do not predict the presence of pulmonary hypertension in advanced idiopathic pulmonary fibrosis.Chest2007;132,773-779. [PubMed]
 
Zisman, DA, Ross, DJ, Belperio, JA, et al Prediction of pulmonary hypertension in idiopathic pulmonary fibrosis.Respir Med2007;101,2153-2159. [PubMed]
 
Zisman, DA, Karlamangla, A, Kawut, SM, et al Validation of a method to screen for pulmonary hypertension in advanced idiopathic pulmonary fibrosis.Chest2008;133,640-645. [PubMed]
 

Figures

Figure Jump LinkFigure 1. Reproduced with permission from Zisman et al.1 MPAP = mean pulmonary artery pressure; NS = not significant.Grahic Jump Location

Tables

References

Zisman, DA, Karlamangla, AS, Ross, DJ, et al (2007) High-resolution chest CT findings do not predict the presence of pulmonary hypertension in advanced idiopathic pulmonary fibrosis.Chest132,773-779. [PubMed] [CrossRef]
 
Tan, RT, Kuzo, R, Goodman, LR, et al Utility of CT scan evaluation for predicting pulmonary hypertension in patients with parenchymal lung disease: medical College of Wisconsin Lung Transplant Group.Chest1998;113,1250-1256. [PubMed]
 
Chen, H, Gould, MK, De Marco, T, et al Utility of CT for predicting pulmonary hypertension in patients with parenchymal lung disease: similar results, different conclusion?Chest2008;133,1053-1054. [PubMed] [CrossRef]
 
Zisman, DA, Karlamangla, A, Ross, DJ, et al High-resolution chest computed tomography findings do not predict the presence of pulmonary hypertension in advanced idiopathic pulmonary fibrosis.Chest2007;132,773-779. [PubMed]
 
Zisman, DA, Ross, DJ, Belperio, JA, et al Prediction of pulmonary hypertension in idiopathic pulmonary fibrosis.Respir Med2007;101,2153-2159. [PubMed]
 
Zisman, DA, Karlamangla, A, Kawut, SM, et al Validation of a method to screen for pulmonary hypertension in advanced idiopathic pulmonary fibrosis.Chest2008;133,640-645. [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

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

Related Content

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

  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543