http://www.jhltonline.org

Usefulness of pulmonary capillary wedge pressure as a correlate of left ventricular filling pressures in pulmonary arterial hypertension Rudolf K.F. de Oliveira, MD,a Eloara V.M. Ferreira, MD, PhD,a Roberta P. Ramos, MD, PhD,a Carolina M.S. Messina, MD,a Carlos E.B. Kapins, MD,b Célia M.C. Silva, MD, PhD,b and Jaquelina S. Ota-Arakaki, MD, PhDa From the aPulmonary Vascular Group, Division of Respiratory Medicine, Federal University of São Paulo, São Paulo, Brazil; and b Hemodynamic Unit, Division of Cardiology, Federal University of São Paulo, São Paulo, Brazil.

KEYWORDS: pulmonary arterial hypertension; right heart catheterization; pulmonary capillary wedge pressure; left ventricular end-diastolic pressure; left ventricular filling pressure

BACKGROUND: Pulmonary arterial hypertension (PAH) is characterized by a pulmonary capillary wedge pressure (PCWP) of r15 mm Hg, given a normal left ventricular filling pressure (LVFP). However, recent studies have shown that, in PAH patients, diagnosis based on PCWP can erroneously classify a significant number of patients compared with diagnosis based on left ventricular end-diastolic pressure (LVEDP). Therefore, we sought to compare the diagnostic accuracy of end-expiratory PCWP and LVEDP measurements in patients suspected of having pulmonary hypertension (PH). METHODS: We reviewed the hemodynamic data from 122 patients suspected of having PH who underwent simultaneous right- and left-side heart catheterizations at a PH referral center from 2006 to 2011. RESULTS: PH was diagnosed in 105 patients, 79% of whom (n ¼ 83) showed a pre-capillary pattern according to the LVEDP measurement. Ninety percent of patients with PCWP r15 mm Hg were correctly classified as having pre-capillary PH. However, 39% of patients with a PCWP 415 mm Hg had LVEDP r15 mm Hg and would have been erroneously diagnosed with pulmonary venous hypertension based on their PCWP measurements alone. The sensitivity and specificity was 0.89 and 0.64, respectively. A Bland-Altman analysis of the PCWP and LVEDP measurements revealed a mean bias of 0.3 mm Hg with 95% limits of agreement of 7.2 to 7.8 mm Hg. CONCLUSIONS: A PCWP r15 mm Hg was found to be a reliable indicator of normal LVFP in precapillary PH patients. When measured properly and analyzed in the clinical context, PCWP is a valuable tool for accurate diagnosis of PAH. J Heart Lung Transplant 2014;33:157–162 r 2014 International Society for Heart and Lung Transplantation. All rights reserved.

Pulmonary arterial hypertension (PAH) is defined as a mean pulmonary arterial pressure (mPAP) of Z25 mm Hg at rest and a pulmonary capillary wedge pressure (PCWP) of r15 mm Hg,1 assuming a normal left ventricular filling Reprint requests: Jaquelina S. Ota-Arakaki, MD, PhD, Pulmonary Vascular Group, Division of Respiratory Medicine, Federal University of São Paulo, Rua Botucatu 740, 31 andar, São Paulo, SP, CEP: 04023-062, Brazil. Telephone: þ55-11-5576-4238. Fax: þ55-11-5082-5105. E-mail address: [email protected]

pressure (LVFP). These hemodynamic values are common to all etiologies of Group 1 pulmonary hypertension (PH) and are indicative of a pre-capillary hemodynamic pattern.2 Identifying this pattern is essential in correctly evaluating patients suspected of having PAH, as PAH-specific therapies are ineffective for pulmonary venous hypertension caused by left-side heart disease (Group 2 PH), which is characterized by a PCWP of 415 mm Hg,3 and is indicative of a post-capillary hemodynamic pattern. Importantly, such

1053-2498/$ - see front matter r 2014 International Society for Heart and Lung Transplantation. All rights reserved. http://dx.doi.org/10.1016/j.healun.2013.10.008

158

The Journal of Heart and Lung Transplantation, Vol 33, No 2, February 2014

misclassifications can deprive PAH patients of potentially beneficial treatment. PCWP represents an alternative measure to left ventricular end-diastolic pressure (LVEDP),4–6 which is the “gold standard” for determining LVFP. However, recent studies have shown that, in PAH patients, PCWP can lead to erroneous PH classification compared with a diagnosis based on LVEDP.7,8 Many factors can interfere with the accurate measurement of PCWP,9 including factors related to the measurement techniques themselves10,11 and patient-specific factors, such as obesity, hypoxemia12 and ventilation.13,14 Therefore, it is important to use other tools to confirm the accuracy and reliability of PCWP measurements during right-heart catheterization (RHC). In this context, clinical characteristics are often helpful for distinguishing between PAH patients and patients with PH caused by left-side heart disease.15,16 Determining the likelihood of each disease may help interpret the RHC results and improve the reliability of PCWP measurements. Although PCWP is a widely used diagnostic measurement, LVEDP is not analyzed in the majority of hemodynamic laboratories. Therefore, we sought to compare the diagnostic accuracy of end-expiratory PCWP and LVEDP in patients suspected of having PH.

Methods Patients We reviewed the hemodynamic data from all patients who underwent simultaneous right- and left-side heart catheterizations as part of their evaluations for suspected PH at the Federal University of São Paulo PH referral center between January 2006 and August 2011. According to the institutional protocols in place during this period, all patients suspected of having PH were referred for simultaneous right- and left-side heart catheterizations. All patients followed a standardized diagnostic algorithm in accordance with current PH guidelines,1,2 prior to RHC indication. Echocardiography, pulmonary function tests, blood-gas analyses, serum brain natriuretic peptide (BNP) levels, human immunodeficiency virus (HIV) serology and other specific tests guided by clinical suspicion were carried out systematically. Based on RHC, patients were stratified according to their hemodynamic pattern, on which the pre-capillary pattern included patients from Groups 1, 3, 4 and 5. None of the patients had mitral stenosis, or prosthetic mitral valves, or was under mechanical ventilation during the examination. One patient had a prosthetic aortic valve. Patients with unrepaired congenital heart disease were excluded from the analysis. This study was approved by the institutional medical ethics committee.

Hemodynamic measurements Catheterizations were performed by 6 experienced physicians according to the operational protocols of the hospital’s cardiac catheter laboratory. All physicians were under the supervision of the same hemodynamicist (C.M.C.S.). Right- and left-side heart measurements were taken via the right femoral vein and artery, respectively. A 6-French catheter with a terminal balloon was inserted into the pulmonary artery, and a 5-French pigtail catheter was inserted into the aorta. Hemodynamic recordings were taken while placing the catheter in the pulmonary artery, and its

progression to the wedge position was guided by fluoroscopic visualization and confirmed by identifying the characteristic waveform. PCWP was measured by determining the average of the a wave, which was identified as the atrial systole based on a simultaneous electrocardiogram. Next, the pigtail catheter was advanced into the left ventricle (LV) to measure LVEDP, which was defined as the LV pressure immediately preceding the onset of the ventricular contraction and was measured at the C point, representing the rise in the ventricular pressure resulting from atrial contraction.17 Zero pressure calibration was performed at the mid-thoracic level with the patient placed in a supine position. Pressure measurements were taken at end-expiration and were the average of at least 3 ventilatory cycles. No patient had a heart rate of 4130 beats/min and all were in sinus rhythm. Cardiac output was determined using Fick’s method with an estimated oxygen consumption (125  body surface area).18 No patient had cardiac output determined by the thermodilution method. Pulmonary vascular resistance (PVR) was expressed in Wood units (WU).

Statistical analysis Values are expressed as the mean ⫾ standard deviation (SD), unless otherwise stated. The diagnostic performance of PCWP vs LVEDP was evaluated using a 2  2 contingency table. Between-group comparisons were performed using one-way analysis of variance (ANOVA). Analyses of sensitivity, specificity, predictive value and likelihood ratio were performed on the PCWP measurements, considering an LVEDP value of r15 mm Hg as the diagnostic standard. The accuracy of using a PCWP measurement of r15 mm Hg as the threshold for correctly diagnosing patients with precapillary PH was determined by calculating the proportion of patients who would have been reclassified had they been diagnosed based on LVEDP (r15 mm Hg). We used Pearson's correlation coefficient to determine the correlation and Bland-Altman analysis to determine the agreement between the two measurements.19 p o 0.05 was considered significant. Statistical analyses was performed using SPSS software, version 19.0.0 (IBM, Armonk, NY).

Results Baseline characteristics A total of 206 patients underwent simultaneous right- and left-side heart catheterizations at our institution between 2006 and 2011. All patients had LVEDP measurements; 69 patients had no PCWP measurements to consult. Fifteen patients had unrepaired congenital heart disease and were excluded from the analysis. There were 122 patients with both LVEDP and PCWP measurements. Among these patients, 105 (86%) were diagnosed with PH, 83 (79%) of whom were diagnosed with a pre-capillary pattern (Figure 1). The different etiologies for PH included idiopathic PAH (IPAH) (n ¼ 19), collagen vascular disease (n ¼ 20), schistosomiasis (n ¼ 5), HIV infection (n ¼ 4), portal hypertension (n ¼ 2) and congenital heart disease (n ¼ 18). Eleven patients had PH because of parenchymal lung diseases, and 16 had chronic thromboembolic disease. Two patients had multifactorial etiologies (chronic renal failure on hemodialysis). Eight patients had left-heart disease as their sole etiology for PH.

de Oliveira et al.

Wedge Pressure in PAH

PH with a post-capillary hemodynamic pattern was found in 21% of all patients with PH. Of note, some of these patients with a high PVR and a high transpulmonary pressure gradient (TPG), were treated with specific medication (7 of the 22 patients with post-capillary hemodynamic pattern) based on their overall clinical context. Those 7 patients had an mPAP of 63 ⫾ 7 mm Hg, PCWP of 20 ⫾ 8 mm Hg, LVEDP of 22 ⫾ 5 mm Hg, PVR of 14.1 ⫾ 4.1 WU and CI of 2.2 ⫾ 1.0 liters/min/m2, as compared with those who were not treated whose corresponding data were 42 ⫾ 10 mm Hg, 17 ⫾ 4 mm Hg, 19 ⫾ 3 mm Hg, 7.4 ⫾ 4.5 WU and 2.5 ⫾ 1.2 liters/min/m2, respectively. The overall mean patient age was 48 ⫾ 16 years, 71% of whom were women. The average mPAP was 49 ⫾ 21 mm Hg, and average PVR was 11.3 ⫾ 9.5 WU. Baseline characteristics of the patients are presented in Table 1.

Hemodynamic parameters The correlation between PCWP and LVEDP measurements in patients with PH is shown in Figure 2 (r ¼ 0.74). A Bland-Altman analysis of PCWP and LVEDP revealed a mean bias of 0.3 mm Hg with 95% limits of agreement of 7.2 to 7.8 mm Hg (Figure 3). The majority of patients had a ⫾3.7 mm Hg difference between their PCWP and LVEDP measurements. Among the patients who did not have PH, the mean PCWP and LVEDP measurements were 10.0 ⫾ 3.2 and 10.0 ⫾ 2.8 mm Hg, respectively. Similar correlations between PCWP and LVEDP were found in this subgroup of patients.

Hemodynamic classification Among the 105 patients with PH, 79% (n ¼ 83) showed a pre-capillary hemodynamic pattern according to the LVEDP measurement. Ninety percent of the patients with a PCWP r15 mm Hg were correctly classified as having

159 Table 1

Baseline Characteristics of Patients (n ¼ 122)

Female Age (years) Body surface area (m2) Resting arterial oxygen saturation (%) Mean systemic arterial pressure (mmHg) No pulmonary hypertension Pulmonary hypertension Pre-capillary pattern Post-capillary pattern Right atrial pressure (mm Hg) Mean pulmonary arterial pressure (mm Hg) Pulmonary vascular resistance (WU) Cardiac index (liters/min/m2) PCWP (mm Hg) LVEDP (mm Hg) Mixed venous oxygen saturation (%)

87 (71%) 48 ⫾ 16 1.68 ⫾ 0.24 92.7 ⫾ 4.2 104 ⫾ 17 17 (14%) 105 (86%) 83 (79%) 22 (21%) 10 ⫾ 5 49 ⫾ 21 11.3 ⫾ 9.5 2.6 ⫾ 1.2 12.7 ⫾ 5.1 12.4 ⫾ 4.7 63.4 ⫾ 11.4

LVEDP, left ventricular end-diastolic pressure; PCWP, pulmonary capillary wedge pressure; WU, Wood units.

pre-capillary PH. However, 39% of patients with PCWP 415 mm Hg had LVEDP r15 mm Hg, and, therefore, would have been erroneously classified as having post-capillary PH based on their PCWP measurements alone (Table 2). The sensitivity, specificity, positive and negative predictive value, accuracy and positive and negative likelihood ratio were 0.89, 0.64, 0.90, 0.61, 0.84, 2.47 and 0.17, respectively. We did not identify any physical characteristics that significantly increased the chance of misclassification. However, right atrial pressure (RAP) and diastolic pulmonary vascular pressure gradient were associated with the chance of error (Table 3). The misclassified patients and their baseline characteristics are shown in Table 3.

Discussion The major finding in this study is that, similar to LVEDP, PCWP r15 mm Hg proved to be a reliable indicator of normal LVFP in patients from a PH referral center who were

Figure 1 Study flow diagram. PH, pulmonary hypertension; RHC, right heart catheterization; PCWP, pulmonary capillary wedge pressure; LVEDP, left ventricular end-diastolic pressure.

Figure 2

Correlation between PCWP and LVEDP.

160

The Journal of Heart and Lung Transplantation, Vol 33, No 2, February 2014

Figure 3 Bland-Altman analysis of PCWP and LVEDP. Mean bias of 0.3 mm Hg with 95% limits of agreement of 7.2 to 7.8 mm Hg.

suspected of having pre-capillary PH. This finding is of much interest as LVEDP is not assessed in most hemodynamics laboratories because left-heart catheterization can increase risk for patients , mostly related to vascular assessment and arrhythmias.20 In addition, left-side heart catheterization requires more extended anti-coagulation suspension, prolonged monitoring and increased examination costs. Approximately 61% (n ¼ 64) of patients had similar PCWP and LVEDP measurements, with differences within ⫾2 mm Hg, and 29% (n ¼ 30) of these patients had identical values. Remarkably, based on the Bland-Altman analysis, PCWP showed a heteroscedastic behavior starting at 11 mm Hg. Twenty-one patients with PCWP Z11 mm Hg had a difference Z4 mm Hg in their LVEDP values, and the majority of these cases corresponded to overestimations of LVFP based on PCWP. This aspect of PCWP has been reported previously,6 suggesting a need to measure LVEDP in this subgroup of patients. An alternative would be to perform additional interventions during the RHC, such as a rapid volume challenge21 or exercise challenge,22 based on preliminary clinical tests. In addition, in cases with PCWP 415 mm Hg, patients without a clinical predisposing condition to left-heart disease (such as systemic sclerosis, obstructive sleep apnea, etc.) or echocardiographic suspicion of left-heart dysfunction should undergo LVEDP measurement, especially in the presence of an elevated RAP (Table 3). Our results do not coincide with previous reports that claimed PCWP underestimates LVFP.7,8,23 Instead, the results suggest that PCWP, when properly measured, can continue to be used for PH classification. One of the major differences between this study and previous studies is the diagnostic frequency of the pre-capillary pattern. In the Halpern and Taichman study, 89.2% of all patients were diagnosed as Group 2 PH according to their post-capillary hemodynamic pattern.8 Their findings may have been influenced by the local cardiac catheter laboratory

characteristics, which may have played a major role in the final conclusions. After recalculating the Halpern and Taichman data using an LVEDP of r15 mm Hg as the diagnostic standard, we arrived at sensitivity, specificity and positive and negative predictive values of 0.64, 0.91, 0.47 and 0.95, respectively.24 Furthermore, the recalculated accuracy was 0.88, similar to our own findings (0.84) and those of others,12,25 which reinforces the idea that the reliability of this measure depends on the pre-test probability of the population in the study. Note that positive and negative predictive values are dependent on disease prevalence,24 and the efficiency of a diagnostic test should not be analyzed in a population in which the condition being tested does not frequently occur. Doing so in this case would almost certainly decrease the apparent reliability of PCWP measurements and inappropriately disqualify a valuable tool for PAH diagnosis and treatment. Although designed with another purpose in mind, one recent study reached conclusions similar to ours. Ryan and colleagues compared digitally measured PCWP and endexpiratory PCWP in patients from a PH referral center.12 In the end-expiratory PCWP subgroup, their results were similar to those in our study: the sensitivity, specificity, positive and negative predictive values and accuracy were 0.86, 1.0, 1.0, 0.74 and 0.90, respectively. The baseline characteristics of the patients studied may justify these study findings. The value of the clinical context and pre-test probability for a specific diagnosis has already been demonstrated26 and may be useful for interpreting PCWP results.27 Taken together, symptoms, medical history, electrocardiograms, echocardiograms and chest computed tomography (CT) scans can suggest the likelihood of PAH, and furthermore they may indicate that elevated LVFP is the cause of patients’ symptoms.15,16,28,29 Consistent with this hypothesis, in this study we have demonstrated that clinical context is essential for accurate PH hemodynamic diagnosis. All patients referred for RHC were suspected of having precapillary PH, or out-of-proportion post-capillary PH. We did not used to refer patients suspected of having proportional pulmonary venous hypertension to RHC. This approach may have contributed to the reliability of PCWP as a method for identifying normal LVFP patients. Among the patients with a post-capillary hemodynamic pattern (n ¼ 22), 7 were treated with PAH-specific medication based on their overall clinical context. They had a high PVR and a high TPG. These patients are missed by the guidelines, which would suggest that they do not have PAH

Table 2 Classification of PH: LVEDP vs PCWP, considering LVEDP r15 mm Hg as Diagnosis Standard

PCWP r15 mm Hg PCWP 415 mm Hg Total

LVEDP r15 mm Hg

LVEDP 415 mm Hg

Total

74 9 83

8 14 22

82 23 105

LVEDP, left ventricular end-diastolic pressure; PCWP, pulmonary capillary wedge pressure.

de Oliveira et al.

Wedge Pressure in PAH

161

Table 3 Pulmonary Hypertension Hemodynamic Patterns and Misclassifications by PCWP Measurement, Considering LVEDP r15 mm Hg as Diagnosis Standard Pre-capillary PH

Female Age (years) BSA (m2) SaO2 (%) mSAP (mm Hg) PH etiology IPAH Collagen vascular disease Schistosomiasis HIV Portal hypertension Congenital heart disease Left heart disease Parenchymal lung disease CTEPH Multifactorial etiologies Echocardiogram Left atrium LVEF o55% MR (4 mild) RHC RAP (mmHg) mPAP (mmHg) PVR (WU) CI (liters/min/m2) PCWP (mmHg) LVEDP (mmHg) TPG (mmHg) TPG 415 mmHg DPG (mmHg)

Post-capillary PH

Correctly classified PCWP r15 mm Hg LVEDP r15 mm Hg (n ¼ 74)

Erroneously classified PCWP 415 mm Hg LVEDP r 15 mm Hg (n ¼ 9)

Correctly classified PCWP 415 mm Hg LVEDP 415 mm Hg (n ¼ 14)

Erroneously classified PCWP r15 mm Hg LVEDP 415 mm Hg (n ¼ 8)

56 (76%) 45 ⫾ 16 1.68 ⫾ 0.25 92.4 ⫾ 4.7 103 ⫾ 20

6 (67%) 45 ⫾ 12 1.75 ⫾ 0.19 93.6 ⫾ 2.1 103 ⫾ 17

9 (64%) 56 ⫾ 12 1.76 ⫾ 0.21 92.2 ⫾ 4.1 107 ⫾ 14

6 (75%) 57 ⫾ 18 1.65 ⫾ 0.12 92.6 ⫾ 3.3 118 ⫾ 15

15 13 4 3 2 16 0 7 12 2

1 2 1 0 0 1 0 2 2 0

2 4 0 0 0 0 5 1 2 0

1 1 0 1 0 1 3 1 0 0

37 ⫾ 6 4 (5%) 1 (1%)

37 ⫾ 2 0 0

40 ⫾ 7 1 (7%) 2 (14%)

42 ⫾ 10 6 (75%) 2 (25%)

10 ⫾4 55 ⫾ 19 13.8 ⫾ 10.3 2.6 ⫾ 1.2 11 ⫾ 3 10 ⫾ 3 44 ⫾ 20 73 (99%) 24 ⫾ 15

13 ⫾ 5* 54 ⫾ 16 11.4 ⫾ 6.4 2.4 ⫾ 0.6 19 ⫾ 3* 13 ⫾ 2* 35 ⫾ 15 8 (89%) 15 ⫾ 12

17 ⫾ 6 51 ⫾ 15 9.1 ⫾ 5.4 2.5 ⫾ 1.4 22 ⫾ 5 21 ⫾ 5 29 ⫾ 13 11 (79%) 10 ⫾ 8

10 ⫾ 5† 48 ⫾ 13 11.5 ⫾ 6.2 2.2 ⫾ 0.5 13 ⫾ 3† 18 ⫾ 1† 35 ⫾ 13 8 (100%) 20 ⫾ 9†

BSA: body surface area; CI: cardiac index; CTEPH: chronic thromboembolic pulmonary hypertension; DPG: diastolic pulmonary vascular pressure gradient; HIV: human immunodeficiency virus; IPAH: idiopathic pulmonary arterial hypertension; LVEDP: left ventricular end-diastolic pressure; LVEF: left ventricular ejection fraction; mPAP: mean pulmonary arterial pressure; MR: mitral regurgitation; mSAP: mean systemic arterial pressure; PCWP: pulmonary capillary wedge pressure; PH: pulmonary hypertension; PVR: pulmonary vascular resistance; RAP: right atrial pressure; RHC: right heart catheterization; SaO2: resting arterial oxygen; TPG: transpulmonary gradient; WU: Wood units. n p o 0.05 vs correctly classified pre-capillary pulmonary hypertension patients. † p o 0.05 vs correctly classified post-capillary pulmonary hypertension patients.

and should not be treated. The ventricular interdependence and the presence of comorbidities could justify an elevated LVFP in these patients.30 Furthermore, these patients have similar morbidity and mortality when compared with PAH patients with normal LVFP.31 Our study is limited by its retrospective design and size of the population studied, and represents a biased patient population selected from a unique center; therefore, these findings may not apply to patients undergoing hemodynamic evaluation in general. Another limitation is that we included patients from Groups 3, 4, and 5. However, as hemodynamic pattern was the end goal of the PH classification, this limitation was unlikely to have interfered with the final results, which reflected a real-life cohort with different PH etiologies.

Furthermore, LVEDP and PCWP were not measured simultaneously, and the positions of the PCWP measurements were not confirmed by oxygen saturation. It is possible that these factors affected the accuracy of the measurements, although we do not believe they significantly influenced our results. These practices are not routinely performed at most centers, and our protocols reflect common hemodynamic practice. In our experience, PCWP measurements guided by fluoroscopy and characteristic waveform patterns can be used to consistently and accurately position catheters for measuring PCWP. In addition, we were unable to determine why certain patients did not have PCWP data, although the baseline characteristics of these individuals did not differ from the entire study population. We were also unable to review all

162

The Journal of Heart and Lung Transplantation, Vol 33, No 2, February 2014

hemodynamic tracings. Nevertheless, all examinations were conducted by experienced physicians and in accordance with the operational protocols of the institutional cardiac catheter laboratory, which we consider quite stringent. In conclusion, we have demonstrated that a PCWP measurement of r15 mm Hg is a reliable indicator of normal LVFP in pre-capillary PH patients. When properly performed and analyzed in the clinical context, PCWP is a valuable tool for accurate diagnosis of PAH.

Disclosure statement

13.

14.

15.

16.

The authors have no conflicts of interest to disclose. 17.

References 1. Badesch DB, Champion HC, Sanchez MAG, et al. Diagnosis and assessment of pulmonary arterial hypertension. J Am Coll Cardiol 2009;54(suppl):S55-66. 2. Galiè N, Hoeper MM, Humbert M, et al. Guidelines for the diagnosis and treatment of pulmonary hypertension: the task force for the diagnosis and treatment of pulmonary hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Eur Heart J 2009;30:2493-537. 3. Fang JC, DeMarco T, Givertz MM, et al. World Health Organization Pulmonary Hypertension group 2: pulmonary hypertension due to left heart disease in the adult–a summary statement from the Pulmonary Hypertension Council of the International Society for Heart and Lung Transplantation. J Heart Lung Transplant 2012;31:913-33. 4. Hellens H, Haynes F, Dexter L. Pulmonary capillary pressure in man. J Appl Physiol 1949;2:24-9. 5. Connolly D, Kirklin J, Wood E. The relationship between pulmonary artery wedge pressure and left atrial pressure in man. Circ Res 1954; 2:434-40. 6. Walston A, Kendall ME. Comparison of pulmonary wedge and left atrial pressure in man. Am Heart J 1973;86:159-64. 7. Soto F, Siegel R, Marks D, et al. Performance of pulmonary capillary wedge pressure (PCWP) vs left ventricular end diastolic pressure (LVEDP) in the diagnosis/classification of patients with suspect pulmonary arterial hypertension. Chest 2005;128(suppl): 137S. 8. Halpern SD, Taichman DB. Misclassification of pulmonary hypertension due to reliance on pulmonary capillary wedge pressure rather than left ventricular end-diastolic pressure. Chest 2009;136:37-43. 9. Bernstein W, Fierer E, Laszlo M, et al. The interpretation of pulmonary artery wedge (pulmonary capillary) pressures. Br Heart J 1960;22: 37-44. 10. Komadina K, Schenk D, LaVeau P, et al. Interobserver variability in the interpretation of pulmonary artery catheter pressure tracings. Chest 1991;100:1647-54. 11. Tonelli AR, Mubarak KK, Li N, et al. Effect of balloon inflation volume on pulmonary artery occlusion pressure in patients with and without pulmonary hypertension. Chest 2011;139:115-21. 12. Ryan J, Rich J, Thiruvoipati T, et al. Current practice for determining pulmonary capillary wedge pressure predisposes to serious errors in the

18. 19. 20.

21.

22.

23.

24.

25.

26. 27. 28.

29.

30. 31.

classification of patients with pulmonary hypertension. Am Heart J 2012;163:589-94. Cengiz M, Crapo R, Gardner R. The effect of ventilation on the accuracy of pulmonary artery and wedge pressure measurements. Crit Care Med 1983;11:502-7. Al-kharrat T, Zarich S, Amoateng-adjepong Y, et al. Analysis of observer variability in measurement of pulmonary artery occlusion pressures. Am J Respir Crit Care Med 1999;160:415-20. Drazner MH, Hellkamp AS, Leier CV, et al. Value of clinician assessment of hemodynamics in advanced heart failure: the ESCAPE trial. Circ Heart Fail 2008;1:170-7. Leung CC, Moondra V, Catherwood E, et al. Prevalence and risk factors of pulmonary hypertension in patients with elevated pulmonary venous pressure and preserved ejection fraction. Am J Cardiol 2010; 106:284-6. Joung B, Ha J-W, Ko YG, et al. Can pro-brain natriuretic peptide be used as a noninvasive predictor of elevated left ventricular diastolic pressures in patients with normal systolic function? Am Heart J 2005;150:1213-9. Wilkinson J. Haemodynamic calculations in the catheter laboratory. Heart 2001;85:113-20. Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10. West R, Ellis G, Brooks N, et al. Complications of diagnostic cardiac catheterisation: results from a confidential inquiry into cardiac catheter complications. Heart 2006;92:810-4. Fujimoto N, Borlaug BA, Lewis GD, et al. Hemodynamic responses to rapid saline loading: the impact of age, sex, and heart failure. Circulation 2013;127:55-62. Borlaug BA, Nishimura RA, Sorajja P, et al. Exercise hemodynamics enhance diagnosis of early heart failure with preserved ejection fraction. Circ Heart Fail 2010;3:588-95. Forrester J, Diamond G, McHugh T, et al. Filling pressures in the right and left sides of the heart in acute myocardial infarction. A reappraisal of central-venous-pressure monitoring. N Engl J Med 1971;285: 190-3. Raschke R, Gerkin R, Curry S, et al. Operating characteristics of pulmonary capillary wedge pressure for pulmonary arterial hypertension. Chest 2010;137:740. Sharma GVRK Woods P, Lambrew CT, et al. Evaluation of a noninvasive system for determining left ventricular filling pressure. Arch Intern Med 2002;162:2084-8. Pauker S, Kassirer J. The threshold approach to clinical decision making. N Engl J Med 1980;302:1109-17. Soto F. Pulmonary venous hypertension: a pulmonologist’s perspective. Adv Pulm Hypertens 2011;10:18-23. Dadfarmay S, Berkowitz R, Kim B, et al. Differentiating pulmonary arterial and pulmonary venous hypertension and the implications for therapy. Congest Heart Fail 2010;16:287-91. Thenappan T, Shah SJ, Gomberg-Maitland M, et al. Clinical characteristics of pulmonary hypertension in patients with heart failure and preserved ejection fraction. Circ Heart Fail 2011;4:257-65. Shapiro BP, McGoon MD, Redfield MM. Unexplained pulmonary hypertension in elderly patients. Chest 2007;131:94-100. Frost AE, Farber HW, Barst RJ, et al. Demographics and outcomes of patients diagnosed with pulmonary hypertension with pulmonary capillary wedge pressures 16 to 18 mm Hg: insights from the REVEAL Registry. Chest 2013;143:185-95.