Progressive Changes in Right Ventricular Geometric Shortening and Long-term Survival in Pulmonary Arterial Hypertension

CHEST

Original Research PULMONARY VASCULAR DISEASE

Progressive Changes in Right Ventricular Geometric Shortening and Long-term Survival in Pulmonary Arterial Hypertension Gert-Jan Mauritz, MSc; Taco Kind, MD; J. Tim Marcus, PhD; Harm-Jan Bogaard, MD, PhD; Mariëlle van de Veerdonk, MD; Pieter E. Postmus, MD, PhD, FCCP; Anco Boonstra, MD, PhD; Nico Westerhof, PhD; and Anton Vonk-Noordegraaf, MD, PhD

Background: Until now, many investigators have focused on describing right ventricular (RV) dysfunction in groups of patients with pulmonary arterial hypertension (PAH), but very few have addressed the deterioration of RV function over time. The aim of this study was to investigate time courses of RV geometric changes during the progression of RV failure. Methods: Forty-two patients with PAH were selected who underwent right-sided heart catheterization and cardiac MRI at baseline and after 1-year follow-up. Based on the survival after this 1-year run-in period, patients were classified into two groups: survivors (26 patients; subsequent survival of . 4 years) and nonsurvivors (16 patients; subsequent survival of , 4 years). Four-chamber cine imaging was used to quantify RV longitudinal shortening (apex-base distance change), RV transverse shortening (septum-free wall distance change), and RV fractional area change (RVFAC) between end diastole and end systole. Results: Longitudinal shortening, transverse shortening, and RVFAC measured at the beginning of the run-in period and 1 year later were significantly higher in subsequent survivors than in nonsurvivors (P , .05). Longitudinal shortening did not change during the run-in period in either patient group. Transverse shortening and RVFAC did not change during the run-in period in subsequent survivors but did decrease in subsequent nonsurvivors (P , .05). This decrease was caused by increased leftward septal bowing. Conclusions: Progressive RV failure in PAH is associated with a parallel decline in longitudinal and transverse shortening until a floor effect is reached for longitudinal shortening. A further reduction of RV function is due to progressive leftward septal displacement. Because transverse shortening incorporates both free wall and septum movements, this parameter can be used to monitor the decline in RV function in end-stage PAH. CHEST 2012; 141(4):935–943 Abbreviations: CMR 5 cardiac magnetic resonance imaging; EDV 5 end-diastolic volume; EF 5 ejection fraction; LV 5 left ventricular; PAH 5 pulmonary arterial hypertension; RHC 5 right-sided heart catheterization; RV 5 right ventricular; RVFAC 5 right ventricle fractional area change; TAPSE 5 tricuspid annular plane systolic excursion

arterial hypertension (PAH) is a proPulmonary gressive and devastating disease characterized by

pathologic lesions of the pulmonary arteries that lead to increased pulmonary artery pressure.1 Although the pulmonary pressure rise is the distinctive characteristic of this disease, it is right ventricular (RV) function that dominates prognosis and long-term survival.2,3 Several geometrical measures exist that quantify RV function. RV longitudinal shortening (tricuspid annular plane systolic excursion [TAPSE]) is a good reflection of RV systolic function.4-6 However, TAPSE only accounts for the longitudinal RV shortening and ignores

www.chestpubs.org

the septal contribution to RV ejection.7 Recently, our group showed that RV transverse shortening reflects global RV function at least as much as longitudinal shortening.8 Furthermore, RV fractional area change (RVFAC) has been shown to be a robust measure of RV systolic function and has a good correlation with invasive hemodynamic data in patients with PAH.9,10 RVFAC has the advantage of combining transverse and longitudinal RV shortening into one measure. However, it is largely unknown how these different RV parameters change during follow-up in patients with progressive RV failure. Therefore, the aim of the CHEST / 141 / 4 / APRIL, 2012

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

935

present study was to evaluate the pathophysiologic changes of RV geometric shortening in patients with progressive PAH who died between 1 and 5 years after initial diagnosis and to compare these results with patients with stable PAH surviving . 5 years. To reach this objective, RV longitudinal and transverse shortening as well as RVFAC were measured in a group of patients with PAH at baseline and after 1 year of follow-up.

Materials and Methods Patient Population The present study was performed in an observational cohort of patients with PAH and was part of a prospective ongoing research project aimed to evaluate the RV in PAH by means of MRI. Between May 2003 and May 2005, 287 patients were referred to our hospital for the evaluation of pulmonary hypertension. A diagnosis of PAH was established while following a standard protocol that included right-sided heart catheterization (RHC).11 We selected treatment-naive patients with PAH who underwent cardiac MRI (CMR) at the time of diagnosis and again after 1 year of follow-up. Patients with congenital systemic-topulmonary shunts were excluded from the study. In total, 42 patients were selected and grouped according to subsequent survival after the 1-year run-in period as follows: Survivors (n 5 26) were those patients with a transplant-free survival period of at least 5 years after the baseline CMR, and nonsurvivors (n 5 16) were those patients who died between 1 and 5 years after baseline CMR because of progression of the disease. Although the choice of 5 years is arbitrary, this period was chosen to make a clear distinction between long- and shortterm survivors. In addition, previous reports and clinical experience show that patients surviving 5 years can be classified as in a stable clinical condition because the annual mortality rate after 5 years is low for this group.12,13 Twenty-five patients met the diagnostic criteria for the study and survived for . 1 year after diagnosis but could not be selected Manuscript received January 3, 2011; revision accepted August 18, 2011. Affiliations: From the Department of Pulmonary Diseases (Mr Mauritz and Drs Kind, Bogaard, van de Veerdonk, Postmus, Boonstra, Westerhof, and Vonk-Noordegraaf), Department of Physics and Medical Technology (Dr Marcus), and Department of Physiology (Dr Westerhof), Institute for Cardiovascular Research, VU University Medical Center, Amsterdam, The Netherlands; and the Department of Medicine (Dr Bogaard), Division of Pulmonary and Critical Care, Virginia Commonwealth University, Richmond, VA. Mr Mauritz and Dr Kind contributed equally to this article. Funding/Support: This study was financially supported by The Netherlands Organisation for Scientific Research (NWO) Toptalent grant [021.001.120 to Dr Kind] and the NWO Vidi Grant [91.796.306 to Dr Vonk-Noordegraaf]. Correspondence to: Anton Vonk-Noordegraaf, MD, PhD, Department of Pulmonary Diseases, VU University Medical Center, Boelelaan 1117, 1007 MB, Amsterdam, The Netherlands; e-mail: [email protected] © 2012 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (http://www.chestpubs.org/ site/misc/reprints.xhtml). DOI: 10.1378/chest.10-3277

because they did not undergo follow-up CMR measurements (eg, dropout, withdrawal, protocol deviation). These patients were analyzed for potential selection bias. The Institutional Research Board approved the study protocol (registration number NL30766.029.10). All patients consented to participate in the study. Cardiac MRI CMR examination was performed with a 1.5-T Siemens Sonata MRI system (Siemens Medical Solutions) equipped with a six-element phased-array coil. ECG-gated cine imaging was performed using a balanced steady-state free precession pulse sequence during breath-holds. Long-axis slices were acquired in the four-, three-, and two-chamber views. Additionally, shortaxis slices were obtained with a typical slice thickness of 5 mm and a slice gap of 5 mm, fully covering both ventricles from base to apex. MRI parameters used were temporal resolution, 35 to 45 milliseconds; voxel size, 1.5 3 1.8 3 5.0 mm3; flip angle, 60°; receiver bandwidth, 930 Hz/pixel; repetition time/inversion time, 3.2/1.6 milliseconds; and matrix, 256 3 156. Image Analysis Global Left Ventricular and RV Function Analysis: From the stack of short-axis cine images, endocardial surfaces were carefully traced manually using Mass Analysis software (MEDIS Medical Imaging Systems Bv) to obtain RV and left ventricular (LV) end-diastolic volume (EDV), end-systolic volume (ESV), stroke volumes, and ejection fractions (EFs). All volumetric CMR measures were corrected for body surface area. RV Geometric Shortening Analysis: The four-chamber cine images were analyzed according to a method we described previously.8 End diastole was defined as the onset of the R wave of the ECG. End systole was determined as the moment of end shortening of the RV free wall (using MR cine imaging). RV longitudinal shortening was calculated as the distance change between end diastole and end systole of the tricuspid annulus-to-apex distance. RV transverse shortening was calculated as the change between end diastole and end systole of the RV free wall-to-septum distance (Fig 1). This transverse shortening was measured at seven different levels covering the whole RV cavity from base (level 1) through apex (level 7), with the mid-level (level 4) exactly halfway through the right ventricle. RV free wall displacement was defined as the displacement of the RV free wall with respect to the tricuspid annulus-to-apex line. Septal displacement was defined as the displacement of the interventricular septum to the tricuspid annulus-to-apex line (Fig 1). Positive displacement means toward the tricuspid annulus-to-apex line, and negative displacement means away from this line. RVFAC was calculated from the RV end-diastolic and end-systolic areas measured from the fourchamber view as follows: RVFAC 5 100 3 [(RV end-diastolic area 2 RV end-systolic area)/RV end-diastolic area]. The procedural duration of postacquisition geometric assessments depends on the number of contours, which have to be drawn. To calculate RVEF, endocardial contours are drawn at end diastole and end systole on all short-axis slices (generally eight to 12). RVFAC is obtained from a single four-chamber view at end diastole and end systole; this process is up to eight to 12 times faster than the determination of RVEF. For the longitudinal and transverse measures, the procedure is even simpler because it requires only single dimensions at end diastole and end systole. Statistical Analysis Normal distribution of the data were verified using a normal probability plot and log transformed if necessary. All data are

936

Original Research

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

to age and sex. The nonsurvivor group comprised more patients with connective tissue disease, a greater number of patients in New York Heart Association functional class IV, and a poorer average 6-min walk distance than survivors. Furthermore, cardiac output was significantly lower in nonsurvivors, but pulmonary artery pressure and pulmonary vascular resistance at baseline were similar in both groups. All patients were optimally treated, and there was no significant difference in the use of the pulmonary hypertension medication between survivors and nonsurvivors. Figure 2 shows the hemodynamics in the subset of patients (22 survivors and 12 nonsurvivors) with RHC at baseline and follow-up. Although statistically not significant, both the survivors and the nonsurvivors showed a reduction in pulmonary vascular resistance after 1-year follow-up. Heart rate increased significantly during follow-up in the nonsurvivors. Selection bias was tested on patients without follow-up measurements (n 5 25) (see “Materials and Methods” section). Baseline characteristics of these patients were not significantly different, except for age, compared with the total study population. Included patients were significantly younger than the patients without follow-up measurements (e-Table 1). Global RV and LV Function

Figure 1. Schematic depiction of RV transverse shortening and RV free wall and septal displacement. RV transverse shortening is defined as a change in the distance from the RV FW to Sep between RV end diastole and RV end systole. Displacement of the FW and Sep is defined as the displacement relative to the tricuspid annulus-to-apex line between RV end diastole and RV end systole. FW 5 free wall; RA 5 right atrium; RV 5 right ventricle; Sep 5 septum. presented as mean ⫾ SD, unless stated otherwise. Comparisons between and within groups were calculated using unpaired and paired Student t tests. A one-way analysis of variance was used when more than two groups were tested. Linear regression analysis was used to estimate the correlation between RVEF and geometric shortening measures. The Fisher exact test was used for categorical data. Statistical tests were performed with SPSS version 18 software (SPSS Inc). Statistical significance was considered when P , .05.

Results Patient Characteristics The baseline demographic and hemodynamic data of both the survivors and nonsurvivors are summarized in Table 1. In four survivors and four nonsurvivors, no RHC was performed at 1-year follow-up within 2 weeks of CMR assessment. The nonsurvivors had a mean survival of 2.5 ⫾ 1 years. There was no difference between the survivors and nonsurvivors with respect www.chestpubs.org

Figure 3 illustrates a long-axis view of a survivor and a nonsurvivor at baseline in end diastole. Compared with the survivor, the RV configuration in the nonsurvivor showed a markedly dilated atrium and ventricle. Volumetric and geometric CMR measurements at baseline and follow-up for the subsequent survivors and nonsurvivors are presented in Table 2. RVEF at the beginning and the end of the run-in period was significantly lower in nonsurvivors than in survivors. LVEDV index and RV and LV stroke volume index were significantly lower at the end of the run-in period in nonsurvivors than in survivors. Although baseline RVEDV index was not significantly different between the survivors and nonsurvivors, the RV dilated progressively during follow-up in only nonsurvivors. RVEF remained stable in the survivors, whereas it further decreased in the nonsurvivors. RV Geometric Shortening Longitudinal Shortening: There was a significant difference between subsequent survivors and nonsurvivors as shown in Figure 4A; the nonsurvivors had a reduced RV longitudinal shortening compared with the survivors (14 ⫾ 7 mm vs 20 ⫾ 5 mm, P , .05). At follow-up, the change in RV longitudinal shortening was not significantly different between baseline and follow-up for both survivors (P 5 .25) and nonsurvivors (P 5 .13). In the nonsurvivors, there was no CHEST / 141 / 4 / APRIL, 2012

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

937

Table 1—Baseline Characteristics of the Study Population Variable Age, y Female sex PAH diagnosis Idiopathic Heritable Drug and toxin induced Connective tissue disease Associated HIV infection Associated portal hypertension NYHA functional class II III IV 6MWT distance, m Hemodynamics Heart rate, beats/min mSAP, mm Hg mPAP, mm Hg CO, L/min PVR, dyne/s/cm5 RAP, mm Hg PCWP, mm Hg Sao2 Svo2 Medication use IV prostacyclin Endothelin receptor antagonist Sildenafil Calcium antagonist

Total Population (N 5 42)

Survivors (n 5 26)

Nonsurvivors (n 5 16)

P Value

44 ⫾ 14 31

43 ⫾ 11 18

46 ⫾ 18 13

.46 .61

28 (67) 3 (7) 1 (2) 8 (19) 1 (3) 1 (3)

20(77) 2 (8) 1 (4) 1 (4) 1 (4) 1 (4)

8 (50) 1 (6) 0 (0) 7 (44) 0 (0) 0 (0)

.19 1.00 1.00 .02a 1.00 1.00

13 (31) 21 (50) 8 (19) 434 ⫾ 136

10 (38) 15 (58) 1 (4) 465 ⫾ 142

3 (18) 6 (38) 7 (44) 380 ⫾ 106

.31 .22 .001a .04a

80 ⫾ 17 90 ⫾ 17 52 ⫾ 15 4.6 ⫾ 1.4 878 ⫾ 455 8⫾6 8⫾5 95 ⫾ 3 64 ⫾ 9

78 ⫾ 14 91 ⫾ 9 54 ⫾ 14 5.0 ⫾ 1.5 821 ⫾ 413 7⫾4 8⫾6 95 ⫾ 2 66 ⫾ 9

83 ⫾ 13 89 ⫾ 13 50 ⫾ 16 3.9 ⫾ 0.9 970 ⫾ 516 10 ⫾ 8 7⫾4 94 ⫾ 4 60 ⫾ 11

.30 .38 .46 .017a .31 .08 .34 .37 .08

12 (29) 24 (58) 15 (36) 5 (12)

6 (16) 17 (45) 9 (24) 4 (11)

6 (33) 7 (39) 6 (33) 1 (6)

.25 .15 .70 .37

Data are presented as mean ⫾ SD or No. (%). 6MWT 5 6-min walk test; CO 5 cardiac output; mPAP 5 mean pulmonary arterial pressure; mSAP 5 mean systemic arterial pressure; NYHA 5 New York Heart Association; PAH 5 pulmonary arterial hypertension; PCWP 5 pulmonary capillary wedge pressure; PVR 5 pulmonary vascular resistance; RAP 5 mean right atrial pressure; Sao2 5 arterial oxygen pressure; Svo2 5 mixed venous oxygen saturation. aSignificant at P , .05.

significant correlation between the change in longitudinal shortening and change in RVEF (r 5 0.4, P 5 .13). Transverse Shortening: The results of the transverse shortening, measured both at baseline and at follow-up, for the subsequent survivors and nonsurvivors are shown in Figure 4B. For the survivors, the RV transverse shortening did not change significantly at all levels between baseline and follow-up (eg, at mid-RV level, 9.7 ⫾ 3.6 mm vs 9.5 ⫾ 4.7 mm; P 5 .82). For the nonsurvivors, RV shortening at baseline was significantly reduced for levels 2 through 7 compared with the survivors (eg, at mid-RV level, 5.1 ⫾ 5.5 mm vs 20.5 ⫾ 5.2 mm; P , .001). During follow-up, the RV transverse shortening further decreased significantly from baseline at levels 2 to 6 and even showed lengthening at levels from mid to apex. There was a significant correlation between the change in RVEF and the change in RV transverse shortening obtained at mid-level (r 5 0.66, P 5 .005) in the nonsurvivors.

RV Wall and Septal Displacement: As mentioned previously, the decrease in transverse shortening at follow-up is a feature of nonsurvivors. This shortening is determined by the displacements of RV free wall and septum. Note that shortening is a change of the distance between the RV free wall and the septum, whereas displacement is an isolated property of the RV free wall and septum separately. Figure 5 illustrates the transverse displacements of RV free wall and septum in nonsurvivors separately. For the RV free wall displacement, there was no significant change at follow-up for any level. In contrast, for the septum, there was a significant increase of leftward displacement for levels 2 through 7, demonstrating that the decrease in transversal shortening in progressive PAH is mainly due to worsening of leftward septum bowing. RV Fractional Area Change: The change in RVFAC is illustrated in Figure 6A. Similar to RV transverse shortening, the RVFAC in the survivors remained stable during 1-year follow-up. The nonsurvivors already

938

Original Research

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

Figure 3. A, Cardiac MRI image of a pulmonary arterial hypertension survivor. B, Cardiac MRI image of a pulmonary arterial hypertension nonsurvivor. The survivor and nonsurvivor encountered equally raised mean pulmonary artery pressure (47 mm Hg vs 48 mm Hg, respectively) but differed in their response to afterload. Note in the nonsurvivor the enlarged RA and RV as signs of a failing RV. LV 5 left ventricle. See Figure 1 legend for expansion of other abbreviations.

Figure 2. A-D, Hemodynamic changes for pulmonary vascular resistance (A), cardiac output (B), heart rate (C), and stroke volume index (D) in 34 patients with pulmonary arterial hypertension (22 survivors, 12 nonsurvivors) with right-sided heart catheterization at baseline and 1-year follow-up. Survivors vs nonsurvivors at baseline (black bars) and follow-up (white bars). Data are presented as mean ⫾ SEM.

showed a decreased RVFAC at baseline compared with the survivors (24% ⫾ 10% vs 31% ⫾ 9%, respectively, P 5 .03). At follow-up, the nonsurvivors showed a further significant (P , .001) decrease to 17% ⫾ 10%. Additionally, there was a moderate correlation between the change in RVFAC and change in RVEF (r 5 0.67, P 5 .005) (Fig 6B). Discussion We investigated the pathophysiologic changes in RV geometry in patients with PAH in the first year after

diagnosis and related the geometric changes occurring during this initial year to subsequent survival. The major finding of this study is the following characterization of RV properties in subsequent nonsurvivors: (1) Longitudinal shortening and transverse shortening are already reduced at baseline; (2) both longitudinal shortening and RV free wall motion stay the same over time in nonsurvivors, whereas transverse shortening shows a further decline over time; and (3) the end-stage decline in RV function is due to a progressive leftward septal displacement rather than to a further decrease in RV free wall transverse or longitudinal displacement. RV Geometric Shortening At baseline, longitudinal and transverse shortening were significantly lower in subsequent nonsurvivors. The longitudinal shortening is comparable with TAPSE as used in echocardiography.14 Forfia et al15 showed that patients with an echocardiographically derived TAPSE of , 18 mm had a significantly reduced survival,

Table 2—Global RV and LV Function at Baseline and Follow-up Survivors Variable LV EDVI, mL/m2 ESVI, mL/m2 SVI, mL/m2 EF, % RV EDVI, mL/m2 ESVI, mL/m2 SVI, mL/m2 EF, %

Nonsurvivors

Comparison Survivors vs Nonsurvivors

Baseline

Follow-up

Pintra

Baseline

Follow-up

Pintra

Pinter Baseline

Pinter Follow-up

43 ⫾ 12 15 ⫾ 5 28 ⫾ 9 66 ⫾ 9

44 ⫾ 15 16 ⫾ 9 27 ⫾ 10 64 ⫾ 12

.84 .31 .39 .41

39 ⫾ 13 14 ⫾ 9 24 ⫾ 7 65 ⫾ 14

32 ⫾ 11 13 ⫾ 7 19 ⫾ 7 62 ⫾ 16

.03a .27 .01a .46

.14 .33 .15 .76

.02a .28 .008a .62

74 ⫾ 21 49 ⫾ 18 25 ⫾ 5 35 ⫾ 11

71 ⫾ 21 48 ⫾ 21 22 ⫾ 8 33 ⫾ 12

.34 .96 .07 .21

80 ⫾ 31 57 ⫾ 24 22 ⫾ 10 28 ⫾ 9

94 ⫾ 39 75 ⫾ 34 19 ⫾ 7 19 ⫾ 7

.04a .01a .21 .002a

.56 .15 .24 .03a

.02a .003a .03a .001

Data are presented as mean ⫾ SD. EDVI 5 end-diastolic volume index; EF 5 ejection fraction; ESVI 5 end-systolic volume index; LV 5 left ventricle; Pinter 5 P values between the survivor and nonsurvivor groups; Pintra 5 P values between baseline and follow-up within the group of survivors and nonsurvivors; RV 5 right ventricle; SVI 5 stroke volume index. aSignificant at P , .05. www.chestpubs.org

CHEST / 141 / 4 / APRIL, 2012

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

939

Figure 4. Right ventricular longitudinal and transverse shortening at baseline and follow-up, showing the averaged RV longitudinal (A) and transverse shortening (B) of survivors and nonsurvivors for baseline (black bars) and follow-up (white bars). A, The longitudinal shortening in the nonsurvivors is significantly (*P , .05) decreased at baseline compared with survivors, whereas no significant change during follow-up is seen in either group. B, The transverse shortening in the nonsurvivors is significantly decreased for levels 2 to 7 at baseline compared with survivors (*P , .05). During follow-up, the right ventricular transverse shortening for levels 2 to 6 further decreases significantly (†P , .05) from baseline and even shows lengthening at the midventricular to apical level, whereas in the survivors, it did not decrease. Data are presented as mean ⫾ SEM. Base 5 baseline; fol 5 follow-up.

and an especially poor outcome was shown in those patients with a TAPSE of , 15 mm. The reduced TAPSE of the nonsurvivors in the current study (mean, 15 ⫾ 6 mm) corresponds to the observations by Forfia et al.15 During a 1-year run-in period, we observed no further changes in longitudinal shortening in subsequent nonsurvivors, whereas stroke volume and RVEF showed a progressive decline in these patients. The absence of a further loss in TAPSE in the group of nonsurvivors may be explained by a floor effect beyond a mean TAPSE of 15 mm in the group of nonsurvivors. We suggest that RV dysfunction may start with a loss of TAPSE and that TAPSE continues to decrease until a lower limit, or floor, is reached.

During further progression of the disease, RV function may continue to deteriorate through a loss of transverse shortening. This floor effect also may explain the absence of a significant correlation between the change in RVEF and change in TAPSE. However, it has been shown that clinical improvement is associated with improved TAPSE.16 During the initial 1-year follow-up, there was a significant reduction in transverse shortening in nonsurvivors, which was related to a reduction in RVEF (r 5 0.66, P 5 .005). The transverse shortening comprises displacement of both the RV free wall and the septum. As for longitudinal shortening, a lower limit, or floor, also was reached for the RV free wall displacement in the group of nonsurvivors (Fig 5). Thus, increased leftward septal bowing is the main explanation for a further decline in RV transverse shortening in progressive PAH. Marcus et al17 demonstrated that maximal leftward septal bowing coincides with peak RV myocardial strain. This supports the explanation that just at the moment of peak displacement of the RV free wall, the septum bows maximally to the left, and thus, the effective transverse shortening is low or even negative (Fig 4). Because leftward septal bowing is a consequence of left-right dyssynchrony,18-20 the present finding is in line with López-Candales et al,21 who demonstrated in patients with PAH that the presence of left-right dyssynchrony correlates well with markers of disease severity.

Figure 5. RV free wall and septal transverse displacement in nonsurvivors at baseline and 1-year follow-up. For the RV free wall, all the gray bars are to the right, which means displacement toward the RV cavity. For the septum (black bars), a bar to the left means displacement toward the RV cavity, and a bar to the right means displacement outward from the RV cavity. For the free wall movements, there was no significant change at follow-up compared with baseline at any level, but there was a significant increase of leftward septum displacement for levels 2 through 7 at follow-up compared with baseline (*P , .05). ED 5 end diastole; ES 5 end systole. See Figure 1 legend for expansion of other abbreviation.

940

Original Research

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

Figure 6. A, Average RV fractional area change at baseline and follow-up of survivors and nonsurvivors for baseline (black bars) and follow-up (white bars). B, For the nonsurvivors, the change in RV fractional area change correlated with the change in RVEF (*P , .05) compared with baseline survivors (†P , .001) and nonsurvivors. RVEF 5 right ventricular ejection fraction. See Figure 1 legend for expansion of other abbreviation.

RVFAC combines the effect of both longitudinal and transverse shortening into one measure and, therefore, includes septal motion. RVFAC obtained both by CMR and by echocardiography has been shown to correlate well with CMR-derived RVEF.8,9,22 The present study also showed that there is a good correlation between the change in RVFAC and the change in RVEF over a 1-year period, suggesting that a decrease in RVFAC is an accurate reflection of RV deterioration in patients with severe PAH. Although the change in transverse shortening correlated well with the change in RV function, it has to be noted that other factors also may contribute to RV dysfunction. Among these factors are tricuspid regurgitation and the effect of the LV contraction on RV ejection.23 Recently, the importance of the RV outflow tract in the determination of RV function also has been demonstrated.24 Clinical Implications An important current challenge in PAH is to improve the monitoring of treatment outcomes in clinical practice and clinical trials.25 Because prognosis in PAH is mainly determined by RV function, there is a need for simple and reproducible measures of RV function that reflect RV deterioration over time. The present study shows that both longitudinal measures and RV free wall movements reach a lower limit in nonsurvivors. Clinically, the lack of improvement in longitudinal and transverse shortening of the RV free wall during treatment can be regarded as a sign of end-stage RV failure. On the other hand, it is still possible that an improvement of these parameters is associated with an improvement in RV function and clinical signs.16,26 RVFAC has the advantage that it quantifies shortening of both the RV free wall and the septum and www.chestpubs.org

is shown to correlate well with global RV function. However, the tracing of the RV endocardial area by two-dimensional echocardiography is difficult because of the limited range of the echocardiographic window and the trabeculated endocardial surface, thereby limiting the use of this measurement in clinical practice. Another parameter that quantifies leftward septal bowing is the eccentricity index, which also might be very useful.27 Limitations Some general methodologic considerations apply when discussing the findings. Both the RV free wall and the septal movements were determined in a fourchamber view. It should be noted that septum bowing, if present, is less distinct in this view than in a shortaxis view because the septum is intersected near its posterior attachment to the right ventricle and left ventricle. In addition, the number of patients who met the diagnostic criteria but who did not receive a follow-up CMR was high. There were a variety of reasons for patients being lost to follow-up, and baseline characteristics of these patients were not different from the other patients in this study. Therefore, we believe that it is unlikely that the outcome of the study was affected by selection bias.

Conclusions Progressive RV failure in PAH is associated with a parallel decline in RV longitudinal and transverse free wall displacement until a floor effect is reached for both. A further reduction of RV function is due to progressive leftward septal displacement. Because transverse shortening incorporates both free wall and septum displacement, this parameter can be used to CHEST / 141 / 4 / APRIL, 2012

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

941

monitor the decline of RV function in end-stage PAH. Acknowledgments Author contributions: Mr Mauritz and Dr Kind had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Mr Mauritz: contributed to the study design; data collection, analysis, and interpretation; and manuscript preparation, revision, and final approval. Dr Kind: contributed to the study design; writing the Matlab software; data collection; and manuscript preparation, revision, and final approval. Dr Marcus: contributed to reading and approving the final manuscript. Dr Bogaard: contributed to the study design, data interpretation, and manuscript revision and final approval. Dr van de Veerdonk: contributed to the data collection, analysis, and interpretation and manuscript revision and final approval. Dr Postmus: contributed to the study design, data interpretation, and manuscript revision and final approval. Dr Boonstra: contributed to the study design, data interpretation, and manuscript revision and final approval. Dr Westerhof: contributed to the study design, data interpretation, and manuscript revision and final approval. Dr Vonk-Noordegraaf: contributed to the study design, data interpretation, and manuscript revision and final approval. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Vonk-Noordegraaf receives lecture fees from Actelion Pharmaceuticals Ltd, Bayer Healthcare Pharmaceuticals, GlaxoSmithKline plc, Eli Lilly and Company, and Pfizer Inc; serves on the industry advisory board for Actelion Pharmaceuticals Ltd and Bayer Healthcare Pharmaceuticals; and serves on steering committees for Actelion Pharmaceuticals Ltd, Bayer Healthcare Pharmaceuticals, and Pfizer Inc. Mr Mauritz and Drs Kind, Marcus, Bogaard, van de Veerdonk, Postmus, Boonstra, and Westerhof have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Role of sponsors: The sponsors had no role in the design of the study, the collection and analysis of the data, or in the preparation of the manuscript. Additional information: The e-Table can be found in the Online Supplement at http://chestjournal.chestpubs.org/content/ 141/4/935/suppl/DC1.

References 1. Humbert M, Sitbon O, Simonneau G. Treatment of pulmonary arterial hypertension. N Engl J Med. 2004;351(14): 1425-1436. 2. McLaughlin VV, Presberg KW, Doyle RL, et al; American College of Chest Physicians. Prognosis of pulmonary arterial hypertension: ACCP evidence-based clinical practice guidelines. Chest. 2004;126(suppl 1):78S-92S. 3. 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(5):343-349. 4. Ueti OM, Camargo EE, Ueti AdeA, de Lima-Filho EC, Nogueira EA. Assessment of right ventricular function with Doppler echocardiographic indices derived from tricuspid annular motion: comparison with radionuclide angiography. Heart. 2002;88(3):244-248. 5. Arce OX, Knudson OA, Ellison MC, et al. Longitudinal motion of the atrioventricular annuli in children: reference values, growth related changes, and effects of right ventricular volume and pressure overload. J Am Soc Echocardiogr. 2002;15(9):906-916.

6. Kaul S, Tei C, Hopkins JM, Shah PM. Assessment of right ventricular function using two-dimensional echocardiography. Am Heart J. 1984;107(3):526-531. 7. Lindqvist P, Calcutteea A, Henein M. Echocardiography in the assessment of right heart function. Eur J Echocardiogr. 2008;9(2):225-234. 8. Kind T, Mauritz GJ, Marcus JT, van de Veerdonk M, Westerhof N, Vonk-Noordegraaf A. Right ventricular ejection fraction is better reflected by transverse rather than longitudinal wall motion in pulmonary hypertension. J Cardiovasc Magn Reson. 2010;12:35. 9. Anavekar NS, Gerson D, Skali H, Kwong RY, Yucel EK, Solomon SD. Two-dimensional assessment of right ventricular function: an echocardiographic-MRI correlative study. Echocardiography. 2007;24(5):452-456. 10. Hinderliter AL, Willis PW IV, Barst RJ, et al; Primary Pulmonary Hypertension Study Group. Effects of long-term infusion of prostacyclin (epoprostenol) on echocardiographic measures of right ventricular structure and function in primary pulmonary hypertension. Circulation. 1997;95(6):1479-1486. 11. Barst RJ, McGoon M, Torbicki A, et al. Diagnosis and differential assessment of pulmonary arterial hypertension. J Am Coll Cardiol. 2004;43(12 suppl S): S40- S47. 12. McLaughlin VV, Shillington A, Rich S. Survival in primary pulmonary hypertension: the impact of epoprostenol therapy. Circulation. 2002;106(12):1477-1482. 13. Sitbon O, Humbert M, Nunes H, et al. Long-term intravenous epoprostenol infusion in primary pulmonary hypertension: prognostic factors and survival. J Am Coll Cardiol. 2002;40(4):780-788. 14. Nijveldt R, Germans T, McCann GP, Beek AM, van Rossum AC. Semi-quantitative assessment of right ventricular function in comparison to a 3D volumetric approach: a cardiovascular magnetic resonance study. Eur Radiol. 2008;18(11): 2399-2405. 15. Forfia PR, Fisher MR, Mathai SC, et al. Tricuspid annular displacement predicts survival in pulmonary hypertension. Am J Respir Crit Care Med. 2006;174(9):1034-1041. 16. Brown SB, Raina A, Katz D, Szerlip M, Wiegers SE, Forfia PR. Longitudinal shortening accounts for the majority of right ventricular contraction in normal subjects and patients with pulmonary arterial hypertension. Chest. 2010:40(1):27-33. 17. Marcus JT, Gan CT, Zwanenburg JJ, et al. Interventricular mechanical asynchrony in pulmonary arterial hypertension: left-to-right delay in peak shortening is related to right ventricular overload and left ventricular underfilling. J Am Coll Cardiol. 2008;51(7):750-757. 18. Kalogeropoulos AP, Georgiopoulou VV, Howell S, et al. Evaluation of right intraventricular dyssynchrony by twodimensional strain echocardiography in patients with pulmonary arterial hypertension. J Am Soc Echocardiogr. 2008; 21(9):1028-1034. 19. López-Candales A, Rajagopalan N, Dohi K, Edelman K, Gulyasy B. Normal range of mechanical variables in pulmonary hypertension: a tissue Doppler imaging study. Echocardiography. 2008;25(8):864-872. 20. López-Candales A, Dohi K, Bazaz R, Edelman K. Relation of right ventricular free wall mechanical delay to right ventricular dysfunction as determined by tissue Doppler imaging. Am J Cardiol. 2005;96(4):602-606. 21. López-Candales A, Dohi K, Rajagopalan N, et al. Right ventricular dyssynchrony in patients with pulmonary hypertension is associated with disease severity and functional class. Cardiovasc Ultrasound. 2005;3:23. 22. Lang RM, Bierig M, Devereux RB, et al; American Society of Echocardiography’s Nomenclature and Standards Committee; Task Force on Chamber Quantification; American College of

942

Original Research

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

Cardiology Echocardiography Committee; American Heart Association; European Association of Echocardiography, European Society of Cardiology. Recommendations for chamber quantification. Eur J Echocardiogr. 2006;7(2):79-108. 23. Rushmer RF. Cardiovascular Dynamics. Philadelphia, PA: WB Saunders Company; 1976. 24. Calcutteea A, Chung R, Lindqvist P, Hodson M, Henein MY. Differential right ventricular regional function and the effect of pulmonary hypertension: three-dimensional echo study. Heart. 2011:97(12):1004-1011.

www.chestpubs.org

25. McLaughlin VV, Suissa S. Prognosis of pulmonary arterial hypertension: the power of clinical registries of rare diseases. Circulation. 2010;122(2):106-108. 26. Basil A, Liu T, Arkles J, et al. Serial changes in TAPSE correlate with functional assessment and clinical events in patients with PAH. Am J Respir Crit Care Med. 2009;179: A4882. 27. Raymond RJ, Hinderliter AL, Willis PW, et al. Echocardiographic predictors of adverse outcomes in primary pulmonary hypertension. J Am Coll Cardiol. 2002;39(7):1214-1219.

CHEST / 141 / 4 / APRIL, 2012

Downloaded from chestjournal.chestpubs.org by Kimberly Henricks on April 5, 2012 © 2012 American College of Chest Physicians

943