Quantitative Analysis of Right Ventricular Wall Motion

In normal subjects, RV volumes are usually higher than LV volumes. There is an increase of RV volume indices (normalized to the body surface area) with age. Considerable interindividual differences of normal RV volumes, geometry, and contraction indicate again the variability of RV performance and the wide range of normality.

The complex and variable shape and geometry of the RV hinder simple and correct measurements of RV volumes and ejection fractions. The greatest error in estimating RV volumes from RV an-giograms is probably due to the need for geometric assumptions, when only two-dimensional data are available (Table 16.2).

Methods for Quantitative Assessment

Several methods have been described using geometric models and formulas to determine end diastolic

Fig. 16.9 • Right ventricular (RV) angiogram (30° RAO view, systole) showing global RV enlargement and severe dysfunction in ARVC/D. There is pronounced dilatation of the RV outflow tract, akinesia of the inferior and subtricuspid walls and functional tricuspid regurgitation (grade 2) due to annulus dilatation

Table 16.2 • RV volumes and ejection fractions in normal control subjects

Author

Year

(%)

RV-EDV (ml/m2)

RV-ESV (ml/m2)

RV-SV (ml/m2)

Graham [30]

1973

9**

AP, LAT

64±9

70±13

NR

NR

Genzler [29]

1974

9

AP, LAT

51±8

81±12

39±9

NR

Ferlinz [25]

1975

NR

66±6

76±11

NR

NR

Lange [23]

1978

100**

AP, LAT

63±7

73±12

27±7

46±9

Redington [31]

1988

10

AP, LAT

62±6

62±13

NR

43±8

Daubert [17]

1988

10

RAO, LAO

59±6

79±10

32±6

47±8

Chioin [32]

1989

22

AP, LAT

60±6

94±15*

37±9*

57±11*

Chiddo [24]

1989

16

RAO

55±5

80±8

NR

NR

Daliento [16]

1990

18

NR

60

95*

38*

NR

Hebert [20]

2004

11

RAO, LAO

58±5

82±21

35±11

46±13

* ml (not indexed to body surface),** children AP, anteroposterior; EDV, end diastolic volume, lateral (sagittal); NR, not reported; Pts, patients;

; EF, ejection fraction; ESV, end systolic volume; LAO, left anterior oblique; LAT, RAO, right anterior oblique; RV, right ventricle; SV, stroke volume

* ml (not indexed to body surface),** children AP, anteroposterior; EDV, end diastolic volume, lateral (sagittal); NR, not reported; Pts, patients;

; EF, ejection fraction; ESV, end systolic volume; LAO, left anterior oblique; LAT, RAO, right anterior oblique; RV, right ventricle; SV, stroke volume and end systolic volumes from RV angiograms. The majority of these techniques rely on a surface-length method based on Simpson's rule and using biplane models and computer-assisted analysis [23,24]. Another method was described by Ferlinz [25] who used a model of a pyramid with a triangular base. However, the various techniques to calculate RV volumes from angiograms give different results and lack interobserver reproducibility.

Boak's formula (RV volume = [n/4] x [Area RAO X Area LAO/max. height of LAO image]) assumes that RV volumes can be obtained by summing cross sections of a given thickness from the apex to the outflow tract [26]. This model has shown close correlation between calculated RV volumes and true volumes of casts measured by water volume displacement, irrespective of RV end-diastolic volume and RV morphology.

Peters et al. [27] tested six different geometric models for the calculation of RV volumes in 20 patients and also found that Boak's formula [26] and a parallelepiped-model produced the most reliable results with a high correlation factor and low systemic error when biplane views (30° RAO and 60° LAO) were used.

The accuracy of any RV volume calculation by contrast angiography is dependent on how contours are defined. While both manual and semiautomat-ed methods have similar precision and variability, each method requires a different calibration [28-32]. However, only few of the proposed methods were validated by correlation of the calculated RV volumes with measurements from casts of human hearts.

Right Ventricular Volumes

Most studies investigating RV volumes in ARVC/D used cut-off values of the mean plus two standard deviations in the control group to define abnormal RV volumes. To correct for differences in body size, it is crucial to calculate volumes indexed for body surface area (ml/m2). In addition, physical activity (i.e., trained athlete) should be taken into account when RV volumes are assessed and interpreted.

Most of the early angiographic studies on ARVC/D demonstrated increased RV volumes and diameters and reduced global ejection fraction. However, this was in part due to a selection bias because these studies mainly included patients with advanced manifestations of ARVC/D and clearly enlarged RVs [16,17].

The systematic study by Daliento et al. [16] also showed increased RV volumes in ARVC/D when compared with normal control subjects. However, all volume parameters overlapped when patients with ARVC/D were compared to those with atrial septal defect and dilative cardiomyopathy. Thus, increased global RV volumes and reduced ejection fractions are not specific for ARVC/D but rather indicate RV dilatation due to volume overload or reduced pump function of any cause.

Therefore, global RV sizes and volumes have only limited value in the detection and the diagnosis of ARVC/D. This is indicated by data showing that in mild or moderate manifestations of ARVC/D with only regional RV dysfunction, the sensitivity and specificity of global RV volumes and ejection fraction remain low because of the significant overlap with normality. This was confirmed in more recent and larger studies, where 68% of patients with ARVC/D had a normal global RV ejection fraction [20]. In addition, the diagnostic value of global RV volumes and ejection fraction is also reduced when ARVC/D is compared with the many other conditions that may underlie mild or moderate RV enlargement.

Quantitative analyses of regional RV size, volume, structure, and systolic as well as diastolic wall motion from RV angiograms are required to improve the diagnostic evaluation, particularly in mild forms of regionally reduced RV function.

New Computer Software

To meet this requirement of regional quantification of RV function, a new computer-based software for frame-by-frame evaluation of RV wall motion in systole and diastole is currently under development at the University of Arizona.

This new quantitative analysis program is expected to significantly improve the detection and evalua tion of localized mild to moderate abnormalities of RV wall motion, particularly in ARVC/D [19]. It compares the contour movement through the contraction (systolic) phase at any specified region in the anterior-posterior, lateral, RAO, and LAO views. In addition, the program can also compare contour movement between nearby points. This relative movement is analogous to the concept of strain used in echocar-diography, and is analyzed to quantify the nonuni-formity of contour movement. Strain can be computed for the test subject in a given region, and then compared to the strain in that region for normal subjects. The combination of contour area movement and strain can identify areas of subtle hypokinesia, akinesia, and dyskinesia. Furthermore, the software provides analysis tools for the calculation of global RV volumes and ejection fraction, and for quantification of global and regional RV diastolic relaxation.

An example of the quantitative analysis of contour area movement is illustrated in Figure 16.11. This angiogram was obtained from the asymptomatic but affected father of twin sisters who were both diagnosed with ARVC/D. Both the twins dis

Fig. 16.11 • Quantitative analysis of regional right ventricular (RV) wall motion (30°RAO view). Overall RV size in this subject appears mildly enlarged.The insert in the upper left shows contrast filling the RV, imaged in the RAO view. The left panel shows the superimposed contours of the RV image at end diastole (black) and end systole (green) to show total contour movement through systole. Letters identify the boundaries of 60° sectors in a clockwise fashion beginning at the pulmonic valve region. In the right panel, contour area movement is computed every 3° and normalized to the projected area of the end diastolic contour. This normalized contour area change is then displayed as a function of angle, where 0° is at the pulmonic valve (A) and proceeding clockwise. For instance,C corresponds to 120°, located just inferior to the RV apex.The normalized contour area movement of the subject is then compared to a database of normal subjects, with the green hatched area corresponding to ± two standard deviations from the mean of normal subjects. In this subject, motion in the inferior wall (between C and D) was markedly abnormal and fell well below two standard deviations of the normal database played extreme RV enlargement accompanied by severe right heart failure, and both required cardiac transplantation. The father's RV was minimally enlarged. The anteroseptal area (region between lines A and B) shows little movement, but is well within two standard deviations of the normal database for this region. However, the inferior wall (in the sector between letters C and D) was markedly hypokinetic compared to the normal database, falling well below two standard deviations.

Was this article helpful?

0 0

Post a comment