Background In pulmonary arterial hypertension (PAH) a prolonged time interval between pulmonary valve closure and tricuspid valve opening is found. This period is interpreted as prolonged right ventricular (RV) relaxation, and thus a reflection of diastolic dysfunction. This concept recently has been questioned, since it was shown that RV contraction continues after pulmonary valve closure causing a post-systolic contraction period.
Objectives To investigate in PAH whether the increased RV post-systolic isovolumic period is caused by either an additional post-systolic contraction period, or an increased relaxation period (diastolic dysfunction).
Methods 23 patients with PAH (mean pulmonary arterial pressure 54±12 mm Hg), and 18 healthy subjects were studied using cardiac MRI. In a RV two-chamber view, times of pulmonary valve closure (TPVC) and tricuspid valve opening (TTVO) were measured, defining the total post-systolic isovolumic period. Time to peak of RV free wall contraction (TpeakRV) was determined with myocardial tagging. Post-systolic contraction and relaxation periods were defined as the time intervals between TPVC and TpeakRV and between TpeakRV and TTVO, respectively. These periods were normalised to an RR interval.
Results The total post-systolic isovolumic period was longer in patients than in healthy subjects (0.15±0.04 vs 0.04±0.02, p<0.001), but the relaxation period was not different (0.06±0.02 vs 0.05±0.02, p=0.09). The post-systolic contraction period in patients was strongly related to the total post-systolic isovolumic period (y=0.98x–0.05; r=0.89, p<0.001), and was associated with disease severity.
Conclusion In PAH, the prolonged post-systolic isovolumic period is caused by an additional post-systolic contraction period, rather than by an increased relaxation period.
- Pulmonary hypertension
- heart failure
- isovolumic relaxation time
- myocardial contraction
- diastolic dysfunction
- pulmonary arterial hypertension (PAH)
Statistics from Altmetric.com
- Pulmonary hypertension
- heart failure
- isovolumic relaxation time
- myocardial contraction
- diastolic dysfunction
- pulmonary arterial hypertension (PAH)
In pulmonary arterial hypertension (PAH), a prolonged post-systolic isovolumic time interval, here called post-systolic isovolumic period, between pulmonary valve closure and tricuspid valve opening is seen.1–4 Several reports have shown that this post-systolic isovolumic period is related to disease severity,5 6 and is assumed to mirror prolonged right ventricular (RV) isovolumic relaxation3 and therefore interpreted as a reflection of RV diastolic dysfunction.2 7–10
However, Marcus et al11 demonstrated that in PAH the shortening (contraction) of the RV free wall continues after pulmonary valve closure, as an indication of a post-systolic (or post-ejection) contraction (see figure 1).
Consequently, this raises the question whether the current interpretation of the increased post-systolic isovolumic period as prolonged RV relaxation (diastolic dysfunction) is valid in PAH. Alternatively, a post-systolic contraction period followed by a normal isovolumic relaxation period might explain this prolonged time interval.
Therefore, the aim of this study was to explore in PAH whether the increased post-systolic isovolumic period is caused by either a post-systolic contraction, or an increased relaxation period (diastolic dysfunction), or both. To answer this question a group of patients with PAH of different disease severity were studied using cardiac MRI.
This study was conducted within the Pulmonary Hypertension Program at the VU University Medical Center. The institutional review board approved the conduct of this study and all patients gave informed consent before enrolment. We recruited 23 consecutive patients referred for right heart catheterisation in the follow-up of their PAH. All patients also underwent cardiovascular magnetic resonance (CMR) imaging as part of their disease evaluation.
A mean pulmonary artery pressure (PAP) of >25 mm Hg with a pulmonary capillary wedge pressure of <15 mm Hg was considered to be PAH.12 13 Patients with any form of cardiopulmonary disease and those who were unable to undergo CMR were excluded. Results were compared with 18 age- and gender-matched, non-smoking controls subjects, without a history of cardiopulmonary diseases.
CMR imaging was performed on a 1.5T scanner (Magnetom Sonata, Siemens, Erlangen, Germany). Complementary tagged (CSPAMM) myocardial images with high temporal resolution (14 ms) were acquired in all patients using steady-state free precession (SSFP) imaging and a multiple brief expiration breath-hold scheme as described by Zwanenburg et al.14 Parameters were: three phase-encoding lines/beat, repetition time 4.7 ms, echo time 2.3 ms, no view sharing, flip angle 20°, voxel size 1.2×3.8×6.0 mm3. Images for two-dimensional strain analysis were acquired in the mid-ventricular short-axis planes. With these tagging cine images, contraction and relaxation within the myocardial wall can be assessed.
In patients and control subjects SSFP cine imaging (without tagging) was performed with full coverage of the left ventricle and right ventricle (stack of short-axis slices) to assess ventricular volumes and ejection fraction (EF).
In addition, to determine the moments of the tricuspid and pulmonary valve closures, a long-axis cine image in the RV two-chamber view was acquired. For this purpose, SSFP imaging with view sharing was used to obtain a temporal resolution of 15 ms in a single breath-hold.
CMR data analysis
The tagged images were analysed with the harmonic phase procedure.15 Circumferential shortening, assumed to reflect contraction, was calculated as a function of time over the cardiac cycle. For the left ventricular (LV) and RV free wall the strains and strain timing parameters were derived as described previously.11 For the whole RV and LV free wall at mid-level, the time to peak of RV circumferential shortening (TpeakRV) was calculated related to the ECG R wave by automated routines.14
The times to pulmonary valve closure (TPVC) and tricuspid valve opening (TTVO) were assessed from RV two-chamber cine images. In three patients pulmonary valve timing was derived from the most basal short-axis cine that showed the valves during the last part of systole.
The stack of short-axis cine images was used for the calculation of the LV and RV end-diastolic volumes and the RV end-systolic volume. Since stroke volume (SV) from RV volumes has limited accuracy,16 LV volumes were used to calculate SV. RVEF was calculated using RV end-diastolic volumes and LV SVs. The time to maximal leftward septal bowing was measured at the most basal short-axis cine slice that still showed the LV and RV myocardium through the cardiac cycle.
Definitions of cardiac phases
In patients and healthy controls the post-systolic isovolumic period was defined as the time interval between TPVC and TTVO. In the healthy subjects this time interval was designated as isovolumic relaxation period.17
In the patients this post-systolic isovolumic period was found to consist of a contraction period and a relaxation period (figure 1). The isovolumic contraction period was defined as the time interval between TPVC (end systole) and TpeakRV. The RV isovolumic relaxation period was defined as the time interval between TpeakRV and TTVO. To minimise the influence of the heart rate, all periods were normalised to the heart period, the R to R interval.
Results are expressed as mean±SD. Statistical significance was set at a value of p<0.05. Comparisons between patients and control subjects were made with unpaired t-tests, without correction for multiple comparisons. The relations between the post-systolic contraction period versus the post-systolic isovolumic period and the post-systolic relaxation period versus the post-systolic isovolumic period were tested by linear regression. Linear correlations between the post-systolic contraction period and pulmonary vascular resistance (PVR), systolic PAP, stroke volume index and RVEF, were also calculated.
The interobserver variation was determined for the time to pulmonary valve closure and time to tricuspid valve opening by Bland–Altman analysis, for a subset of 10 patients. All statistical analyses were performed with SPSS Statistics 15.0 (SPSS Inc)
There were no differences between the 23 patients with PAH and 18 controls with respect to age (PAH=43±14 years vs controls=38±9 years) and proportion of male to female subjects (PAH=6/17 vs controls=6/12, Fisher's exact test, p=0.8). Patients' characteristics and haemodynamic variables are presented in table 1. On the basis of the ECG morphology, three patients had an incomplete, and two patients a complete, right bundle branch block (RBBB). Seventeen patients were diagnosed as having idiopathic PAH, whereas six had chronic thromboembolic PAH. The majority of patients were in New York Heart Association functional class III. Haemodynamics yielded characteristics of RV pressure overload. MRI data of healthy controls and patients with PAH are shown in table 2.
Images, strains and valves timing
Figure 2 shows in a patient with PAH, RV two-chamber images at the time of pulmonary valve closure and at the time of tricuspid valve opening, a short-axis cine image at TpeakRV and the circumferential strain curves during the cardiac cycle for the LV and RV free walls. The contraction of the left and right ventricles starts simultaneously, but the right ventricle reaches its peak later than the left ventricle. As can be seen, RV contraction continues after TPVC (post-systolic contraction) and reaches its peak before TTVO, dividing the post-systolic isovolumic period into a contraction period and a relaxation period. The post-systolic isovolumic period divided by RR interval in the patients was longer than in the control subjects (0.15±0.04 vs 0.04 ±0.017, p<0.001). However, the time to pulmonary valve closure, divided by RR interval, was not earlier in the patients (table 2).
Relation of contraction and relaxation with the post-systolic isovolumic period
As shown in figure 3, in the patients with PAH a significant relation exists between the RR-normalised post-systolic contraction period and the post-systolic isovolumic period (r=0.89, y=0.98x–0.05 p=0.0001), but no significant relation is found for the relaxation period and the post-systolic isovolumic period (r=0.05, p=0.82). Additionally, in table 2 and figure 4, a comparison of the relaxation period between patients and healthy controls (0.057±0.018 and, 0.047±0.017, respectively) shows no significant difference (p=0.09).
Regression analysis of the post-systolic contraction period and disease severity
As shown in figure 5, there was an association between the post-systolic contraction period and PVR (p <0.001, r=0.75) and with systolic PAP (p=0.002, r=0.61). Furthermore, the post-systolic contraction period was negatively related to stroke volume index (r=−0.46, p=0.03) and RVEF (r=−0.58, p=0.003). None of these associations was found for the post-systolic relaxation period.
The interobserver variation in the time to pulmonary valve closure was given by a correlation coefficient of 0.87 with p <0.001, and a bias of 9 ms with 95% confidence limits of agreement of −25 and 45 ms, respectively.
For the time to tricuspid valve opening, interobserver variation was given by a correlation coefficient of 0.89 with p< 0.001, and a bias of 4 ms with 95% confidence limits of agreement of −37 and 44 ms, respectively.
In normal physiology, the period between pulmonary valve closure and tricuspid valve opening represents the RV isovolumic relaxation period only.17 18 Until now, prolongation of this period has been interpreted as RV diastolic dysfunction.4 10 19–21
In this study the time of tricuspid valve opening was measured, which made it possible to analyse two separate time intervals in the RV post-systolic isovolumic period: (a) the post-systolic RV contraction time, between pulmonary valve closure and time to peak strain, and (b) the true isovolumic relaxation time, between time to peak strain and tricuspid valve opening. Thus, in PAH the RV post-systolic isovolumic period cannot be interpreted as a measure of RV diastolic function only, and thus should not be labelled as an isovolumic relaxation time. Second, the post-systolic RV contraction is associated with PAP and PVR. And finally, the true RV isovolumic relaxation time is not associated with PAP and PVR, and not prolonged.
Therefore in patients with PAH the increased post-systolic isovolumic period results from prolonged RV contraction duration rather than slower relaxation.
Figure 6 shows a pathophysiological concept, integrating our observations with previous findings in pulmonary hypertension. RV contraction starts simultaneously with LV contraction. Then follows the ejection period until pulmonary valve closure. After pulmonary valve closure (defined as end systole), RV contraction continues with a post-systolic contraction. This prolonged RV contraction, still continuing after the peak of LV contraction11 22 (see also figure 2) leads to a reversal of the trans-septal pressure gradient, which causes the post-systolic leftward bowing of the septum into the left ventricle and thereby impairs LV filling.20 23 24 Thus the post-systolic isovolumic period starts with an abnormal contraction period, during which the RV contraction energy is wasted in the non-functional septum bowing, which causes inefficiency in both RV systole and LV diastole.25 Only after peak RV contraction, which coincides with maximum leftward septum bowing, does a normal RV relaxation period begin.
Post-systolic contraction period and disease severity
Previous findings have shown that in pulmonary hypertension prolongation of the post-systolic isovolumic period, is related to PVR, PAP systolic and RVEF.2 4 5 These findings are in line with our results. However, we show here that the interpretation needs revision. Based on our results, it is the post-systolic contraction period that depends on disease severity (figure 5), in contrast to the isovolumic relaxation period, which appears to be disease independent
The mechanism of the prolonged contraction period is probably the increased afterload—that is, higher systolic PAP (figure 5). Increased contraction duration with increased load has been shown in whole heart26 27 and cardiac muscle28 29 studies. At the level of the sarcomeres, an increased load leads to prolonged time of shortening, as found in isolated cardiac trabeculae.30 At the organ level of the right ventricle, the prolonged shortening with constant RV volume is made possible by the early pressure decline in the left ventricle, and the subsequent leftward septal bowing.11
As mentioned above, the observed increase in the post-systolic isovolumic period is a reflection of increased RV contraction period. A prolongation of this period is a manifestation of the inefficient post-systolic RV contraction which is an important sign of RV pressure overload in PAH. This is in contrast with the current concept so far that this period is completely a relaxation period. This study therefore shows that in the right ventricle of patients with PAH, this prolonged isovolumic period is not a sign of diastolic dysfunction.
Because of the twofold disadvantageous effect of prolonged RV shortening both on RV and LV function, treatment could focus on reduction of the prolonged RV post-systolic contraction. Dual pacing could be attempted to reduce the L-R delay in peak contraction, as shown earlier in an animal model of PAH31 and by a computer simulation analysis.32 Another option is to reduce the imbalance in contraction duration between the chambers with pharmacological methods that limit the variability and load dependence of contraction. Such treatments may improve both RV systolic efficiency and LV diastolic filling.25 33
How can we measure the RV post-systolic isovolumic period in daily practice, and thereby the effect of any treatment? It is the time interval between pulmonary valve closure and tricuspid valve opening, which can easily be measured by echocardiography.1 3–6 17 Therefore, measurement of this period provides an easy tool for monitoring individual patients during treatment.
We acknowledge some limitations of our study. The temporal resolution of 15 ms limits the accuracy of the timing of valves and peak strain. Furthermore, the tagged myocardial cine images and the RV two-chamber cine images could not be acquired simultaneously. However, the acquisitions of both cine images were obtained within the same MRI session, with no discernible difference in cardiac conditions, such as heart rate. The lack of longitudinal and radial strain is a limitation. However, it is unlikely that these other strain components would have a different time to peak value.
Finally, patients with and without targeted medication were included, but, nevertheless, a prolonged contraction was found for all patients .
In PAH, the time interval between pulmonary valve closure and tricuspid valve opening (called post-systolic isovolumic period) is increased and consists of a contraction period and a relaxation period. The increase in post-systolic isovolumic period is only caused by increased RV contraction duration, and is not a reflection of diastolic dysfunction.
The authors thank Professor Dirk-Jan Duncker, MD PhD, of the Department of Experimental Cardiology, Erasmus Medical Center, Rotterdam, for critical comments on the manuscript.
Funding Anton Vonk-Noordegraaf was supported by Netherlands Organisation for Scientific Research (NWO)-VIDI.
Competing interests None.
Ethics approval This study was conducted with the approval of the the institutional review board.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.