To the Editors:
Pulmonary arterial hypertension (PAH) is a rare, life-threatening dyspnoea–fatigue syndrome, caused by progressive increase in pulmonary vascular resistance (PVR) and eventual right ventricular failure [1]. The heritable form of PAH has been shown to be associated with mutations of the gene encoding the bone morphogenetic protein receptor-2 (BMPR2). Asymptomatic carriers of BMPR2 mutations are at high risk of developing PAH [2]. Careful follow-up of these subjects might help to detect early-stage disease with a more favourable response to targeted therapies. However, there is uncertainty about the optimal screening method. A recent European multicentre study showed that relatives of patients with idiopathic PAH (IPAH) present with an increased prevalence of abnormally high pulmonary artery pressure (PAP) during exercise or during low-oxygen breathing [3]. In that study, measurements of pulmonary vascular function were limited to a systolic PAP (sPAP) estimated from the maximum tricuspid regurgitation velocity (TRV) assessed by Doppler echocardiography. Here, we report on additional measurements performed in one of the participating centres, providing insight into abnormal pulmonary vascular distensibility and hypoxia-induced PVR in healthy BMPR2 carriers.
This study was part of a larger multicentre European project, which included 291 relatives of 109 IPAH patients and 191 age-matched controls [3]. The participating PAH centre in Brussels, Belgium, contributed with 35 relatives of 10 index patients with IPAH and 38 healthy controls. The 35 relatives were aged mean±sd 35±14 yrs. The 38 controls were aged mean±sd 36±10 yrs, and matched for sex and body surface area. Five of the asymptomatic relatives and two of the index IPAH patients were carriers of BMPR2 mutations. All relatives and controls underwent complete standard Doppler echocardiography at rest, followed by repetitive measurements of sPAP during exercise, with workload increased by 25 W every 2 min up to the maximum tolerated, and after 2 h of breathing an inspiratory oxygen fraction of 0.12, as previously described [3]. sPAP was calculated from maximum TRV [4].
Additional analysis specific to this study consisted of right ventricular outflow tract time–velocity integral (RVOTVTI) during the resting normoxic and hypoxic measurements, to estimate a mean PAP (mPAP) from the acceleration time of pulmonary flow velocity [5], and PVR calculated as 10×TRV/RVOTVTI+0.16 Wood units [6]. mPAP was also calculated from sPAP using the formula mPAP=0.6×sPAP+2 mmHg [7]. Cardiac output (Q) was estimated from the left ventricular outflow tract (LVOT) diameter, LVOTVTI and cardiac frequency (fC) at rest [8], and exercise–associated changes estimated from proportional changes in fC. Linear regressions were calculated on multipoint mPAP–Q plots, with adjustment for individual variability [9]. As multipoint mPAP–Q relationships actually disclose a slight curvilinearity, a distensibility coefficient α accounting for it was calculated using the equation mPAP=(((1+α×LAP)5+5αR0Q)1/5-1)/α, where R0 is total PVR at rest and LAP is left atrial pressure, which is assumed to be unchanged from the resting state, as previously described [9]. Data are presented as mean±sd. The statistical analysis consisted of a repeated-measures ANOVA and modified paired t-tests when the F-ratio of the ANOVA reached a critical p<0.05 value.
The main results with measurements at rest, at the maximum workload achieved and after 2 h of hypoxia, are shown in table 1. Maximum workloads were 103±45 W in the BMPR2 carrier relatives, 113±38 W in the non-BMPR2 carrier relatives and 119±33 W in the controls. Resting mPAP calculated from TRV or from the acceleration time of RVOTVTI were not different: 14±2 versus 12±6, 15±2 versus 14±3 and 15±3 versus 15±3 mmHg in normoxia, and 28±5 versus 29±5, 27±5 versus 25±6 and 25±4 versus 22±5 mmHg in hypoxia in BMPR2 carrier relatives, non-BMPR2 carrier relatives and controls, respectively (p-values were always nonsignificant between the two methods). Blood pressure, fC, mPAP, Q, PVR and the slopes of mPAP–Q relationships were not different between the three groups. PVR did not decrease during exercise in the BMPR2 carriers. Individual mPAP–Q relationships showed a marked interindividual variability (fig. 1). However, α was markedly decreased in the BMPR2 carriers, at rest and during exercise. Furthermore, the BMPR2 carriers also had an enhanced hypoxic pressure response when expressed as PVR versus arterial oxygen saturation measured by pulse oximetry. Hypoxia was associated with a decrease in α, which was of the same magnitude in the three study groups. There was a correlation between mPAP after 120 min of hypoxia and at maximal exercise (r=0.49, p<0.001).
The present results show the potential relevance of refined analysis of stress echocardiography PAP responses to exercise and hypoxia in terms of pressure–flow relationships and PVR–Sp,O2 plots. Previous invasive studies have shown that multipoint mPAP–Q relationships conform to a distensible mathematical model of the pulmonary circulation better than the PVR equation [10]. Thus, adding a measure of the compliance of pulmonary resistive vessels allows for an improved prediction of pressures at given levels of flow. It is interesting that α calculated at 2% change in diameter per mmHg transmural pressure from invasively as well as noninvasively measured multipoint pressure–flow plots agrees with in vitro measurements in isolated arterioles [9, 10]. Whether decreased α in the asymptomatic BMPR2 mutation carriers in the present study reflects early pathological changes with potential for progression is not known.
Our data also show that healthy BMPR2 mutation carriers present with an enhanced hypoxic pressure response to hypoxia. This result agrees with recent report that BMPR2 mutant mice are more susceptible to hypoxic pulmonary hypertension [11]. Hypoxic breathing was associated with the expected decrease in α in the three study groups [10], suggesting that any intrinsic difference in pulmonary vascular compliance is overwhelmed by hypoxic pulmonary vasoconstriction.
There are obvious limitations to the present preliminary report. The number of healthy BMPR2 mutation carriers was small, and pressure and flow calculations from noninvasive measurements rested on several assumptions. IPAH is a rare disease, with an incidence of one to two cases per million per year, only a minority of these patients is carrying a BMPR2 mutation, and asymptomatic relatives carrying a BMPR2 mutation are difficult to identify [1–3]. Exercise stress echocardiographic estimates of PAP and Q have previously been reported to be associated with realistic α calculations [9]. Left atrial pressure was not measured, but assumed constant, in keeping with previous estimates in normal subjects at exercise [9] and invasive studies showing that significant increases in LAP occur at exercise only at high workloads >150 W associated with cardiac outputs over 15–20 L·min−1, much higher than in the present study [12]. The assumption in the present study that fC would be the main determinant of exercise-induced increase in Q may be more debatable. We therefore compared distensibility α and Q calculated from changes in fC only or from repetitive LVOTVTI and fC measurements during exercise stress echocardiography in a total of 20 healthy subjects included in a recently reported study [13]. The results of both methods were highly correlated. The fC-only method underestimated Q only at values >15 L·min−1 and α was underestimated by ∼10 %.
It is hoped that the present echocardiographic approach for the study of pulmonary vascular function will be applied with more direct measurements and further validation in future studies on patients with risk factors for the development of PAH.
Footnotes
Support Statement
This study was funded by a grant from the European Union Fifth Framework programme "Disposition to PPH, QLGI-CT-2002-01116".
Statement of Interest
Statements of interest for J-L. Vachiéry, E. Grunig and R. Naeije can be found at www.erj.ersjournals.com/site/misc/statements.xhtml
- ©ERS 2012