Acute vasodilator response to vardenafil and clinical outcome in patients with pulmonary hypertension
Abstract
Purpose Acute vasodilator testing is recommended in patients with pulmonary arterial hypertension to identify individuals who may benefit from long-term treatment with oral calcium channel blockers. The aim of this study was to investigate the use of vardenafil in acute vasoreactivity testing compared to adenosine. Methods A total of 20 patients eligible for right heart cathe- terisation were enrolled. Acute vasoreactivity testing was car- ried out with intravenous (iv) adenosine (n=18) followed by oral vardenafil (n=20). Haemodynamic responses were re- corded at baseline and after 60 min (vardenafil). Responders were defined according to consensus guideline criteria.
Results Both vardenafil and adenosine significantly decreased mean pulmonary arterial pressure (mPAP, p<0.001 and p= 0.026, respectively) and pulmonary vascular resistance (p<0.001 and p>0.001, respectively), and significantly in- creased cardiac output (p=0.001 and p=0.005, respectively).
Vardenafil reduced mPAP more than adenosine (p=0.044), while adenosine resulted in higher responses of cardiac index (p = 0.009) and pulmonary arterial oxygen saturation (p = 0.042). Acute adverse reactions were common with adeno- sine, while no side effects were observed after a single oral dose vardenafil. Vardenafil identified five responders (out of 20), while adenosine identified three responders (out of 18). During a 7-year follow-up, vardenafil responders had signifi- cantly lower NT-proBNP levels compared to non-responders. Conclusions Vardenafil may be safely used for acute vasoreactivity testing in patients with PH. A single oral dose of vardenafil is better tolerated than iv adenosine and may identify additional responders who could benefit from long- term vasodilator treatment.
Keywords : Adenosine . Vardenafil . Acute vasodilator test . Haemodynamics . Pulmonary hypertension . Right heart catheterisation
Introduction
Pulmonary hypertension (PH) is a chronic disease characterised by endothelial dysfunction of the small pulmo- nary arteries, resulting in progressive increase of pulmonary vascular resistance (PVR) ultimately leading to right ventric- ular heart failure and death [1]. The World Health Organisa- tion (WHO) divides PH into five clinical groups, and pulmo- nary arterial hypertension (PAH) is one of them [2, 3]. PAH is characterised as a persistent increase in mean pulmonary arte- rial pressure (mPAP) to ≥25 mmHg at rest, with a mean pul- monary artery wedge pressure (mPAWP) ≤15 mmHg and a normal or reduced cardiac output (CO). Acute pulmonary vasoreactivity testing is performed as part of diagnostic work- up during right heart catheterisation (RHC). Currently, three agents are recommended for acute vasodilator testing: inhaled nitric oxide (NO) and intravenous (iv) adenosine or epoprostenol. Diagnostic acute vasodilator testing allows identification of responders and may guide therapeutic strate- gies in patients with PAH. Among responders, long-term treat- ment with calcium channel blockers (CCBs) has resulted in improved survival [4–7].
More recently, novel treatment options including prostacyclin analogues, endothelin receptor antagonists, phosphodiesterase type 5 (PDE5) inhibitors and soluble guanylate cyclase have become available. PDE5 inhibitors prevent the degradation of cyclic guanosine monophosphate (cGMP) [7–9]. cGMP triggers the enzyme cGMP kinase, which, in turn, activates mitochondrial KATP channels, resulting in decreased intracellular calcium levels followed by smooth muscle relaxation and vasodilatation. Among the three available PDE5 inhibitors: vardenafil, sildenafil and tadalafil, only the latter two are approved for PAH treatment. Vardenafil appears to be an attractive alternative for PAH treat- ment, as it is a potent and highly selective PDE5 inhibitor causing rapid vasodilation and exhibiting relative pulmonary selectivity [10–14]. Indeed, vardenafil has recently been proven effective and well-tolerated in a long-term randomised controlled trial of PAH patients [15]. Inhibition of calcium fluxes especially in the pulmonary artery may serve as an explanation for more effective vasorelaxation compared to sildenafil and tadalafil [16–18]. Due to the relative pulmonary selectivity and comparatively short half-life (3.4 h) [19], vardenafil could also be an attractive can- didate for acute vasoreactivity testing. However, the diagnostic utility of vardenafil for identifying subjects who may benefit from CCB and/or PDE5 inhibitor treatment has not been established so far.
The primary aim of the present study was to explore the clinical usefulness of vardenafil in acute vasoreactivity testing and to compare haemodynamic responses to oral vardenafil with iv adenosine. A secondary aim was to evaluate the pre- dictive value of haemodynamic responsiveness to vardenafil in relation to long-term clinical outcome.
Materials and methods
Subjects and study design
Between 2006 and 2007, 20 consecutive patients with PAH or suspicious of PAH who underwent RHC at the Regional Cen- tre for PAH in Uppsala were enrolled in the study. Patients over 18 years of age having resting mPAP ≥25 mmHg as measured by RHC were eligible for the study. Exclusion criteria were severe liver dysfunction (Child-Pugh class C) or hypotension (<90/50 mmHg). Diagnostic workup included echocardiography, chest X-ray, lung function tests, 6-min walk distance (6MWD), and in selected cases computed to- mography scan of the lung and pulmonary angiography. The estimated absolute glomerular filtration rate (eGFR) was cal- culated with the Cockcroft-Gault formula. Two patients re- fused vasoreactivity testing with adenosine due to fear of side effects but consented to oral vardenafil testing. The study was approved by the Independent Ethics Com- mittee, Uppsala, Sweden, and was conducted in accordance with the International Conference on Harmonization Good Clinical Practice guidelines and the Helsinki Declaration. Written informed consent was obtained from all patients en- rolled in the study. Haemodynamic assessment After an overnight fast, all patients underwent RHC. All ex- aminations were performed by the same physician (G.W.), except for one patient. Details of the procedure were presented in an earlier publication [19]. Patients were hospitalised 24 h before and after RHC intervention and were in a resting supine position during the investigation. Haemodynamic data and blood gases, central venous pressure (CVP), mPAP, mPAWP, mean right atrial pressure (mRAP), CO, cardiac index (CI) and mixed venous oxygen saturation were measured at base- line and in connection with vasodilator tests during ongoing RHC. PVR and systemic vascular resistance (SVR) as well as PVR/SVR were calculated based on measured heamodynamic parameters. The PVR/SVR ratio served as an indicator of the pulmonary selectivity of the vasodilatory effect. Vasodilator drug administration After measurement of baseline haemodynamic parameters, aden- osine infusions were started at a dose of 70 μg kg−1 min−1. Doses were gradually increased in 2-min intervals and titrated up to the maximum tolerated dose, but not higher than 210 μg kg−1 min−1. Six patients reached the pre-defined maximum dose of 210 μg kg−1 min−1; 12 patients received 140 μg kg−1 min−1 and one patient received 100 μg kg−1 min−1. Haemodynamic measurements and blood gases were continuously monitored during the administration. After haemodynamic parameters had returned to baseline, each patient received a single pre-defined oral dose of either 5 mg (n =1), 10 mg (n = 2) or 20 mg (n = 17) vardenafil (Levitra®, Bayer Schering Pharma, Berlin, Germany) depend- ing on age and liver function. Haemodynamic measurements and blood gas analyses were carried out at baseline and 60 min after vardenafil administration. All patients, except two, had received adenosine infusions prior to vardenafil testing. A positive acute vasoreactivity response was defined in accor- dance with the consensus statement of the European Society of Cardiology [2] and American College of Chest Physicians [3] as a reduction in mPAP ≥10 mmHg and an absolute value of mPAP ≤40 mmHg, with an increased or unchanged CO. Follow-up All patients were followed-up for up to 7 years after the vasoreactivity testing in order to evaluate the long-term clini- cal outcome as judged by survival, changes in WHO function- al class, measurement of 6MWD and NT-proBNP levels. Statistics and data analysis Baseline data and haemodynamic parameters are expressed as median (interquartile range) or median (range). Bland-Altman plots were used to compare the haemodynamic responses to adenosine with vardenafil. Due to the small sample size and non-normal distribution of data, non-parametric tests were applied. The comparison of the haemodynamic effects of adenosine and vardenafil was based on the percentage change from baseline to maximal dose, and the Wilcoxon signed rank test was used for evaluation of statistical significance. The Mann–Whitney U test was used to compare the responders with non-responders. To account for multiplicity bias due to multiple testing, Bonferroni-Holm correction was performed. P values <0.05 were considered statistically significant. Sta- tistical analyses were carried out with SAS 9.3 (SAS Institute Inc, Cary, NC). Results Patient population The patient group consisted of 20 patients (13 women, 7 men) with a median age of 63.0 years (interquartile range (IQR) 32). Seven subjects had idiopathic PAH (IPAH), five associated PAH (APAH), five PH due to left heart disease, two PH due to lung disease and one chronic thromboembolic pulmonary hypertension (CTEPH) ac- cording to the latest clinical classification system [20]. Sixteen of these patients had been included in earlier pub- lished PK/PD studies [14, 19]. Based on clinical criteria, the patients were classified as WHO functional class II (n = 4) and class III (n = 16) and received at least one of the conventional standard therapies (angiotensin- converting enzyme inhibitors, angiotensin receptor blockers, β-adrenoreceptor blockers, diuretics, warfarin, digoxin or calcium channel blockers). Six patients earlier diagnosed with PAH already received treatment with PAH-specific medications at enrolment (three with bosentan, two with sildenafil and one with sildenafil in combination with epoprostenol). The remaining 14 pa- tients were newly diagnosed with PH or PAH after diag- nostic RHC. The baseline clinical characteristics and hae- modynamic data are shown in Table 1. Haemodynamic responses Changes in haemodynamic variables during adenosine and vardenafil administration are shown in Table 2. During aden- osine infusion, mPAP, PVR, SVR and mSAP were significant- ly reduced. CO, CI, pulmonary artery (PA) saturation and the arterial O2 saturation increased significantly. No statistical sig- nificant change in PVR/SVR ratio was observed for adenosine compared to baseline. The association became non-significant for mSAP, mPAP, SVR and arterial saturation after Bonferroni-Holm correction and was still significant for PVR, CO, CI and PA saturation. Sixty minutes after vardenafil administration, mPAP, PVR, SVR, PVR/SVR and mSAP significantly decreased (Table 2). A significant increase in CO and CI was observed. The asso- ciation became non-significant for PVR/SVR after Bonferroni-Holm correction; however, it was still significant for mSAP, mPAP, PVR, SVR, CO and CI. Significant differences between adenosine and vardenafil in the extent of haemodynamic responses were observed, as shown in Fig. 1. Vardenafil treatment resulted in a larger re- duction of mPAP compared to adenosine. Contrary, the impact of adenosine on CI and arterial O2 saturation was larger than for vardenafil, respectively. Bland-Altman plots were used for comparison of mPAP and CO haemodynamic response be- tween adenosine and vardenafil exposure (see Fig. 1 in the supplementary material). Positive response to acute vasodilator testing In accordance with the criteria for positive response in acute vasoreactivity testing, five out of 20 patients (25 %) were judged as responders after vardenafil exposure and three out of 18 patients (17 %) using adenosine. Three patients responded both to adenosine and vardenafil, and two addition- al patients showed a positive acute response to vardenafil (one of those patients did not get adenosine). Responders to vardenafil belonged to the following groups according to the Nice classification; APAH (n=3), IPAH (n=1) and PH due to left heart disease (n=1). Responders versus non-responders A comparison between clinical characteristics and baseline haemodynamics among responders and non-responders is given in Table 1. There was a statistical significant difference for median NT-proBNP, S-creatinine and eGFR. Safety and tolerability Single oral dose vardenafil was well tolerated, and no adverse reactions were observed in this study. Adenosine infusion caused headache and flush, well-known side effects, and only 6 out of 18 patients tolerated the maximum dosage (210 μg kg−1 min−1). Long-term follow-up and survival Irrespective of the outcome of vardenafil acute vasoreactivity testing, all patients received standard PAH-specific therapy according to the outcome of RHC. The five responders to vardenafil received the following treatment after RHC: silden- afil and CCB (n=2), sildenafil and bosentan (n=1) and CCB (n=1) (supplementary Table 1). The patient diagnosed with PH due to left heart disease was not eligible to receive PAH treatment. One of the responders was lost to follow-up, and another committed suicide (Fig. 2). Still, survival data of these patients were included in the analysis. Data of non-responders to PAH-specific treatment after RHC are shown in supplementary Table 1. Survival data for all patients are shown in Fig. 2. All patients, both responders and non-re- sponders, were re-evaluated for WHO functional class, 6MWD and NT-proBNP at years 1, 3, 5 and 7 after vasoreactivity testing (Table 3). Marked differences between responders and non-responders were seen; however, the num- ber of observations was too small to allow for significance testing. Discussion This study is the first to investigate the diagnostic utility of oral vardenafil in acute vasoreactivity testing. Acute effects on haemodynamic parameters after a single oral dose of vardenafil were similar to those observed after iv adenosine at maximum tolerated dose. Furthermore, more positive re- sponders were identified with vardenafil than with adenosine testing. During a follow-up time of up to 7 years, vardenafil responders had significantly lower NT-proBNP levels and in general a relatively higher proportion of survivors compared to non-responders. Both vardenafil and adenosine significantly decreased mSAP, mPAP, PVR and SVR. Further, vardenafil, but not adenosine, significantly reduced the PVR/SVR ratio. A de- crease in the PVR/SVR ratio is considered a favourable hae- modynamic outcome parameter of vasodilator treatment [21]. This is in agreement with our earlier published study of 16 patients where vardenafil decreased PVR/SVR by 16.9 % [14]. In the present study, both vardenafil and adenosine sig- nificantly increased CO and CI. However, opposed to vardenafil, adenosine also significantly increased pulmonary and systemic oxygen saturation. Similar results were found in the above mentioned study of 16 vardenafil-treated PAH pa- tients, where arterial oxygenation remained unchanged during RHC [9]. According to literature, only 10–15 % of IPAH patients are expected to be acute responders to short-acting vasodilating drugs such as adenosine [2]. In a register study from the USA enrolling 578 patients with PAH, only 4.5 % were acute re- sponders to adenosine [22]. This response rate was consider- ably lower than observed in the present study where 17 % responded to adenosine. However, despite the high response rate in our study, only few patients tolerated the maximum recommended test doses of adenosine. Normally, the sug- ge sted s t arting dose o f a deno sine infusio n is 50 μg kg−1 min−1, which is increased every 2 min to a max- imum dose of 350–500 μg kg−1 min−1. Only six out of 18 patients tolerated the targeted dose of 210 μg kg−1 min−1 with- out experiencing adverse effects. In the literature, doses of 200 μg kg−1 min−1 are considered sufficient to detect acute positive responders in patients with PH and congestive heart failure [23–25]. The majority of our patients only tolerated adenosine doses only up to 140 μg kg−1 min−1. Moreover, two patients in our study refused taking adenosine due to earlier negative experience. The lower than recommended maximum doses might have compromised the diagnostic util- ity of adenosine testing. Indeed, the three adenosine re- sponders in our study received doses of ≥140 μg kg−1 min−1. In our study, five out of 20 patients (25 %) were responders to vardenafil. Only one of these patients had received PAH treatment with sildenafil earlier. In addition, all patients show- ing an acute positive response to adenosine also responded to vardenafil, but two additional responders could be detected with vardenafil. The clinical and statistical significance of the higher response rate to vardenafil remains unclear and could be a chance finding due to the limited number of obser- vations. On the other hand, the potential of vardenafil to block calcium fluxes in addition to PDE5 inhibition could represent a possible explanation why additional subjects who might benefit from CCB treatment could be identified through vardenafil testing [16, 17]. An additional advantage of vardenafil is lower incidence of adverse effects compared to adenosine. Therefore, vardenafil, especially if available as iv preparation, would appear an attractive alternative for acute vasodilator testing during RHC. Further, this approach would allow continuing with the same vasodilator for long-term treatment of PAH after RHC. The general usefulness of acute vasoreactivity testing has been questioned [2]. At present, however, there are no feasible alternative clinical or diagnostic parameters at hand that could replace invasive vasoreactivity testing. In our study, we inves- tigated potential biochemical parameters for non-invasive di- agnostic tests. Responders had statistically significant lower baseline values of NT-proBNP, S-creatinine and eGFR as compared to non-responders. Among the younger patients, lower NT-proBNP values and higher eGFR could indicate less severe disease. Earlier observations also demonstrated that younger subjects in general are more likely to show positive acute vasodilator response [23]. A strong correlation between NT-proBNP, stage of functional impairment and disease pro- gression in PAH patients has been indicated in earlier studies [26, 27]. Repeated measurements of NT-proBNP could also be helpful for follow-up of disease progression in patients with PH [28]. Other novel biochemical markers are asymmet- ric dimethylarginine (ADMA), symmetric dimethylarginine (SDMA) and arginine. In our material, their concentrations tended to be lower, while arginine concentrations were higher in responders compared to non-responders. Elevated ADMA and SDMA levels can inhibit NO production and thus endothelium-dependent vasodilatation. Changes in arginine and arginine/ADMA ratio have recently been associated with improved CI [29]. Increased ADMA has been reported in patients with IPAH [30], systemic sclerosis (SSc)-related PAH [31] and CTEPH [32]. In our study, the number of observations was too limited to allow a definite answer to the question whether vardenafil re- sponders have a better survival prognosis compared to non- responders. Among the five vardenafil responders, one patient unfortunately committed suicide during follow-up. The 15 non- responders showed significantly higher NT-proBNP values compared to responders at baseline and after 3 years of fol- low-up (Table 3). It has been shown in an earlier study that NT-proBNP levels above 1256 ng L−1 predict poorer outcome in patients with PAH [28]. In our material, none of the re- sponders had NT-proBNP values above 1256 ng L−1 at baseline and after 1 year of follow-up, while among non-responders, median NT-proBNP was above this cut-off (Table 3). The present study has several limitations. The number of observations was limited and the study population inhomoge- neous. It would have been desirable to enrol more patients suf- fering from the same type of pulmonary hypertension to allow studying more homogenous groups. However, PH is a rare dis- ease with poor prognosis making it difficult to recruit larger numbers of patients within a reasonable time frame. A complete set of haemodynamic parameters was not available in all patients due to logistic reasons. All patients, except two, had received adenosine prior to vardenafil, and this could theoretically have had impact on vardenafil response. However, Bpriming effects^ of adenosine are unlikely due to the short plasma half-life less than of 10 s and an expected duration of pharmacodynamic effects of maximum 2 min [2, 3]. Further, we made sure that haemodynamic parameters had returned to baseline values be- fore vardenafil was administered. Still, conditioning effects of pre-treatment with adenosine cannot be totally excluded. The response to vardenafil could be monitored first after 60 min. From a pharmacokinetic perspective, iv administration of vardenafil would have been preferable due to faster haemody- namic response. Moreover, oral vardenafil showed variable and in some subjects delayed absorption. However, iv vardenafil formulations were not available at the time of the study. Current- ly, adenosine is no longer first choice for acute vasodilator test- ing due to its limited pulmonary selectivity and has been substituted by inhaled NO. However, at the time when the pres- ent study was performed, adenosine was still the standard vaso- dilator agent used at the heart catheterisation centre of the Insti- tution of Cardiology, Uppsala University Hospital, Sweden. In conclusion, vardenafil may be safely used as a vasodilatory agent to identify responders among patients with PH. Acute haemodynamic responses to oral vardenafil are comparable to iv adenosine, but a single oral dose of vardenafil resulted in greater reductions of mPAP and caused fewer adverse effects. Further studies in larger and more ho- mogenous patient populations are warranted to further study the utility of vardenafil in acute vasodilator testing and for prediction of long-term clinical outcome in PH patients.