Abstract
In patients undergoing right heart catheterisation, the timing of haemodynamic measurement after sheath insertion (immediately or after a short resting period) influences the diagnosis, classification and risk stratification of pulmonary hypertension https://bit.ly/2yHN9xq
To the Editor:
Haemodynamic measurements from right heart catheterisation (RHC) are used for diagnosis and risk stratification in pulmonary hypertension (PH) [1]. The risk stratification scheme implemented in the guidelines for pulmonary arterial hypertension (PAH) has been validated in real-life cohorts [1–5]. However, the timing of measurements during RHC is not specified. We aimed to investigate the influence of the timing of measurements on RHC parameters and the consequences for risk stratification and diagnosis.
We retrospectively investigated patients in the Giessen PH Registry [6]. Incident patients referred for diagnostic RHC between February 2010 and August 2017 were included, with most of the relevant RHC measurements (listed below) available at two consecutive time points: immediately after placing the sheath correctly (baseline-1) and after the patient had rested for a short period (baseline-2). Some patients had further RHC measurements after a second resting period (baseline-3). To exclude methodological issues, zero levelling was re-evaluated for all baselines and, if necessary, corrected [7–9]. However, in most cases the position of the pressure transducer was not changed. RHC was performed by experts who were not blinded to the clinical data or previous measurements. Almost all procedures were performed using the right jugular vein with ultrasound and local anaesthesia. All patients were diagnosed according to the current guidelines [1] using RHC measurements obtained at baseline-2. All diagnoses of PH were assessed by a multidisciplinary board. Patients were followed until December 2018 for survival analysis. The investigation conforms with the Declaration of Helsinki and was approved by the ethics committee of the Faculty of Medicine at the University of Giessen. All participating patients gave written informed consent to be enrolled in the Giessen PH Registry.
RHC measurements included right atrial pressure (RAP), cardiac output (CO, direct Fick method or thermodilution if direct Fick was not available), cardiac index (CO/body surface area), mean pulmonary arterial pressure (mPAP), pulmonary artery wedge pressure (PAWP), pulmonary vascular resistance (PVR) and mixed venous oxygen saturation (SvO2). All vascular pressures were measured at end-expiratory breath hold for 5–7 s. Risk stratification was based on cardiac index, SvO2 and RAP, to calculate a cumulative risk score [2, 3]. Statistical analyses were performed using R (The R Foundation, Vienna, Austria) and SPSS (Version 26, IBM, Armonk, USA).
In total, 1093 patients were included (median age 72.5 years, interquartile range 60.8–84.2 years; pulmonary venous hypertension (PVH), n=341; PH due to lung diseases (LD-PH), n=253; chronic thromboembolic PH (CTEPH), n=132; PAH, n=114; miscellaneous PH, n=44; PH excluded by RHC, n=209). The mean±sd duration between baseline-1 and the start of baseline-2 measurements was 21±9 min. 66 patients had baseline-3 values (mean duration between the start of baseline-2 and baseline-3 measurements: 25±17 min).
The difference in mPAP between baseline-1 and baseline-2 was 1.17±3.40 mmHg, while the difference in PAWP was 0.47±2.85 mmHg (both p<0.001). Diagnosis or exclusion of PH changed between baseline-1 and baseline-2 in 61 patients (5.6%). Of 773 patients with mPAP ≥25 mmHg at baseline-1, 51 (6.6%) had mPAP <25 mmHg at baseline-2. Of 320 patients with mPAP <25 mmHg at baseline-1, 10 (3.1%) had mPAP ≥25 mmHg at baseline-2 (PVH, n=3; LD-PH, n=5; CTEPH, n=1; miscellaneous PH, n=1). Moreover, the categorisation of PH as pre- or post-capillary was influenced by the waiting period. Of 223 patients classified as having post-capillary PH (PAWP >15 mmHg [1, 10]) at baseline-1, 45 patients (20.2%) were classified with pre-capillary PH at baseline-2. Conversely, 21 of 596 patients (3.5%) with pre-capillary PH according to baseline-1 values were classified with post-capillary PH at baseline-2. Between baseline-2 and baseline-3, no patient changed between pre- and post-capillary PH.
The cumulative risk score changed between baseline-1 and baseline-2 in 154 patients (14.1%; PVH, n=54; LD-PH, n=28; CTEPH, n=25; PAH, n=20; miscellaneous PH, n=10; PH excluded by RHC, n=17). Using baseline-2 instead of baseline-1 data resulted in the reclassification of 6.1% of low-risk patients (n=636) as intermediate-risk, 3.8% and 12.6% of intermediate-risk patients (n=365) as high-risk and low-risk, respectively, and 20.7% of high-risk patients (n=92) as intermediate-risk. Overall, 9.4%, 15.6% and 16.1% of the patients changed their risk class based on RAP, cardiac index and SvO2, respectively. Regarding RAP, 23.3% of high-risk patients (n=73) and 21.7% of intermediate-risk patients (n=295) improved their risk class, whereas 2.2% and 1.4% of low- (n=723) and intermediate-risk patients, respectively, worsened. Regarding cardiac index, 13.7% and 16.3% of high- (n=182) and intermediate-risk patients (n=300) improved whereas 11.0% and 8.0% of low- (n=608) and intermediate-risk patients worsened. Regarding SvO2, 20.7% and 12.6% of high- (n=202) and intermediate-risk patients (n=192) improved, whereas 6.1% and 3.8% of low- (n=695) and intermediate-risk patients worsened. In patients with PAH, the cumulative risk class showed higher prognostic power at baseline-2 than at baseline-1; only the cumulative risk class at baseline-2 significantly predicted mortality in intermediate-risk patients (figure 1a). In patients who changed their risk class, baseline-2 values predicted 5-year survival more accurately than baseline-1 values (area under the receiver operating characteristic curve 0.722 (95% CI 0.514–0.930) versus 0.624 (95% CI 0.374–0.873)). The predictive values of RAP, cardiac index and SvO2 are shown in figure 1a.
Bland–Altman plots showed no meaningful biases but wide limits of agreement between baseline-1 and baseline-2 (figure 1b–e). The precision of PAWP, cardiac index and SvO2 improved between baseline-2 and baseline-3 (figure 1f), with no directionality in risk class changes (three patients improved and three worsened).
Our results indicate that haemodynamic measurements taken following a short resting period differ from those taken immediately after sheath insertion, altering diagnosis/exclusion of PH and categorisation of PH as pre- or post-capillary in subsets of patients and influencing risk stratification in PAH. Although Bland–Altman analysis revealed imprecision at each timepoint, these changes could be clinically meaningful in individuals. Baseline-2 measurements showed higher prognostic and discriminatory power than baseline-1. Of note, only arbitrary changes were observed between baseline-2 and baseline-3.
Spontaneous variability of RHC measurements was described decades ago [11, 12]. Rich et al. [11] assessed 12 patients hourly for six consecutive hours and observed spontaneous haemodynamic variability, analogous to our results, concluding that particularly mPAP, PVR and CO are time-dependent. Furthermore, variability of PAWP is a well-known reason for misclassification of PH [1, 13]. Our results showed a predominant direction of change between baseline-1 and baseline-2 in addition to the previously described intrinsic variability. Time-dependent differences of variability may be one possible explanation for these changes. However, due to imprecision at each time point, the usefulness of multiple measurements (as suggested in the guidelines [1]) should be questioned.
Our results suggest that defining a waiting period after sheath placement with consideration of the clinical context may help to standardise diagnosis, risk stratification and treatment decisions. However, our study is limited by its retrospective design, small number of patients with baseline-3 data and lack of blinding. Prospective studies are needed to identify the optimum time point of measurement. Nevertheless, we believe that allowing a short resting period between sheath placement and haemodynamic assessment in defined environmental conditions leads to an enhanced stratification of haemodynamic function and risk in patients with PH. In practice, this could be implemented by remeasuring mPAP, RAP and PAWP after measuring CO. We conclude that the timing of haemodynamic measurements after RHC sheath placement is of major importance and should be clarified in future guidelines.
Shareable PDF
Supplementary Material
This one-page PDF can be shared freely online.
Shareable PDF ERJ-01892-2019.Shareable
Acknowledgements
Editorial assistance was provided by Claire Mulligan (Beacon Medical Communications Ltd, Brighton, UK), funded by the University of Giessen.
Footnotes
Conflict of interest: A. Yogeswaran reports editorial support from University of Giessen, during the conduct of the study.
Conflict of interest: M.J. Richter reports grants from German Research Foundation, editorial support from University of Giessen, during the conduct of the study; grants from United Therapeutics, grants and personal fees for lectures and consultancy from Bayer, personal fees for lectures from Actelion, Mundipharma, Roche and OMT, outside the submitted work.
Conflict of interest: N. Sommer reports personal fees from Actelion, outside the submitted work.
Conflict of interest: H.A. Ghofrani reports grants from German Research Foundation, editorial support from University of Giessen, during the conduct of the study; personal fees for consultancy and advisory board work from Bayer and Pfizer, personal fees for consultancy, lectures and advisory board work from Actelion and GSK, personal fees for consultancy from Merck, grants and personal fees for consultancy from Novartis, grants and personal fees for lectures from Bayer HealthCare and Encysive/Pfizer, grants from Aires, German Research Foundation, Excellence Cluster Cardiopulmonary Research and German Ministry for Education and Research, personal fees for advisory board work from Takeda, outside the submitted work.
Conflict of interest: W. Seeger reports grants from German Research Foundation, editorial support from University of Giessen, during the conduct of the study; personal fees for lectures and consultancy from Pfizer and Bayer Pharma AG, outside the submitted work.
Conflict of interest: H. Gall reports grants from German Research Foundation, editorial support from University of Giessen, during the conduct of the study; personal fees from Actelion, AstraZeneca, Bayer, BMS, GSK, Janssen-Cilag, Lilly, MSD, Novartis, OMT, Pfizer and United Therapeutics, outside the submitted work.
Conflict of interest: K. Tello reports grants from German Research Foundation, editorial support from University of Giessen, during the conduct of the study; personal fees for lectures from Actelion and Bayer, outside the submitted work.
Support statement: This work was funded by the Excellence Cluster Cardio-Pulmonary System (ECCPS) and the Collaborative Research Center (SFB) 1213 - Pulmonary Hypertension and Cor Pulmonale, grant number SFB1213/1, project B08 (German Research Foundation, Bonn, Germany). Funding information for this article has been deposited with the Crossref Funder Registry.
- Received September 23, 2019.
- Accepted April 3, 2020.
- Copyright ©ERS 2020