Pressure record analysis method parameters during weaning success and failure from mechanical ventilation in prolonged ventilated patients after cardiac surgery

Introduction

Hemodynamic compromise remains the principal cause of adverse outcomes among patients undergoing cardiac surgery (1,2), yet accurate assessment of hemodynamic status continues to present a major clinical challenge. Although biomarkers such as mixed venous oxygen saturation (SvO₂) and lactate have been associated with patient prognosis, their inability to provide continuous monitoring limits their clinical utility. Continuous hemodynamic monitoring modalities, such as pulse indicator continuous cardiac output (PiCCO), are available; however, their invasive nature and high cost restrict widespread application. Recently, energy-based variables have emerged as promising early indicators of hemodynamic compromise and have shown potential utility as prognostic markers (3,4).

A variety of techniques have been established to quantify cardiac efficiency. Among these, echocardiography is widely regarded as a pivotal tool for assessing cardiac performance, primarily via the measurement of ventricular-arterial coupling in clinical practice (5). Nonetheless, echocardiography, despite its numerous advantages, is constrained by dependence on operator skill and limited to intermittent monitoring, particularly in complex clinical scenarios.

To address these limitations, cardiac cycle efficiency (CCE) has been introduced as a novel, minimally invasive parameter derived from pulse contour analysis using the pressure recording analytical method (PRAM). Although the clinical utility of PRAM remains a subject of ongoing debate, it has gained considerable acceptance for the measurement of cardiovascular function in various clinical settings. The high signal sampling rate of 1,000 Hz afforded by PRAM enables the detection of even subtle changes in arterial pressure waveforms; however, this increased sensitivity may also raise the potential for artefactual signal capture.

The present study aimed to investigate the trends in PRAM-derived parameters during both successful and unsuccessful weaning from mechanical ventilation in patients following cardiac surgery. We present this article in accordance with the TRIPOD reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1194/rc).

Methods

The study was prospectively registered in ClinicalTrials.gov with the registration number NCT06230497 on January 27, 2024.

Patient enrollment

A prospective observational cohort study was conducted in a 30-bed adult intensive care unit (ICU) from February 2024 to July 2024. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Research and Ethics Committee of Peking Union Medical College Hospital (No. K24C0570). Written informed consent was provided by the next of kin of all enrolled patients.

The inclusion criteria were: (I) age between 18 and 80 years; (II) admission to ICU following cardiac surgery; (III) mechanical ventilation for >24 hours and considered eligible for a spontaneous breathing trial (SBT) by the ICU physician; (IV) ScvO₂ ≥60%; (V) oxygenation index >250 mmHg; and (VI) norepinephrine <0.2 µg/kg/min.

Patients were excluded if they presented pathological conditions potentially interfering with arterial pulse transmission (e.g., aortic valve disorders, aortic aneurysm or dissection, and thoracic outlet syndrome).

Study protocol

To minimize potential bias, clinicians performing SBT were blinded to PRAM parameters. MostCare monitor displays were concealed, with only conventional vital signs accessible for clinical decision-making. During data analysis, group allocation was masked through coded labels (Group A/B), which were revealed only after completion of all statistical analyses.

All patients received ventilation via the SV800 Ventilator (Mindray, Shenzhen, China). In accordance with clinical protocol, sedatives were discontinued prior to SBT, and patients’ responsiveness was confirmed.

Baseline respiratory data were collected during pressure support ventilation (support pressure 8 cmH₂O, positive end-expiratory pressure 5 cmH₂O) for 10 minutes. One minute after baseline data collection, SBT was conducted using a T-tube and supplemental oxygen, lasting up to 60 minutes. The duration of SBT was recorded in cases of failure.

SBT failure was defined by the occurrence of any of the following: respiratory rate >35 breaths/min, SpO₂ <90%, heart rate (HR) >140 bpm, systolic blood pressure >180 or <90 mmHg, or clinical signs of agitation, diaphoresis, or anxiety. Only patients who completed the 60-minute SBT without failure were extubated. Weaning failure was defined as unsuccessful SBT or the need for reintubation within 48 hours post-extubation.

Given the variability in SBT duration, data were analyzed at eight predefined time points: two minutes prior to SBT initiation (PSV), the first and last minute of SBT (1-min, last-min), and five equally distributed intervals (T1–T5) between these endpoints (Figure 1). This approach was adopted to ensure a comprehensive and rigorous analysis.

Figure 1 Protocol of the study. MV, mechanical ventilation; PSV, pressure support ventilation; SBT, spontaneous breathing trial.

Measurement

PRAM parameters

The methodology and configuration of PRAM have been described previously (6-8). In this study, a MostCare transducer (Vygon, Padova, Italy) was connected to an indwelling arterial catheter for data acquisition.

Fundamental arterial blood pressure parameters were sampled at a frequency of 1,000 Hz, providing high-fidelity data. Measurement parameters for each variable were collected and stored at 30-second intervals using dedicated software, then downloaded to spreadsheets for offline analysis.

CCE

Two previous studies have described the design and setup of CCE measurements using MostCare (3,7). CCE represents the ratio of energy consumed by the cardiovascular system during cardiac systole to that consumed throughout the entire cardiac cycle. It serves as a reflection of the energy efficiency of the entire circulatory system, including the heart, arterial system, venous system, and pulmonary circulation. CCE enables the estimation of energy expenditure required by the body to maintain hemodynamic stability under various conditions. This indicator signifies the heart’s effectiveness in maintaining hemodynamic balance under physiological or pathophysiological conditions. Consequently, CCE emerges as an independent and objective parameter unaffected by operator bias, expressing the cardiovascular system’s capacity to sustain dynamic equilibrium across different energy levels through the synchronized interplay of pumping function, arterial system, venous return, and pulmonary circulation (3,9). Hemodynamic readings were excluded if CCE fluctuated by more than 0.10 during the minute preceding data collection.

The maximal slope of systolic upstroke (dp/dt)

The positive and negative values of dp/dt represent the maximal rates of change in the ascending and descending phases of left ventricular pressure, respectively, reflecting ventricular contractility and systolic function. Previous studies have demonstrated good agreement between dp/dt values measured by echocardiography and MostCare.

Statistical analysis

The sample size was calculated to detect a significant difference in the CCE between patients who succeeded and those who failed weaning from mechanical ventilation. Assuming a two-sided significance level (alpha) of 0.05 and a statistical power of 80%, the calculation was based on a two-independent-sample t-test. According to preliminary data and relevant literature, the expected mean difference in CCE between groups was 0.1, with an estimated standard deviation of 0.12. Considering a weaning success to failure ratio of 2:1, the required sample size was determined to be 36 patients in the success group and 18 patients in the failure group, for a total of no less than 54 patients.

Statistical analyses were conducted using IBM SPSS Statistics 25.0 (IBM Analytics, USA), and data visualization was performed using Prism 10 (GraphPad Software, La Jolla, CA, USA). Continuous variables were expressed as mean ± standard deviation or median (interquartile range), as appropriate. Normality was assessed with the Kolmogorov-Smirnov test. Group comparisons were performed using independent-samples t-tests, Mann-Whitney U tests, or Chi-squared tests, as appropriate.

Changes in PRAM variables during SBT were evaluated using generalized estimating equations (GEE) for repeated measures, assessing main effects of time (P_time), group (P_group), and their interaction (P_{time × group}). When significant effects were detected, post hoc pairwise comparisons were performed using paired t-tests with Bonferroni correction. The discriminative ability of CCE was assessed via receiver operating characteristic (ROC) analysis. All statistical tests were two-tailed and a P<0.05 was considered statistically significant.

Results

Demographic and clinical characteristics

A total of 58 post-cardiac surgery patients were enrolled during the study period, with a mean age of 58±13 years. Surgical procedures included valve surgery (63.8%, n=37), coronary artery bypass grafting (CABG) (22.4%, n=13), and combined CABG plus valve surgery (13.8%, n=8). The mean cardiopulmonary bypass (CPB) time was 102±17 minutes. Of the enrolled patients, 43 were successfully weaned and extubated (Success group), whereas 15 experienced weaning failure (Failure group). Patient demographic and clinical characteristics are summarized in Table 1.

PRAM parameters

A significant interaction between group and time was observed for CCE (P<0.001), warranting separate analyses for each group. In the success group, temporal differences in CCE were not significant, whereas in the failure group, a significant decline was noted (P=0.007), with CCE decreasing from PSV to the last minute of SBT (−0.011±0.170 to −0.118±0.226, P<0.05 with Bonferroni correction). Significant between-group differences in CCE were identified at specific time points, with more pronounced trends in the failure group. Similarly, a significant interaction was found for dp/dt (P=0.02; Table 2, Figure 2).

Figure 2 Pressure recording analytical method parameters during spontaneous breathing trial. *, P<0.05. CCE, cardiac contractility efficiency; CI, cardiac index; dp/dt, change in pressure over change in time; HR, heart rate; MAP, mean arterial pressure; SVI, stroke volume index.

Analysis revealed no significant effect of time on HR (P_time =0.65), but a significant between-group difference (P_group =0.005). A significant interaction between time and group (P_{time × group} <0.001), indicated differential effects of time on HR across groups. For mean arterial pressure (MAP), a significant time-group interaction (P_time×group <0.001) was observed, with time exerting a significant effect (P_time =0.009), but no significant group difference (P_group =0.26) (Table 2 and Figure 2).

The area under the curve (AUC) in predicting weaning success using the CCE_PSV was 0.6752 (95% CI: 0.5345–0.8159, P=0.045). At a threshold of CCE_PSV <0.1145, the sensitivity was 80.00%, the specificity was 58.14%, and likelihood ratio was 1.911 (Figure 3).

Figure 3 ROC of CCE value before spontaneous breathing trial to predict success weaning. AUC, area under the curve; CCE, cardiac cycle efficiency; ROC, receiver operating characteristic curves.

Patients were stratified into low-CCE (CCE_PSV ≤25th percentile) and high-CCE (CCE_PSV ≥75th percentile) groups. Baseline characteristics, including age, body mass index (BMI), European System for Cardiac Operative Risk Evaluation II (EuroSCORE II), pre-operative left ventricular ejection fraction (pre-LVEF), duration of mechanical ventilation at weaning, and type of surgery, were comparable between groups. The High-CCE group had a 0% failure weaning rate, while the Low-CCE group had a 40% success rate (P=0.02). The Low-CCE group had a longer ICU stay (median 6, IQR 3–7 days) compared to the High-CCE group (median 4, IQR 3–5 days), P=0.02 (Table 3).

Discussion

This study sought to evaluate the application of PRAM monitoring during SBT in patients with prolonged weaning after cardiac surgery, focusing on CCE and dp/dt as key parameters. Our findings indicate that CCE and dp/dt exhibit distinct trends between patients who successfully weaned and those who did not.

The temporal change in CCE differed significantly between the success and failure groups. Stable CCE values were observed in patients who successfully weaned, whereas a significant decline was noted in the failure group. This divergence was confirmed by a significant group × time interaction in the GEE model (P<0.001), and the within-failure-group decline persisted after Bonferroni correction (P=0.007).

Similarly, dp/dt demonstrated significant temporal changes between groups, highlighting its potential as an ancillary marker for identifying patients at risk of weaning failure.

CCE as a dynamic, integrative hemodynamic marker

CCE is an unconventional indicator that encapsulates comprehensive hemodynamic function, reflecting the complex interplay among cardiac, arterial, and pulmonary compartments.

A previous study has demonstrated an association between compromised ventricular-arterial coupling and reduced CCE, particularly in pediatric patients following congenital heart surgery (10). Additionally, Han et al. (3) have discovered that a reduction in CCE correlated with early adverse outcomes in pediatric cardiac surgery patients and is a potentially influential predictor of postoperative outcomes in adults undergoing cardiovascular procedures (11-13).

The present study’s ROC analysis showed that CCE_PSV had moderate predictive accuracy (AUC =0.6752), with high sensitivity (80%) but moderate specificity (58.14%). These findings indicate that while CCE_PSV may be valuable for identifying patients likely to wean successfully, its predictive utility may be enhanced when combined with other clinical parameters.

Notably, there were important baseline imbalances between the success and failure groups. The failure group had a lower pre-LVEF (45%±10% vs. 52%±8%, P=0.008) and a longer CPB time (111±16 vs. 99±16 min, P=0.01). The prevalence of COPD was also higher in the failure cohort, although this difference did not reach statistical significance in subgroup comparisons (P=0.25). These observations align with ICU Physician experience and prior epidemiologic data showing that patients with poorer baseline cardiopulmonary reserve are at greater risk of extubation failure. Crucially, however, our analysis prioritized the dynamic behavior of CCE during the SBT rather than reliance on a single baseline measurement. The significant group × time interaction identified by the GEE indicates that CCE trajectories during weaning capture real-time pathophysiologic cardiac responses that are not apparent from baseline values alone.

Group comparisons based on CCE_PSV

Baseline characteristics of patients in low-CCE and high-CCE groups were similar, ensuring that observed differences in outcomes could be attributed to CCE_PSV differences rather than baseline disparities.

The low-CCE_PSV cohort experienced a significantly higher rate of weaning failure and a longer ICU stay compared with the high-CCE_PSV group (median ICU length of stay, P=0.02). These findings imply that patients with lower CCE_PSV may require more prolonged intensive care and face a greater risk of failed extubation. However, this hypothesis requires confirmation in larger, prospective studies that can adequately control for potential confounders.

Clinical implications

Predicting weaning failure in patients with delayed weaning after cardiac surgery is challenging due to their complex and variable physiological status (14,15). During the weaning process, multiple organ systems interact, and changes in one system may unpredictably affect others. Under these circumstances, management must go beyond identification of obvious respiratory causes: clinicians need to detect the underlying drivers of weaning difficulty quickly and institute targeted interventions without delay.

In addition to standard bedside parameters such as HR, blood pressure, respiratory rate and SpO₂, modern continuous hemodynamic monitoring provides dynamic indices that may reveal decompensation earlier than routine measures. Of these, cardiac energy metrics, exemplified by CCE, deserve particular attention, since shifts in CCE often precede changes in conventional physiologic variables. A falling or negative CCE trajectory during a SBT can therefore serve as an early warning signal, prompting investigation of the precipitating haemodynamic disturbance and timely corrective measures to avert weaning failure.

If these observations are corroborated, bedside CCE monitoring during SBT could help identify patients whose cardiopulmonary reserve fails under the stress of spontaneous breathing. A decline in CCE might justify postponing weaning, prioritizing optimization of cardiovascular and respiratory support, and then reattempting weaning once stability is restored. Such a tailored approach has the potential to reduce premature extubation and downstream complications in high-risk cardiac surgery patients.

To establish clinical utility, future research should consist of adequately powered, multicenter prospective studies with multivariable adjustment for baseline differences, direct comparisons of CCE against established predictors, and ultimately randomized controlled trials that assess whether CCE-guided weaning strategies improve patient-centered outcomes relative to standard care.

Limitations

There are some limitations in this study: (I) the sample size of this study was small, particularly in the failure group. The study was conducted at a single center, which may limit the generalizability of the results. Future studies should include larger, multicenter cohorts to validate these findings. (II) The potential limitation of PRAM’s reliability in patients may stem from factors such as the positioning of the arterial catheter and abnormal signal conditions arising from the arterial transducer, including both over- and underdamping. Prior to data acquisition, we implemented a rapid flush test to discern the presence of signal artifacts. Furthermore, while stringent measures were taken to mitigate disruptions from tubing and transducers, there is a likelihood that PRAM’s sensitivity to even subtle signal fluctuations might inadvertently introduce errors through spurious signal analysis. Many studies have authenticated PRAM against the esteemed Fick methodology in patients undergoing cardiac catheterization, revealing a substantial level of concurrence in cardiac output assessments between the two techniques (3,6,11). Importantly, PRAM’s capability to monitor trends in hemodynamic variables and its intergroup comparisons retains significance, regardless of absolute measurement precision. (III) While both CCE and dp/dt exhibited significant interaction effects, weaning outcomes are multifactorial, and this study did not construct a multivariable predictive model. Future work should integrate CCE with other established predictors to facilitate more comprehensive risk stratification and guide clinical decision-making.

Conclusions

This study demonstrates that CCE and dp/dt follow distinct trajectories during successful and failed SBTs in prolonged ventilated patients after cardiac surgery. CCE values obtained during SBT, together with their dynamic decline, may hold promise as indicators of weaning outcome; however, these observations are preliminary. Well-powered, prospective studies with multivariable adjustment are warranted to establish whether real-time CCE monitoring can reliably guide weaning and extubation decisions.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1194/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1194/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1194/prf

Funding: This work was supported by the National High-Level Hospital Clinical Research Funding (No. 2022-PUMCH-B-115), the National High-Level Hospital Clinical Research Funding (No. 2022-PUMCH-A-216).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1194/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Research and Ethics Committee of Peking Union Medical College Hospital (No. K24C0570). Written informed consent was provided by the next of kin of all enrolled patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sankar A, Rotstein AJ, Teja B, et al. Prolonged mechanical ventilation after cardiac surgery: substudy of the Transfusion Requirements in Cardiac Surgery III trial. Can J Anaesth 2022;69:1493-506. [Crossref] [PubMed]
2. Fernandez-Zamora MD, Gordillo-Brenes A, Banderas-Bravo E, et al. Prolonged Mechanical Ventilation as a Predictor of Mortality After Cardiac Surgery. Respir Care 2018;63:550-7. [Crossref] [PubMed]
3. Han D, Pan S, Li H, et al. Prognostic value of cardiac cycle efficiency in children undergoing cardiac surgery: a prospective observational study. Br J Anaesth 2020;125:321-9. [Crossref] [PubMed]
4. Thomsen KK, Kouz K, Saugel B. Pulse wave analysis: basic concepts and clinical application in intensive care medicine. Curr Opin Crit Care 2023;29:215-22. [Crossref] [PubMed]
5. Tuinman PR, Shi Z, Heunks L. How we use ultrasound in the management of weaning from mechanical ventilation. Intensive Care Med 2025;51:393-6. [Crossref] [PubMed]
6. Greiwe G, Luehsen K, Hapfelmeier A, et al. Cardiac output estimation by pulse wave analysis using the pressure recording analytical method and intermittent pulmonary artery thermodilution: A method comparison study after off-pump coronary artery bypass surgery. Eur J Anaesthesiol 2020;37:920-5. [Crossref] [PubMed]
7. Xie F, Jin M, Ma T, et al. Effect of high-flow nasal oxygen therapy on cardiac cycle efficiency in coronary artery disease patients undergoing endoscopy: study protocol for randomised controlled trial. BMJ Open 2025;15:e092321. [Crossref] [PubMed]
8. Seker M, Aktas Yildirim S, Ulugol H, et al. Cardiovascular Effects of Tourniquet Application with Cardiac Cycle Efficiency: A Prospective Observational Study. J Clin Med 2024;13:2745. [Crossref] [PubMed]
9. Kılınç E, Yildirim SA, Ulugöl H, et al. The evaluation of cardiac functions in deep Trendelenburg position during robotic-assisted laparoscopic prostatectomy. Front Med (Lausanne) 2023;10:1273180. [Crossref] [PubMed]
10. Marinari E, Rizza A, Iacobelli R, et al. Ventricular-Arterial Coupling in Children and Infants With Congenital Heart Disease After Cardiopulmonary Bypass Surgery: Observational Study. Pediatr Crit Care Med 2019;20:753-8. [Crossref] [PubMed]
11. Saugel B, Kouz K, Scheeren TWL, et al. Cardiac output estimation using pulse wave analysis-physiology, algorithms, and technologies: a narrative review. Br J Anaesth 2021;126:67-76. [Crossref] [PubMed]
12. Li M, Wang S, Zhang H, et al. The predictive value of pressure recording analytical method for the duration of mechanical ventilation in children undergoing cardiac surgery with an XGBoost-based machine learning model. Front Cardiovasc Med 2022;9:1036340. [Crossref] [PubMed]
13. Greiwe G, Balfanz V, Hapfelmeier A, et al. Pulse Wave Analysis Using the Pressure Recording Analytical Method to Measure Cardiac Output in Pediatric Cardiac Surgery Patients: A Method Comparison Study Using Transesophageal Doppler Echocardiography as Reference Method. Anesth Analg 2022;135:71-8. [Crossref] [PubMed]
14. Zhou Y, He H, Cui N, et al. Elevated pulsatility index of the superior mesenteric artery indicated prolonged mechanical ventilation in patients after cardiac valve surgery. Front Surg 2022;9:1049753. [Crossref] [PubMed]
15. Zhou Y, He H, Wang X, et al. Resistance Index of the Superior Mesenteric Artery: Correlation With Lactate Concentration and Kinetics Prediction After Cardiac Surgery. Front Med (Lausanne) 2021;8:762376. [Crossref] [PubMed]