Beat-to-beat blood pressure measurement using a cuffless device does not accurately reflect invasive blood pressure

Accuracy of non-invasive cuffless blood pressure in the intensive care unit: Promises and challenges

Objective

Continuous non-invasive cuffless blood pressure (BP) monitoring may reduce adverse outcomes in hospitalized patients if accuracy is approved. We aimed to investigate accuracy of two different BP prediction models in critically ill intensive care unit (ICU) patients, using a prototype cuffless BP device based on electrocardiogram and photoplethysmography signals. We compared a pulse arrival time (PAT)-based BP model (generalized PAT-based model) derived from a general population cohort to more complex and individualized models (complex individualized models) utilizing other features of the BP sensor signals.

Methods

Patients admitted to an ICU with indication of invasive BP monitoring were included. The first half of each patient's data was used to train a subject-specific machine learning model (complex individualized models). The second half was used to estimate BP and test accuracy of both the generalized PAT-based model and the complex individualized models. A total of 7,327 measurements of 15 s epochs were included in pairwise comparisons across 25 patients.

Results

The generalized PAT-based model achieved a mean absolute error (SD of errors) of 7.6 (7.2) mmHg, 3.3 (3.1) mmHg and 4.6 (4.4) mmHg for systolic BP, diastolic BP and mean arterial pressure (MAP) respectively. Corresponding results for the complex individualized model were 6.5 (6.7) mmHg, 3.1 (3.0) mmHg and 4.0 (4.0) mmHg. Percentage of absolute errors within 10 mmHg for the generalized model were 77.6, 96.2, and 89.6% for systolic BP, diastolic BP and MAP, respectively. Corresponding results for the individualized model were 83.8, 96.2, and 94.2%. Accuracy was significantly improved when comparing the complex individualized models to the generalized PAT-based model in systolic BP and MAP, but not diastolic BP.

Conclusion

A generalized PAT-based model, developed from a different population was not able to accurately track BP changes in critically ill ICU patients. Individually fitted models utilizing other cuffless BP sensor signals significantly improved accuracy, indicating that cuffless BP can be measured non-invasively, but the challenge toward generalizable models remains for future research to resolve.

Cuffless blood pressure measuring devices: review and statement by the European Society of Hypertension Working Group on Blood Pressure Monitoring and Cardiovascular Variability

BACKGROUND

Many cuffless blood pressure (BP) measuring devices are currently on the market claiming that they provide accurate BP measurements. These technologies have considerable potential to improve the awareness, treatment, and management of hypertension. However, recent guidelines by the European Society of Hypertension do not recommend cuffless devices for the diagnosis and management of hypertension.

OBJECTIVE

This statement by the European Society of Hypertension Working Group on BP Monitoring and Cardiovascular Variability presents the types of cuffless BP technologies, issues in their validation, and recommendations for clinical practice.

STATEMENTS

Cuffless BP monitors constitute a wide and heterogeneous group of novel technologies and devices with different intended uses. Cuffless BP devices have specific accuracy issues, which render the established validation protocols for cuff BP devices inadequate for their validation. In 2014, the Institute of Electrical and Electronics Engineers published a standard for the validation of cuffless BP devices, and the International Organization for Standardization is currently developing another standard. The validation of cuffless devices should address issues related to the need of individual cuff calibration, the stability of measurements post calibration, the ability to track BP changes, and the implementation of machine learning technology. Clinical field investigations may also be considered and issues regarding the clinical implementation of cuffless BP readings should be investigated.

CONCLUSION

Cuffless BP devices have considerable potential for changing the diagnosis and management of hypertension. However, fundamental questions regarding their accuracy, performance, and implementation need to be carefully addressed before they can be recommended for clinical use.

Improving Cuff-Less Continuous Blood Pressure Estimation with Linear Regression Analysis

In this work, the authors investigate the cuff-less estimation of continuous BP through pulse transit time (PTT) and heart rate (HR) using regression techniques, which is intended as a first step towards continuous BP estimation with a low error, according to AAMI guidelines. Hypertension (the ‘silent killer’) is one of the main risk factors for cardiovascular diseases (CVDs), which are the main cause of death worldwide. Its continuous monitoring can offer a valid tool for patient care, as blood pressure (BP) is a significant indicator of health and, using it together with other parameters, such as heart and breath rates, could strongly improve prevention of CVDs. The novelties introduced in this work are represented by the implementation of pre-processing and by the innovative method for features research and features processing to continuously monitor blood pressure in a non-invasive way. Currently, invasive methods are the only reliable methods for continuous monitoring, while non-invasive techniques measure the values every few minutes. The proposed approach can be considered the first step for the integration of these types of algorithms on wearable devices, in particular on those developed for the SINTEC project.

Elucidation of obstructive sleep apnoea related blood pressure surge using a novel continuous beat-to-beat blood pressure monitoring system

BACKGROUND

Obstructive sleep apnoea (OSA) episode related blood pressure (BP) surge may mediate the association of OSA with cardiovascular disease. However, BP is not measured during a clinical sleep study.

METHOD

We tested the feasibility of incorporating the Caretaker physiological monitor, which utilizes a novel continuous beat-to-beat (b-b) BP monitoring technology, into polysomnography (PSG) and aimed to characterize BP surges related to obstructive respiratory events. B-b BP was concurrently collected and merged with PSG data on a posthoc basis. We compared BP surge between mean respiratory (apnoea, hypopnea and desaturation-alone events) and nonrespiratory events (spontaneous or leg movement-related arousals). We examined the association of the degree of oxygen desaturation with BP surge in a given respiratory event combining all events. A total of 17 consecutive patients (12 men, mean 52 years old, nine diagnostic and eight split-night PSGs) undergoing clinically indicated PSG were included after excluding one patient with poor signal quality due to excessive movement.

RESULTS

Caretaker was well tolerated. Mean respiratory BP surge ranged from 5 to 19 mmHg [Median (IQR) = 13.9 (9.5--16.2)]. Mean BP surge between the respiratory and nonrespiratory events was similar [13.8 (4.5) vs. 14.9 (5.3) mmHg, P = 0.13]. Accounting for the count distribution of desaturation/BP surge data pair events, there was a linear correlation between the degree of oxygen desaturation and BP surge (R = 0.57, P < 0.001). In eight patients undergoing split-night sleep studies, the number of BP surge events (≥10 mmHg/h) decreased during continuous positive airway pressure in all but one patient.

CONCLUSION

We demonstrated highly variable OSA-related BP surge patterns using the Caretaker's b-b BP monitoring technology that has the potential to be integrated into sleep studies.

Tracking of the beat-to-beat blood pressure changes by the Caretaker physiological monitor against invasive central aortic measurement

OBJECTIVE

There is an unmet need for noninvasive continuous blood pressure (BP) monitoring technologies in various clinical settings. Continuous and noninvasive central aortic BP monitoring is technically not feasible currently, but if realized, would provide more accurate and real-time global hemodynamic information than any form of peripheral arterial BP monitoring in an acute care setting. As part of our efforts to develop such, herein we examined the tracking correlation between noninvasively-derived peripheral arterial BP by Caretaker device against invasively measured central aortic BP.

METHODS

Beat-to-beat BP by Caretaker was recorded simultaneously with central aortic BP measured in patients undergoing cardiac catheterization. Pearson's correlation was also derived for SBP and DBP. A trend comparison analysis of the beat-to-beat BP change was performed using a four-quadrant plot analysis with the exclusion zones of 0.5 mmHg/s to determine concordance, (i.e. the direction of beat-to-beat changes in SBP and DBP).

RESULTS

A total of 47 patients were included in the study. A total of 31 369 beats representing an average of 17.3 min of recording were used for analysis. The trend analysis yielded concordances of 84.4 and 83.5% for SBP and DBP, respectively. Respective correlations (Pearson's r) for SBP and DBP trends were 0.87 and 0.86 (P < 0.01). Tracking of beat-to-beat BP by Caretaker showed excellent concordance and correlation in the direction and the degree of BP change with central aortic BP, respectively.

CONCLUSION

This study supports the satisfactory performance of the Caretaker device in continuous tracking of central aortic BP beat-to-beat BP and provides a basis to develop an algorithm for absolute central aortic BP estimation in the future.

Interpretation of Heart Rate Variability: The Art of Looking Through a Keyhole

The heart may be a mirror of the soul, but the human mind is more than its heart rate variability (HRV). Many techniques to quantify HRV promise to give a view of what is going on in the body or even the psyche of the subject under study. This "Hypothesis" paper gives, on the one hand, a critical view on the field of HRV-analysis and, on the other hand, points out a possible direction of future applications. In view of the inherent variability of HRV and the underlying processes, as lined out here, the best use may be found in serial analysis in a subject/patient, to find changes over time that may help in early discovery of developing pathology. Not every future possibility is bright and shining, though, as demonstrated in a fictional diary excerpt from a future subject, living in a society geared toward preventive medicine. Here implanted biochips watch over the health of the population and artificial intelligence (AI) analyses the massive data flow to support the diagnostic process.