Acoustic analysis of a mechanical circulatory support

Detecting Oxygenator Thrombosis in ECMO: A Review of Current Techniques and an Exploration of Future Directions

Extracorporeal membrane oxygenation (ECMO) is a life-support technique used to treat cardiac and pulmonary failure, including severe cases of COVID-19 (coronavirus disease 2019) involving acute respiratory distress syndrome. Blood clot formation in the circuit is one of the most common complications in ECMO, having potentially harmful and even fatal consequences. It is therefore essential to regularly monitor for clots within the circuit and take appropriate measures to prevent or treat them. A review of the various methods used by hospital units for detecting blood clots is presented. The benefits and limitations of each method are discussed, specifically concerning detecting blood clots in the oxygenator, as it is concluded that this is the most critical and challenging ECMO component to assess. We investigate the feasibility of solutions proposed in the surrounding literature and explore two areas that hold promise for future research: the analysis of small-scale pressure fluctuations in the circuit, and real-time imaging of the oxygenator. It is concluded that the current methods of detecting blood clots cannot reliably predict clot volume, and their inability to predict clot location puts patients at risk of thromboembolism. It is posited that a more in-depth analysis of pressure readings using machine learning could better provide this information, and that purpose-built imaging could allow for accurate, real-time clotting analysis in ECMO components.

Detection of inflow obstruction in left ventricular assist devices by accelerometer: A porcine model study

BACKGROUND

Left ventricular assist devices (LVAD) provide circulatory blood pump support for severe heart failure patients. Pump inflow obstructions may lead to stroke and pump malfunction. We aimed to verify in vivo that gradual inflow obstructions, representing prepump thrombosis, are detectable by a pump-attached accelerometer, where the routine use of pump power (PLVAD) is deficient.

METHOD

In a porcine model (n = 8), balloon-tipped catheters obstructed HVAD inflow conduits by 34% to 94% in 5 levels. Afterload increases and speed alterations were conducted as controls. We computed nonharmonic amplitudes (NHA) of pump vibrations captured by the accelerometer for the analysis. Changes in NHA and PLVAD were tested by a pairwise nonparametric statistical test. Detection sensitivities and specificities were investigated by receiver operating characteristics with areas under the curves (AUC).

RESULTS

NHA remained marginally affected during control interventions, unlike PLVAD. NHA elevated during obstructions within 52-83%, while mass pendulation was most pronounced. Meanwhile, PLVAD changed far less. Increased pump speeds tended to amplify the NHA elevations. The corresponding AUC was 0.85-1.00 for NHA and 0.35-0.73 for PLVAD.

CONCLUSION

Elevated NHA provides a reliable indication of subclinical gradual inflow obstructions. The accelerometer can potentially supplement PLVAD for earlier warnings and localization of pump.

Current and future strategies to monitor and manage coagulation in ECMO patients

Extracorporeal membrane oxygenation (ECMO) can provide life-saving support for critically ill patients suffering severe respiratory and/or cardiac failure. However, thrombosis and bleeding remain common and complex problems to manage. Key causes of thrombosis in ECMO patients include blood contact to pro-thrombotic and non-physiological surfaces, as well as high shearing forces in the pump and membrane oxygenator. On the other hand, adverse effects of anticoagulant, thrombocytopenia, platelet dysfunction, acquired von Willebrand syndrome, and hyperfibrinolysis are all established as causes of bleeding. Finding safe and effective anticoagulants that balance thrombosis and bleeding risk remains challenging. This review highlights commonly used anticoagulants in ECMO, including their mechanism of action, monitoring methods, strengths and limitations. It further elaborates on existing anticoagulant monitoring strategies, indicating their target range, benefits and drawbacks. Finally, it introduces several highly novel approaches to real-time anticoagulation monitoring methods including sound, optical, fluorescent, and electrical measurement as well as their working principles and future directions for research.

Detection of inflow obstruction in left ventricular assist devices by accelerometer: An in vitro study

Inflow obstruction in left ventricular assist devices (LVAD) may lead to embolic stroke and pump malfunction. We investigated if an accelerometer detected graded LVAD inflow obstructions. Detection performances were compared to the current continuous surveillance routine based on the pump power consumption (P_LVAD). In ten mock circuit experiments, four different-sized pendulating balloons obstructed HVAD™ inflow conduits cross-section areas by 14%-75%. Nonharmonic amplitudes (NHA) of continuous signals from a triaxial accelerometer attached to the LVAD were compared against single-point P_LVAD values, using load and speed alterations as control interventions. We analyzed the NHA band power with a pairwise nonparametric statistical test. The detection performances were analyzed by receiver operating characteristics with areas under the curves (AUC). The NHA remained unaffected during load alterations. In contrast, NHA increased significantly from the 27% obstruction level (AUC≥0.82), an effect amplified by increased pump speed. P_LVAD did not change significantly below the maximal 75% obstruction level (AUC≤0.36). In conclusion, NHA detected the inflow obstructions much better than P_LVAD. The technique may provide a future monitoring modality of any pendulating obstructive inflow pathology.

Acoustic Properties of Axial and Centrifugal Flow Left Ventricular Assist Devices and Prediction of Pump Thrombosis

OBJECTIVE

To characterize the properties of the audible tones produced by current left ventricular assist device (LVAD) pumps approved for use, and to ascertain if changes in those may be present in the setting of pump thrombosis.

PATIENTS AND METHODS

From August 31, 2016, to January 16, 2020, LVAD recipients consented to have surface recordings obtained using a high-fidelity digital stethoscope. Audio data were analyzed using digital recording and editing software to produce an acoustic spectrogram by Fast Fourier transformation.

RESULTS

Recordings were obtained in 53 patient encounters (27 HeartMate II, 19 HeartWare and 7 HeartMate 3). In 12 patients (9 HeartMate II, 3 HeartWare) there was a clinical concern for pump thrombosis. In all patients and pump models, a fundamental frequency was noted, and the second and third harmonics were also clearly detectable. Where thrombosis occurred in the HeartMate II pump, the absolute (normal -46.9 [-57.5,-42.9] dB vs thrombosis -41.4 [-49.8,-26.8] dB; P=.08) and relative (normal 0.72 [0.62, 0.92] vs thrombosis 0.95 [0.86, 1.24]; P=.01) third harmonic frequencies were increased in amplitude. Where paired data were available, an increase in the absolute and relative third harmonic frequencies was observed in all patients. In the case of the HeartWare device, a consistent difference in harmonic amplitudes in the setting of thrombosis could not be identified.

CONCLUSION

A consistent pattern of fundamental and harmonic frequencies is common to all LVADs currently approved for use. Alterations in the amplitude of higher order harmonics may signal the onset of pump thrombosis in axial flow LVADs.

Evaluation of real-time thrombus detection method in a magnetically levitated centrifugal blood pump using a porcine left ventricular assist circulation model

Pump thrombosis induces significant complications and requires timely detection. We proposed real-time monitoring of pump thrombus in a magnetically levitated centrifugal blood pump (mag-lev pump) without using additional sensors, by focusing on the changes in the displacement of the pump impeller. The phase difference between the current and displacement of the impeller increases with pump thrombus. This thrombus detection method was previously evaluated through simulated circuit experiments using porcine blood. Evaluation of real-time thrombus detection in a mag-lev blood pump was performed using a porcine left ventricular assist circulation model in this study. Acute animal experiments were performed five times using five Japanese domestic pigs. To create thrombogenic conditions, fibrinogen coating that induces thrombus formation in a short time was applied to the inner surfaces of the pump. An inflow and an outflow cannula were inserted into the apex of the left ventricle and the carotid artery, respectively, by a minimally invasive surgical procedure that allowed minimal bleeding and hypothermia. Pump flow was maintained at 1 L/min without anticoagulation. The vibrational frequency of the impeller (70 Hz) and its vibrational amplitude (30 μm) were kept constant. The thrombus was detected based on the fact that the phase difference between the impeller displacement and input current to the magnetic bearing increases when a thrombus is formed inside a pump. The experiment was terminated when the phase difference increased by over 1° from the lowest value or when the phase difference was at the lowest value 12 hours after commencing measurements. The phase difference increased by over 1° in three cases. The pump was stopped after 12 hours in two cases. Pump thrombi were found in the pump in three cases in which the phase difference increased by over 1°. No pump thrombus was found in the other two cases in which the phase difference did not increase. We succeeded in real-time thrombus monitoring of a mag-lev pump in acute animal experiments.

Prediction of aortic valve regurgitation after continuous-flow left ventricular assist device implantation using artificial intelligence trained on acoustic spectra

Significant aortic regurgitation (AR) is a common complication after continuous-flow left ventricular assist device (LVAD) implantation. Using machine-learning algorithms, this study was designed to examine valuable predictors obtained from LVAD sound and to provide models for identifying AR. During a 2-year follow-up period of 13 patients with Jarvik2000 LVAD, sound signals were serially obtained from the chest wall above the LVAD using an electronic stethoscope for 1 min at 40,000 Hz, and echocardiography was simultaneously performed to confirm the presence of AR. Among the 245 echocardiographic and acoustic data collected, we found 26 episodes of significant AR, which we categorized as “present”; the other 219 episodes were characterized as “none”. Wavelet (time–frequency) analysis was applied to the LVAD sound and 19 feature vectors of instantaneous spectral components were extracted. Important variables for predicting AR were searched using an iterative forward selection method. Seventy-five percent of 245 episodes were randomly assigned as training data and the remaining as test data. Supervised machine learning for predicting concomitant AR involved an ensemble classifier and tenfold stratified cross-validation. Of the 19 features, the most useful variables for predicting concomitant AR were the amplitude of the first harmonic, LVAD rotational speed during intermittent low speed (ILS), and the variation in the amplitude during normal rotation and ILS. The predictive accuracy and area under the curve were 91% and 0.73, respectively. Machine learning, trained on the time–frequency acoustic spectra, provides a novel modality for detecting concomitant AR during follow-up after LVAD.

Real-time, non-invasive thrombus detection in an extracorporeal circuit using micro-optical thrombus sensors

Introduction:

Real-time, non-invasive monitoring of thrombus formation in extracorporeal circuits has yet to be achieved. To address the challenges of conventional optical thrombus detection methods requiring large devices that limit detection capacity, we developed a micro-optical thrombus sensor.

Methods:

The proposed micro-optical thrombus sensor can detect the intensity of light scattered by blood at wavelengths of 660 and 855 nm. Two thrombus sensors were installed on in vitro circuit: one at the rotary blood pump and one at a flow channel. To evaluate the variation in the ratio of incident light intensity at each wavelength of the two sensors, R_fluct (for 660 nm) and I_fluct (for 855 nm) were defined. Using fresh porcine blood as a working fluid, we performed in vitro tests of haematocrit (Hct) and oxygen saturation (SaO₂) variation and thrombus detection. Thrombus tests were terminated after R_fluct or I_fluct showed a larger change than the maximum range of those in the Hct and SaO₂ variation test.

Results:

In all three thrombus detection tests, I_fluct showed a larger change than the maximum range of those in the Hct and SaO₂ variation test. After the tests, thrombus formation was confirmed in the pump, and there was no thrombus in the flow channel. The results indicate that I_fluct is an effective parameter for identifying the presence of a thrombus.

Conclusion:

Thrombus detection in an extracorporeal circuit using the developed micro-optical sensors was successfully demonstrated in an in vitro test.

Detecting Suspected Pump Thrombosis in Left Ventricular Assist Devices via Acoustic Analysis

OBJECTIVE

Left ventricular assist devices (LVADs) fail in up to 10% of patients due to the development of pump thrombosis. Remote monitoring of patients with LVADs can enable early detection and, subsequently, treatment and prevention of pump thrombosis. We assessed whether acoustical signals measured on the chest of patients with LVADs, combined with machine learning algorithms, can be used for detecting pump thrombosis.

METHODS

13 centrifugal pump (HVAD) recipients were enrolled in the study. When hospitalized for suspected pump thrombosis, clinical data and acoustical recordings were obtained at admission, prior to and after administration of thrombolytic therapy, and every 24 hours until laboratory and pump parameters normalized. First, we selected the most important features among our feature set using LDH-based correlation analysis. Then using these features, we trained a logistic regression model and determined our decision threshold to differentiate between thrombosis and non-thrombosis episodes.

RESULTS

Accuracy, sensitivity and precision were calculated to be 88.9%, 90.9% and 83.3%, respectively. When tested on the post-thrombolysis data, our algorithm suggested possible pump abnormalities that were not identified by the reference pump power or biomarker abnormalities.

SIGNIFICANCE

We showed that the acoustical signatures of LVADs can be an index of mechanical deterioration and, when combined with machine learning algorithms, provide clinical decision support regarding the presence of pump thrombosis.

Sound analysis of the magnetically levitated left ventricular assist device HeartMate 3™

Introduction:

The HeartMate 3™ has shown lower rates of adverse events compared to previous devices due to the design and absence of mechanical bearings. For previous devices, sound analysis emerged as a way to assess pump function. The aims of this study were to determine if sound analysis can be applied to the HeartMate 3 in vivo and in vitro and to evaluate an electronic stethoscope.

Method:

Sound recordings were performed with microphones and clinical accessible electronic stethoscope. The recordings were studied in both the time and the frequency domains. Recordings from four patients were performed to determine if in vivo and in vitro recordings are comparable.

Results:

The results show that it is possible to detect sound from HeartMate 3 and the sound spectrum is clear. Pump frequency and frequency of the pulsatile mode are easily determined. Frequency spectra from in vitro and in vivo recordings have the same pattern, and the major proportion (96.7%) of signal power is located at the pump speed frequency ±40 Hz. The recordings from the patients show low inter-individual differences except from location of peaks originating from pump speed and harmonics. Electronic stethoscopes could be used for sound recordings, but the dedicated equipment showed a clearer sound spectrum.

Discussion:

The results show that acoustic analysis can also be performed with the HeartMate 3 and that in vivo and in vitro sound spectrum is similar. The frequency spectra are different from previous devices, and methods for assessing pump function or thrombosis need further evaluation.

Accelerometer Detects Pump Thrombosis and Thromboembolic Events in an In vitro HVAD Circuit

Pump thrombosis and stroke are serious complications of left ventricular assist device (LVAD) support. The aim of this study was to test the ability of an accelerometer to detect pump thrombosis and thromboembolic events (TEs) using real-time analysis of pump vibrations. An accelerometer sensor was attached to a HeartWare HVAD and tested in three in vitro experiments using different pumps for each experiment. Each experiment included thrombi injections sized 0.2-1.0 mL and control interventions: pump speed change, afterload increase, preload decrease, and saline bolus injections. A spectrogram was calculated from the accelerometer signal, and the third harmonic amplitude was used to test the sensitivity and specificity of the method. The third harmonic amplitude was compared with the pump energy consumption. The acceleration signals were of high quality. A significant change was identified in the accelerometer third harmonic during the thromboembolic interventions. The third harmonic detected thromboembolic events with higher sensitivity/specificity than LVAD energy consumption: 92%/94% vs. 72%/58%, respectively. A total of 60% of thromboembolic events led to a prolonged third harmonic amplitude change, which is indicative of thrombus mass residue on the impeller. We concluded that there is strong evidence to support the feasibility of real-time continuous LVAD monitoring for thromboembolic events and pump thrombosis using an accelerometer. Further in vivo studies are needed to confirm these promising findings.

Acoustic Signatures of Left Ventricular Assist Device Thrombosis

Left ventricular assist devices (LVADs) are life-saving, surgically implanted mechanical heart pumps used to treat patients with advanced heart failure (HF). While life-saving, LVAD support is associated with a high incidence of complications, making early recognition and management of LVAD complications a critical need. Blood clot formation within the LVAD, known as LVAD thrombosis, is a catastrophic complication of LVAD therapy that often requires LVAD exchange due to delayed diagnosis and treatment. Using digital stethoscopes, we identified differences in acoustic spectra from two patients presenting with LVAD thrombosis compared with normally functioning LVAD pumps within the same patient. Importantly, these acoustic changes were present even in the absence of typical signs of HF that are often present in LVAD thrombosis patients. Our work suggests that acoustic spectral analysis of digital stethoscope signals could be used for early detection and mitigation of LVAD complications.

An in vitro acoustic analysis and comparison of popular stethoscopes

Purpose

To compare the performance of various commercially available stethoscopes using standard acoustic engineering criteria, under recording studio conditions.

Materials and methods

Eighteen stethoscopes (11 acoustic, 7 electronic) were analyzed using standard acoustic analysis techniques under professional recording studio conditions. An organic phantom that accurately simulated chest cavity acoustics was developed. Test sounds were played via a microphone embedded within it and auscultated at its surface by the stethoscopes. Recordings were made through each stethoscope’s binaurals and/or downloaded (electronic models). Recordings were analyzed using standard studio techniques and software, including assessing ambient noise (AMB) rejection. Frequency ranges were divided into those corresponding to various standard biological sounds (cardiac, respiratory, and gastrointestinal).

Results

Loudness and AMB rejection: Overall, electronic stethoscopes, when set to a maximum volume, exhibited greater values of perceived loudness compared to acoustic stethoscopes. Significant variation was seen in AMB rejection capability. Frequency detection: Marked variation was also seen, with some stethoscopes performing better for different ranges (eg, cardiac) vs others (eg, gastrointestinal).

Conclusion

The acoustic properties of stethoscopes varied considerably in loudness, AMB rejection, and frequency response. Stethoscope choice should take into account clinical conditions to be auscultated and the noise level of the environment.

Left ventricular assist device exchange for the treatment of HeartMate II pump thrombosis

Background

Pump thrombosis is the most severe and acute complication of left ventricular assist device (LVAD) therapy and treatment remains challenging. Whilst lysis therapy is often not successful, the exchange of the occluded LVAD is currently the most applied therapeutic treatment for this event. With this study we examine the effects of minimal-invasive LVAD exchange on the rate re-thrombosis and outcomes as well as adverse events in the study group.

Methods

Between February 2004 and December 2015 more than 600 LVADs were implanted at our institution. We retrospectively studied a patient cohort of 41 patients who underwent LVAD exchange because of pump thrombosis at a single institution. Outcomes, rates of re-thrombosis and adverse events were analyzed.

Results

Between February 2004 and December 2015, 87 exchanges of LVADs were performed at a single center. In 41 cases pump thrombosis was the reason for LVAD exchange. A total of 28 patient years (10,276 days) were analyzed. Average ICU stay was 15.8±20.4 days and average in-hospital stay 38.1±37.3 days after LVAD exchange. After thirty days the survival rate was 80.5%, 75.6% after 6 months and 70.7% one year after LVAD exchange. Out of the study cohort, three patients have successfully undergone heart transplantation. Twelve patients suffered a stroke postoperatively (29%). Twelve patients needed postoperative dialysis (29%). No technical complications of the VAD were recorded in the study group. Two patients underwent successful LVAD explantation due to myocardial recovery. One year after LVAD exchange, 14 patients underwent re-exchange due to pump thrombosis (34%). Eight patients suffered from a LVAD related infection out of which two patients were treated by pump exchange. A total of 12 patients died during the complete one year follow up of this study (29%). Four patients died in the second, two in the third and one in the fourth year after LVAD exchange. The remaining 17 patients are still ongoing on the device.Conclusions: It is generally feasible to treat pump thrombosis via LVAD exchange. Yet, the exchange procedure is not without risk and the risk of re-thrombosis (34%), stroke (29%), postoperative dialysis (29%) and perioperative complications remains high.

Sound analysis of a left ventricular assist device: A technical evaluation of iOS devices

Introduction:

The use of left ventricular assist device (LVAD) has grown rapidly. Adverse events do continue to occur. In recent years, analysis of LVAD sound recordings emerged as a means to monitor pump function and detect pump thrombosis. The aim of this study was to characterize the sounds from HeartMate II and to evaluate the use of handheld iOS devices for sound recordings.

Method:

Signal analysis of LVAD sound recordings, with dedicated recording equipment and iOS devices, was performed. Two LVADs running in mock loop circuits were compared to an implanted LVAD. Spectral analysis and parametric signal models were explored to quantify the sound and potentially detect changes in it.

Results:

The sound recordings of two LVADs in individual mock loop circuits and a third one implanted in a patient appeared to be similar. Qualitatively, sound characteristics were preserved following changes in pump speed. Recordings using dedicated equipment showed that HeartMate II sound comprises low-frequency components corresponding to pump impeller rotation, as well as high-frequency components due to a pulse width modulation of the electric power to the pump. These different signal components interact and result in a complicated frequency spectrum. The iPhone and iPod recordings could not reproduce the sounds as well as the dedicated equipment. In particular, lower frequencies were affected by outside disturbances.

Discussion:

This article outlines a systematic approach to LVAD sound analysis using signal processing methods to quantify and potentially detect changes, and describes some of the challenges, for example, with the use of inexpensive recording devices.

Artificial hearts-recent progress: republication of the article published in the Japanese Journal of Artificial Organs

This review was created based on a translation of the Japanese review written in the Japanese Journal of Artificial Organs in 2015 (Vol.44, No. 3, pp.130-135), with some modifications regarding several references published in 2015 or later.

Epidermal mechano-acoustic sensing electronics for cardiovascular diagnostics and human-machine interfaces

Physiological mechano-acoustic signals, often with frequencies and intensities that are beyond those associated with the audible range, provide information of great clinical utility. Stethoscopes and digital accelerometers in conventional packages can capture some relevant data, but neither is suitable for use in a continuous, wearable mode, and both have shortcomings associated with mechanical transduction of signals through the skin. We report a soft, conformal class of device configured specifically for mechano-acoustic recording from the skin, capable of being used on nearly any part of the body, in forms that maximize detectable signals and allow for multimodal operation, such as electrophysiological recording. Experimental and computational studies highlight the key roles of low effective modulus and low areal mass density for effective operation in this type of measurement mode on the skin. Demonstrations involving seismocardiography and heart murmur detection in a series of cardiac patients illustrate utility in advanced clinical diagnostics. Monitoring of pump thrombosis in ventricular assist devices provides an example in characterization of mechanical implants. Speech recognition and human-machine interfaces represent additional demonstrated applications. These and other possibilities suggest broad-ranging uses for soft, skin-integrated digital technologies that can capture human body acoustics.

Detection of Thrombosis in the Extracorporeal Membrane Oxygenation Circuit by Infrasound: Proof of Concept

As of today, there exist no reliable, objective methods for early detection of thrombi in the extracorporeal membrane oxygenators (ECMO) system. Within the ECMO system, thrombi are not always fixed to a certain component or location in the circuit. Thus, clot fragments of different shapes and consistencies may circulate and give rise to vibrations and sound generation. By bedside sound measurements and additional laboratory experiments (although not detailed herein), we found that the presence of particles (clots or aggregates and fragments of clots) can be detected by analyzing the strength of infra-sound (< 20 Hz) modes of the spectrum near the inlet and outlet of the centrifugal pump in the ECMO circuit. For the few patients that were considered in this study, no clear false positive or negative examples were found when comparing the spectral approach with clinical observations. A laboratory setup provided insight to the flow in and out of the pump, confirming that in the presence of particles a low-amplitude low-frequency signal is strongly amplified, enabling the identification of a clot.

Acoustic Characterization of Axial Flow Left Ventricular Assist Device Operation In Vitro and In Vivo

The use of left ventricular assist devices (LVADs), implantable pumps used to supplement cardiac output, has become an increasingly common and effective treatment for advanced heart failure. Although modern continuous-flow LVADs improve quality of life and survival more than medical management of heart failure, device malfunction remains a common concern. Improved noninvasive methods for assessment of LVAD function are needed to detect device complications. An electronic stethoscope was used to record sounds from the HeartMate II axial flow pump in vitro and in vivo. The data were then uploaded to a computer and analyzed using two types of acoustic analysis software. Left ventricular assist device acoustics were quantified and were related to pump speed, acoustic environment, and inflow and outflow graft patency. Peak frequency values measured in vivo were found to correlate strongly with both predicted values and in vitro measurements (r > 0.999). Plots of the area under the acoustic spectrum curve, obtained by integrating over 50 Hz increments, showed strong correlations between in vivo and in vitro measurements (r > 0.966). Device thrombosis was found to be associated with reduced LVAD acoustic amplitude in two patients who underwent surgical device exchange.

Pump thrombosis-A riddle wrapped in a mystery inside an enigma

This manuscript reviews the state of the art regarding the subject of pump thrombosis (PT). The historical context of PT and the clinical data are described, the etiologic factors are elucidated, preventive strategies are explored, diagnostic modalities are reviewed, and management principles are defined. There clearly remains much work to be done towards solving this riddle wrapped in a mystery inside an enigma, but promising foundations are being established.