Safety and Accuracy of a Novel Bioimpedance System for Real-Time Detection and Monitoring of Endovascular Procedure-Related Bleeding in a Porcine Model

Abstract: Objectives. The aim of this study was to determine the safety and accuracy of a novel bleed detection system, the Early Bird Bleed Monitoring System (EBBMS; Saranas) for the detection of simulated internal bleeding and the monitoring of bleed progression associated with endovascular procedures. Background. Periprocedural bleeding events during endovascular procedures are frequent and are associated with increased morbidity, mortality, and healthcare costs. Methods. This study was a prospective, self-controlled, acute animal study including 20 Yorkshire cross swine undergoing endovascular procedures involving cannulation of both femoral artery and vein. Extravascular bleeding was simulated by a continuous and controlled subcutaneous injection of a blood solution proximal to the access site. The capacity of the EBBMS to detect bleed occurrence and to characterize its progression in three levels of severity (level 1, level 2, level 3) was assessed. Sensitivity and specificity in bleed detection were determined. Results. Forty EBBMS devices were inserted in 20 animals. During these 40 procedures, bleeding was appropriately detected in all of them. The EBBMS achieved a sensitivity of 100% and specificity of 100% in detection of bleeding. Detection of bleeding progression at level 1 severity occurred at 31.5 ± 12.7 mL, level 2 severity at 77.8 ± 53.5 mL, and level 3 severity at 145.5 ± 100.5 mL, with a significant difference in blood volume (P<.001). No significant difference in bleed detection was seen when the EBBMS was inserted in the femoral vein or artery. Conclusion. The EBBMS accurately detected access-related bleeding onset and progression during a simulated endovascular procedure.

J INVASIVE CARDIOL 2020;32(7):249-254. Epub 2020 June 8.

Key words: bleeding, catheter, percutaneous intervention, surgery

Periprocedural bleeding events in patients undergoing endovascular procedures are frequent and are associated with increased morbidity and mortality.^1-5 Emergent percutaneous interventions using large-bore endovascular access, such as transcatheter aortic valve replacement, percutaneous ventricular assist device, and endovascular aneurysm repair, are intrinsically associated with higher bleeding event rates compared with procedures involving smaller catheters.^1,6 Large-bore catheter-related bleeding has also been shown to be associated with significant increases in mortality, hospital length of stay, and healthcare costs.¹ Despite major technological improvements in device profile and deliverability, major bleeding still occurs and compromises the benefit of these new therapies.^7,8 Unfortunately, early detection of these bleeds remains challenging, and clinical recognition of periprocedural bleeding mainly relies on the occurrence of signs and symptoms (hematoma, pain, hypotension, and sometimes death) when the severity of the blood loss has already reached a level of clinical significance.^9,10 Ideally, the detection of procedural bleeding events should occur early in the process, at the time of bleed initiation, where the magnitude of the bleed is not yet associated with symptoms, potentially leading to subsequent diagnostic or therapeutic maneuvers that could mitigate the severity of the bleed and its associated detrimental effects.

The Early Bird Bleed Monitoring System (EBBMS; Saranas) was designed to measure the bioimpedance across the blood vessel, allowing for an early and real-time detection of internal bleeding complications at low volumes, prior to patients becoming symptomatic post procedure. The system consists of: (1) a standard 6 Fr or 8 Fr endovascular access sheath (Figure 1A); (2) a bleed detection array with embedded electrodes; and (3) a user interface display (Figure 1B), which houses a printed circuit board assembly running an algorithm that analyzes bioimpedance and can trigger visible and audible indicators to communicate the state of change in bioimpedance.

Briefly, bleed monitoring is accomplished via a proprietary algorithm, which monitors and interrogates changes in regional bioimpedance. It has been demonstrated that there is a consistent correlation between a decrease in bioimpedance and an increase in extravascular fluid accumulation. Bioimpedance measurements are obtained through a series of electrodes, which provide a means of electrical contact with body fluids and tissue and are located on the sheath cannula.¹¹ The two outer electrodes drive a fixed frequency, alternating current to establish an electrical field, which is measured by the 2 inner electrodes. Extraneous signals are filtered out through a series of high and low pass filters integrated on the printed circuit board assembly and digital filters employed in the firmware. 

After EBBMS insertion and power up, bioimpedance measurements are taken. As the procedure progresses, the bioimpedance will change in response to an internal bleeding complication, presented as fluid extravasation, which is detected by the device. The user interface display features a 3-level bleed indicator system (level 1, level 2, and level 3) that sequentially illuminates LED indicators to show an increase in internal bleed complication progression (Figure 1B). An audible notification is also generated for each level. Upon vessel trauma and/or postprocedural vessel access bleeding, extravascular fluid can accumulate. Due to this accumulation, the bioimpedance begins to slope downward. If the downward slope reaches a prescribed slope threshold, the algorithm will trigger the level 1 indicator. At this moment, the algorithm will log the current bioimpedance measurement as a reference from which bleed progression will be measured against. If the progression of the bleed persists and the impedance continues to fall, then after a fixed percent change from the level 1 bioimpedance reference point, the EBBMS will trigger the level 2 indicator. If bleeding continues to progress, the EBBMS will trigger the level 3 indicator at a predetermined bioimpedance percent change threshold.

The present study reports the initial animal data to support the safety and efficacy of the EBBMS to detect the onset and progression of internal bleeding events associated with endovascular procedures.

Methods

This prospective, self-controlled, acute animal study utilized 20 Yorkshire cross swine, (commercially available, ages 6-12 months, and weighing 75-90 kg) and 40 EBBMS devices. Animal testing was performed at American Preclinical Services, an accredited United States Department of Agriculture research facility in Minneapolis, Minnesota (#41-R-0074). Institutional Animal Care and Use Committee approval was obtained for the protocol.

The aim of this study was to demonstrate the ability of the EBBMS to detect extravascular fluid accumulation at the access-site level. The 8 Fr EBBMS devices were implanted into the femoral artery or femoral vein using the standard Seldinger technique and ultrasound guidance. An additional standard sheath (8 Fr) was also inserted via Seldinger technique and ultrasound guidance into the ipsilateral vessel (either vein or artery) next to the EBBMS system. This additional sheath was inserted to assess the interoperability of the EBBMS to function properly while an adjacent sheath was in place, involving manipulation and passage of endovascular material within this sheath. Each animal underwent 2 successive, simulated percutaneous coronary interventions (1 EBBMS per leg) for a total of 40 procedures (Figure 2). 

Following sheath insertions, a baseline non-bleed phase lasting 10 minutes prior to simulated bleed was incorporated to mimic the time from when the introducer sheath is first inserted into the vasculature to the time of primary intervention. Once the non-bleed phase was complete, the infusion needle was placed relative to the proximal electrode, under fluoroscopic guidance, to simulate an arterial access-site bleed with blood solution injection, initiating the simulated bleed period (Figure 3). After needle placement, 500 mL of blood solution, consisting of porcine blood:saline:contrast (3:2:1) to best simulate blood impedance and to allow for fluoroscopic imaging, was injected through the infusion needle at a constant rate of 10 mL/min to simulate the access bleed (Figure 3). 

Standard equipment used during percutaneous procedures, such as 6 Fr guide catheters (Merit Medical Systems) and 0.014˝ guidewires (Boston Scientific Corporation), were advanced through the EBBMS catheter and the commercially available sheath (8 Fr Prelude Pro; Merit Medical Systems) during the non-bleed and simulated bleed phases of the procedure. Fluoroscopic images were collected for every EBBMS bleed level indication during the study period and at specific volumetric injection points during the bleed simulation. Gross necropsies were performed, and histological analysis of collected tissues was performed.

Statistical analysis. Sensitivity and specificity were determined based on the capability of the EBBMS to appropriately detect bleeding (level 1 indicator triggered) during blood infusion or, if bleeding was detected during the non-bleed phase, confirmed by gross necropsy. To characterize bleed volume for each stage, descriptive statistics (mean ± standard deviation, median with interquartile range, and range) were used. Since the paired differences were not normally distributed, the Wilcoxon signed-rank test, a non-parametric method, was used to compare the paired differences in volumes between bleed indication levels (level 1, level 2, and level 3).

Results

Forty EBBMS devices were inserted in 20 animals (2 legs per animal). During these 40 procedures, bleeding events were detected in all of them. Among them, 10 bleeding events (25%) triggered a level 1 bleed indication on the EBBMS during the non-bleeding phase. After the non-bleed phase concluded, immediate gross necropsy confirmed the presence of bleeding in all of these animals at the access-site level, thus ruling out a false positive signal from the EBBMS in these 10 events. These bleeds occurred because of vessel injuries during the access and sheath insertion processes. Among the 30 procedures with no bleed detected during the non-bleeding phase, the EBBMS successfully detected the initiation (level 1) and the progression (level 2 and level 3 indicators) of all the simulated bleeds. Table 1 shows the EBBMS sensitivity (100%; no false negatives) and specificity (100%; no false positives) for bleed detection. Table 2 summarizes the volume of blood detected for each level. Detected volume of blood significantly increased through each level (P<.001) (Figure 4). 

Once the EBBMS indicated a level 1 bleed (bleed detection), an additional 46.3 ± 46.5 mL of blood solution was required to trigger a level 2 indicator (bleed progression). Once the EBBMS triggered at level 2 indicator, an additional 67.7 ± 77.2 mL of blood solution was required to trigger the level 3 indicator (Figure 4). In the samples that did not trigger a bleed-indicator notification during the non-bleed period, the bioimpedance response remained stable during percutaneous intervention simulations and decreased systematically upon infusion of the blood solution (Figure 5). 

The volume of blood detected by the EBBMS inserted in the femoral artery versus the femoral vein is summarized in Table 3. No significant difference in blood-detection volume was seen between EBBMS arterial insertion versus venous insertion (P=.83 for level 1; P=.13 for level 2; and P=.81 for level 3).

Among all sections of arteries and veins instrumented in this study, including the vessels receiving the EBBMS and the vessels receiving the standard endovascular sheath, all showed some degree of endothelial loss, with only rare sections showing evidence of damage to the underlying wall of the vessel. There was no significant damage to the vessel walls that was specifically associated with the EBBMS device compared with the standard sheath introducer. No significant thrombus or endothelium damage were detected. No major adverse events to animal health were noted during the study.

Discussion

The current report, derived from 20 animals undergoing 40 endovascular procedures with simulated access-related bleeding, demonstrated for the first time the capacity of the Saranas EBBMS to appropriately detect bleed initiation and bleed progression. The EBBMS demonstrated excellent sensitivity (100%) and specificity (100%). Bleed detection was independent of whether the EBBMS was inserted in the vein or artery.

Access-related bleeding complications are the most frequent complications associated with endovascular procedures. A recent report from more than 17,000 real-world endovascular procedures demonstrated that approximately 1 out of 5 patients experienced major bleeding events directly related to endovascular access.¹ The occurrence of major bleeding after procedures has been associated with an increase in morbidity, including transfusion, acute kidney injury, dialysis, infection, increased hospital length of stay, and mortality. It has been determined that each major bleeding event costs approximately $18,000, increasing the cost burden on the healthcare system. Incorporating a technology that could mitigate the severity of these major bleeding events, such as the early detection of any bleeding events, could potentially lead to an improvement in patient outcomes and a decrease in healthcare cost by addressing the potential bleeding complication prior to the patient becoming symptomatic. 

Interestingly, the Saranas EBBMS demonstrated the capacity to detect the early onset of bleeding, with level 1 triggering the alarm at a mean of approximately 30 mL of extravascular blood accumulation. While this amount of bleeding could be considered clinically insignificant, early detection of any bleed is crucial in order to intervene early to prevent the progression to more significant bleeding and its detrimental consequences. The current paradigm for detection of bleeding complications, such as groin hematoma and retroperitoneal bleed, revolves around the development of symptoms, which has been shown to be associated with increased mortality.^1,4,12 Symptoms associated with bleeding usually appear late when hypotension or shock occurred, usually after approximately 500-1000 mL of blood loss. If blood loss and the need for transfusion are to be prevented, early detection of bleeding must occur below ~50 mL of extravascular accumulation. The Saranas EBBMS demonstrated the required sensitivity to potentially mitigate and modify the natural progression of access-related bleeding complications and their detrimental effect on patient prognosis.

Importantly, the Saranas EBBMS demonstrated the capacity to detect bleeding when inserted in either the femoral vein or artery. This finding is important and provides the versatility of using this system in a variety of procedures, either venous or arterial. Actually, the strategy used in the current animal study mirrors a potential use in large-bore arterial procedures, such as transcatheter aortic valve replacement or use of a hemodynamic mechanical support device, where the large-bore device is inserted in the arterial axis and the Saranas EBBMS is inserted in the venous axis, parallel and adjacent to the large arterial device. This strategy has the advantage of being able to monitor for the onset and progression of bleeding not only during the procedure, but also post removal of the arterial large-bore device. Indeed, many bleeds occur within 4 hours of removal of the large arterial sheath, where adequate hemostasis remains challenging. Having the capacity to continuously monitor the extravascular space (access-site and retroperitoneal level) via a venous system that could stay in place during the postprocedural/recovery phase, even after the removal of the arterial devices, represents a novel approach that could have substantial benefits.

Conclusion

Results from this animal validation study demonstrated the capability of the Saranas EBBMS to accurately detect internal bleeds, with 100% sensitivity and 100% specificity, and the capability to actively detect and alert when significant changes in bleed progression occur. The clinical utility of such novel technology in humans remains to be determined and is currently ongoing (NCT03621202). 

From the ¹Gagnon Cardiovascular Institute, Morristown Medical Center, Morristown, New Jersey; ²Saranas, Inc., Houston, Texas; ³Ruiz Department of Ophthalmology and Visual Science, McGovern Medical School at The University of Texas Health Science Center at Houston, Houston, Texas; and ⁴Texas Heart Institute, Houston, Texas.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Généreux reports speaker fees from Edwards Lifesciences, Cordis, Medtronic; consultant fees from Abiomed, Boston Scientific, Cardiovascular Systems, Inc, Cordis, Edwards Lifesciences, Medtronic, Opsens, Soundbite Medical Solutions, Pi-Cardia, Saranas, Siemens, SIG.NUM; shareholder in Soundbite Medical Solutions, SIG.NUM, Pi-Cardia, Puzzle Medical. Mr Bueche and Ms Vondran are employees of Saranas, Inc. Dr Razavi is the founder of Saranas. Dr Chuang reports no conflicts of interest regarding the content herein.

Manuscript submitted April 21, 2020, final version accepted April 29, 2020.

Address for correspondence: Philippe Généreux, MD, Gagnon Cardiovascular Institute, Morristown Medical Center, 100 Madison Avenue, Morristown, NJ 07960. Email: philippe.genereux@atlantichealth.org

1. Redfors B, Watson BM, McAndrew T, et al. Mortality, length of stay, and cost implications of procedural bleeding after percutaneous interventions using large-bore catheters. JAMA Cardiol. 2017;2:798-802.
2. Redfors B, Genereux P, Witzenbichler B, et al. Bleeding severity after percutaneous coronary intervention. Circ Cardiovasc Interv. 2018;11:e005542.
3. Kapur NK, Alkhouli MA, DeMartini TJ, et al. Unloading the left ventricle before reperfusion in patients with anterior ST-segment-elevation myocardial infarction. Circulation. 2019;139:337-346.
4. Kwok CS, Khan MA, Rao SV, et al. Access and non-access site bleeding after percutaneous coronary intervention and risk of subsequent mortality and major adverse cardiovascular events: systematic review and meta-analysis. Circ Cardiovasc Interv. 2015;8:e001645.
5. Bhardwaj B, Spertus JA, Kennedy KF, et al. Bleeding complications in lower-extremity peripheral vascular interventions: insights from the NCDR PVI registry. JACC Cardiovasc Interv. 2019;12:1140-1149.
6. Beohar N, Kirtane AJ, Blackstone E, et al. Trends in complications and outcomes of patients undergoing transfemoral transcatheter aortic valve replacement: experience from the PARTNER Continued Access Registry. JACC Cardiovasc Interv. 2016;9:355-363.
7. Vemulapalli S, Dai D, Hammill BG, et al. Hospital resource utilization before and after transcatheter aortic valve replacement: the STS/ACC TVT Registry. J Am Coll Cardiol. 2019;73:1135-1146.
8. Grover FL, Vemulapalli S, Carroll JD, et al; STS ACC TVT Registry. 2016 annual report of the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry. Ann Thorac Surg. 2017;103:1021-1035.
9. Fruhwirth J, Pascher O, Hauser H, Amann W. Local vascular complications after iatrogenic femoral artery puncture [German]. Wien Klin Wochenschr. 1996;108:196-200.
10. Kent KC, Moscucci M, Mansour KA, et al. Retroperitoneal hematoma after cardiac catheterization: prevalence, risk factors, and optimal management. J Vasc Surg. 1994;20:905-910.
11. Arevalos CA, Nathan J, Razavi M. Use of a functionalized introducer sheath and bioimpedance spectroscopy for real-time detection of vascular access complications. J Med Eng Technol. 2015;39:191-197.
12. Eisen A, Kornowski R, Vaduganathan M, et al. Retroperitoneal bleeding after cardiac catheterization: a 7-year descriptive single-center experience. Cardiology. 2013;125:217-222.