Research Project: Evaluation of Blood Damage Caused by Medical Materials and Devices
The Blood Damage Assessment Laboratory was established to assist FDA and manufacturers in the performance and interpretation of in vitro blood damage testing of medical devices. CDRH requires premarket bench testing for most blood-contacting cardiovascular materials and devices to determine their potential for causing blood damage. Moreover, in vitro testing is also important in determining the source of blood damage during postmarket adverse patient event investigations. However, uncertainty in the performance and use of the data can occur since standardized testing and reporting procedures, assays for quantifying blood damage (i.e. hemolysis of erythrocytes, platelet activation, and thrombosis), and objective pass/fail criteria have not been established for evaluating many of these devices. Furthermore, it has not been shown that current preclinical hemocompatibility testing can adequately and reliably predict a device’s clinical performance. This laboratory addresses these deficiencies by contributing to standards and developing guidelines on how to perform, report, and interpret hemolysis testing, by evaluating the differences between testing with animal and human blood, and by exploring how flow visualization and computational flow dynamic simulations can be used to validate and predict blood damage in medical devices (i.e. relating important engineering flow parameters (shear forces, cell exposure times) to adverse biologic responses (hemolysis, platelet activation)). In addition, to better assess the preclinical safety of medical devices and materials, in vitro methods to characterize device-induced platelet activation and thrombus formation are also being developed in this lab. Through better evaluation techniques, FDA will help to improve upon device design, testing, and regulatory review that will lead to improved safety and efficacy throughout the premarket and postmarket stages of the products’ life cycle.
Task: Important Parameters during In Vitro Hemolysis Testing
In vitro hemolysis tests, which measure the release of hemoglobin from damaged erythrocytes, are commonly used for evaluating blood cell damage caused by materials and devices. In patients, hemolysis resulting from blood-device interactions has been linked to renal failure, anemia, arrhythmias, and death. Although FDA has generated guidance documents to promote standardized testing and reporting procedures for making relative comparisons between new and predicate medical devices, assays for blood damage and objective pass/fail criteria have not been established for evaluating many devices used in cardiovascular applications. Moreover, in vitro testing is most often performed using bovine blood, which limits use of the results since it is less fragile and behaves differently than human blood. Due to the complexity and uncertainty in current blood trauma testing, preclinical evaluations of new devices using in vitro laboratory and animal testing have not always predicted how the devices will perform in actual patients. Thus, there is an immediate need to better understand the limitations in extrapolating laboratory hemolysis results to clinical use. Previous laboratory accomplishments related to this project include comparative testing of the common measurement techniques used to assess hemolysis, development of standardized testing for evaluating hemolysis caused by medical materials, evaluation of some of the physical parameters which affect medical device hemolysis testing, and the determination of how undetected kinked hemodialysis tubing can cause hemolysis. Recent work includes designing and fabricating experimental flow models and test systems for use with animal and human blood that simulate fluid stresses encountered in medical devices, such as from sudden changes in flow geometry or flow in narrow gaps. Flow-through recirculating blood loop models, a single-pass orifice model, a rotating cone-plate rheometer, and a simple rocker-bead method will be used to identify the differences in blood fragility between bovine, porcine, sheep and human blood. Experiments will be designed and carried out to identify how blood properties (e.g. anticoagulation, species, hematocrit, temperature, blood age and deformability) affect the test results. It is hypothesized that the results of the hemolysis testing will depend on both the blood species and the flow characteristics of the device being tested.
Task: Platelet Activation in Medical Devices
Although hemolysis testing provides a valuable and easily obtainable surrogate measure of damage to blood elements, it does not provide information about localized platelet activation and thrombus formation, which may occur in devices and lead to device malfunction and/or serious embolic complications. While several techniques are currently available for quantifying platelet activation and aggregation, no standardized methods have been established for the in vitro testing of medical devices, especially when performing tests with bovine blood. Hence, it is extremely difficult to determine the safety of a medical device based on platelet data provided in a company’s premarket submission. To evaluate thrombogenicity of biomaterials used in blood-contacting medical devices, current FDA/CDRH policy recommends an in vivo test involving a non-heparinized canine venous implant model. However, it has been suggested that this canine model may not be predictive of human clinical use because canine blood is more susceptible to thrombus formation than human blood. In addition, this model may not capture some potential thrombogenic events if formed thrombi are easily detached from the material surfaces. The widely referenced hemocompatibility standard ISO 10993-4 suggests a list of in vitro tests to assess thrombogenicity, but there are no detailed protocols to perform these tests, and their usefulness has not been established. The goal of this project is to develop in vitro assays for both animal and human blood to assess platelet activation, coagulation activity, and thrombus formation due to the exposure of blood to medical materials/devices in relevant flow models.
Task: Computational Fluid Dynamic predictions of Blood Damage: An FDA Critical Path Initiative Project
Flow patterns within medical devices have been directly linked to blood damage. Regions of high shear stress can cause mechanical damage to red and white blood cells and promote platelet activation, whereas regions of low velocity are susceptible to blood clot formation. To evaluate fluid transport in medical devices, companies use particle flow visualization and computational fluid dynamic simulation techniques. However, owing to the use of simplifications in computational simulations, the lack of rigorous validation techniques, and the complexity in relating shear forces and cell exposure times to blood damage, these tools have limited utilization in current device review. Since future submissions to FDA will increasingly rely on computational flow simulations for designing and characterizing devices, research which integrates blood-loop bench experiments, flow visualization, and computational techniques is needed to identify the critical variables associated with mechanical hemolysis and thrombosis in medical devices.
For the past two years, our laboratory has directed the FDA Critical Path Initiative project “Standardization of computational fluid dynamic (CFD) techniques used to evaluate performance and blood damage safety in medical devices”. Through this science initiative, we established a technical working group consisting of academic consultants to better understand how CFD is being used in the medical device industry. In Phase 1 of the project, participants from academia, industry, and the FDA performed an interlaboratory study on a benchmark flow model (nozzle design) to determine the applicability and limits of current CFD simulations in evaluating medical devices. Through our partnering with biomedical and computational societies, 28 groups from around the world performed fluid dynamic simulations; 11 of these groups also predicted levels of blood damage in the nozzle model under different flow conditions. The large variations seen in the results of the simulations, compared to those obtained by quantitative flow visualization performed in three independent laboratories, demonstrate an industry-wide need to standardize the use and validation of CFD in medical device evaluation. In Phase 2 of the project, lessons learned from Phase 1 will be utilized by participants to investigate flow through a ventricular assist device model. Actual blood damage testing of the benchmark devices is currently underway in three labs to support a database that will help researchers develop predictive models of blood damage. This project will help to relate the engineering results provided by CFD simulations to predictions of the biological response to the device. Through the Critical Path Initiative program, OSEL can engage a large number of researchers to address these complex problems and develop best-practice guidelines. Besides assisting the medical device community, this important project has also helped OSEL to develop internal expertise in the areas of quantitative flow visualization, CFD simulations, and blood damage assessment.