Research Papers

Thrombogenicity Testing of Medical Devices in a Minimally Heparinized Ovine Blood Loop

[+] Author and Article Information
Kent Grove

American Preclinical Services,
8960 Evergreen Boulevard, NW,
Coon Rapids, MN 55433
e-mail: kgrove@sjm.com

Steve M. Deline

American Preclinical Services,
8960 Evergreen Boulevard, NW,
Coon Rapids, MN 55433
e-mail: sdeline@apsemail.com

Tim F. Schatz, Sarah E. Howard

American Preclinical Services,
8960 Evergreen Boulevard, NW,
Coon Rapids, MN 55433
e-mail: tschatz@apsemail.com

Deanna Porter

St. Jude Medical,
177 East County Road B,
St. Paul, MN 55117
e-mail: dporter@sjm.com

Mark E. Smith

American Preclinical Services,
8945 Evergreen Boulevard, NW,
Coon Rapids, MN 55433
e-mail: msmith@apsemail.com

1Present address: St. Jude Medical, 177 East County Road B, St. Paul, MN 55117.

2Corresponding author.

Manuscript received July 20, 2016; final manuscript received January 5, 2017; published online May 3, 2017. Assoc. Editor: Marc Horner.

J. Med. Devices 11(2), 021008 (May 03, 2017) (8 pages) Paper No: MED-16-1273; doi: 10.1115/1.4035724 History: Received July 20, 2016; Revised January 05, 2017

ISO 10993-4 in vivo thrombogenicity testing is frequently performed for regulatory approval of many blood-contacting medical devices and is often a key part of submission packages. Given the current state of in vivo thrombogenicity assays, a more robust and reproducible assay design, including in vitro models, is needed. This study describes an in vitro assay that integrates freshly harvested ovine blood containing minimal heparin in a closed pumped loop. To confirm the reproducibility of this assay, control materials were identified that elicited either a positive or a negative thrombogenic response. These controls demonstrated reproducibility in the resulting thrombogenicity scores with median scores of 5 and 0 for the positive and negative controls, respectively, which also demonstrated a significant difference (p < 0.0001). For a direct comparison of the in vitro blood loop assay to the traditional in vivo nonanticoagulated venous implant (NAVI) assay, seven sheep were used as blood donors for the loop and then as subjects for an NAVI assay. In each assay—loop or NAVI—three study articles were used: the positive and negative controls and a marketed, approved catheter. The resulting thrombogenicity scores were similar when comparing the loop to the NAVI results. For each study article, the median thrombogenicity scores were the same in these two different assays, being 0, 1, and 5 for the negative control, the marketed catheter, and the positive control, respectively. These data suggest that the in vitro assay performs similarly to the in vivo NAVI assay. This in vitro blood loop method has the potential to predict a materials' in vivo thrombogenicity, can substantially de-risk the materials or coating selection process, and may eventually be able to replace the in vivo models currently in use.

Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.


ISO, 2002, “ Biological Evaluation of Medical Devices—Part 4: Selection of Tests for Interactions With Blood,” Association for the Advancement of Medical Instrumentation, Arlington, VA, Standard No. ISO 10993-4.
ASTM, 2009, “ Standard Practice for Assessment of Hemolytic Properties of MATERIALS, Medical and Surgical Materials and Devices; Anesthetic and Respiratory Equipment; Manufacture of Pharmaceutical Products,” American Society for Testing and Materials, ASTM International, West Conshohocken, PA, Standard No. F7560-00.
Wolf, M. F. , and Anderson, J. M. , 2012, “ Practical Approach to Blood Compatibility Assessments: General Considerations and Standards,” Biocompatibility and Performance of Medical Devices, J.-P. Boutrand , ed., Woodhead Publishing, Oxford, UK.
U.S. Food and Drug Administration, 2014, “ Methods for Thrombogenicity Testing,” Public Workshop, U.S. Food and Drug Administration, White Oak Campus, Silver Spring, MD, Apr. 14.
ISO/DIS, 2016, “ Biological Evaluation of Medical Devices—Part 4: Selection of Tests for Interactions With Blood,” Draft International Standard, International Organization for Standardization, Geneva, Switzerland, Standard No. ISO/DIS 10993-4.
Krajewski, S. , Neumann, B. , Kurz, J. , Perle, N. , Avci-Adali, M. , Cattaneo, G. , and Wendel, H. P. , 2015, “ Preclinical Evaluation of the Thrombogenicity and Endothelialization of Bare Metal and Surface-Coated Neurovascular Stents,” Am. J. Neuroradiol., 36(1), pp. 133–139. [CrossRef]
van Oeveren, W. , Tielliu, I. F. , and de Hart, J. , 2012, “ Comparison of Modified Chandler, Roller Pump, and Ball Valve Circulation Models for In Vitro Testing in High Blood Flow Conditions: Application in Thrombogenicity Testing of Different Materials for Vascular Applications,” Int. J. Biomater., 2012(2012), p. 673163. [PubMed]
Walker, E. K. , Nauman, E. A. , Allain, J. P. , and Stanciu, L. A. , 2015, “ An In Vitro Model for Preclinical Testing of Thrombogenicity of Resorbable Metallic Stents,” J. Biomed. Mater. Res. A., 103(6), pp. 2118–2125. [CrossRef] [PubMed]
Sukavaneshvar, S. , 2016, “ Device Thrombosis and Pre-Clinical Blood Flow Models for Assessing Antithrombogenic Efficacy of Drug-Device Combinations,” Adv. Drug Delivery Rev. (in press).
Hastings, S. M. , Deshpande, S. R. , Wagoner, S. , Maher, K. , and Ku, D. N. , 2016, “ Thrombosis in Centrifugal Pumps: Location and Composition in Clinical and In Vitro Circuits,” Int. J. Artif. Organs, 39(4), pp. 200–204. [CrossRef] [PubMed]
Miller, R. , Fainerman, V. B. , Wustneck, R. , Kragel, J. , and Trukhin, D. V. , 1998, “ Characterization of the Initial Period of Protein Adsorption by Dynamic Surface Tension Measurements Using Different Drop Techniques,” Colloids Surf. A, 131(1–3), pp. 225–230. [CrossRef]
Grove, K. , Deline, S. , Schatz, T. , Howard, S. , and Smith, M. E. , 2016, “ Thrombogenicity Testing for Blood Contacting Medical Devices in an In Vitro Ovine Blood Loop,” ASME J. Med. Devices, 10(2), p. 020904. [CrossRef]
Muller, H. R. , and Hinn, G. , 1990, “ Buser, Internal Jugular Venous Flow Measurement by Means of a Duplex Scanner,” J. Ultrasound Med., 9(5), pp. 261–256. [CrossRef] [PubMed]
Byrom, M. J. , Bannon, P. G. , White, G. H. , and Ng, M. K. , 2010, “ Animal Models for the Assessment of Novel Vascular Conduits,” J. Vasc. Surg., 52(1), pp. 176–195. [CrossRef] [PubMed]
Bianco, R. W. , Grehan, J. F. , Grubbs, B. C. , Mrachek, J. P. , Schroeder, E. L. , Schumacher, C. W. , Svendsen, C. A. , and Lahti, M. , 2004, “ Large Animal Models in Cardiac and Vascular Biomaterials Research and Testing,” Biomaterials Science: An Introduction to Materials in Medicine, 2nd ed., B. D. Ratner , A. S. Hoffman , F. J. Schoen , J. E. Lemons , eds., Academic Press, San Diego, CA, pp. 379–395.
Pinto, D. S. , Lorenz, D. P. , Murphy, S. A. , Marble, S. J. , DiBattiste, P. M. , Demopoulos, L. A. , Cannon, C. P. , and Gibson, C. M. , 2003, “ TIMI Study Group. Association of an Activated Clotting Time < or =250 Seconds With Adverse Event Rates After Percutaneous Coronary Intervention Using Tirofiban and Heparin (A TACTICS-TIMI 18 Substudy),” Am. J. Cardiol., 91(8), pp. 976–978. [CrossRef] [PubMed]
St. Jude Medical, 2016, “ St. Jude Medical Complaint Handling Database; Search Performed for Thromboembolic Events for the St. Jude Medical™ Response™ Electrophysiology Catheter May 2011–April 2016,” St. Jude Medical Data Analytics, St. Paul, MN, Internal Communication.


Grahic Jump Location
Fig. 1

Two operating blood loops are shown with the peristaltic head from each pump: loop 1 has three devices deployed (positions indicated by overlaid white bars) and loop 2 has one device (position also depicted by overlaid white bars). The total length of the tubing for each loop is 140 cm; the two linear segments (outflow and return) are ∼40 cm in length. The radius of the tubing turn within the pump head is ∼7.5 cm, and the radius of the opposite turn in the loop (outside the pump) is ∼10 cm.

Grahic Jump Location
Fig. 2

Activated clotting times from the blood loop preblood draw, at the time of loop filling and after 4 h of blood loop pumping. Normal range for ACT from APS animals is 110 ± 20.2 s. ANOVA shows that there is an effect of time on overall ACT scores, and a post-hoc analysis using Dunnett's multiple comparisons test showed differences between baseline ACT levels and both the preheparinization and the 4 h values (P < 0.05).

Grahic Jump Location
Fig. 3

Test and control articles in situ after 4 ± 0.25 h in the flowing blood loop. (a)–(c) Examples of the negative control, the positive control, and the test article with the associated thrombogenicity scores. The tubing segments of the loop were drained of blood, opened longitudinally, rinsed with saline, and opened to view. Digital images were obtained prior to removal of the devices for scoring.

Grahic Jump Location
Fig. 4

Fluoroscopic images show sheep jugular veins (a) with deployed radiopaque test articles (arrows) and (b) iliac veins both with no apparent disruptions due to anatomy or venous valves. A radiolucent control catheter was deployed in a right iliac vein (inset in b) showing placement of a test article in the vein. All the images were taken with the aid of intravenously injected IsoVue™ angiogenic contract solution to enhance the ability to demonstrate and qualitatively assess blood flow.

Grahic Jump Location
Fig. 5

Activated clotting times from the in vivo NAVI model taken at baseline, postimplantation, and pretermination (4 ± 0.25 hr after implant). ANOVA shows no effect of time on the overall ACT scores, and a post-hoc analysis did not detect any differences between baseline and either pre-implantation or pretermination values.

Grahic Jump Location
Fig. 6

Test and control articles are shown in situ after dissection and removal of the implanted veins. (a)–(c) Examples of the negative control, the positive control, and the test article, respectively, in situ after removal of the isolated vein and longitudinal opening of the entire implant site. The thrombogenicity scores for these representative samples were 0, 5, and 1 for the negative control, the positive control, and the test article, respectively. These implant sites were rinsed in situ, and digital images were obtained. The devices were removed and scored for thrombogenicity.

Grahic Jump Location
Fig. 7

Representative examples of scoring in the positive and negative controls (a) and in the test articles (b). As described in Table 1, the scores are based on overall surface area of thethrombus covering the device. Thrombus thickness was not accounted for in this scoring. The arrows and circle indicate examles of insertion site thrombus, which was excluded from the scoring.

Grahic Jump Location
Fig. 8

Thrombogenic activity of positive and negative controls in the in vitro blood loop assay. This chart shows all the data with medians for each group: negative control median = 0 and positive control median = 5. The positive and negative control articles used in the head-to-head comparison assay were also used in previous routine in vitro thrombogenicity assessments assays testing a variety of other catheter-type test articles. Although the test article data are not shown, the cumulative results of the positive and negative controls are shown here. Seventeen sets of blood loop assays were performed with up to three positive and three negative controls per assay. Overall, this compilation contains results for 47 positive and negative controls run over a period of 13 months. Using the one-tailed Mann–Whitney test for nonparametric data, the median values for the two groups are significantly different (p < 0.0001).

Grahic Jump Location
Fig. 9

Comparisons of thrombogenicity score of controls and test articles in the blood loop and the NAVI model (sheep numbers 15S1023–15S1029) for each individual animal are shown. This summary data is provided to compare individual animal results between the two assays. With the low number of data points within each individual, statistical analyses are not appropriate.

Grahic Jump Location
Fig. 10

Cumulative scores of the positive and negative scores (median with interquartile range) are shown comparing the in vitro blood loop to the results from the in vivo NAVI model. Both assays show a significant difference between the both negative control and test article medians compared to the positive control median (P < 0.05). There is no significant difference between the negative control median and the test article median in either assay.



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In