Open Access
Add to BibSonomyAbstract
A blood clot (thrombus) that becomes lodged in the arterial network supplying the brain can cause an ischemic stroke, depriving the brain of oxygen and often resulting in permanent disability. As an alternative to conventional clot-dissolving drug treatment, we are developing an intravascular laser-activated therapeutic device using shape memory polymer (SMP) to mechanically retrieve the thrombus and restore blood flow to the brain. Thermal imaging and computer simulation were used to characterize the optical and photothermal behavior of the SMP microactuator. Deployment of the SMP device in an in vitro thrombotic vascular occlusion model demonstrated the clinical treatment concept.
©2005 Optical Society of America
1. Introduction
Each year ischemic stroke strikes more than 600,000 people in the United States [1]. Roughly two-thirds of those ischemic strokes are caused by thrombotic vascular occlusion (formation or lodging of a blood clot) in the arterial network supplying the brain [2]. Interruption of blood flow in these vessels deprives the brain of oxygen, often resulting in permanent disability [1,3]. Traditionally, treatment has been limited to administration of recombinant tissue plasminogen activator (t-PA), a thrombolytic (clot-dissolving) drug which is infused over a 1-hour period into the systemic circulation. Clinical protocol requires that treatment may only be initiated within 3 hours of the onset of symptoms [4]. Unfortunately, most patients arrive at the hospital after the 3-hour treatment window has expired [1]. Due to delays in obtaining treatment and strict exclusion criteria aimed to temper the risk of intracranial hemorrhage, a potential side effect of t-PA [2], less than 10% of ischemic stroke patients are treated with t-PA [5].
Alternative treatment modalities have exhibited the potential to overcome the drawbacks of systemic infusion of t-PA [6]. In an effort to extend the 3-hour treatment window of conventional t-PA therapy, clinical studies investigating localized intra-arterial thrombolytic drug treatment demonstrated neurological benefit in patients treated up to 6 hours after symptom onset [7,8]; however, the risk of cerebral hemorrhage persisted [9]. More recently, the Food and Drug Administration approved the use of an intravascular device to mechanically retrieve the thrombus up to 8 hours after the onset of stroke [10]. Using a helical nitinol wire to physically capture the clot under fluoroscopic guidance, blood flow is restored in a matter of minutes rather than hours, unlike the process of chemically dissolving the clot. In addition to providing faster treatment, the significantly longer treatment window combined with elimination of the risk of intracranial hemorrhage should increase the number of patients eligible for treatment. Other mechanical thrombectomy methods, including photoacoustic thrombolysis [11], capture of the clot using a snare device [12] or a nitinol basket [13], and a saline jet device to fragment and aspirate the clot [14], are under investigation. In addition to expanding the patient population eligible for treatment, these and other non-pharmacological means of vessel reperfusion may offer a faster, safer alternative to thrombolytic drug therapy after ischemic stroke.
We are developing an intravascular device using shape memory polymer (SMP) to mechanically retrieve the thrombus and restore blood flow following ischemic stroke. SMP has been identified as a material suitable for therapeutic microactuator fabrication [15,16]. This unique polyurethane-based material can maintain a stable secondary shape and, upon controlled heating, will return to its primary shape. In our device, the primary shape is a tapered corkscrew and the secondary shape is a straight rod. The proposed thrombectomy concept is illustrated in Fig. 1. We propose to deliver the SMP microactuator in its secondary straight rod form through a catheter distal to the vascular occlusion, which is located by fluoroscopy routinely used in stroke treatment. The microactuator, which is mounted on the end of an optical fiber, is then transformed into its primary corkscrew shape by laser heating. Once deployed, the microactuator is retracted and the captured thrombus is removed from the body to restore blood flow. In this paper, we describe the fabrication techniques and the optical, photothermal, and thermomechanical properties of the device. We also demonstrate clinical application of the device using an in vitro thrombotic vascular occlusion model.
Fig. 1. Depiction of endovascular thrombectomy using the laser-activated SMP microactuator coupled to an optical fiber. (a) In its secondary straight rod form, the microactuator is delivered through a catheter distal to the thrombotic vascular occlusion. (b) The microactuator is then transformed into its primary corkscrew form by laser heating. (c) The deployed microactuator is retracted to capture the thrombus.
2. Materials and methods
2.1 Shape memory polymer
SMP has the ability to resume a primary or “memorized” shape after being deformed into a secondary shape. The secondary shape is maintained until the SMP is heated above its soft phase glass transition temperature (Tgs). Our device was fabricated using commercially available thermoplastic SMP materials, MM6520 (nominal Tgs = 65°C) and MM5520 (nominal Tgs = 55°C) (DiAPLEX, a subsidiary of Mitsubishi Heavy Industries Ltd., Tokyo, Japan). While heated above its highest glass transition temperature (approximately 130°C), the SMP is formed into the primary shape and cooled to memorize the shape. At a temperature above Tgs, the SMP can be deformed into a secondary shape and then cooled to maintain the new shape. The primary shape is recovered by heating again to Tgs.
2.2 Device fabrication
Fig. 2. SMP microactuator coupled to an optical fiber shown in its (a) secondary straight rod and (b) primary corkscrew forms. The maximum diameter of the SMP corkscrew is approximately 3 mm.
The device consists of an SMP microactuator coupled to an optical fiber for delivery of 810 nm laser light for activation. The device is shown in its straight and corkscrew forms in Fig. 2. The fabrication process began by extruding the raw SMP material (MM6520) into a strand with circular cross-section approximately 380 μm in diameter. A cylindrical rod segment was soaked in methanol for 1 hour and vacuum dried at 50°C for 24 hours. The rod was then dip-coated in a solution consisting of SMP material (MM5520) and EpolightTM 4121 platinum dye (Epolin, Inc., Newark, NJ) dissolved in tetrahydrofuran (THF) and vacuum dried at 50°C, yielding a coating approximately 10–15 μm thick (measured using calipers). The dye, which has an absorbance peak centered at 803 nm (full-width at half-maximum is ~90 nm) [17], was used to enhance the heating efficiency of the laser. The concentration of dye in the coating solution was approximately 114 parts dye per 106 parts SMP (by weight). Localization of the dye in a thin outer layer enabled more light to reach (and thus heat) the distal end of the microactuator than if the dye had been distributed throughout the SMP volume.
In one end of the rod, a socket (diameter = 279 μm, depth = 1 mm) was drilled coaxially with the longitudinal axis of the rod using a Levin® lathe (Louis Levin and Son, Inc., Santa Fe Springs, CA). The socket was filled with a transparent ultraviolet (UV) light curable epoxy (EPO-TEK® OG603, Epoxy Technology, Inc., Billerica, MA) with a refractive index of 1.49. A multimode ultra low-OH 200 μm core diameter (240 μm diameter including the outer polyimide buffer) step-index silica core/silica clad optical fiber (FIP200220240, Polymicro Technologies, Phoenix, AZ) was cleaved on one end (core refractive index = 1.4571, numerical aperture = 0.22). The other end was coupled to a continuous-wave 810 nm diode laser (UM4500 Dental, Unique Mode, Jena, Germany). The cleaved end of the fiber was inserted into the drilled socket in the SMP rod and the epoxy was cured using a UV light source (Green Spot, American Ultraviolet Company, Lebanon, IN). Epoxy that was forced out of the socket as the fiber was inserted formed a collar around the fiber at the SMP shoulder, providing extra strength to the joint. A microscope image of the joint is shown in Fig. 3.
Fig. 3. Socket joint between the optical fiber and the SMP microactuator. The polyimide buffer was burned off and the optical fiber was cleaved prior to insertion into the epoxy-filled socket. The epoxy, whose refractive index is between that of the optical fiber core and that of the SMP, was chosen to provide high coupling efficiency while maintaining a strong bond.
After the epoxy cured, the SMP rod was wrapped around a custom-made conical mandrel (maximum diameter = 3 mm) to set the primary corkscrew shape. The corkscrew shape was chosen to maintain the waveguiding ability of the SMP microactuator while providing a means of capturing a thrombus. The mandrel, shown in Fig. 4, was fabricated from temperature-resistant plastic (Vespel SP-21, DuPont, Wilmington, DE) using a 4-axis milling machine (MDX-650, Roland DGA Corp., Irvine, CA). An aluminum cap placed over the mandrel captured the wrapped SMP rod and maintained the corkscrew form as it was heated using a hot air system (Model 210-A, Beahm Designs, Los Gatos, CA) at 150°C for 15 minutes and then cooled to room temperature to set the primary shape. The SMP corkscrew was finally dip-coated in Sylgard® 184 silicone elastomer (Dow Corning, Midland, MI) to provide a low-index cladding (refractive index = 1.41) for consistent waveguiding regardless of the physiological environment. Prior to deployment, the SMP corkscrew was placed in a hot air stream at approximately 75°C and manually straightened into its secondary rod shape, and then cooled to set the secondary shape.
2.3 Optical characterization
The refractive index of the SMP (MM5520) was measured at 589 nm using a digital refractometer (Abbe Mark II, Reichert Analytical Instruments, Depew, NY) and corrected to 810 nm (laser wavelength) using Herzberger’s dispersion equation [18]:
where n is the refractive index, λ. is the wavelength in microns, and A through E are empirically determined constants. The equation was first fit to refractive index dispersion data for styrene acrylonitrile [19], an optical polymer with refractive index similar to the SMP, to obtain values for the constants. These values were then used to estimate the refractive index of the SMP at 810 nm, with the exception of A which was changed such that the dispersion curve was shifted vertically to intercept the SMP refractive index data point at 589 nm. The estimated SMP refractive index was used to calculate the coupling loss between the optical fiber and the SMP, and was also used in the computer simulation of laser light propagation through the SMP corkscrew. The same value of the refractive index was used for both the MM5520 and MM6520 SMP materials.
Coupling loss due to Fresnel reflection at the optical fiber-epoxy boundary and the epoxy-SMP boundary was calculated using the formula
where nf, ne, and ns are the refractive indices of the optical fiber core, epoxy, and SMP, respectively. The squared terms represent the fraction of light reflected at each boundary.
In its straight form, the cylindrical SMP microactuator transmits light like an optical fiber. However, as the microactuator transforms into its primary corkscrew form, the light paths of the propagating rays are altered. Because laser heating must continue as the microactuator changes shape to achieve complete actuation, the corkscrew shape must permit light to reach the distal end of the microactuator. ZEMAX optical design software (ZEMAX Development Corp., Bellevue, WA) was used to model the light transmission from the optical fiber through the SMP microactuator in its corkscrew form, representing the worst-case scenario for light propagation. Absorption of 810 nm light by the SMP was assumed to be negligible for our purposes, and hence was not included in the simulation. Scattering due to imperfections in the SMP was not included. Also, since the laser absorbing dye was added at such a low concentration and was confined to a very thin outer layer, it was not included in the simulation. The total power and power per unit area (irradiance) were computed and the cross-sectional spatial light distribution was determined at several locations along the corkscrew to ascertain the effects of the change in shape.
Laser light leaking from the SMP microactuator was measured using an integrating sphere (Model 70491, Oriel Instruments, Stratford, CT) with a photodiode detector (PIN-10DF, UDT Sensors, Inc., Hawthorne, CA). These measurements were done to estimate the absorption losses due to the SMP and dye, and to determine the relative fractions of light leaking radially from the microactuator and transmitted to the distal end of the microactuator.
2.4 Photothermal characterization
The temperature of the SMP device was monitored during laser-activation in air using a cooled infrared video camera (Thermacam PM250, Inframetrics, Billerica, MA) fitted with a close-up lens. Changes in the SMP temperature were observed as the microactuator transformed from a straight rod to a corkscrew.
2.5 Thrombotic stroke model
A carotid artery model with a 60° bifurcation (Shelley Medical Imaging Technologies, Ontario, Canada) was used to test the feasibility of the SMP device for intravascular thrombectomy. The experimental setup is shown in Fig. 5. Water at body temperature (37°C) was pumped into the main vessel segment (Segment A, lumen diameter = 8 mm) using a peristaltic pump (Model 505Du, Watson-Marlow, Ltd., Cornwall, England) at a flow volume of 33 ml/min as measured by a flow probe connected to a dual-channel small animal blood flow meter (Model T206, Transonic Systems, Inc., Ithaca, NY). With no obstructing thrombus, the flow volume exiting the narrower vessel after the bifurcation (Segment B, lumen diameter = 5 mm) was measured as 7 ml/min. The flow volume exiting the remaining vessel segment (Segment C, lumen diameter = 7 mm) was, therefore, 26 ml/min (not measured).
Fig. 5. Flow system used to test feasibility of the SMP device for intravascular thrombectomy. Flow is in the counterclockwise direction. A: main vessel of carotid bifurcation model; B and C: branches of carotid bifurcation model; TB: Touhy Borst valve; FP: flow probe; FM: flow meter; PP: peristaltic pump; H2O: water reservoir.
An artificial thrombus (made in-house) based on an acrylamide hydrogel was inserted into the main vessel (Segment A) and became lodged in Segment B immediately distal to the bifurcation point, mimicking a thrombotic stroke event. Red food coloring was added to the otherwise transparent hydrogels to aid visualization. The gels had elastic moduli in the range of 2000 to 3000 Pa as measured by dynamic mechanical testing at 1 Hz and 10% strain using a rheometer (Ares LS2, TA Instruments, New Castle, DE), which is similar to the elastic moduli of actual blood clots reported in the literature [20,21,22].
A 4 Fr catheter (Radiofocus Glidecath, Terumo Medical Corp., Somerset, NJ) was inserted into the main vessel segment via a Touhy Borst valve and was positioned proximal to the thrombus. The SMP microactuator was delivered in its secondary straight rod form through the catheter and pushed distal to the thrombus. After laser activation, the SMP microactuator was retracted to capture the thrombus.
2.6 Estimation of blood temperature
To determine the thermal impact of the device during clinical deployment, the temperature rise ΔT of the surrounding blood induced by absorption of the light escaping from the device was estimated using the equation
where P is the absorbed laser power in Watts, t is the laser duration in seconds, c is the specific heat capacity of the blood in J/g-°C, ρ is the density of blood in g/cm3, and V is the volume of blood in the vicinity of the SMP microactuator in cm3. This estimate assumes the heat load is evenly distributed in the volume of blood surrounding the device and does not include conductive heat losses to the blood and arterial wall adjacent to the heated volume.
3. Results
3.1 Optical and photothermal behavior
Fig. 6. Computer simulation of laser light propagation through the SMP corkscrew in air. The laser power exiting the optical fiber is 1.00 W. Virtual detectors (a-p) indicate the spatial light distribution (irradiance) at various cross-sectional locations along the corkscrew. Total power and peak irradiance are noted for each detector. The first detector (a) is positioned at the optical fiber tip.
The refractive index of the SMP was measured as 1.5823 at 589 nm and was estimated to be 1.57 at 810 nm using Eq. 1. Using Eq. 2, the coupling loss was calculated to be 0.08%. According to this calculation, over 99.9% of the laser light is transmitted from the optical fiber into the SMP microactuator (assuming zero absorption by the epoxy as noted by the manufacturer).
Light transmission through the SMP microactuator in its corkscrew form is illustrated by the computer simulation in Fig. 6. This simulation shows the SMP microactuator in an air environment. The total power drops with distance due to light leakage from the corkscrew turns as the light paths exceed the critical angle for total internal reflection; approximately 93% of the light is transmitted to the distal end of the corkscrew and 7% of the light is lost due to leakage. However, the irradiance increases as the light enters the first turn of the corkscrew and becomes confined to the outer region. Despite the steady drop in total power due to light leakage, the irradiance remains relatively constant with distance after peaking in the first turn. This optical behavior is confirmed in the thermal camera video of laser actuation shown in Fig. 7. As the corkscrew is formed, the temperature increases due to increasing irradiance as predicted by the computer simulation. The highest temperatures correspond to the regions of highest irradiance: the optical fiber tip and the first turn of the corkscrew. The temperature is fairly constant after the first turn.
For comparison, another simulation was conducted for the SMP microactuator in a blood environment (not shown in figures). Though only 43% of the light is transmitted to the distal end of the corkscrew (57% loss due to leakage) as a result of the relatively high refractive index of blood (n=1.38 [23]), the irradiance followed a similar trend as it did in an air environment; the light became confined to the outer region causing the irradiance to peak in the first turn (peak irradiance = 4.7 kW/cm2) and remain fairly constant thereafter (peak irradiance around 3.3 kW/cm2).
Fig. 7. (1.2 MB) Real-time thermal camera video of laser actuation of the SMP microactuator in air. The laser power was 0.60 W. The maximum diameter of the SMP corkscrew is approximately 3 mm (see Fig. 2).
Integrating sphere measurements indicated that most (approximately 92%) of the laser light entering the SMP microactuator escaped (including the light leaking radially from the microactuator and the light transmitted from the distal end of the microactuator). Therefore, only around 8% of the light was absorbed by the SMP and dye. Further measurements using an SMP device without the dye layer showed that about 5% was absorbed by the SMP and 3% was absorbed by the dye. Of the light escaping from the device without dye (comparable to the ZEMAX simulation), approximately 37% leaked radially from the corkscrew. This value is higher than that yielded by the simulation (7%) due to scattering from imperfections in the SMP device which were not modeled.
3.2 Device deployment
Prior to attempting thrombectomy in the vascular model, the SMP device was deployed in static water (Fig. 8). The water temperature was approximately 37°C (human body temperature) and the laser power was set to 4.89 W. Cooling by the water required a much higher laser power to achieve thermal actuation than was required for actuation in air. Actuation was complete within 3 seconds after the laser was turned on.
Fig. 8. (448 KB) Real-time video of laser actuation of the SMP device in static water at body temperature.
In vitro thrombectomy of the artificial blood clot was performed in the bifurcated vessel model (Fig. 9). The SMP device was pushed through the occlusion in its straight form and then actuated at a laser power of 4.89 W, causing the SMP microactuator to resume its corkscrew form. After the laser was turned off, the SMP device was retracted to capture the clot and pull it back against the flow past the bifurcation to the catheter.
Fig. 9. (2.2 MB) Real-time video of in vitro thrombectomy using the SMP device in a bifurcated vessel model. Water flow is from left to right.
3.3 Blood temperature
Equation 3 was used to estimate the temperature rise of the surrounding blood during clinical deployment caused by blood absorption of the laser light escaping from the SMP device. Based on the integrating sphere measurements of escaping light and the laser power used to activate the device in body temperature water, and assuming all of the light is absorbed by the blood, the laser power P was set to 4.5 W (92% of 4.89 W). The laser duration t was set to 3 seconds (time required to achieve complete actuation in water at body temperature). The specific heat capacity c and density ρ of blood were set to 3.85 J/g-°C and 1.05 g/cm3, respectively [24]. The blood volume surrounding the SMP microactuator was calculated to be 0.28 cm2 for a blood vessel lumen diameter of 0.6 cm (length of vessel occupied by SMP corkscrew = 1 cm). Using these values, the temperature rise of the blood near the device is approximately 12°C. Adding this value to 37°C (body temperature) yields the final blood temperature reached during laser activation, 49°C.
Tissue damage is a kinetic process related to a combination of temperature rise and duration of that temperature rise. Irreversible thermal tissue damage occurs after 3 seconds for a sustained temperature of ~62°C [25]. A temperature of 49°C can be sustained for ~103 seconds without causing permanent damage [25].
4. Conclusion
We have designed, built, and tested a prototypical optomechanical stroke treatment device using shape memory polymer and an infrared diode laser coupled together by an optical fiber. Computer simulation and thermal imaging were used to characterize the optical and photothermal properties, aiding the initial design effort. Preliminary feasibility was demonstrated in vitro by the successful capture of an artificial blood clot within a water-filled bifurcated vessel model under flow.
Though the device served to illustrate the basic treatment concept, further modifications are necessary prior to deployment in vivo, including the incorporation of (i) a torqueable sheath around the optical fiber to allow the surgeon to maneuver the device through the tortuous paths of the neurovasculature, (ii) radio-opacity for visualization under fluoroscopic guidance, and (iii) a means of securing the captured clot after retraction for withdrawal from the body. In addition, the diameter of the SMP corkscrew may have to be scaled down slightly for human use, depending on the clot location. Notwithstanding the lack of these design elements in the initial prototype, the laser-activated SMP thrombectomy device is a promising tool for treating ischemic stroke without the need for systemic infusion of thrombolytic drugs.
Acknowledgments
The authors thank V. Sperry for helping to fabricate the SMP devices and D. Schumann for providing the SMP refractometer data. This work was performed under the auspices of the U.S. Department of Energy by University of California, Lawrence Livermore National Laboratory under Contract W-7405-ENG-48 and supported by the National Institutes of Health/National Institute of Biomedical Imaging and Bioengineering, Grant R01EB000462.
References and links
1. American Heart Association, Heart Disease and Stroke Statistics-2005 Update (American Heart Association, Dallas, 2005), http://www.americanheart.org/downloadable/heart/1105390918119HDSStats2005Update.pdf.
2. D. O. Kessel and J. V. Patel, “Current trends in thrombolysis: Implications for diagnostic and interventional radiology,” Clin. Radiol. 60, 413–424 (2005). [CrossRef] [PubMed]
3. Center for Disease Control and Prevention, “Prevalence of disabilities and associated health conditions among adults-United States, 1999,” Morbidity and Mortality Weekly Report50, 120–125 (2001). [PubMed]
4. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group., “Tissue plasminogen activator for acute ischemic stroke,”N. Engl and J. Med, 333, 1581–1588 (1995).
5. D. M. Bravata, “Intravenous thrombolysis in acute ischaemic stroke: optimising its use in routine clinical practice,” CNS Drugs 19, 295–302 (2005). [CrossRef] [PubMed]
6. M. C. Leary, J. L. Saver, Y. P. Gobin, R. Jahan, G. R. Duckwiler, F. Vinuela, C. S. Kidwell, J. Frazee, and S. Starkman, “Beyond tissue plasminogen activator: mechanical intervention in acute stroke,” Ann. Emerg. Med. 41, 838–846 (2003). [CrossRef] [PubMed]
7. G. J. del Zoppo, R. T. Higashida, A. J. Furlan, M. S. Pessin, H. A. Rowley, and M. Gent, “PROACT: a phase II randomized trial of recombinant pro-urokinase by direct arterial delivery in acute middle cerebral artery stroke.Stroke 29, 4–11 (1998). [CrossRef] [PubMed]
8. A. Furlan, R. Higashida, L. Wechsler, M. Gent, H. Rowley, C. Kase, M. Pessin, A. Ahuja, F. Callahan, W. M. Clark, F. Silver, and F. Rivera, “Intra-arterial pro-urokinase for acute ischemic stroke. The PROACT II study: a randomized controlled trial,” JAMA 282, 2003–2011 (1999). [CrossRef] [PubMed]
9. C. S. Kase, A. J. Furlan, L. R. Wechsler, R. T. Higashida, H. A. Rowley, R. G. Hart, G. F. Molinari, L. S. Frederick, H. C. Roberts, J. M. Gebel, C. A. Sila, G. A. Schulz, R. S. Roberts, and M. Gent, “Cerebral hemorrhage after intra-arterial thrombolysis for ischemic stroke: the PROACT II trial,” Neurology 57, 1603–1610 (2001). [CrossRef] [PubMed]
10. Y. P. Gobin, S. Starkman, G. R. Duckwiler, T. Grobelny, C. S. Kidwell, R. Jahan, J. Pile-Spellman, A. Segal, F. Vinuela, and J. L. Saver, “MERCI 1: A phase 1 study of mechanical embolus removal in cerebral ischemia,” Stroke 35, 2848–2854 (2004). [CrossRef] [PubMed]
11. A. Berlis, H. Lutsep, S. Barnwell, A. Norbash, L. Wechsler, C. A. Jungreis, A. Woolfenden, G. Redekop, M. Hartmann, and M. Schumacher, “Mechanical thrombolysis in acute ischemic stroke with endovascular photoacoustic recanalization,” Stroke 35, 1112–1116 (2004). [CrossRef] [PubMed]
12. C. W. Kerber, J. D. Barr, R. M. Berger, and B. W. Chopko, “Snare retrieval of intracranial thrombus in patients with acute stroke,” J. Vasc. Interv. Radiol. 13, 1269–1274 (2002). [CrossRef] [PubMed]
13. T. E. Mayer, G. F. Hamann, and H. J. Brueckmann, “Treatment of basilar artery embolism with a mechanical extraction device: necessity of flow reversal,” Stroke 33, 2232–2235 (2002). [CrossRef] [PubMed]
14. R. J. Bellon, C. M. Putman, R. F. Budzik, R. S. Pergolizzi, G. F. Reinking, and A. M. Norbash, “Rheolytic thrombectomy of the occluded internal carotid artery in the setting of acute ischemic stroke,” AJNR Am. J. Neuroradiol. 22, 526–530 (2001). [PubMed]
15. M. F. Metzger, T. S. Wilson, D. Schumann, D. L. Matthews, and D. J. Maitland, “Mechanical properties of mechanical actuator for treating ischemic stroke,” Biomedical Microdevices 4, 89–96 (2002). [CrossRef]
16. D. J. Maitland, M. F. Metzger, D. Schumann, A. Lee, and T. S. Wilson, “Photothermal properties of shape memory polymer micro-actuators for treating stroke,” Las. Surg. Med. 30, 1–11 (2002). [CrossRef]
17. Data from Epolin website, http://www.epolin.com/p4000.html.
18. W. J. Tropf, M. E. Thomas, and T. J. Harris, “Properties of crystals and glasses,” in Handbook of Optics Volume II,M. Bass, E. W. Van Stryland, D. R. Williams, and W. L. Wolfe, eds. (McGraw-Hill, Inc., New York, 1995), pp. 33.3–33.101.
19. J. D. Lytle, “Polymeric optics,” in Handbook of Optics Volume II, M. Bass, E. W. Van Stryland, D. R. Williams, and W. L. Wolfe, eds. (McGraw-Hill, Inc., New York, 1995), pp. 34.1–34.21.
20. M. E. Carr, A. Krishnaswami, and E. J. Martin, “Platelet contractile force (PCF) and clot elastic modulus (CEM) are elevated in diabetic patients with chest pain,” Diabet. Med. 19, 862–866 (2002). [CrossRef] [PubMed]
21. M. E. Carr Jr. and S. L. Zekert, “Abnormal clot retraction, altered fibrin structure, and normal platelet function in multiple myeloma,” Am. J. Physiol. 266, H1195–H1201 (1994). [PubMed]
22. D. F. Brophy, E. J. Martin, T. W. B. Gehr, and M. E. Carr Jr., “A hypothesis-generating study to evaluate platelet activity in diabetics with chronic kidney disease,” Thrombosis Journal 3, 3 (2005), http://www.thrombosisjournal.com/content/3/1/3. [CrossRef] [PubMed]
23. D. K. Sardar and L. B. Levy, “Optical properties of whole blood,” Lasers Med Sci 13, 106–111 (1998). [CrossRef]
24. Biomaterials Science: An Introduction to Materials in Medicine, B. D. Ratner, A. S. Hoffman, F. J. Schoen, and J. E. Lemons, eds. (Academic Press, San Diego, 1996), p. 469.
25. M. J. C. van Gemert, A. J. Welch, J. W. Pickering, and O. T. Tan, “Laser treatment of port wine stains,” in Optical-Thermal Response of Laser-Irradiated Tissue, A. J. Welch and M. J. C. van Gemert, eds. (Plenum Press, New York, 1995), pp. 789–829.
