Improving Stroke Treatment Using Magnetic Nanoparticle Sensors to Monitor Brain Thrombus Extraction

Abstract

(1) Background: Mechanical thrombectomy (MT) successfully treats ischemic strokes by extracting the thrombus, or clot, using a stent retriever to pull it through the blood vessel. However, clot slippage and/or fragmentation can occur. Real-time feedback to a clinician about attachment between the stent and clot could enable more complete removal. We propose a system whereby antibody-targeted magnetic nanoparticles (NPs) are injected via a microcatheter to coat the clot, oscillating magnetic fields excite the particles, and a small coil attached to the catheter picks up a signal that determines the proximity of the clot to the stent. (2) Methods: We used existing simulation code to model the signal from NPs distributed on a hemispherical clot with three orthogonally applied magnetic fields. An in vitro apparatus was built that applied fields and read out signals from a 1.5 mm pickup coil at a variable distance and orientation angle from a sample of 100 nm iron oxide core/shell NPs. (3) Results: Our simulations suggest that the sum of the voltages induced in the pickup coil from three orthogonal applied fields could localize a clot to within 180 µm, regardless of the exact orientation of the pickup coil, with further precision added via rotation-correction formulae. Our experimental system validated simulations; we estimated an in vitro distance recovery precision of 41 µm with a pickup coil 1 mm from the clot. (4) Conclusions: Magnetic NP sensing could be a safe and real-time method to estimate whether a clot is attached to the stent retriever during MT.

Keywords: clot; magnetic dipole field; magnetic nanoparticles; mechanical thrombectomy; simulations; stroke; thrombus.

Figure 1

Schematic of novel sensing modality. Three orthogonal pairs of driving magnetic field coils (yellow), centered approximately around the thrombus. The thrombus (clot) is coated by injected targeted NPs. The resulting NP magnetization signal is picked up by a fine-gauge magnetization pickup coil attached to the stent and inserted into the artery (yellow). Thrombus breakage or slippage is detectable via a drop in the pickup coil signal. The change in signal is high with any movement of the thrombus relative to the pickup coil because the signal drops as one over the distance cubed.

Figure 2

Schematics of (a) NP injection and thrombus coating via microcatheter and (b) thrombus slippage with applied field orientation. The fine-gauge probe (pickup coil) (green) is near the thrombus (shaded). In (a), only one pair out of the three pairs of applied field coils in the full setup is shown.

Figure 3

Model assumption of 25 NPs symmetrically distributed on half of a spherical thrombus with a 1 mm diameter. Vertices (dots) indicate the locations of Amperian loops that simulate the NP magnetic field.

Figure 4

Schematic of geometry of NPs, the induced field, and pickup coil probe (not to scale). Only one pair out of the three pairs of applied field coils in the full setup is shown.

Figure 5

Photographs of the apparatus. (Left) A 3 mm diameter hand-wound pickup coil glued to a thin carbon fiber rod probe (black) and NP solution (black) inside a standard test tube. The setup exists inside the applied field coils and is mounted on a 2D translation stage (not shown). (Right) A 1.5 mm diameter pickup coil hand-wound on a 1 mm carbon fiber rod with a 36-gauge AWG fine copper wire (diameter 0.1 mm).

Figure 6

Simulation of the NP magnetization induced by a combination of 10 mT, 1 kHz sinusoidal AC field (z-direction), and a 1 mT DC field (y-direction), oriented perpendicular to each other. The voltage produced in the pickup coil resulting from the magnetization shown in panel (a) is shown in panel (b). For thrombus-sensing applications, we only care about the amplitude of the voltage, not the shape of the signal.

Figure 7

Simulated sum of fluxes, Φ_3, (a) as a function of one-dimensional translation distance, and (b) two-dimensional translation. In panel (a), both an inverse quadratic function with an offset and an inverse cubic function with an offset are fitted to the data. Both fits for the data are decent approximations but neither is perfect. This is expected from the multipole series expansion of a magnetic dipole field as a function of distance, which lacks a closed-form solution.

Figure 8

Simulated error in distance recovery with large changes in pickup orientation. In (a), we see ${\mathsf{\Phi }}_3$ as a function of both distance between thrombus and pickup, and pickup coil orientation angle. We note that changing the orientation angle does not change the ${\mathsf{\Phi }}_3$ as dramatically as does the distance. In (b), we estimate the change in ${\mathsf{\Phi }}_3$ as the coil orientation is changed by 90 degrees about either axis. We find these large rotations corresponds to about a $\pm 180$ micron variation (black lines) in distance. This is our upper-bound estimate for distance recovery error using the sum of 3 fluxes, demonstrating its suitability for orientation-independent distance estimation.

Figure 9

Simulated estimation of the precision of distance recovery from the sum of fluxes with variable pickup orientation angle (hemispherical distribution of NPs). The two sets of black lines demonstrate how the error in distance estimation increases for larger distances where the ${\mathsf{\Phi }}_3$ has dropped significantly. This figure again highlights how the pickup coil orientation angle (shown in the legend) affects ${\mathsf{\Phi }}_3$, but the impact of angle is much gentler than the impact of distance.

Figure 10

Simulated asymmetric, quadraspherical distribution of NPs, panel (a). Simulated flux versus distance when pickup orientation angle changes, panel (b). Upon comparison of (b) with the corresponding results for the hemispherical case (Figure 9), we find that the results are nearly identical. This suggests that the details of the distribution of NPs on the clot do not affect ${\mathsf{\Phi }}_3$ significantly. This robustness is a desirable feature for our distance-sensing application.

Figure 11

Simulated functional dependence of flux on rotation at multiple distances between the clot and the pickup coil. We see that if we normalize ${\mathsf{\Phi }}_3$, the orientation dependence is identical at all distances. This is obvious from the separable nature of the two effects.

Figure 12

Simulated rotation correction. Since orientation and distance are separable, we can explicitly correct for orientation changes using the direct measurement of $\widehat{n}$ discussed above. Panel (b) shows the result of the correction process on the data in (a).

Figure 13

Experimental (in vitro) flux versus distance measurements for (a) 1.5 mm and (d) 3 mm diameter pickup coils. Panel (b) shows two-dimensional translations. Panel (c) shows the error in distance recovery as a function of the distance between the clot and the pickup coil. Panel (a) has an inverse quadratic and an inverse cubic fit, with computed $R^2$ values of 0.9965 and 0.9982, respectively, showing that the inverse cubic fit has a better fit. Panel (d) similarly shows the same two fits with $R^2$ values of 0.992 and 0.997, respectively, again showing that the inverse cubic fit agrees with the data slightly better. We note that visually, the inverse cubic fit is better in the long-distance limit compared to the inverse quadratic fit, consistent with the theoretical expectations for a magnetic dipole field.

Figure 13

Experimental (in vitro) flux versus distance measurements for (a) 1.5 mm and (d) 3 mm diameter pickup coils. Panel (b) shows two-dimensional translations. Panel (c) shows the error in distance recovery as a function of the distance between the clot and the pickup coil. Panel (a) has an inverse quadratic and an inverse cubic fit, with computed $R^2$ values of 0.9965 and 0.9982, respectively, showing that the inverse cubic fit has a better fit. Panel (d) similarly shows the same two fits with $R^2$ values of 0.992 and 0.997, respectively, again showing that the inverse cubic fit agrees with the data slightly better. We note that visually, the inverse cubic fit is better in the long-distance limit compared to the inverse quadratic fit, consistent with the theoretical expectations for a magnetic dipole field.