Histological assessment of thermal damage in the brain following infrared neural stimulation

Abstract

Background: Infrared neural stimulation (INS) is a novel technique for modulating neural function. Its advantages over electrical stimulation include high spatial specificity, lack of electrical artifact and contact-free stimulation. INS acts via a rapid, focal increase in temperature. However, in order to become a viable experimental and therapeutic tool, the safety of INS must be demonstrated.

Objective/hypothesis: Our aim was to determine the upper limit for the radiant exposure of INS in the brain without causing damage, using an INS sequence previously shown to induce both behavioral and electrophysiological effects in rodents and non-human primates.

Methods: We stimulated the brains of anesthetized rodents and two squirrel monkeys using an infrared laser, depositing radiant energies from 0.3 to 0.9 J/cm2 per pulse in 0.5 s-long 200 Hz trains. At the end of the experiment, the animals were euthanized, perfused and the brains processed using standard histological techniques.

Results: Radiant exposures greater than or equal to 0.4 J/cm2 resulted in identifiable lesions in brain sections. The lesions had a shape of a parabola and could further be subdivided into three concentric zones based on the type of damage observed.

Conclusions: The thermal damage threshold following our INS paradigm was between 0.3 and 0.4 J/cm2 per pulse. This value is lower than the one found previously in peripheral nerve. The differences are likely due to the structure of the INS sequence itself, particularly the repetition rate. The results warrant further modeling and experimental work in order to delimit the INS parameter space that is both safe and effective.

Keywords: Infrared neural stimulation; Non-human primate; Thermal damage.

Copyright © 2014 Elsevier Inc. All rights reserved.

Figures

Figure 1

INS pulse sequence. We used the INS pulse sequence shown above throughout the study, with the exception of a few trials using single high energy pulses. It consisted of high-frequency trains of laser pulses expected to modulate brain function followed by periods when the laser was off and the cortex was allowed to recover. To simplify interpretation of results, we kept the timing of the sequence constant and only varied the power (by adjusting the laser diode current) to alter the dose of radiant energy delivered to the cortex.

Figure 2

INS stimulation of the rat brain in vivo. The figure on the right depicts the general arrangement of the stimulation setup. A craniotomy was performed over the right hemisphere and the brain is clearly visible through a transparent silicone membrane secured in place with agar. A 200 micron core optical fiber enclosed in a steel cannula to improve rigidity is positioned over the membrane,. On the right is a magnified image taken after focal INS stimulation, showing three areas subjected to INS. The radiant exposures per pulse were: A- 0.7 J/cm2, B- 0.8 J/cm2, C- 0.5 J/cm2, D- 0.4 J/cm2. Spots A and B clearly show blanching of the brain, indicating the presence of thermally induced damage.

Figure 3

Thermally induced lesion in a rat brain (A). Schematic representation of a thermally induced cortical lesion. Three visually identifiably zones of damage are labeled as 1, 2, 3. Area outside of the lesions is labeled as zone 4. B Stained sections of rat brain showing a thermally induced lesion. (B). Hematoxylin and eosin stained sections reveal a parabola-shaped lesion at the location of stimulation, with pulse energy of 0.8 J/cm2. The three zones of damage are labeled as in (A). (C). The areas marked on (B) are shown on the right at greater magnification. The arrow in panel 4 shows the typical appearance of a neuron outside the lesion area. In 1, 2, 3, neurons have an abnormal appearance (arrows), as described in the text in greater detail.

Figure 4

The spread of thermally induced INS lesions as a function of radiant exposure. (A). Hematoxylin and eosin stained sections of thermally induced lesions following INS in the rat brain. The radiant exposures were 0.4, 0.7 and 0.8 J/cm2 per pulse, as labeled. The panel on the lower right is an overlay of the three exposures, with the photographs colored green, red and blue, respectively. (B). Quantification of the maximal spread below cortical surface of the lesions as a function of radiant exposure.

Figure 5

Thermally induced lesions following INS in the rat brain: high energy single pulses. Shown are hematoxylin and eosin stained sections of a rat brain exposed to single INS pulses of 2 (A) and 5 (B) J/cm2. The lower radiant exposure lesion is similar in appearance to the ones obtained using pulse sequences (see figures 2 and 3). At 5 J/cm2, some new features emerge, such as the distortion and appearance of cavities in zone A and the filling of zone B with blood, presumably due to rupture of the vasculature.

Figure 6

Thermally induced lesions following INS of a squirrel monkey brain. Shown are representative cresyl violet- stained frozen sections of a squirrel monkey brain subjected to INS. (A–C) Histological brain slice, photographed using a 4x objective. (D–F) Magnified view of area of the lesion (red squares), taken with a 20x objective.