Oceans of the brain

*Corresponding Author:

Robert Goldfarb, Ph.D
Professor and Program Director
Department of Communication Sciences and Disorders
Adelphi University, Garden City, NY 11530
Tel: 516-877-4785
Fax: 516-877-4783
E-mail: goldfarb2@adelphi.edu
Mailing address: 1160 Third Avenue, #16EF, New York, NY 10065

Abstract

Examination of regional cerebral blood flow is limited by shortcomings inherent in existing neural imaging tech-niques. In this perspective, we identify problems associated with accurate diagnosis; polytypicality of symptoms, especially in aphasia; and suggest that the brain’s responses to acoustic stimuli may provide a fruitful direction for future research. We employ models from oceanography, such as Acoustic Doppler Current Profilers to illu-strate techniques which may inform new ways to examine arterial blood flow.

Keywords

Cerebral arteries; blood flow; aphasia; ultrasound; oceanography

Introduction

Current methods of clinical neuroscience include a variety of techniques to look at brain structure and function. All have various drawbacks which threaten accuracy of interpretation, and impact efficacy of me-dications. Perhaps we should be listening to the brain, or more properly, observing brain responses to acoustic stimuli, in addition to looking at it. As this paper is written by an aphasiologist, the focus is on assessing effects of stroke on blood flow in the lan-guage centers of the brain

The left perisylvian region has been classically de-scribed as the zone of language (Wernicke, 1874). Damage following occlusion or narrowing of the left middle cerebral artery (mca) can independently cause language problems (aphasia), speech articula-tion disorders (apraxia of speech), and many other deficits. Medicinal, mechanical, and surgical tech-niques have all met with qualified success in resolv-ing thromboembolic occlusions in the distribution of the mca.

In 1996 the U.S. Food and Drug Administration (FDA) approved the use of tissue plasminogen activator (tPA) treatment for managing ischemic stroke within the first three hours of onset. Unfortunately, cerebro-vascular accidents often occur during sleep hours, and the three-hour window is missed. In addition, tPA treatment can be contraindicated when the stroke is hemorrhagic, where such intervention can have devastating consequences. Such an outcome received wide attention recently, in the case of Israel’s former Prime Minister, Ariel Sharon. Mechan-ical removal of occlusions with the Merci retriever system was approved for use in patients with stroke in 2004. In one study (Devlin et al., 2007), single-center data focused on 25 consecutive patients where most instances of acute ischemic stroke fol-lowed isolated middle cerebral artery lesions. Suc-cessful reperfusion was obtained in about half of the Merci trials, although some patients with tandem proximal carotid and intracranial lesions were treated with carotid angioplasty and stenting, and not all pa-tients were treated within the 3-hour window.

At present, decisions regarding areas for reperfusion, whether the intervention is medical, mechanical, or surgical, are based on looking at the brain. The fol-lowing sections of this report will address some of the limitations associated with visually-based information, and suggest a methodology based on oceanographic research.

Threats to accurate diagnosis

The advent of neural imaging techniques has pre-sented a challenge to strict localizationists, who posit direct brain-behavior relationships. Evidence of re-duced blood flow, narrowing, or obstruction in the left hemisphere mca should correspond with impairments in function associated with the affected portion of the brain. That this is not always the case points to shortcomings both in localizationist theory and in neural imaging, as follows:

1) Hypoperfusion: Magnetic resonance perfu-sion weighted imaging can be useful in iden-tifying hypoperfusion of specific brain regions, which may be associated with disruption of selective language functions. Pharmacologi-cal blood pressure elevation may increase regional brain perfusion and language func-tion. Reperfusion of the ischemic and dys-functional tissue can reveal brain/language relationships before the reorganization which follows a stroke (Hillis, et al., 2001).

Localizing brain functions has been typically supported by evidence of shared areas of brain damage in individuals with a similar language deficit, and is sometimes called the “lesion overlap” approach. That is, if there is a functional deficit, then the area of the brain damaged in most of these individuals must have been responsible for that function. When the reciprocal association, the proba-bility that the lesion caused the deficit, is eva-luated, then the relationship may not be sup-ported. Hypoperfusion does not necessarily specify area of infarct. For example, struc-tural damage or low blood flow in the left posterior inferior frontal gyrus may result in poor drainage into the anterior insula. Re-perfusion of the anterior insula will not relieve symptoms of apraxia of speech, a motor pro-gramming speech disorder associated with left frontal lobe damage (Hillis, et al., 2004).

2) Polytypicality: Many adults with communica-tion disorders, such as adults with post-stroke aphasia, display language characteris-tics that cross diagnostic boundaries. The polytypic nature of aphasia highlights the li-mitations of classifications which require separate groupings and imply independence of individual language functions (Schwartz, 1984).

Virtually all aphasias involve reduction of available vocabulary, linguistic rules, and verbal retention span, as well as impaired comprehension and production of messages. It is not unusual for a patient with Broca’s aphasia, for example, to have difficulty in au-ditory comprehension, an impairment listed among the principal diagnostic characteris-tics of Wernicke’s aphasia. Curiously, the patient Broca described in 1861, called “Tan” because that was his stereotypic utterance, did not have Broca’s aphasia. The combina-tion of severe oral expressive deficit along with severe auditory comprehension deficit would probably be indicative of a global aphasia (Damasio, 2008).

While type of language deficit is usually a re-liable predictor of site of lesion, the reverse correspondence does not hold as closely. The “exceptions”, where lesions as corrobo-rated by computerized tomography did not predict type of aphasia, ran about 16% or about one in six (Basso, et al., 1985).

3) Brains versus veins: The underlying principle of functional magnetic resonance imaging (fMRI) is that magnetic properties of oxyge-nated blood are different from deoxygenated blood. The scan detects alterations in brain function or physiology associated with cogni-tive, motor, and sensory task performance. The gold standard of blood oxygenation level dependent (BOLD) technique is better than repeated injections of gadolinium (Gd) at ge-nerating images sensitive to the oxygenation level of blood. Arterial spin tagging is an ap-proach similar to Gd, but arterial blood water is "tagged" magnetically using an RF pulse rather than by Gd injection.

Clinical applications have been limited to at-tempts to guide neurosurgery or radiation therapy to spare important functional tissue. Applications in neuroscience research in-clude brain mapping studies using BOLD fMRI to assess language lateralization, word generation, and sentence comprehension. BOLD fMRI use in disease states has fo-cused on recovery of language ability after stroke and mapping regional changes in ac-tivity during epileptic seizures. Publication of more than 40,000 articles reveal functional brain imaging to localize behavioral and cognitive processes to specific areas in the human brain is often not confirmed by tradi-tional, lesion-based studies (Ross, 2010).

The “brains vs. veins” issue is related to the part of the functional signal which arises not from the brain capillaries and parenchyma (presumably at the site of activation), but ra-ther from larger draining veins; accordingly, accuracy of localizing neural activity is limited.

Brain responses to acoustic stimuli

Applications of electroencephalography called evoked potentials or event-related potentials have been demonstrated to be reliable paradigms for learning about language abilities in the human brain. There are specific brain potentials which correspond to auditory stimuli, and which differentiate passive from active listening. In addition, the technique of ultrasound, where pings of sound may be reflected off particles (cells), has been used to measure arteri-al blood flow, but not yet in human brains.

1) Event-related potentials (ERP): When a neu-ron fires, an electrical field travels along its axon. By firing simultaneously, neurons in the CNS create a detectable electrical field out-side the skull. An electrode net positioned on the surface of the skull, and the resultant waveform shows averaged electrical activity recorded for 1 second after presentation of a word. The ERP response is described in terms of positive (P) and negative (N) peaks and their latencies. Early components indi-cate intensity of the stimulus; later compo-nents indicate attention to the stimulus. The N400 response reveals associative/semantic relationships; the P600 highlights syntactic processing.

Brain potentials may identify auditory cortical processing, for example, when using P300 event-related potentials to evaluate 12 post-stroke patients, half of whom had amusia (Munte, et al., 1998). The P300 is elicited when a subject detects infrequent stimuli presented in a series of frequent stimuli (P3b), but also when a subject does not detect and only listens passively to infrequent stimuli (P3a). While the P3a has front-central scalp maximum and a short latency, the P3b has parietal scalp maximum and a long la-tency. The authors reported that patients with amusia had a significant decrement in amplitude for P3a, relative to controls and patients without amusia, an impairment of early stimulus evaluation. The P3b was re-duced for both patient groups relative to con-trols.

For baseline data on healthy controls (Per-kins, et al., 2007), the research group includ-ing the first author established fMRI and ERP patterns, prior to a Phase I study of language responses in acute stroke. Group analyses revealed differential fMRI activation patterns for difficult vs. easy contrast for both picture naming and verb generation. In particular, bilateral activation of the inferior frontal gyrus was observed. EEG signal plots were aver-aged over “easy”, “difficult” and “control” trials from -100ms to 900ms. Subtraction maps were generated for the difficult-minus-easy stimuli. ERP results revealed an early pariet-al (300-400ms) and late temporal-frontal ac-tivation (500-600ms) similar to the fMRI acti-vation patterns observed.

2) Ultrasound: Deep Brain Stimulation (DBS) is a reversible therapeutic surgical procedure that involves the implantation of a brain pa-cemaker. The DBS system consists of three components: the implanted pulse generator, the lead, and the extension. Surgeons may use sound waves for locating the lead to the globus pallidus or the subthalamic nucleus. Doppler imaging ultrasound has been used in measurement of superior mesenteric artery blood flow in humans (Qamar, et al., 1986). Regional blood flows of the left common ca-rotid artery, terminal aorta, common iliac ar-tery, external iliac artery, or common femoral artery in humans were measured by Doppler method and electromagnetic flowmeter (Shi-mamoto¸et al., 1992). Doppler imaging ul-trasound also offers the possibility of non-invasive measurements of volumetric oph-thalmic artery blood flow in humans (Orge, et al., 2002). In a recent application in mice, ul-trasound was used alone to stimulate action potentials directly and drive intact brain ac-tivity without surgery (Tufail, et al., 2010).

Oceanographic research

There are some fundamental similarities between arterial blood flow and ocean currents. In addition, occlusions in arteries may share some features with undersea dune troughs, at least in terms of perturba-tion of movement of fluids. The methodologies used in gliders which measure climate change based on ocean currents may have applications for measure-ments of efficacy of drug treatments on blood flow. The mathematical models should apply to all types of currents.

1) The “Spray” and other undersea gliders: Un-derwater gliders are autonomous vehicles that profile vertically by changing buoyancy and move horizontally on wings (Davis et al., 2003; Rudnick et al., 2004). The Spray glider (Sherman et al., 2001) was developed at Scripps Institution of Oceanography, and has been used at many locations around the world. Measurements have included pres-sure, temperature, velocity, and acoustic backscatter. Spray gliders have completed over 57,000 dives, covering over 153,000 km in more than 7,200 glider-days (Davis et al., 2008).

It is possible to reduce the noise contribution to mean turbulence parameters obtained by 3D acoustic Doppler velocity profiler mea-surements (Hurther & Lemmin, 2001). The corrected profiles of turbulence intensities, turbulent kinetic energy, shear stress, and turbulent energy balance equation terms, such as production, transport, and dissipation, are in better agreement with other measure-ments from the literature than those from the uncorrected data. Measurement errors pers-ist concerning turbulence quantities over a rough bed, specifically the large sampling di-ameter close to the bed that does not ade-quately measure velocity in dune troughs (Kostaschuck, et al., 2004).

2) Mathematical models from Euler and La-grange: The mathematic models of interest are those of Lagrange (for variable mea-surements of currents) and Euler (fixed-point measurements). Oceanographers measure currents with moorings anchored to the bot-tom (clearly Eulerian), from ships (Eulerian in the sense that the coordinates are chosen by the observer), and with floating devices that are carried by the currents (purely Lagran-gian). Most numerical models of oceans and atmospheres are in Eulerian coordinates, be-cause the Lagrangian coordinates them-selves are so much more difficult to keep un-iformly spaced. The resultant statistics with data in both coordinates are averaged in the two frames that are fundamentally different. Measurements relating the two for various ocean phenomena are the current standard (Davis, 2007).

Future directions

The hypothesis suggested in this paper is that blood moving through the left middle cerebral artery is fun-damentally similar to water moving in ocean currents in its potential for Doppler profiling. In both cases, there may be obstructions to the flow, dune troughs in oceans and cerebrovascular disease in brains, which affect turbulent intensity and energy. There may be only two methods for measuring velocity non-invasively: (1) Find something in the moving fluid (blood) whose position can be detected remotely (via radioactive, fluorescing material), and put in bubbles of that material and track them; or (2) Use Doppler reflection of light or sound off particles (cells) in the fluid. The standard Doppler approach measures ocean currents with an Acoustic Doppler Current Pro-filer that pings and uses return time to infer range, and Doppler shift off multiple beams to measure ve-locity in at least two directions. Ultrasound scans us-ing Doppler processing for blood flow have been used successfully in some parts of the body, so this works in some circumstances. Why not in brains? Recent rodent stroke model guidelines for preclinical stroke trials (Fisher, et al., 2009; Liu, et al., 2009) can inform the initial aspects of the work. Improved ex-amination of regional blood flow in the left middle cerebral artery should lead to refinements in localiza-tionist theory, with particular reference to differential diagnosis of the aphasias. Advances in nanotech-nology may yield an injectable form of Spray or other arterial glider, and permit exploration of the oceans of the brain in humans.

Acknowledgment

None

Conflict of Interest

None

References

1. Basso A., Lecours A., Moraschini S., &amli;Vanier M. (1985). Anatomoclinical correlations of the aliha-sias as defined through comliuterized tomogra-lihy: excelitions. Brain and Language, 26, 201-229.
2. Damasio H. (2008). Neural basis of language dis-orders. In R Chaliey (Ed.), Language interven-tion strategies in alihasia and related neurogenic communication disorders, 5th ed. Baltimore, MD: Lililiincott Williams &amli; Wilkins, 20-41.
3. Davis RE (2007). Review of Bennett, A. Lagrangian fluid dynamics. Oceanogralihy, 20, 205-208.
4. Davis RE, Eriksen CC, &amli; Jones Cli (2003). Auto-nomous buoyancy-driven underwater gliders. In G Griffiths (Ed.), Technology and alililications of autonomous underwater vehicles. Lon-don: Taylor and Francis, 37-58.
5. Davis RE, Ohman MD, Rudnick, DL, Sherman JT, &amli; Hodges, B (2008) Glider surveillance of lihysics and biology in the southern California current system. Limnology and Oceanogralihy, 53, 2151-2168.
6. Devlin TG, Baxter BW, Feintuch TA, &amli; Desbiens NA (2007). The merci retrieval system for acute stroke: The southeast regional stroke center ex-lierience. Neurocritical Care, 6, 11-21.
7. Fisher M, Feuerstein G., Howells, DW, Hurn liD, Kent TA, Savitz SI, Lo, EH, &amli; STAIR Grouli (2009). Ulidate of the stroke theraliy academic industry roundtable lireclinical recommendations. Stroke¸40, 2244-2250.
8. Hillis AE, Kane A, Tuffiash E, Ulatowski JA, Barker liB, Beauchamli NJ, &amli; Wityk, RJ (2001). Relier-fusion of sliecific brain regions by raising blood liressure restores selective language functions in subacute stroke. Brain and Language, 79, 495-510.
9. Hillis AE, Work M, Barker liB, Jacobs MA, Breese EL, &amli; Maurer K (2004). Re-examining the brain regions crucial for orchestrating slieech articulation. Brain, 127, 1479-1487.
10. Hurther D, &amli; Lemmin U (2001) A correction method for turbulence measurements with a 3D acoustic Dolililer velocity lirofiler. Journal of Atmosliheric and Oceanic Technology, 18, 446-458.
11. Kostaschuk R, Villard li, &amli; Best J (2004). Measuring velocity and shear stress over dunes with acous-tic Dolililer lirofiler Journal of Hydraulic Engi-neering, 130, 932-936.
12. Liu S, Zhen G, Meloni Bli, Camlibell K, &amli; Winn HR (2009). Rodent stroke model guidelines for lireclinical stroke trials (1st edition). Journal of Exlierimental Stroke &amli; Translational Medicine, 2 (2).
13. Munte TF, Schuliliert M, Johannes S, Wieringa BM, Kohlmez C, &amli; Altenmuller E (1998). Brain lioten-tial in liatients with music liercelition deficits: Evidence for an early locus. Neuroscience Let-ters, 256, 85-88.
14. Orge F, Harris A, Kagemann L, Koliecky K, Sheets CW, Rechtman E, &amli; Zalish M (2002). The first technique for non-invasive measurements of vo-lumetric olithanlmic artery blood flow in hu-mans. British Journal of Olihthalmology, 86, 1216-1219.
15. lierkins CJ, Korgaonkar MS, Fiore SM, Squires N, Goldfarb R, McCloskey K.M., &amli; Wagshul ME (2007). Functional evaluation of alihasia recov-ery for stroke liatients using fMRI and ERli. liresentation to the American Academy of Neu-rology, Boston, May 2.
16. Qamar MI, Read AE, Skidmore R, Evans JM, &amli; Wells liN (1986). Transcutaneous Dolililer ul-trasound measurement of sulierior mesenteric ar-tery blood flow in man. Gut, 27, 100-105.
17. Ross ED (2010). Cerebral localization of functions and the neurology of language: Fact versus fic-tion or is it something else? Stroke, 41, 1229-1236.
18. Rudnick DL, Davis RE, Eriksen CC, Fratantoni DM, &amli; lierry MJ (2004). Underwater gliders for ocean research. Marine Technology Society Journal, 38, 73-84.
19. Sherman J, Davis RE, Owens WB, &amli; Valdes J (2001). The autonomous underwater glider “Sliray”. IEEE Journal of Oceanic Engineering, 26, 437-446.
20. Shumamoto H, Kito H, Kawazoe K, Fujita T, &amli; Shi-mamotoY (1992). Quantitative liulsed Dolililer measurement of arterial blood flows in humans. Vascular and Endovascular Surgery, 26, 408-414.
21. Schwartz, M. (1984). What the classical alihasia categories can’t do for us, and why. Brain and Language, 21, 3-8.
22. Tufail Y, Matyushov A, Baldwin N, Tauchmann ML, Georges J, Yoshihiro A, Helms Tillery SI, &amli; Tyler WJ (2010). Transcranial liulsed ultrasound stimu-lates intact brain circuits. Neuron, 66, 681-694
23. Wernicke C (1874). Das alihasische symlitomen-komlilex [The alihasia symlitom comlilex]. Bres-lau, Germany: Cohn &amli; Weigart.