Under construction !!!
 

SPECT difference analysis for localization of language activity: correlation with fMRI

The purpose of the study is to assess whether language activation results in meaningful brain perfusion changes as detected using SPECT difference analysis.

Introduction
Localizing areas of the brain involved in language processing is an important part in many neurosurgical procedures, e.g. for epilepsy. The standard technique for language localization has been by intra-operative stimulation or functional cortical mapping, but this procedure requires dedicated equipment and consumes precious time in the OR thereby increasing risks and costs. Functional imaging techniques, such as PET and fMRI, can be applied for non-invasive language localization. PET .... fMRI is a very promising new technique that has been extensively evaluated in activation studies in several centers in the world. More and more the technique is applied to localize language in the presurgical evaluation [Carpentier 1999]. Thus far, however, fMRI can only be applied in a few dedicated centers in the world, the costs are high and the patient is required to keep the head as fixed as possible during the task (usually head fixating material or devices are applied). An alternative non-invasive imaging modality to assess function is known as Single-Photon Emission Tomography (SPECT).

In most hospitals in the western world SPECT is a standard imaging modality usually applied to study bloodflow. These studies are typically performed with the patient in a resting state so as to minimize confounding effects of activity. On the other hand, SPECT has also been applied to study activation. A distinct advantage of SPECT compared to other functional modalities is that the subject is not required to keep still while performing a task, since only the iv-bolus-injection of the radiotracer is done during the task. The images are acquired approx. 30-45 minutes after the injection; during the actual scanning, the subject is required to keep the head still. SPECT has been previously applied to study brain perfusion changes caused by activation. Bouras published an abstract on brain perfusion changes caused by rectal distention in a group of 5 normal subjects using Tc99m-ECD [Bouras 1997]. Kawaguchi used motor activation Tc99m-HMPAO SPECT scans to evaluate the effect of a cerebral artery bypass on pure motor function [Kawaguchi 1997]. Okumura performed fMRI and Tc99m-ECD SPECT on 5 normals and 5 tumor patients to evaluate changes in the sensorimotor area [Okumura 1998]. Montaldi used ROIs and SPM to analyze Tc99m-HMPAO SPECT in 10 normal subjects and showed that associative encoding of pictures activates the medial temporal lobe [Montaldi 1998]. For this project we apply 99TcHMPAO-SPECT to image brain perfusion.

Comparing activity-related perfusion to non-activated perfusion provides the most sensitive determination and functional localization. This requires two SPECT scans to be analyzed and compared. Visual comparison of the two scans is difficult because of the relatively low resolution of SPECT, the low signal-to-noise ratio and the fact that both brain scans are (almost) never in the same geometrical position. Furthermore, SPECT image intensities depend on several different aspects, e.g. dose injected, attenuation correction and uncalibrated collimator sensitivities. This renders SPECT very difficult to quantify. The visual detection of perfusion changes as a result of activation has proven to be unreliable. Computer assisted difference analysis of SPECT data results in vastly superior interpretation of the functional data. SPECT difference analysis has been mainly applied for the evaluation of patients with medically refractory epilepsy [Zubal 1995, Avery 2000]. We demonstrated that the interpretation and application of Tc99m-HMPAO SPECT ictal and interictal images can be substantially improved through a series of image processing algorithms [Zubal 1995].  By registering, normalizing, and subtracting the interictal scan from the ictal scan, differences in brain perfusion can be calculated. Thesedifferences can finally be superimposed onto the MRI scans for anatomical localization.  These "difference images" make it possible to accurately localize perfusion changes, which often are associated with the epileptogenic area, and not detected with visual analyses.

Using a group of normals we aim to evaluate the information obtained using SPECT difference analysis for a language activation task and correlate the findings with the corresponding fMRI data. This serves two purposes:
1.      to determine whether SPECT may be an alternative method for language mapping. We aim to evaluate whether another functional modality, SPECT, may be a worthwhile, more easily implemented and less costly alternative for language mapping.
2.      to obtain an indication of the level of perfusion changes due to activation of the brain. This is very relevant for the analysis of SPECT difference data of epilepsy patients since activation may produce similar perfusion changes as those resulting from a seizure thereby seriously interfering with the localization of an epileptogenic region.

Material and Methods

This study was approved by the Human Investigation Committee and the Radiation Safety Committee of Yale University School of Medicine.

A group of 12 normal healthy adults was recruited by word of mouth. A brief medical physical exam and medical history were performed to establish health status. No pregnant or potentially pregnant women were included which was verified for all female subjects by asking specifically and by performing a urine pregnancy test prior to the SPECT injection.

Tasks

The Visual language task (VLTask) compared visually presented language sentences (typical examples: "Filling tanks", "Ears can heard", "Drill Wells", "Arranging meetings") against cross-match line decision (2 lines, each line contains symbols like: "//////", "<><<<<", ")((((("). (243 seconds task, nine alternated cycles, 4On-5Off, 25sec each, 10 trials each). TR was 1848msec.  OFF-periods consisted of two (identical or different) rows of lines presented every 3sec.

Analysis
Correlation of the SPECT difference data and the fMRI data will be performed using the following methods:

1. visual analysis of the SPECT difference results by three independent, blinded observers using the subject's high-resolution MR images for anatomical reference. This step includes the definition of ROIs and obtaining measures from these ROIs.
2. analysis of the fMRI data and definition of areas of activation.
3. statistical analysis of the SPECT difference data using SPM96 according to the protocol applied in [Chang, 2001]. Also, the subjects will be grouped so as to evaluate whether SPM96 presents a meaningful group effect.
4. the areas of fMRI activation are correlated with the SPECT difference data and the SPECT difference measures from these areas are obtained.

Step no. 3 requires 12 subjects for sufficient statistical power in comparing levels of sensitivity and specificity. For sample sizes up to 12 adding additional subjects gives considerable gain in terms of detectable normalized change. After 20 subjects there is a diminishing return in terms of power.

Subsequently, the results from the original SPECT difference analysis, the fMRI analysis, the SPM96 analysis, and the secondary SPECT difference analysis will be compared.
 

REFERENCES

[Avery 2000] Avery RA, Spencer SS, Studholme C, Stokking R, Morano G, Corsi M, Seibyl JP, Spencer DD, Zubal IG. Reproducibility of serial peri-ictal single-photon emission tomography difference images in epilepsy patients undergoing surgical resection. European Journal of Nuclear Medicine 2000; 27(1): 50-55.

[Baciu 1999] Baciu MV, Rubin C, Decorps MA, Segebarth CM. FMRI assessment of hemispheric language dominance using a simple inner speech paradigm. NMR in Biomedicine 1999; 12: 293-298.

[Bouras 1997] Bouras EP, O'Brien TJ, O'Conner MK, Mullan BP, Camilleri M. A pilot study of cerebral topographic representation of rectal distention in humans using Tc99m-ECD single photon emission computed tomography (SPECT). Gastroenterology 1997; 112: A703.

[Carpentier 1999] Carpentier AC, Pugh K, Westerveld M, Spencer DD, Constable RT, A Functional MRI Study of Visual versus Auditory Language Processing in Epileptic Patients and Normal Volunteers, Epilepsia, 40(7), Abstract D12, 1999.

[Kawaguchi 1997] Kawaguchi S, Noguchi H, Sakaki T, Morimoto T, Hoshida T, Yonezawa, T Imai T, Ohishi H. Evaluating the effect of superficial temporal artery to middle cerebral artery bypass on pure motor function using motor activation single photon emission computed tomography. Neurosurgery 1997; 41(5): 1065-1071.

[Montaldi 1998] Montaldi D, Mayes AR, Barnes A, Pirie H, Hadley DM, Patterson J, Wyper DJ. Associative encoding of pictures activates the medial temporal lobes. Human Brain Mapping 1998; 6: 85-104.
 

[Okumura 1998] Okumura A, Kuwata K, Nishimura Y, Kawaguchi M, Takenaka K, Sakai N, Era S, Hoshi H. Clinical applicability of functional magnetic resonance imaging and activation study with single photon emission computerized tomography for functional mapping. Neurological Research 1998; 20: 191-197.

[Villanueva-Meyer90] Villanueva-Meyer J, Thompson D, Mena I, Marcus CS. Lacrimal Gland Dosimetry for the Brain Imaging Agent Technetium-99m-HMPAO. Journal of Nuclear Medicine 1990; 31: 1237-1239.

[Zubal 1995] Zubal IG, Spencer SS, Imam K, Seibyl J, Smith EO, Wisniewski G, Hoffer PB. Difference-Images Calculated from Ictal and Interictal Tc99m-HMP
AO SPECT Scans of Epilepsy.  Journal of Nuclear Medicine 1995; 36: 684-689.

EXTRA:
[Baciu 1999]
Baciu MV, Rubin C, Decorps MA, Segebarth CM.
FMRI assessment of hemispheric language dominance using a simple inner speech paradigm.
NMR in Biomedicine 1999; 12: 293-298.

[Beversdorf 1995]
David Beversdorf and Stan Metzger and David Nelson and Robert Alonso and Jerry Knight.
Single-word auditory stimulation and regional cerebral blood flow as studied by SPECT.
Psychiatry research: Neuroimaging 1995; 61: 181-189

[Bouras 1997]
Bouras EP, O'Brien TJ, O'Conner MK, Mullan BP, Camilleri M.
A pilot study of cerebral topographic representation of rectal distention in humans using Tc99m-ECD single photon emission computed tomography (SPECT).
Gastroenterology 1997; 112: A703.

[Carpentier 1999]
Carpentier AC, Pugh K, Westerveld M, Spencer DD, Constable RT,
A Functional MRI Study of Visual versus Auditory Language Processing in Epileptic Patients and Normal Volunteers,
Epilepsia, 40(7), Abstract D12, 1999.

[Crosson 1994]
Bruce Crosson and David J. G. Williamson and Shailendra S. Shukla and Janice C. Honeyman and Stephen E. Nadeau.
A technique to localize activation in the human brain with Technetium-99m-HMAPO SPECT: A validation study using visual stimulation.
JNM 1994; 35: 755-763

[George 1992]
M. S. George and H. A. Ring and D. C. Costa and K. Kouris and P. J. Ell.
Demonstration of human cortex activation using SPECT.
J of Neural Transmission 1992; 87: 231-236

[Goldenberg 1992]
G. Goldenberg and W. Oder and J. Spatt and I. Podreka.
Cerebral correlates of disturbed executive function and memory in survivors of severe closed head injury: a SPECT study.
J of Neurology, Neurosurgery, and Psychiatry 1992: 55; 362-368

[Gomez-Tortosa 1994]
Estrella Gomez-Tortosa and Eileen M. Martin and Jerry J. Sychra and Manuel Dujovny.
Language-activated Single-Photon Emission Tomography imaging in the evaluation of language lateralization - evidence from a case of crossed aphasia: Case report.
Neurosurgery; 35: 515-520 (no. 3)

[Holm 1994]
Soren Holm and Peter L. Madsen and Bjorn Sperling and Niels A. Lassen.
Use of 99m-Tc-Bicisate in activation studies by split-dose technique.
JCBFM 1994; 14 (suppl 1): S115-S120.

[Hurtig 1994]
Richard R. Hurtig and Richard D. Hichwa and Daniel S. O'Leary and Laura L. Boles Ponto and Shalini Narayana and G. Leonard Watkins and Nancy C. Andreasen.
Effects of timing and duration of cognitive activation in [15-O]water PET studies.
JCBFM 1994; 14: 423-430

[Kawaguchi 1997]
Kawaguchi S, Noguchi H, Sakaki T, Morimoto T, Hoshida T, Yonezawa, T Imai T, Ohishi H.
Evaluating the effect of superficial temporal artery to middle cerebral artery bypass on pure motor function using motor activation single photon emission computed tomography.
Neurosurgery 1997; 41(5): 1065-1071.

[Montaldi 1998]
Montaldi D, Mayes AR, Barnes A, Pirie H, Hadley DM, Patterson J, Wyper DJ.
Associative encoding of pictures activates the medial temporal lobes.
Human Brain Mapping 1998; 6: 85-104.

[Okumura 1998]
Okumura A, Kuwata K, Nishimura Y, Kawaguchi M, Takenaka K, Sakai N, Era S, Hoshi H.
Clinical applicability of functional magnetic resonance imaging and activation study with single photon emission computerized tomography for functional mapping.
Neurological Research 1998; 20: 191-197.

[Pantano 1992]
Patrizia Pantano and Vittorio Di Piero and Monica Ricci and Cesare Fieschi and Luigi Bozzao and Gian Luigi Lenzi.
Motor stimulation response by technetium-99m hexamethylpropylene amine oxime split-dose method and single photon emission tomography.
EJNM 1992; 19: 939-945

[Robert 1995]
Philippe H. Robert and Octave Migneco and Michel Benoit and Jacques Darcourt and Elizabeth Ellul and Jos\'{e} Benoliel and Pierre Bonhomme and Valerie Aubin and Francoise Bussiere and Guy Darcourt.
A single photon emission computed tomography study of cerebral regional perfusion changes induced by a learning task in subjects with mild memory impairment.
Psychiatry research: Neuroimaging 1995; 61: 231-241

[Shedlack 1991]
K. J. Shedlack and R. Hunter and D. Wyper and R. McLuskie and G. Fink and G. M. Goodwin
The pattern of cerebral activity underlying verbal fluency shown by split-dose single photon emission tomography SPET or SPECT) in normal volunteers.
Psychological Medicine 1991; 21: 687-696

[Woods91]
Scott W. Woods and Irene M. Hegeman and I. George Zubal and John H. Krystal and Kenneth Foster and Eileen O. Smith and George R. Heninger and Paul B. Hoffer.
Visual stimulation increases Technetium-99m-HMPAO distribution in human visual cortex.
JNM 1991; 32: 210-215
 

[Constable1999] and [Studholme1999]