Epilepsia, 54(5):793–800, 2013
doi: 10.1111/epi.12139
FULL-LENGTH ORIGINAL RESEARCH
Optimizing SPECT SISCOM analysis to localize seizure-onset
zone by using varying z scores
*Christopher R. Newey, †‡Chong Wong, †Z. Irene Wang, §Xin Chen, †¶Guiyun Wu, and
†Andreas V. Alexopoulos
*Department of Neurology General Neurology, Cleveland Clinic Foundation, Cleveland, Ohio, U.S.A.; †Department of Neurology,
Epilepsy Center, Cleveland Clinic Foundation, Cleveland, Ohio, U.S.A.; ‡Department of Neurology, Westmead Hospital,
Sydney, New South Wales, Australia; §Toshiba Medical Imaging Research, Mayfield Village, Ohio, U.S.A.; and
¶Department of Nuclear Medicine, Cleveland Clinic Foundation, Cleveland, Ohio, U.S.A.
SUMMARY
Purpose: Subtraction ictal single photon emission com-
puted tomography (SPECT) co-registered to magnetic
resonance imaging (MRI) (SISCOM) is a useful modality
to identify epileptogenic focus. Using this technique, sev-
eral studies have generally considered the area of highest
ictal hyperperfusion, as outlined by thresholding the dif-
ference images with a standard z score of 2, to be highly
concordant to the epileptogenic focus. In clinical prac-
tice, several factors influence ictal hyperperfusion and
using different SISCOM thresholds can be helpful. We
aimed to systematically evaluate the localizing value of
various z scores (1, 1.5, 2, and 2.5) in a seizure-free
cohort following resective epilepsy surgery, and to exam-
ine the localizing information of perfusion patterns
observed at each z score.
Methods: Twenty-six patients were identified as having
ictal-interictal SPECT images, preoperative and postope-
rative MRI studies, and having remained seizure free for
at least 6 months after temporal or extratemporal surgi-
cal resection. SISCOM analysis was performed using pre-
operative MRI studies, and then blindly reviewed for
Ictal single-photon emission computed tomography
(SPECT) is frequently used to aid presurgical work-up.
Analysis of ictal SPECT is usually performed in comparison
with an interictal SPECT image. Ictal-interictal SPECT with
co-registration to magnetic resonance imaging (MRI) (SIS-
COM) has proven to be a highly valuable diagnostic tool in
noninvasive localization of the seizure-onset zone (Kaibori-
boon et al., 2002; Lee et al., 2006; Wichert-Ana et al.,
2008), especially in challenging patients with nonlesional
and/or extratemporal epilepsies (O’Brien et al., 2000), and
Accepted January 29, 2013.
Address correspondence to Andreas V. Alexopoulos, Department of
Neurology, Epilepsy Center, Cleveland Clinic Foundation, 9500 Euclid
Avenue, Cleveland, OH 44195, U.S.A. E-mail: alexopa@ccf.org
Wiley Periodicals, Inc.
© 2013 International League Against Epilepsy
793
localization of hyperperfused regions. With the added
information from postoperative, coregistered MRI, perfu-
sion patterns were determined.
Key Findings: Using pair-wise comparisons, we found that
the optimal z score for SPECT-SISCOM localization of
the epileptogenic zone was 1.5, not the commonly used z
score of 2. The z score of 1.5 was 84.8% sensitive and 93.8%
specific. The z score of 1.5 had a moderate interrater
agreement (0.70). When an hourglass configuration hyp-
erperfusion pattern was present, a trend toward correctly
localizing the seizure onset region was suggested (100% of
the 11 observed occurrences). Nonetheless this trend was
not statistically significant, possibly reflecting the small
number of occurrences in our study.
Significance: SISCOM is a useful modality in evaluating
patients for epilepsy surgery. This study shows that the
z score of 1.5 represents a highly sensitive and specific
SISCOM threshold that should be examined in conjunc-
tion with the traditionally used z score of 2 to enhance the
chances of correct localization. Further prospective inves-
tigations are needed to confirm this finding in large
patient series.
KEY WORDS: Ictal SPECT, SISCOM, z score, Epilepsy.
patients who have failed previous resections (Wetjen et al.,
2006). Recent prospective studies have shown that SPECT
can be used to generate implantation hypothesis and
improve surgical outcome in patients referred for invasive
electroencephalography (EEG) evaluations (Matsuda et al.,
2009; von Oertzen et al., 2011).
Proper selection of a z score (or threshold) is critical for
consistent, reliable, and successful analysis of SISCOM
data (O’Brien et al., 1998b; Van Paesschen, 2004; Van
Paesschen et al., 2007). Traditionally, a z score of 2 has
been proposed (O’Brien et al., 1998a). However, SISCOM
images often display multiple areas of hyperperfusion,
which include both the areas of ictal onset and seizure prop-
agation (Van Paesschen, 2004). Given the dynamic nature
of SPECT, a fixed, conservative threshold may not be opti-
mal in all patients. Some users may actually vary the z score
794
C. R. Newey et al.
for different patients, but the sensitivity and specificity at
each z score have not been systematically studied. Further-
more, different hyperperfusion patterns emerge at each z
score, which may provide additional localizing information
(Dupont et al., 2006). It is the purpose of this study to sys-
tematically evaluate the localizing value of z scores of 1,
1.5, 2, and 2.5 and to examine the localizing information of
hyperperfusion patterns observed at each z score.
Methods
Patient screening
After obtaining approval by the institutional review
board, we screened 94 consecutive patients who underwent
either temporal or extratemporal resection from 2007 to
2009 at the Cleveland Clinic Epilepsy Center. Patients were
included if they: (1) underwent ictal and interictal SPECT
studies preoperatively; (2) had preoperative and postopera-
tive MRIs; (3) became seizure-free postoperatively (mini-
mum 6-month follow-up). Patients with previous history of
epilepsy surgery were excluded. In all patients, resective
surgery strategy and extent of resection were determined by
consensus at patient management conference on the basis of
integration of all available clinical information and investi-
gations such as scalp and/or intracranial electroencephalog-
raphy, MRI [18F]fluorodeoxyglucose positron emission
tomography (FDG-PET), SPECT, magnetoencephalogra-
phy (MEG), and neuropsychological assessment. During
patient management conference, SPECT SISCOM threshol-
ded with a z score of 2 had been used for clinical interpreta-
tion and decision-making.
SPECT injection and imaging
Epilepsy-trained nurses sat by the bedside and injected
the radiopharmaceutical tracer in all studies. 99mTc-ECD
(ethyl-cysteinate dimer; 20–40 mCi) was injected into a
forearm vein immediately after noting seizure onset.
SPECT images were acquired on a Siemens (Erlangen,
Germany) Symbia dual-head camera (SPECT: 15 s per stop
960 stops, 128 9 128 matrix, iterative reconstruction with
attenuation correction, six iteration, eight subsets; CT:
A B
Epilepsia, 54(5):793–800, 2013
doi: 10.1111/epi.12139
30 mAs, 5 mm slice) within 2 h after radiopharmaceutical
injection. Interictal SPECT studies were accomplished after
minimum of 24 h of no seizures. Timing of injection was
determined on retrospective video and EEG review. Injec-
tion time was defined as the time point at which the radio-
pharmaceutical-containing syringe plunger had been
completely depressed. Seizure onset was defined as the ear-
liest ictal EEG or clinical evidence of seizure activity, and
seizure cessation when seizure activity was no longer pres-
ent on EEG (Wong et al., 2010).
SISCOM
Ictal and interictal SPECT studies were coregistered
using an automatic registration algorithm based on mutual
information (Maes et al., 1997). The interictal image was
subtracted from the ictal study after normalization, and then
the subtracted image was smoothed using three-dimensional
(3D)–Gaussian smoothing kernel (full width at half maxi-
mum = 12 mm). This information was transformed into a z
score using the mean and standard deviation of the differ-
ences in all brain voxels. The mean activation map was then
used for coregistration to the preoperative MRI for anatomic
localization (O’Brien et al., 1998b; Dupont et al., 2006).
Blinded SISCOM review
All SISCOM images were then reviewed for cortical
localization by two reviewers (CW, CN) blinded to clinical
data. The whole brain was divided into 10 regions including
left and right frontal, temporal, perirolandic, parietal, and
occipital based on previous studies (McNally et al., 2005).
Concordance between the SPECT hyperperfusion and
extent of resection was determined at the level of these 10
regions. Regions of hyperperfusion were superimposed on
the presurgical MRI (Fig. 1A), using MRICRO software
(Georgia Institute of Technology, Atlanta, GA, U.S.A.) with
varying z scores of 1, 1.5, 2, and 2.5. Reviewers were then
asked to localize, if possible, each individual interictal-ictal
SPECT study to one or more of the 10 cerebral regions.
Emulating the approach applied to patient management con-
ference, the criteria used to determine localization included:
SISCOM image intensity, location of hyperperfusion, and
Figure 1.
SISCOM analysis and MRI coregis-
tration. Panel A shows SISCOM
analysis performed on preoperative
MRI (Z = 2). Panel B shows the
coregistered postoperative MRI,
which presents a resection consis-
tent with SISCOM localization.
Epilepsia ILAE

number of regions of hyperperfusion. Final localization was
based on agreement of the two reviewers. If disagreement
occurred, a third reviewer (GW) evaluated the images.
Agreement between this reviewer and one of the two other
reviewers was the final determination. If no agreement
occurred, the study was considered nonlocalizing.
Determination of SISCOM localization value and
hyperperfusion patterns
We unblinded the data after completing the above blinded
portion of the study. We then judged correctness of localiza-
tion by introducing the patient’s postoperative MRI, which
was obtained 3–6 months after resection and was coregis-
tered with the corresponding preoperative MRI (Fig. 1B). If
the area determined by the final decision of localization in
the previous step fell within the surgical resection, then the
localization decision was considered correct.
Reviewers further identified the hyperperfusion patterns
as a separate unblinded step. By modifying the previously
published descriptions by Dupont et al. (2006), hyperperfu-
sion patterns were classified as follows: Pattern 1 describes
the pattern with the largest and most intense (highest z
score) hyperperfusion falling within the surgical resection
site (Fig. 2A). Pattern 2 describes SISCOM hyperperfusion
with a bilobulated, that is, hourglass appearance with the
brightest lobule being outside the resection and the less
intense of the two lobules residing within the resection site,
Figure 2.
Representative SPECT images at varying z scores superimposed on the postoperative MRI. In both examples, the z score of 1.5 shows
optimal sensitivity. Panel A shows the largest and most intense hyperperfusion overlapping with the resected site (perfusion pattern 1).
Panel B shows an hourglass pattern of hyperperfusion overlapping with the resected area (perfusion pattern 2). Center of resection is
denoted by blue cross.
Epilepsia ILAE
795
Varying z Score in SPECT SISCOM
but not vice versa (Fig. 2B). Pattern 3 consists of multiple
SISCOM hyperperfusion areas not within or near the surgi-
cal resection site.
Statistics In this statistics section is described quite nicely
Continuous measures were described as means, standard
deviations, and percentiles. Categorical measures were
summarized using frequencies and percentiles. Generalized
estimating equations (GEEs) were performed to evaluate
the relationship between agreement and z score, and assess
the association between correct localization and injection
time as percent seizure time across all z scores. Because
there are four z scores and multiple measurements for each
patient (due to multiple reviewers), the correlation of the
measurements within patient was adjusted by using a
compound symmetry covariance structure during the GEE
model selection procedure. Logistic regression was per-
formed to assess the association between correct localiza-
tion and injection time (or percent injection time over total
seizure time) for each z score. All analyses were performed
at a significance level of p < 0.05 using SAS 9.2 software
(SAS Institute, Cary, NC, U.S.A.).
Sensitivity and specificity analyses were performed
based on a modified version of the methodology by McNal-
ly et al. (2005). For each predetermined threshold, the 10
cerebral regions (left and right frontal, temporal, perirolan-
dic, parietal, and occipital) were evaluated to assess whether
Epilepsia, 54(5):793–800, 2013
doi: 10.1111/epi.12139
796
C. R. Newey et al.

one or several lobar area(s) contained areas of hyperperfu-
sion, based on the reviewers’ agreed decision. By definition
the site of resection should encompass the true epilepto-
genic zone in patients who experience sustained seizure
freedom following epilepsy surgery. Therefore, for a given
region, the sensitivity and specificity parameters in this
cohort were defined as follows: true-positive (hyperperfu-
sion and region resected), false-positive (hyperperfusion but
not resected), true-negative (no hyperperfusion and no
resection), and false-negative (absence of hyperperfusion in
resected region). Sensitivity and specificity values were cal-
culated at all four z scores.
Results
Patient profile
Twenty-six patients (nine male, 34.6%) met inclusion
criteria. Mean age was 36.5 years (standard deviation [SD]
15.2 years; range 5–67 years). Preoperative mean seizure
burden was 9.5 seizures/week with 18 (69.2%) having a his-
tory of generalized motor seizures. MRI was abnormal in 19
(73%). Twenty-three (88.5%) had PET and 17 (65.4%)
underwent intracranial EEG. Half of these patients had
extratemporal epilepsy, and the other half had temporal lobe
epilepsy (nonmesial temporal in the majority of patients).
Surgical pathology was cortical dysplasia (n = 10, 38.5%),
infarct (n = 7, 26.9%), vascular malformation (n = 2,
7.7%), hippocampal sclerosis (n = 3, 11.5%), tumor
(n = 3), and nonspecific (n = 1, 3.8%). Mean postoperative
follow up was 20 months (range 7–35 months). Seven of
the 26 patients had secondarily generalized seizures at the
time of ictal SPECT.
1.5 is significantly superior (p = 0.005), when compared to
the traditional z score of 2, as well as 1 and 2.5. There are no
significant differences in terms of localization value
between z scores 2 and 1, 2.5 and 1, and between 2 and 2.5.
Figure 2 shows representative SPECT images at varying z
scores superimposed on the patient’s postoperative MRI. In
both examples, a 1.5 z score shows optimal sensitivity.
Agreement between reviewers was poor with SISCOM
threshold of 1; Kappa coefficient was 0.29 (95% confidence
interval [CI] 0.16, 0.74). A z score of 1.5 and 2 had moder-
ate Kappa coefficients, 0.70 (95% CI 0.44, 0.96) and 0.60
(95% CI 0.29, 0.91), respectively. A z score of 2.5 had a
high Kappa coefficient of 0.90 (95% CI 0.70, 1), where the
reviewers agreed on all but one case.
Localizing value of each hyperperfusion pattern
The bilobulated pattern 2 was encountered the least
(n = 11); when present, it always correctly localized the
epileptogenic focus (100%; Fig. 2B). It was most frequently
observed at z score 1.5 (n = 7). Pattern 1 was present in 41
instances and correctly localized in 38 (92.7% of occur-
rences). It was also encountered most frequently at z score
1.5 (n = 12). Pattern 3 was observed most frequently at z
score of 1. Of 52 encounters of pattern 3, reviewers had
localized correctly the epileptogenic zone in two cases
(3.8%). Results are summarized in Table 2. Patterns 1 and
2, exemplified in Fig. 2A,B, respectively, are both clearly
more localizing than 3 (odds ratio >1), as confirmed using
the GEE model. We were unable to compare pattern 1
against pattern 2, as the GEE model did not converge due to
the much less frequent occurrence of pattern 2.

Injection time
Average injection time was 18.4 s (range 3–63 s), with Table 1. Pair-wise comparisons of each z score with
the majority of patients (61.4%) having injections under regards to its localization accuracy
20 s. Average ictal duration was 75.6 s (range 17–260 s). z score pairs Odds ratio 95% CI p-Value
With an estimated peak brain uptake of 30 s (Andersen, 1.5 vs. 1 4.55 1.78 11.67 0.002
1989; Kaminska et al., 2003; Dupont, 2008; Pastor et al., 1.5 vs. 2 4.23 1.54 11.62 0.005
2008), the majority of SPECT studies (17 of 26) occurred 1.5 vs. 2.5 7.53 2.56 22.18 0.0002
ictally; 6 patients had short seizures lasting <30 s. Despite 2 vs. 1 1.08 0.42 2.78 0.88
2 vs. 2.5 1.78 0.88 3.59 0.11
expected heterogeneity of seizures in this group, we found
2.5 vs. 1 0.60 0.22 1.65 0.33
no significant correlation between injection time
(p = 0.36), or percentage of injection time over total seizure
time (p = 0.39), and localization accuracy as indicated by Table 2. Localization accuracy of each pattern indicated
the GEE model. by number of correct localization at each z score
z score Total
Localizing value of each z score Total no. of correct no. of
At z score of 1, 34.6% (n = 9) of studies were correctly 1 1.5 2 2.5 localization (%) occurrence
localized. The correctly localized studies increased to Perfusion 7 12 10 9 38 (92.7) 41
76.9% (n = 20) at a z score of 1.5. Higher z scores were Pattern 1
associated with a lower percentage of correctly localized Perfusion 1 7 3 0 11 (100) 11
studies: 50% (n = 13) at 2% and 34.6% (n = 9) at 2.5. Pattern 2
Perfusion 1 1 0 0 2 (3.8) 53
Table 1 shows pair-wise comparisons of each individual z
Pattern 3
score with regard to its localization correctness. A z score of
Epilepsia, 54(5):793–800, 2013
doi: 10.1111/epi.12139

Table 3. Sensitivity and specificity at each z score
z score Sensitivity Specificity
1 87.8 86.6
1.5 84.8 93.8
2 78.8 92.8
2.5 76.5 93.8
Sensitivity and specificity
Sensitivity and specificity values, calculated from all 26
patients, are shown in Table 3. Optimizing sensitivity and
specificity by maximizing the sum of the two, we found that
the 1.5 z score represents optimal performance (specificity
93.8%, sensitivity 84.8%). When compared to a z score of 2
(92.8% specific, 78.8% sensitive), a 1.5 z score is both more
sensitive and more specific.
Discussion
Study rational
SISCOM is a valuable noninvasive tool in the presurgi-
cal evaluation of patients with medically intractable epilep-
sies. Traditionally, SISCOM results above z score of 2
(i.e., changes >20%; O’Brien et al., 1998a) have been gen-
erally regarded as meaningful and have been used to deter-
mine localization of the epileptogenic focus by many
epilepsy centers including ours (O’Brien et al., 1999;
Dupont et al., 2006; Chassagnon et al., 2009; Aboian
et al., 2011; Lee et al., 2011; von Oertzen et al., 2011).
However, ictal perfusion changes are influenced by several
factors. As a result SISCOM foci with lesser ictal increase
in intensity can provide important localizing information
(O’Brien et al., 1998b; Van Paesschen, 2004). It should be
noted that the “traditional” z score of 2 was arbitrarily
selected as an objective and readily defined threshold
(O’Brien et al., 1998a), but was not rigorously compared
with other thresholds for analysis of SISCOM data. We
designed this study to systematically evaluate the localiz-
ing value of varying z scores in a consecutive cohort of sei-
zure-free patients in whom we can infer the epileptogenic
zone. Although this is a highly selected cohort, without this
inference, it would not be possible to calculate sensitivity
and specificity.
Sensitivity and specificity at each z score
Overall, our results strongly suggest that it is essential to
not only analyze SISCOM images at the commonly set z
score of 2, but also to vary the z score on an individual case-
by-case basis, and particularly to attend to the lower z score
of 1.5. In fact, our statistical analysis indicates that the 1.5 z
score is best suited in predicting correct localization. There
were no significant differences between the generally
accepted z score of 2 versus the less specific z score of 1, or
the more restrictive z score of 2.5, with regards to their
797
Varying z Score in SPECT SISCOM
localizing value. It may seem surprising at first that z score
of 1 had a lower localization rate of 34.6%, compared to
76.9% at z score of 1.5. Although the brightest regions
would be the same in both, the lower z score images would
show more and larger regions of hyperperfusion. Because
our criteria used to determine localization included hyper-
perfusion intensity and location, as well as number of
regions of hyperperfusion, many studies at z score of 1 were
judged to be nonlocalizable, which explains the low locali-
zation rate and the consequent low Kappa coefficient show-
ing poor agreement among reviewers.
In a previous study by McNally et al. (2005), receiver
operating characteristic (ROC) curve analysis was per-
formed to determine the optimal p-value threshold based on
ictal-interictal SPECT analysis using statistical parametric
mapping. They computed a range of p-values from 0.00001
to 0.3 and found a peak around p = 0.01 (i.e., z = 2) both in
mesial temporal and neocortical cases. In our study, we used
a much narrower range of discrete z scores of 1, 1.5, 2, and
2.5, equivalent to p-values of 0.1, 0.03, 0.01, and 0.003. We
sought to evaluate these four clinically relevant z scores
with a methodology that can be applied to daily clinical
practice. Taking into account these differences, sensitivity
and specificity values in our study largely correspond to the
region that showed peak ROC performance, as described in
the McNally study.
We found z score of 1.5 to have an optimal sensitivity of
84.8% and specificity of 93.8%. Using ROC analysis,
McNally et al. reported sensitivity of 93% and 77% and
specificity of 87% and 93% for mesial and neocortical epi-
lepsies, respectively. These differences may be attributable
to different patient population and SPECT methodology.
In the McNally study the majority of injections occurred
postictally. Localization of the epilepsy was determined
based on seizure-free outcome (similar to our study) in all
patients with mesial temporal, but not all patients with neo-
cortical epilepsy (McNally et al., 2005).
Interrater agreement at each z score
We found the highest interrater Kappa coefficient at z
score of 2.5 indicating substantial agreement (the reviewers
agreed in all but one case). Because the number of hyperper-
fused areas became markedly smaller at the highest 2.5
threshold, reviewers had a relatively easier task in arriving
to a final decision, although their decision was not always
correct (only 34.6% correctly localized). In practice, an
inexpert SISCOM user would tend to apply a high z score
such as 2.5, because the obtained image is cleaner and pro-
vides better agreement among reviewers. However, as our
data indicated, strict agreement statistics do not always
yield the right localization. Both z scores of 1.5 and 2 were
associated with a moderate Kappa coefficient. The poorest
Kappa coefficient was seen at the lowest threshold of 1
reflecting the higher number and extent of regions of hyper-
perfusion at this threshold.
Epilepsia, 54(5):793–800, 2013
doi: 10.1111/epi.12139
798
C. R. Newey et al.
Patterns of ictal hyperperfusion
We found that all occurrences of the hour-glass configu-
ration (pattern 2) led to a correct decision regarding locali-
zation of the ictal focus. This bilobulated hourglass
configuration is thought to indicate ictal propagation, with
the lobule of highest z score residing outside the seizure-
onset zone, whereas the less intense lobule lies closest to the
area of seizure generation (Dupont et al., 2006). Because of
its infrequent occurrence, we were unable to determine sta-
tistically if pattern 2 has higher localizing value when com-
pared to pattern 1. Nonetheless, our study suggests that the
identification of an hourglass-appearing SISCOM pattern
can be highly valuable and leads to more accurate localiza-
tion of the ictal-onset zone.
For our study, we used a modified version of the images
proposed by Dupont et al. to determine patterns of hyper-
perfusion. The previous study examined hyperperfusion
patterns in relationship to a preoperatively identified/
known dysplastic lesion on MRI (Dupont et al., 2006).
In our study, initial determination of SISCOM localization
was made based on agreement among blinded reviewers.
Subsequent analysis of perfusion patterns was performed
as a separate unblinded step, after coregistering the SIS-
COM images on the patient’s postoperative MRI. Despite
the different approaches, the two studies yielded similar
conclusions. It is important to note areas of hyperperfusion
identified by ictal SPECT may point not only to the
presumed ictal onset, but also to the pathways and regions
of seizure propagation (Noachtar et al., 1998; Shin et al.,
2002; Dupont et al., 2006). Hyperperfusion patterns that
involve the largest area and/or exhibit the highest z score
may in fact represent propagated activity. Therefore, the
area of highest z score may not need to be completely
resected or included in the resection to achieve seizure
freedom (Van Paesschen et al., 2007). Rather, the resection
site should always be determined by a multimodal
approach, in the context of a full presurgical evaluation
(Van Paesschen, 2004; Kimura et al., 2012).
Injection time
We did not find injection time, in its absolute value or its
percentage over total seizure time, to have a significant cor-
relation with the probability of correct localization. This
may be due to the small sample size and/or the relatively
rapid radioisotope injection time. Indeed the average injec-
tion time in this series was 18.4 s, with seizure duration
averaging 75.6 s, and 50% of patients having completed the
tracer injection within 15 s of seizure onset. As a result the
majority of patients (17/26) in our study had a true ictal
SPECT study, even after allowing for the generally accepted
minimum of 30 s of seizure duration postinjection for tran-
sit time to occur (Andersen, 1989; Kaminska et al., 2003;
Dupont, 2008; Pastor et al., 2008). Six of the remaining nine
patients had short seizures lasting <30 s. On the basis of
these data we did not find correlation between the existence
Epilepsia, 54(5):793–800, 2013
doi: 10.1111/epi.12139
of the hourglass perfusion pattern and true ictal versus
“post-ictal” injection. Earlier injections have been sug-
gested to be more localizing compared to later injections
(Noachtar et al., 1998; Hong et al., 2008). This is particu-
larly important in extratemporal epilepsies, where propaga-
tion is more likely to occur early after ictal onset (Noachtar
et al., 1998). Theoretically, the most useful ictal SPECT is
one that has the radioisotope injected seconds before ictal
onset. This is demonstrated in few “preictal” SPECT studies
that were captured either by serendipity or in special cir-
cumstances when seizures can be provoked (Hong et al.,
2008; Pastor et al., 2008; Barba et al., 2009). These studies
showed a high degree of colocalization between the area of
“preictal” SPECT hyperperfusion and the corresponding
epileptogenic focus. Lastly, 7 of 26 patients in our study had
secondarily generalized seizures at the time of SPECT
study. This observation may decrease localization value of
SISCOM images and confound the correlation analysis
(Varghese et al., 2009). Nonetheless we did not find any
correlation between the existence of the hourglass perfusion
pattern and the presence or absence of secondarily general-
ized seizures.
Limitations and future directions
Inherent limitations exist due to the retrospective nature
of our current study. Although we included all ictal SPECT
studies performed in a consecutive patient series, it should
be noted that not all patients undergoing presurgical evalua-
tion are referred for ictal SPECT studies, especially when
there is clear concordance between video-EEG and MRI
abnormalities, such as in patients with unilateral hippocam-
pal sclerosis (Velasco et al., 2011). This is likely the reason
that very few patients with mesial temporal lobe epilepsy
were included in this study (only three). Moreover, ictal
SPECT is a cost-intensive and labor-intensive procedure
reserved for the more challenging patients undergoing pre-
surgical evaluation.
Overall, our seizure-free cohort comprises a mixture of
temporal (50%) and extratemporal patients (50%), and it
included patients with both lesional and nonlesional preopera-
tive MRI studies. Although we screened a large number of
patients undergoing their first resective epilepsy surgery, we
included only those who clearly achieved documented and
sustained seizure freedom. The relatively small sample size
of 26 eligible patients did not allow for subgroup analyses.
Nonetheless, our study is the largest study to date to evaluate
the localizing value of various z scores in patients who under-
went SISCOM analysis and successful epilepsy surgery.
Earlier studies evaluating SISCOM localization judged
the efficacy of SPECT localization by comparison with sei-
zure foci suggested by other clinical information such as
EEG (Newton et al., 1995). In this situation, it could be dif-
ficult to judge whether false SPECT localization was truly
false, or attributable to incorrect localization of seizure
focus. In our study, we selected patients who became

seizure free after surgery to eliminate this ambiguity. Our
premise is that the surgically resected area by definition
encompasses the ictal-onset zone in patients who experience
sustained postoperative seizure freedom. Based on this pre-
mise we were able to calculate sensitivity and specificity
values. Although this approach is presented as a strength to
the study, it also represents a potential weakness because
the selection of seizure-free patients may exert a bias toward
more localizing SISCOM. Furthermore, SPECT was an
integral part of the presurgical evaluation and did inform the
process of localization and decision for surgery in this retro-
spective study. It should also be noted that the final extent of
surgical resection is based on a number of factors that can-
not be standardized, and the seizure-onset zone, as localized
by a multimodal noninvasive approach, cannot always be
resected in its entirety. With these limitations in mind we
conclude that a threshold of 1.5 provides useful and some-
times unique information, when compared to the conven-
tionally used z-score of 2. It is our hope that this specifically
designed retrospective study will pave the way for future
larger scale studies in consecutive surgical patients, to
examine the localization value of varying z scores on a finer
scale, concordance of SPECT localization with other
modalities, as well as prospective studies to determine the
additive value of varying z scores in presurgical evaluation.
Conclusions
Our study presents evidence to suggest that analysis of
subtracted SPECT studies should not be restricted to only
identifying the largest area of highest hyperperfusion at a
conventional z score of 2. In some patients, varying the SIS-
COM threshold and analyzing SISCOM data at a z score of
1.5 may be helpful in providing complementary or unique
localizing information. Therefore, our study may provide
the impetus for a more dynamic, individualized, and likely
more fruitful review of subtracted SPECT images in
patients with difficult-to-localize intractable focal
epilepsies.
Acknowledgment
We would like to acknowledge Xiaobo Liu at Quantitative Health Sci-
ences for statistical assistance. We would also like to thank the reviewers
for their helpful critique and suggestions.
Disclosure
Christopher Newey, Chong Wong, Guiyun Wu, and Z. Irene Wang have
no disclosures to report. Xin Chen is a senior research scientist at Toshiba
Medical Research Institute U.S.A., Inc. Andreas V. Alexopoulos serves on
the editorial board of Epileptic Disorders, and has received research sup-
port from the National Institutes of Health, Department of Defence, and
American Epilepsy Society, and from UCB and Pfizer Inc. Financial sup-
port for this project was provided by the Cleveland Clinic Department of
Epilepsy. We confirm that we have read the Journal’s position on issues
involved in ethical publication and affirm that this report is consistent with
those guidelines.
799
Varying z Score in SPECT SISCOM
References
Aboian MS, Wong-Kisiel LC, Rank M, Wetjen NM, Wirrell EC, Witte RJ.
(2011) SISCOM in children with tuberous sclerosis complex-related
epilepsy. Pediatr Neurol 45:83–88.
Andersen AR. (1989) 99mTc-D, L-hexamethylene-propyleneamine oxime
(99mTc-HMPAO): basic kinetic studies of a tracer of cerebral blood
flow. Cerebrovasc Brain Metab Rev 1:288–318.
Barba C, Di Giuda D, Fuggetta F, Colicchio G. (2009) Provoked ictal
SPECT in temporal and extratemporal drug-resistant epileptic patients:
comparison of Statistical Parametric Mapping and qualitative analysis.
Epilepsy Res 84:6–14.
Chassagnon S, Namer IJ, Armspach JP, Nehlig A, Kahane P, Kehrli P,
Valenti MP, Hirsch E. (2009) SPM analysis of ictal-interictal SPECT in
mesial temporal lobe epilepsy: relationships between ictal semiology
and perfusion changes. Epilepsy Res 85:252–260.
Dupont P. (2008) Multi-modal imaging in epilepsy: SPECT and PET.
JBR-BTR 91:258–261.
Dupont P, Van Paesschen W, Palmini A, Ambayi R, Van Loon J, Goffin J,
Weckhuysen S, Sunaert S, Thomas B, Demaerel P, Sciot R, Becker AJ,
Vanbilloen H, Mortelmans L, Van Laere K. (2006) Ictal perfusion
patterns associated with single MRI-visible focal dysplastic lesions:
implications for the noninvasive delineation of the epileptogenic zone.
Epilepsia 47:1550–1557.
Hong SB, Joo EY, Tae WS, Cho JW, Lee JH, Seo DW, Suh YL, Hong SC.
(2008) Preictal versus ictal injection of radiotracer for SPECT study in
partial epilepsy: SISCOM. Seizure 17:383–386.
Kaiboriboon K, Lowe VJ, Chantarujikapong SI, Hogan RE. (2002) The
usefulness of subtraction ictal SPECT coregistered to MRI in single-
and dual-headed SPECT cameras in partial epilepsy. Epilepsia 43:408–
414.
Kaminska A, Chiron C, Ville D, Dellatolas G, Hollo A, Cieuta C, Jalin C,
Delalande O, Fohlen M, Vera P, Soufflet C, Dulac O. (2003) Ictal
SPECT in children with epilepsy: comparison with intracranial EEG
and relation to postsurgical outcome. Brain 126:248–260.
Kimura Y, Sato N, Ito K, Kamiya K, Nakata Y, Saito Y, Matsuda H, Sugai
K, Sasaki M, Sugimoto H. (2012) SISCOM technique with a variable z
score improves detectability of focal cortical dysplasia: a comparative
study with MRI. Ann Nucl Med Mar 17. [Epub ahead of print].
Lee SK, Lee SY, Yun CH, Lee HY, Lee JS, Lee DS. (2006) Ictal SPECT in
neocortical epilepsies: clinical usefulness and factors affecting the
pattern of hyperperfusion. Neuroradiology 48:678–684.
Lee JY, Joo EY, Park HS, Song P, Young Byun S, Seo DW, Hong SB.
(2011) Repeated ictal SPECT in partial epilepsy patients: SISCOM
analysis. Epilepsia 52:2249–2256.
Maes F, Collignon A, Vandermeulen D, Marchal G, Suetens P. (1997)
Multimodality image registration by maximization of mutual
information. IEEE Trans Med Imaging 16:187–198.
Matsuda H, Matsuda K, Nakamura F, Kameyama S, Masuda H, Otsuki T,
Nakama H, Shamoto H, Nakazato N, Mizobuchi M, Nakagawara J,
Morioka T, Kuwabara Y, Aiba H, Yano M, Kim YJ, Nakase H, Kuji I,
Hirata Y, Mizumura S, Imabayashi E, Sato N. (2009) Contribution
of subtraction ictal SPECT coregistered to MRI to epilepsy surgery:
a multicenter study. Ann Nucl Med 23:283–291.
McNally KA, Paige AL, Varghese G, Zhang H, Novotny EJ Jr, Spencer SS,
Zubal IG, Blumenfeld H. (2005) Localizing value of ictal-interictal
SPECT analyzed by SPM (ISAS). Epilepsia 46:1450–1464.
Newton MR, Berkovic SF, Austin MC, Rowe CC, McKay WJ, Bladin PF.
(1995) SPECT in the localisation of extratemporal and temporal seizure
foci. J Neurol Neurosurg Psychiatry 59:26–30.
Noachtar S, Arnold S, Yousry TA, Bartenstein P, Werhahn KJ, Tatsch K.
(1998) Ictal technetium-99m ethyl cysteinate dimer single-photon
emission tomographic findings and propagation of epileptic seizure
activity in patients with extratemporal epilepsies. Eur J Nucl Med
25:166–172.
O’Brien TJ, O’Connor MK, Mullan BP, Brinkmann BH, Hanson D,
Jack CR, So EL. (1998a) Subtraction ictal SPET co-registered to
MRI in partial epilepsy: description and technical validation of the
method with phantom and patient studies. Nucl Med Commun 19:
31–45.
O’Brien TJ, So EL, Mullan BP, Hauser MF, Brinkmann BH, Bohnen NI,
Hanson D, Cascino GD, Jack CR Jr, Sharbrough FW. (1998b)
Epilepsia, 54(5):793–800, 2013
doi: 10.1111/epi.12139
800
C. R. Newey et al.
Subtraction ictal SPECT co-registered to MRI improves clinical
usefulness of SPECT in localizing the surgical seizure focus. Neurology
50:445–454.
O’Brien TJ, So EL, Mullan BP, Hauser MF, Brinkmann BH, Jack CR Jr,
Cascino GD, Meyer FB, Sharbrough FW. (1999) Subtraction SPECT
co-registered to MRI improves postictal SPECT localization of seizure
foci. Neurology 52:137–146.
O’Brien TJ, So EL, Mullan BP, Cascino GD, Hauser MF, Brinkmann BH,
Sharbrough FW, Meyer FB. (2000) Subtraction peri-ictal SPECT is
predictive of extratemporal epilepsy surgery outcome. Neurology
55:1668–1677.
Pastor J, Dominguez-Gadea L, Sola RG, Hernando V, Meilan ML, De Dios
E, Martinez-Chacon JL, Martinez M. (2008) First true initial ictal
SPECT in partial epilepsy verified by electroencephalography.
Neuropsychiatr Dis Treat 4:305–309.
Shin WC, Hong SB, Tae WS, Kim SE. (2002) Ictal hyperperfusion patterns
according to the progression of temporal lobe seizures. Neurology
58:373–380.
Van Paesschen W. (2004) Ictal SPECT. Epilepsia 45(Suppl. 4):35–
40.
Van Paesschen W, Dupont P, Sunaert S, Goffin K, Van Laere K. (2007)
The use of SPECT and PET in routine clinical practice in epilepsy.
Curr Opin Neurol 20:194–202.
Varghese GI, Purcaro MJ, Motelow JE, Enev M, McNally KA, Levin AR,
Hirsch LJ, Tikofsky R, Paige AL, Zubal IG, Spencer SS, Blumenfeld H.
Epilepsia, 54(5):793–800, 2013
doi: 10.1111/epi.12139
(2009) Clinical use of ictal SPECT in secondarily generalized tonic–
clonic seizures. Brain 132:2102–2113.
Velasco TR, Wichert-Ana L, Mathern GW, Araujo D, Walz R, Bianchin
MM, Dalmagro CL, Leite JP, Santos AC, Assirati JA Jr, Carlotti CG Jr,
Sakamoto AC. (2011) Utility of ictal single photon emission computed
tomography in mesial temporal lobe epilepsy with hippocampal
atrophy: a randomized trial. Neurosurgery 68:431–436; discussion 436
von Oertzen TJ, Mormann F, Urbach H, Reichmann K, Koenig R,
Clusmann H, Biersack HJ, Elger CE. (2011) Prospective use of
subtraction ictal SPECT coregistered to MRI (SISCOM) in presurgical
evaluation of epilepsy. Epilepsia 52:2239–2248.
Wetjen NM, Cascino GD, Fessler AJ, So EL, Buchhalter JR, Mullan BP,
O’Brien TJ, Meyer FB, Marsh WR. (2006) Subtraction ictal single-
photon emission computed tomography coregistered to magnetic
resonance imaging in evaluating the need for repeated epilepsy surgery.
J Neurosurg 105:71–76.
Wichert-Ana L, de Azevedo-Marques PM, Oliveira LF, Fernandes RM,
Velasco TR, Santos AC, Araujo D, Kato M, Bianchin MM, Sakamoto
AC. (2008) Ictal technetium-99 m ethyl cysteinate dimer single-photon
emission tomographic findings in epileptic patients with polymicrogyria
syndromes: a subtraction of ictal-interictal SPECT coregistered to MRI
study. Eur J Nucl Med Mol Imaging 35:1159–1170.
Wong CH, Mohamed A, Larcos G, McCredie R, Somerville E, Bleasel A.
(2010) Brain activation patterns of versive, hypermotor, and bilateral
asymmetric tonic seizures. Epilepsia 51:2131–2139.
This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy.
Users should refer to the original published version of the material.