Study designs of evaluations included in the review
The authors did not state whether any specific study designs were included. The studies had to include at least six participants.
Specific interventions included in the review
No inclusion criteria relating to the intervention were specified. Most reports used technetium-99m-labelled hexamethylpropyleneamineoxime (99m-Tc-HMPAO). For this reason, SPECT studies using radiopharmaceuticals were analysed as a separate group. SPECT images were classified as unifocal ipsilateral to electroencephalography (EEG), unifocal contralateral to EEG, multifocal ipsilateral, multifocal bilateral, and normal.
Reference standard test against which the new test was compared
No inclusion criteria relating to the reference standard test were specified. The diagnostic performance of SPECT was evaluated separately, in the review, with respect to EEG data and surgical outcome. Surgical outcome was classified as good or poor, where good was defined as the patient being either seizure-free or experiencing a greater than 90% reduction in seizure frequency (studies were only included in this analysis if a minimum of 12 months' surgical follow-up was reported).
Participants included in the review
To be included in the meta-analysis, the patients had to have had a localisation-related epileptic syndrome, and at least an interictal EEG-documented epileptiform abnormality. The patients were categorised as:
medically refractory, medically responsive, or if not explicitly stated, not refractory;
adults or paediatrics (aged less than 13 years); or
of general or special patient populations, e.g. those with normal CT scans, those with unilateral EEG findings, or those excluded due to ambiguous EEG or SPECT results.
Outcomes assessed in the review
No outcome-related inclusion criteria were specified. The review calculated the sensitivities and false-positive rates for SPECT localisation were assessed relative to diagnostic evaluation and surgical outcomes.
With respect to unilateral EEG, sensitivity was defined as: the fraction of unifocal ipsilateral SPECT findings among all unilateral EEGs; unifocal or multifocal SPECT findings ipsilateral to EEG localisation. False positives with respect to EEG were defined as: unifocal SPECT finding contralateral to EEG; unifocal or multifocal SPECT finding contralateral to EEG.
With respect to surgery, sensitivity was defined as the fraction of unifocal SPECT findings ipsilateral to the operative side, among all good surgical outcomes. As with the comparison to EEG, both narrow (unifocal SPECT finding), and broad (unifocal and multifocal SPECT findings) were used to define sensitivities and false positive rates.
Wherever possible, patient-level information was used to classify results, rather than reported summary measures.
How were decisions on the relevance of primary studies made?
Decisions on the relevance of the primary studies were made by consensus.