This clearly reported review concluded that single-photon emission computed tomography appears a valuable method in the detection of tumour recurrence in patients treated with radiotherapy for supratentorial glioma. However, as the authors acknowledged, poor methodological quality and small sample sizes of the included studies limit these conclusions. The reliability of the conclusions is potentially further undermined by the limited literature search.

To evaluate the accuracy of single-photon emission computed tomography (SPECT) for the diagnosis of recurrent glioma in patients who have previously been treated with radiotherapy for supratentorial glioma.

PubMed and EMBASE were searched from 1980 to March 2003 for English language studies. The search terms, which were described in full, included a diagnostic filter. The references of relevant publications were screened to identify additional studies.

Study designs of evaluations included in the review

Studies that included at least 6 patients with confirmed tumour recurrence were eligible for inclusion.

Specific interventions included in the review

Studies that evaluated 201-Tl SPECT for the detection of tumour recurrence were eligible for inclusion. The 201-Tl dose ranged from 111 to 148 MBq. The time interval from tracer injection to scanning ranged from 5 to 10 minutes to 3 hours. The reference region of interest was either the mirror image (contralateral hemisphere) or contralateral scalp. The definition of a positive SPECT scan was based on a tumour-non tumour ratio ranging >=0.5 to >=3.5.

Reference standard test against which the new test was compared

Studies in which histopathological confirmation of the suspected brain lesion or cytological examination of cerebrospinal fluid (CSF) for tumour cells was used as the reference standard were eligible for inclusion. Clinical follow-up was not considered to be a valid reference standard.

Participants included in the review

Studies of patients who had previously received radiotherapy for supratentorial glioma and who were clinically suspected to have recurrent tumour growth were eligible for inclusion. Glioma was considered to include astrocytic, oligodendroglial or ependymal tumours, or mixed gliomas, based on the World Health Organization classification of tumours of the nervous system. Patients with various types of glioma were included, although most had high-grade gliomas. The patients had received different types and dosages of radiotherapy. Some patients had also previously received chemotherapy. The mean age of the participants ranged from 43 to 57 years and the overall age from 9 to 84 years. The prevalence of tumour recurrence ranged from 33 to 100%.

Outcomes assessed in the review

Studies had to report sufficient data to construct a 2x2 table of test performance to be included. The outcomes reported in the review were the sensitivity, specificity and diagnostic odds ratio (DOR).

How were decisions on the relevance of primary studies made?

Two reviewers independently selected studies for inclusion in the review. Any disagreements were resolved through consensus, or by referral to a third reviewer if necessary.

Two reviewers independently assessed the methodological quality of the studies based on the following criteria: use of a valid reference test; blind interpretation of the index test and reference standard relative to the results of the other; interpretation of the index test blind to clinical information; verification bias avoided; avoidance of treatment paradox; valid selection of the patients; appropriate clinical setting and patient spectrum; description of imaging information. Any disagreements were resolved through consensus, or by referral to a third reviewer if necessary.

Two reviewers independently extracted relevant data using standardised data extraction forms. The data were extracted as 2x2 tables of test performance. For each study, the sensitivity, specificity and DOR, together with 95% confidence intervals (CIs), were calculated. Researchers were contacted for additional information if insufficient data were reported in the published studies.

How were the studies combined?

A summary receiver operating characteristic (ROC) curve was constructed using the Moses-Littenberg method. Forest plots of the sensitivity and specificity were also constructed. The presence of a threshold effect between sensitivity and specificity was assessed using the Spearman rank test.

How were differences between studies investigated?

Differences between the studies were discussed in the text.

Eight studies (171 patients) were included. The sample size ranged from 6 to 47.

The quality of the included studies was poor and the sample sizes were small. Studies failed to fulfil between 1 and 5 of the internal validity criteria, most commonly those relating to blinding, patient selection, and whether the reference standard was applied before treatment was started. None of the studies provided full information on the imaging protocol.

Summary ROC curves were not constructed because of methodological and statistical heterogeneity. The sensitivity ranged from 43 to 100% and the specificity from 25 to 100%.

SPECT appears a valuable method for the detection of tumour recurrence in patients treated with radiotherapy for supratentorial glioma. However, the poor methodological quality and small sample sizes of the included studies limits these conclusions.

This was generally a well-conducted and clearly reported review. The objective was clearly stated and was supported by defined inclusion criteria. The literature search was limited to two electronic databases, restricted to English-language studies, included a diagnostic filter, and did not include attempts to locate unpublished studies. It is therefore likely that relevant studies have been missed and the review may be subject to language and publication bias. Details of the review process were reported clearly; these included appropriate steps to minimise bias at all stages of the review. A detailed quality assessment was conducted using appropriate criteria and the results were reported and discussed in full.

The decision not to pool the studies was appropriate given the heterogeneity between the studies. It would, however, have been helpful to have included a summary ROC plot to display the trade-off between sensitivity and specificity. The authors' conclusions take account of the poor methodological quality of the included studies and seem appropriate given the data. The reliability of the conclusions is potentially further undermined by the limited literature search.

Practice: The authors did not state any implications for practice.

Research: The authors stated that future studies should assess the accuracy of 201-Tl SPECT and determine optimal thresholds for the detection of tumour recurrence in patients who have previously received radiotherapy for supratentorial glioma.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.