The review compared the diagnostic performance of 2-deoxy-2-[F-18]fluoro-D-glucose-positron emission tomography (FDG-PET) with and without attenuation correction in cancer patients. No significant difference was found between attenuation-corrected and nonattenuation-corrected full-ring PET, but attenuation correction was found to improve lesion detection for dual-head coincidence PET. These conclusions are likely to be reliable.

To compare the diagnostic accuracy of attenuation-corrected (AC) and nonattenuation-corrected (NAC) 2-deoxy-2-[F-18]fluoro-D-glucose-positron emission tomography (FDG-PET) in cancer patients.

MEDLINE and EMBASE were searched to February 2006; the search strategy was reported. The bibliographies of included studies were reviewed to identify additional studies. No language or date restrictions were applied. Abstracts were excluded.

Study designs of evaluations included in the review

No inclusion criteria were specified for the study design.

Specific interventions included in the review

Studies evaluating FDG-PET with and without attenuation correction and using full-ring (FR) PET or dual-head coincidence (DH) PET scanners were eligible for inclusion.

Reference standard test against which the new test was compared

No inclusion criteria were specified for the reference standard. Measures of diagnostic accuracy (sensitivity and specificity) were calculated for studies using an independent reference standard (histopathology) to confirm diagnosis.

Participants included in the review

Studies of at least 10 oncology patients undergoing imaging were eligible for inclusion.

Outcomes assessed in the review

Included studies were required to report sufficient data to construct 2x2 contingency tables by disease status, or to derive relative lesion detection rates of AC versus NAC imaging.

How were decisions on the relevance of primary studies made?

Two reviewers independently screened for relevance the titles and abstracts of retrieved articles. In cases of uncertainty, the full article was reviewed.

Each reviewer independently assessed the included studies forinternal and external validity using criteria based upon the recommendations of the Cochrane Methods Group in Screening and Diagnostic Tests. The criteria for internal validity were valid reference standard, blinding, avoidance of verification bias and prospective study; the criteria for external validity were participant characteristics, disease spectrum, description of inclusion and exclusion criteria, avoidance of selection bias and standardised index test. Reproducibility was also considered.

Two reviewers independently extracted the data, including those to populate 2x2 contingency tables. Any disagreements were resolved by consensus. Data on the sensitivity and specificity of individual tests and concordance between AC and NAC imaging ('relative lesion detection', defined as the percentage of lesions scoring equally positive or negative with NAC versus AC images) were presented in the review.

How were the studies combined?

Pooled estimates of sensitivity, specificity for AC and NAC, and relative lesion detection for AC versus NAC were calculated. A random-effects model was used where there was evidence of statistical heterogeneity between the studies and a fixed-effect model where there was no evidence of such heterogeneity. The z-test was used to compare measures of diagnostic accuracy between AC and NAC (p<0.05 considered significant).

How were differences between studies investigated?

For the meta-analyses, studies were stratified by imaging technique (FR-FDG-PET or DH-FDG-PET). Between-study heterogeneity was examined using the chi-squared test. Subgroup analyses were conducted for different locations of lesions.

Twelve studies were included in the review. The total numbers of participants and lesions studied were not clear.

The results of the quality assessment were reported in full. Common areas of concern were: description of exclusion criteria; avoidance of selection bias; blinding of test interpretation to other test results and clinical information; reporting of reproducibility.

Diagnostic accuracy of AC and NAC FDG-PET.

The pooled per-patient sensitivities of AC and NAC FR-PET were 64% (95% confidence interval, CI: 52, 74) and 62% (95% CI: 51, 73), respectively (3 studies, 182 patients). The pooled per-patient specificities of AC and NAC FR-PET were 97% (95% CI: 92, 99) and 99% (95% CI: 95, 100), respectively (2 studies, 155 patients). One study (not reported) provided sensitivity and specificity data for DH-PET.

Relative lesion detection with NAC versus AC FDG-PET.

Relative lesion detection of NAC FR-PET versus AC FR-PET was 98% (95% CI: 96, 99; 7 studies, 1,012 lesions). Relative lesion detection of NAC DH-PET versus AC DH-PET was 88% (95% CI: 81, 94; 4 studies, 288 lesions). For FR-PET and DH-PET individually, the relative lesion detection of NAC versus AC was similar across body sites (head/neck, chest and abdomen/pelvis). However, relative lesion detection (NAC versus AC) in the chest and in the abdomen/pelvis was significantly lower with DH-PET than with FR-PET.

No significant difference in sensitivity, specificity, or relative lesion detectability was found between AC and NAC FR-FDG-PET. However, attenuation correction significantly improved lesion detection for DH-FDG-PET.

The review addressed a clearly stated research question and appropriate inclusion criteria were defined. The search strategy was adequate and no language or date restrictions were applied. Appropriate measures to reduce the potential for error and bias in the review process were described and the methodological quality of the included studies was assessed; the quality assessment was reported in full (though not considered in the analysis). Details of the included studies were very limited and the results of individual studies were not reported, which severely hampers interpretation. The methods used to generate pooled estimates of diagnostic performance and to compare imaging methods were broadly appropriate and, taken overall, the authors' conclusions are likely to be reliable.

Practice: The authors stated that the evaluation of both NAC and AC images should remain a part of PET image interpretation.

Research: The authors recommended further research in the area of discrepancies between NAC and AC PET/computed tomography images.

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.