The authors concluded that transbronchial needle aspiration for mediastinal staging of non-small-cell lung cancer is highly specific, but its sensitivity depends on the prevalence of mediastinal metastasis and study methods. Although the search was limited, the review was otherwise well-conducted and the authors' conclusions are likely to be reliable.

To estimate the accuracy of transbronchial needle aspiration (TBNA) for mediastinal staging of non-small-cell lung cancer (NSCLC).

MEDLINE (from inception to April 2004) and EMBASE (from inception to July 2003) were searched. Reference lists of included studies and review articles were screened for additional relevant studies. Authors were contacted for details of unpublished studies. No language restrictions were applied.

Study designs of evaluations included in the review

No inclusion criteria relating to the study design were specified.

Specific interventions included in the review

Studies of TBNA using a flexible bronchoscope for mediastinal staging were eligible for inclusion. The size and type of TBNA needle used varied between studies, as did the number of aspirate passes per lymph node station.

Reference standard test against which the new test was compared

No inclusion criteria relating to a reference standard were specified. The reference standards used in the included studies were mediastinoscopy, mediastinotomy and thoracotomy. Some studies used real-time imaging (computed tomography or endobronchial ultrasound) to guide needle placement during TBNA.

Participants included in the review

Studies of patients with NSCLC that involved at least 10 patients with and/or 10 patients without mediastinal metastasis were eligible for inclusion. Studies were more than 10% of the patients had a diagnosis other than NSCLC were excluded unless they reported data separately for patients with NSCLC. Where reported, the median age of the patients was 60 years and the median proportion of men was 89%. The prevalence of mediastinal lymph node metastasis ranged from 29 to 100% (median 70%).

Outcomes assessed in the review

The studies had to report sufficient data to calculate the sensitivity and specificity to be eligible for inclusion. The review also assessed complications.

How were decisions on the relevance of primary studies made?

Two reviewers selected studies reported in English for inclusion in the review. Any disagreements were resolved with the aid of a third reviewer if required. A single reviewer selected non-English language studies. Authors of abstracts and studies reporting insufficient data on which to assess inclusion were contacted for additional data.

Validity was assessed by considering the technical quality of TBNA and the reference test, application of the reference test, independence of interpretation, clinical description, cohort population and sample size. The authors used criteria adapted from those described by Kent et al. (see Other Publications of Related Interest), along with additional criteria developed by themselves based on clinical experience and by reviewing guidelines. High-quality studies were those that enrolled at least 10 patients with and 10 patients without mediastinal lymph node involvement, surgically confirmed all TBNA results, and used the patient as the unit of analysis.

At least two reviewers assessed the validity of studies reported in English and any disagreements were resolved by discussion. One reviewer assessed all non-English language studies.

One reviewer extracted data on patient characteristics and the sensitivity and/or specificity of TBNA. If possible, data for TBNA patients with and without enlarged lymph nodes on computed tomography scan, and for biopsies performed at hilar, subcarinal, paratracheal or other lymph node stations, were extracted separately. Test characteristics were tabulated separately for studies using real time imaging. A 2x2 table of test performance was extracted for each study, and the true and false positive rates were calculated together with the log diagnostic odds ratio (log DOR). If necessary, 0.5 was added to cells with zero events to enable the DOR to be calculated. Many studies of TBNA did not confirm positive test results surgically and so were unable to report false positive rates or specificity; a weighted inverse kappa was therefore calculated.

How were the studies combined?

A random-effects model was used to pool the sensitivity, specificity, log DOR and inverse kappa. Studies that involved fewer than 10 patients with mediastinal lymph node involvement were excluded when calculating the pooled sensitivity, while those involving fewer than 10 patients without mediastinal lymph node involvement were excluded when calculating the pooled specificity. Summary receiver operating characteristic curves were calculated using the Moses-Littenberg method. Publication bias was investigated using funnel plots based on log DORs and their standard errors.

How were differences between studies investigated?

Sensitivity analysis, discriminate function analyses and meta-regression were used to investigate sources of heterogeneity. The sensitivity analyses included stepwise single study elimination, adjusting the correction factor, and varying the reference standard result in studies that used a suboptimal reference standard. The sensitivity and specificity were compared jointly in studies grouped by quality and prevalence of mediastinal metastases using discriminant function analysis. Multivariate analysis of variance (ANOVA) was used to compare sensitivities and log DORs in studies with high and low prevalence of metastasis and year of publication.

Thirteen studies were included in the review. The median number of patients in each study was 44 (range: 10 to 183).

Studies met between 12 to 23 of the 34 quality criteria.

High-quality studies (5 studies).

The median prevalence of mediastinal metastases was 34% (range: 29 to 60); prevalence was significantly lower than in low-quality studies (p=0.002). The pooled sensitivity was 39% (95% confidence interval, CI: 17, 61) and the pooled specificity was 99% (95% CI: 96, 100). The pooled kappa-1 coefficient showed poor to fair accuracy for false-negative results (30%, 95% CI: 15, 46).

Low-quality studies (8 studies).

The median prevalence of mediastinal metastases was 81% (range: 55 to 100) for the 6 studies that did not use real-time imaging and 83% for the 3 studies that did. These studies did not verify positive results, thus no data on false-positive results were available. The pooled sensitivity was 78% (95% CI: 71, 84). The pooled kappa-1 coefficient showed fair accuracy for false-negative results (40%, 95% CI: 19, 62).

Complications (12 studies).

Two studies reported major complications; these included two major bleeds and one pneumothorax requiring a chest tube. The overall rate of major complications was 0.26% (95% CI: 0.01, 4).

Sensitivity analyses.

The discriminant function analysis showed a statistically significant difference in the joint sensitivity and specificity between the high- and low-quality studies (p=0.002).

Multivariate ANOVA.

This analysis showed that sensitivity was higher in studies with a higher prevalence of lymph node metastasis (less than 60% versus at least 60%): the difference was 60% (95% CI: 51, 69). Sensitivity was also higher in more recently published studies (before 1995 versus 1995 onwards): the difference was 10% (95% CI: 1, 18). However, it should be noted that only one of the high-quality studies showed a disease prevalence that was 60% or more, so this difference could be a result of study quality rather than an association with prevalence.

TBNA is highly specific for the identification of mediastinal metastasis in patients with NSCLC, but its sensitivity depends on the prevalence of mediastinal metastasis and study methods. Sensitivity was lower in populations with a low prevalence of mediastinal metastasis.

The review addressed a clearly focused question that was supported by defined inclusion criteria. The literature search was limited to two electronic databases, but attempts were made to identify unpublished studies and non-English language studies. The authors did assess publication bias, but the methods used were inappropriate for DORs and so should be interpreted with caution. Methods were used to minimise reviewer error and bias in the selection of studies and assessment of validity, but the data extraction was not undertaken in duplicate and this could have resulted in reviewer error. A detailed quality assessment was carried out, the details and results of which are available in additional tables on the journal website (accessed 18/06/2007). See Web Address at end of abstract.

The methods used to pool the results were appropriate and were supported by graphical presentation. The sensitivity analysis should be interpreted with caution given the small number of studies included in the review. Although there was a statistically significant difference in the diagnostic accuracy of high- and low-quality studies, study quality was apparently not taken into account in the multivariate ANOVA; however, the authors clearly discussed the various factors that might have contributed to this difference in accuracy. Despite the limited search, the review was otherwise well-conducted and the authors' conclusions are likely to be reliable.

Practice: The authors stated that TBNA continues to be an appropriate diagnostic test in the sampling of mediastinal lymph nodes, especially if performed concurrently with routine bronchoscopic examination for suspected lung cancer.

Research: The authors stated that large multicentre prospective studies of TBNA should be performed in consecutively enrolled patients with NSCLC, and that future studies should require thoracotomy with systematic sampling of both normal and abnormal appearing lymph nodes at all accessible mediastinal stations to exclude the presence of lymph node metastasis.

Kent DL, Haynor DR, Larson EB, Deyo RA. Diagnosis of lumbar spinal stenosis in adults: a meta-analysis of the accuracy of CT, MR, and myelography. Am J Roentgenol 1992;158:1135-44.

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.