Pulmonary Cryobiopsy: A Modality With Diverse Applications; Systematic Review and Meta-analysis

SESSION TITLE: Procedures 1 - EMN/BT/Rigid/Cryo SESSION TYPE: Original Investigation Poster PRESENTED ON: Wednesday, October 26, 2016 at 01:30 PM - 02:30 PM PURPOSE: Bronchoscopic lung biopsies typically employ forceps for the retrieval of tissue. Recently investigators have studied the utility of cryoprobes for sampling lung tissue during bronchoscopies. In this study we have performed a systematic review and meta-analysis to assess the diagnostic yield and safety of pulmonary cryobiopsies in 3 different clinical settings. METHODS: An Ovid MEDLINE and PubMed search was performed to look for studies that had assessed the performance of pulmonary cryobiopsies. The quality of the eligible studies was assessed using the Quality Assessment, Data Abstraction and Synthesis-2 tool. Data was collected by two authors (MSA, DM) and differences were sorted out with consensus. Meta-analysis was performed using Comprehensive Meta-Analysis software (Version 3). A P-value of <0.05 was considered statistically significant. Inverse variance weighting was used to aggregate diagnostic yield proportions across studies, with the number of subjects in each study representing its weight. Study heterogeneity was assessed using the I2 index. We set I2 index value of >40% as being indicative of significant heterogeneity. If significant heterogeneity was found, random effects model was used. Duval and Tweedie′s test was used to look for publication bias. RESULTS: A total of 19 studies (August 2009 to April 2016) with 1571 patients were selected. 12 of these studies had assessed the diagnostic performance of cryobiopsies for lung parenchymal lesions (10 studies included patients with diffuse parenchymal lung disease, one study included pulmonary nodules and one study included immunocompromised patients with infiltrates). Meta-analysis for these 12 studies revealed an inverse variances weighted diagnostic yield of 79.6% (95% CI 71-85, P <0.001). 7 out of these 12 studies had provided adequate information about mean sizes of biopsy samples. The pooled mean biopsy sample size was 25.6 mm2 (95% CI 15.2-37.9, P <0.001). 4 studies only assessed diagnostic utility of cryobiopsies for endobronchial lesions. For these studies, the inverse variance weighted diagnostic yield was 93.9% (95% CI 91.1-95.9, P<0.001). Finally, 3 studies evaluated the utility of cryobiopsies for evaluation of lung allografts. For these, the pooled mean biopsy sample size was 50.3 mm2 (95% CI 15.9-84.6, P=0.004). All but two studies provided safety data. Overall complication rate was 16%. Most of these complications included self-limited bleeds and small pneumothoraces requiring no further intervention. Only 1 patient died after cryobiopsy. CONCLUSIONS: Pulmonary cryobiopsies have a high diagnostic yield and allow for retrieval of large biopsy specimens. They have been studied and found to be effective for both endobronchial lesions and lesions outside the airways. They are also effective for sampling lung allografts. CLINICAL IMPLICATIONS: Cryobiopsies have a higher diagnostic yield than biopsy forceps and can be considered for first line testing in pathologies such as interstitial lung diseases. They can potentially obviate the need for surgical lung biopsies. However, the high overall complication rate in our analysis is a concern, even though the most of these complications were minor. Therefore, decision regarding employing cryobiopsies needs to be individualized. DISCLOSURE: The following authors have nothing to disclose: Muhammad Ali, Jaskaran Sethi, Divyanshu Mohananey, Naser Yamani, Ahya Ali No Product/Research Disclosure Information