A systematic review and meta-analysis of the diagnostic accuracy of anti-heat shock protein 70 antibodies for the detection of autoimmune hearing loss

OBJECTIVE:\r\nWe reviewed case-control studies concerning the diagnostic accuracy of Heat Shock Protein 70 (Hsp-70) auto antibodies in the detection of immunomediated inner ear disease.\r\nMATERIALS AND METHODS:\r\nWe searched for relevant articles published in English language on PubMed and Scopus up to December 2011. A quality assessment of the retrieved articles was performed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) 2 tool. Pooled data on the accuracy of the test were calculated, where possible.\r\nRESULTS:\r\nThree articles were deemed eligible. Among them, 2 evaluated the relationship between Hsp-70 and immunomediated inner ear disease by using the Western blot, whereas one report used the enzyme-linked immunosorbent assay method. Pooled sensitivity of Western blot test for Hsp-70 was 0.70 (95% confidence interval [CI], 0.59-0.80), with a large heterogeneity (I = 72.7%), and pooled specificity was 0.98 (95% CI, 0.87-1.00), with an I of 61.0%. Pooled positive likelihood ratios (LR) was 14.7 (95% CI, 2.1-104.1; I = 31.4%), and pooled negative LR was 0.32 (95% CI, 0.10-0.70; I = 78.8%). Sensitivity and specificity of enzyme-linked immunosorbent assay test for Hsp-70 auto antibodies was 0.85 (95% CI, 0.55-0.98) and 0.98 (95% CI, 0.86-1.00). Risk of bias was performed by using QUADAS 2 tool, with high scores obtained for patient selection and index test domains and low for the applicability criterion.\r\nDISCUSSION:\r\nThis review shows that studies on autoimmune hearing loss diagnosis based on the detection of Hsp-70 autoantibodies used different inclusion and methodologic criteria and are affected from potential bias. Additional studies are actually required to identify an accurate laboratory diagnostic method for the autoimmune hearing loss.