RESUMEN
INTRODUCTION: Several studies have reported alterations in the cerebrospinal fluid biomarkers (Abeta-42, T-tau and P-tau proteins), both in Alzheimer's disease (AD) and in mild cognitive impairment (MCI). AIM: To perform a meta-analysis of the diagnostic yield of this technique for the prediction of patients with MCI who are going to progress to AD. MATERIALS AND METHODS: A search was conducted in PubMed and Embase of papers published between 1999 and September 2008, and as a result only prospective studies were included for the systematic review. The sensitivity and specificity for each biomarker were studied separately and also jointly. RESULTS: Of the 12 studies that were included, 6 quantified the Abeta-42 protein, 11 the T-tau protein and seven the P-tau protein. In three of the studies data was obtained from the three biomarkers in combination. The sensitivity of the quantification of the T-tau and P-tau proteins is 82%, with a diagnostic odds ratio of 12.09 (confidence interval 95%, CI 95% = 7.71-18.99; p = 0.1) and 16.29 (CI 95% = 9.69-27.4; p = 0.9), respectively. Alteration of any of the three biomarkers has a specificity of 87%, with a diagnostic odds ratio of 35.97 (CI 95% = 7.8-164.6; p = 0.04). CONCLUSIONS: The isolated alteration of T-tau or P-tau levels in cerebrospinal fluid is very sensitive for differentiating between patients with MCI who are going to develop AD and those who are going to remain stable. Normality of the three biomarkers is a very reliable way of ruling out the progression of AD in patients with MCI.
