We assessed the diagnostic utility of 3 CSF biomarkers-14-3-3 protein, total tau (T-tau), and neuron-specific enolase (NSE)-from the same lumbar puncture to distinguish between participants with neuropathologically confirmed sporadic Creutzfeldt-Jakob disease (sCJD, n = 57) and controls with nonprion rapidly progressive dementia (npRPD, n = 41). Measures of diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value, as well as logistic regression and area under the receiver operator curve (AUC), were used to assess the ability of these CSF biomarkers, alone or concomitantly, to predict diagnosis. In a subcohort with available MRI (sCJD n = 57, npRPD = 32), we compared visual assessment of diffusion-weighted imaging MRI sequences to these CSF biomarkers. MRI was the best predictor, with an AUC of 0.97 (confidence interval [CI] 0.92-1.00) and a diagnostic accuracy of 97% (CI 90%-100%). Of the CSF biomarkers, T-tau had a higher diagnostic accuracy (79.6%) than 14-3-3 (70.4%, CI for difference 8.7%, 9.7%; p = 0.048) or NSE (71.4%, CI for difference 7.6%, 8.7%; p = 0.03).