Magnetic resonance imaging (MRI) and clinical examination are tools commonly used in the diagnosis of meniscus tears. It has been suggested routine MRI before therapeutic arthroscopy for clinically diagnosed meniscus tears will reduce the number and cost of unnecessary invasive procedures. We designed a systematic review of prospective cohort studies comparing MRI and clinical examination to arthroscopy to diagnosis meniscus tears. Thirty-two relevant studies were identified by a literature review. Careful evaluation by an experienced examiner identifies patients with surgically treatable meniscus lesions with equal or better reliability than MRI. MRI is superior when indications for arthroscopy are solely diagnostic. However, the methods by which such a clinician arrives at a conclusion have not been identified. To create an evidence-based algorithm for the diagnosis of a meniscus tear future investigations should prospectively assess the value of commonly used aspects of the patient history and meniscus tests. MRI is useful, but should be reserved for situations in which an experienced clinician requires further information before arriving at a diagnosis. Indications for arthroscopy should be therapeutic, not diagnostic in nature.