Validity and generalizability of findings of randomized controlled trials on arthroscopic partial meniscectomy of the knee.

Abstract

The validity and generalizability of evidence from trials on arthroscopic partial meniscectomy (APM) for a ruptured meniscus of the knee has been debated in scientific journals. The aim of this paper was to assess the validity and generalizability of evidence for trials on effectiveness of APM by a novel Benchmarking method; to compare this assessment with established assessment methods; and to make an overall assessment of the current evidence on validity and generalizability of findings. Literature search was undertaken to find all randomized controlled trials. The benchmarking method, the Cochrane method for the assessment of validity of RCTs, and CONSORT method for the assessment of generalizability of findings were used. The data were extracted and checked for accuracy twice. Seven RCTs were found, of which 2 were double blinded. The benchmarking method indicated that only one trial, assessing real-world effectiveness, had recruited patients comprehensively from the catchment area and showed clinically important effectiveness for APM. All trials had deficiencies in reporting of baseline characteristics and adherence to interventions. In 5 trials, the crossover to surgery from conservative treatment arm was between 19 and 36 percent. The benchmarking method indicates that APM may be an effective treatment for meticulously selected patients, or when APM is targeted to those patients who do not respond to the conservative treatment. There is a need for more sham-controlled and real-world effectiveness trials reporting comprehensively patient characteristics and adherence to interventions, preferably in a representative sample of patients living in the recruitment area.