In August 2007 NICE issued its guidance for the treatment of patients with knee osteoarthritis (OA) with arthroscopic lavage. The recommendations stated that referral for arthroscopic lavage and debridement should not be offered as part of treatment for osteoarthritis, unless the person has osteoarthritis with a clear history of ‘mechanical locking’ (not gelling, giving way, or x-ray evidence of loose bodies). The aim of this study was to assess both the application of these guidelines over a four month period and whether this procedure had improved symptoms at first follow-up.

This was a retrospective review from August-December 2011. The total number of arthroscopies performed during this period was obtained from theatre records. Further data was obtained through the hospital's electronic database. The diagnosis of OA was made through the analysis of referral and clinic letters, plain radiographs, MRI reports and operation notes. Only those patients with persisting OA symptoms were included, those with OA and recent history of injury or trauma were excluded. During this time period, 222 knee arthroscopies were performed in total, 99 were identified with persistent OA symptoms.

Having identified these patients, referral letters were further analysed to identify the initial presenting symptom. Of the 99, 50 presented with pain, 28 presented with pain plus another symptom other than locking e.g. stiffness/swelling/giving-way, 21 presented with pain plus mechanical locking. According to current guidelines only these 21 patients should have been offered arthroscopic lavage as a form of treatment. In addition to these findings we identified what procedures had been carried out during arthroscopy for each symptom. Of those presenting with pain, 82% had a washout and debridement, 8% had washout, 4% had partial medial meniscectomy, 4% had lateral patellar release and 2% had partial lateral meniscectomy. Those with pain plus other symptoms not including locking, 82% had washout and debridement, 11% had partial medial meniscectomy, and 7% had a washout. Of those presenting with pain plus mechanical locking, 81% had washout and debridement and 19% had partial medial meniscectomy.

Following the procedure, we analysed the outcome of symptoms at first-follow up. The mean follow-up time was 8 weeks. Of those presenting with just pain, 44% showed improvement, 52% had no change/on-going symptoms, 2% were unknown. Of those with pain plus other symptoms other than locking, 57% showed improvement, 35% had no change/on-going symptoms, 8% unknown. Of those with pain plus mechanical locking, 80% showed improvement, 10% had no change/on-going symptoms, 10% unknown.

The results of this study support the current evidence that unless there are clear mechanical symptoms of locking, the use of arthroscopy in arthritic knee joints should be judicious and the reasons should be clearly documented.