Complications of arthroscopy of the knee.

Abstract

Although arthroscopy of the knee is a minimally invasive procedure with less risk than endoscopy of the visceral cavities complications can and do occur. The operation is performed in very large numbers worldwide and although the percentage of patients experiencing problems is relatively small the total number is not insignificant. Complications are more likely to occur with more complex procedures such as meniscal repair, synovectomy and intraarticular reconstruction of both the anterior (ACL) and posterior (PCL) cruciate ligaments. DeLee 1 reported a retrospective review of 118 590 arthroscopies carried out by the Arthroscopy Association of North America (AANA). There were 930 complications (0.8%). It was felt that this probably underestimated the total number of complications since it was a retrospective review based on surgeons’ memories. In 1986 Small 2 carried out a retrospective review of 395 566 arthroscopies. There was an overall complication rate of 0.56%. The percentage was higher in the more complex procedures, 2.4% in meniscal repair and 1.8% in reconstruction of the ACL. In a further prospective study in 1998 3 reviewing 10 262 procedures he found an overall complication rate of 1.68%. The most common complications in this study were haemarthrosis (60.1%), infection (12.1%), thromboembolic disease (6.9%), anaesthetic complications (6.4%), instrument failure (2.9%), complex regional pain syndrome 1 (CRPS 1) (2.3%), ligament injury (1.2%) and fracture or neurological injury (0.6% each). Sherman et al 4 retrospectively reviewed 2640 arthroscopies performed by four surgeons. Overall, there were 216 complications (8.2%) of which 126 were designated as major and 97 as minor. The major complications were infection, haemarthrosis, adhesions, effusions, cardiovascular and neurological problems, CRPS 1 and instrument breakage, and the minor were difficulties with wound healing and ecchymoses. A number of predisposing factors were recognised. There was a higher incidence of neurological complications and CRPS 1 in patients with an industrial injury. As would be expected the lowest complication rate occurred with diagnostic arthroscopy. A higher overall complication rate occurred with partial medial meniscectomy and the incidence of haemarthrosis was highest in this procedure. Instrument breakage occurred most frequently during partial lateral meniscectomy. The largest number of problems with wound healing occurred after abrasion arthroplasty and lateral release was the commonest cause of adhesions. Slightly surprisingly, the complication rate was unrelated to the experience of the surgeon. The gender of the patient and the use of a tourniquet also had no influence. The length of time of inflation of the tourniquet did appear to be significant, with a higher risk in those patients who had a tourniquet time of 60 minutes or longer. In addition patients over 50 years of age were more likely to have complications.