Factors associated with symptomatic pseudotumors following metal-on-metal total hip arthroplasty

BACKGROUND: Pseudotumors associated with metal-on-metal hips can be symptomatic or asymptomatic. The purpose of this study was to identify the characteristics of pseudotumors associated with pain. METHODS: A total of 239 large-diameter, metal-on-metal total hip arthroplasties (THAs) were performed in 222 patients. Screening for pseudotumors was performed using magnetic resonance imaging (MRI) in all patients who underwent metal-on-metal THA, and 57 patients with 62 affected hips showed pseudotumors. There were 45 women with 49 hips and 12 men with 13 hips affected, with a mean age of 64 years and a mean body mass index (BMI) of 23.9 kg/m(2). Sixteen hips had symptomatic pseudotumors with pain, and 46 hips were asymptomatic. Pseudotumor size was determined. The anatomical position of pseudotumors was divided into anterior position and posterolateral position. Types of pseudotumors were divided into two types: cystic type; and mixed solid cystic and solid type without a cystic component. The follow-up study of pseudotumors was determined using MRI in 33 patients. The serum cobalt and chromium ion levels were measured in 38 patients after unilateral THA. Univariate and multivariate analyses were performed comparing symptomatic and asymptomatic patients to identify the characteristics of symptomatic pseudotumors. RESULTS: The mean BMI was 25.4 kg/m(2) in symptomatic patients and 23.4 kg/m(2) in asymptomatic patients; a higher BMI was associated with symptoms (P = 0.036). Symptomatic pseudotumors were significantly larger (three-fold) than asymptomatic pseudotumors (1812 mm(2) vs 642 mm(2), P = 0.003). Pseudotumors located in the anterior position were associated with symptoms (P = 0.032), and mixed solid cystic and solid type pseudotumors were associated with symptoms (P = 0.007). A multivariate analysis showed significant differences only in size (R (2) = 0.298, P = 0.031). No asymptomatic patients with pseudotumors became symptomatic during the follow-up period of MRI evaluation. CONCLUSION: Larger size was a significant factor for pain on multivariate analysis.