Letter — elucidation and erratum

This article refers to:

Primary arthroplasty is better than internal fixation of displaced femoral neck fractures: A meta-analysis of 14 randomized studies with 2,289 patients

Sir—After the late Professor Johnell and I published our metaanalysis on treatment of displaced femoral neck fractures in Acta Orthopaedica 3/06 (Rogmark and Johnell 2006), a discussion arose amongst colleagues cited in the paper regarding what kind of surgery should be defined as reoperation after internal fixation. One opinion is that only secondary arthroplasty should be reported. In our meta-analysis, we defined reoperation after internal fixation as any kind of secondary open surgery performed on an individual with a failed internal fixation, i.e. THA, hemiarthroplasty, Girdlestone procedure, drainage of deep infections and extraction of screws or pins. The reason for including screw or pin extraction is that in both our randomized clinical trial (Rogmark et al. 2002) and the one performed by Röden et al. (2003), some patients were found to be treated with this procedure only after developing avascular necrosis or nonunion. The choice of a smaller procedure was because of frailty of the patient. From the patient's point of view, the event of secondary surgery itself—and not the extent of the procedure—is experienced as a strain. As stated in the paper, removal of screws or pins after uneventful healing of the fracture is not defined as a reoperation.

Regrettably, there is a typographical error in Table 1; the failure rate after internal fixation in the study of Tidermark et al. (2003) is 19/53 (not 49/53). The misprint occurred during the final layout, and the the correct number is used in the calculations.