Surgical approaches for cam femoroacetabular impingement: the use of multicriteria decision analysis

Abstract

Background: Currently, three surgical approaches are available for the treatment of cam femoroacetabular impingement (FAI), namely surgical hip dislocation (SHD), hip arthroscopy (HA), and the miniopen anterior approach of the hip (MO). Although previous systematic reviews have compared these different approaches, an overall assessment of their performance is not available.

Questions/purposes: We therefore executed a multidimensional structured comparison considering the benefits, opportunities, costs, and risk (BOCR) of the different approaches using multicriteria decision analysis (MCDA).

Methods: A MCDA using analytic hierarchical process (AHP) was conducted to compare SHD, HA, and MO in terms of BOCR on the basis of available evidence, institutional experience, costs, and our understanding of pathophysiology of FAI. A preclinical decision-making model was created for cam FAI to establish the surgical approach that better accomplishes our objectives regarding the surgical treatment. A total score of an alternative's utility and sensitivity analysis was established using commercially available AHP software.

Results: The AHP model based on BOCR showed that MO is the best surgical approach for cam FAI (normalized score: 0.38) followed by HA (normalized score: 0.36) and SHD (normalized score: 0.25). The sensitivity analysis showed that HA would turn into the best alternative if the variable risks account for more than 61.8% of the priority during decision-making. In any other decision-making scenario, MO remains as the best alternative.

Conclusions: Using a recognized method for decision-making, this study provides supportive data for the use of MO approach as our preferred surgical approach for cam FAI. The latter is predominantly derived from the lower cost of this approach. Our data may be considered a proxy performance measurement for surgical approaches in cam FAI.

Fig. 1

The basic structure of the analytic hierarchical process. It includes the goal, criteria, and alternatives levels.

Fig. 2

The structure of the BOCR model used for the current study is depicted.

Fig. 3

AHP analysis results are shown. Raw values: priority values for each alternative are retrieved directly from the matrix of the network with no additional calculations. Normal values: the raw priority values for each one of the alternatives are summed and then each element is divided by the sum. The elements in the normalized priority values sum to 1. Ideal values: each raw priority value is divided by the largest so that the best alternative gets a priority of 1 and the others get their proper proportion less than 1.