The Journal of Arthroplasty, ISSN: 0883-5403, Vol: 37, Issue: 3, Page: 495-500

Direct Anterior Approach for Primary Total Hip Arthroplasty Lowers the Risk of Dislocation Compared to the Posterior Approach: A Single Institution Experience

Jacob A. Haynes; Robert H. Hopper Jr.; Henry Ho; James F. McDonald III; Nancy L. Parks; William G. Hamilton
Hip

Background

One purported benefit of the direct anterior approach (DAA) for total hip arthroplasty (THA) is a lower rate of postoperative dislocation.

Methods

An institutional database was used to identify 8840 primary THAs performed from 2003 to 2020 including 5065 (57%) performed using the DAA and 3775 (43%) performed via the posterior approach (PA). Direction and mechanism of dislocation were determined from chart review. Outcomes were compared using Kaplan-Meier survivorship with dislocation as the endpoint and a Cox multivariate regression was used to investigate factors associated with dislocation. The mean follow-up was 1.7 ± 2.0 years for the DAA and 3.1 ± 3.3 years for the PA.

Results

The 0.5% (26/5065) incidence of dislocation among DAA hips was significantly less than the 3.3% (126/3775) among PA cases (P < .001). The majority of dislocations were posterior (DAA 57%, PA 79%) and occurred during activities of daily living (DAA 82%, PA 77%). Five-year survivorship was significantly higher for the DAA group compared to the PA group (99.1% vs 95.4%, P < .001). Dislocation risk was 4.9 times higher for the PA compared to the DAA (hazard ratio = 4.9, 95% confidence interval = 3.2-7.5, P < .001). Increasing head diameter reduced the risk (hazard ratio = 0.70, 95% confidence interval = 0.57-0.86, P < .001). The 0.2% incidence (10/5065) of revision for instability among the DAA group was significantly lower than the 1.1% (43/3775) rate for the PA group (P < .001).

Conclusion

Compared to primary THAs performed with the PA, DAA cases had a lower risk of dislocation, higher survivorship with dislocation as an endpoint, and a lower risk of revision for instability in this single institution cohort.

Link to article