Single-bundle versus double-bundle reconstruction for anterior cruciate ligament rupture: a meta-analysis--does anatomy matter?

PURPOSE

To determine whether double-bundle anterior cruciate ligament reconstruction leads to better restoration of anterior and rotational laxity and range of motion than single-bundle reconstruction.

METHODS

A search was performed in the Medline, Embase, CINAHL, and Cochrane databases. All randomized, quasi-randomized, and observational clinical trials that reported the outcome of double- versus single-bundle anterior cruciate ligament reconstruction were included in our meta-analysis. The primary outcomes were anterior laxity (KT arthrometer; MEDmetric, San Diego, CA), pivot shift, and range of motion. Subgroup analyses were performed for more than 2 years' follow-up, anatomic reconstruction, and nonanatomic reconstruction. The quality of the included studies was scored by use of the GRADE Checklist.

RESULTS

Included 12 studies in this meta-analysis, 5 of which were randomized. There was a statistically significant difference in favor of double-bundle reconstruction for anterior laxity (KT arthrometer difference, -0.6 mm), Lachman test (64% risk reduction of positive Lachman), and pivot-shift test (69% risk reduction of positive shift). Similar results were found for the subgroup with more than 2 years' follow-up and anatomic reconstructions. There were no significant differences for the subgroup with nonanatomic reconstructions, except a 2.6 times risk increase of extension deficit with nonanatomic double-bundle reconstruction compared with nonanatomic single-bundle reconstruction. Most of the included studies were found to have at least one serious limitation in study design.

CONCLUSIONS

In comparison with single-bundle reconstruction, double-bundle reconstruction showed less anterior laxity, as measured by the KT arthrometer and Lachman test, and better rotational laxity, as measured by the pivot-shift test. The majority of the included studies had at least one major limitation in study design that decreased the quality of the study.

LEVEL OF EVIDENCE

Level I, meta-analysis.