Summary

Both HP and CCLR offer high union rate and improvement in post-operative pain and function. However complications are higher with hook plates.

Abstract

Background

HP are widely accepted modality in treatment of DDCF, albeit associated with postoperative complications that necessitate re-operation for removal of hardware. However, reconstruction techniques with coracoclavicular ligament is a successful alternative that avoids hardware removal We performed a review to compare clinical outcomes and complications of both HP vs CCLR in DDCF.

Method

This study was conducted in accordance with PRISMA guidelines using Medline and Embase, databases. Only comparative studies of patients with displaced distal third clavicle fractures undergoing HP or CCLR were included. Non-randomised studies were appraised using the NOS tool, and randomised trials were appraised using the RoB 2 tool. Meta-analysis was performed on outcomes reported in three or more studies.

Results

Ten studies met the eligibility criteria with 426 patients (HP 221 vs 205 CCLR) with 61% males. Mean follow-up ranged between five months to 6 years across both groups.
Union rate was reported in nine studies and was excellent in both HP and CCLR (98.5% vs 92.6%). Meta-analysis showed no difference in union rates between HP and CCLR groups [OR 3.03, 95% CI 0.87 – 10.60, p=0.08].

Postoperative complications were reported in all studies. The mean overall complication rate in HP group was 26.7% (n=59) and 6.3%% (n=13) in the CCLR group. The commonest complications reported in the HP group were implant-related discomfort in 16.7% (n=37), soft-tissue-related (functional) problems in 4.9% (n=11) and periprosthetic fractures in 2.7% (n=6). Other complications reported included loss of reduction and implant mal positioning (n=2). The commonest complications reported in the CCLR group were soft-tissue-related problems in 3.4% (n=7), symptomatic non-union in 1.3% (n=3), and failure of fixation in 0.5% (n=1). Other complications reported included suture sinus formation (n=1). The rate of infection was similar in the CCLR group, 1.95% and with HP group, 1.35%. There was a statistically significant difference in the incidence of postoperative complications, favouring CCLR group [OR 5.01, 95% CI 2.31 – 10.84, p<0.0001].

There was a significant improvement in all PROMs post-operatively. The Constant-Murley Score (CMS) was the most reported outcome in five studies. The meta-analysis showed a statistically significant difference in post-operative CMS score, favouring HP group [MD -3.88, 95% CI -6.97 – -0.78, P=0.01]. The Visual Analogue Scale (VAS) was reported in four studies and meta-analysis showed a statistically significant difference in postoperative VAS score, favouring CCLR group [MD 0.33, 95% CI 0.03 – 0.62, p=0.03]. The American Shoulder and Elbow Surgeons (ASES) score was reported in three studies and meta-analysis showed a statistically significant difference in postoperative ASES score, favouring HP group [MD -3.96, 95% CI -6.30 – -1.62, p=0.0009].

Conclusion

DDCF have high union rate with both HP and CCLR. However, incidence of postoperative complications and subsequently re-operation for removal of hardware was significantly higher with hook plates. When discussing treatment options with patients, it is important to highlight complication profile when using HP.