Objectives To compare the outcomes between autograft and allograft reconstruction in patients with PCL deficiency. During the extraction of data and its comparison and interpretation in the development of this meta-analysis, the lack of standardisation in patient follow-up with regards to length of follow-up, modalities measured, and reports of adverse events were notable. Many studies were thus excluded due to failure to meet preset inclusion criteria. The subsequent data analysis therefore became limited in its translation towards guiding clinical and surgical practice. The development of a standardised pre and post-operative assessment and follow-up criteria will not only benefit patients, but will also ensure that future systematic reviews conducted will carry a higher impact towards guiding clinical practice.

Method Medline, EMBASE, and the Cochrane Library databases were searched from January, 1980 until December 1 st, 2016 to identify all relevant articles. Clinical outcomes including International Knee Documentation Committee (IKDC), Tegner and Lysholm scores, joint laxity and posterior tibial displacement were evaluated. Dichotomous outcomes were pooled into odds ratios while continuous outcomes were pooled into weighted mean differences (MD) using random effects meta-analysis.

Results We conducted a systematic review looking at outcomes of isolated PCL reconstruction comparing autograft vs allograft sources. Clinical outcomes including International Knee Documentation Committee (IKDC), Lysholm and Tegner scores, joint laxity, and posterior tibial displacement were evaluated. Amongst the 145 unique articles found through the screening process, 25 studies, with a combined patient population of 900, were deemed eligible for inclusion in this study. Post-operative improvement was observed regardless of graft source. Pooled findings revealed that autografts demonstrated a statistically significant post-operative activity as measured by Tegner scores (MD: 0.5, 95% CI 0.03, 0.9; p=0.04) and a reduced posterior laxity (MD: −1.2, 95% CI −1.6,–0.8; p<0.00001).

Conclusions However, despite a statistically significant improvement, there is difficulty establishing a clinically significant improvement. This stems from the absence of a standardised guideline of measuring pre-operative and post-operative functions. One example is the inconsistent usage of IKDC scores between studies, a failure of reporting both pre and post-operative IKDC scores, and subjective reporting as either ‘normal’ or ‘abnormal’ rather than following a protocol. Thus, the development of a systematic approach to assess patients before and after operations, along with establishment of an agreed minimally clinical important difference will lend to more impactful data analysis and ease of generating guidelines.