Summary
This study found a 35% incidence of bacterial contamination in ACL reconstruction grafts, with both gram-negative and gram-positive bacteria frequently present. Bacterial DNA presence was linked to graft exposure time and manipulation. While vancomycin pre-soaking reduces infection rates, it may not prevent long-term graft failure due to a broader range of bacterial species.
Abstract
Introduction
The incidence of postoperative infections in anterior cruciate ligament reconstruction (ACLR) is low, ranging from 0.37 to 2.0%. Recent studies have detected bacterial DNA in failed ACLR (Everhart JS et all 2018, Hiller NL et all 2015). Moreover, an association has been described between the presence of bacterial DNA and graft failure and tunnel widening, highlighting the potential impact of BC on long-term success (Flanigan et all, Burnett et all).
The aim of this study was to detect the incidence of autograft BC and bacteria type at the genus level in ACLR. We hypothesize that BC is frequent in the graft.
Methods
A prospective descriptive study of adult patients who underwent ACLR with hamstrings. Three samples were taken from every surgery; (N1) from the graft at the time it was obtained, (N2) from the graft after tibial fixation, and (N3) saline solution on the instrument’s table (control).
Bacteria were detected using 16S rRNA gene next-generation sequencing.
Descriptive statistics were used to summarize data. Fisher’s exact test was used to compare group samples (p<0.05).
Results
Thirty-one patients (21 males) were included, with an average age of 32.2 years. Mean follow-up was 150 days (range 90-208). During the follow-up, no patient had infection, graft failure, or required surgical re-intervention.
Out of the 31 patients, BC was detected in 11(35.4%). BC was found in sample N1 for one patient (1/11) and in sample N3 for another patient (1/11). All 11 patients showed BC in sample N2 (11/11), with statistical significance (p<0.05).
In the N2 samples with BC; 47 types of gram-negative bacteria (GNB) were identified, with an average of 6,8 bacteria per sample and an average relative abundance of 3,97% per bacteria. Moreover, 33 types of gram-positive bacteria (GPB) were found, with an average of 6 bacteria per sample and an average relative abundance of 9,62% per bacteria. All the colonized samples exhibited the presence of both GPB and GNB.
Among the GNB, Pseudomonas was present in 5 out of 11 samples, with a relative abundance of 6.76%. We found Paracoccus colonization in 3 out of 11 samples, with a relative abundance of 17.38%. Photobacterium had the highest relative abundance (23.66%), being present in 2 out of 11 samples. Among the GPB, Streptococcus was present in 8 out of 11 samples, with a relative abundance of 26.23%. We found Staphylococcus in 8 out of 11 samples, with a relative abundance of 20.41%. Lactobacillus ranks third, present in 6 out of 11 samples, with a relative abundance of 12.89%.
Discussion And Conclusion
Graft pre-soaking with vancomycin has demonstrated a decrease in clinical infection rate. However, it might be insufficient for broader bacterial species, that could play a role in long-term graft failure and tunnel widening.
This study found 35% incidence of bacterial contamination. Both GNB and GPB are frequent. The presence of bacterial DNA is related to graft exposure time and manipulation.