Summary
The article investigates the interpretation of positive Cutibacterium cultures in revision shoulder arthroplasty. The authors conducted a multi-institutional study to assess differences in culture methodology and reporting across institutions. The primary objective was to determine the rate of positivity for negative control (NC) samples and positive control (PC) samples with varying bacterial loads of Cutibacterium. The secondary objective was to evaluate whether time to culture positivity and strength of culture positivity could differentiate true positives from NC samples. The study hypothesized that the rate of positivity would vary across institutions, and time to positivity and strength of positivity would differ between PC and NC samples.
Methods:
Participating institutions: 11 American Shoulder and Elbow Surgeons (ASES) Periprosthetic Joint Infection Multicenter Group institutions.
Study design: Each institution received 12 blinded samples, including 10 PC samples and 2 NC samples. The samples were tested within a calendar week using standard culturing methods.
PC isolates: A C. acnes strain was isolated from a revision shoulder arthroplasty patient. Five dilutions were prepared, and the middle dilution represented the clinical sample.
Reporting: Institutions reported culture positivity, time to culture positivity, and strength of culture positivity.
Results:
Rate of culture positivity: All PC samples showed positive growth, with 100% positivity in the four highest dilutions and 91% in the lowest dilution. NC samples grew Cutibacterium in 14% of samples at 2 out of 11 institutions.
Time to culture positivity: PC samples had a mean time to positivity of 4.0 days, with all samples showing growth within 7 days. NC samples had a mean time to positivity of 8.3 days, significantly longer than PC samples.
Strength of culture positivity: The strength of positivity correlated with bacterial load. NC samples had significantly lower strength of positivity compared to PC samples.
Variation between institutions: There were significant differences in time to positivity and strength of positivity between institutions.
GRADE: Low
Risk of bias: How the institutions were selected is unclear, potentially leading to a selection bias. Moreover, laboratories were not asked to alter their standard culturing methods, which might introduce methodological differences and bias.
Indirectness: The study results are limited to the specific settings (11 institutions across the U.S.) and Cutibacterium cultures. The generalizability of the results to other settings or organisms is unclear.
Imprecision: The study provides averages, standard deviations, and p-values, which help us gauge the precision of the results. However, confidence intervals are missing.