Session III - Polytrauma Fracture Healing

Effect of Commonly Used Locally Applied Antibiotics on Human Fibroblast, Lymphocyte-like, and Osteoblast-like Cells In-vitro

Theodore Miclau, MD; Gayle E. Lester, PhD; Matthew L. Edin, BS; David P. Fitzgerald, BS; Eytan Alpern, MD, Department of Orthopaedic Surgery, University of California, San Francisco, CA

Purpose: There has been growing clinical enthusiasm for local antibiotic administration. Locally applied antibiotics have achieved levels orders of magnitude greater than toxic serum levels without exceeding deleterious systemic concentrations. However, recent studies demonstrated that commonly used antibiotics for implantation, i.e. tobramycin, vancomycin, cefazolin, and ciprofloxacin, all inhibit osteoblast-like cell replication and DNA synthesis at certain concentrations. Reports also have shown that antibiotics may be detrimental to cells of other organ systems including those involved in wound repair and the immune response. The goal of this study is to evaluate the effects of different concentrations of commonly used locally applied antibiotics on human fibroblast, lymphocyte-like, and osteoblast-like cell proliferation.

Materials and Methods: Human dermal fibroblasts (HDF), Jurkat lymphocyte-like cells, and MG-63 osteoblast-like cells were obtained and cultured in antibiotic-free medium. The cells were plated in 96 well culture plates at pre-confluent densities and allowed 24 hours to attach. Dose response curves were established at various concentrations for vancomycin (0, 1000, 2500, 5000, and 7500 µg/ml), tobramycin (0, 200, 400, 800, 1600, and 3200 µg/ml), cefazolin (0, 250, 500, 750, and 1000 µg/ml), and ciprofloxacin (0, 25, 50, 100, 250, and 500 µg/ml). The cells were exposed to the drug for 72 hours at 37° C and analyzed for cell number at 72 hours. After evaluating the dose response curves, a second study was performed to assess the ability of the cells to recover proliferation following the removal of antibiotics. Using the same cell lines and concentrations as the previous study, the cells were treated for 72 hours and allowed to recover in antibiotic-free media for 72 hours. Cell number was again determined. For each incubation time and concentration, 4 wells were individually analyzed for each experiment, and the experiments were repeated 3 times. Within each experiment, outcome measures at each dose were compared to those at zero dose using t-tests that allow for unequal variances. The level of significance for this study was chosen to be p < 0.05 for each experiment. A concentration was considered significant if two of the three experiments demonstrated statistical significance.

Results: The concentrations at which cell number was significantly decreased for the treatment and recovery groups are listed below:

Cell Type	Antibiotic	Treatment (µg/ml)	Recovery (µg/ml)
Dermal Fibroblasts	Vancomycin Tobramycin Cefazolin Ciprofloxacin	5000 1600 250 25	7500 1600 500 25
Lymphocyte-like cells	Vancomycin Tobramycin Cefazolin Ciprofloxacin	2500  400 500 25	5000 1600 500 25
Osteoblast-like cells	Vancomycin Tobramycin Cefazolin Ciprofloxacin	5000 800 250 50	5000 1600 250 25

Discussion: Each of the cell types exhibited levels at which cell proliferation was affected during treatment and recovery periods. Although cell numbers were affected similarly by high antibiotic concentrations across cell types, the levels were dependent upon the cells exposed and antibiotic used. In general, each cell type tolerated higher levels of vancomycin and tobramycin than cefazolin and ciprofloxacin. In addition, the cells appeared to have the ability to recover better upon exposure to higher levels of the former two antibiotics, indicating that they may have less long-term toxicity. Thus, vancomycin and tobramycin may be preferrable to cefazolin and ciprofloxacin for local application. These results are consistent with those results previously reported for the effects of these antibiotics on MG-63 cells. Future study is necessary to confirm these findings invivo.

Conclusion: With at least three of the antibiotics tested, the reported concentrations achieved using locally applied methods exceed the levels found in this study to be detrimental to cells associated with bone and wound healing and infection prevention. These high antibiotic concentrations may not only be detrimental during periods of exposure but after exposure as well. The clinician should, therefore, be aware of the potentially adverse effects of high levels of antibiotics achieved with local application. In addition, as the minimal inhibitory concentration for most sensitive bacteria are well below those levels found to be toxic, efforts should be made to maintain non-toxic but therapeutic levels during local delivery.