Single-Dose, Preoperative Vitamin-D Supplementation Decreases Infection in a Mouse Model of Periprosthetic Joint Infection

Abstract

Background: Despite recent advances, infection remains the most common etiology of arthroplasty failure. Recent work suggests that 25-hydroxyvitamin D (25D) deficiency correlates with the frequency of periprosthetic joint infection (PJI). We endeavored to examine whether 25D3 deficiency leads to increased bacterial burden in vivo in an established mouse model of PJI and, if so, whether this effect can be reversed by preoperative 25D3 supplementation.

Methods: Mice (lys-EGFP) possessing fluorescent neutrophils were fed a vitamin D3-sufficient (n = 20) or deficient (n = 40) diet for 6 weeks. A group of 25D3-deficient mice (n = 20) were "rescued" with 1 intraperitoneal dose of 25D3 at 3 days before surgery. A stainless steel implant was inserted into the knee joint and the joint space was inoculated with bioluminescent Staphylococcus aureus (1 × 10 colony forming units [CFUs]). In vivo imaging was used to monitor bacterial burden and neutrophil infiltration. Blood was drawn to confirm 25D3 levels 3 days before surgery and on postoperative days (PODs) 0 and 14. Mice were killed at POD 21, and CFUs were quantified after culture. Myeloperoxidase (MPO) and β-N-acetylglucosaminidase (NAG) were assayed to look at neutrophil infiltration and activated tissue macrophage recruitment, respectively.

Results: Serum values confirmed 25D3 deficiency and repletion of the 25D3-rescued group. Bacterial bioluminescence and neutrophil fluorescence were significantly greater (p < 0.05) in the 25D3-deficient group. CFU counts from the joint tissue and implant were also significantly greater in this group (p < 0.05). Rescue treatment significantly decreased bacterial burden and neutrophil infiltration (p < 0.05). Compared with the 25D3-sufficient and 25D3-rescued groups, MPO activity was higher (p < 0.02) and NAG activity was lower (p < 0.03) in the 25D3-deficient group.

Conclusions: This study demonstrated in vivo in a mouse model of PJI that (1) 25D3 deficiency results in increased bacterial burden and neutrophil infiltration, and (2) this effect can be reversed with preoperative repletion of 25D3.

Clinical relevance: Considering that >65% of patients undergoing arthroplasty have insufficient or low levels of total 25D and that 25D levels can be replenished with ease using a U.S. Food and Drug Administration (FDA)-approved, oral 25D3 product, 25D deficiency may be an important modifiable risk factor in humans undergoing joint replacement.

Fig. 1

Experimental protocol for 25-hydroxyvitamin D (25D₃) supplementation. Four-week old male lys-EGFP mice were randomly fed either a vitamin D₃-sufficient (n = 20) or D₃-deficient (n = 40) diet. A group of 25D₃-deficient mice (n = 20) were given 25D₃ by intraperitoneal (IP) injection 3 days prior to surgery and switched over to a vitamin D₃-sufficient diet for the remainder of the study to act as a “rescued” group. The remainder of the mice in the 25D₃-sufficient and deficient groups were fed their respective diets for the remainder of the study. POD = postoperative day.

Fig. 2

Figs. 2-A and 2-B Measurement of bacterial burden in vivo using live-animal bioluminescence in 25D₃-deficient (def), sufficient (suff), and rescued (resc) mice. A stainless steel implant was inserted into the right knee joint of the mice (n = 20 per group), and the joint space was inoculated with Xen36 Staphylococcus aureus (1 × 10³ colony-forming units) possessing the bioluminescent construct in a stable plasmid. Fig. 2-A Bacterial counts as measured by S. aureus bioluminescence in vivo (mean maximum flux and standard error of the mean [logarithmic scale]). Fig. 2-B Representative in vivo S. aureus bioluminescence on a color scale overlaid on a grayscale image of the mouse. POD = postoperative day.

Fig. 3

Figs. 3-A and 3-B Confirmation of bacterial burden using colony-forming unit (CFU) counts. At postoperative day 21, mice were killed and bacteria from the implant and surrounding joint tissue were processed for culture. Fig. 3-A Box-and-whisker plots (logarithmic scale) of bacterial counts as measured by CFUs in the surrounding joint tissue of the right knee (p < 0.05 for the 25D₃-deficient mice compared with both 25D₃-sufficient and rescued mice). Fig. 3-B Box-and-whisker plots (logarithmic scale) of bacterial counts as measured by CFUs adherent to the implant (p < 0.04 for the 25D₃-deficient mice compared with both 25D₃-sufficient and rescued mice). The horizontal line within the boxes indicates the median, the outside borders indicate the medians of the upper half and lower half of the data, and the whiskers indicate the minimum and maximum values of the data set.

Fig. 4

Figs. 4-A and 4-B In vivo neutrophil EGFP fluorescence induced by the Xen36 Staphylococcus aureus strain in 25D₃-deficient (def), sufficient (suff), and rescued (resc) mice. A stainless steel implant was inserted into the right knee joint of the mice (n = 20 per group), and the joint space was inoculated with Xen36 S. aureus (1 × 10³ colony-forming units) possessing the bioluminescent construct in a stable plasmid. Fig. 4-A Neutrophil infiltration (neutrophil EGFP fluorescence) as measured by fluorescence in vivo (mean maximum radiant efficiency and standard error of the mean). Fig. 4-B Representative in vivo neutrophil EGFP fluorescence on a color scale overlaid on a grayscale image of the mouse. POD = postoperative day.

Fig. 5

Figs. 5-A and 5-B Myeloperoxidase (MPO) and β-N-acetylglucosaminidase (NAG) activity induced by the Xen36 Staphylococcus aureus strain in 25D₃-deficient, sufficient, and rescued mice. Fig. 5-A Mean MPO activity of the infected joint-tissue specimens (and standard error of the mean [SEM]) (n = 3 mice per group; p < 0.02 for the 25D₃-deficient mice compared with both 25D₃-sufficient and rescued mice). Fig. 5-B Mean NAG activity of the infected joint tissue specimens (and SEM) (n = 3 mice per group; p < 0.03 for the 25D₃-deficient mice compared with both 25D₃-sufficient and rescued mice).