Hyperbaric oxygen therapy augments ciprofloxacin effect against Pseudomonas aeruginosa biofilm infected chronic wounds in a mouse model

Abstract

Introduction: Chronic wounds have a compromised microcirculation which leads to restricted gas exchange. The majority of these hypoxic wounds is infested with microorganisms congregating in biofilms which further hinders the antibiotic function. We speculate whether this process can be counteracted by hyperbaric oxygen therapy (HBOT).

Methodology: Twenty-eight BALB/c mice with third-degree burns were included in the analyses. Pseudomonas aeruginosa embedded in seaweed alginate beads was injected under the eschar to mimic a biofilm infected wound. Challenged mice were randomized to receive either 4 days with 1 x ciprofloxacin combined with 2 × 90 min HBOT at 2.8 standard atmosphere daily, 1 x ciprofloxacin as monotherapy or saline as placebo. The mice were clinically scored, and wound sizes were estimated by planimetry daily. Euthanasia was performed on day 8. Wounds were surgically removed in toto, homogenized and plated for quantitative bacteriology. Homogenate supernatants were used for cytokine analysis.

Results: P. aeruginosa was present in all wounds at euthanasia. A significant lower bacterial load was seen in the HBOT group compared to either the monotherapy ciprofloxacin group (p = 0.0008), or the placebo group (p < 0.0001). IL-1β level was significantly lower in the HBOT group compared to the placebo group (p = 0.0007). Both treatment groups had higher osteopontin levels than the placebo group (p = 0.002 and p = 0.004). The same pattern was seen in the S100A9 analysis (p = 0.01 and p = 0.008), whereas no differences were detected between the S100A8, the VEGF or the MMP8 levels in the three groups.

Conclusion: These findings show that HBOT improves the bactericidal activity of ciprofloxacin against P. aeruginosa wound biofilm in vivo. HBOT in addition to ciprofloxacin also modulates the host response to a less inflammatory phenotype.

Fig. 1

Experimental setup (created with biorender.com) 32 acclimatized mice were given a full thickness necrosis on the lumbar region under general anesthesia. Post-procedure s. c. injections with saline and IVC cages with easily accessible Nutella/buprenorphine, pre-watered chow and heat were allotted to avoid pain, hypothermia and dehydration. P. aeruginosa embedded in seaweed alginate beads was injected under the eschar to mimic a biofilm infected wound. Mice were randomized to receive either ciprofloxacin combined with 2 × 90 min HBOT at 2.8 atm. daily (n = 10), ciprofloxacin as monotherapy (n = 11) or placebo (n = 7) for 4 consecutive days (4 mice were excluded from the study due to unexpected death or as a humane endpoint). Mice were clinically scored, and wound sizes were estimated by planimetry. Euthanasia was performed day 8. Wounds were surgically removed in toto, homogenized and plated for quantitative bacteriology. Homogenate supernatants were used for cytokine analysis.

Fig. 2

Adjunctive HBOT reduces wound bacteriology Quantitative bacteriology differed significantly between the three groups. The lowest median was seen in the group treated with both ciprofloxacin and adjunctive HBOT (2.83·10⁶ CFU/wound (1.05·10⁶; 6.75·10⁶)) and was significantly lower (p < 0.0008) than the median of the single treated group (1.17·10⁷ CFU/wound (3.90·10⁶; 8.25·10⁷)). Both intervention groups differed significantly from the placebo group (1.25·10⁹ CFU/wound (1.07·10⁸; 1.74·10⁹) (p ≤ 0.0001). Each data point represents the CFU in one wound. Data are represented with median and 95% confidence interval.

Fig. 3

Interleukin-1 beta is reduced in the double treated group When the groups were compared only the group treated with both HBOT and ciprofloxacin (3.18·10³ (2.23·10³;5.44·10³) and the placebo group (1.94·10⁴ (8.62·10³;5.12·10⁴) were significantly different with a p-value of 0.0007. Data are represented as the median + 95%CI.

Fig. 4

Osteopontin levels are elevated in the treated groups OPN levels were higher in both treatment groups compared to the placebo group (p = 0.002 and p = 0.004), median and 95%CI were respectively (3.18·10³ (1.28·10⁵;6.13·10⁵), 1.26·10⁵ (8.19·10⁴;1.72·10⁵) and 4.50·10⁴ (1.25·10⁴;1.17·10⁵). Data are represented as median + 95%CI.

Fig. 5

S100A9 levels are elevated in the treated groups The S100A9 levels in both the treatment groups were significantly higher than the level in the placebo group (p = 0.01 and p = 0.008 (1.32·10⁶ (1.11·10⁶;1.46·10⁶), 1.36·10⁶ (1.14·10⁶;1.43·10⁶) and 1.03·10⁶ (7.16·10⁵;1.29·10⁶)). Data are represented as median and 95%CI.

Fig. 6

Wound sizes on sacrifice day were not significantly different between the groups. The group receiving both HBOT and ciprofloxacin had a wound area of 1.9 cm² (1.5;2.2), the group receiving ciprofloxacin had an area of 1.9 cm² (1.6;2.04) and the placebo group had a wound size of 1.6 mm² (0.9;2.39), p = 0.6). Data are represented as median and 95%CI.