Experimental beneficial effect of rosuvastatin on burn wound healing in a rat model

Abstract

BACKGROUND

In addition to their primary lipid-lowering effects, statins also exhibit pleiotropic properties, including anti-inflammatory, immunomodulatory, antimicrobial, antioxidative, and angiogenic effects, all of which promote wound healing. Rosuvastatin, a synthetic hydrophilic statin, has recently been proposed to enhance wound healing by stimulating angiogenesis and accelerating tissue regeneration. Its hydrophilic nature, longer half-life, greater hepatoselectivity, and better efficacy/safety than other statins are believed to contribute to its superior efficacy in wound repair, as suggested by promising preliminary results. However, current data on its use remain limited, necessitating further preclinical and clinical studies to thoroughly investigate this novel treatment option for burns.

AIM

To investigate the effects of rosuvastatin and its mechanism of action on burn wound healing process in an experimental study.

METHODS

Ninety male Wistar albino rats aged 12-16 weeks were randomly assigned to three groups of 30, subjected to burn using specific stainless steel sealer. Burn eschar was removed the following day applying topical rosuvastatin cream (study), Eucerin cream (placebo), normal saline (control), and sterile wound dressing. Each group was divided into three subgroups of ten according to sacrifice day (3^rd, 6^th, 9^th). C-reactive protein (CRP), tumor necrosis factor-alpha (TNF-α), interleukin (IL)-1β, IL-6, digital assessment of burn healing, and histopathology were performed.

RESULTS

Using the value on the third day in the control group as the baseline, reductions were measured on the sixth and ninth days. Similarly, reduction values were recorded on the third, sixth, and ninth days in the study group. A statistically significant reduction was observed in the study group compared to the control group, with greater reductions corresponding to later sacrifice days (3^rd, 6^th, and 9^th) in CRP (P < 0.01), TNF-α (P < 0.01), IL-1β (P < 0.01), and IL-6 (P < 0.01) levels and burn size (P < 0.01). Histopathology revealed a statistically significant reduction in inflammatory infiltration, coagulative necrosis, and microhemorrhage (P < 0.01). Conversely, statistically significant increases in neovascularization (P = 0.012) and fibroblastic reactions (P = 0.023) were noted. No adverse effects or deaths were observed.

CONCLUSION

Rosuvastatin reduces inflammation by lowering TNF-α, IL-1β, IL-6, CRP while increasing neo-angiogenesis, fibroblast reactions and microenvironmental protection in burn wounds. These effects promote repair and positively impact burn wound healing.

Key Words: Experimental skin burns; Statins; Rosuvastatin; Burn wound healing; Skin burn treatment; Inflammation; Anti-inflammatory rosuvastatin action

Core Tip: Overall, the data from this original preclinical study confirm that rosuvastatin does not act unidimensionally but simultaneously influences the inflammatory, angiogenic, and reparative phases of burn wound healing. These findings are consistent with data from other statins and expand our understanding of the mechanisms of action of rosuvastatin, highlighting it as a promising therapeutic agent for treating skin burn wounds. Its superiority is attributed to its unique effect profile. The administration of rosuvastatin was safe in this experimental model, with no observed adverse effects. However, further preclinical and clinical studies to assess its safety and efficacy are warranted.

INTRODUCTION

In addition to their primary role in lipid reduction, statins have pleiotropic effects, including anti-inflammatory, immunomodulatory, antimicrobial, antioxidative, and proangiogenic effects, all of which contribute to enhanced wound healing[1-5]. Rosuvastatin, a synthetic hydrophilic statin, has recently been suggested to facilitate skin wound repair by promoting angiogenesis and accelerating tissue regeneration[6-9]. Its hydrophilic properties, extended half-life, increased hepatoselectivity, and favorable efficacy and safety profile compared with other statins are hypothesized to explain its superior performance in wound healing, as indicated by preliminary findings[8,10]. Nevertheless, existing evidence remains limited, underscoring the need for further preclinical and clinical investigations to comprehensively evaluate this emerging therapeutic approach for burn injuries.

Burn injuries pose a major worldwide public health challenge and can result in serious complications. The link between the delayed healing of burn wounds and higher mortality rates has motivated many researchers to develop new treatments focused on accelerating burn wound recovery[8,10]. The healing process of wounds, particularly burns, is a complex, multistage biological process involving hemostasis, inflammation, cell proliferation, extracellular matrix formation, new blood vessel growth, and tissue remodeling[11-14].

Statins not only reduce cholesterol levels but also promote wound healing[15-19]. Statins studied include lovastatin[20,21], simvastatin[22-26], pravastatin[27,28], fluvastatin[29], and atorvastatin[30-33], and more recently, rosuvastatin, a synthetic hydrophilic statin shown to be effective in wound healing[34-38] and bone healing[39,40]. Specifically, simvastatin[41-45] and atorvastatin[46,47] have been used in burn wound healing, whereas rosuvastatin has been applied only in skin wound healing[10,12,34]. Because the pathophysiology, healing process, and treatment of burn wounds differ from those of other types of wounds[48] and considering the involvement of autophagy and apoptosis[49], examining the effects of rosuvastatin specifically on burn wound healing is important. Hydrophilic rosuvastatin is reported to be more effective than the lipophilic statins simvastatin and atorvastatin[20]. Its strong bioactivity and extended half-life enable prolonged regulation of endothelial and inflammatory reactions[30].

Although rosuvastatin primarily targets the liver, there are concerns about its multiple effects in other tissues where its uptake is low. Nevertheless, various studies have demonstrated the pleiotropic effects of rosuvastatin, such as the inhibition of the isoprenylation of Rho and Ras family proteins. These proteins are crucial for conveying numerous intracellular signals from the cell membrane to the nucleus. Therefore, by blocking mevalonic acid synthesis, the production of isoprenes is reduced, which subsequently prevents the activation of Rho and Ras family proteins[20].

Recently, stem cells[26,33,41] and platelet-rich plasma[10,35] combined with rosuvastatin have shown promising results in enhancing wound and bone healing[50]. Different statin delivery methods have been utilized to improve absorption and extend release duration[13,15,16,24,34], including the incorporation of rosuvastatin into biological scaffolds[17,22,25,51].

This preclinical study aimed to examine the effects of rosuvastatin and elucidate its underlying mechanism of action on the burn wound healing process using an experimental rat model. Simvastatin and, to a lesser extent, atorvastatin are perhaps the only statins for which the effects on burn wound healing have been investigated. These studies have yielded some positive effects and encouraging results.

To date, the effect of rosuvastatin on partial-thickness burn wounds has not been studied. This research is of great interest because rosuvastatin, as the newest generation of statins, is characterized by fewer side effects and increased efficacy. Therefore, this study is original because it is the first to investigate the effect of rosuvastatin in an experimental model of partial-thickness skin burn wounds in rats.

MATERIALS AND METHODS

Animals

The animals were handled according to the science-based guidelines for Laboratory Animal Care published by the National Research Council Institute for laboratory animal research in November 2003. The experimental protocol was approved by the Medical School of Aristotle University of Thessaloniki (approval No. 11525-12/05/2023) and the Department of Animal Health, Directorate of Veterinary Policy, Prefecture of Thessaloniki, Greece [approval No. 608344(2695)].

The experimental part of the study was conducted at the Laboratory of Development-Breeding of Animal Models and Biomedical Research, School of Veterinary Medicine, Aristotle University of Thessaloniki (registration number EL-54-BIOexp-10). Ninety male Wistar albino rats, aged 12-16 weeks and weighing approximately 350-400 g, were used in the study. The rats were individually housed in ventilated cages connected to a central ventilation circuit. The cages were maintained at a stable room temperature of 23 °C, with humidity levels between 50% and 60%, noise levels below 50 dB, and an artificial 12-hour light/dark cycle. The rats had ad libitum access to tap water and a standard chow diet. The experimental protocol was designed to minimize pain, discomfort, and distress in the animals. The sample size was calculated using G*Power software on the basis of sample sizes from earlier studies with comparable experimental designs and methods.

Animal groups

The rats were randomly assigned to three main groups (the study group, placebo group and control group) of 30 rats each according to the use or absence of rosuvastatin cream. The animals in all the groups were subjected to burn-causing procedures. These main groups were divided into three subgroups of 10 rats each according to the day of sacrifice (3 days, 6 days, or 9 days after surgery). This study focused on the healing effects of the substance on the 3^rd, 6^th, and 9^th days following the burn. After the burn was induced, the animals were divided on specific days to observe the wound healing process. In the study group, the burn surface was treated with rosuvastatin cream into the Eucerin base, whereas the placebo group received Eucerin cream alone. In the control group, the wounds were washed with normal saline (NaCl 0.9%).

Preparation of cream formulations

For the study group, the formulation was prepared by incorporating twenty tablets of 20 mg rosuvastatin (Rosuvastatin/Actavis F.C., Reykjavik, Iceland) into 20 g of Eucerin cream base (Beiersdorf AG, Hamburg, Germany), followed by thorough mixing and homogenization to ensure uniform distribution. In contrast, the placebo formulation consisted exclusively of 20 g of the Eucerin cream base without the addition of any active pharmaceutical ingredient.

Operative procedure, following care and euthanasia

All the rats were anesthetized with intraperitoneal injections of ketamine hydrochloride (Ketamin-Actavis, Institute of Pharmaceutical Research and Technology, Athens, Greece) and dexmedetomidine (Dexmedetomidine, Ever Valinject, Oberburgau, Austria) at doses of 75 mg/kg and 0.5 mg/kg, respectively. Analgesia was provided with an intramuscular injection of tramadol (Tramal, Vianex, Athens, Greece) at a dosage of 25 mg/kg.

The animals were immobilized and weighed, and their dorsums were shaved on a surface of at least 5 cm². The skin was then prepared with 10% povidone iodine. To induce the burn, a specialized metal stamp measuring 17 mm in diameter and 2.5 mm in thickness and weighing 200 g was utilized. These dimensions facilitate efficient heat transfer. The stamp is made of stainless steel and features a wooden handle. It was heated to 85 °C and then applied to the skin on the back for 10 seconds to produce a deep partial-thickness burn. The next step involved washing the wound with normal saline, taking photographs of the burn wounds, and documenting the burn surface. Afterward, the wounds were covered with sterile dressings. All surgical procedures were conducted by the same surgeon on a warm, stable dissection table in a sterile environment using sterile surgical instruments. On the first day after the burn, the eschar was removed, and topical treatments were applied. Each day, 200 μL of a mixture of Eucerin cream with rosuvastatin, Eucerin cream, or normal saline was applied to the burn wounds on the basis of the group assignment of each rat. The wounds were then covered with sterile dressings.

Deep tissue excision at the burn site was performed on the scheduled day (3 days, 6 days, or 9 days after surgery) under the same anesthesia as the initial surgery. Blood and burn tissue samples were collected for biochemical and histopathological examinations, and the burn size was photographed for assessment. The rats were then euthanized. A portion of each blood sample was placed in a tube with ethylenediaminetetraacetic acid and centrifuged at 5000 rpm for 15 minutes at +4 °C. The serum samples were then transferred to Eppendorf tubes, immediately frozen in liquid nitrogen, and stored at -80 °C until biochemical analysis.

Biochemical analysis

The serum levels of oxidative stress markers and inflammatory response C-reactive protein (CRP) [rat CRP enzyme-linked immunosorbent assay (ELISA) kit, BD Biosciences, San Diego, CA, United States], tumor necrosis factor-alpha (TNF-α) (rat TNF alpha ELISA kit, Antibodies.com Europe AB, Stockholm, Sweden), interleukin (IL)-1β (rat IL-1β ELISA kit, Antibodies.com Europe AB, Stockholm, Sweden), and IL-6 (rat IL-6 ELISA kit, Antibodies.com Europe AB, Stockholm, Sweden) were assessed using ELISA.

Wound size assessment

The size of the wound was assessed using digital photography. The burn surface area was measured using ImageJ software, version 1.45 (National Institutes of Health, United States). The burn areas captured in the photographs were utilized to calculate the wound healing rate using the following formula based on the planimetric method: Wound healing rate = (burn surface area on day x/burn surface area on day 0) × 100.

Histopathological examination

The burn tissue samples were fixed in 10% neutral buffered formalin and embedded in paraffin blocks. Histological sections 2-3 μm in thickness were obtained and then subjected to hematoxylin-eosin staining. The histopathology assessment was performed using a light microscope (Nikon eclipse 50i, Nikon Corporation, Tokio, Japan) and a digital camera (Nikon DS-5 ML1, Nikon Corporation, Tokio, Japan) at 200 × magnification. Inflammatory infiltration, neovascularization and fibroblast activity were assessed using scores ranging from 0 to 3 (0 = absent, 1 = moderate, 2 = severe).

Statistical analysis

Linear regression analyses were conducted to examine the effects of treatment (rosuvastatin, Eucerin cream or saline) and time (sacrifice days 3, 6, and 9) on the levels of the four inflammation markers (CRP, TNF, IL-1β, and IL-6) as well as on burn size. For each dependent variable, an ordinary least squares linear regression model was employed, using dummy variables to categorize both treatment and sacrifice day. Specifically, the following dummy variables were employed: (1) x₁ = 1, if rosuvastatin cream is administered; otherwise, x₁ = 0; (2) x₂ = 1, if Eucerin cream is administered; otherwise, x₂ = 0; (3) The reference group for the treatment was the administration of normal saline. x₃ = 1 on sacrifice day 6; otherwise, x₃ = 0; and (4) x₄ = 1 on sacrifice day 9; otherwise, x₄ = 0. The reference group for time is sacrifice day 3. Therefore, the complete reference group in the models consisted of the experimental animals that received normal saline and were sacrificed on the third day (x₁ = x₂ = x₃ = x₄ = 0).

In cases where statistically significant interactions between the independent variables were identified, the corresponding interaction terms were included in the final model. Conversely, when the interaction terms were not statistically significant or resulted in a poorer fit on the basis of the Akaike information criterion (AIC) and Bayesian information criterion (BIC) statistical criteria, an additive model without interactions was used.

For each model, the regression equation is presented along with the estimated coefficients, the values of the coefficient of determination (R² and adjusted R²), the F test statistic for the overall significance of the model, and the corresponding P values. The interpretation of the results is based on calculating the predicted mean responses for each experimental condition relative to the reference group.

For the variables of inflammatory infiltration, neovascularization, and fibroblastic reaction - each recorded on an ordered categorical scale (0 = absent, 1 = moderate, 2 = severe) - multinomial ordinal logistic regression was applied. The group (control as the reference, placebo, study), the day of sacrifice (day 3 as the reference), and the interaction terms between group and day were included as independent variables. All statistical analyses were conducted using RStudio statistical software (Boston, MA, United States). Statistical significance was set at P < 0.05. A biomedical statistician performed the statistical review of the study.

RESULTS

CRP

The linear model for CRP does not include interaction terms because these terms were not statistically significant, and the model fit improved without them, as indicated by lower AIC and BIC values. The model equation is CRP = 21.0028 - 6.0037x₁ - 2.4567x₂ - 4.5233x₃ - 9.5600x₄. The R² of the model above is 0.92, indicating that 92% of the variance in the CRP is explained by the independent variables. The adjusted R² is 0.9163, and the overall model fit is statistically significant [F (4, 85) = 244.5, P < 0.001].

In more detail, the expected CRP concentration for the reference group (normal saline, day 3) was 21.0028 units (P < 0.001). On the same day, the administration of rosuvastatin reduced the CRP concentration by 5.75 units (P < 0.001). Correspondingly, the administration of Eucerin cream reduced the CRP concentration by 2.4567 units (P < 0.001). When normal saline treatment was continued, sacrifice on day 6 was associated with a reduction in CRP levels of 4.8700 units (P < 0.001), whereas sacrifice on day 9 was associated with a reduction of 9.6900 units (P < 0.001). Overall, the results indicate that both rosuvastatin administration and the timing of sacrifice are associated with a statistically significant reduction in CRP levels, with the greatest effect observed when rosuvastatin is administered and when sacrifice occurs later. Changes in CRP levels according to the rat group and day of sacrifice are depicted graphically in Figure 1A, with corresponding descriptive statistics presented in Table 1.

Figure 1 Graphical presentation. A: C-reactive protein changes by rat group and day of sacrifice. The study group showed a significant and gradual decrease in C-reactive protein levels on later sacrifice days compared to the control and placebo groups (P < 0.01), confirming a reduction in the inflammatory response; B: Graphical presentation tumor necrosis factor-alpha changes by rat group and day of sacrifice. The study group showed a significant and gradual decrease in tumor necrosis factor-alpha levels on later sacrifice days compared to the control and placebo groups (P < 0.01), confirming a reduction in the inflammatory response; C: Graphical presentation interleukin-1β changes by rat group and day of sacrifice. The study group showed a significant and gradual decrease in interleukin-1β levels on later sacrifice days compared to the control and placebo groups (P < 0.01), confirming a reduction in the inflammatory response; D: Graphical presentation interleukin-6 changes by rat group and day of sacrifice. The study group showed a significant and gradual decrease in interleukin-6 levels on later sacrifice days compared to the control and placebo groups (P < 0.01), confirming a reduction in the inflammatory response; E: Graphical presentation burn wound size changes by rat group and day of sacrifice. The study group showed a significant and gradual decrease in burn wound size on later sacrifice days compared to the control and placebo groups (P < 0.01), confirming a better healing process. CRP: C-reactive protein; TNF-α: Tumor necrosis factor-alpha; IL-1β: Interleukin-1β; IL-6: Interleukin-6.

Table 1 Descriptive statistics analysis of C-reactive protein changes by rat group and day of sacrifice.

Group	Day	mean	Range	Median	IQR	SD
Group placebo	3	18.22	5.85	18.26	1.56	3.85
Group control	3	21.04	1.77	21.10	0.82	4.22
Group study	3	15.29	5.29	15.52	1.70	4.52
Group placebo	6	13.71	5.65	13.40	1.24	3.85
Group control	6	16.53	6.40	16.57	1.59	4.22
Group study	6	10.74	4.37	10.48	2.00	4.52
Group placebo	9	9.63	2.06	9.42	1.30	3.85
Group control	9	11.35	4.52	11.48	0.87	4.22
Group study	9	4.89	3.43	4.98	1.66	4.52

IQR: Interquartile range.

TNF-α

The linear model for TNF-α includes both main effects and interaction terms because statistically significant interactions between the independent variables were identified. The model equation is as follows: TNF-α = 35.288 - 11.020x₁ - 4.587x₂ - 7.832x₃ - 12.513x₄ - 2.398x₁x₃ - 4.679x₁x₄ - 0.252x₂x₃ - 3.405x₂x₄. The R² of the model is 0.9609, indicating that 96.09% of the variation in TNF-α is explained by the model. The adjusted R² is 0.957, and the overall fit of the model is statistically significant [F (8, 81) = 248.7, P < 0.001].

In more detail, For the reference group (normal saline, day 3), the expected TNF-α concentration was 35.2880 pg/mL (P < 0.001). On the same day, the administration of rosuvastatin reduced the TNF-α concentration by 11.0200 pg/mL (P < 0.001), whereas the administration of the Eucerin cream resulted in a reduction of 4.5870 pg/mL (P < 0.001). Compared with normal saline treatment, sacrifice on day 6 resulted in a reduction of 7.8320 pg/mL (P < 0.001), whereas sacrifice on day 9 resulted in a reduction of 12.5130 pg/mL (P < 0.001). Owing to interactions, the changes in the combinations of treatment and sacrifice day are more complex. Regarding rosuvastatin on sacrifice day 6, the expected TNF-α concentration decreased by 11.0200 + 7.8320 + 2.3980 = 21.2500 pg/mL (P = 0.036). Regarding rosuvastatin on sacrifice day 9, the total decrease was 11.0200 + 12.5130 + 4.6790 = 28.2120 pg/mL (P < 0.001). Regarding Eucerin cream on sacrifice day 9, the decrease was 4.5870 + 12.5130 + 3.4050 = 20.5050 pg/mL (P = 0.0032). Overall, these results demonstrate that both treatment and time significantly affect TNF-α levels, with rosuvastatin producing the greatest reduction, especially as the day of sacrifice progresses. Changes in TNF-α levels by the rat group and the day of sacrifice are depicted graphically in Figure 1B, with corresponding descriptive statistics presented in Table 2.

Table 2 Descriptive statistics analysis of tumor necrosis factor-alpha changes by rat group and day of sacrifice.

Group	Day	mean	Range	Median	IQR	SD
Group placebo	3	30.70	4.69	30.64	1.65	6.76
Group control	3	35.29	6.80	34.39	2.26	5.65
Group study	3	24.27	5.50	23.91	3.10	7.34
Group placebo	6	22.62	5.58	22.89	1.83	6.76
Group control	6	27.46	6.07	27.55	2.52	5.65
Group study	6	14.04	4.93	13.72	1.74	7.34
Group placebo	9	14.78	3.91	14.24	1.63	6.76
Group control	9	22.78	6.21	23.82	3.49	5.65
Group study	9	7.08	4.06	6.94	1.16	7.34

IQR: Interquartile range.

IL-1β

The linear model for IL-1β does not include interaction terms because these terms were not statistically significant, and the model demonstrated a better fit without them, as indicated by lower AIC and BIC values. The model equation is as follows: IL-1β = 17.955 - 6.573x₁ - 2.623x₂ - 3.986x₃ - 7.55x₄. The R² of the model is 0.9452, indicating that 94.52% of the variation in IL-1β is explained by the model. The adjusted R² is 0.9426, and the overall fit is statistically significant [F (4, 85) = 366.7, P < 0.001].

In detail: The expected value of IL-1β for the reference group (normal saline, day 3) was 17.9552 pg/mL (P < 0.001). On the same day, the administration of rosuvastatin reduced the IL-1β concentration by 6.1900 pg/mL (P < 0.001). Correspondingly, the administration of Eucerin cream reduced the IL-1β concentration by 2.6227 pg/mL (P < 0.001). When normal saline treatment was continued, sacrifice on day 6 resulted in a reduction of 3.7200 pg/mL (P < 0.001), whereas sacrifice on day 9 was associated with a reduction of 6.9300 pg/mL (P < 0.001). Overall, the results revealed that the administration of rosuvastatin and an advanced day of sacrifice were associated with a significant reduction in IL-1β levels, with no significant interaction between these factors. Changes in IL-1β levels by the rat group and the day of sacrifice are depicted graphically in Figure 1C.

IL-6

The linear model for IL-6 includes both main effects and interaction terms, as statistically significant interactions between the independent variables were identified. The model equation is as follows: IL-6 = 40.241 - 9.712x₁ - 4.677x₂ - 5.755x₃ - 14.371x₄ - 5.418x₁x₃ - 5.760x₁x₄ - 2.728x₂x₃ - 3.461x₂x₄. The R² of the model is 0.9633, indicating that 96.33% of the variation in IL-6 is explained by the model. The adjusted R² is 0.9597, and the overall fit of the model is statistically significant [F (8, 81) = 266.1, P < 0.001].

In detail: The expected value of IL-6 for the reference group (normal saline, day 3) is 40.2410 pg/mL (P < 0.001). On the same day, the administration of rosuvastatin reduced the IL-6 concentration by 9.7120 pg/mL (P < 0.001), whereas the administration of Eucerin cream reduced the IL-6 concentration by 4.6770 pg/mL (P < 0.001). In the normal saline group, the mean IL-6 concentration decreased by 5.7550 pg/mL on day 6 (P < 0.001) and by 14.3710 units on day 9 (P < 0.001).

When interactions are taken into account: For rosuvastatin on day 6, the total reduction was 9.7120 + 5.7550 + 5.4180 = 20.8850 pg/mL (P < 0.001). For rosuvastatin on day 9, the total reduction was 9.7120 + 14.3710 + 5.7600 = 10.400 pg/mL (P < 0.001). For the Eucerin cream on day 6, the total reduction was 4.6770 + 5.7550 + 2.7280 = 13.1600 pg/mL (P = 0.023). For the Eucerin cream on day 9, the total reduction was 4.6770 + 14.3710 + 3.4610 = 22.5090 pg/mL (P = 0.004). Overall, the results revealed that rosuvastatin administration was associated with a significant reduction in IL-6 levels, which became more pronounced over time, especially when rosuvastatin was combined with sacrifice at later time points. Changes in the level of IL-6 by the rat group and the day of sacrifice are depicted graphically in Figure 1D, with corresponding descriptive statistics presented in Table 3.

Table 3 Descriptive statistics analysis of interleukin-6 changes by rat group and day of sacrifice.

Group	Day	mean	Range	Median	IQR	SD
Group placebo	3	35.56	6.00	35.52	3.62	7.72
Group control	3	40.24	6.17	39.37	2.99	6.25
Group study	3	30.53	5.21	30.87	0.71	8.50
Group placebo	6	27.08	6.54	27.75	2.77	7.72
Group control	6	34.49	5.58	34.11	2.40	6.25
Group study	6	19.36	5.63	19.56	1.45	8.50
Group placebo	9	17.73	7.43	17.82	3.19	7.72
Group control	9	25.87	4.44	25.72	1.99	6.25
Group study	9	10.40	4.30	10.25	1.75	8.50

IQR: Interquartile range.

Burn size

The linear model for burn size includes both main effects and interaction terms, as these were found to be statistically significant. The model equation is as follows: Burn size = 235.171 - 30.526x₁ - 15.858x₂ - 31.476x₃ - 71.617x₄ - 20.995x₁x₃ - 31.699x₁x₄ - 8.548x₂x₃ - 8.538x₂x₄. The R² of the model is 0.9438, indicating that 94.38% of the variance in burn size is explained by the model. The adjusted R² is 0.9383, and the overall fit is statistically significant [F (8, 81) = 170.1, P < 0.001].

In detail: The expected burn size for the reference group (normal saline, day 3) is 235.171 (P < 0.001). On the same day, the use of rosuvastatin reduced the burn size by 30.526 units (P < 0.001), whereas the use of Eucerin cream reduced it by 15.858 units (P < 0.001). In the normal saline group, sacrifice on day 6 resulted in a reduction of 31.476 units (P < 0.001), whereas sacrifice on day 9 resulted in a reduction of 73.62 units (P < 0.001).

Combined with interaction terms: For rosuvastatin on day 6, the total reduction is 30.526 + 31.476 + 20.995 = 82.997 units (P = 0.0019). For rosuvastatin on day 9, the total reduction is 30.526 + 71.617 + 31.699 = 133.842 units (P < 0.001).

Overall, the results show that rosuvastatin administration and an advanced day of sacrifice are associated with a significant reduction in burn size, with a more pronounced effect noted when the two factors are combined. Changes in burn size by rat group and day of sacrifice are depicted graphically in Figure 1E, with corresponding descriptive statistics presented in Table 4. Representative macroscopic digital images of burn wound healing for the 9 study subgroups are displayed in Figure 2, whereas representative microscopic burn wound healing images are displayed in Figure 3.

Figure 2 Representative macroscopic burn wound healing digital images. The rosuvastatin group showed a gradual increase in granulomatous tissue growth, epithelialization, and wound shrinkage on later sacrifice days compared to the control and placebo groups, indicating enhanced burn wound healing.

Figure 3 Representative microscopic burn wound healing images. A: Coagulative necrosis involving the epidermis and superficial dermis, characterized by loss of tissue architecture and magenta staining (arrowheads). Control group, day 9; haematoxylin-eosin (HE) staining, magnification × 40; B: Focal coagulative necrosis observed in skin tissue (arrowheads). Rosuvastatin group, day 9; HE staining, magnification × 40; C: Scant neoplastic capillaries (arrow) corresponding to a low vascular score) observed in a healing area, control group, day 9; HE staining, magnification × 200; D: Highly vascularized connective tissue (arrows) corresponding to a high vascular score, rosuvastatin group, day 9; HE staining, magnification × 20; E: Granulation tissue showing mild to moderate inflammatory cell infiltration (score 1), rosuvastatin group, day 9; HE staining, magnification × 200; F: Granulation tissue showing severe inflammatory cell infiltration (score 2), placebo group, day 9; HE staining, magnification × 200; G: Mild fibroblastic response with moderate fibroblast proliferation and minimal collagen deposition, control group, day 9; HE staining, magnification × 200; H: Marked fibroblastic reaction with extensive fibroblast proliferation and increased collagen deposition, rosuvastatin group, day 9; HE staining, magnification × 100; I: Μicrohemorrhage (small, well-defined area of extravasated red blood cells), placebo group, day 9; HE staining, magnification × 100; J: Multifocal scattered microhemorrhages (small foci of red blood cell extravasation), rosuvastatin group, day 9; HE staining, magnification × 100.

Table 4 Descriptive statistics analysis of burn wound size changes by rat group and day of sacrifice.

Group	Day	mean	Range	Median	IQR	SD
Group placebo	3	219.31	18.77	217.29	6.66	34.53
Group control	3	235.17	25.88	236.17	5.08	31.40
Group study	3	204.65	57.20	207.37	13.06	44.26
Group placebo	6	179.29	34.08	179.73	20.22	34.53
Group control	6	203.70	35.32	203.36	14.04	31.40
Group study	6	152.17	19.42	155.25	8.73	44.26
Group placebo	9	139.16	29.02	139.29	13.26	34.53
Group control	9	163.55	38.28	164.22	9.25	31.40
Group study	9	101.33	34.99	99.85	8.05	44.26

IQR: Interquartile range.

Histopathological examination of inflammatory infiltration

For the inflammatory infiltration variable, which was recorded on an ordinal categorical scale (0 = absent, 1 = moderate, 2 = severe), ordinal logistic regression analysis was performed. The independent variables included group (control as the reference, placebo, study), day of sacrifice (day 3 as the reference), and the interaction term group × day. The regression analysis results of the inflammatory infiltration model coefficients are presented in Table 5.

Table 5 Coefficients of the inflammatory Infiltration model: Regression analysis results.

Predictor	Coefficient (β)	SE	t value	P value
Group study	-20.9163	0.3231	-64.73	< 0.001
Group placebo	-0.4418	0.9449	-0.47	0.640
Day 6	1.7917	1.0206	1.76	0.079
Day 9	1.2528	0.9449	1.33	0.185
Group study × day 6	-3.5835	1.6030	-2.24	0.025
Group study × day 9	-21.1606	Approximately 0.0000	-3.94e + 08	< 0.001
Group placebo × day 6	-0.9444	1.3849	-0.68	0.495
Group placebo × day 9	0.9808	1.4121	0.69	0.487
Cut-point 0|1	-20.5108	0.3224	-63.62	< 0.001
Cut-point 1|2	0.4055	0.6455	0.63	0.530

Model equations: Transition from 0 to 1: Log [P (Y ≤ 0)/P (Y > 0)] = -20.5108 + 20.9163 × group study + 0.4418 × group placebo - 1.7917 × day 6 - 1.2528 × day 9 + 3.5835 × (group study × day 6) + 21.1606 × (group study × day 9) + 0.9444 × (group placebo × day 6) - 0.9808 × (group placebo × day 9). Transition from 1 to 2: Log [P (Y ≤ 1)/P (Y > 1)] = 0.4055 + 20.9163 × group study + 0.4418 × group placebo - 1.7917 × day 6 - 1.2528 × day 9 + 3.5835 × (group study × day 6) + 21.1606 × (group study × day 9) + 0.9444 × (group placebo × day 6) - 0.9808 × (group placebo × day 9).

Commentary on the results: Compared with the control group, the study group demonstrated a highly significant reduction in the likelihood of developing a higher grade of inflammatory infiltrate (β = -20.92; P < 0.001), indicating a strong protective effect of the therapeutic intervention. In contrast, the placebo group did not significantly differ from the control group (β = -0.44, P = 0.64). With respect to the observation period, days 6 and 9 had positive coefficients (β = 1.79 and β = 1.25, respectively), indicating a tendency toward an increase in inflammation; however, the result was not statistically significant (P = 0.079 and P = 0.185, respectively).

The interaction terms highlighted the dynamics of the treatment. Specifically, the interaction between study group and day 6 showed a significant reduction (β = -3.58, P = 0.025), whereas the interaction between study group and day 9 demonstrated a highly significant and strong negative effect (β = -21.16, P < 0.001), indicating almost complete suppression of the inflammatory response.

In contrast, the interactions between the placebo group and days 6 and 9 were not statistically significant (P > 0.48), reinforcing the absence of a therapeutic effect. Finally, analysis of the intercepts revealed that the model clearly differentiated between absent and moderate inflammation (0|1, P < 0.001); however, the difference between moderate and severe inflammation (1|2) was not statistically significant (P = 0.53). This finding suggests that while the model effectively captures early differences in inflammatory infiltration, it has limitations in distinguishing between higher levels of intensity.

Histopathological examination of neovascularization

Neovascularization was assessed as an ordered categorical variable (0 = absent, 1 = moderate, 2 = intense). To investigate the effect of the therapeutic intervention, multinomial logistic regression was applied with the independent variables group (control as reference, placebo, study), observation day (day 3 as reference) and the interaction term group × day. The regression analysis results of the neovascularization model coefficients are presented in Table 6.

Table 6 Coefficients of the neovascularization model: Regression analysis results.

Predictor	Coefficient (β)	SE	t value	P value
Group study	2.3072	0.9218	2.50	0.012
Group placebo	0.3138	0.8292	0.38	0.705
Day 6	-0.7534	0.9046	-0.83	0.405
Day 9	-1.2683	0.9872	-1.28	0.199
Group study × day 6	3.0884	1.5367	2.01	0.044
Group study × day 9	2.9249	1.4193	2.06	0.039
Group placebo × day 6	1.7872	1.2369	1.44	0.148
Group placebo × day 9	1.3427	1.3110	1.02	0.306
Cut-point 0|1	0.1451	0.5973	0.24	0.808
Cut-point 1|2	2.5504	0.7012	3.64	< 0.001

Commentary on results: Compared with the control group, the study group had a significantly increased probability of more intense neovascularization (β = 2.31, P = 0.012), demonstrating the strong angiogenic effect of the therapeutic intervention. The interaction terms study group × day 6 (β = 3.09, P = 0.044) and study group × day 9 (β = 2.92, P = 0.039) were statistically significant, confirming that the positive effect of the treatment progressively increased with time and was maintained until day 9. In contrast, the placebo group did not show statistically significant differences in either the main effects or the interactions with days (P > 0.14), which reinforces the indication of the absence of a biological effect. The individual effects of days (day 6, day 9) were not statistically significant (P > 0.19), indicating that the enhancement of neovascularization is attributed mainly to the therapeutic intervention and not to the simple effect of time.

Finally, analysis of the thresholds revealed that the model does not adequately distinguish between absent and moderate neovascularization (0|1, P = 0.808) but clearly distinguishes moderate from intense (1|2, P < 0.001). This finding confirms its ability to capture the highest levels of angiogenesis. Overall, the administration of the substance to the study group was associated with significantly increased neovascularization, especially on days 6 and 9, compared with that in the control and placebo groups. The results clearly show time dependence, as the effect is enhanced over time and remains strong. The modeling proved effective in distinguishing the levels of neovascularization, especially between moderate and intense, enhancing the reliability of the findings and confirming the strong angiogenic effect of the therapeutic intervention.

Histopathological examination of fibroblastic response

Fibroblastic response was recorded as an ordinal categorical variable (0 = absent, 1 = moderate, 2 = intense) and analyzed using multinomial logistic regression. The effects of group (control as reference, placebo, study), observation day (day 3 as reference) and the interaction term group × day were examined. The regression analysis results of the neovascularization model coefficients are presented in Table 7.

Table 7 Coefficients of the fibroblastic response model: Regression analysis results.

Predictor	Coefficient (β)	SE	t value	P value
Group study	2.0683	0.9107	2.27	0.023
Group placebo	0.2975	0.8148	0.37	0.715
Day 6	-0.6730	0.8562	-0.79	0.432
Day 9	-1.0816	0.8970	-1.21	0.228
Group study × day 6	3.0282	1.5111	2.00	0.045
Group study × day 9	3.5531	1.5307	2.32	0.020
Group placebo × day 6	1.7481	1.2163	1.44	0.151
Group placebo × day 9	1.2231	1.2378	0.99	0.323
Cut-point 0|1	-0.1883	0.5850	-0.32	0.748
Cut-point 1|2	2.3328	0.6766	3.45	< 0.001

Commentary on results: Compared with the control group, the study group presented a significantly increased probability of intense fibroblastic response (β = 2.07, P = 0.023). These findings demonstrate that therapeutic intervention strongly stimulates fibroblast activation and collagen production. In contrast, compared with the control group, the placebo group did not significantly differ (P = 0.715), which reinforces the indication of the absence of biological activity.

The individual effects of time (days 6 and 9) were negative but not statistically significant (P > 0.22), suggesting that they do not lead to a reduction in response by themselves unless combined with therapeutic intervention. The interaction terms study group × day 6 (β = 3.03, P = 0.045) and study group × day 9 (β = 3.55, P = 0.020) were statistically significant. This finding indicates that not only did the treatment increase fibroblast activity but also that this phenomenon became more pronounced and more consistent over the following days, confirming the time-dependent enhancement of the healing process. The placebo × day interactions were not significant (P > 0.15), excluding any therapeutic contribution of the placebo.

With respect to the thresholds, the model was unable to satisfactorily distinguish between no response and a moderate response (0|1, P = 0.748), but it accurately distinguished moderate from intense response (1|2, P < 0.001), which reinforces the validity of the conclusions about the higher levels of fibroblast activation.

Overall, therapeutic intervention applied to the study group led to a clear and statistically significant increase in the fibroblast response, especially on days 6 and 9, which is consistent with the activation of tissue repair and remodeling mechanisms. In contrast, the placebo group did not significantly differ. The model reliably distinguished between moderate and intense responses, confirming that the intervention substantially increased the level of fibroblast activation.

Histopathological examination of coagulative necrosis and microhemorrhage

In all samples examined in the control group, diffuse coagulative necrosis (corresponding to the area where the special metal sealer was applied) and microhemorrhage were observed. This involved all layers of the epidermis and the middle and deep dermis; however, in some cases, it extended to the muscle layer. In the study group, rosuvastatin induced a significant decrease in both lesions (P < 0.001). The healing process, including hyperemia, microhemorrhages, inflammatory infiltration and the formation of neogranulation tissue and neovascularization, involves the deep dermis, muscularis mucosa and subcutaneous tissue.

Adverse event outcomes

No adverse effects or deaths were observed in any experimental group during the study.

DISCUSSION

Rosuvastatin, a synthetic hydrophilic statin, is distinguished within its class based on its unique pharmacokinetic and pharmacodynamic properties. It has the longest half-life among statins because of its molecular structure, which allows it to bind to multiple sites on 3-hydroxy-3-methylglutaryl-CoA reductase inhibitor that produces cholesterol[7]. This enables effective enzyme inhibition at lower doses of the drug[52]. The primary role of rosuvastatin is associated with two genes: Solute carrier organic anion transporter family member 1B1 and ATP-binding cassette transporter G2[6]. These genes regulate proteins responsible for transporting the drug throughout the body. Data on ATP-binding cassette transporter G2 genotype polymorphisms could be valuable for tailoring rosuvastatin treatment to individual patients because these polymorphisms are associated with the drug’s pharmacokinetics[52,53]. Its liver selectivity leads to reduced bioavailability in other tissues, thereby decreasing the risk of muscle toxicity after administration[7]. Additionally, rosuvastatin is metabolized mainly by the cytochrome P450 2C9 enzyme, with minimal involvement of cytochrome P450 3A4[6], which accounts for its minimal interactions with drugs that inhibit cytochrome P450 enzymes.

The hydrophilic properties of rosuvastatin and its selective uptake by liver cells lead to minimal passive diffusion across cell membranes and almost complete absorption by the liver[7]. As a result, compared with simvastatin and pravastatin, rosuvastatin has superior lipid-lowering effects[6], as well as enhanced wound healing ability. Its hydrophilicity, strong liver selectivity, and long half-life contribute to improved wound repair[8,10,12,34,54]. Rosuvastatin specifically targets the endothelium and microvasculature, limiting passive diffusion into non-liver tissues and thereby reducing potential side effects[8,10,12,34]. This targeted action supports better microcirculation and stimulates angiogenesis, both of which are essential for wound healing. Additionally, rosuvastatin has multiple beneficial effects, including anti-inflammatory effects through the inhibition of TNF-α, IL-6, and IL-1β and immunomodulatory local effects[55]; cardioprotective effects against chemotherapy toxicity[56]; vascular endothelial growth factor and endothelial nitric oxide synthase upregulation[2]; antioxidant protection through nicotinamide adenine dinucleotide phosphate oxidase inhibition[2]; and antiapoptotic effects on endothelial cells and fibroblasts[8,10,37,39,54]. Together, these properties foster an optimal microenvironment for rapid tissue regeneration[38,40].

Rosuvastatin improves microvascular function in the skin of individuals with type 2 diabetes mellitus[57] and helps in the healing of diabetic wounds[58]. However, long-term use of statins, particularly atorvastatin followed by rosuvastatin, may increase the risk of developing diabetes mellitus[59].

Compared with oral intake, the topical application of rosuvastatin for wound healing is beneficial. It helps prevent gastrointestinal issues and reduces side effects. Moreover, topical delivery allows the medication to cover a wider area of the skin, promoting faster healing by improving epithelial growth and reducing the presence of inflammatory cells[60]. Additionally, rosuvastatin has antimicrobial effects[4].

Histopathological changes in burn tissue can be classified into three concentric zones: The coagulative necrosis zone (characterized by ischemic necrosis and hypoxia), the stagnation zone (representing borderline viability), and the hyperemia zone. Burns compromise skin integrity and impair both cellular and humoral immunity. Heat-induced vascular damage increases vascular permeability, resulting in the extravasation of plasma and electrolytes into the interstitial space. Dead tissues decrease the number of immune cells in the burned region, allowing pathogens to penetrate deeper into the underlying tissues and facilitating their spread[61].

Considering the aforementioned literature, we chose to evaluate rosuvastatin in partial-thickness burn wounds using a well-established experimental rat model. This model was selected because it accurately replicates the local and systemic pathophysiological processes characteristic of human burns, thereby enabling the extraction of clinically relevant findings. In an experimental study in rats, the optimal topical application dose of rosuvastatin was 20 mg/kg body weight to prevent intra-articular adhesions of the knee[37]. In our study, we used 200 μL of cream once daily, which equates to approximately 7-8 mg of rosuvastatin per rat weighing approximately 350-400 g.

The originality of this study lies in the fact that although previous research has focused on the effects of other statins (e.g., simvastatin, atorvastatin, and pravastatin), with encouraging results in healing various burn wound models, rosuvastatin - as a newer, more potent statin with a better safety profile - has not yet been investigated in this specific context. Therefore, to our knowledge, this is the first study to examine the use of rosuvastatin for burn wound healing in rats.

The study focused on three main areas: (1) The systemic inflammatory response was assessed by quantifying the levels of proinflammatory cytokines (IL-1β, IL-6, and TNF-α) in serum to investigate the extent of inflammatory activation and the potential role of rosuvastatin in modulating this response; (2) Histopathological lesions in the skin were evaluated by examining parameters such as inflammatory infiltration, neovascularization, fibroblastic reactions, coagulative necrosis, and hemorrhages to assess the effects of rosuvastatin at the morphological and cellular levels; and (3) Macroscopic evaluation (size and contraction of the burn wound) was performed.

This integrated approach enabled the simultaneous assessment of both morphological changes in the tissue and the underlying molecular mechanisms, thereby enhancing the validity and interpretative power of the findings. Consequently, this study aims to link histopathological data with biochemical and immunological parameters, providing a more comprehensive framework for understanding the effects of rosuvastatin on the healing of burn lesions.

The findings of the present study can be summarized in six key points: (1) The induction of a partial-thickness burn wound is accompanied by an intense inflammatory response, an increase in proinflammatory cytokines, and significant histological alterations in the skin; (2) Rosuvastatin significantly reduced inflammatory infiltration and cytokine levels, indicating an anti-inflammatory effect and a decrease in burn wound size. Using the third-day values from the control group as baseline, the study group demonstrated statistically significant decreases in serum CRP (5.75-10.30-16.15 vs 21.40-4.87-9.69, P < 0.01), TNF-α (11.02-21.25-28.21 vs 35.29-7.83-12.51, P < 0.01), and IL-1β (6.19-10.28-14.41 vs 17.83-3.72-6.93, P < 0.01), and IL-6 levels (9.71-20.88-29.84 vs 40.24-5.75-14.37, P < 0.01) as well as a significant reduction in burn wound size (30.52-83.00-135.84 vs 235.17-31.47-73.62, P < 0.01) across the later sacrifice days. These findings clearly demonstrate a distinct difference in the inflammatory response between the groups during the follow-up period. The study group showed significant and gradual decreases in the levels of the inflammatory markers CRP, TNF-α, IL-1β, and IL-6. This reduction in inflammation correlated with a gradual decrease in burn wound size in the study group, indicating a more effective therapeutic response to rosuvastatin; (3) Rosuvastatin administration was associated with enhanced neovascularization (P = 0.012) and a fibroblastic response (P = 0.023), processes critical for the organization of granulation tissue and the promotion of healing; (4) The presence of coagulative necrosis and hemorrhages was less pronounced in treated animals (P < 0.001), suggesting that rosuvastatin has a protective effect on the wound microenvironment; (5) The combination of the above findings demonstrates that rosuvastatin does not act unidimensionally; rather, it affects both the inflammatory and reparative phases of healing, achieving a better balance between the two; and (6) The administration of rosuvastatin was safe in this experimental model, with no adverse effects or mortality observed.

Overall, our results suggest that rosuvastatin represents a potentially promising therapeutic agent for the treatment of burn wounds, as it combines anti-inflammatory, angiogenic, and reparative effects. This study presents a series of significant findings regarding the effects of rosuvastatin on burn wound healing. These findings not only confirm previous observations about the pleiotropic actions of statins but also provide new data with potential clinical and pathophysiological significance.

CONCLUSION

The findings of our study support the conclusion that partial-thickness burn induction triggers an intense inflammatory response characterized by elevated levels of proinflammatory cytokines (IL-1β, IL-6, TNF-α) and significant histological alterations, which is consistent with the established pathophysiology of burn wounds. The administration of rosuvastatin resulted in a significant reduction in the levels of these cytokines, demonstrating the anti-inflammatory effects of the substance, which is consistent with the existing international literature. Histopathologically, therapeutic intervention was associated with decreased inflammatory infiltration, reduced coagulative necrosis, and fewer hemorrhages, highlighting a protective effect on the wound microenvironment. Rosuvastatin also significantly enhanced neovascularization and fibroblastic activity - processes critical for tissue remodeling and collagen synthesis - confirming the angiogenic and regenerative effects reported in recent studies. Macroscopically, burn wound contraction was faster and more pronounced in the rosuvastatin-treated group, reflecting the improved molecular and histopathological profiles and functional benefits. Rosuvastatin administration was proven to be safe in this experimental model, with no adverse effects. Additional extensive experimental and clinical research is necessary to confirm the safety and effectiveness of using rosuvastatin topically for burn wound treatment prior to its use in clinical settings.

ACKNOWLEDGEMENTS

The authors extend their appreciation to Ioannis Taitzoglou, Professor, School of Veterinary Medicine, Faculty of Health Sciences, Aristotle University of Thessaloniki, for his valuable assistance and guidance in the experimental performance; and Vasilis Chasiotis, Assistant Professor of Official Statistics and Statistical Methodology, Department of Statistics, Athens University of Economics and Business, for performing the statistical analysis and review.