Introduction
Diabetes mellitus (DM) is a chronic disease characterized by the inability of the body to regulate blood sugar [1]. In 2019, it was estimated that 463 million people or 9.3% of the global population suffer from DM [2]. The incidence of DM is increasing in the younger population, causing heavier economic and social burdens [3]. The high morbidity and mortality of DM are caused by its complications. More than 30% of DM patients presented to the Danish health service with evidence of peripheral neuropathy or vascular disease, and approximately 40% had non-trauma amputation for diabetic ulcers [4]. The incidence of diabetic foot in a tertiary hospital in Bandung was 44% of all DM patients [5].
Impaired wound healing, including diabetic ulcers and burns, poses a significant global healthcare threat. Accelerating healing requires biological factors that promote angiogenesis, promote cellular proliferation, and reduce inflammation [6-9]. Therapy for diabetic wounds includes surgery, education, nutrition, physical exercise, and pharmacology. The American Diabetes Association advises nutritional therapy to be given to all diabetic patients [10-13]. Herman et al. [14] applied low-carbohydrate dietary intervention on DM type 2-induced rats for a month and demonstrated improved blood sugar levels, body weight, and insulin levels, which reduces the risks of vasculopathy, neuropathy and immunopathy in DM.
A ketogenic diet is a high-fat and low-carbohydrate diet, restricting carbohydrate intake up to 5%–10% of overall diet necessity and replacing it with sufficient fat and protein (1 g/kg) [15,16]. A study by Kesl et al. [17] showed that a ketogenic diet could significantly increase blood flow in young and old rats. Another study stated that ketone supplements increased wound closing 10% to 14% faster on day three and 36% more quickly on day 10. To this day, the role of nutrition in wound healing is still controversial. In this experimental study, we attempt to compare the wound closure rate between diabetic rat models on a conventional diabetic diet versus those on a ketogenic diet.
Methods
Experimental animals
Inclusion criteria were male white rats (Rattus norvegicus), aged 16–20 weeks old, acclimated for 7 days before the experiment was conducted, with body weights of 200–300 g. Exclusion criteria were ill and inactive rats during the adaptation period. Case and control groups were collected with a completely randomized design method. The minimum sample size was nine per group. One extra subject was added to anticipate dropout, resulting in a total of 30 models. We had implemented the 3R (replacement, reduction, refinement) and 5F (free of hunger and thirst, pain, trauma and disease, long-term fear and stress also expressing natural behavior) principles. All experimental protocols employed in this study were approved by the Medical Research Ethics Committee of Padjadjaran University Bandung, Indonesia (211/UN6.KEP/EC/2002).
Induction of diabetes
Diabetes was induced by a high-fat diet for 1 month (53% fat, 25% carbohydrate, and 22% protein) with intraperitoneal injection of streptozotocin (40 mg/kg). Materials used for the high-fat diet were standard pellet, egg yolk, wheat flour, coconut oil, goat fat, quail egg juice, and water [14]. During the streptozotocin injection process, the rat was taken with a clean towel and then carefully held on its back. The abdominal region was disinfected with an alcohol swab, and streptozocin was injected. Blood sugar was measured on days 3, 4 and 5 after injection. Blood was drawn from the tail region after disinfection, and then the blood was put into glucometer sticks. The injection wound was pressed with an alcohol swab until the bleeding stopped. If fasting blood glucose was >140 mg/dL in three measurements, the rat would be established as a DM model [16].
Induction of wound
The wound was made with a solder (1 cm diameter). We used the Mason-Walker model as a guideline. Intramuscular ketamine 50 mg/kg was used as anesthesia, after which thin metal cylinders were heated for 60 seconds above an open fire, then pressed gently on the shaved and cleaned dorsal area (i.e., the back) for 10 seconds to produce a second-degree, 1 cm-diameter burn injury [16]. The rats were then placed in different cages (25×20×16 cm) to prevent infections and widening of wounds.
Experimental design
The animals were divided into three groups. Group I was given a ketogenic diet (60% fat, 24% carbohydrate, and 16% protein) composed of 12.16 g standard feed, 8.33 g wheat flour, one egg yolk, 5.42 g coconut oil, 7 g goat fat, and 250 mL water. Group II was given a conventional diabetic diet (43% carbohydrate, 15% fat, and 42% protein) composed of 25 g wheat flour and 25 g standard pellet for each rat. Group III was the control group with a normal diet consisting of 25 g of standard pellets (13% water, 18.5%–20.5% protein, fat ≥4%, fiber ≤6%, ash ≤8%, 0.9% calcium, and 0.7% phosphor) per day [14].
Wound healing was assessed by the size of wound closure (cm2) on each group. Stress level, activity level and fasting blood sugar level during the experimental process were monitored to avoid bias. To prevent infection, the wound was treated every day and the rats were given 4 mg/kg of intramuscular gentamycin. Five mg/kg of intramuscular ketorolac was given to reduce pain. The groups were then given ketogenic, conventional, or normal diets. The wound closure rates were measured on days 1, 7, 14, and 21 to represent each wound healing phase. The measurement of the wound was calculated by tracing the wound on tracing paper and measuring the longest and shortest length with a ruler. It was then calculated using an ellipse equation [16]:
The white rats were kept in cages and placed in a quiet room with adequate lighting. Room temperature was maintained in the range of 20–25°C. The animals were given food three times a day and drank ad libitum. Cages were kept well with daily cleaning. The rats were then placed in different cages (25×20×16 cm). Sunlight was given for 12 hours per day with a total cycle of light:dark at 14:10. After the trial was done, the animals were euthanized and the remains were burned. Euthanasia was done by injecting 100 mg/kg subcutaneous phenobarbital.
Statistical analysis
Editing, coding and entry of data were done with Microsoft Excel 2010 and Statistical Package for the Social Sciences (SPSS) 26 for Windows. The comparisons of wound area between groups were performed using one-way analysis of variance (ANOVA) due to the normal distribution of data. Multiple comparisons were done with post-hoc Bonferroni analysis.
Results
From the induction of DM, we obtained 30 diabetic rat models. Wound closure was measured on days 7, 14, and 21 (Fig. 1). The mean wound closure in all groups increased in every measurement.
Diabetic wound closure on day 7
The mean wound area on day 7 in the ketogenic diet group was 0.664 cm2, the diabetic conventional diet group 0.717 cm2, and the normal diet group 0.747 cm2 (Fig. 1). The ketogenic diet had a more extensive wound closure area than the conventional diet (Fig. 2). Statistic analysis using one-way ANOVA comparing the wound area in ketogenic diet, diabetic conventional diet, and normal diet resulted in P=0.051 (Table 1). One-way ANOVA comparing wound area differences in ketogenic diet, conventional diet, and normal diet resulted in in P=0.035 (Table 1).
Table 2 and Fig. 3 show post-hoc Bonferroni analysis test comparing the effect of the ketogenic diet, diabetic conventional diet, and normal diet on wound area in diabetic rat models. There was no significant statistical difference in the models on the ketogenic diet compared with those on conventional diet, and conventional diet models compared with normal diet models. Analysis of ketogenic diet compared to normal diet results in P<0.05, indicating a statistically significant difference between the two groups.
Table 2 and Fig. 4 show post-hoc Bonferroni analysis test comparing the effect of the ketogenic diet, diabetic conventional diet, and normal diet on wound area difference in diabetic rat models. There was no significant statistical difference in rats on the ketogenic diet compared with those on the conventional diet, or rats on the conventional diet compared with those on a normal diet. Analysis of the ketogenic diet compared to normal diet results in P<0.05, indicating a statistically significant difference between the two groups.
Diabetic wound closure on day 14
Table 3 and Fig. 1 show that the mean wound closure on day 14 in the ketogenic group was 0.564 cm2, while for the diabetic conventional diet the closure was 0.817 cm2 and for the normal diet 0.647 cm2. The mean difference in wound closure in ketogenic groups was found to be more extensive than in the conventional and normal diet groups. Statistical analysis using one-way ANOVA comparing the wound area in the ketogenic diet, conventional diet, and normal diet showed a significant difference (P<0.05). One-way ANOVA comparing the wound area difference in the ketogenic diet, conventional diet, and normal diet showed a significant difference of P<0.05.
The post-hoc Bonferroni test (Table 4, Fig. 3) resulted in a significant difference in the ketogenic diet compared to the conventional diet (P<0.001). The same result was found in the comparison between the ketogenic diet and the normal diet (P=0.051), more so than between the conventional diet and normal diets (P<0.001).
Table 4 and Fig. 4 show the post-hoc Bonferroni analysis test comparing the ketogenic diet, conventional diet, and normal diet on wound area differences in the diabetic rat models. There was no significant statistical difference in the ketogenic diet compared with the conventional diet and conventional diet compared with the normal diet. Analysis of ketogenic diet compared to normal diet resulted in P<0.05, indicating statistically significant difference between the two groups.
Diabetic wound closure on day 21
Table 5 and Fig. 1 demonstrate that the mean wound closure on day 21 in the ketogenic diet group was 0.164 cm2, while for the diabetic diabetic conventional diet it was 0.517 cm2, and 0.447 cm2 for the normal diet. Based on the data, the difference in wound closure was greater in the ketogenic diet group than in the conventional and normal diet group. Statistic analysis using one-way ANOVA resulted in a significant difference of wound area between groups (P<0.05). One-way ANOVA comparing the wound area difference in ketogenic diet, conventional diet, and normal diet showed a significant difference of P<0.05.
Post-hoc Bonferroni test on day 21 (Table 6, Fig. 4) showed a significant difference between the ketogenic diet and conventional diet on wound area (P<0.001). There was also a significant difference in the ketogenic diet compared to the normal diet group in the wound area (P<0.001). There was no significant difference found in the conventional diet group compared with the normal diet group P>0.05. Fig. 5 displays the serial progression of wound size comparison in the three groups.
Discussion
The pathophysiology of wound healing impairment in DM involves vascular, neural, immune, and biomechanical components. The wound healing process in diabetes is affected by chronic forms of inflammation due to disturbed angiogenesis, decreased endothelial progenitors, and imbalance of extracellular matrix (ECM) regulation. Prolonged hyperglycemia state correlates with stiff blood vessels resulting in vasculopathy and microangiopathy, where circulation slows down and microvascular dysfunction occurs. The pathogenesis of DM and atherosclerosis is closely intertwined. Systemic inflammation in type 2 DM results in hemodynamic disruption with vascular permeability disruptions forming a thrombus, increasing the risk of vessel obstruction. A ketogenic diet might benefit healing by providing an alternative source of energy to glucose and improving blood flow [18]. DM is a significant risk factor for cardiovascular disease, particularly ischemic heart disease (IHD). The pathophysiology of myocardial ischemia in diabetic patients is complex, with genetic variants for adenosine triphosphate (ATP)-dependent potassium channels potentially playing a role in IHD determinism [19]. In ischemic conditions, perfusion will decrease, potentially altering oxygenation, and ATP will decrease as much as 90%, which will have a negative effect on wound healing [20,21]. Every phase of the wound healing process demands cellular energy in large quantities [20].
In DM, there is an increase in chemotactic chemokines, causing infiltration of neutrophils and macrophages. Increased interleukin-1β and tumor necrosis factor-α will maintain continuous inflammation. Meanwhile, growth factors related to impaired healing in DM are insulin-like growth factor-1 and transforming growth factor-β [22]. There is an imbalance in the promotion of new vessel formation and maturation, and capillary density is not sufficient for normal wound healing, due to angiogenesis dysfunction of endothelial cells exposed to high levels of sugar. Hypoxia induced factor-1α with vascular endothelial growth factor (VEGF) as a gene target is also disturbed in hyperglycemia. Adipose derived cells from diabetic patients show reduced VEGF secretion and impaired angiogenic capacity, potentially resulting in impaired healing due to dysregulation of monocytes/macrophages. The last phase of healing is impaired in DM due to the disruption of angiopoietin (ANG)-1, ANG-2, and platelet derived growth factors. Matrix metalloproteinase-1 levels are also increased while tissue inhibitor of metalloprotease-1 decreases in diabetic wounds, disrupting the modulation of ECM [23].
In this study, it was found that a ketogenic diet improves diabetic wound healing compared to normal and conventional diabetic diets. Since day 7, significant differences were found between wound areas in the ketogenic group compared to those of the conventional and normal diets. On the 14th day, there was a significant difference in wound areas between those of the ketogenic diet and those of the conventional diet and normal diet. Day 14 was when the proliferation phase occurred, in which granulation and maturation could be delayed due to reduced oxygen and energy. The ketogenic diet may provide alternatives in oxygen and energy. On day 21, the difference in wound area was significant in the ketogenic diet group compared to those of the conventional and normal diets.
A study by Moriconi et al. [24] showed that ketogenic diet in DM patients was safe and effective as a means of lifestyle management. The body weight, blood pressure, and hemoglobin A1c levels were improved with a ketogenic diet. Satiety promoted by this diet also provides better patient compliance in maintaining the diet. Elswaidy et al. [25] conducted a study on ketogenic diet in 24 rats with delayed wound healing. It was found that the ketogenic diet improved wound healing by suppressing oxidative stress, modulating inflammation and collagen deposition, promoting proliferation, and increasing angiogenesis. In DM, high glucose levels could result in reactive oxygen species (ROS) formation in hyperglycemia sensitive cells such as endothelial cells. Excess ROS are associated with continuous inflammation, limiting efficiency in wound repair. Ketone bodies such as β-hydroxybutyrate strengthen antioxidants’ capacity and lower the damage of free radicals. Ketone also inhibits the production of mitochondrial ROS through glutamate excitotoxicity and increasing nicotinamide adenine dinucleotide oxidation [21]. A ketogenic diet increases mitochondrial protein release and activity and increases ATP production [23]. This diet will lower inflammation and increase blood perfusion. Ketone could also elevate vasodilator adenosine levels which will help repair vascularization [20,23]. A ketogenic diet also prevents pathogenic microorganism growth and inhibits inflammation and immune reaction [25].
A diabetic conventional diet was created as an early intervention for the obesity and diabetes cycle. This diet is helpful for insulin therapy due to the low glycemic index. The level of carbohydrates in this diet is constructed to avoid ketosis. The conventional diabetic diet also has a higher protein and fat composition than the normal diet. A high protein diet could help the immune system and tissue growth in tissue repair. One of the vital proteins in wound healing is arginine, which acts as an antioxidant and has a critical role in inflammation, proliferation, epithelization, and remodeling in wound healing [25]. From day 7 to day 21, there was no significant difference in the wound area between the conventional diabetic and normal diets in this study. Protein in the conventional diabetic diet is lower than in the ketogenic diet, which can explain the insignificant difference between groups.
This study proved that the ketogenic diet had a more significant relation with wound healing than the conventional diabetic and normal diets. However, this study has several limitations. The wounds produced in these diabetic rat models need to be improved. The wound is expected to describe the pathogenesis for impaired wound healing in diabetic models, implying vasculopathy, neuropathy, and immunopathy, contributing to slower wound healing. As animal models differ in metabolism, physiology, and genetics from humans, human studies are necessary to confirm the effect of the ketogenic diet on wound closure. Histopathological studies could also be conducted to ensure the roles of ketogenic diet in each wound-healing phase.
In conclusion, a ketogenic diet significantly improved wound healing compared to conventional diabetic diet in diabetic rat model. The difference of wound closure in ketogenic diet and diabetic conventional diet was found to be significant since day 7. Further studies on the different animal subjects or human subjects are needed for better results.
