A total of 100 healthy male rats with a weight of 280±20 g were purchased from Experimental Animal Center of Hebei Medical University (Hebei, China), and the animal certificate number was 1607209. The use of animals was in accordance with the specification for laboratory animal use of People's Hospital of Hebei Province. The study was approved by the Ethics Committee of the Hebei General Hospital (Hebei, China).

Feeds with different protein intake were prepared (Table I). The protein levels of the feed were referred to Medical Laboratory Animal Standards promulgated by the Ministry of Health of the People's Republic of China in 1992. The standards stipulated that the protein level of the feed was 18–25%.

The rats were housed at a constant room temperature (20–25°C) and humidity (40–60%) with standard feed (20% protein level) for one week. After being numbered according to body weight, the rats were randomly divided into 4 groups (groups A, B, C and D) using the random number table. The number of rats in each group was 25.

All the rats were subjected to the same surgical procedure. Before surgery, they were fasted and water deprived for 8 h. Each rat was subjected to depilatory treatment with a skin area of 4×4 cm after they were anesthetized by intraperitoneal injection of 10% chloral hydrate (0.3 ml/100 g body weight). The depilatory area was at the buttock on the left side of the back center line. After disinfection of the skin with 5% povidone-iodine, an incision with a length of 2 cm was performed. Fascia and tissue were separated by tweezers and a sterile steel disk (diameter 15 mm, thickness 0.3 mm, weight 0.6 g, autoclave sterilization) was implanted under the muscle. The incision was stitched with polyglycolic acid suture and 5% povidone-iodine was used to disinfect the suture to prevent infection. Three days after surgery, a magnetic disk (diameter 15 mm, thickness 2.5 mm, weight 2.6 g, with magnetic flux up to 1,500 gauss) was placed outside the skin in the area where the steel disk was placed (14). The pressure between the steel disk and magnet disk could cause pressure on the muscles and skin, which potentially led to local tissue ischemia. Two hours after pressurization, the magnets were removed for 0.5 h for ischemia reperfusion. This pressurization process was cycled 5 times a day and the stage II pressure ulcer was formed after 2–3 days of cyclic pressure operation. The criteria of successful stage II pressure ulcer models were: The color of the pressure ulcer was dark red, the skin around the pressure ulcer was redish, bleeding and exudate did not occur, obvious pain was showed by the rats.

Rats were fed with standard feed before and during the construction of stage II pressure ulcer models. The pressure ulcer area and body weight of the rats were measured and recorded immediately after the pressure ulcer was formed. After stage II pressure ulcer models were constructed, the rats in each group were fed with different protein levels of feed. The protein levels of the feed in groups A, B, C and D were 10, 15, 20 and 25%, respectively. The amount of feed was 20 g/rat/day and water was adequate.

Healing time, pressure ulcer area, and rat body weight were monitored at different times. Evaluation criteria for pressure ulcer healing were as follows: The redness of the skin around the pressure ulcer had subsided, the scab of the pressure ulcer was shed, the new skin color was consistented with the surrounding normal skin color.

Before surgery, after surgery, and 3, 7 and 14 days after rat models were constructed, 3 rats in each group were randomly selected to obtain abdominal aortic blood samples. Albumin (ALB) and hemoglobin (Hb) levels were measured by spectrophotometry after the blood samples were diluted with Drabkin's solution and Bromocresol green, respectively.

After the pressure ulcer models were established, pressure ulcer tissue samples were obtained. Pressure ulcer tissue samples of each group were obtained after 14 days feeding with different feeds. Conventional H&E staining was performed to observe the histological structure.

Data were analyzed by SPSS 19.0 software (SPSS, Inc., Chicago, IL, USA) and expressed as mean ± SD. Multiple comparisons were performed by analysis of variance (ANOVA) followed by post hoc test (LSD test). P<0.05 was considered to indicate a statistically sgnificant difference.

In each group, 22 rats were modeled successfully (Fig. 1) and the pressure ulcer healing time of each group was measured (Fig. 2). Rats in group A and B needed 21.63±1.12 and 19.27±1.22 day, respectively, for their ulcers to heal. However, the healing time of group C was only 16.36±1.57 days, which was significantly lower than that of group A and B (P<0.05). Furthermore, when compared with group C, a little longer healing time was found in group D (17.58±1.31 days), but without significant difference.

The pressure ulcer area was measured over time. No significant difference was found in the pressure ulcer area among groups after the models were constructed. Seven days after surgery the pressure ulcer area of each group was significantly reduced when compared with that of 1 day after surgery (P<0.05). The same situation applied to 14 days after surgery. Of note was that, at these times, the pressure ulcer area of group C (0.19±0.11 cm²) and group D (0.23±0.07 cm²) was significantly lower than that of group A (0.38±0.10 cm²) and group B (0.33±0.17 cm²) (P<0.05). Group D had a larger pressure ulcer area compared to group C, but without significant difference (Fig. 3).

The rat body weights of each group were recorded at different times (Fig. 4). Before surgery, there was no significant difference in body weight among the 4 groups, and it was significantly decreased in all groups after the models were constructed, but again without significant difference among groups. Three days after surgery, all the rat body weights were back to normal, and 7 days after surgery, a significantly higher body weight in groups C and D was found when compared with group A and B at the same time (P<0.05). Fourteen days after surgery, compared to after surgery, the body weight of group A and B was significantly increased, but it was significantly lower than that of groups C and D. Rats in group D had a higher body weight compared to group C, but with no significant difference.

Changes of ALB and Hb levels over time are shown in Fig. 5, respectively. No significant difference was found in ALB or Hb levels among groups before surgery. From 3 days after surgery, a significant decrease ALB and Hb levels was found in groups A and B when compared with that of before surgery. Moreover, these two levels of group C and D were significantly lower than that of group A and B. No significant difference was found in the ALB or Hb levels between group C and D, however, both of these levels of group C were closer to normal level and the ALB level of group D at 14 days after surgery was obviously a little higher than that before surgery.

After the pressure ulcer models were formed, the structural integrity of the pressure ulcer tissue was very poor and tissue fracture phenomenon was observed. In addition, broken cell structure occurred (Fig. 6A). Fourteen days after feeding with the different feeds, the damage degree of all rat pressure ulcer tissues was significantly improved. However, the structural integrity of the pressure ulcer tissue of rats in group A and B remained very poor, and the cell structure still displayed a certain degree of damage (Fig. 6B and C). At the same time, the pressure ulcer tissue of rats in group C and D had significantly better structural integrity, and the integrity of the cell structure had significantly improved, and the tissue fracture phenomenon had disappeared (Fig. 6D and E).