Adnexal torsion is an important cause of gynecological emergency surgery with a prevalence of 2.7%, and it is more common after menarche, although it can be seen at any age [1, 2, 3]. Today, preserving the adnexa is essential, but salpingo-oophorectomy or oophorectomy was previously performed in cases of ovarian torsion [3]. Therefore, for the preservation of ovarian function, early diagnosis and treatment of torsion is very important. However, the goal of ischemia therapy is not only to restore blood flow but also to improve tissue perfusion. When detorsion is applied to the adnexa, some local and systemic effects occur due to the reperfusion of the ovaries [4].

Paradoxically, the reperfusion of ischemic tissue causes much more serious damage to the tissue than damage caused by ischemia alone [5, 6]. The restoration of circulation and reperfusion following ischemia presents a new physiopathological process called “reperfusion injury”, which itself results in various degrees of tissue damage. Total tissue damage is accepted as damage caused by both ischemia and reperfusion [7, 8, 9, 10]. Therefore, the prevention of reperfusion injury will further increase the success of ischemia treatment [7, 8].

Total antioxidant status (TAS) represents the whole effect of all antioxidants, while total oxidant status (TOS) expresses the total effect of all oxidants in body fluids and plasma [11]. The oxidative stress index (OSI) is considered a more proper index of oxidative stress and is calculated with the TOS/TAS formula. Malondialdehyde (MDA) and total thiol (t-SH) are other biomarkers that measure the level of oxidative stress [12]. Catalase is also a major indicator of the antioxidant defense system [13]. These substances can be a marker of reperfusion damage in various organs. Many antioxidant and anti-inflammatory treatments have been examined in terms of their efficacy in preventing ischemia/reperfusion (I/R) injury [14, 15, 16].

Granulocyte-colony stimulating factor (G-CSF) is a glycoprotein that mediates the production, differentiation and function of macrophages and neutrophils. It is used to provide peripheral stem cell mobilization in allogeneic or autologous stem cell transplantation and to increase the neutrophil count in neutropenic patients [17]. G-CSF also has protecting effects against I/R in certain tissues, such as cerebral injury [18], myocardial infarction [19, 20], kidney injury [21], and retinopathy [22].

Since rat experiments, at least to some extent, do indeed reflect human ovarian torsion and I/R injury, many studies have been conducted in this regard. In this study, we aimed to investigate the role of G-CSF as an antioxidant in the prevention of ovarian I/R injury and to examine its histo-pathological effects on rat ovarian tissue.

After obtaining approval from the Animal Experiments Local Ethics Committee of Health Sciences University Ankara Health Research Application Center (approval number: 552/2019), the study was conducted in accordance with the international guidelines for the ethical use of animals. The sample consisted of a total of 28 healthy adult non-pregnant female Wistar rats (weighs 220–260 g). The rats were housed in cages of 3–4 animals under optimum conditions (50–60% humidity, 20–24 °C and 12-hour light/dark cycle) and during the experimental period, rats were fed ad libitum cubes and tap water. No feed and water restrictions were applied, and all the rats were allowed to acclimate to their environment one week before the experiment. Four groups of seven rats each were formed: Group 1, sham; Group 2, torsion/detorsion; Group 3, torsion/detorsion + saline; and Group 4, torsion/detorsion + G-CSF.

All the rats were anesthetized by administering 40 mg/kg ketamine hydrochloride (Ketalar, Pfizer, Istanbul, Turkey) and 7 mg/kg xylazine (Rompun, Bayer, Istanbul, Turkey) under sterile conditions. The abdominal hair of the rats was shaved and then 10% povidone iodine solution was used for antiseptic. A longitudinal incision was made in the midline region of the lower abdomen by placing the rats in the supine position. Then, the uterine horns and adnexa were detected by making a peritoneal incision. In Group 1 (sham), after the uterus and adnexa were observed, the incision was repaired using 3/0 silk sutures. Three hours later, both ovaries were excised. In group 2, the ovarian pedicles on both sides were rotated 360${}^{\circ }$ clockwise and 3/0 silk sutures were used for fixation to the abdominal wall. After three hours of torsion, bilateral adnexal detorsion was performed. After three more hours of reperfusion period after suturing the surgical fields, both ovaries were removed. According to the protocol applied by Bostancı [16], intraperitoneal saline and G-CSF (100 IU/kg, Leucostim 30 MIU; Dem İlaç, Istanbul, Turkey) were administered in Groups 3 and 4 after 2.5 hours of torsion, respectively. After 30 minutes, bilateral adnexal detorsion was performed. Both ovaries were removed at the end of a total period of six hours, one for the analysis of biochemical markers and the other for histopathological evaluation.

Although TOS was lower in the G-CSF group (11.1 $\mu$mol H${}_2$O${}_2$ equivalent/L) compared to the torsion/detorsion group (13.76 $\mu$mol H${}_2$O${}_2$ equivalent/L), it was not statistically significant (p $>$ 0.05). OSI was also lower in the G-CSF group (37.0 $\pm$ 17.2 vs 28.8 $\pm$ 11.7), but this was also not significant (p $>$ 0.05). There was no statistically significant difference between the groups in terms of the TAS, catalase, MDA and t-SH values. Biochemical parameters are shown in Tables 1,2.

No primary follicle was observed in any of the rats in the torsion/detorsion group. In the G-CSF group, the number of primary follicles was similar to the sham group (3.14 vs 3.43) but significantly higher compared to the torsion/detorsion group (p = 0.001). The number of secondary follicles was also significantly higher in the G-CSF group than in the torsion/detorsion group (2.29 vs 0.14, p = 0.001). There was no statistically significant difference between the groups in terms of the numbers of tertiary and Graafian follicles (p $>$ 0.05) (Table 3) (Fig. 1).

Fig. 1.

Images of histopathological examination obtained by light microscope. (a) Primordial follicle in the ovary of a rat in the sham group. (b) A normal primary follicle in the ovary of a rat in the torsion/detorsion + G-CSF group. (c) A normal secondary follicle in the ovary of a rat in the torsion/detorsion + G-CSF group. (d) A degenerated primary follicle in the ovary of a rat in the torsion/detorsion group. (e) A degenerated secondary follicle in the ovary of a rat in the torsion/detorsion + saline group. (f) A degenerated Graafian follicle in the ovary of a rat in the torsion/detorsion + saline group.

Ovarian torsion treatment should be aimed at protecting the ovarian reserve, as well as relieving symptoms. The duration of ischemia in patients with ovarian torsion is an important factor to reduce possible ovarian damage. In this study, 3 hours of I/R was performed. Previous studies showed morphological and biochemical changes after I/R injury at different time intervals and revealed that a 3-hour torsion-3-hour reperfusion time is sufficient for I/R injury to occur [27, 28]. It is known that ovarian tissue can be seriously damaged by not only ischemia but also reperfusion. After reperfusion in tissue, free oxygen radicals can cause protein denaturation, lipid peroxidation and DNA damage, which are mainly causing tissue damage. Therefore, I/R in ovarian torsion is an important issue that has often been the subject of many studies, and various substances have been used to reduce reperfusion injury in rat ovaries [29, 30, 31, 32]. The presence of free radicals produced due to reperfusion suggests that the effects of most agents used in I/R models are due to their antioxidant properties.

G-CSF is thought to have a protective effect in reperfusion after ischemia due to its inhibitory effect on inflammation and oxidative stress. For this reason, it has been used for treatment and protection in I/R injury of many organs [18, 19, 20, 21, 22]. However, G-CSF using does not ameliorate in all types of injury and all tissues [33]. Although the antioxidant effect of G-CSF has also been demonstrated in I/R in rats both biochemically and histopathologically [16, 34], we found no significant difference between the groups in relation to biochemical markers evaluated in our study. However, the primary and secondary follicle counts were significantly higher in the histopathological examination in the group that received G-CSF compared to the rats that underwent torsion/detorsion alone.

In cases of ovarian torsion treated with detorsion, histopathological changes are first observed in the Graafian follicles, while primary follicles are affected last [35]. In the current study, the use of G-CSF before detorsion prevented the reduction of primary and secondary follicles numbers in the rat ovary, similar to previous studies indicating histopathological improvement in the rat ovary following detorsion with the use of G-CSF [16, 34]. It has been previously shown that G-CSF has an anti-apoptotic effect on vascular endothelial cells, and it was suggested to be the most effective mechanism to reduce reperfusion injury [22].

In the literature, the presence of G-CSF receptors has been demonstrated in ovarian and endometrial tissue [36, 37]. However, there is no study that detected receptors in tissue in an I/R model established with G-CSF, and we did not measure them in our study either. We speculate that differences in tissue response across studies may be related to the number of these receptors. The number of receptors can affect the antiapoptotic response in tissue, thus reducing tissue reperfusion injury and resulting in a lower level of biochemical markers.

Reactive oxygen species (ROS), such as hydrogen peroxide, hydroxyl radical and superoxide are essential components of oxidative damage in I/R. A large number of biochemical markers are used to identify the overproduction of ROS and reveal the degree of associated damage. Although we investigated several markers in this study, we found no significant difference between the groups with and without torsion/detorsion. The main mechanism in torsion is the complete or partial cessation of blood flow to the tissue, causing damage. The main determining factors in this damage are the degree and duration of torsion of the vascular pedicle. Blood flow has been observed in some cases of ovarian torsion even torsion was applied up to 360 degrees and it has also been shown that the damage increases as the duration of torsion increases [38]. Therefore, retrospectively reviewing the technique we applied in our model, we consider that 360-degree torsion may not be sufficient.

As a limitation of our study, we did not investigate different doses of G-CSF or examine the effect of G-CSF on healthy rat ovarian tissue. In addition, although we used many markers were used to determine antioxidant activity, unlike previous studies on the subject, the use of G-CSF in the I/R model did not result in a significant difference in relation to these markers. However, there was significant histopathological improvement in the group where G-CSF was administered.

In conclusion, although oxidative stress markers did not significantly differ between the groups, primary and secondary follicle numbers in rat ovaries were significantly higher in those using G-CSF.

SAyd—Protocol/project development, data analysis, manuscript writing/editing. MG—Protocol/project development, data analysis, manuscript writing/editing. MCA—Data collection/management. SAyh—Data collection/management. MES—Protocol/project development, data collection management. MK—Data collection/management. MÇ—Data collection management. MŞ—Data collection management. YÜ—Protocol/project development, data analysis. All authors read and approved the final manuscript.

After obtaining approval from the Animal Experiments Local Ethics Committee of Health Sciences University Ankara Health Research Application Center (approval number: 552/2019), the study was conducted in accordance with the international guidelines for the ethical use of animals.

We would like to express our gratitude to all those who helped us during the writing of this manuscript.

This research received no external funding.

The authors declare no conflict of interest.