Bilirubin provides perforator flap protection from ischaemia‐reperfusion injury in a rat model: a preliminary result

Abstract

The use of bilirubin, a well‐known and powerful antioxidant, has gained popularity in recent years because of its role in the prevention of ischaemic heart disease in patients with Gilbert's syndrome. We investigate the effects of bilirubin on ischaemia‐reperfusion (I/R) injury using a rat perforator flap model.

Forty‐eight rats were randomly divided into two groups: experimental (bilirubin) group (n = 24) and control group (n = 24). In each group, elevated bilateral deep inferior epigastric perforator (DIEP) flaps were created. The right (no ischaemia side) and left (ischaemia side) DIEP flaps were separated according to the presence of ischaemia induction. Ischaemia was induced in anaesthetised rats by perforator clamping for 15 or 30 minutes. After surgery, the flap survival was assessed daily on postoperative days 0 to 5, and overall histological changes of DIEP flaps above the perforator were analysed at postoperative day 5.

The flap survival rate in the bilirubin group was significantly higher than that in the control group at the ischaemia side following perforator clamping for 15 or 30 minutes (93·42 ± 4·48% versus 89·63 ± 3·98%, P = 0·002; and 83·96 ± 4·23% versus 36·46 ± 6·38%, P < 0·001, respectively). The difference in flap survival between the two groups was the most prominent on the ischaemic side following 30 minutes of perforator clamping.

From a morphologic perspective, pre‐treatment with bilirubin was found to alleviate perforator flap necrosis caused by I/R injury in this experimental rat model.

Introduction

Perforator flaps have evolved from musculocutaneous/fasciocutaneous flaps without muscle or fascial carriers, and were first described by Dr Koshima in 1989. Since then, various perforator flaps have been designed and these have largely replaced the traditional flaps, owing to the versatile flap design, reduced donor site morbidity and short postoperative recovery period associated with perforator flaps. For these reasons, the introduction of perforator flaps has greatly expanded our options in the field of plastic surgery. Nevertheless, perforator flaps require meticulous dissection, and are associated with an unreliable perforator size and possible damage of the perforators. In a recent study regarding perforator flaps, a rat model was used and was demonstrated to be a highly reliable model to predict the effects of clinical application of various materials 1.

The use of bilirubin, a well‐known and powerful antioxidant, has gained popularity in recent years because of its role in the prevention of ischaemic heart disease in patients with Gilbert's syndrome. Moreover, mild elevated serum bilirubin has been demonstrated to be clinically helpful for patients with atherosclerosis in terms of preventing myocardial ischaemia in a cohort study 2, 3, 4, 5; and, in addition, a recent study reported that bilirubin supplementation appeared to result in a significant decrease in the myocardial infarct size during the period of ischaemia 6.

While studies using various materials to counteract ischaemia‐reperfusion (I/R) injuries are becoming increasingly common, the mechanisms underlying the same are not completely understood. Therefore, we here performed an I/R experimental study to evaluate the effects of bilirubin on the survival of deep inferior epigastric perforator (DIEP) flaps using a rat model.

Materials and methods

Experiment protocol

All animal protocols used in this study were approved by the Institutional Animal Care and Use Committee of Konkuk University (KU14086). Forty‐eight male Sprague–Dawley rats weighing 240–280 g (7 weeks of age) were housed separately in an animal resources facility in a room at a controlled temperature (20–22·8°C) and a light/dark cycle of 12 hours, and were provided food and water ad libitum. The rats were randomly assigned into two groups each consisting of 24 rats: the bilirubin group and the control group.

Using a chamber, the animals underwent general anaesthesia with 5% isoflurane (Aerane^®; Ilsung Pharmaceuticals, Seoul, Korea) initially, and were maintained with 1·5% isoflurane until the end of the procedure using a nasal cone. After shaving off their ventral hair, a rectangular‐shaped flap was marked. Bilateral flaps, 5 × 3 cm in size, were designed on the abdomen of each rat with the superior border being the costal margins and the inferior border located just above the horizontal line between the bilateral anterosuperior iliac spines (Figure 1). Bilirubin stock solutions were prepared to a final concentration of 2 mM by dissolving 0·5 ml of 0·2 N NaOH. The total volume was increased by addition of RPMI 1640 to a total volume of 50 ml. Adjustment of the pH to 7·4 was achieved by adding hydrochloric acid. Aliquots of 2 mM bilirubin in RPMI were stored at −80°C and were protected from light until being thawed and used for each experiment. The bilirubin group (n = 24) was injected with bilirubin by subdermal injection distributed evenly in each area (0·2–0·3 ml each) immediately after flap elevation. Meanwhile, the control group (n = 24) was pre‐treated with the same dose of only RPMI.

Figure 1.

Design of the bilateral deep inferior epigastric perforator (DIEP) flaps, 5 × 3 cm in size.

Twenty minutes after injection, a plastic surgeon, who was blinded to the injected material, made a vertical midline incision and elevated the bilateral encompassed skin territories from the midline towards the lateral side, slightly lateral to the linea albae, originating from the anterior rectus sheath. During the whole procedure, the perforators were carefully preserved. After assessing the patency and size of the perforators, the most reliable perforator emerging from the anterior rectus sheath in each side was selected and preserved (Figure 2). The skin paddle of the control side was returned to its bed using 4‐0 nylon. In the ischaemia side, the selected perforator was clamped for 15 minutes in 12 rats, and 30 minutes in the remaining 12 rats from each group, after which the flap was re‐inset. A schema of the study design is presented in Figure 3.

Figure 2.

Elevated right deep inferior epigastric perforator (DIEP) flap based on the most reliable perforator emerging from the anterior rectus sheath.

Figure 3.

Schema of the study design.

Evaluation of perforator flap

The flap survival was evaluated on postoperative days 0 to 5. On each day, a digital photograph was taken, and the flap survival rate was calculated on postoperative day 5 for each animal using a transparent sheet and ImageJ^® software (NIH, Bethesda, MD). The survival rate was assessed independently by two investigators who were blinded to the treatment groups, and was expressed as a percentage of the total flap area [survival rate (%) = viable area/total area × 100]. The calculated ratios were used for statistical analysis to examine the significance of the differences between the groups. Fresh tissue samples were taken from the skin flap above the perforator of the DIEP flaps 5 days after flap elevation in both the control and bilirubin groups. The samples were fixed with 10% formaline, processed with routine tissue preparation methods, embedded into paraffin blocks and stained with haematoxylin and eosin stain (H&E). A pathologist who was blinded to the treatment group evaluated the representative samples taken from both groups. Histological changes including cellularity of fibroblasts, reparative vascular proliferation, degree of inflammatory cell infiltration and stromal fibrosis were evaluated.

Statistical analysis

The results of the experiments are expressed as mean ± SD. For comparison of the flap survival rates, Student's t‐test was used. A P‐value <0·05 was considered statistically significant.

Results

The overall flap survival rate of the no ischaemia side in the control group was 94·46 ± 3·68%, and the flap survival rates of the ischaemic side after 15 and 30 minutes of perforator clamping were 89·63 ± 3·98% and 36·46 ± 6·38%, respectively.

In the bilirubin group, the overall flap survival rate of the no ischaemia side was 96·29 ± 3·72%, and the flap survival rates of the ischaemic side after 15 and 30 minutes of perforator clamping were 93·42 ± 4·48% and 83·96 ± 4·23%, respectively.

The flap survival rate following 30 minutes of perforator clamping in the bilirubin group was significantly higher than that in the control group (83·96 ± 4·23% versus 36·46 ± 6·38%, P < 0·001); and this was also the case following 15 minutes of perforator clamping (bilirubin versus control group: 93·42 ± 4·48% versus 89·63 ± 3·98%, P = 0·002). However, the difference in the flap survival rate in the no ischaemia sides between the bilirubin and control groups was not significant (96·29 ± 3·72% versus 94·46 ± 3·68%, P = 0·097). The above‐mentioned flap survival data are presented in Figure 4, and the representative gross images are presented in Figures 5, 6, 7, 8.

Figure 4.

Gross survival rates of the flaps according to the degree of ischaemia in the bilirubin and control groups.

Figure 5.

Gross findings of the perforator flaps in representative animals of the control group on postoperative days 0, 1, 2, 3 and 5 [right deep inferior epigastric perforator (DIEP) flap: no ischaemia; left DIEP flap: 15 minutes of perforator clamping].

Figure 6.

Gross findings of the perforator flaps in representative animals of the control group on postoperative days 0, 1, 2, 3 and 5 [right deep inferior epigastric perforator (DIEP) flap: no ischaemia; left DIEP flap: 30 minutes of perforator clamping].

Figure 7.

Gross findings of the perforator flaps in representative animals of the bilirubin group on postoperative days 0, 1, 2, 3 and 5 [right deep inferior epigastric perforator (DIEP) flap: no ischaemia; left DIEP flap: 15 minutes of perforator clamping].

Figure 8.

Gross findings of the perforator flaps in representative animals of the control group on postoperative days 0, 1, 2, 3 and 5 [right deep inferior epigastric perforator (DIEP) flap: no ischaemia; left DIEP flap: 30 minutes of perforator clamping].

There was no significant histological difference between the bilirubin and control groups without perforator clamping. However, there was more stromal haemorrhage, fibrosis and inflammatory cell infiltration in general in the control group than that in the bilirubin group both after 15 and 30 minutes of perforator clamping (Figure 9). These changes were more diffuse and severe in both the bilirubin and control groups after 30 minutes of perforator clamping than those after 15 minutes of perforator clamping. Other histological parameters such as degree of reparative vascular proliferation and cellularity of fibroblasts were generally similar between both groups although they varied partly in one animal in each group.

Figure 9.

Histological findings of the representative animals of the bilirubin and control groups after 15 and 30 minutes of perforator clamping (H&E stain, ×100).

Discussion

It is widely acknowledged that ischaemia can cause tissue injuries, and the concept of reperfusion injury after revascularisation following ischaemia has been well investigated. I/R injury induces a cascade of pathophysiological changes, including neutrophil influx, interstitial oedema and increased permeability, resulting in flap necrosis or loss 7. This injury may increase the morbidity and mortality in patients with trauma or tumours that require local or free perforator flap reconstruction, and the effective treatment of this type of injury is an area requiring further investigation. Recently, numerous studies aimed at determining how to alleviate I/R injury using various interventions have been reported 7, 8, 9, 10, 11, 12, 13, 14. In the current study, the hypothesis that preconditioning rats with bilirubin protects perforator flaps against subsequent I/R injury was investigated.

In this study, to induce I/R injury, inferior epigastric artery perforator flaps were designed, in which the feeding perforator of the flaps were clamped and removed 15 minutes (24 rats/group) and 30 minutes (24 rats/group) later. The perforator flap design used in the current study was originally described by Oksar et al. 15 in 2001, and this model is considered reliable for evaluating the effects of pharmacologic manipulations in terms of augmentation of perforator flap survival. According to our previous pilot study using the two 5 × 3 cm‐sized DIEP flaps used in the present study, clamping for more than 30 minutes can cause severe flap congestion, and no prominent recovery is gained following revascularisation. For this reason, it seemed that outcome comparisons were not easy to perform.

Studies on the beneficial effects of mild hyperbilirubinaemia on various diseases are ongoing. Mildly increased serum bilirubin has been suggested as a protective factor, and it may also be able to reduce the risk of coronary artery disease by acting as an antioxidant 16. In addition, we reported on the important role of the bilirubin system in ageing and age‐related diseases in our previous review 17.

In this study, the overall flap survival rate of the 5 × 3 cm‐sized DIEP flap without any further ischaemic intervention was 94·63% in the control group. This finding is consistent with the outcomes in the control groups of two prior studies, in which 97·78% of nine perforator flaps (3 × 3 cm) based on a single cranial epigastric perforator survived at postoperative day 7, and 99·93% of ten perforator flaps (4 × 4 cm) based on a single cranial epigastric perforator survived at postoperative day 5 18, 19. Similarly, our pilot experiment showed that the overall survival rate of a 5 × 5 cm‐sized DIEP flap was over 90% in a rat model. Accordingly, we believe that the perforator flap based on a single perforator is a reliable model to simulate relevant clinical circumstances.

In this study, we made some interesting observations. Many previous studies using rat models to study perforator flaps assessed the final flap survival rate at postoperative day 7; and most of those studies did not apply a silicone sheet under the flap bed to block revascularisation from the bed, as perforators in the plane between the panniculus carnosus and anterior rectus sheath might be injured by the silicone sheet. Our supplementary study showed that prominent vessel ingrowth was visible on postoperative day 7, but not on postoperative day 5. On the contrary, necrosis is not reversed, but rather shortened because of the tissue contracture. Therefore, unless the study is intended to evaluate the angiogenic potential of the novel material, our results suggest that the appropriate assessment timing of perforator flaps is on postoperative day 5 rather than day 7.

Another novel observation is the early effects of bilirubin on perforator flaps following perforator clamping. Moderate inflammation was always seen in the ischaemia side of the bilirubin group, and redness and swelling were more frequently observed in this group compared with that in the control group at postoperative days 2 and 3. However, at postoperative day 4, abrupt regression of inflammation was seen in the bilirubin group, while ongoing inflammation and subsequent necrosis ensued in the control group. We speculate that the active inflammation in this period was associated with better flap survival in the bilirubin group; however, this gross observation warrants further molecular studies to determine the exact mechanisms of bilirubin in the wound healing process in the future.

The protocols applied in I/R injury models in the previous studies on the topic are largely variable, with different durations of ischaemia and reperfusion being used. Our study used a relatively small perforator compared with these previous studies. Additionally, we performed a literature search using the Medline database to review the relevant articles on various flaps in rat models, with the focus on I/R injury, and we found that most of the identified articles used the inferior epigastric vessels as the pedicle, which is larger than our perforator 8, 18, 20, 21.

The duration of ischaemia ranged from 4 to 10 hours depending on the pedicle they used (4 hours of ischaemia for superior epigastric vessel in transverse rectus abdominis myocutaneous (TRAM) flap or 10 hours of ischaemia for superficial epigastric vessels in skin flap) 20, 22. Because of our very small perforator, the duration of ischaemia can be decreased. In addition, some previous studies evaluating various surgical, pharmacologic or anaesthesiologic interventions on I/R injury used muscle flaps 18, 23. While this is theoretically sound, because the muscle tissue is more vulnerable to ischaemia than skin flaps as a result of its lower critical ischaemic time and high metabolic requirement, plastic surgeons more frequently use versatile perforator flaps rather than conventional muscle flaps to reconstruct tissue defects; our study hence simulate the more relevant clinical trends.

The major limitation of our study is that it is a preliminary one to figure out whether the bilirubin has a positive effect on I/R injury based on gross and histopathologic findings. In addition, we still have not performed the molecular work to understand the respective roles of bilirubin, biliverdin reductase and haeme‐oxygenase 1. And another limitation of our study was that we did not investigate the molecular factors associated with I/R injury. These series of limitations are considered as an interesting area to investigate in the near future.

Recently, the beneficial effects of bilirubin in reducing the oxidant status in the wounds of diabetic rats have been reported 24. In line with this finding, our present study showed that the effect of bilirubin on normoxic tissue was not as strong as that on hypoxic tissue. We induced the I/R injury model by perforator clamping for 15 or 30 minutes. I/R‐induced reactive oxygen species contributes to the activation of adhesion molecules, which in turn leads to leukocyte infiltration and causes a series of changes that impair microcirculation. Thus, our results suggest that there may be an important network between bilirubin and these cascades of I/R injury. In other words, undiscovered molecular networks between hypoxic molecules and the bilirubin cycle may be responsible for these results, and further studies are warranted to better explain this novel finding.