Hyperglycemia Increases Susceptibility to Ischemic Necrosis

Diabetic patients are at risk for spontaneous foot ulcers, chronic wounds, infections, and tissue necrosis. Current theories suggest that the development and progression of diabetic foot ulcers are mainly caused by arteriosclerosis and peripheral neuropathy. Tissue necrosis plays a primordial role in the progression of diabetic foot ulcers but the underlying mechanisms are poorly understood. The aim of the present study was to investigate the effects of hyperglycemia per se on the susceptibility of ischemic tissue to necrosis, using a critical ischemic hind limb animal model. We inflicted the same degree of ischemia in both euglycemic and streptozotocin-induced hyperglycemic rats by resecting the external iliac, the femoral, and the saphenous arteries. Postoperative laser Doppler flowmetry of the ischemic feet showed the same degree of reduction in skin perfusion in both hyperglycemic and euglycemic animals. Nevertheless, we found a significantly higher rate of limb necrosis in hyperglycemic rats compared to euglycemic rats (71% versus 29%, resp.). In this study, we revealed that hyperglycemia per se increases the susceptibility to limb necrosis in ischemic conditions. Our results may help to better understand the physiopathology of progressive diabetic wounds and underline the importance of strict glycemic control in patients with critical limb ischemia.


Introduction
Diabetes mellitus (DM) favors the development of spontaneous foot and leg ulcers, chronic wounds, infections, and gangrene [1,2]. Approximately 85% of amputations in diabetic patients are preceded by a foot ulcer, which subsequently deteriorates into a severe infection or gangrene [3]. ese amputations signi�cantly contribute to the high morbidity and mortality rates as well as the high health care costs in patients with diabetic foot ulcers [4,5]. e most important factors related to the development of diabetic foot ulcers are peripheral neuropathy and arteriosclerosis. Furthermore, diabetic peripheral neuropathy not only leads to the loss of protective sensation, increasing the risk of unnoticed minor traumas, but also results in the degeneration of the sympathetic nerves innervating thermoregulatory arteriovenous shunt vessels of the skin [6]. ese denervated shunt vessels lose their normal capacity of contraction and the nutritional skin capillaries are bypassed leading to the development of chronic capillary ischemia [7]. At the same time, DM is associated with an acceleration of arteriosclerotic changes of the greater arteries resulting in decreased blood �ow, further contributing to a limb threatening ischemic condition [8].
However, it remains elusive whether hyperglycemia itself has a direct in�uence on the development of foot ulcers and its complications, especially in ischemic conditions. e aim of the present study was to evaluate the role of hyperglycemia per se in the physiopathology of spontaneous ischemic ulcers and on the susceptibility for ischemic necrosis, using a critical ischemic hind limb animal model with streptozotocininduced hyperglycemic rats.

Animals. Male Wistar rats (
) aged 90 days and weighing 300-350 g (Charles River Laboratories, L' Arbresle Cedex, France) were fed a standard diet and given water ad libitum. e local ethics committee and veterinary authority approved all procedures according to Swiss guidelines.
Rats were randomly assigned to two groups of equal size ( ): control group and hyperglycemic group. Hyperglycemia was induced by single i.p. injection of streptozotocin (STZ) (Sigma-Aldrich, St.Louis, MO, USA) three weeks before surgery. STZ was dissolved in sterile 0.1 M citrate buffer (pH 4.5) and injected intraperitoneally (50 mg/kg body weight) within 10 minutes of the preparation. Blood glucose levels were measured from tail venous blood by blood glucose test strips and just before sacri�cing from carotid blood using glucose oxidase method (Glu; Roche Diagnostics, Rotkreuz, Switzerland). Rats with glucose levels >11.1 mM were included in the hyperglycemic group.
2.2. Ischemic Foot Model. As previously described [9], through a longitudinal incision in the inguinal region that was shaved, the external iliac and femoral arteries were dissected from the common iliac to the saphenous arteries. To provoke an ischemic condition, the dissected arteries were resected from the common iliac in the le limb (ischemic limb, Figure 1) while in the right limb arteries were conserved and limbs considered being nonischemic (sham surgery). All surgical procedures were performed under an operating microscope (Carl Zeiss, Jena, Germany), and animals were anesthetized by inhalation of iso�urane 5% for induction and 3% for maintenance of anesthesia.

Macroscopic Limb Evaluation.
Limbs were assessed macroscopically every two days during the �rst week and T 1: Skin blood perfusion (PU) before (baseline) and aer resection of the arteries or sham surgery (days 1 and 7), in euglycemic and hyperglycemic animals.

Euglycemic animals
Hyperglycemic animals Sham surgery Resection Sham surgery Resection

Blood Flow Measurement.
Blood perfusion levels, measured by laser Doppler �owmetry, were similar in euglycemic and hyperglycemic animals in both ischemic and nonischemic conditions. In ischemic limbs the blood �ow signi�cantly decreased immediately aer resection of the arteries, and in animals which did not present necrosis the blood �ow remained low during the �rst week (Table 1).

Limb Necrosis.
In nonischemic conditions, no foot necrosis was observed neither in euglycemic nor in hyperglycemic animals. In ischemic conditions, a signi�cantly higher occurrence of limb necrosis was observed in hyperglycemic animals compared to euglycemic animals.
In hyperglycemic animals, �ve out of seven ischemic limbs showed total necrosis (71%), affecting the feet and lower limbs up to the level of the knee. is rate of limb necrosis was signi�cantly higher than in ischemic limbs of euglycemic rats (2 out of 7, 29%, , Table 2). In limbs that would eventually become necrotic, we observed signs of critical ischemia starting from day three (cyanotic aspect, stiffness, and coldness) that rapidly progressed in necrosis from proximal to distal, motivating the immediate sacri�ce of the animal.

Discussion
Every 30 seconds, somewhere in the world, a lower limb is amputated due to DM [10]. e progression of critical ischemic limbs to necrosis and gangrene is reportedly 40% in diabetic patients compared to 9% in nondiabetic patients [11]. Since current therapeutic approaches of diabetic wounds oen are unsatisfactory, it is of crucial importance to better understand the pathophysiology of the progressive necrosis at the wound margins that leads to the characteristic extension of diabetic wounds. In one of our precedent studies, we developed a new model of ischemic foot ulcers. We aimed at creating a severe ischemic condition without provoking limb necrosis to study wound healing in ischemic conditions [9]. We assessed different degrees of ischemic conditions by resecting arteries on different levels. We concluded that in euglycemic rats, resection of the external iliac artery, the femoral artery, and the saphenous artery down to the level of the knee resulted in a severe ischemic condition with a delayed wound healing process, but with low occurrence of limb necrosis.
In the present study, we aimed at assessing the effects of hyperglycemia on tissue susceptibility for ischemic necrosis. We used the same model as described above in euglycemic and streptozotocin-induced hyperglycemic rats. e same degree of ischemia was in�icted to euglycemic and hyperglycemic rats by resecting the external iliac, the femoral and the saphenous artery. To avoid the effects of longterm complications of DM, in particular of vasculopathy and neuropathy, we induced ischemia only three weeks aer injection of streptozotocin. By laser Doppler �owmetry we con�rmed that the skin perfusion at the dorsal surface of the feet was indeed comparable in hyperglycemic and euglycemic animals. Nevertheless, we found a signi�cantly higher rate of limb necrosis in hyperglycemic rats compared to euglycemic rats (71% versus 29%, resp.).
Our results demonstrate a direct relation between hyperglycemia and acute ischemic necrosis. We hypothesize that the decrease of tissue tolerance to ischemia induced by hyperglycemia could be responsible for the initial skin lesions, which would lead to the development of diabetic ulcers. is could explain the speci�c evolution of diabetic wounds toward peripheral progressive necrosis.
e phenomenon of decreased tolerance to ischemia caused by hyperglycemia has already been investigated in other ischemic tissues, notably in heart and brain [12][13][14][15].
Hyperglycemia has been associated with impaired outcome and expanded infarct volume in patients with ischemic strokes and myocardial infarctions [15,16]. A recent metaanalysis of the association between hyperglycemia and stroke size in animal models revealed that hyperglycemic animals presented 94% larger infarct areas (140% in streptozotocin induced hyperglycemia and 48% aer dextrose infusion) compared to euglycemic animals [17]. Several mechanisms have been identi�ed through which hyperglycemia could aggravate cerebral and cardiac injury aer ischemia, whereas a mechanistic link between hyperglycemia and susceptibility to necrosis is yet to be established. We propose the following hypotheses to explain how hyperglycemia per se could sensitize to ischemic tissue necrosis: (1) increased blood viscosity and hypercoagulability worsening progressively the level of ischemia, and (2) direct cellular toxicity due to the development of increased oxidative stress and lactic acidosis.
Hyperglycemia increases plasma �brinogen levels and thereby blood viscosity and coagulability [18][19][20]. In nondiabetic patients with acute ischemic stroke, high glucose levels are related to a hypercoagulant state and a worsened outcome [21]. Furthermore, it has been shown that increased blood viscosity is associated with decreased tcPO2 levels in diabetic feet [22]. e increased blood viscosity in hyperglycemic conditions could therefore further aggravate the oxygen deprivation caused by the initial ischemia. In fact, it has been shown that dalteparin, a low molecule weight heparin, improves local tissue oxygenation in patients with diabetes, severe vascular disease, and foot ulcers [22].
Oxidative stress induced by hyperglycemia has been suggested as another possible mechanism of tissue injury exacerbation aer strokes and myocardial infarctions [23,24]. Indeed, both ischemia and hyperglycemia leads to excessive production of reactive oxygen species (ROS), and it is possible that the two factors accelerate each other. On the one hand, ischemia leads to oxidative stress through the generation of ROS through NADPH oxidase and other oxidases [25]. On the other hand, glucose has been shown to be the requisite electron donor for superoxide production by NADPH oxidase [23]. Additionally, hyperglycemia leads to the generation of free oxygen radicals also through the accumulation of advanced glycosylation end products (AGEs) [26,27]. Further research is needed in order to evaluate whether the oxidative stress provoked by ischemia and exacerbated by hyperglycemia is a major player in the pathophysiology of ischemic cell necrosis.
Lactic acidosis could be another important contributor to ischemic cell death in hyperglycemic conditions [28]. When blood �ow and oxygen supply decrease to critical levels, cells generate energy predominantly through anaerobic glycolysis. As a result, blood concentration of lactic acid rises and the pH level drops [29,30]. Under hyperglycemic conditions, there is more substrate available for anaerobic glycolysis and consequently lactic acid production rises [31]. In fact, in an ischemic stroke animal model, it was shown that bolus infusion of glucose resulted in a severe intracellular acidosis and increased tissue death [32]. We suggest that when combined with elevated glucose levels, the ischemia-induced shi to anaerobic glycolysis leads to elevated tissue acidosis, which may accentuate cellular death related to ischemia.
In hyperglycemic conditions, the combination of oxidative stress, acidosis, and hypercoagulability creates a nocuous cellular environment that favors progressive ischemic necrosis. As the mechanisms through which both ischemia and hyperglycemia induce cellular toxicity are similar, it is possible that they amplify each other.

Conclusions
e unfavorable evolution of many diabetic feet is not only a result of neurogenic and arteriosclerotic long-term effects of diabetes, but could also be, as we observed, a direct effect of hyperglycemia itself. In this study, we demonstrated for the �rst time that at similar levels of ischemia, limb necrosis appears signi�cantly more oen in hyperglycemic conditions compared to euglycemic conditions. is �nding suggests that hyperglycemia sensitizes to the effects of ischemia, revealing a direct involvement of hyperglycemia in the events leading to ischemic necrosis. Our results may help to better understand the physiopathology of progressive diabetic wounds and underline the importance of strict glycemic control in patients with critical limb ischemia.

Authors' Contribution
D. Lévigne and M. Tobalem have equally contributed to this work.

Con�ict of �nterests
e authors have no �nancial interests to declare in relation to the content of this paper. e Swiss National Science Foundation provided the only �nancial support.