INTRODUCTION
Wound healing is an integral part of recovery of critically ill patients[1]. They often are at risk for impaired healing because of their compromised body systems and multiple risk factors. Disruption of healing may result in delayed healing, dehiscence, infection, and death. Understanding healing and the factors that may impede it provides the physician with the foundation on which care is planned to facilitate wound healing.
IMPEDIMENTS TO WOUND HEALING
The treatment of chronic wounds presents a challenge to physicians, caregivers, and patients[2]. A familiarity with the factors that impede healing is necessary if treatment of chronic wounds is to be effective. These wounds often heal in a short period of time if the factors inhibiting wound healing are adequately identified and managed. Recombinant growth factor therapy may provide an additional stimulus to healing in certain types of chronic wounds. However, there remains no substitute for a physiologic environment conductive to tissue repair and regeneration, without which the efficacy of growth factor therapy is questionable. Some of the most commonly encountered and clinically significant impediments to wound healing include tissue hypoxia, infection, presence of debris and necrotic tissue, use of anti-inflammatory medications, a diet deficient in vitamins or minerals, or general nutritional deficiencies, tumors, environmental factors, and metabolic disorders such as diabetes mellitus. Treatment of chronic wounds should be directed against the main etiologic factors responsible for the wound. Moreover, factors that may impede healing must be identified and corrected for healing to occur if possible.
Especially, regarding diabetes mellitus, there have been numerous experimental and clinical studies, not only establishing its negative impact on wound restoration but also studying its metabolic pathophysiology in comparison with every stage of wound healing. This is not only because diabetes mellitus is one of the most common clinical entities found, but also because it coexists with other metabolic disorders, such as hypothyroidism. While hypothyroidism alone, or as part of a clinical condition, like myxedema, is a well established disease, little research has been made to determine its exact effect on tissue healing and the biomechanical profile of wounds in hypothyroid patients.
COMPLICATIONS OF DIABETES MELLITUS
The long term complications of both types of diabetes mellitus are usually categorized into macrovascular, microvascular, and neuropathic diseases[3]. Macrovascular complications are associated with an accelerated rate of atherosclerosis and an increased propensity for peripheral vascular disease, myocardial infarction, and cerebrovascular accidents. Microvascular complications include retinopathy and nephropathy and are associated with thickening of the capillary basement membrane. Also, the current suggestion by investigators states that microcirculatory damage is apparent in the skin and subcutaneous tissue, leading to impaired wound healing. An intact microcirculation is required for tissue nutrition, removal of waste products, inflammatory responses and temperature regulation therefore, logically any defect in microvascular function adversely affects tissue repair. One undisputed structural change that occurs in people with diabetes is thickening of the basement membrane, the extracellular matrix (ECM) bellow the cells lining the vessels. This is a consistent finding in the capillaries and arterioles of people with diabetes[4]. Studies have demonstrated that in early diabetes, microvascular blood flow is increased with prolonged exposure to hyperglycemia[5]. This is explained by increased production of nitric oxide by the endothelial cells as a result of vessel wall damage and heightened viscosity of the blood due to glycosylation of hemoglobin[6]. Hyperemia and capillary hypertension ensue. Precapillary resistance is reduced and further vasodilatation occurs in response to increased metabolic demand and oxygen requirements of damaged tissue. In turn, shear stress on the vascular wall increases leading to accumulation of ECM and the protein fibronectin in response to this chronic injury. Further basement membrane thickening occurs, altering membrane charge and as a result affecting capillary permeability. Microangiopathy-capillary disease has progressive pathology, particularly affecting cells in which glucose transport and metabolism are independent of insulin, such as endothelial cells[7]. Persistent hyperglycemia encourages the conversion within the endothelial cells of glucose to sorbitol. Cellular edema occurs, as sorbitol cannot diffuse across the cell membrane. This results in metabolic alterations, membrane function alteration and basement membrane thickening. It is also thought that hyperglycemia leads to increased protein synthesis, deposition of abnormal ECM proteins and further basement membrane thickening. However, there is a lack of human experimental evidence to support this[6]. As microvascular disease progresses usually in parallel with the duration of diabetes, impaired vascular reactivity and limited hyperemia result in the loss of normal autoregulatory function[4]. Homeostatic balance relies on the microvascular circulation maintaining a constant blood supply despite fluctuations in perfusion pressure. However, a thickened and rigid basement membrane restricts the normal hyperemic response to tissue damage and hypoxia ensues. Skin tissue breakdown inevitably occurs.
Neuropathic complications include damage to sensory, motor, and autonomic nerves.
WHY BLOOD GLUCOSE AFFECTS HEALING
Patients with diabetes often have wounds that are difficult to heal[8]. The initial barrier to healing is an increased blood glucose level, which causes the cell walls to become rigid, impairing blood flow through the critical small vessels at the wound surface and impeding red blood cell permeability and flow. Impairment release of oxygen by hemoglobin results in oxygen and nutrient deficits in the wound. A less optimal immune function also contributes to poor wound healing in patients with diabetes. When blood glucose levels are persistently elevated, chemotaxis and phagocytosis are compromised. Chemotaxis is the process by which white cells are attracted to the site of infection, while phagocytosis is the ingestion of bacteria by white cells. Both processes are important in controlling wound infections. Diabetic infections take a longer time to heal because of delayed macrophage introduction and diminished leukocyte migration, which causes a prolonged inflammatory phase in the wound healing cascade.
Protein-calorie malnutrition and the resultant body composition changes are an additional consideration in wound healing. Patients with diabetes often have a progressive loss of lean body mass, which is replaced with a metabolically inactive fat mass.
The absence or deficiency of insulin in diabetes mellitus causes impaired metabolism of carbohydrates, fat and proteins, which are necessary for cellular activities and tissue synthesis in wound healing[9]. Insulin is required for glucose to enter cells as to provide a source of energy for uptake of amino acids to synthesize proteins and for inhibition of adipose tissue lipolysis. Glucose is also needed to supply energy for fibroblastic and polymorphonuclear (PMN) activities during wound healing. Altered glucose metabolism, as seen in diabetes mellitus, leads to defective metabolism of these nutrients and reduces fibroblastic/PMN activity, causing impaired wound healing.
Altered protein metabolism
Protein is essential for the synthesis of collagen structures that establish wound tensile strength. Without adequate tensile strength of collagen structures, dehiscence and evisceration can occur. Protein deficiencies also impair formation of new capillaries, fibroblastic activity, and the bactericidal activity of PMNs. Insulin, which is lacking in diabetes, is the anabolic hormone that stimulates protein synthesis, so to promote wound healing, both insulin and protein need to be provided in adequate amounts for protein tissue rebuilding.
Altered lipid metabolism
Without insulin and glucose for energy, fat is the primary source of body fuel. So, when fatty acids, which are used to form cell membranes, are used for energy, as in poorly controlled diabetes mellitus, healing is impaired, since there is delayed cell membrane synthesis.
Decreased chemotaxis, phagocytosis, bacterial killing[10,11], decreased heat shock protein expression[12], less antioxidant synthesis , and increased oxygen free radical generation[13,14] during the early phase of wounding healing all have been implicated in impaired diabetic healing. Additionally, growth factor depletion[15,16], increased glucocorticoid concentration[17], decreased cell proliferation[18], and up-regulation of apoptosis[19] characterize the later phases of diabetic healing resulting in poorer granulation tissue formation. Diabetes mellitus lowers extracellular matrix synthesis following wounding[20].
While the relation of diabetes mellitus to wound healing is well elicited in numerous other studies and papers[21-25] demonstrating that it is a complex metabolic disorder whose components have several direct and indirect effects on the healing of wounds, Komesu
et al[26] evaluated the healing process during early phases of experimental diabetes on rat skin and found that not only the initial healing phase has a slow beginning and tends to last longer, but also that healing areas show lower density of neutrophils after surgery, and in addition, after three days, when the neutrophils leave the healing area and are replaced by macrophages, diabetic animals show a higher number of neutrophils. The principal conclusion is that although diabetes is a chronic progressive disease, acute diabetes can be associated to subclinical alteration, and is considered responsible for deficiencies in defense cells and in repair tissue failures.
METABOLIC PATHOPHYSIOLOGY OF DIABETES MELLITUS
There are three possible ways that hyperglycemia contributes to the metabolic pathophysiology of diabetes-related complications[27].
The first hypothesis is that abnormal glucose levels alter the actual control of cellular Na+/K+ ATPase activity[28]. Hyperglycemia leads to increased polyol pathway activity, which leads to a depletion of myo-inositol stores. This involves the biochemical pathway in which glucose is converted to sorbitol by aldose reductase. Because the conversion of sorbitol to fructose (by sorbitol dehydrogenase) is slow, sorbitol tends to collect in cells. The end-product of the sorbitol to fructose is NADH. The increased conversion to fructose leads to decreased NADPH and increased risk for oxidative stress. The increased levels of sorbitol may increase the osmotic load in the cells. Sorbitol inhibits myo-inositol uptake, which in turn leads to the alteration in Na+/K+ ATPase activity.
A second metabolic abnormality that results from hyperglycemia may be related to the activity of protein kinase C (PKC). Diabetes appears to increase the synthesis of diacyl-glycerol, which in turn leads to increased PKC activity. PKC is a key signaling receptor for many cellular activities including proliferation, contraction, calcium influx and others.
Thirdly, hyperglycemia also leads to the production of pathologic by-products. Hyperglycemia leads to ‘advanced glycosylation end products’ (AGEs), which are large aggregates of aldoses covalently bound to reactive amino groups. AGEs may lead to increased oxidative stress and activate a key transcription factor, nuclear factor (NF)-κB. NF-κB is involved in many cytokine-related cell responses. For instance, AGEs appear to induce the production of platelet-derived growth factor (PDGF), tumor necrosis factor-α, and interleukin-1α. AGES may also lead to collagen cross-linking and inhibit normal collagen degradation.
Fahey et al[29] examined leukocyte infiltration and the appearance of tumor necrosis factor-α (TNFα) and interleukin-6 (IL-6) in wound chambers implanted in normal and streptozotocin-induced diabetic mice. The data suggest that delayed healing on diabetes is associated with altered leukocyte infiltration and wound fluid IL-6 levels during the late inflammatory phases of wound healing. These observations suggest that while the initial inflammatory response is intact, diabetes can impair the late inflammatory response to wound healing, data that corroborate previous reports from Goodson and Hunt[30-32] that implicate a defective inflammatory response to wounds in the pathophysiology of delayed wound healing are associated with diabetes. They have shown that obesity, insulin resistance, hyperglycemia, and depressed leukocyte function interfere with collagen synthesis and thus impair wound healing. Their studies of skin and subcutaneous wounds in diabetic mice showed that the granulocyte influx is slow and is associated with depressed synthesis of protocollagen and collagen. Hennessey et al[33] postulated that altered collagen metabolism seen in experimental diabetes could be related to the presence of increased tissue levels of advanced glycosylation end products. Others, like McMurry[34] suggested that impaired leukocyte function may contribute to the impaired healing in diabetes based on abundant evidence that neutrophil chemotaxis and phogocytosis are inhibited by hyperglycemia and poorly-controlled diabetes mellitus[35]. Robson et al[36] have addressed the issue of wound infections in diabetic patients by evaluating the relative growth of various bacteria in the presence of hyperglycemia, and found that gram-positive bacteria thrive in hyperglycemic serum and that Gram-negative bacteria grow less well in hyperglycemic serum, which could partially explain the clinical observation that diabetic patients are prone to staphylococcal infection.
Additionally, defects in wound healing of experimental diabetes may be corrected, at least in part, by admini-stration of large amounts of vitamin A[37] or zinc[38].
Last but not least, uremia is another factor that alters healing[39]. Patients with diabetes mellitus are prone to silent or overt renal disease. Increased urinary protein loss in these diseases predisposes the patient to edema, which contributes to impaired tissue repair. In addition, uremia is another factor that independently contributes to the altered healing.
WOUND HEALING AND GROWTH FACTORS
Recent advances in wound treatment include topical growth factor therapy, which has been successful in diabetic wounds. Platelet-derived growth factor (PDGF), fibroblast growth factor, and epidermal growth factor have been shown to accelerate tissue repair in an experimental wound model. Insulin is one of the primary anabolic hormones in the body, and numerous studies have shown beneficial effects of insulin therapy on wound healing. Insulin increases wound tensile strength and stimulates protein anabolism in skin and muscle. One possible reason for their success is the relative deficiency of growth factors in chronic wound fluid due to decreased supply, increased binding, or increased degradation of the naturally occurring growth factors. A study[40] was performed to examine the level of the insulin-degrading enzyme (IDE) in diabetic wound fluid and to relate this to glucose control [hemoglobin A1c (HbA1c) levels]. An excess of this enzyme could contribute to reduced levels of growth factors in wound fluid. The authors showed that insulin-degrading activity is present in human wound fluid, and that the activity is higher in subjects with diabetes. The biochemical properties of the wound fluid insulin-degrading activity are consistent with the properties of the insulin-degrading enzyme (insulysin). Insulin-degrading activity correlates with HbA1c levels, suggesting a mechanism for the relationship between glucose controlling and wound healing. Improvement in glucose control is a critical factor in wound healing, but a reduction in wound fluid insulin-degrading activity is also a potential therapeutic approach.
PDGF levels have been examined by radioimmunoassay in wound tissue of normal and diabetic rats, while immunohistochemical analysis can be utilized to localize and characterize PDGF immunopositive cells at the wound site of normal and diabetic animals[41]. The findings suggest that absence of an initial increase in PDGF may play an important role in poor wound healing observed in diabetic animals, and that the reduction in PDGF could be related to decreased cellular PDGF production rather than lack of PDGF-producing cells. The authors concluded that perhaps the diabetic state inhibits cellular PDGF gene expression signaled by wounding or interferes with normal PDGF expression at the wound site.
To examine the effects of recombinant growth factors in vivo, impaired wound healing has been studied in genetically diabetic C57BL/KsJ-db/db mice[42]. Large full-thickness skin wounds, made on the backs of these mice, exhibited significant delays in the entry of inflammatory cells into the wound, the formation of granulation tissue, and in wound closure, when compared to their nondiabetic littermates. Recombinant human platelet-derived growth factor (rPDGF-BB), recombinant human basic fibroblast growth factor (rbFGF), or combinations of both were applied topically to the wounds for five to fourteen days after wounding. Combinations of rPDGF-BB and rbFGF improved all parameters of healing but not to a greater extent than either growth factor alone.
Numerous other studies have held to elucidate the effect of growth factors in the process of healing of diabetic wounds. Hennessey et al[43] studied the interactions of topically applied insulin and epidermal growth factor (EGF) in diabetic rats. The EGF and insulin promoted a 202% increase over controls in collagen synthesis after fifteen days, while diabetic rats receiving EGF or insulin alone had significantly less collagen than controls. The individual effects of insulin and EGF added synergistically for a net gain in wound collagen content are a gain that has not been observed with either EGF or insulin alone. Bitar et al[44] have proved that diabetes induces not only a marked reduction in insulin-like growth factor-I (IGF-I) levels both in wound fluid and serum of diabetic rats, but also a reduction in the levels of IGF binding proteins and transforming growth factor-β (TGF-β) in diabetic wound fluid. The investigators reported that the diabetic wound fluid increased levels of extracellular matrix gelatinases, and that a single dose of TGF-β partially reversed the diabetes-related decrease in the tensile strength of standardized incisions.
Furthermore, elevated levels of glucose have been shown to affect insulin signaling in various ways. Hyperglycemia can alter glucose-induced insulin secretion, a phenomenon referred to as glucose toxicity. Glucose is known to affect insulin action as well by regulating the expression of several genes, including the IGF-I receptor (IGFR) and insulin receptor (IR) genes, at both the transcriptional and translational levels[45]. Moreover, hyperglycemia can inhibit insulin action. This inhibition is thought to be a result of serine phosphorylation through a PKC-mediated mechanism as well as by activation of protein tyrosine phosphatases, which deactivates the IR function. In addition to its possible involvement in inducing complications of chronic diabetes, glucose down-regulates its own transport and metabolism. In this study, the authors investigated the relative roles of hyperglycemia, insulin, and IGF-I, all of which are abnormal in diabetes, in primary murine skin keratinocytes. The results showed that in the presence of high glucose (20 mmol/L), the glucose transport rate of primary proliferating or differentiating keratinocytes was downregulated, whereas at 2 mmol/L glucose the transport rate was increased. These changes are associated with changes in the GLUT1 expression and in the affinity constant (km) of the transport. Exposure to high glucose is associated with changes in cellular morphology, decreased proliferation and enhancement of Ca-induced differentiation of keratinocytes. Furthermore, in the presence of high glucose, ligand-induced IGF-I receptor, but not insulin receptor (IR) autophosphorylation, is decreased. Consequently, in high glucose, the effects of IGF-I on glucose uptake and keratinocyte proliferation are inhibited. Interestingly, lack of IR expression in IR-null keratinocytes abolishes insulin-induced glucose uptake and partially decreases insulin- and IGF-I-induced proliferation, demonstrating the direct involvement of the IR in these processes. The results demonstrate that hyperglycemia and impaired insulin signaling might be directly involved in the development of chronic complications of diabetes, like wound healing, by impairing glucose utilization of skin keratinocytes as well as skin proliferation and differentiation.
CONFLICTING RESULTS
In a recent retrospective study though[46] that was designed to assess the contribution of several factors that could be incriminated in the occurrence of abdominal wound dehiscence in major abdominal operations, diabetes mellitus was found not to be a significant risk factor, in contrast with advanced age (> 65 years), emergency operation, cancer, haemodynamic instability, intra-abdominal sepsis, wound infection, hypoalbuminemia, ascites, obesity, and steroids. The mortality and the possibility of dehiscence seem to correlate directly with the number of those risk factors.
HISTOLOGICAL, HISTOCHEMICAL AND BIOCHEMICAL STUDY OF DIABETIC WOUNDS
Prakash et al[47,48] have studied surgical wounds in albino rats. Uncontrolled diabetics, controlled diabetics and normal hypoglycemic rats were used for study and compared with normal control animals. The parameters of histology, histochemistry and biochemistry were assessed. Healing was delayed in the uncontrolled diabetic group and was significantly ameliorated by the administration of insulin, but not to an extent as to be better than normal controls. More specifically, the uncontrolled diabetic group showed a more marked and prolonged inflammatory response. The first appearance of scar was after seven days as contrasted at five days in the other groups, while at the end of fifteen days the scar was significantly more vascular and cellular in the uncontrolled diabetic rats. Histochemically, the precollagen (reticulin) fibers were present in moderate amounts on the 5th post-operative d, increasing up to the 10th d, after which a decline was obvious. As a contrast, practically no stainable reticulin was seen on the 5th d in the diabetic group. It appeared on the 7th d and increased till the 15th d. The reticulin fibers in that group were sparse and poorly staining. The uncontrolled diabetic group showed delay in the appearance of collagen, increasing up to the 15th d, when it was less deeply staining and less dense, in contrast with the other groups where collagen was present in minimal amounts on the 5th d. The acid and neutral mucopolysaccharides were also under study. Traces were present on the second day after which there was an increase till the seventh day. A decline in stain of this substance was present thereafter. This response was observed in the other three groups. The uncontrolled diabetic group differed by showing the presence of mucopolysaccharides till the 15th d, indicating activity of the healing processes. The hydroxyproline content was seen to register a fall on the second day after wounding as compared with the first day. Subsequently, a steady rise was seen up to the 7th d. In the uncontrolled diabetic rat group there was a rise obvious from the 5th d onwards after which there was a steady gain in hydroxyproline content, to values lower than in the other three groups.
INFLUENCE OF DIABETES ON GI WOUND HEALING
While it is widely accepted that changes in the extracellular matrix, particularly due to collagen metabolism and related disturbances like diabetes mellitus, are a prognostic factor for anastomotic leakages[49,50], Gottrup et al[51] have investigated the influence of experimental diabetes on the healing of incisional wounds in stomach and duodenum. Experimental diabetes impaired the mechanical strength of healing wounds in stomach and duodenum. The reduction was most pronounced for breaking energy, while only duodenum showed a decrease for breaking strength of the tissues, an impairment that became more pronounced by increasing healing time. The investigators showed that there is a relation between mechanical strength and total collagen content and that insulin treatment prevents these retardations.
The influence of diabetes on the healing of HCL-induced gastric lesions and the healing promoting effect of basic fibroblast growth factor (bFGF) on these lesions under diabetic conditions induced in rats by streptozotocin has been reported[52]. Diabetic conditions could not affect the development of HCI-induced gastric lesions but could significantly delay the healing of these lesions. Daily administration of insulin could return high blood glucose levels to normal ranges and significantly antagonize the delayed healing of these lesions. The delayed healing in diabetic rats could also be significantly promoted by recombinant human basic fibroblast growth factor without any effect on blood glucose level. The mucosal bFGF levels in streptozotocin-diabetic rats are significantly lower under basal conditions before HCI treatment and do not increase after injury, yet such dysregulation of bFGF production could be partially restored by insulin treatment. These results suggest that diabetic conditions have deleterious influences on the healing of acute gastric lesions in both an insulin- and bFGF-sensitive manner, and that the administration of exogenous bFGF antagonizes the delayed healing of gastric lesions observed in diabetic animals.
Verhofstad and Hendriks[53] demonstrated in an experimental study in rats that diabetes indeed lower bursting strength in both ileal and colon anastomosis. However, this effect is not secondary to decreased collagen deposition bur rather due to increased abscess formation around the anastomosis. Incubation of fibroblasts isolated from ileum or skin of nondiabetic rats with diabetic and nondiabetic serum only affects collagen synthesis in skin fibroblasts, suggesting that fibroblasts from different organs are distinctly affected by diabetes[54]. Postoperative restoration of blood glucose by insulin[55] or islet transplantation[56] only partially restores bursting pressure. Enhanced matrix metalloproteinase activity has been shown to be present in diabetic colon anastomosis, possibly explaining the weaker breaking strength. In contrast to dermal healing, diabetic anastomoses are characterized by an elevated number of neutrophils, which could serve as a source of matrix metalloproteinases[57].
Martens et al[58], studying the postoperative changes in collagen synthesis in intestinal anastomosis of the rat, investigated the differences between small and large bowel. The results showed that the ileum responds more quickly and strongly to wounding than the colon, at least as far as the production of new collagen is concerned. Possibly, this phenomenon contributes to the lower failure rate apparent for anastomoses in the small bowel.
Verhofstad et al[59] have studied the cellular and architectural parameters of anastomotic healing, this time in diabetic rats. The results showed that although anastomotic necrosis, edema, and epithelial recovery are not affected by diabetes, in diabetic rats, the number of polymorphonuclear cells and macrophages is significantly increased, both in ileal and colonic anastomosis. Repair of the submucosal-muscular layer in colonic anastomoses from diabetic rats is impaired, but in ileal anastomoses no difference is found. In the anastomotic area, collagen deposition remains unaffected by diabetes, thus proving that experimental diabetes leads to alterations in cellular components involved in the early phase of repair of intestinal anastomoses, but not to a reduced accumulation of wound collagen.
Most recently, Onodera et al[60] studied anastomotic leakage in diabetic rats. The purpose of the study was to determine whether collagen synthesis was correlated with the anastomotic strength in diabetic animals in special reference to collagen type differences. The investigators showed that although there is no statistical significance between the control and diabetic groups in hydroxyproline concentrations and that although the expressions of mRNA of collagen type III in both groups are not statistically different, the expression of collagen type I in the diabetes group merely increases and is clearly less than that of the control group. The increase of type I collagen is much more pronounced and long-lasting than that of type III collagen. Brasken et al[61] suggested that type I collagen synthesis is enhanced for at least 2 wk after anastomosis whereas type III collagen synthesis is increased only during the first week. Taken together with previous experimental results, the weakness of anastomotic healing with diabetic rats is not attributed to the total contents of collagen but to changes of newly formed collagen, i.e. mainly the blockage of the type I collagen synthesis.
CONFLICTING RESULTS
Although, other investigators like Witte and Barbul[62] and Wagner and Egger[50] have recognized that colon healing is a structured cascade of different phases that can be affected by a multitude of local (infection, ischemia) and systemic (diabetes, malnutrition, anemia, hypothermia, trauma) factors, a recent retrospective study done to identify the risk factors for anastomotic leakage after left-sided colorectal resections with rectal anastomosis, showed that diabetes is a non-significant variable[63].
HYPOTHYROIDISM AND WOUND HEALING
Hypothyroid patients have long been thought to have a higher risk of surgical morbidity and mortality[64,65]. Thyroid hormone deficiency does cause derangements of cardiovascular[66], pulmonary[67], renal[68] and central nervous system functions[69], and alters drug metabolism[70] in ways that could predispose to surgical complications. Surgery in hypothyroid patients is associated with an increased risk of minor perioperative complications, which should be anticipated and preemptively managed in the course of their anesthetic and surgical care[71], a conclusion which was validated by Ladenson PW et al[72], who reported that postoperatively, hypothyroid patients more commonly have, among others, gastrointestinal and neuropsychiatric complications than control patients.
Since collagen is the only protein in the body containing hydroxyproline in significant amounts, urinary excretion of hydroxyproline has been considered an index of collagen metabolism since the 60’s[73]. Hormones concerned with body growth and protein metabolism particularly affect collagen metabolism[74]. Thyroid hormone can stimulate hydroxyproline excretion and correct the reduction of hydroxyproline excretion in hypothyroidism[75].
Kowalewski and Yong[76] have shown a significant increase of a saline-extractable, and total soluble hydro-xyproline, and a reduction of insoluble fraction in the normal wound. This might be due to vigorous biosynth-esis of collagen. In contrast, no evidence of the synthesis of new collagen has been found in the wounds of hypothyroid rats. Insoluble fractions are higher in the wounds of hypothyroid rats than in normal control animals, suggesting that biosynthesis, solubility, and overall metabolism of collagen in hypothyroid rats are deficient.
Experimental work by Lennox and Johnston[77] has shown that wound healing is accelerated by a mean of 2.5 d in a hyperthyroid group of rats and delayed by a mean of 2.0 d in the hypothyroid group as compared with control rats. Mehregan and Zamick[78] demonstrated that T3 has a beneficial effect on the healing of deep dermal burns in rats. There is better organization of collagen bundles, fewer retraction spaces, and smoother scars. Another experimental study by Herndon et al[79] showed that levels of thyroid hormone have a profound effect on the rate of healing of burn wounds in thyroidectomized guinea pigs. They showed that epithelization is inhibited at low and high doses of levothyroxine sodium. However, administration of levothyroxine sodium at the intermediate dose of 30 μg/kg results in improved wound closure relative to euthyroid control subjects. This finding is consistent with the results of Kivirikko et al[80], who showed that the rate of collagen synthesis is decreased both in hyperthyroidism and hypothyroidism. On the other hand, there are clinical reports of kelloid development during the healing process of wounds of patients with hyperthyroidism[81], but no change in wound healing with euthyroid hamsters receiving ip T4[82]. A retrospective study[83] concluded that the conditions of patients with laryngeal tumors treated with surgery or surgery and radiation should be evaluated for hypothyroidism. When patients become hypothyroid, they should be treated promptly. Treatment of thyroid dysfunction can accelerate healing in patients with complications such as flap edema, flap necrosis, or fistula. Cannon[84] reported in an experimental study that although hypothyroidism alone, occurring in patients undergoing treatment for head and neck neoplasms, has no significant unfavorable impact on wound tensile strengths or flap survival, when combined with preoperative radiation, there are statistically deleterious effects on both wound tensile strengths and flap survival. Histologically, collagen fibers within the wound appear shorter and thinner, which probably account for decreased wound tensile strengths. It is recommended that thyroid function tests should be done routinely before and after therapy in all patients with head and neck cancer undergoing combined surgery and radiation.
Another study also demonstrated that the hormonal replacement therapy in hypothyroidism cases is beneficial with regard to wound healing and the results are more satisfactory if zinc is added to the therapy[85]. A decrease in serum zinc level in hypothyroidism has been reported in many clinical and experimental studies[86]. As noted above, experimental and clinical studies on hypothyroidism have shown that L-thyroxine replacement therapy has a positive effect on wound healing, which is generally delayed in this disorder. Elevation of wound breaking strength is one of the most important phenomena observed during the wound healing. In addition, the important factors for the increase of breaking strength are the amount of collagen of the wound and the increase in intramolecular-intermolecular cohesive bounds of collagen present in the wound (cross-linking), the later being most important[84]. Lysyl oxidase, a copper-dependent enzyme, is responsible for the development of covalent bounds during collagen aggregation, in the activation process of which, it was recently shown that zinc has an important function[87]. However, there are some studies suggesting that zinc has no effect on the amount of collagen synthesis during wound healing[88]. Zinc given alone to the hypothyroid animals provides a small rise in tissue hydroxyproline level (not significant). These findings, consistent with the current recognition of collagen metabolism mentioned above, point out that zinc probably does not have a direct effect on the amount of collagen synthesis, but rather affects more directly the cross-linking of formed collagen.
Natori et al[89], in order to analyze the relationship between hypothyroidism and wound healing, induced a state of severe hypothyroidism in rats by surgically resecting their thyroid glands, and then assaying the levels of hydroxyproline and procollagen peptide (types I and III), which are the precursors of collagen (type IV: 7S) in wounds. The results of the assay of collagen and hydroxyproline in wounds indicate a significant decrease in type IV collagen and hydroxyproline in the surgical hypothyroidism rat group during the inflammatory phase and extending to the proliferative phase. These findings suggest that thyroid hormone is associated with the proliferation and secretion of fibroblasts in the process of wound healing. In the state of hypothyroidism, it appears that the suppression of thyroid hormone secretion causes a disturbance in the metabolical activation in tissues and the synthesis of collagen extending from the inflammatory phase to the proliferative phase.
In vitro, keratinocyte proliferation is retarded in T3-deficient medium relative to T3-replete medium[90]. In vivo, topical T3 stimulates epidermal proliferation[90,91] and topical triac, the mild TH analog, thickens skin[92]. The keratin genes encode the intermediate filaments, making about 30% of the protein of the epidermis. Some associations between the keratin genes and specific phases of skin growth have been made[93,94]. Fox example, K6a, K16, and K17 are associated with epidermal proliferation and wound repair. In K6a knockout mice[95], the absence of K6a results in diminished superficial wound healing, but no change in full thickness wound healing. When K16 is over-expressed in cultured human keratinocytes, proliferation is enhanced[96]. However, human K16 overexpressed in mice results in delayed wound healing in the transgenic animals[97]. Ex vivo investigation suggested that the K16 overexpression inhibits keratinocyte migration. Although TH stimulates expression of K6a both in vivo and in vitro[98], only negative TH response elements have been identified for the keratin genes associated with proliferation[99].
The following investigation has been undertaken to ascertain the need for TH in optimal wound healing and to clarify the effect of TH on the expression of wound healing-associated keratin genes[98]. The study showed that K6a, K16, and K17 gene expressions are decreased in T3 deficient keratinocytes. To determine whether the in vitro T3-mediated keratin gene expression could be reproduced in vivo, and whether that pattern is associated with significantly poorer wound healing, the investigators contrasted the impact of surgical hypothyroidism on wound healing and proliferation-associated keratin gene expression. The result showed that hypothyroid mice have indeed retarded wound healing, suggesting that T3 is necessary for optimal wound healing. T3 exerts its influence by stimulating the proliferation of cytokeratins 6a, 16, and 17.
CONCLUSION
It is well established that both diabetes mellitus and hypothyroidism are considered to have a negative impact on wound healing, not only as part of complex metabolic disorders, but also as clinical entities. Especially for the diabetic part, numerous studies have been undertaken, not only to provide explanations in regards to its metabolic pathophysiology, but also to provide evidence for its effect on wound integrity and dehiscence. As for the hypothyroid status and its exact role in wound healing, much more studies are still needed, so as to enlighten its close relationship with collagen, and therefore wound leakage or not, but also as clinical entities, as previous studies of impaired blood flow and its relationship to wound healing, using various suturing techniques[100], and of the evaluation of adhesion formation among different types of intraperitoneal catheters used in surgical practice[101], have proven.
