Abstract

Mammalian target of rapamycin inhibitors (mTOR-I) lacks nephrotoxicity, has antineoplastic effects, and reduces viral infections in kidney transplant recipients. Earlier studies reported a significant incidence of wound healing complications and lymphocele. This resulted in the uncomfortable willingness of transplant clinicians to use these agents in the immediate posttransplant period. As evidence and experience evolved over time, much useful information became available about the optimal use of these agents. Understandably, mTOR-I effects wound healing through their antiproliferative properties. However, there are a lot of other immunological and nonimmunological factors which can also contribute to wound healing complications. These risk factors include obesity, uremia, increasing age, diabetes, smoking, alcoholism, and protein-energy malnutrition. Except for age, the rest of all these risk factors are modifiable. At the same time, mycophenolic acid derivatives, steroids, and antithymocyte globulin (ATG) have also been implicated in wound healing complications. A lot has been learnt about the optimal dose of mTOR-I and their trough levels, its combinations with other immunosuppressive medications, and patients’ profile, enabling clinicians to use these agents appropriately for maximum benefits. Recent randomized control trials have further increased the confidence of clinicians to use these agents in immediate posttransplant periods.

1. Introduction

The combination of calcineurin inhibitor (CNI), mycophenolic acid derivatives, and steroids has reduced acute rejection by 12% in 1 year [1]. However, it has not been translated into long-term survival. Functional graft loss due to cardiovascular diseases, malignancies, and infections is still the main reason for graft loss [26]. Another potential reason for failure to achieve long-term survival benefits is the nephrotoxicity of the calcineurin inhibitors. mTOR-I has also been successfully used to minimize CNI in various randomized control trials. Lack of nephrotoxicity, antineoplastic, and antiviral effects make mTOR-I a good choice for transplant nephrologists to combine them with low-dose CNI [4, 6]. This maintains efficacy and reduces nephrotoxicity and viral infections in kidney transplant recipients. Unfortunately, mTOR-I’s earlier use resulted in more wound healing complications and lymphocele. Wound healing complications occurred in 5–47% of the patient on sirolimus (SRL) [7, 8] and 6–40% inpatients on everolimus (EVL) [9, 10]. Incidence of lymphocele is around 4–24% in SRL [11, 12] and 7–21% in EVL [13, 14]. Wound healing complications and lymphocele formation can cause significant morbidity, longer hospital stays, more radiological investigations, and resurgical exploration or radiological intervention. This leads to an increased overall cost of transplantation. Recent randomized control trials on SRL and EVL have increased our insight into using these agents to gain maximum benefits and minimize its adverse events, including wound healing complications. Accumulating evidence has shown that other immunosuppressive medications, when used concurrently with mTOR-I, have a synergistic effect on wound healing complications. Medications implicated in wound healing complications other than mTOR-I include mycophenolic acid derivatives [15, 16], steroids [17], and ATG [18]. Similarly, various nonimmunological risk factors can also lead to wound healing complications. Nonimmunological factors include obesity, uremia, increasing age, diabetes, smoking, and protein-energy malnutrition. NEVERWOUND study is a randomized control trial that describes wound healing complications as fluid collection, including hematoma and lymphocele, prolonged lymphatic drainage (lymphorrhea), wound dehiscence, wound infection, urine leak, and incisional hernia [18]. Randomized studies which looked at these wound healing complications were thoroughly reviewed. This review focuses on the mechanism of these wound healing complications, how mTOR-I affects wound healing, risk factors for wound healing, and the way forward for optimal use of mTOR-I in light of the evidence available from randomized control trials.

2. Mechanism of Action of mTOR-I

The primary mechanism of action of mTOR-I is the inhibition of the mammalian target of rapamycin. It is a regulatory protein kinase involved in lymphocyte proliferation. When mTOR-I enters the cell, it binds a cytoplasmic receptor called FKBP-12. This receptor blocks serine-threonine kinase known as mTOR. This kinase (mTOR) is a downstream regulator of phosphatidyl inositol 3-kinase (PI13K) and protein kinase B (Akt). Both PI13K and AKT are activated by interleukin (IL-12, I-L15), oncogenes, vascular endothelial growth factor (VEGF), and cytomegalovirus, which activates mTOR leading to the proliferation of lymphocytes, tumor cells, cytomegalovirus, and endothelial cells. mTOR-I blocks mTOR and leads to reduced lymphocytes, endothelial cells, tumor cells, and cytomegalovirus [19, 20]. These agents cause interaction inhibition among mTOR Complex 1 (mTORC1), mTOR Complex 2 (mTORC2), and PI3K. Currently, there are two mTOR-I available. Sirolimus (SRL) is a macrolide lactone produced by Streptomyces hygroscopicus and has a long half-life of 62 hours. EVL is one of the derivatives of SRL and has a similar structure but has a covalently attached 2-hydroxyethyl group at position 40, leading to improved bioavailability and reducing the half-life of 26 hours [21, 22].

3. mTOR-I and Mechanism of Impaired Wound Healing and Lymphocele Formation

The wound healing process consists of four phases: hemostasis, inflammation, proliferation, and tissue remodeling or resolution [23]. Hemostasis consists of vascular constriction, platelet aggregation, degranulation, and fibrin formation (thrombus). The phase of inflammation includes infiltration of neutrophils, lymphocytes, and monocyte and its differentiation to macrophages. The proliferation phase includes reepithelialization, angiogenesis, collagen synthesis, and extracellular matrix formation. The final remodeling phase includes collagen remodeling, vascular maturation, and regression [24].

T cells and various cytokines play an essential role in wound healing. Various studies showed that late infiltration and reduced T cells at wound sites are associated with wound healing problems [24]. Similarly, impaired angiogenesis and reduced fibroblast activity have been implicated in wound healing [25, 26]. Vascular endothelial growth factor (VEGF) and nitrous oxide are essential mediators for angiogenesis and collagen synthesis and play a critical role in wound healing [27, 28]. mTOR-I binds FK binding protein (FKBP) and acts on the mTOR. mTOR regulates the phosphoinositide 3-kinase/Akt pathway, which is stimulated by interleukin-2 and other cytokines [29]. It also affects cell cycle progression and angiogenesis. As a result, mTOR inhibition will cause inhibition of lymphocyte, endothelial, and fibroblast proliferation. mTOR-I also causes a reduction of VEGF and NO [30]. Inhibition of endothelial and fibroblast cells by mTOR-I leads to impaired angiogenesis and fibroblastic activity [25, 31].

There are various animal studies on the pathophysiology of wound healing. It has been shown that hypoxia increases DNA synthesis and proliferative effects of platelet-derived growth factor (PDGF) and fibroblast growth factor (FGF) in rat and human smooth muscle and endothelial cells. This effect is dependent on mTOR activation downstream enzyme, phosphatidyl inositol 3-kinase. Rapamycin has been shown in rats to impair wound healing by blocking this enzyme [31]. Intraepithelial lymphocytes, γδT cells in the skin, help in wound healing, and depletion of these cells with rapamycin results in delayed wound healing in rats [32]. EVL has been shown to cause a reduction in hydroxyproline and collagen deposition in wounds resulting in reduced breaking strength and bursting pressure of ileal and colonic anastomosis in the rat model [33]. Similar effects were seen in the abdominal wound in rats in another study [34]. Bladder healing was assessed in rats in another study. It showed that eosinophil and neutrophil infiltration and myofibroblast proliferation were significantly lower in the bladder, fascia, and dermis of the rats who received rapamycin compared to the control group. Mean microvessel density and the percentage of cells expressing vascular endothelial growth factors in the bladder, fascia, and dermis were also significantly lower among rapamycin [35]. In a study done in pigs to assess ureteric anastomosis, the tensile strength and the hydroxyproline levels in the ureter and fascia were lower in the rapamycin-treated group [36]. Yet, in another study in pigs, although rapamycin derivatives prevented the development of bronchiolitis obliterans, it impaired the healing of bronchial anastomosis [37]. These animal studies suggest that mTOR-I impairs the ability of wound healing. Summaries of all these studies are included in Table 1.

Lymphocele l is a pseudocyst with lymph inside with an outside hard, fibrous capsule. It is usually adjacent to the graft [38, 39]. mTOR-I has been shown to have antilymphoangiogenic effects during surgical wound healing both in vitro and in vivo. It was demonstrated that VEGF-C plays an essential role in lymphangiogenesis. EVL and SRL inhibit this intracellular mediator of lymphangiogenesis [40].

4. Is It mTOR-I Only?

Besides mTOR-I, various studies conducted in animal and clinical settings have implicated other immunological medications and nonimmunological factors in wound healing complications. Therefore, it is important to investigate these factors and do a fair analysis of all these factors to reach the root cause analysis of wound healing events. Figure 1 shows all the risk factors of wound healing complications.

These immunological and nonimmunological factors which result in impaired wound healing are as follows.

4.1. Mycophenolic Acid Derivatives

Mycophenolic acid derivatives include mycophenolate mofetil (MMF) and its metabolites, mycophenolic acid (MPA). It is a highly selective, noncompetitive, and reversible inhibitor of the inosine monophosphate dehydrogenase. It is the rate-limiting enzyme for de novo biosynthesis of guanosine nucleotides [41]. Guanosine nucleotides are important for DNA replication and RNA and protein synthesis. Various experimental studies have shown that MMF/MPA affects various body cells and collagen synthesis. These agents inhibit the proliferation of both T and B lymphocytes. MMF has been shown to cause downregulation of cytoskeleton proteins vinculin, actin, and tubulin in fibroblasts exposed to pharmacological doses of MPA. Skin biopsies of patients treated with MPA expressed less vinculin, actin, and tubulin than control biopsies, which could be a potential explanation for impaired wound healing [42]. Willem et al. showed in the rodent model that MMF may negatively affect the abdominal wall wound healing but had no effect on colonic anastomosis [43]. The bladder wound of rats treated with tacrolimus (TAC) and MMF has more immature collagen (type III) as compared to the control group, which has mature collagen (type I) [44]. MMF may cause inhibition of fibroblast by depletion of guanosine. Human tenon fibroblasts were cultured with various concentrations of MMF with and without guanosine, and it was shown that growth of tenon fibroblast was inhibited in a concentration-dependent way. These effects were reversed with guanosine [45]. MMF has been shown to inhibit the growth of nonimmune cells, including tubular cells [46], mesangial cells [47, 48], and myointerestitial fibroblasts [49] in the kidneys and has a potential role in proliferative glomerulonephritides and slowing down interstitial fibrosis and tubular atrophy in kidney transplant patients. In clinical studies, MMF has been implicated in causing wound healing complications in 16.6% of kidney transplant recipients [50]. In a retrospective analysis, more lymphoceles (OR = 2.6;  = 0.03), fluid drainage (17 vs. 5 interventions), and sclerotherapies (8 vs. 0) were observed in MMF group as compared to azathioprine (AZA) [51]. MMF has been implicated along with SRL in wound healing complications in several randomized control trials. However, it is difficult to assess the individual agent’s impact on wound healing because of its use in combination with mTOR-I. In a prospective randomized control trial, hernial eventration/wound evisceration was 7/71 in the SRL-MMF group compared to 0/71 in the ciclosporin (CsA)-MMF group [52]. In the ORION study, SRL-MMF had significantly higher wound healing complications than the SRL-TAC elimination group (23% vs.16.4%,  < 0.05). Similarly, the incidence of lymphocele was also significantly higher in the SRL-MMF group [16]. In the SYMPHONY trial, 17% of patients had delayed wound healing in low-dose SRL and MMF groups, significantly higher than other groups ( value = 0.006). The incidence of lymphocele was 15.8% which was also significantly higher in the SRL-MMF group when compared to other groups ( value <0.001) [1]. In the TRANSFORM trial, wound healing complications were 19.8% in EVL compared to 16.2% in the MPA group, with relative risk between the two groups being 1.22 (1.01 to 1.47) [53]. In a meta-analysis of randomized control trials, the incidence of wound healing complications (OR 3.00, CI 1.61–5.59) and lymphocele (OR 2.13, CI 1.57–2.90) were significantly higher in mTOR-I and MMF as compared to mTOR-I and calcineurin inhibitor [54]. Table 2 shows the summary of experimental studies and studies conducted in a clinical setting on mycophenolic derivatives on wound healing complications.

4.2. Steroids

Corticosteroids cause wound healing complications by a variety of mechanisms. Corticosteroids reduce inflammation, fibroblast proliferation, collagen synthesis, angiogenesis, and reepithelialization [55]. In vitro studies conducted in an animal model have shown that steroids cause impaired wound healing through various mechanisms. Methyl prednisolone treatment has been shown to decrease transforming growth factor beta (TGF-beta) and insulin-like growth factor I (IGF-I) in the wound fluid and hydroxyproline content in the tissue ( < 0.05) in rats’ model [56]. In another study, administration of hydrocortisone in mice reduced the skin wound healing resistance during the first postoperative week [57]. Steroids have also been implicated as risk factors in a retrospective analysis of abdominal wounds complicated by dehiscence in the general population [58].

In several randomized control trials, steroids avoidance led to fewer wound complications. Sandrini et al. showed that overall wound complications were significantly lower in the off-steroids group than those on steroids (18.8% vs. 45.6%, respectively,  < 0.0004). Similarly, incidence of lymphocele (5.0% vs. 32.3%,  < 0.0001) and dehiscence (0% vs. 10.3%,  < 0.009) were significantly lower in steroids avoidance group [17]. The addition of steroids to SRL increases 4.2-fold the risk for wound complications [17]. In another to randomized control trial, the incidence of lymphocele was higher in steroid-free regimens than low-dose steroids (1.5% vs. 5.9%), but it was not statistically significant [59]. Roger et al. compared 109 patients treated with a corticosteroid avoidance regimen with a historical control group (n = 72) that received CsA, MMF, and steroids. The corticosteroids avoidance group has lower incidence of wound healing complications (13.7% vs. 28%,  = 0.03) and lymphoceles (5.5% vs. 16%,  = 0.02) than the control group [60].

Steroids use in humans has shown that high-dose corticosteroid administration for <10 days has no clinically significant effect on wound healing. In patients taking chronic corticosteroids for at least 30 days before surgery, their rates of wound complications may be increased 2 to 5 times compared with those not taking corticosteroids [61].

4.3. Antithymocyte Globulin (ATG)

Various studies have also implicated ATG in wound healing problems. Benavides et al. [62] studied wound healing complications in patients receiving rabbit antithymocyte globulin (rATG) induction for a maximum of two weeks postoperatively. Patients receiving ATG: 39.1% patients have significant wound healing complications compared to 26.0% basiliximab induction ( = 0.025). Pourmand et al. found a significant relationship between ATG therapy and wound complications ( = 0.034) [63]. These findings were confirmed in the NEVERWOUND study, which reported an increased risk of wound healing >60% while using ATG induction [18].

4.4. Obesity

Obesity is another important risk factor accounting for wound complications. Around 34.5% of kidney transplant recipients have a body mass index (BMI) greater than 30 [64]. Wound infection and dehiscence are more when BMI is  > 30 [65]. The risk of wound healing complications goes up with the severity of obesity. Andrade et al. assessed the effect of weight on wound complications in underweight (BMI < 20 kg/m2), normal weight (20 ≤ BMI < 25), overweight (25 ≤ BMI < 30), class I obese (30 ≤ BMI < 35), class II obese (35 ≤ BMI < 40), and class III obese (BMI ≥ 40). There was a significantly increased risk of wound complications by 1.9-fold for every 5 points increase in BMI ( < 0.001), and wound complications were observed 17.5, 29.0, 45.0, and 60% with BMIs of 30, 35, 40, and 45, respectively, in each group [64]. In an analysis of data of 869 kidney transplant recipients, Lynch et al. [64] reported a graded increase in the frequency of wound infection from 8.5% among those with BMI 20–25 to 40% among those with BMI > 40 [66]. In another study conducted on SRL to assess risk factors for wound healing, obesity was an important contributor. The authors compared SRL-MMF patients with complications within three months of transplantation with SRL-MMF patients without complications and matched renal transplant recipients receiving TAC-MMF. Obesity (BMI ≥ 30 kg/m2) was significantly associated with wound problems. The mean BMI of SRL cases with complications was 29.9 kg/m2 compared to 25.4 kg/m2 for SRL patients without complications ( = 0.047). Seventy-one percent of obese SRL patients experienced complications compared with 24.3% ( = 0.025) of nonobese SRL patients [67]. Another retrospective analysis assessed risk factors for wound healing complications in patients receiving de novo SRL, low-dose CsA, and corticosteroid. Multivariate analysis showed that body mass index (BMI) > 26 (odds ratio 2.498,  = 0.027) was a significant risk factor for wound healing complications in patients taking SRL. The risk was even higher with BMI >30 (odds ratio 3.738,  = 0.007) [68]. In a prospective randomized trial using high-dose SRL (15–20 ng/mL), wound healing complications increased across all BMI, except patients with a BMI less than or equal to 24 kg/m2. In the second phase of the same trial, after excluding BMI >32 kg/m2 and using a low level of SRL (10 to 15 ng/mL), the complication rate in patients with BMI 28.1 to 32.0 was 33% in the SRL group as compared with 78% in phase I of the same trial [8]. Recently, TRANSFORM study excluded patients with BMI greater than 35. The mean BMI of EVL and MPA arm was 25.6 between the two groups. EVL targeting a trough concentration of 3–8 ng/ml avoided the increased rates of lymphocele seen previously, though wound healing events/complications were still slightly higher as compared to MPA (19.8% vs.16.2%) [53]. Later in-depth analysis of TRANSFORM data by Tedesco et al. found no significant association when wound healing complications were compared with mean EVL concentration during the periods from day 4 to week 4, day 4 to month 2, and day 4 to month 12 [69].

4.5. Uremia and Renal Dysfunction

Unlike other surgeries done on patients with normal renal functions, kidney transplant patients have preceding uremia, which has a negative impact on wound healing. There are over 100 uremic toxins in patients with end-stage renal disease [70]. It has been shown that uremia impairs fibroblast proliferation and hydroxyproline level [7174]. Other factors that make a chronic kidney disease patient prone to impaired wound healing include uremic itch, calcemic uremic arteriopathy, malnutrition, edema, and propensity for infections [75]. Cadaveric transplantation being an unplanned event, it is always difficult to ensure adequate dialysis in the preceding past. Live transplantation being a preplanned event always provides the opportunity to provide adequate dialysis in the preceding month.

4.6. Age

Increasingly a greater number of elderly populations is being transplanted nowadays. Age-related skin changes affect all stages of wound healing [76]. Platelets’ adherence to injured endothelium and release of various cytokines (PDGF, TGF) is enhanced in the elderly population [77]. As a result, inflammatory cells are recruited to the wound healing site. There is early infiltration of neutrophils but delayed infiltration of monocytes-macrophages compared to the young population. Macrophages played an important role in wound healing, and their late infiltration may be one reason for impaired wound healing in this population [78]. In rat models, angiogenesis is reduced in aged rats [79] and has reduced macrophage content [80]. Wound remodeling may be impaired due to reduced collagen turnover and increased fibroblast senescence [76]. There is a paucity of aging data and its effect on impaired wound healing in the kidney transplant population.

4.7. Diabetes

There is paucity of data on the impact of diabetes on wound healing complications in kidney transplantation. In diabetics, there is a delayed response to injury due to impaired functioning of the leukocytes and fibroblast and reduced insulin in the face of hyperglycemia [81]. Experimental studies in the acute diabetic pig model have shown that reduced insulin-like growth factors rather than hyperglycemia resulted in impaired wound healing [82]. Another in vitro study on mice showed that diabetic fibroblasts show selective impairments in cellular responses needed for tissue repair, impaired VEGF production, and impaired response to hypoxia [83]. Osmotic diuresis and catabolism associated with uncontrolled diabetes may also impair wound healing [84, 85]. Keeping these facts in mind, it is crucial to have meticulous diabetes control pre, peri, and postoperative time for better wound healing.

4.8. Alcohol

Both acute alcohol intoxication and chronic alcoholism impaired wound healing. Critical alcohol consumption reduces proinflammatory cytokines in the face of inflammatory challenges. It also reduces the infiltration of neutrophils and their phagocytic function at the site of inflammation. This impairs the initial inflammatory response and increases the risk of infection [86, 87]. Alcohol also affects the proliferative phase of wound healing. It has been shown in experimental studies that epithelial healing, new blood vessel formation, collagen production, and wound closure are all reduced even with a single dose of alcohol [88, 89]. Single ethanol exposure in both in vitro and in vivo settings before the injury can cause a significant decrease in wound breaking strength due to impaired fibroblast function and collagen production [90].

4.9. Smoking

Smoking has been shown to affect the migration of white blood cells to the site of inflammation. There is a reduced number of monocytes and macrophages at the wound sites, while the ability of neutrophils to kill bacteria is also impaired. Smoking affects lymphocytes and natural killer cells’ functional ability at the site of inflammation [91, 92]. Smoking impairs epithelization and reduces the ability of fibroblasts to migrate and proliferate, resulting in an impaired proliferative phase of wound healing [91]. Nicotine causes peripheral vasoconstriction and increases the blood’s viscosity through reduced fibrinolytic activity and increased platelet aggregations. Carbon monoxide in smoker binds hemoglobin more efficiently and reduces oxygen saturation. These factors result in reduced oxygen and blood supply leading to impaired wound healing [91, 93]. It has been shown that quitting smoking improves wound healing and reduces infection [94]. It is important to stop smoking six weeks before surgery, including transplantation [95].

4.10. Protein-Energy Malnutrition

Malnutrition of dialysis patients is multifactorial. Inadequate protein and calorie intake, loss of appetite, inflammation, loss of residual renal function, inadequate dialysis, insulin resistance, and superimposed comorbid conditions are the various causes for malnutrition [9698]. The prevalence of malnutrition in dialysis patients has been reported between 18% and 56% [99, 100]. The recommended dietary protein intake for clinically stable maintenance hemodialysis patients is 1.2 g/kg body weight/day. At least 50% of the dietary protein should be of high biological value. Dietary protein intake for patients on peritoneal dialysis who are clinically stable is 1.2 to 1.3 g/kg body weight/day [101].

Protein is essential for wound healing, capillary formation, fibroblast proliferation, proteoglycan, and collagen synthesis. It is also necessary for optimal phagocytic activities of leukocytes. As a result, protein-energy malnutrition (PEM) results in impaired wound healing and reduced phagocytic infection [102]. Collagen is the major protein component of connective tissue and is composed primarily of glycine, proline, and hydroxyproline. Collagen synthesis requires hydroxylation of lysine and proline and cofactors such as ferrous iron and vitamin C. Impaired wound healing results from deficiencies in any of these cofactors [103]. KDOQI guidelines also suggested that nPCR should be between 1.0 and 1.2 g/kg/d, and serum albumin should be equal to or greater than 4.0 g/dL [104]. It is important to achieve these parameters before any surgical intervention to avoid wound healing complications.

5. Review of Randomized Control Trials on mTOR-I

In NEVERWOUND study, a randomized control trial, wound healing complications included fluid collection, including hematoma and lymphocele, prolonged lymphatic drainage (lymphorrhea), wound dehiscence, wound infection, urine leak, and incisional hernia [18]. Ueno et al. [105] described wound healing complications such as wound dehiscence, wound infection, incisional hernia, lymphorrhea, fluid collections, peri graft hematoma, and urine leak. All fluid collections were diagnosed by either ultrasound or computed tomography (CT). We reviewed all randomized control trials which looked at these wound healing complications.

With the advent of SRL in 1972 [106], SRL was being evaluated since earlier 1996 in randomized control trials [107]. Unfortunately, not all trials looked at wound healing complications or lymphocele formations as a primary or secondary outcome. Some of the initial randomized control trials reported more wound infections, wound healing complications, and lymphocele formation [108111]. Earlier case series and retrospective data also point to wound healing complications and lymphocele formation [52, 68, 112]. Various randomized control trials from 1999 to 2017 over the last 2 decades looked at either wound healing complications or lymphocele formations [1, 7, 8, 11, 15, 16, 108111, 113127]. Most of the earlier trials reported a positive association of SRL with either wound healing complications or lymphocele formations. However, most of these trials used a loading dose ranging from 6 mg to 30 mg and maintained very high trough level of 10–30 ng/mL [1, 5, 7, 12, 108111, 113, 115, 116, 118125]. In RCT by Kandasamy et al. [116], wound healing complications were significantly reduced when loading dose was avoided in the second phase of the trial. Various randomized control trials which compared low-dose SRL with high-dose SRL reported a smaller number of wound healing complications in low-dose SRL [6, 8, 9, 1214, 110, 113, 116, 117] numerically. Vitko et al. compared low-dose SRL (1.5 mg first dose followed by 0.5 mg once a day vs. 6 mg first dose followed by 2 mg per day) and found a significantly lower incidence of lymphocele ( = 0.022) [11]. In a recent RCT, SRL was used with extended-release tacrolimus (ER-TAC). SRL-ER TAC was compared with MMF-TAC. SRL level was kept at 3–5 ng/mL, and no difference was found between the two groups in terms of wound healing and lymphocele formation [127]. Table 3 shows summaries of randomized control trials which looked into wound healing complications and lymphocele formations [1, 7, 8, 11, 12, 16, 108127].

Since earlier 2000 multiple randomized control trials were conducted on EVL [9, 10, 13, 14, 18, 69, 105, 128139] as shown in table [4]. Various studies compared low-dose EVL (1.5 mg/day) with high-dose EVL (3 mg per day). Wound healing complications were numerically higher in the high-dose EVL group [2, 3, 8] but were not statistically significant [129, 130, 135]. In 2013, Cooper et al. [134] showed that the higher blood level of EVR (>8 ng/mL) was also associated with increased risk (HR, 1.69; 95% CI, 1.20–2.38;  = 0.002) of wound healing complications. Therefore, the initial dose of 1.5 mg seems safer and more reasonable than 3 mg. Most studies used level between 3–8 ng/mL [10, 14, 69, 131133, 135, 138, 139] without any significant impact on wound healing.

It is important to consider the patient’s weight, induction therapy, and combinations of other immunosuppressive medications with mTOR-I. Obesity is an important risk factor that can potentially augment mTOR-I wound healing complications. BMI of >26 has significantly been associated with wound healing complications in patients taking SRL (odds ratio 2.498,  = 0.027). The risk is even larger if BMI >30 (odds ratio 3.738,  = 0.007) [68]. The risk increases by 1.9-fold for every 5 points of BMI across a range of BMI from 20 to greater than 40 BMI in kidney transplant recipients [64]. In a systemic approach to minimize wound healing complications, multivariate analysis of recipients treated with de novo SRL showed that a BMI more than 30 to 32 kg/m2 was the most significant variable related to delayed wound healing (OR 3.01, 0.02) and the need to repair a transplant wound surgically (OR 8.05,  = 0.0001) [140]. Kandasamy et al., in the second phase of their trial, showed that exclusion of BMI >32 significantly reduced wound healing complications in SRL groups [116]. In the NEVERWOUND study, BMI of <25 kg/m2 was identified as a predictor of WHC-free status at 12 months [18]. Besides, consideration of obesity induction with ATG and subsequent use of mTOR-I also increase wound healing complications [18, 62, 63]. Another important fact is that using SRL and MMF may have a synergetic effect on wound healing complications [1, 16, 53, 54]. However, few studies on EVL combined with MPS or MMF did not show this synergism. de Fijter et al. [9] compared EVL-MPS with EVL-CNI and found no difference in wound healing complications. Similarly, in the CALLISTO study, no difference was observed in the incidence or severity of wound healing complications in kidney transplant recipients receiving either MMF or EVR as de novo immunosuppressive drug [10]. Nashan et al. [137] also compared EVL-MPS with EVL-CNI and found no difference in wound healing complications in BMI category ≤25 percentile (EVR, 0.9 vs. CNI, 0.8%;  = 0.846) and in BMI category of >25–≤50 percentiles (2.6 vs. 1.1%,  = 0.271). However, wound healing complications were significantly higher in >50–≤75 categories (2.0 vs. 0.6%,  = 0.049). Majorities of the earlier de novo studies on SRL showed a positive association between wound healing complications and lymphoceles [1, 5, 7, 12, 108111, 113, 115, 116, 118125]. In contrary to most studies on EVL, which kept trough level 3–8 ng/mL [10, 14, 69, 131133, 135, 138, 139], we did not find significance on wound healing or lymphocele formations. These differences could be due to a shorter half-life or higher bioavailability of EVL, or it could be due to loading doses and a very high trough level of 10–30 ng/mL used in the case of SRL. Avoidance of loading dose [116, 127] and use of low-dose SRL have been shown to reduce wound healing complications [115, 127]. No randomized control trial has made a head-to-head comparison between SRL and EVL. An open label RCT is going at the moment, which will compare three arms (EVL-TAC, SRL-TAC, and MMF-TAC). The study will be completed by the end of 2021 and will look into safety profile including wound healing complications between SRL and EVR [141].

The thought that delayed administration of mTOR-I may reduce wound complications and delayed graft function was evaluated in a few RCTs. Albano et al. [132] were the first to assess this strategy in 2009. They compared immediate EVL from day 1 with delayed EVL from week five and found no difference in delayed graft function and wound healing complications. Similarly, the CALLISTO study [10] did not find the difference in wound healing complications and delayed graft functions between immediate or delayed use of EVL. These findings were reinforced in 2020 by Manzia et al. [18], who also found no difference in wound healing complications between immediate or delayed use of EVL.

A couple of the recent studies on mTOR-I further increased the insight into using these agents in de novo transplantation. Schäffer et al. [28] in 2018 used ER-TAC with SRL and compared it with ER-TAC and MMF. This group kept trough level 3–5 ng/mL. Wound healing and risk of lymphocele were not significantly different between the two groups. In the TRANSFORM study, the use of EVL aiming for a trough concentration of 3–8 ng/ml avoided the increased rates of lymphocele though the wound healing complication were slightly higher [53]. Later on, an in-depth analysis of TRANSFORM data was performed by Tedesco et al. They compared wound healing complications with mean EVL concentration during the periods from day 4 to week 4, day 4 to month 2, and day 4 to month 12. They found no significant association of the mean concentration of EVL with wound healing complications [69]. The ATHENA randomized control trial was published in 2019 and compared three arms consisting of EVR/TAC, EVR/CsA, and MPA/TAC and found no difference in wound healing complications among the three groups [139]. NEVERWOUND study was another RCT published in 2020 which compared immediate use of EVL-CsA-Pred with delayed use and found no difference in wound healing complications and lymphoceles between the two arms [18].

6. Emergency or Elective Surgery in Patients on mTOR-I

Clear guidelines for continuing mTOR-I in the wake of any emergency or elective surgery after kidney transplantation are lacking. This is simply due to the lack of randomized control trials. Most of the data available are case reports, retrospective studies, or prospective case series. SRL and obesity have been risking factors for hernia recurrence in liver transplant patients [142]. Different approaches have been reported in the literature for patients undergoing surgery. Scheuerlein et al. switched SRL to calcineurin inhibitors in patients undergoing laparoscopic incisional hernia repair after solid-organ transplantation [143]. On the other side, immunosuppression, including mTOR-I, was maintained postoperatively in patients undergoing laparoscopic incisional hernia repair after solid-organ transplantation and aortic valve replacement in kidney transplant patients [144, 145]. Hebel et al., in their retrospective analysis of 13 pediatric cardiac patients who underwent surgery, found that only 1/13 (7.7%) has wound complications [146]. Schwarz et al. studied six liver transplant recipients who underwent nine major abdominal or thoracic surgical procedures without mTOR-I discontinuation or specific dosage adjustment. They found no evisceration, incisional surgical site infection, or lymphocele [147]. However, one has to bear in mind that patients in this retrospective analysis did not include obese patients and the overall mean SRL trough concentration was 4.8 ng/mL. Campistol et al. made recommendations for minor surgery, major surgery, and emergency surgery in patients on SRL [148]. They kept into consideration nonmodifiable risk factors (age, African American) and modifiable risk factors (obesity >26 kg/m2, use of steroids, and use of ATG and anticoagulant) while deciding mTOR-I in the event of surgery. They suggested that no change is required in minor surgery or laparoscopic surgery without risk factors. In major surgery, including those who required chemotherapy, the group recommended holding SRL 5–10 days before the operation and restarting 1–3 months later. They suggested stopping mTOR-I immediately and restarting five days later in emergency surgery.

In the absence of robust data, it is challenging to advise about the withdrawal of mTOR-I in the wake of surgery. While planning for elective surgeries, it is crucial to look for risk factors of wound healing and reduce the dose of mTOR-I to ensure lesser chances of wound healing complications and prevent rejection at the same time. In emergency surgery in patients with risk factors for wound healing or postoperative wound complications, a decision of withdrawal may be considered.

7. Way Forward for the Use of mTOR-I

Minimization of CNI while using mTOR inhibitor provides synergistic immunosuppressive effects and reduces nephrotoxicity. mTOR-I has an antiviral and antitumor effect [4, 6]. The incidence of cytomegalovirus and BK virus infections in EVL is significantly lower when used with minimized CNI and steroids compared to the combination of MMF [53]. Therefore, it is important to use these agents wisely to achieve their maximum potential benefits and to keep its side effects minimum possible level.

Since the introduction of SRL in 1972 and multiple randomized control trials on EVL since the start of 2000 with the availability of significant amount of evidence, much has been known about these agents. Events of cadaveric kidney transplantation are not planned events and clinicians do not have enough time to optimize nonimmunological risk factors for wound healing complications. In contrast to cadaveric transplantation, live kidney transplantation gives clinicians a chance to optimize the kidney transplant recipients before surgery to ensure a better outcome. Therefore, live kidney transplant recipients should be optimized before the planned surgery. Nonimmunological risk factors should be identified and discussed with the recipients to avoid wound healing complications and optimize recipient and graft survival. Those selected to be a candidate for kidney transplantation should be thoroughly evaluated for wound healing risks. Those with BMI greater than 30 kg/m2 should be encouraged to lose weight. BMI in lower ranges has been significantly associated with reduced wound healing complications [64]. Detailed smoking history should be obtained, and it is essential to stop smoking six weeks before kidney transplantation [95]. Similarly, alcoholics should be encouraged to quit drinking to reduce wound healing complications. Potential live recipients with a history of diabetes should have optimal diabetes control before transplant to minimize the perceived complications [84]. Patients undergoing renal replacement therapy must have adequate dialysis in the preceding months before transplantation to minimize the effect of uremia on wound healing complications [75]. Clinical evaluation should be performed along with an estimation of protein catabolic rate to identify malnourished patients. These patients should be treated with dietitians to improve their nutritional status [103, 104]. These patients must have nPCR between 1.0 and 1.2 g/kg/d. The serum albumin should be equal to or greater than 4.0 g/dL [104]. Figure 2 shows the way forward to minimize wound healing complications.

Previous higher wound healing complications were attributed to higher loading doses of SRL (ranging from 6 mg to 30 mg) along with higher trough level of 10–30 ng/mL [1, 5, 7, 12, 108111, 113, 115, 116, 118125]. Avoidance of loading has been shown to reduce wound healing complications significantly [116]. Use of low-dose SRL has been shown to reduce incidence of lymphocele significantly when compared with higher dose [11]. Therefore, we suggest to avoid loading dose and keep trough level between 5–10 ng/mL [149, 150] to minimize adverse events and wound healing complications. If one contemplates using SRL with TAC-ER, SRL level can be kept even low at 3–5 ng/mL [127]. Low-dose EVL when compared with high-dose EVL (1.5 mg/day vs. 3 mg/day) led to numerically a smaller number of wound healing complication [2, 3, 8]. Higher blood level of EVL (>8 ng/mL) has been shown with increased risk (HR, 1.69; 95% CI, 1.20–2.38;  = 0.002) [134]. Therefore, we suggest EVL level to be kept between 3–8 ng/mL [10, 14, 69, 131133, 135, 138, 139]. ATG induction in patients with mTOR-I should be avoided to reduce wound healing complications [18, 62, 63]. Since most candidates for mTOR-I are of low immunological risk, induction with basiliximab will be a reasonable option. The combination of SRL with MMF has synergetic effects on wound healing, as reported by the SYMPHONY trial. Therefore, mTOR-I, especially SRL, should be avoided with MMF or MPA [1]. Steroid’s dose should be minimized to reduce the risk of wound healing [151].

Planning surgery also plays a vital role in preventing wound healing complications. Surgeons must be aware of the potential use of mTOR-I in the posttransplant period. Tiong et al. analyzed a systemic approach to minimize wound healing complications in de novo SRL [140]. Their approach included patient selection (body mass index) [BMI] < 32 kg/m2, the use of closed suction drains, modifications of surgical technique, and avoidance of a loading dose of SRL. Surgical wound closure was performed via a multilayer closure approach using nonabsorbable interrupted sutures in the fascia. The skin closure was performed through interrupted nonabsorbable monofilament sutures. The drains were removed after 2–3 weeks or when drain volume was less than 50 ml for two days. The sutures were usually left for three weeks in the majority of patients. Using this approach, a significant reduction was found in cumulative wound complications (7.8% vs. 19.6%,  < 0.007) and lymphocele (22.3% vs. 47.1%,  < 0.0001) as compared to the historical cohort [140], leaving staples for 3–4 weeks and draining till drainage minimizes wound dehiscence and collection [148]. Ligation of lymphatic meticulously, peritoneal fenestration, and minimizing dissection will reduce lymphocele formation [151].

8. Conclusion

mTOR-I can be used immediately after kidney transplantation. Loading doses and high trough levels for SRL lead to more wound healing complications and should be avoided. EVL trough level of 3–8 ng/mL maintains its efficacy and avoids most adverse events, including wound healing complications. Induction with ATG may be avoided. mTOR-I should be used with low-dose CNI, and its combination with mycophenolic acid derivatives should be avoided. Patients with BMI ≥30 kg/m2 should be encouraged to lose weight before surgery. Adequate nutrition, cessation of smoking and alcoholism, controlling diabetes, and adequate dialysis before transplant surgery can minimize wound healing complications. Surgical wound closure in multilayers using interrupted suture, meticulous ligation of lymphatics, leaving staples for 3–4 weeks, and using close suction drain decrease wound healing complications [152] (Table 4).

Abbreviations

Pred:Prednisolone
mTOR-I:Mammalian target of rapamycin inhibitors
ATG:Antithymocyte globulin
BMI:Body mass index
CNI:Calcineurin inhibitor
CsA:Cyclosporine
ER-TAC:Extended-release tacrolimus
MMF:Mycophenolate Mofetil
MPA:Mycophenolic acid
SRL:Sirolimus
TAC:Tacrolimus.

Data Availability

This is a review article with no data.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The corresponding author acknowledges all the coauthors for their valuable input and drafting of this manuscript.