Safety of Deep Sedation in Patients Undergoing Full- Thickness Skin Graft Harvesting and Skin Graft Reconstruction for Limb Salvage. An Outcome Analysis

Abstract

Background: Studies on obtaining donor skin graft using intravenous sedation for patients undergoing major foot surgeries in the same operating room visit have not previously been reported. The objective of this retrospective study is to demonstrate that intravenous sedation in this setting is both adequate and safe in patients undergoing skin graft reconstruction of the lower extremities in which donor skin graft is harvested from the same patient in one operating room visit. Methods: Medical records of 79 patients who underwent skin graft reconstruction of the lower extremities by one surgeon at the Yale New Haven Health System between November 1, 2008, and July 31, 2014, were reviewed. The patients' demographic characteristics, American Society of Anesthesiologists class, comorbid conditions, intraoperative analgesic administration, estimated blood loss, total operating room time, total postanesthesia care unit time, and postoperative complications within the first 72 hours were reviewed. Results: This study found minimal blood loss and no postoperative complications, defined as any pulmonary or cardiac events, bleeding, admission to the intensive care unit, or requirement for invasive monitoring, in patients who underwent major foot surgery in conjunction with full-thickness skin graft. Conclusions: We propose that given the short duration and peripheral nature of the procedures, patients can safely undergo skin graft donor harvesting and skin graft reconstruction procedures with intravenous sedation regardless of American Society of Anesthesiologists class in one operating room visit.

Obtaining donor skin graft with intravenous sedation for patients undergoing major foot surgery in the same operating room visit has not been reported previously. Donor skin grafts have been taken in the past with the help of intravenous opioids and topical application of lidocaine,[1] skin infiltration with bupivacaine with epinephrine at the donor site,[2] and blocks such as continuous fascia iliaca compartment block with bupivacaine 0.2% over 48 hours, and all have been described to be effective.[3] By performing this retrospective study, we aimed to demonstrate that sedation can be safe and sufficient for patients undergoing skin graft reconstruction of the lower extremities with donor skin graft harvesting being from the same patient in the same operating room visit.

A universally used method by anesthesia providers for evaluation of the physical status of patients preoperatively is the American Society of Anesthesiologists (ASA) classification system.[4] It has also been used to gauge perioperative risk,[5] although studies have questioned its validity given lack of consistency of application across anesthetists.[6] There have been mixed findings of the efficacy of the use of this classification system. In a multivariate logistic regression analysis, Wolters et al[7] found that the risk of complications in patients was reflected by ASA III (risk odds ratio, 2.2) and ASA IV (risk odds ratio, 4.2). A previous finding showed that ASA III and IV patients undergoing limb-salvage procedures are a unique exception to the risks commonly found in these classes of patients, making them suitable for monitored intravenous sedation and local anesthesia rather than general anesthesia with controlled airway, as has traditionally been practiced.[8] To our knowledge, the use of sedation for foot surgery for the skin graft harvesting donor procedure and skin reconstruction procedure safely in the same operating room visit has not been reported previously.

Methods

After institutional review board approval was obtained, the anesthetic records and operative reports of 79 patients who underwent skin graft procedures with monitored total intravenous sedation and foot blocks using ankle, Mayo, or digital block were reviewed for complications occurring within 72 hours postoperatively. All of the surgeries were performed by a single surgeon (P.A.B.) between November 1, 2008, and July 31, 2014. The ankle blockade consisted of administering local anesthetic to the ankle, thereby hindering conduction of the sural, deep peroneal, superficial peroneal, tibial, and saphenous nerves. The digital block was performed from the dorsal aspect such that two separate sites at the base of each digit were chosen, one on the lateral aspect and the other on the medial aspect of the toe, blocking conduction of the lateral dorsal, medial dorsal, and corresponding lateral and medial plantar proper digital nerves. The Mayo block consisted of blocking neural conduction to the entire portion of the forefoot from the level of the base of the involved metatarsal(s). These local anesthetic blockades were administered with 0.5% plain bupivacaine. Sedation was obtained via subanesthetic infusion as an adjunct to the ankle and foot blocks by intravenously administering midazolam 0 to 2 mg, fentanyl 0 to 100 mg, and propofol 0 to 5 mg/kg per hour. No anesthetic block technique was used at the donor site.

The procedures conducted with the skin graft reconstruction included lower-limb amputations and debridement of bone and soft tissue. Of specific interest for the present study was any unexpected adverse event, including any cardiac or pulmonary complications such as myocardial infarction, pulmonary edema, stroke, any bleeding, or requirement for invasive monitoring or admission to the intensive care unit within the initial 72 hours postoperatively. The donor skin graft was obtained from the upper part of the thigh in the distribution of the femoral nerve and the lateral femoral nerve.

Results

Of the 79 patients included in the analyses of this study, one (1.32%) was ASA I, four (5.26%) were ASA II, 62 (81.58%) were ASA III, and nine (11.84%) were ASA IV physical status (Table 1). Of the 79 patients, 22 (27.85%) were female and 57 (72.15%) were male. The mean ± SD age of the patient group was 58.35 ± 11.95 years. Of the 79 patients, 53 (67.09%) were white, 17 (21.52%) were black, seven (8.86%) were Latino or Hispanic, and two were (2.53%) other. Regarding comorbidities, 69 patients (87.34%) had diabetes mellitus, 64 (81.01%) had hypertension, 33 (41.77%) had hyperlipidemia, 18 (22.78%) had coronary artery disease, 29 (36.71%) had chronic kidney disease or end-stage renal disease, 13 (16.46%) had peripheral vascular disease, one (1.27%) had chronic obstructive pulmonary disease, ten (12.66%) had congestive heart failure, 25 (31.65%) had peripheral neuropathy, ten (12.66%) had myocardial infarction, and five (6.33%) had cancer. The median operative time was 72.5 min (interquartile range [IQR], 61.0–90.5 min), and the median time spent in the recovery room was 70.5 min (IQR, 52.0–100.0 min). The median estimated blood loss recorded in operating room reports was 0 mL (IQR, 0.0–11.0 mL) and based on anesthesia records was 10.0 mL (IQR, 0.0–20.0 mL). All of the patients in this study were followed postoperatively for 72 hours. None of the patients had any postoperative complications, defined as any event of stroke, pulmonary edema, respiratory complications, myocardial infarction, bleeding, any requirement for invasive monitoring or admission to the intensive care unit. None of the patients had uncontrolled pain at the donor skin graft site in the 72 hours after surgery.

Table 1. Demographic Features, Clinical Characteristics, and Postoperative Outcomes of the 79 Study Participants

Table 1. Demographic Features, Clinical Characteristics, and Postoperative Outcomes of the 79 Study Participants

Discussion

Given the recent advancements in anesthetic technique and the practice of peripheral nerve blocks, the use of regional anesthesia has been considered the preferred method of analgesia for most ankle and foot surgeries.[9] However, there is a lack of data regarding the appropriate anesthetic techniques for patients with multiple comorbidities and physical status of ASA III and ASA IV undergoing foot surgery procedures needing skin graft donor harvesting and skin graft reconstruction procedures.

ASA I was defined as a healthy patient, ASA II as a patient with mild systemic disease with no functional limitations, ASA III as a patient with moderate systemic disease with function limitations, ASA IV as a patient with severe systemic disease that is a constant threat to life, and ASA V as a moribund patient who is not expected to survive another 24 hours with or without surgery. Several studies have explored the interrater reliability of the classification system in adults and have found that there was considerable variability in the assignment of ASA classification in the adult patient population.[6,10-12] It was, therefore, determined that although the ASA classification system serves as a simple method to categorize, it lacks precision. Sixty percent of respondents stated that they continue to use the classification system as a tool to predict perioperative risk. Its sustained use for this purpose is possibly due to the fact that it continues to be one of the few constantly recorded measures of general physical status preoperatively.[5,13]

On the other hand, other the literature supports a relationship between the rate of perioperative complications and ASA class in adults.[7,14-19] Still, ASA class is not as strong of a predictor of perioperative outcomes as are procedure-associated and patient-associated factors.[14] A study conducted by Klotz et al[20] reported several factors affecting complications postoperatively, namely, malignancy, severity of surgical procedure, symptoms of respiratory disease, and severity of ASA class. Additional factors with predictive value for surgical mortality include male sex, urgency of procedure, and increasing age.[16]

The inconsistency in the findings supporting use of the ASA classification system continues to warrant studies that clarify its usefulness in its role as a tool for assessing perioperative risk. There is little in the literature that explores the recommendations for administering anesthetics in patients with multiple comorbidities and ASA III and IV classifications who undergo peripheral limb procedures. Most patients needing major foot surgeries have comorbid conditions such as diabetes, hypertension, cardiac and renal problems, and obesity. The comorbid conditions lead to the classification of these patients as ASA III or IV. The findings of this study indicated that in the subset of patients undergoing skin graft reconstruction, even with comorbidities (diabetes mellitus, hypertension, hyperlipidemia, coronary artery disease, chronic kidney disease, etc), ASA class was not an effective indicator of increased risk of perioperative complications.

In the present study, major foot surgery and donor full-thickness skin graft harvesting and reconstruction were performed in the same operating room visit in patients with ASA III or ASA IV classification. The study included 62 ASA III patients and nine ASA IV patients. We found that regardless of the severity of ASA classification, patients in the present study were found to have no complications intraoperatively or postoperatively in the recovery room and within the initial 72 hours after skin graft reconstruction. Of note, this was accomplished using regional nerve blockade with monitored intravenous sedation rather than general anesthesia with controlled airway.

Comparatively, in a study by Wolters et al[7] of 6,301 patients who underwent any surgery largely using general anesthesia, there was a greater prevalence of complications within the initial 72-hour postoperative period: 13% in ASA III patients and 46% in ASA IV patients due to cardiac and pulmonary complications. These surgical patients predominantly underwent general anesthesia (97.5%), although the rest were given regional (1.4%) or combined (1.1%) anesthesia. Of these 6,301 patients, 57% underwent major surgery, including thoracic surgery, peripheral vascular interventions, bowel resection, and abdominal interventions; 27% underwent moderate surgery, including cholecystectomy, colostomy formation, and herniotomy; and 16% underwent perineal surgery or repair of soft-tissue wounds. Of note, the patients in this study contrast considerably from those in the present study in terms of procedures and anesthetic administered, thereby making a direct comparison of the results incongruous. However, given the absence of complications in even the patients in the more serious ASA classifications (ASA III and IV), we found no adverse events. We, thereby, conclude that the patients in the present study managed better in comparison.

In addition to donor skin graft harvesting being achievable and a viable option with intravenous sedation because of the briefness of the procedure, we propose that harvesting donor skin graft with skin graft reconstruction in this patient population is efficacious and safe for several reasons. First, ASA III and IV patients often have reduced circulation and peripheral vascular disease or diabetes, which may contribute to minimizing the spread of anesthetics systemically, effectively keeping it localized to the site of administration. In addition, skin graft reconstruction is performed at sites distant from vital organs such that these patients can be given peripheral blocks without adverse hemodynamic instability. Furthermore, in patients with diabetic neuropathy, reduced amounts of medication can be administered to achieve sufficient analgesia compared with patients without diabetic neuropathy. This factor may also contribute to the lack of hemodynamic changes in the present study, which may not have been so otherwise. Finally, the administration of regional blocks can block the potentially unfavorable stress hormone and catecholamine responses that are associated with surgery.[21]

Given the simple technique and the ease with which peripheral nerve blocks can be administered and the efficacy with which it provides patients with analgesic relief, we recommend that it can be used in adjunct with monitored intravenous sedation for patients undergoing skin graft donor harvesting and skin graft reconstruction regardless of ASA class. The advantages of monitored intravenous sedation include facilitation of block administration, improved patient acceptance, and gauging of patient comfort and safety.[22] One of the major limitations of this study is the lack of a group to compare whether there is a difference in the rate of postoperative complications between general or local anesthesia. Previous studies have demonstrated favorable effects of regional anesthesia, namely, with peripheral nerve blocks, in terms of the duration and quality of postanesthetic recovery time after orthopedic surgeries in the outpatient setting.[23-27]

The literature on monitored anesthesia care with intravenous sedation has been largely promising in general surgery, gynecologic surgery, outpatient plastic surgery, urologic surgery, and vascular surgery.[28-33] The primary concerns regarding monitored intravenous sedation continue to be the risk of oversedating and the ensuing respiratory depression.[34] However, it lends numerous advantages over general anesthesia, including decreased hospital stay duration, more rapid recovery, and reduced adverse effects associated with anesthesia, including confusion and nausea,[35] in addition to cost-effectiveness.[36] It has been proposed that monitored anesthesia care with intravenous sedation is at least as safe as general anesthesia given that the patients are appropriately chosen.[37] Given the literature that supports the gradual increase in the utilization of the monitored anesthesia care with intravenous sedation method, we suggest that it be considered as a management option in appropriate patient populations, such as those undergoing peripheral procedures.

Topical anesthesia, skin infiltration with local anesthetic, and delivery of local anesthetic by block, such as regional blockade of the lateral cutaneous nerve of the thigh, provide analgesia, but pain relief can be inadequate. The common problems are inadequate area and inadequate depth of anesthesia, especially if large areas of skin have to be harvested for skin graft.[38] The intravenous sedation provided comfortable analgesia in the operating room for skin graft donor harvest and skin graft reconstruction in foot surgery procedures in the present study. Toxicity from systemic absorption of local anesthetics by skin infiltration to donor skin graft site and infection by any indwelling catheter in the skin graft donor site is eliminated. There are no potential adverse effects given the lack of local anesthetic block at the site of the graft harvest. More sedation was not necessary to keep the patients comfortable, and the surgeon did not have any issues with patient movement during the harvest secondary to inadequate analgesia.

The results of the present study should be viewed with caution given the limitations of this study and the retrospective study design, which introduces inherent bias. Foremost, this study needs follow-up with a larger patient group to be able to more precisely sample the population so that the findings may be more accurately generalized. The size of the skin graft donor site was not measured, and the location of the skin graft site (medial or lateral thigh) was also not documented. Another limitation of this study stems from its reliance on the accuracy of the data, and postoperative complications, in the medical record. Furthermore, the lack of an appropriate control group in this study necessitates a randomized controlled study with patients assigned to receive either local anesthesia and intravenous sedation or general anesthesia with controlled airways.

Conclusions

The results of this study suggest that skin graft reconstructions can be conducted with intravenous sedation and local anesthesia regardless of ASA class in these patients. We propose an exception to the previous practice of using general anesthesia with controlled airway in ASA III and IV patients undergoing procedures based on our findings of negligible blood loss and no perioperative complications in any of the patients who underwent skin graft reconstruction. Despite the elevated ASA III and IV classes in some patients and various comorbidities, we propose that this subset of patients is suitable for skin graft reconstruction given the distal nature of the site of the skin graft from vital organs and the brevity of the procedure. Furthermore, associated peripheral vascular disease likely keeps the anesthetic localized and further reduces the risk of adverse outcomes.