STUDY. Intra-Incisional Prophylactic Antibiotics for Dermatologic Surgery

Transcription

1 Intra-Incisional Prophylactic Antibiotics for Dermatologic Surgery Robert D. Griego, MD; John A. Zitelli, MD STUDY Objective: To determine the efficacy of intra-incisional antibiotics in decreasing the risk of wound infections in cutaneous surgery. Design: Prospective, blinded, randomized, placebocontrolled trial conducted during an 8-month period. Setting: A private practice Mohs micrographic surgery referral center. : Seven hundred ninety consecutive patients referred for Mohs surgery or other dermatologic surgery were randomized to receive anesthesia either with study compound or placebo. The 2 groups were equivalent with respect to age and sex distribution and the lesions treated were similar in character. No patients were withdrawn for adverse effects. Interventions: received local anesthesia before surgery with either buffered lidocaine hydrochloride or a solution consisting of nafcillin sodium in buffered lidocaine. Main Outcome Measures: All surgical wounds were evaluated in a blinded fashion at the time of suture removal (5-7 days) and scored according to a standardized assessment chart based on erythema, edema, and the presence of purulent discharge. Results: Seven hundred ninety consecutive patients with 908 surgical wounds were enrolled in this study. A total of 12 wound infections were recorded. Eleven (2.5%) of these occurred in the control group, while only 1 (0.2%) occurred in the nafcillin group. This difference was highly significant (P =.003). Observers were blinded to patient groupings particularly for surgical wound scoring. Conclusions: This study offers strong supporting data for the use of a single intra-incisional dose of an antibiotic administered immediately before dermatologic surgery. The use of nafcillin and buffered lidocaine solution is inexpensive, safe, convenient, and effective. Arch Dermatol. 1998;134: Drs Griego and Zitelli are in private practice in Mesa, Ariz, and Pittsburgh, Pa, respectively. POSTOPERATIVE wound infections are serious complications of cutaneous surgery. They cause pain, delay healing, and most important, they result in noticeable scarring. For these reasons, prophylactic antibiotics are often used in cutaneous surgery, particularly for facial reconstruction. No guidelines exist for the use of prophylactic antibiotics in dermatologic surgery. For the prevention of endocarditis, bacteremia may be minimized by the use of oral antibiotics 1 hour before surgery or intravenous (IV) antibiotics. Wound infections in head and neck or general surgery may be reduced by the use of IV antibiotics 1 hour before surgery, but neither oral nor IV antibiotics have been well tested for preventing infections in cutaneous surgery. Furthermore, both IV administration and a 1-hour delay from the time of administering oral antibiotics to the beginning of surgery are significant inconveniences for elective office based surgery. Therefore, these regimens are rarely used, and instead a common practice is to prescribe postoperative antibiotics when the patient is discharged from the office, or alternatively administering oral antibiotics within minutes of beginning surgery. The efficacy of these practices in decreasing wound infection is unknown. The purpose of this article is to report the efficacy of intra-incisional antibiotics for decreasing the risk of wound infections in cutaneous surgery. This route of administration should immediately achieve high antibiotic levels in tissue and is easily incorporated into an office-based surgical setting. RESULTS A total of 397 patients with 461 wounds received nafcillin as intra-incisional pro- This article is also available on our Web site: 688

2 PATIENTS, MATERIALS, AND METHODS A prospective, blinded, randomized, placebo-controlled study was conducted during an 8-month period from October 25, 1995, to June 30, A total of 790 consecutive patients (411 males and 379 females) with 908 surgical wounds were enrolled in this study. were randomly allocated to 2 groups. The control group (393 patients) received local anesthesia injected approximately 15 minutes before surgery by infiltration into the dermis and subcutaneous tissues, with 1% buffered lidocaine hydrochloride with epinephrine 1: The treatment group (397 patients) received local infiltration of anesthesia with a solution consisting of 0.5 mg of nafcillin sodium per milliliter of 1% buffered lidocaine hydrochloride with epinephrine 1: The surgical procedures performed in this study included the reconstruction following Mohs micrographic excisions using complex linear closure, flaps, and skin grafts; the second stage of any staged repair; and excision and repair of benign dermatologic neoplasms. Preoperative skin preparation was performed using antiseptic scrub (Technicare, Care Tech Laboratories, St Louis, Mo) (3.0% chloroxylenol and 3.0% cocamidopropyl PG-dimonium chloride phosphate). All linear closures and flaps were sutured in 2 or more layers. Deep subcutaneous suturing was performed with absorbable materials (either polyglactin 910 or monofilament polyglyconate), and superficial suturing was performed using either rapidly absorbable plain gut or nylon sutures. All surgical procedures were performed using sterile techniques in a licensed ambulatory surgical facility operating room. who had received any systemic antibiotic within 2 weeks before surgery were excluded from the study; no prophylactic perioperative systemic antibiotics were administered. with a known history of allergy to penicillin were assigned to the control group. MATERIALS The nafcillin sodium in buffered lidocaine solution used in this study was prepared as follows: 5 ml of sodium bicarbonate (84 mg/ml) was injected into a 50-mL vial of 1% lidocaine hydrochloride with epinephrine 1: ; 3.4 ml of the buffered lidocaine solution was used to reconstitute the vial containing 1 g of nafcillin sodium for injection; 0.1 ml of the reconstituted nafcillin solution was withdrawn from the vial and injected into the 50-mL vial of buffered lidocaine with epinephrine; the remainder of the vial of reconstituted nafcillin was placed in a freezer and kept frozen for later use. The 3-mL syringes were filled from the 50-mL vial of the preparation and kept refrigerated for use in the study as needed; any syringe not used within 48 hours of refrigeration was discarded. We tested our nafcillin lidocaine study solution against laboratory strains of penicillinase-producing Staphylococcus aureus by incubation of equal volumes of Mueller Hinton broth inoculated with bacteria in concentrations of 10 5 organisms per milliliter and the drug mixture at 4 different ratios of drug-to-bacteria broth (1:1, 1:2, 1:4, and 1:8). The inoculated test tubes were incubated at 35 C for 24 hours. The presence of growth was determined visually by a microbiologist. Any tube found to exhibit bacterial growth by 48 hours was considered positive, any without growth was considered negative. We found complete and consistent bactericidal activity of all dilutions of the drug mixture as follows: after 11 days under refrigeration, after 60 days frozen, and after 7 days at room temperature. The drug mixture demonstrated no antibacterial effect in identical experiments after being stored at room temperature for 20 days, thus demonstrating some chemical instability of the solution. The control solution exhibited no antimicrobial activity by identical broth inoculation experiments. WOUND SCORING We enlisted the use of a recently published scoring system for assessment of the postoperative wound condition (Table 1). 1 All surgical wounds were evaluated in a blinded fashion at the time of suture removal (5-7 days) and scored by the physician or nurse according to this assessment chart. A score of 4 or higher was classified as a wound infection. Specimens were obtained for culture from all infected surgical wounds, as well as from those with equivocal evidence of infection (wound scores 3). Wounds with scores of 3 that demonstrated heavy culture growth of a single pathogenic bacterium were also considered infections. All isolates recovered were identified by standardized methods of subculturing. phylaxis, while 393 patients with 447 wounds received buffered lidocaine with epinephrine only. The 2 groups were similar with respect to age and sex distribution, and the treated lesions were similar in character (Table 2 and Table 3). A total of 908 surgical procedures were examined. Of these, 877 were Mohs excisions followed by various reconstructive procedures (Table 3). The number of stages of Mohs surgery required per tumor for each group were similar, as were the necessary reconstructive procedures; however, there were significantly more non Mohs excisions in the control group. There was no difference in mean volume of anesthetic required per lesion between the 2 groups (control, 4.15 ml; nafcillin, 4.09 ml); this volume corresponds to the size of the defect repaired. Wound scores were higher in the control group (average, 0.62) compared with the nafcillin group (average, 0.37) (Table 4). There were no allergic reactions or complications attributable to the use of nafcillin. A total of 12 wound infections were recorded. Eleven (2.5%) of these occurred in the control group, while only 1 (0.2%) occurred in the nafcillin group. This difference was highly significant (P =.003). All infections in the control group were caused by S aureus. None of these patients were assigned to the control group because of a penicillin allergy. Five of these infections occurred on the nose, 2 on the ear, 2 on the 689

3 Table 1. Scoring System of Postoperative Wound Condition Score Condition Normal Healing 1 Normal healing but with 1 of the following signs of infections: erythema, edema, or increased pain 2 Normal healing but with 2 of the following: erythema, edema, increased pain, or hemoserous discharge alone 3 Normal healing but all 3 of the following: increased erythema, edema, and increased pain; or 1 of the latter combined with hemoserous discharge Infection 4 Pus; or hemoserous discharge combined with 2 of the following: erythema, edema, or increased pain 5 Pus combined with 1 of the following: erythema, edema, or increased pain; or hemoserous discharge combined with erythema, edema, and increased pain 6 Pus combined with 2 of the following: erythema, edema, or increased pain 7 Pus combined with erythema, edema, and increased pain Table 3. Characteristics of Lesions Treated Characteristics Control (n = 447) (n = 461) Location Scalp, palm, or sole 10 8 Face Neck 8 7 Trunk Extremities 14 5 Hands or feet 1 2 Repair, No. (%) Primary 323 (72) 331 (72) Skin graft 93 (21) 82 (18) Flap 31 (7) 47 (10) Mohs stages, No. (%) Non-Mohs cases 33 (7) 8 (2) (54) 227 (49) (24) 155 (34) 3 65 (15) 71 (15) Table 2. Characteristics of Study Characteristics Control (n = 393) cheek, and 1 each on the abdomen and forehead. Six of the infections involved skin graft repairs, 2 were second stages of paramedian forehead flaps, 2 were cheek advancement flaps, and 1 was a primary closure. The infection in the nafcillin group was caused by Enterobacter aerogenes. Although this case was classified as an infection because of the heavy growth of a single organism combined with a clinical picture suspicious for an infection (wound score, 3), the lack of purulent discharge resulted in a wound score slightly below the threshold for the definition of infection using the scoring system shown in Table 1. The growth of such an unusual bacterium for facial skin, resolution within days of treatment with an antibiotic shown to be ineffective by in vitro sensitivity testing, and the clinical setting of a tight flap closure make it likely that this case was actually only bacterial colonization associated with wound-edge necrosis from the tight sutures. COMMENT (n = 397) Mean age, y % Female Multiple tumors, % Nonsmoker In this study we have shown that a single intra-incisional dose of a local anesthetic preparation containing nafcillin can lower the rate of postoperative wound infections compared with local anesthetic alone. The difference in infection rates between the treatment group (0.2%) and the control group (2.5%) was highly significant (P =.003). The rate of infection in our treatment group compares favorably with the lowest published rates for cutaneous surgery (range, 0.70% 2-43% 3 ). Table 4. Wound Scores* Score Control (n = 447) (n = 461) (70) 340 (74) (24) 106 (23) 2 12 (3) 12 (3) 3 4 (1) 3 (1) 4 7 (2) 0 (0) 5 3 (1) 0 (0) *All values are number (percentage). 2 (df = 3) = 8.04; P =.04. Our study also provides the first data on wound infection rates in Mohs surgery without prophylaxis followed by various reconstructive procedures. Our infection rate for the 447 procedures performed in the control group was 2.5%, compared with the only other published rate of 4.3% for cutaneous surgery without prophylaxis. 3 This 2.5% infection rate is nearly identical to the previously published rate of 2.29% for Mohs surgery and excisions that included the use of prophylactic perioperative antibiotics for selected high-risk wounds. 2 The intra-incisional delivery of antibiotics offers the advantage of immediate achievement of effective levels in tissue. Conversely, IV, intramuscular, or oral administration requires a significant delay to attain effective levels of an antimicrobial in tissue. Intravenous prophylactic antibiotics for surgical wound infections have been shown to be most effective when administered within the 2 hours before surgery, 4 and are commonly given within 30 minutes of an operation. 5,6 The optimal timing for intramuscular or oral administration for wound infection prophylaxis is not as clear; no data exist to support the common recommendation of administering oral prophylaxis 1 or 2 hours before a procedure. 5,7 The translation of the timing recommendations of oral antibiotics for endocarditis prophylaxis to cutaneous surgery is complicated by the associated variabilities in absorption, dis- 690

4 tribution, metabolism, and bioavailability at the site of the wound. Four studies 8-11 using intra-incisional prophylactic antibiotics have been published. The first 8 of those demonstrated a significant reduction in the incidence of wound infection with the use of preoperative intra-incisional cefoxitin for abdominal surgery. Others 10 have demonstrated a significantly lower rate of infection in those with a single intra-incisional dose of cefamandole than in those who received the same IV dose. A less well-designed study 11 reported that the infiltration of cefamandole along with the anesthetic solution reduced the infection rate from 4.6% to 0% in a series of 308 patients undergoing primary inguinal hernia repair under local anesthesia. Intra-incisional administration may also offer advantages associated with reducing exposure of other body sites to antibiotics. These could include reduced risks for toxic and some allergic adverse effects; interactions with other medications; gastrointestinal intolerance; variable absorption; secondary yeast or bacterial overgrowth complications; and the potential to contribute to antibiotic resistance. Although no scientific data exist to support these theoretical benefits, even complete systemic absorption of the small amount of nafcillin required for prophylaxis in our study (an average of 2 mg per patient) would logically be associated with fewer complications than the much higher doses required for systemic prophylaxis. However, the risk for anaphylaxis should not be overlooked in patients allergic to penicillin or allergic to cephalosporin. Several different mechanisms are involved in the development of bacterial resistance to antimicrobials. These include factors that encourage the development, facilitate the persistence, and augment the transmission of resistant organisms. 12 Among these is the use of antimicrobials. 13 Although their use does not appear to be the major cause of the development of resistant bacterial mutants (these are most likely to develop through spontaneous genetic mutations or transfer of genetic material from one resistant species to another), 12,14,15 the selective pressures associated with antimicrobial use creates ideal environments for resistant strains to multiply. Correspondingly, their use is the predominant cause of the proliferation of resistant strains, and has led to the rapidly increasing prevalence of multidrugresistant pathogens. 16,17 Therefore, it is intuitive that by minimizing the number of ideal environments created for resistant bacteria to multiply, their proliferation should be diminished. The administration of an antimicrobial agent only to the desired site of infection prophylaxis removes the exposure of common body sites that harbor bacteria (nasopharynx, respiratory tract, gastrointestinal tract, and genitourinary tract), and may decrease the proliferation or development of these strains. Nafcillin was chosen for this study because of its spectrum of activity (coverage for S aureus and Streptococcus pyogenes), suitability for intracutaneous and subcutaneous administration, chemical stability in lidocaine hydrochloride, wide availability, and low cost. 18 Unpublished trials conducted by Wyeth Laboratories, Philadelphia, Pa (oral communication, September 15, 1995) have shown sustained antimicrobial activity of an unbuffered solution of reconstituted nafcillin in lidocaine for up to 48 hours under refrigeration and 14 days frozen. 1 However, our in vitro studies clearly demonstrate chemical stability for much longer periods. It would be safe to keep refrigerated solutions for up to 1 week, frozen solution for 1 month, and solution may be stored at room temperature for up to 3 days, if necessary. The dosage of nafcillin used in this study was based on peak serum levels associated with IV administration of 500 mg of nafcillin sodium. To achieve similar concentrations in a lidocaine solution, ml of the reconstituted nafcillin sodium solution mixed with 50 ml of lidocaine would be required. This minuscule volume of solution was believed to be impractical for clinical use. It was determined that 0.1 ml was the minimal volume of solution practical for precisely withdrawing from the vial of reconstituted nafcillin sodium. This volume, when diluted in 50 ml of buffered lidocaine, produces a concentration of 0.5 mg of nafcillin sodium per milliliter of lidocaine, which is 45 times the peak serum levels obtained after IV administration of 500 mg of nafcillin sodium. This was the concentration used for this study. The monetary cost of the nafcillin used in this study was minimal. Our cost for ten 1-g vials of nafcillin sodium for injection was $39.75; $3.98 per gram. When reconstituted, each 1-g vial is capable of formulating 40 separate 50-mL vials of our nafcillin-buffered lidocaine solution. The average cost of nafcillin per study case was less than 1 cent. The labor cost of formulating and administering our nafcillin-lidocaine solution was also minimal. The additional time required for preparation of the study solution was insignificant compared with that for the standard (control) buffered lidocaine solution. The routine use of antibiotic prophylaxis for cutaneous surgery is controversial. Some Mohs surgeons have reported the use of prophylactic antibiotics in up to 77% of patients, and many use prophylaxis for all repairs of facial defects. 19 We have demonstrated the effectiveness of a single intra-incisional dose of an antibiotic formulation administered immediately before the repair of defects following Mohs surgery; this method of prophylaxis may prove to be a better alternative for other highrisk wounds. In conclusion, we have found the use of nafcillin and buffered lidocaine solution to be inexpensive, safe, convenient, and effective. Because there are scenarios in which antimicrobials should be administered for wound infection prophylaxis, we believe that their local administration is sensible, practical, and in this era of cost containment and increasing drug resistance, it is responsible. Accepted for publication December 12, Presented at the 29th Annual Meeting of the American Society for Dermatologic Surgery, Boston, Mass, May 8, Reprints: John A. Zitelli, MD, 5200 Centre Ave, Suite 303, Pittsburgh, PA

5 REFERENCES 1. Amland PF, Andenaes K, Samdal F, et al. A prospective, double-blind, placebocontrolled trial of a single dose of azithromycin on postoperative wound infections in plastic surgery. Plast Reconstr Surg. 1995;96: Futoryan T, Grande D. Postoperative wound infection rates in dermatologic surgery. Dermatol Surg. 1995;21: Bencini PL, Galimberti M, Signorini M, et al. Antibiotic prophylaxis of wound infections in skin surgery. Arch Dermatol. 1991;127: Classen DC, Evans RS, Pestotnik SL, et al. The timing of prophylactic administration of antibiotics and the risk of surgical-wound infection. N Engl J Med. 1992; 326: Haas AF, Grekin RC. Antibiotic prophylaxis in dermatologic surgery. J Am Acad Dermatol. 1995;32: Antibiotic prophylaxis in surgery. Med Lett Drugs Ther. 1989;1: Rabb DC, Lesher JL. Antibiotic prophylaxis in cutaneous surgery. Dermatol Surg. 1995;21: Taylor TV, Walker WS, Mason RC, et al. Preoperative intraparietal (intraincisional) cefoxitin in abdominal surgery. Br J Surg. 1982;69: Armstrong CP, Taylor TV, Reeves DS. Pre-incisional intraparietal injection of cephamandole: a new approach to wound infection prophylaxis. Br J Surg. 1982;69: Dixon JM, Armstrong CP, Duffy SW, et al. A randomized prospective trial comparing the value of intravenous and preincisional cefamandole in reducing postoperative sepsis after operations upon the gastrointestinal tract. Surg Gynecol Obstet. 1984;158: Lazorthes F, Chiotasso P, Massip P, et al. Local antibiotic prophylaxis in inguinal hernia repair. Surg Gynecol Obstet. 1992;175: Cohen ML. Emerging problems of antimicrobial resistance. Ann Emerg Med. 1994; 24: Mouton RP, Glerum JH, van Loenan AC. Relationship between antibiotic consumption and frequency of antibiotic resistance of four pathogens: a seven-year survey. J Antimicrob Chemother. 1976;2: Stratton CW, Ratner H, Johnston PE, et al. Focused microbiologic surveillance by specific hospital unit: practical application and clinical utility. Clin Ther. 1993; 15: Sanders CC, Sanders WE Jr, Thompson KS. Flouroquinolone resistance in staphylococci: new challenges. Eur J Clin Microbiol Infect Dis. 1995;1: Thornsberry C. Trends in antimicrobial resistance among today s bacterial pathogens. Pharmacotherapy. 1995;15:3S-8S. 17. Goldmann DA, Weinstein RA, Wenzel RP, et al. Strategies to prevent and control the emergence and spread of antimicrobial resistant microorganisms in hospitals. JAMA. 1996;275: Howard RJ, Simmons RL, eds. Antimicrobial agents. In: Surgical Infectious Diseases. 3rd ed. East Norwalk, Conn: Appleton & Lange; 1995: Rudolph R, Miller SH. Reconstruction after Mohs cancer excision. Clin Plast Surg. 1993;20: Submissions for Special Features Critical Situations Readers are invited to submit examples of newly described disorders, the use of new diagnostic technology, dermatologic manifestations of important social disorders such as child abuse, and cases that highlight the complex nature of acute care dermatology to Thomas G. Cropley, MD, Department of Medicine (Dermatology), University of Massachusetts Medical Center, Worcester, MA When appropriate, these should be written in case presentation format with a brief discussion following. The Cutting Edge Clinicians, local and regional societies, residents, and fellows are invited to submit cases of challenges in management and therapetics to this section. Cases should follow the established pattern and be submitted double-spaced and in triplicate. Photomicrographs and illustrations must be clear and submitted as positive color transparencies (35-mm slides) or blackand-white prints. Do not submit color prints unless accompanied by original transparencies. Material should be accompanied by the required copyright transfer statement, as noted in Instructions for Authors. Material for this section should be submitted to George J. Hruza, MD, Cutaneous Surgery Center, 1 Barnes Hospital Plaza, Suite 16411, St Louis, MO Reprints are not available. Issues in Dermatology Issues in Dermatology solicits provocative essays relevant to all aspects of dermatology. Please submit manuscripts to the section editor, A. Bernard Ackerman, MD, 2 E 70th St, New York, NY Off-Center Fold Clinicians, local and regional societies, and residents and fellows in dermatology are invited to submit quiz cases to this section. Cases should follow the established pattern and be submitted double-spaced. Photomicrographs and illustrations must be clear and submitted as positive color transparencies. If photomicrographs are not available, the actual slide from the specimen will be acceptable. Material should be accompanied by the required copyright transfer statement, as noted in Instructions for Authors. Material for this section should be submitted to Lori Lowe, MD, Department of Pathology, University of Michigan, Medical Science 1, M5242, 1301 Catherine Rd, Ann Arbor, MI Reprints are not available. 692