 |
|
Clindamycin for Intraincisional Antibiotic Prophylaxis in Dermatologic Surgery
Michael J. Huether, MD;
Robert D. Griego, MD;
David G. Brodland, MD;
John A. Zitelli, MD
Arch Dermatol. 2002;138:1145-1148.
ABSTRACT
| |
Objective To assess the efficacy of intraincisional clindamycin therapy as an
alternative to nafcillin treatment in decreasing the risk of postoperative
wound infections in dermatologic surgery.
Design Prospective, double-blinded, randomized, placebo-controlled trial conducted
over a 7-month period.
Setting Three private practice Mohs micrographic surgery referral centers.
Patients A total of 1030 consecutive patients who underwent Mohs micrographic
surgery with subsequent reconstruction.
Interventions Prior to reconstruction, patients were randomly assigned to receive
either intraincisional buffered lidocaine with epinephrine containing clindamycin
or buffered lidocaine with epinephrine without clindamycin. Nurses and physicians
who scored the wound at follow-up were blinded to the treatment conditions.
Main Outcome Measures Surgical wounds evaluated at the time of suture removal were scored
according to a standardized assessment based on erythema, edema, and the presence
of purulent discharge. Wounds scored 4 or higher were considered to be infected.
Bacterial cultures obtained when indicated were also compared.
Results Of the 1172 surgical wounds included in the study, 29 had wound scores
of 4 or higher, 6 in the study group and 23 in the control group (P = .001, Fisher exact test). Of these 29, 18 had culture-positive
infections. Four of these occurred in the study group, and 14 occurred in
the control group (P = .02, Fisher exact test).
Conclusions The results of this study further support the efficacy of single-dose
preoperative intraincisional antibiotic treatment for dermatologic surgery.
With the relatively high prevalence of patient-reported penicillin allergies,
buffered lidocaine containing clindamycin offers an inexpensive, safe, convenient,
and effective alternative.
INTRODUCTION
POSTOPERATIVE SURGICAL wound infections can have a significant effect
on the outcome of a dermatologic surgical procedure, including the final appearance
of the wound. In many cases, patients are given preoperative or postoperative
systemic antibiotics in an effort to decrease morbidity, despite the low estimated
incidence of postoperative wound infections.1
Unfortunately, giving either oral or intravenous antibiotics prior to surgery
is not only inconvenient and cumbersome, but the efficacy of such procedures
is largely unstudied in dermatologic surgery.
A report by 2 of us has shown that intraincisional nafcillin treatment
prior to skin surgery resulted in a statistically significant reduction in
the occurrence of postoperative wound infections.2
Unfortunately, there are many patients with a potential or documented allergy
to penicillin or the penicillin class of antibiotics. For these patients,
use of nafcillin is contraindicated, and an alternative antibiotic for intraincisional
prophylaxis would be desirable.
Given the spectrum of antimicrobial effects of nonpenicillin-type antibiotics,
several candidates for use in skin surgery may be appropriate, but the macrolide
clindamycin offers the most appropriate antimicrobial activity against the
most common pathogens in cutaneous wound infections. Other injectable drugs
that were considered (vancomycin and ciprofloxacin) had limitations such as
restricted use recommendations or less appropriate antimicrobial coverage.
The purpose of this study is to examine the efficacy of a single dose of preoperative
intraincisional clindamycin in preventing postoperative wound infections in
dermatologic surgery.
METHODS
A prospective, blinded, randomized, placebo-controlled study was conducted
in our 3 private practice offices from February 12, 1998, to September 18,
1998. Consecutive patients undergoing Mohs micrographic surgery in whom reconstruction
was to be performed (primary closure, local or fasciocutaneous pedicle flap,
or skin graft) and who did not meet any exclusion criteria were included in
the study. Exclusion criteria included history of allergic reaction to clindamycin
or other macrolide, concurrent or perioperative use of systemic antibiotics,
or inability to return for follow-up evaluation. Informed consent was obtained.
CLINDAMYCIN SOLUTION
The concentration of clindamycin used in the study solution was determined
by extrapolation from published data. With intravenous administration of standard
doses of clindamycin, serum levels of 7 to 14 µg/mL are achieved at
steady-state dosing.3 Also, after standard
intravenous dosing of clindamycin, steady-state wound fluid concentrations
of 4 to 5 µg/mL are achieved and maintained within 1 to 5 hours.4 However, taking into account dilution among interstitial
tissue fluid after injection, and to avoid the inaccuracies of delivering
very small volumes when mixing the solutions, we decided to investigate concentrations
that were much higher than the standards described above. Varying concentrations
of clindamycin (272 µg/mL, 408 µg/mL, and 544 µg/mL) were
tested. Using Mueller-Hinton broth inoculated with laboratory strains of nonpenicillinase Staphylococcus aureus bacteria in concentrations of 105 organisms per milliliter, we diluted the study mixtures to 4 different
ratios of solution-to-bacteria broth (1:1, 1:2, 1:4, and 1:8). Incubation
of the sequential dilutions was done at 35°C and was performed and interpreted
by a microbiologist who assessed growth at 24 and 48 hours. The 272-µg/mL
concentration allowed growth of bacteria at 48 hours when tested 7 days after
mixing the study solution at a dilution of 1:8. Further testing of this dilution
was not pursued. At concentrations of 408 µg/mL and 544 µg/mL,
no growth of bacteria occurred at 48-hour culture assessment in any dilution
when tested up to 30 days after mixing.
Storage methods of the study solution were then tested and compared
for bactericidal activity. There was complete bactericidal activity for all
dilutions of preparations (408 µg/mL and 544 µg/mL) that were
stored at room temperature, refrigerated, or frozen for up to 30 days after
mixing the study solutions. In an attempt to avoid the potential tissue irritancy
of higher concentrations, the 408-µg/mL concentration was selected as
the lowest effective concentration for the current study. Our study solution,
at a concentration of 408 µg/mL, is 29 to 58 times the standard steady-state
serum concentration of clindamycin administered by intravenous injection.
PROCEDURE
Patients eligible for the study were randomized to receive local anesthetic
with or without clindamycin. The anesthetic solutions were injected into the
dermis and subcutaneous tissue following tumor extirpation, approximately
15 minutes prior to reconstruction. The volume of the preparations injected
was that required to achieve adequate local anesthesia. After injection, the
skin was treated with a preoperative antiseptic scrub containing 3.0% chloroxylenol
and 3.0% cocamidopropyl PG-dimonium chloride phosphate (Technicare; Care Tech
Laboratories, St Louis, Mo) prior to skin incision. Superficial, rapidly absorbable
plain gut suture or nylon was used in each closure with buried interrupted
absorbable suture used where appropriate in layered closures.
Syringes with study medications were made each week. They contained
either 1% lidocaine with 1:100 000 epinephrine buffered with sodium bicarbonate
(control) or 1% lidocaine with 1:100 000 epinephrine buffered with sodium
bicarbonate containing clindamycin (408 µg/mL). Both solutions were
prepared by adding 5 mL of 8.4% sodium bicarbonate (50 mEq/50 mL) to a 50-mL
vial of 1% lidocaine with 1:100 000 epinephrine. For the clindamycin
solution, 0.15 mL of 150-mg/mL clindamycin phosphate injection was also added
to yield a clindamycin concentration of 408 µg/mL. The 2 types of syringes
were then stored in separate bins, and the medical staff selected one as each
consecutive patient was ready for anesthesia. No note was made in the patient
chart that indicated which solution was used. To facilitate approximately
equal patients in each group, medical staff were instructed to use syringes
from alternating bins.
WOUND SCORING AND ASSESSMENT
Patients were assessed at their follow-up visit for suture removal (5-8
days). At this visit, the wound appearance was scored by a physician or a
nurse blind to the treatment group using a standardized wound scoring scale
(Table 1).5
Patients were also questioned regarding cutaneous and systemic allergic reactions
and bowel symptoms including nausea, vomiting, and diarrhea. When patients
had multiple wounds, each wound was assessed individually and given a separate
wound score. A wound was considered infected if a wound score of 4 or higher
was obtained at postoperative assessment. If infection was suspected, the
wound was cultured, and empiric antibiotics were prescribed, if clinically
indicated, and logged. One patient, who had 2 tumors removed from her nose,
was thought to have tissue necrosis at the wound edges at follow-up. Culture
showed Escherichia coli, but owing to the clinical
impression of wound edge necrosis, antibiotics were not prescribed, and the
erythema resolved without sequelae. Analyses with and without this patient
included yield-equivalent results.
|
|
|
|
Table 1. Scoring System of Postoperative Wound Condition
|
|
|
ANALYSIS
Comparisons between the 2 study groups on demographic and wound characteristics
were done using the Huber-White sandwich estimators for SEs because some participants
contributed multiple observations.6 Comparisons
between the 2 study groups on the outcome variables were done with exact tests
because of the low frequency of infections. The Fisher exact test was used
for dichotomous outcomes and the Freeman-Halton extension of the Fisher exact
test for the 8-level wound score.
RESULTS
The 1172 wounds eligible for the study represented 93.5% of the 1253
wounds assessed for inclusion. Of the 1030 patients enrolled in the study,
910 had a single observation, 94 had multiple observations in which all wounds
received either drug or control, and 26 patients had observations in both
drug and control categories (treatment was obtained at different visits).
The patients in the 2 groups had similar demographic characteristics,
and the lesions treated in each group were similar in character (Table 2). The mean defect diameter following
Mohs micrographic surgery was similar for both groups (study, 1.80 cm; control,
1.72 cm). The mean volume of anesthetic solution injected was similar for
both groups (study, 4.25 mL; control, 4.10 mL) and correlated with lesion
size. There were no statistically significant differences between the 2 groups
in demographic characteristics, number of non–Mohs micrographic surgery
cases, Mohs stages, location of lesions, or type of reconstruction. A total
of 452 patients with 598 surgical wounds received the clindamycin study solution
for intraincisional antibiotic prophylaxis while 458 patients with 574 surgical
wounds received the control solution.
|
|
|
|
Table 2. Characteristics of Study Patients and Tumors*
|
|
|
Wound scores were lower in the clindamycin group (mean, 0.19) than in
the control group (mean, 0.35) (P = .20; Mann-Whitney U). Six wounds (1.0%) in the clindamycin group and 23 wounds
(4.0%) in the control group had scores of 4 or higher (P = .001, Fisher exact test) (Table
3).
|
|
|
|
Table 3. Wound Scores for Surgeries in Clindamycin and Control Groups*
|
|
|
Culture-positive wound infections were less frequent in the clindamycin
group (4 wounds) than in the control group (14 wounds) (P = .02, Fisher exact test). The 4 organisms cultured from the study
group grew S aureus. All of these infected wounds
were on the face and were closed primarily. Twelve of the wound infections
in the control group were S aureus infections, 1
was methicillin-resistant S aureus, and 1 was mixed S aureus and Enterococcus. Of
the infected wounds in this group, 6 occurred on the face (other than the
nose), 4 on the nose, 2 on the trunk, and 1 on the scalp. Twelve of these
wounds were closed primarily, and 2 were closed by local flaps.
No allergic reactions (drug eruption or anaphylactoid reactions) were
noted in either group. Gastrointestinal symptoms were recorded in both groups.
In the clindamycin group, 4 patients noted postoperative nausea. In the control
group, 1 patient noted postoperative cramping. No vomiting or diarrhea was
reported in either group.
COMMENT
This study further supports the efficacy of treatment with intraincisional
antibiotics to prevent postoperative wound infections in dermatologic surgery.
In this study, local anesthetic administered with clindamycin prior to surgery
decreased the incidence of clinical signs of wound infections and decreased
the number of culture-proven postoperative wound infections when compared
with placebo.
These findings concur with previous reports showing that intraincisional
nafcillin treatment resulted in fewer postoperative wound infections. Our
findings suggest that an alternative antibiotic, clindamycin, is also effective
in preventing postoperative wound infections, thus providing an alternative
antibiotic for penicillin-sensitive patients. While several penicillin alternatives
were considered during study design, clindamycin was chosen primarily because
of its antimicrobial activity against the most common skin pathogens, gram-positive
cocci. It was also selected for its easy availability, low cost (pennies per
3-mL syringe of anesthetic), suitability for subcutaneous injection, and stability.
Vancomycin was considered but was excluded owing to issues of use restriction
and concern over promoting resistance to this "last chance" antibiotic. Ciprofloxacin
was also considered, but despite adequate activity against S aureus, it has less activity against Streptococcus species.
The potential benefits of intraincisional antibiotic prophylaxis2 include immediate delivery to the site where it is
needed, ease of use, enhanced compliance, and low cost compared with other
routes of delivery. There are also theoretically decreased potentials for
resistance, drug interactions, intolerance, and bacterial or fungal overgrowth
because miniscule quantities are delivered into a localized site. Though some
of these potential benefits have been questioned,7-8
given the potential morbidity of surgical wound infections (despite their
low 2%-4% incidence), the benefits appear to far outweigh the risks based
on information currently available.
AUTHOR INFORMATION
Accepted for publication December 12, 2001.
Corresponding author: Michael J. Huether, MD, 5501 N Oracle Rd, Suite
161, Tucson, AZ 85704-3851 (e-mail: mjhah{at}gci-net.com).
Dr Huether is in private practice in Tucson, Ariz; Dr Griego, in Mesa,
Ariz; and Drs Brodland and Zitelli, in Pittsburgh, Pa.
REFERENCES
| |
1. Futoryan T, Grande D. Postoperative wound infection rates in dermatologic surgery. Dermatol Surg. 1995;21:509-514.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
2. Griego RD, Zitelli JA. Intra-incisional prophylactic antibiotics for dermatologic surgery. Arch Dermatol. 1998;134:688-692.
FREE FULL TEXT
3. USP DI: Drug Information for the Health Care Professional. Greenwood Village, Colo: Micromedex; 2000:926-929.
4. Bagley DH, Mac Lowry J, Beazley RM, Gorschboth C, Ketcham AS. Antibiotic concentration in human wound fluid after intravenous administration. Ann Surg. 1978;188:202-208.
PUBMED
5. Amland PF, Andenaes K, Samdal F, et al. A prospective, double-blind, placebo-controlled trial of a single dose
of azithromycin on postoperative wound infections in plastic surgery. Plast Reconstr Surg. 1995;96:1378-1383.
WEB OF SCIENCE
| PUBMED
6. Huber PJ. The behavior of maximum likelihood estimators under non-standard conditions. In: The Fifth Berkeley Symposium on Mathematical
Statistics and Probability. Berkley: University of California Press;
1967:221-233.
7. Haas AF, Grekin RC. Practical thoughts on antibiotic prophylaxis. Arch Dermatol. 1998;134:872-873.
FREE FULL TEXT
8. Griego RD, Zitelli JA. Antibiotic prophylaxis. Arch Dermatol. 1999;135:716-717.
FREE FULL TEXT
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter
What's this?
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Intraincisional Clindamycin Prophylaxis
Journal Watch Dermatology 2002;2002:3-3.
FULL TEXT
|
