Documente Academic
Documente Profesional
Documente Cultură
Julio López,1 Gilberto Gómez,1 Karime Rodriguez,2 Julio Dávila,3 José Núñez,2 and Luis Anaya1
Downloaded by National University of Singapore NUS SWETS/24793917 from www.liebertpub.com at 04/21/18. For personal use only.
Abstract
Background: Skin and soft tissue infections are common problems dealt with in emergency departments and
medical offices. It is routine practice to prescribe antibiotic agents after incision and drainage of cutaneous
abscesses. However, current evidence does not support prescribing oral antibiotic agents after surgical de-
bridement. The aim of the present study was to determine the actual role of antibiotic agents after drainage of
cutaneous abscesses.
Patients and Methods: This was a prospective study of patients undergoing incision and drainage (I&D) of a
subcutaneous abscess. Patients were randomly assigned either to receive antibiotic agents (group 1) or placebo
(group 2) after I&D. The primary end point was resolution rate of the abscess at the seventh day. Secondary end
points were pain at the seventh day and total time to full healing of the wound. P value <0.05 was considered
statistically significant.
Results: One hundred sixty-five patients were included for analysis. Age, gender, body mass index (BMI), and
comorbidities did not differ substantially between groups. Chest and peri-anal abscesses were statistically more
frequent in group 2, whereas neck abscesses were more frequent in group 1 (p = 0.02). Leukocyte count was also
statistically higher in group 1 (p = 0.005). Resolution rate was 96% in group 1 and 93% in group 2, with no
statistical difference between both (p = 0.28). Neither pain at seventh day nor time to full healing differed
statistically between groups.
Conclusions: Antibiotic agents are not necessary for uncomplicated subcutaneous abscesses after I&D. These
cases can be managed safely on an outpatient basis without any increase in morbidity.
Keywords: antibiotic agents; cutaneous abscess; incision and drainage; skin abscess; soft tissue infection
1
Department of Surgery, 2Emergency Department, Mexican Institute of Social Security, Delicias, Mexico.
3
Department of Surgery, Mexican Institute of Social Security, Chihuahua, Mexico.
345
346 LÓPEZ ET AL.
May 2016 at General Zone Hospital (HGZ) No. 11 of the Surgical procedure
Mexican Institute of Social Security in Delicias, Mexico
when authorization by the Local Ethical and Health Research Abscess sites were prepped with iodine povacrylex (0.7%
Committee was granted. The study complied with the poli- available iodine) and isopropyl alcohol, 74% w/w (Dur-
cies of Mexico’s General Law of Health Research, as well aPrep; 3M, St. Paul MN) or 0.8% iodopovidone (GER-
as the Declaration of Helsinki, and written informed consent MISIN Espuma; Farmacéuticos Altamirano de México S.A.
was obtained from all participants or legal representatives. De C.V., Ciudad De Mexico, Mexico), as available. Local
Eligible patients included only those older than 15 years anesthesia was achieved using 2% xylocaine, and a skin in-
affiliated with the institute, who were diagnosed with sub- cision was made with a scalpel blade number 21 over a length
cutaneous abscess and had not received antibiotic agents by of at least half the diameter of the abscess as estimated clini-
any route in the recent past (up to 15 days prior) to I&D. cally. The incision was deepened as necessary to access the
Patients with history of malignancy, hematologic diseases, abscess cavity. The content was then evacuated completely
human immunodeficiency virus/acquired immunodeficiency and the cavity debrided thoroughly with several iodopovidone-
syndrome (HIV/AIDS), splenectomy, necrotizing fasciitis, soaked gauze pads. Next, the volume of the abscess cavity was
Fournier gangrene, diabetic foot, toxic shock syndrome, estimated by infusing normal saline until the cavity was full.
extensive cellulitis 5 cm or more, abscesses with peritoneal Finally, the incision was packed with gauze soaked in 0.5%
fistulization, bone/joint involvement, severe circulatory hypochlorite solution and draped. Dressing changes twice
insufficiency, and patients likely to die in the short term daily were prescribed during the first seven days, with re-
were excluded. packing of the incision as described. On the first follow-up
visit, dressing changes were modified to once per day if an
Study design appropriately granulated incision bed was observed.
In this single-blinded, randomized, placebo-controlled Study end points
study, the patients admitted in the emergency department of
HGZ No. 11 with a diagnosis of subcutaneous abscess were The primary end point was the rate of resolution of the abscess
randomly assigned to receive or not to receive oral antibio- at the seventh day (first follow-up visit) in the two randomized
tic treatment after undergoing I&D. Two study arms were de- groups. Resolution of abscess was defined as the presence of
signed as follows. Patients in group 1 (study group) were given three or more of the following criteria: no local hyperthermia;
analgesics and antibiotic treatment after I&D, whereas patients no redness; no induration; and adequate granulation on incision
in group 2 (control) received analgesics and placebo. Analgesic bed. Resolution or failure to resolve at the seventh day was
treatment was provided by dipyrone 1 g and acetaminophen 1 g determined by a non-blinded researcher. Secondary end points
orally three times a day, as needed. Antibiotic coverage with were local pain on the seventh day and time to complete healing.
ciprofloxacin 250 mg orally twice daily for 7 days was given Local pain was evaluated by means of the visual analogue scale
to patients in group 1. Placebo pills were also prescribed every (VAS). Time to complete healing was defined as the time
12 hours for 7 days. elapsed from I&D to full epithelialization of the incision.
All medications were previously stored in plain bottles
Statistical analysis
labeled as ‘‘Analgesic 1,’’ ‘‘Analgesic 2,’’ ‘‘Antibiotic 1, and
‘‘Antibiotic 2,’’ with the latter being the placebo pills. All An analysis with intention-to-treat was done. Normality of
patients were discharged immediately after I&D and follow- data distribution was first determined by Kolmogorov-Smirnov
up visits were scheduled at 7, 14, and 21 days in the outpatient test. In normally distributed data, continuous variables appear
clinic. Patients were instructed to return to the emergency as mean – standard deviation (SD), and were analyzed for
department as soon if they noticed continued signs and homogeneity by using unpaired samples t-test, whereas non-
symptoms of ongoing infection after 48 hours of their initial normal continuous data are expressed as median – interquartile
treatment. Such signs and symptoms included malaise, in- range (IQR) and were tested for differences between two
tense local pain and hyperthermia, fever, induration, and groups by Mann-Whitney U test. Categorical variables are
redness (i.e., ongoing cellulitis). In such cases, the patients expressed as N with percentage and were tested with w2 or
were admitted to the surgical ward and received broad- Fisher exact test, as appropriate. Statistical significance was
spectrum antibiotic agents until culture/antibiogram results defined as p < 0.05. Statistical analyses were performed
were available to adjust to the specific drugs. using SPSS 20.0 for OSX (IBM SPSS Inc., Armonk, NY).
ABSCESS MANAGEMENT 347
FIG. 1. Flow diagram of patient selection. HIV/AIDS = human immunodeficiency virus/acquired immunodeficiency syndrome.
348 LÓPEZ ET AL.
because no pus was evacuated after I&D (i.e., cellulitis study group (group 1, n = 84; group 2, n = 81). Other demo-
without abscess), whereas in group 2, three patients were graphic and peri-operative data such as age, gender, BMI,
excluded for the same reason. Furthermore, seven patients comorbidities, and volume of the abscess were also compa-
were subsequently excluded in group 1 and five patients in rable between both groups (Table 1).
group 2 because they took additional non-prescribed medi- There was, however, statistically significant differences
cations in the forms of pills, creams, ointments, or dry with regard to abscess location (p = 0.02). Post hoc testing
powders in the recent past prior to their abscess treatment. revealed that chest and peri-anal abscesses were statistically
Thirteen patients did not attend the outpatient clinic for more frequent in the placebo arm, whereas neck abscesses
follow-up revision. A total of 165 patients remained suitable were more frequent in the antibiotic group. Leukocyte count
for outcome analysis (Fig. 1). also was found to be higher in patients assigned to group 1
after I&D (p = 0.005).
Patient characteristics
Patient demographics, peri-operative data, and abscess End points
Downloaded by National University of Singapore NUS SWETS/24793917 from www.liebertpub.com at 04/21/18. For personal use only.
characteristics are shown in Table 1. No difference was ob- Resolution rates of the infectious/inflammatory process
served regarding the number of patients allocated to each were not statistically different when testing both study arms
(p = 0.28). Resolution rate in group 1 was 96% (n = 81),
Table 1. Characteristics of Enrolled Patients whereas it was 93% (n = 75) in group 2 (placebo). Overall,
Group 1 Group 2 seven patients returned to the emergency department for
Characteristic (antibiotic) (placebo) p revision after 48 hours because they noticed no improve-
ment of the incision. After a thorough examination, no
N 84 81 signs of ongoing infection or worsening were found in any
Age (y) 0.7a of these patients, who were reassured and encouraged to
Mean – SD 49 – 17 48 – 14 continue with the treatment.
Range 19–91 21–73 Assessment of pain by VAS revealed no statistical dif-
Gender 0.36b ference at the seventh day, nor did the time to full healing of
Male 46 (55%) 50 (62%) the incision (Table 2). Eighty-eight percent of patients in
Female 38 (45%) 31 (38%) group 1 and 90% of patients in group 2 achieved full epi-
BMI (kg/m2) 0.55a thelialization by the 21-day follow-up visit.
Mean – SD 30 – 5 29 – 5
Range 21.8–43.6 20.5–53.3 Discussion
Comorbidities 0.67b
DM2 10 (12%) 10 (13%) A cutaneous abscess is defined as focal, confined purulent
SAH 18 (21%) 17 (21%) infection with a well-defined cavity surrounded by an in-
DM2 + SAH 16 (19%) 17 (21%) flammatory process involving deep subcutaneous tissues. On
DM2 + SAH + ESRD 0 2 ( 2%) physical examination, fluctuation suggests the presence of pus
None 40 (48%) 35 (43%) in the abscess cavity. Cellulitis, however, is a pyogenic in-
Abscess location 0.02b fection of the skin without an organized cavity and is typically
Face 18 (21%) 16 (20%) located in the epidermis, dermis, and shallow subcutaneous
Neck 4 ( 5%)c 0 tissues [14]. Because cellulitis constitutes the acute phase of an
Torso 2 ( 2%) 10 (12%)c SSTI, it is the only setting in which antibiotic agents are truly
Back 4 ( 5%) 2 ( 2%) indicated. An already developed abscess, however, is the result
Axila 2 ( 2%) 4 ( 5%)
Abdomen 6 ( 7%) 2 ( 2%)
Upper extremity 4 ( 5%) 6 ( 8%) Table 2. Outcome Parameters
Lower extremity 10 (12%) 4 ( 5%)
Inguinal 4 ( 5%) 1 ( 1%) Group 1 Group 2
Gluteal 10 (12%) 8 (11%) Parameter (antibiotic) (placebo) p
Peri-anal 8 (10%) 18 (22%)c N 84 81
Scrotal 2 ( 2%) 0
Surgical site 10 (12%) 10 (12%) Resolution at 7th day 0.28a
Yes 81 (96%) 75 (93%)
Leucocyte level 0.005a No 3 ( 4%) 6 ( 7%)
(103/mcL)
Mean – SD 13.9 – 3.7 12.3 – 3.4 Time to full healing 0.94a
Range 8.7–21.5 2.0–22.6 7 days 9 (11%) 10 (12%)
a 14 days 33 (39%) 30 (37%)
Abscess volume (mL) 0.85 21 days 32 (38%) 33 (41%)
Mean – SD 28 – 102 25 – 77 > 21 days 10 (12%) 8 (10%)
Range 2–685 5–480
VAS pain at 7th day 0.6b
a
Independent samples t-test. Mean – SD 1.6 – 0.9 1.7 – 0.9
b
Pearson w2 test. Range 1–4 1–4
c
Statistically different as shown by w2 post hoc testing.
a
SD = standard deviation; BMI = body mass index; DM2 = type 2 Pearson w2 test.
b
diabetes mellitus; SAH = systemic arterial hypertension; ESRD = end- Independent samples t-test.
stage renal disease. VAS = visual analogue scale; SD = standard deviation.
ABSCESS MANAGEMENT 349
of the body’s defense mechanisms. In addition, pus consists abscesses are amenable to I&D alone, whereas in cases of
of a build-up of serum, dead neutrophils, necrotic tissue, recurrent or refractory abscesses the presence of methicillin-
cholesterol and glucose. Most importantly, infective micro- resistant Staphylococcus aureus (MRSA) should be sus-
organisms do not always contribute to the pus composition. pected and treated accordingly [24]. Recent studies, however,
In this regard, we believe that there is a misunderstanding reported that the outcomes of SSTIs treated with antibio-
of the natural process of the development of an abscess, tic agents were much the same regardless the activity of the
leading physicians to believe that the presence of pus within prescribed drug against the infecting organism as confirmed
subcutaneous tissues constitutes the infectious material itself, by culture/antibiogram [25,26]. When cultures of patients
and consequently, to administer antibiotic agents. Also, we with SSTIs reported the presence of MRSA, all of those in-
have noted that medical staff at our institution prefer to fections improved even when b-lactamic antibiotic agents,
‘‘prevent’’ by admitting and prescribing broad-spectrum ineffective against this micro-organism, were administered
antibiotic agents to patients with peri-anal abscesses or sur- [6,27]. However, in a trial involving 1,265 patients with a
gical site infections (SSIs) because they are concerned about drained cutaneous abscess, Talan et al. [28] found that pa-
the medical and legal consequences that may arise should tients who received trimethoprim-sufamethoxazole (at doses
Downloaded by National University of Singapore NUS SWETS/24793917 from www.liebertpub.com at 04/21/18. For personal use only.
those infections progress to Fournier gangrene or necrotizing of 320 and 1600 mg, respectively, twice daily, for 7 d) had a
fasciitis. According to the results of this study, however, this higher cure rate than those who received placebo. The re-
premise is not only unnecessary but also exaggerated because searchers also found that many secondary outcomes were
we managed every single patient on an outpatient basis and better in the trimethoprim-sulfamethoxazole group than
observed no adverse outcomes (Fig. 2). in the placebo group, including fewer subsequent surgical
Although simple cases of cutaneous cellulitis and ab- drainage procedures, new skin infections, and infections
scesses are caused by gram-positive pathogens, it is common among household members six to eight weeks after the end
for physicians to use broad-spectrum antibiotic agents against of the treatment period.
gram-negative and anaerobic micro-organisms [15–18]. Med- Our study challenges the current recommendations of the
ical staff must remain aware that exposing patients to unnec- Clinical Practice Guidelines of Mexican Institute of Social
essary, lengthy, and unwarranted broad-spectrum antibiotic Security (IMSS-074-08) for the medical management of
treatments is unacceptable in the current era of progressive SSTIs and SSIs. Both guidelines endorse the administration
antimicrobial resistance [19–21] as well as the incidence and of a variety of oral/intravenous antibiotic agents for such
severity of infection by Clostridium difficile [22,23]. types of infections [29]. The results of this study may well
Current guidelines of the American Society of Infectious serve for future updates in the aforementioned guidelines.
Diseases (ASID) and the U.S. Centers for Disease Control In this study we sought to determine any possible role of
and Prevention (CDC) state that uncomplicated cutaneous antibiotic agents as adjunct treatment in the management of
FIG. 2. Image depicting resolution of infective process in a patient allocated to the placebo arm after incision and
drainage (I&D). (A) A large amount of pus was drained by I&D. (B) Incision packing with hypochlorite-moistured gauze.
(C) Incision bed appearance by seventh day. (D) Incision bed by 21st day. Note: because of the size of the surgical site, it
did not heal completely by 21st day. It did, however, resolve infective process, as no signs of ongoing infection/cellulitis
were seen by the seventh day (C).
350 LÓPEZ ET AL.
subcutaneous abscesses. No such a role could be identified. listed have agreed to submit in present form and declare that
Patients undergoing I&D resolved their infectious process at there are no conflicts of interests.
a similar rate and time whether or not antibiotic agents were
prescribed. This was true even in patients with diabetes Author Disclosure Statement
mellitus and those older than 70 years. We also wanted to
determine whether antibiotic agents help decrease pain or No competing financial interests exist.
accelerate healing by acting synergistically with the im-
mune system. Such benefits were also ruled out according to References
our results.
1. Frazee BW, Lynn J, Charlesbois ED, et al. High prevalence
We have to recognize that ciprofloxacin was not the most
of methicillin-resistant Staphylococcus aureus in emer-
appropriate choice, even though it has activity against gram- gency department skin and soft tissue infections. Ann
positive as well as gram-negative bacteria and one (not the Emerg Med 2005;45:311–320.
preferred) of their indications is soft tissue infections. As we 2. Hersh AL, Chambers HF, Maselli JH, et al. National
mentioned in the introduction, emergency department phy-
Downloaded by National University of Singapore NUS SWETS/24793917 from www.liebertpub.com at 04/21/18. For personal use only.
soft-tissue infections: Impact of antimicrobial therapy on 25. Fridkin SK, Hageman J, Morrison M, Sanza T, et al.
outcome. Clin Infect Dis 2007;44:777–784. Methicillin-resistant Staphylococcus aureus disease in three
18. Catarrala J, Roson B, Fernandez-Sabe N, et al. Factors communities. N Engl J Med 2005;352:1436–1444.
associated with complications and mortality in adult pa- 26. Lee MC, Rios AM, Aten MF, et al. Management and out-
tients hospitalized for infectious cellulitis. Eur J Clin come of children with skin and soft tissue abscesses caused
Microbiol Infect Dis 2003;22:151–157. by community-acquired methicillin-resistant Staphylococcus
19. Johnson L, Sabel A, Burman WJ, et al. Emergence of aureus. Pediatr Infect Dis 2004;23:123–127.
fluoroquinolone resistance in outpatient urinary Escher- 27. Young DM, Harris HW, Charlebois ED, et al. An epidemic
ichia coli isolates. Am J Med 2008;121:876–884. of methicillin-resistant Staphylococcus aureus soft tissue
20. Lautenbach E, Synnestvedt M, Weiner MG, et al. infections among medically underserved patients. Arch
Imipenem resistance in Pseudomonas aeruginos: Emer- Surg 2004;139:947–951.
gence, epidemiology and impact on clinical and eco- 28. Talan DA, Mower WR, Krishnadasan A, et al. Trimethoprim-
nomic outcomes. Infect Control Hosp Epidemiol 2010; sulfamethoxazole versus placebo for uncomplicated skin ab-
31:47–53. scess. N Engl J Med 2016;374:823–832.
Downloaded by National University of Singapore NUS SWETS/24793917 from www.liebertpub.com at 04/21/18. For personal use only.
21. Chen DK, McGeer A, de Azavedo JC, et al. Decreased sus- 29. www.cenetec.salud.gob.mx/descargas/gpc/CatalogoMaestro/
ceptibility of Streptococcus pneumoniae to fluoroquinolones 074_GPC_Fascitisnecrosante/Fascitis_necrosante_R_CENE
in Canada. N Engl J Med 1999;34:233–239. TEC.pdf (Last accessed January 27, 2018).
22. Dallal RM, Harbrecht BG Boujoukas AJ, et al. Fulminant 30. Korownyk C, Allan GM. Evidence-based approach to abscess
Clostridium difficile: An underappreciated and increasing management. Can Fam Physician 2007; 53:1680–1684.
cause of death and complications. Ann Surg 2002;235;
363–372. Address correspondence to:
23. McDonald LC, Owings M, Jernigan DB. Clostridium Dr. Julio López
difficile infection in patients discharged from US short- Department of Surgery
stay hospitals, 1996–2003. Emerg Infect Dis 2006;12: HGZ/MF 11 Mexican Institute of Social Security
409–415. Avenida Rio Conchos Poniente y 7a. Poniente S/N
24. Stevens DL, Bisno AL, Chambers HF, et al. Practice Delicias, Chihuahua 33000
guidelines for the diagnosis and management of skin and Mexico
soft-tissue infections. Clin Infect Dis 2005;41:1373–
1406. E-mail: juliocesar1701@icloud.com