Trimethoprim-sulfamethoxazole for uncomplicated skin abscess

in

Talan DA, et al. Trimethoprim-sulfamethoxazole versus placebo for uncomplicated skin abscess. N Engl J Med 2016;374:823-32.

 

Clinical Question

In patients with an uncomplicated abscess treated with incision & drainage (I&D), does a 1-week course of trimethoprim-sulfamethoxazole (TMP-SMX) increase the likelihood of abscess cure?

 

Bottom Line

In a population of patients with first uncomplicated abscess and a ~50% prevalence of MRSA, addition of TMP-SMX for 1 week to I&D increased the chance of clinical cure from ~73% to ~80% without any clinically-important adverse effects. 

 

Integrating This Study Into Practice

For patients who would meet this study's eligibility criteria, consider having a shared decision-making discussion with the following options:

  1. I&D alone (no antibiotics): ~75% chance of cure @ 1-2 weeks
  2. I&D + antibiotic x7 days*:
    • Increases chance of cure @ 1-2 weeks to 80%
    • Generally well-tolerated, but could cause headache, GI discomfort, rash in 10-20% of people who take it (estimates based on overall studies)
    • <1/10,000 risk of serious reaction or allergy (e.g. anaphylaxis, Stevens-Johnson Syndrome, DRESS)
    • Costs $0-10 depending on patient's drug plan

*In my opinion, clindamycin, doxycycline & TMP-SMX are all reasonable PO options here despite this study only evaluating TMP-SMX. The choice between these agents should be based on previous allergies/intolerances & interactions with concomitant medications.

 

Design

Allocation-concealed RCT with "everybody" blinded, low loss-to-follow-up (4%), analyzed using a modified intention-to-treat population (mITT).

 

Patients and Setting

  • USA, 5 emergency departments
  • Enrollment timeline: April 2009 - April 2013
  • Inclusion criteria:
    1. >12 y/o
    2. Suspected cutaneous abscess
      • Based on physical exam or ultrasound
      • Found to have purulent material on surgical exploration
    3. Lesion characteristics
      • Present for <1 week
      • Diameter 2+ cm
    4. Expected outpatient management
  • Exclusion criteria:
    • "Complicated" abscess:
      • Indwelling device
      • Suspected osteomyelitis or septic arthritis
      • Diabetic foot, decubitus, or ischemic ulcer
      • Mammalian bite
      • Wound with organic foreign body
      • Infection of another organ system/site
      • Perirectal, perineal or paronychial location
    • PMHx:
      • IVDU within previous month + fever
      • Immunodeficiency (absolute neutrophil count <500/mm^3, on immunosuppressive drugs, active chemotherapy, known AIDS assessed by history)
      • Cardiac condition with risk of endocarditis (e.g. prosthetic valve)
      • CrCl <50 mL/min
      • Allergy/intolerance to trimethoprim-sulfamethoxazole
      • Known G6PD or folate deficiency
      • Underlying skin condition
    • SHx
      • Long-term care residence
      • Incarceration
    • Concurrent meds:
      • Trimethoprim-sulfamethoxazole within 24h
      • Other topical or systemic antibiotic
      • Methotrexate
      • Phenytoin
      • Warfarin
  • 1308 screened -> 1265 randomized -> 1247 analyzed in primary analysis (modified intention-to-treat population 1)
  • Average patient:
    • 35 y/o
    • 42% female
    • PMHx
      • Diabetes 11%
      • Previous MRSA infection 7-8%
    • Clinical presentation
      • 4 days with symptoms before I&D
      • Fever within 1 week of enrollment 18%
      • Abscess location: Head/neck 14%, trunk/abdo/back 21%, groin/buttocks 20%, arms/hands 24%, lower extremities 22%
      • Abscess dimensions: Median 2.5 cm x 2 cm x 1.5 cm (depth)
      • Erythema dimensions: median 7 cm x 5 cm
      • Vital signs
        • Temp >38 degrees Celsius - 0.8%
        • HR >90 - 25%
        • RR >20 - <3%
    • Wound culture result
      1. MRSA ~45% (96% of which were Panton-Valentine leukocidin [PVL]-positive)
      2. MSSA ~15%
      3. Coagulase-negative Staph ~10-15%
      4. Streptococcus ~5%
      5. Other 10-15%

 

Intervention and Control

  • Intervention: TMP-SMX 2 double-strength (DS) tabkets (total 320/1600 mg) PO BID x7 days
    • Adherence: 2/3 took all tablets
  • Control: Matching placebo
  • Co-interventions common to both groups:
    • Incision & drainage of the abscess (standardized training on technique)
    • Culture & susceptibility of drainage sample

 

Outcomes

  • Death: 1 in each group (unrelated to study intervention)
  • Efficacy: 
    • Clinical cure (not meeting clinical failure criteria 1-3) @ day 7-14 (primary outcome): TMP-SMX 80.5%, placebo 73.6% (p=0.005)
      • Statistically significant 6.9% higher probability of clinical cure with TMP-SMX versus placebo (number needed to treat = 15)
      • Clinical failure defined as any of the criteria below:
        1. Study withdrawal, lost-to-follow-up, or missing outcomes
        2. Worsening @ day 3-4: Fever attributable to the infection, increase in erythema >25% from baseline, worsening of wound swelling/tenderness
        3. No improvement @ day 8-10: Fever, no decrease in erythema from baseline, or no decrease in swelling or tenderness
        4. No resolution @ day 14-21: fever or more than minimal erythema, swelling or tenderness
      • Results consistent in analyses of more selective subpopulations (per-protocol and a second miTT)
    • Clinical response (so-called FDA guidance early endpoint) @ day 2-3
      • TMP-SMX 36.3%, placebo 33.7% (p=0.38)
      • Definition: Temp 37.7 degrees Celsius + decrease/no increase in erythema from baseline + no worsening in swelling/induration
    • No statistically significant differences in: Hospitalization, recurrent skin infection, presence of swelling or induration, change in mean area of erythema from baseline, days missed from normal activities or work/school, or days of use of analgesics
  • Safety
    • Discontinuation due to adverse events: TMP-SMX 1.9% versus placebo 0.6%, not statistically significant different
    • Overall adverse events: TMP-SMX 65.4%, placebo 65.2%, not statistically significant different
      • GI-related: TMP-SMX 42.7% versus 36.1%
    • C. difficile infection: 0 in both groups

 

Generalizability & Application

  • Patients/Intervention: 
    • Had no systemic signs of infection (SIRS) or shock
    • High prevalence of community-acquired MRSA (almost half of all positive cultures), for which other PO alternatives include clindamycin or doxycycline.
  • Outcomes
    • Must balance discordance between
      • Statistically & clinically significant improvement in clinical cure at 1-2 weeks (primary outcome)
      • No difference in response @ day 2-3, in speed of resolution of erythema, or in days to return to normal activities
    • Secondary outcomes reported only for per-protocol population, which introduces attrition bias, & therefore any differences between groups are unreliable.

Summary author: Ricky Turgeon BSc(Pharm), ACPR, PharmD
Summary date: 10 May 2016