Effect of Cephalexin Plus Trimethoprim-Sulfamethoxazole vs Cephalexin Alone on Clinical Cure of Unco
“Effect of Cephalexin Plus Trimethoprim-Sulfamethoxazole vs Cephalexin Alone on Clinical Cure of Uncomplicated Cellulitis: A Randomized Clinical Trial” JAMA. 2017;317(20):2088-2096.
Skin infections are common complaints in emergency departments in the United States, especially with the rise of MRSA infections, considered the most common cause of purulent infections. In contrast, cellulitis without evidence of purulent drainage is predominantly caused by beta-hemolytic streptococci. The Infectious Disease Society of America practice guideline recommends treating uncomplicated cellulitis with anti-microbial coverage against only streptococci. However, providers commonly prescribe antibiotics to also cover for MRSA such as Trimethoprim-Sulfamethoxazole. This study aimed to compare clinical cure rates of uncomplicated cellulitis with Cephalexin plus Trimothoprim-Sulfamethoxazole vs Cephalexin alone.
This study was a multi-center, double-blind, superiority trial from April 2009 to June 2012 at five emergency departments in the United States. Participants were enrolled if lesions had been present for less than one week and had a minimum diameter of two centimeters. Exclusion criteria included underlying skin conditions, history of intravenous drug use with fever, concurrent infection at another site, and immunosuppression. Participants were blinded into two groups, each with n= 248, either receiving Cephalexin 500mg QID with Trimothoprim-Sulfamethoxazole 320mg/1600mg BID for seven days or Cephalexin 500mg QID with placebo for seven days. This study’s primary outcome assessed clinical cure with presence and degree of fever, erythema, swelling, and tenderness at days 3-4 (during therapy- allowed up to 25% of erythema without systemic signs), days 8-10 (end of therapy- improvement from initial), days 14-21 (test of clinical cure- minimal to no erythema), and days 49-63 (extended follow up after treatment). Secondary outcomes included overnight hospitalizations, recurrent skin infections, and similar infection in household contacts.
The data showed that Cephalexin alone showed no difference in rates of clinical resolution compared to Cephalexin and Trimothoprim-Sulfamethoxazole in the per-protocol group, however analysis in the modified intention-to-treat group favored use of both antibiotics. The per-protocol group, the primary outcome population, included those who adhered to ≥ 75% of study medications and physical follow up at 14-21 days after enrollment. The researchers considered this group to be more accurate than the modified intention-to-treat group which included those who adhered to ≥ 1 dose of study medication with in-person or telephone assessment at test of cure. In this group, losses to follow up and missing or unassigned outcomes were considered clinical failure in the analysis.
The per-protocol analysis showed a clinical cure of 83.5% (both antibiotics) vs 85.5% (Cephalexin alone) with a confidence interval of -9.7 to 5.7%, showing no difference in rates of clinical resolution. The modified intention-to-treat group showed a clinical cure of 76.2% (both antibiotics) vs 69% (Cephalexin alone) with a confidence interval of -1 to 15.5% which included the minimum clinical importance difference of 10%, determined by consensus, which favors the combination of both antibiotics. Failure rates for both groups were around 15% in the per protocol analysis, moreover, both groups had similar proportions of patients who developed MRSA during the follow up period.
Secondary outcomes revealed no statistical difference between Cephlexin and Trimothoprim-Sulfamethoxazole vs Cephalexin alone. Sub-group analysis also showed similar resolution of fever in both groups (82.6% vs 80%) and similar cure rates in diabetics (84% vs 95%) and in those with an area of erythema > 75 cm2 (82% vs 85.7%). Adverse events were also not significantly different in both groups, 90% of which were graded as “mild”. Failure of treatment were also similar with equal proportions of MRSA found among both treatment groups.
Clinical cure rates for the treatment of uncomplicated cellulitis are similar between Cephalexin alone vs Cephalexin and Trimothoprim-Sulfamethoxazole, based on the per-protocol analysis
This study was unable to definitively show that Cephalexin is superior or equal to the combination of antibiotics by the intention-to-treat analysis
Cellulitis may be caused by both streptococci and staphylococcus
Both groups showed failure rates of around 15% with equal proportions of the development in MRSA in both treatment groups
Sub-group analysis (diabetics, those with fever, those with large areas of erythema) had similar clinical cure rates
Both groups showed similar secondary outcomes, including hospitalization, recurrent skin infection, infections in household contacts
A difference of 2-3% clinical cure rate may not be worth adding a second antibiotic to the treatment of uncomplicated cellulitis