RESUMO
Background: The need for oral, cost-effective treatment for complicated skin and skin structure infections (cSSSIs) due to methicillin-resistant Staphylococcus aureus (MRSA) was addressed by the non-inferiority comparisons of oral minocycline plus rifampicin with linezolid. Methods: In the AIDA multicenter, open label, randomized, controlled clinical trial, hospitalized adults with cSSSI and documented MRSA were randomly assigned at a 2:1 ratio to either oral 600 mg rifampicin qd plus 100 mg minocycline bid or oral 600 mg linezolid bid for 10 days. The primary endpoint was the clinical cure rate in the clinically evaluable (CE) population at the test-of-cure visit (14 days). Non-inferiority was confirmed if the lower confidence limit (CI) did not fall below the accepted error margin of 15%. The study is registered with EudraCT number 2014-001276-56. Findings: 123 patients recruited between November 2014 and January 2017 were randomly assigned to treatment (81 patients to minocycline plus rifampicin and 42 patients to linezolid). Cure rates were 78.% (46/59, 90% CI 67.3-86.5) and 68.6% (24/35, 90% CI 53.4-81.3), respectively (P = 0.337). The percent difference in cure rates was 9.4% (90% CI -7.2 to 26.8%). Minocycline plus rifampicin combination was deemed non-inferior to linezolid as the lower CI was -7.2% i.e. smaller than the accepted error margin of -15%. Although statistically not significant, the overall rate of adverse events was higher in the linezolid group (47.6%, 20/42 versus 38.3%, 31/81). Interpretation: Oral minocycline plus rifampicin was non-inferior to oral linezolid treatment providing alternative oral treatment for cSSSI. Funding: The EU Seventh Research Framework Programme.
RESUMO
Although clinical definitions of acute bacterial skin and skin-structure infection (ABSSSI) are now well established, guidance of the prediction of likely pathogens based on evidence is missing. This was a large survey of the microbiology of ABSSSIs in Greece. During the period November 2014 to December 2016, all admissions for ABSSSI in 16 departments of internal medicine or surgery in Greece were screened to determine the likely bacterial aetiology. Samples were cultured on conventional media. Expression of the SA442, mecA/mecC and SCCmec-orfX junction genes was assessed. Following univariate and forward logistic regression analysis, clinical characteristics were used to develop scores to predict the likely pathogen with a target of 90% specificity. In total, 1027 patients were screened and 633 had positive microbiology. Monomicrobial infection by Gram-positive cocci occurred in 52.1% and by Gram-negative bacteria in 20.5%, and mixed infection by Gram-positive cocci and Gram-negative bacteria in 27.3%. The most common isolated pathogens were Staphylococcus aureus and coagulase-negative staphylococci. Resistance to methicillin was 57.3% (53.5-61.1%). Three predictive scores were developed: one for infection by methicillin-resistant S. aureus, incorporating recent hospitalisation, atrial fibrillation, residency in long-term care facility (LTCF) and stroke; one for mixed Gram-positive and Gram-negative infections, incorporating localisation of ABSSSI in lumbar area, fluoroquinolone intake in last 6 days, residency in LTCF and stroke; and another for Gram-negative infection, incorporating skin ulcer presentation, peptic ulcer and solid tumour malignancy. In conclusion, methicillin-resistant staphylococci are the main pathogens of ABSSSIs. The scores developed may help to predict the likely pathogen.