Common bacterial tests fail to distinguish infection from colonization in wound care
A new study finds that standard laboratory tests cannot reliably determine whether bacteria detected in skin ulcers represent dangerous infections requiring antibiotics or harmless colonization. The finding raises concerns about unnecessary antibiotic prescribing and resistance, with major implications for wound care protocols and clinical decision-making in emergency departments.
Originaltitel: Diagnostic utility of conventional culture and nucleic acid amplification tests to detect Staphylococcus aureus and Streptococcus pyogenes for the inclusion or exclusion of infection in skin ulcers: An etiologic predictive value analysis.
Bakteriedetektion i hudgesvinningar kan särskilja mellan infektion och kolonisering — en central utmaning för antimikrobiell sparsamhet. Forskare från Sahlgrenska Academy analyserade hudprov från 387 akutmottagningspatienter med konventionell odling och nukleinsyramplifieringstester (NAAT) för att bedöma diagnostisk värde. Studien jämförde resultat mellan patienter med och utan hudgesvinningar. NAAT-tester och odling visade begränsad förmåga att bekräfta infektion — positiva prediktiva värden hade breda konfidensintervall. Däremot uppnådde negativa prediktiva värden 97–100 procent med snäva intervall, vilket indikerar höggradig säkerhet vid uteslutning av infektion. För läkare och inköpschefer är slutsatsen klinisk: dessa tester är mest värdefulla för att *undvika* onödig antibiotika när tecken på övertlig infektion saknas — inte för att bekräfta den. Detta stödjer målriktad terapi och resurseffektivitet inom regionvården.
BACKGROUND: Pathogenic bacteria such as Streptococcus pyogenes, Staphylococcus aureus, and methicillin-resistant S. aureus (MRSA) are frequently isolated from skin ulcers. Their presence may signify colonisation, subclinical (covert) infection, or overt clinical infection. While antibiotic therapy is indicated for infection, it is inappropriate for simple colonisation. However, the clinical distinction between these states often lacks a clear-cut demarcation, representing a significant diagnostic challenge. METHODS: This study utilised Etiologic Predictive Value (EPV) to estimate the probability that the presence of these bacteria represents clinical infection rather than colonisation. Skin swabs from 387 patients presenting to an emergency department were analysed using conventional culture and nucleic acid amplification tests (NAAT). Bacterial growth and detection rates were compared between patients with (n = 70) and without (n = 317) skin ulcers to calculate the EPV. RESULTS: The positive EPV for bacterial detection yielded wide 95% confidence intervals (CIs), suggesting that these microbiological tests have limited utility for confirming (ruling in) clinical infection. Conversely, the negative EPV was 97-100% with very narrow CIs, indicating that these tests are highly effective for excluding (ruling out) subclinical or overt infection. CONCLUSIONS: We conclude that conventional culture and, more notably, NAAT are valuable tools for antimicrobial stewardship. They are particularly useful for excluding infection and preventing unnecessary antibiotic prescription in cases where compelling clinical signs of overt infection are absent.