COX-2 fluorescent probe as an adjunctive quantitative tool for assessing inflammation severity in bacterial vaginosis
The COX-2 fluorescent probe detected significantly higher signals in murine and human BV samples compared to controls, with statistical correlation to multiple microbiological parameters, but no sensitivity, specificity, or clinical cut-off values were reported.
| Population | C57BL/6J mice in a murine BV model (preclinical arm) and 45 female outpatients classified as Normal or BV by microecological criteria (clinical arm) |
|---|---|
| Exposure | Measurement of COX-2-targeting nano-probe fluorescence intensity in vaginal lavage (murine) and vaginal secretions (human) |
| Comparator | Sham control mice (murine arm) and Normal group without BV (clinical arm) |
| Outcome | COX-2 fluorescence in murine samples (BV vs. control); COX-2 fluorescence in human patients (BV vs. Normal); Correlation with Nugent score; Correlation with clue cells; IL-1β and IL-10 levels in murine model; Correlation with vaginal microbiota type; Diagnostic accuracy (sensitivity/specificity/AUC) |
Summary of findings
| Outcome | Effect | 95% CI | Certainty | Clinical relevance | Notes |
|---|---|---|---|---|---|
| COX-2 fluorescence in murine samples (BV vs. control) | p < 0.05, in the effect size or CI reported | — | Very low | — | 1 studies |
| COX-2 fluorescence in human patients (BV vs. Normal) | p < 0.05, in the effect size or CI reported | — | Low | — | 1 studies |
| Correlation with Nugent score | p < 0.05, correlation coefficient not reported | — | Low | — | 1 studies |
| Correlation with clue cells | p < 0.05, correlation coefficient not reported | — | Low | — | 1 studies |
| IL-1β and IL-10 levels in murine model | p < 0.05, in the absolute values or CI reported | — | Very low | — | 1 studies |
| Correlation with vaginal microbiota type | p < 0.05, in the effect size or CI reported | — | Low | — | 1 studies |
| Diagnostic accuracy (sensitivity/specificity/AUC) | not reported | — | Very low | — | 1 studies |
Context
Bacterial vaginosis (BV) affects up to 40% of women of reproductive age and is associated with adverse reproductive outcomes. Current diagnostic standards—Amsel criteria and Nugent score—are subjective and cannot quantify inflammatory severity. An objective inflammatory severity tool could complement existing diagnostic workflows.
What the study showed
In the murine model, BV mice showed elevated IL-1β and reduced IL-10 (p < 0.05), with significantly higher COX-2 fluorescence than controls (p < 0.05). In the clinical arm (n = 45), the BV group showed higher COX-2 fluorescence signals than the Normal group (p < 0.05). Fluorescence intensity correlated with vaginal cleanliness, Nugent score, microbiota type, clue cells, and bacterial density (all p < 0.05). No absolute fluorescence values, 95% CIs, effect sizes (OR, RR, SMD), or diagnostic cut-off points were reported in the available text.
How it was done
Hybrid study combining a preclinical arm (murine BV model using vaginally infected, estrogen-pretreated C57BL/6J mice) and a cross-sectional observational clinical arm with 45 outpatients. Patients were classified into Normal or BV groups by microecological criteria without reported randomization or blinding. Follow-up duration was not specified in the available extract.
Effect magnitude
The study reports only p < 0.05 for all primary comparisons. No effect size estimates (SMD, OR, RR), 95% CIs, or absolute fluorescence values were provided in the available extract, precluding assessment of clinical magnitude.
Risk of bias
Small clinical sample (n = 45) with no reported power calculation; absence of evaluator blinding and no risk-of-bias tool applied (e.g., RoB 2, ROBINS-I); no ROC curve, sensitivity, specificity, or cut-off reported; murine model does not fully replicate human vaginal physiology; the available text is a structured extract, not the full article with detailed methods.
What this study does NOT prove
This study does not prove that the COX-2 probe offers superior or complementary diagnostic accuracy over current standards, nor that inflammatory quantification improves clinical outcomes. Statistical correlation with microbiological parameters does not equal validated diagnostic utility.
In clinical practice
No change in clinical practice is justified based on this study. Clinicians should maintain Amsel criteria and Nugent score as standard diagnostic tools. The COX-2 probe remains in an exploratory phase without sufficient clinical validation for adoption.
Limitations
Small clinical sample (n = 45) with no reported power calculation; absence of evaluator blinding and no risk-of-bias tool applied (e.g., RoB 2, ROBINS-I); no ROC curve, sensitivity, specificity, or cut-off reported; murine model does not fully replicate human vaginal physiology; the available text is a structured extract, not the full article with detailed methods.
What is still missing
Larger prospective controlled studies to determine sensitivity, specificity, and cut-off values of the COX-2 probe against the gold standard (Nugent score). Validation in diverse populations including pregnant women and postmenopausal women.
Technical appendix
Version history
- 1.0 · 2026-09-19 — Auto-generated under Evidence Standard v1.0