Cochrane · DTA · CD010276
Diagnostic tests for oral cancer and potentially malignant disorders in patients presenting with clinically evident lesions
Walsh, T, Macey, R, Kerr, AR, et al. · Cochrane Database of Systematic Reviews · 2021 · Cochrane Library
Full textEvery question was extracted from the full review (methods and results), not only its Summary of Findings table.
Questions
9
current version
GRADE rated
8
No studies
1
Included studies
63
as reported
No studies 1 · 11.1%Ungraded 0 · 0%Very low 4 · 44.4%Low 2 · 22.2%Moderate 2 · 22.2%High 0 · 0%
Ungraded means an effect was reported without a GRADE rating; it is not a gap. No studies means the review found nothing to answer the question. The six categories are a display order, not a score.
Review question
| Element | Planned (methods) | Found (results) |
|---|---|---|
| Population | Adults (aged 16 years or over) presenting to primary or secondary care with clinically evident suspicious or innocuous oral lesions | 7942 lesions from 63 studies (India 17, Germany 8, US 7, Australia 3, Brazil 2, Canada 1, China 4, Iran 2, Italy 4, Japan 2, Poland 1, Spain 2, Sri Lanka 3, Taiwan 2, Turkey 1, UK 4); published 1980 to 2020; all in secondary care (hospital) settings |
| Intervention | Index tests used alone or in combination as an adjunct to the conventional oral examination (COE): Conventional oral examination (COE), Vital staining (e.g. toluidine blue, tolonium chloride), Brush cytology (e.g. OralCDx brush biopsy), Light-based detection (chemiluminescence e.g. ViziLite, ViziLite plus, Microlux DL; tissue fluorescence imaging e.g. VELscope, Identafi 3000; tissue fluorescence spectroscopy), Blood and saliva analysis | Vital staining (n=20); Oral cytology (n=20); Light-based detection (n=23); Combined tests (n=9); Blood and saliva analysis (n=0) |
| Comparison | Surgical or punch biopsy with histological assessment | Surgical or punch biopsy with histological examination (n=63) |
| Outcomes | Diagnostic accuracy (sensitivity and specificity) for detection of the target conditions - Oral cancer: oral cavity squamous cell carcinoma (OSCC); Oral potentially malignant disorders (OPMD): leukoplakia, proliferative verrucous leukoplakia, erythroplakia, lupus erythematosus, submucous fibrosis, hereditary disorders (dyskeratosis congenita, epidermolysis bullosa); test quality via QUADAS-2. Any level of dysplasia (mild, moderate or severe) or OSCC classified as the target conditions. | Sensitivity and specificity for detection of oral cancer and potentially malignant disorders, with GRADE certainty (n=63) |
| Study designs | Cross-sectional diagnostic test accuracy studies (or consecutive series) and randomised studies of diagnostic test accuracy. Excluded: studies collecting data retrospectively, studies reported in abstract form alone, case-control studies, uncontrolled reports and RCTs of screening programme effectiveness. | Cross-sectional diagnostic test accuracy studies (n=63; 79 datasets) |
| Timing | Not specified | Studies published between 1980 and 2020 |
Search date: 2020-10-20. Databases: MEDLINE Ovid;Embase Ovid;US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov;World Health Organization International Clinical Trials Registry Platform
Questions and results
Show all 3 versionsVital staining (e.g. toluidine blue, tolonium chloride) vs Surgical or punch biopsy with histological assessment
Population: Adults (aged 16 years or over) presenting to primary or secondary care with clinically evident suspicious or innocuous oral lesions
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Specificity | Specificity 0.68 [0.58, 0.77] 312 [224, 409] false positives per 1000 at 2.5% pre-test probability | 20 studies | Very low Downgraded: risk of bias, imprecision, inconsistency |
Sensitivity | Sensitivity 0.86 [0.79, 0.90] 3 [2, 5] false negatives (missed lesions) per 1000 at 2.5% pre-test probability | 20 studies | Low Downgraded: risk of bias, inconsistency |
Oral cytology (e.g. brush cytology, OralCDx brush biopsy) vs Surgical or punch biopsy with histological assessment
Population: Adults (aged 16 years or over) presenting to primary or secondary care with clinically evident suspicious or innocuous oral lesions
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Sensitivity | Sensitivity 0.90 [0.82, 0.94] 2 [1, 4] false negatives (missed lesions) per 1000 at 2.5% pre-test probability | 20 studies | Moderate Downgraded: risk of bias |
Specificity | Specificity 0.94 [0.88, 0.97] 58 [29, 117] false positives per 1000 at 2.5% pre-test probability | 20 studies | Moderate Downgraded: risk of bias |
Light-based detection and/or oral spectroscopy (e.g. ViziLite, VELscope, Microlux DL, Identafi 3000) vs Surgical or punch biopsy with histological assessment
Population: Adults (aged 16 years or over) presenting to primary or secondary care with clinically evident suspicious or innocuous oral lesions
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Specificity | Specificity 0.50 [0.32, 0.68] 487 [312, 663] false positives per 1000 at 2.5% pre-test probability | 23 studies | Very low Downgraded: risk of bias, imprecision, inconsistency |
Sensitivity | Sensitivity 0.87 [0.78, 0.93] 3 [2, 5] false negatives (missed lesions) per 1000 at 2.5% pre-test probability | 23 studies | Low Downgraded: risk of bias, inconsistency |
Blood and saliva analysis (biomarker tests) vs Surgical or punch biopsy with histological assessment
Population: Adults (aged 16 years or over) presenting to primary or secondary care with clinically evident suspicious or innocuous oral lesions
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Outcome NR | No eligible studies assessed the diagnostic test accuracy of blood and saliva analysis; potentially eligible biomarker studies were excluded as case-control designs. | 0 studies | No studies |
Combined index tests (e.g. vital staining plus light-based detection or cytology) vs Surgical or punch biopsy with histological assessment
Population: Adults (aged 16 years or over) presenting to primary or secondary care with clinically evident suspicious or innocuous oral lesions
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Specificity | Specificity 0.71 [0.53, 0.84] 283 [156, 458] false positives per 1000 at 2.5% pre-test probability | 9 studies | Very low Downgraded: risk of bias, imprecision, inconsistency |
Sensitivity | Sensitivity 0.78 [0.45, 0.94] 5 [1, 14] false negatives (missed lesions) per 1000 at 2.5% pre-test probability | 9 studies | Very low Downgraded: risk of bias, imprecision, inconsistency |