Non-Cochrane · DTA · PM34636266
Enamel Caries Detection and Diagnosis: An Analysis of Systematic Reviews.
Walsh T, Macey R, Ricketts D, et al. · Journal of dental research · 2022 · PubMed · doi:10.1177/00220345211042795
Full textEvery question was extracted from the full text, figures and supplements by two independent LLM extractors, disagreements adjudicated. Rows cite where in the review the value comes from.
Questions
42
current version
GRADE rated
10
No studies
0
Included studies
64
as reported
No studies 0 · 0%Ungraded 32 · 76.2%Very low 0 · 0%Low 10 · 23.8%Moderate 0 · 0%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 | Children or adults presenting asymptomatically or suspected of having enamel caries (clinical studies); extracted teeth (in vitro studies); dentine and frank cavitations excluded | Network meta-analysis: 64 studies, 24,567 tooth surfaces (70% enamel-caries prevalence); fully within-person analysis: 19 studies, 2,849 tooth sites (66% prevalence) |
| Intervention | Five index-test technologies: visual or visual-tactile classification (ICDAS, ERK, other); fluorescence-based devices; imaging (analog and digital radiographs, CBCT); electrical conductance or impedance; transillumination and OCT | All 5 technologies evaluated in Analysis 2 (n=5); 3 technologies (fluorescence, imaging, visual classification) in Analysis 1 (n=3) |
| Comparison | Reference standard: histology, operative exploration/excavation, or enhanced visual assessment (with or without tooth separation) | Reference standards as planned (histology, excavation, enhanced visual examination) |
| Outcomes | Comparative diagnostic test accuracy (summary sensitivity and specificity) at the enamel caries threshold, with certainty assessed by GRADE | Summary sensitivity and specificity with 95% CrI and 95% PrI for each technology (Tables 1 and 2); pairwise differences in Fig. 3 |
| Study designs | General dental practice; primary studies conducted in vitro (extracted teeth) and in vivo (clinical) | Mix of in vitro (extracted teeth) and in vivo studies; indirectness noted from predominance of extracted-teeth studies |
| Timing | Target condition: caries at the enamel threshold (early caries) | Caries at the enamel threshold |
Search date: Not specified. Databases: Not specified (overview of 5 Cochrane DTA reviews: CD013811, CD014546, CD014545, CD013855, CD014547)
Questions and results
Fluorescence-based devices vs Reference standard (histology, operative excavation, or enhanced visual examination)
Population: Asymptomatic children or adults suspected of enamel caries (clinical studies); extracted teeth (in vitro studies)
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Specificity for detecting enamel caries Analysis 1: fully within-person comparison (studies evaluating >=3 index tests) In the fully within-person analysis, fluorescence had a summary specificity of 0.88 (95% CrI 0.78, 0.94); ungraded. Source: Table 1 (Analysis 1); Results, Analysis 1 · same question in other reviews | Specificity 0.88 [0.78, 0.94] | 19 studies · 2849 participants | Ungraded |
Sensitivity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Fluorescence-based devices had a summary sensitivity of 0.76 (95% CrI 0.68, 0.82) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Sensitivity 0.76 [0.68, 0.82] 213 true positives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Sensitivity for detecting enamel caries Analysis 1: fully within-person comparison (studies evaluating >=3 index tests) In the fully within-person analysis, fluorescence had a summary sensitivity of 0.71 (95% CrI 0.59, 0.81); ungraded. Source: Table 1 (Analysis 1); Results, Analysis 1 · same question in other reviews | Sensitivity 0.71 [0.59, 0.81] | 19 studies · 2849 participants | Ungraded |
Specificity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Fluorescence-based devices had a summary specificity of 0.83 (95% CrI 0.75, 0.89) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Specificity 0.83 [0.75, 0.89] 598 [540, 641] true negatives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Imaging (analog or digital radiographs) vs Reference standard (histology, operative excavation, or enhanced visual examination)
Population: Asymptomatic children or adults suspected of enamel caries (clinical studies); extracted teeth (in vitro studies)
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Specificity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Radiographic imaging had a summary specificity of 0.89 (95% CrI 0.83, 0.93) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Specificity 0.89 [0.83, 0.93] 641 [598, 670] true negatives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Sensitivity for detecting enamel caries Analysis 1: fully within-person comparison (studies evaluating >=3 index tests) In the fully within-person analysis, radiographic imaging had a summary sensitivity of 0.48 (95% CrI 0.35, 0.62); ungraded. Source: Table 1 (Analysis 1); Results, Analysis 1 · same question in other reviews | Sensitivity 0.48 [0.35, 0.62] | 19 studies · 2849 participants | Ungraded |
Sensitivity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Radiographic imaging had the lowest summary sensitivity, 0.50 (95% CrI 0.40, 0.59), significantly lower than all other technologies; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Sensitivity 0.50 [0.40, 0.59] 140 [112, 165] true positives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Specificity for detecting enamel caries Analysis 1: fully within-person comparison (studies evaluating >=3 index tests) In the fully within-person analysis, radiographic imaging had a summary specificity of 0.92 (95% CrI 0.84, 0.96); ungraded. Source: Table 1 (Analysis 1); Results, Analysis 1 · same question in other reviews | Specificity 0.92 [0.84, 0.96] | 19 studies · 2849 participants | Ungraded |
Visual or visual-tactile classification (e.g., ICDAS, ERK) vs Reference standard (histology, operative excavation, or enhanced visual examination)
Population: Asymptomatic children or adults suspected of enamel caries (clinical studies); extracted teeth (in vitro studies)
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Specificity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Visual or visual-tactile classification had a summary specificity of 0.81 (95% CrI 0.73, 0.87) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Specificity 0.81 [0.73, 0.87] 583 [526, 626] true negatives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Sensitivity for detecting enamel caries Analysis 1: fully within-person comparison (studies evaluating >=3 index tests) In the fully within-person analysis, visual classification had a summary sensitivity of 0.82 (95% CrI 0.73, 0.89); ungraded. Source: Table 1 (Analysis 1); Results, Analysis 1 · same question in other reviews | Sensitivity 0.82 [0.73, 0.89] | 19 studies · 2849 participants | Ungraded |
Sensitivity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Visual or visual-tactile classification had a summary sensitivity of 0.83 (95% CrI 0.77, 0.87) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Sensitivity 0.83 [0.77, 0.87] 232 [216, 244] true positives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Specificity for detecting enamel caries Analysis 1: fully within-person comparison (studies evaluating >=3 index tests) In the fully within-person analysis, visual classification had a summary specificity of 0.85 (95% CrI 0.74, 0.93); ungraded. Source: Table 1 (Analysis 1); Results, Analysis 1 · same question in other reviews | Specificity 0.85 [0.74, 0.93] | 19 studies · 2849 participants | Ungraded |
Electrical conductance or impedance vs Reference standard (histology, operative excavation, or enhanced visual examination)
Population: Asymptomatic children or adults suspected of enamel caries (clinical studies); extracted teeth (in vitro studies)
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Sensitivity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Electrical conductance or impedance had a summary sensitivity of 0.83 (95% CrI 0.66, 0.92) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Sensitivity 0.83 [0.66, 0.92] 232 [185, 258] true positives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Specificity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Electrical conductance or impedance had a summary specificity of 0.72 (95% CrI 0.44, 0.89) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Specificity 0.72 [0.44, 0.89] 518 [317, 641] true negatives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Transillumination and OCT vs Reference standard (histology, operative excavation, or enhanced visual examination)
Population: Asymptomatic children or adults suspected of enamel caries (clinical studies); extracted teeth (in vitro studies)
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Specificity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Transillumination and OCT had a summary specificity of 0.82 (95% CrI 0.68, 0.91) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Specificity 0.82 [0.68, 0.91] 590 [490, 655] true negatives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Sensitivity for detecting enamel caries Analysis 2: network meta-analysis (all studies evaluating >=2 index tests) Transillumination and OCT had a summary sensitivity of 0.76 (95% CrI 0.63, 0.86) for detecting enamel caries; low certainty. Source: Table 2 (Summary of Findings and GRADE); Table 1 (Analysis 2); Results, Analysis 2; Fig. 2 · same question in other reviews | Sensitivity 0.76 [0.63, 0.86] 213 [176, 241] true positives per 1,000 tooth surfaces (28% prevalence) | 64 studies · 24567 participants | Low Downgraded: risk of bias, inconsistency, indirectness |
Fluorescence-based devices vs Dental imaging (analog or digital radiographs)
Population: Children or adults asymptomatic or suspected of enamel caries; extracted teeth
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Difference in specificity Analysis 1: fully within-person comparison Fluorescence and imaging had similar specificity in the fully within-person analysis. Source: Results, Analysis 1 pairwise differences (JATS XML) · same question in other reviews | Difference in specificity -0.03 [-0.14, 0.06] | 19 studies · 2849 participants | Ungraded |
Difference in sensitivity Analysis 1: fully within-person comparison Fluorescence had higher sensitivity than imaging in the fully within-person analysis. Source: Results, Analysis 1 pairwise differences (JATS XML) · same question in other reviews | Difference in sensitivity 0.23 [0.05, 0.39] | 19 studies · 2849 participants | Ungraded |
Dental imaging (analog or digital radiographs) vs Visual classification
Population: Children or adults asymptomatic or suspected of enamel caries; extracted teeth
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Difference in sensitivity Analysis 1: fully within-person comparison Imaging had lower sensitivity than visual classification in the fully within-person analysis. Source: Results, Analysis 1 pairwise differences (JATS XML) · same question in other reviews | Difference in sensitivity -0.34 [-0.49, -0.17] | 19 studies · 2849 participants | Ungraded |
Difference in specificity Analysis 1: fully within-person comparison Imaging and visual classification had similar specificity in the fully within-person analysis. Source: Results, Analysis 1 pairwise differences (JATS XML) · same question in other reviews | Difference in specificity 0.06 [-0.04, 0.19] | 19 studies · 2849 participants | Ungraded |
Fluorescence-based devices vs Visual classification
Population: Children or adults asymptomatic or suspected of enamel caries; extracted teeth
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Difference in specificity Analysis 1: fully within-person comparison Fluorescence and visual classification had similar specificity in the fully within-person analysis. Source: Results, Analysis 1 pairwise differences (JATS XML) · same question in other reviews | Difference in specificity -0.03 [-0.09, 0.16] | 19 studies · 2849 participants | Ungraded |
Difference in sensitivity Analysis 1: fully within-person comparison Fluorescence and visual classification had similar sensitivity in the fully within-person analysis. Source: Results, Analysis 1 pairwise differences (JATS XML) · same question in other reviews | Difference in sensitivity -0.11 [-0.25, 0.03] | 19 studies · 2849 participants | Ungraded |
Comparison
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Difference in summary specificity (NMA, Analysis 2) Fluorescence vs Transillumination | Specificity difference (95% CrI) 0.01 [-0.10, 0.15] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Fluorescence vs Transillumination | Sensitivity difference (95% CrI) -0.01 [-0.12, 0.13] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Imaging vs Transillumination | Sensitivity difference (95% CrI) -0.27 [-0.40, -0.12] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Electrical conductance vs Imaging | Specificity difference (95% CrI) -0.16 [-0.45, 0.02] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Imaging vs Visual | Sensitivity difference (95% CrI) -0.34 [-0.42, -0.23] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Electrical conductance vs Visual | Sensitivity difference (95% CrI) 0.00 [-0.17, 0.11] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Electrical conductance vs Fluorescence | Specificity difference (95% CrI) -0.11 [-0.40, 0.07] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Imaging vs Transillumination | Specificity difference (95% CrI) 0.07 [-0.03, 0.20] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Electrical conductance vs Fluorescence | Sensitivity difference (95% CrI) 0.06 [-0.09, 0.18] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Fluorescence vs Imaging | Sensitivity difference (95% CrI) 0.26 [0.15, 0.37] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Electrical conductance vs Transillumination | Specificity difference (95% CrI) -0.09 [-0.38, 0.12] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Fluorescence vs Imaging | Specificity difference (95% CrI) -0.05 [-0.14, 0.02] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Electrical conductance vs Imaging | Sensitivity difference (95% CrI) 0.33 [0.15, 0.47] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Fluorescence vs Visual | Sensitivity difference (95% CrI) -0.07 [-0.16, 0.01] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Imaging vs Visual | Specificity difference (95% CrI) 0.08 [0.01, 0.16] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Transillumination vs Visual | Specificity difference (95% CrI) 0.01 [-0.13, 0.12] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Electrical conductance vs Visual | Specificity difference (95% CrI) -0.08 [-0.37, 0.10] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Electrical conductance vs Transillumination | Sensitivity difference (95% CrI) 0.06 [-0.12, 0.22] | NR | Ungraded |
Difference in summary specificity (NMA, Analysis 2) Fluorescence vs Visual | Specificity difference (95% CrI) 0.02 [-0.07, 0.11] | NR | Ungraded |
Difference in summary sensitivity (NMA, Analysis 2) Transillumination vs Visual | Sensitivity difference (95% CrI) -0.06 [-0.19, 0.05] | NR | Ungraded |