Radiology · Medical Literature
Comparative Diagnostic Performance of Cone Beam Computed Tomography and Dual-Energy X-Ray Absorptiometry for Low Bone Density Assessment. A Systematic Review and Meta-Analysis.
Abstract
This systematic review evaluated the diagnostic accuracy of CBCT for BMD assessment using DEXA as the reference standard and aimed to quantify its performance through meta-analysis. This review included prospective and retrospective observational studies assessing CBCT's diagnostic performance in detecting low BMD, with DEXA as the reference. Searches in PubMed, Embase, and Scopus identified 13 eligible studies involving adult participants (≥ 18 years). Two reviewers independently extracted data and assessed study quality using the QUADAS-2 tool. Meta-analysis was conducted using random-effects models. Pooled sensitivity and specificity of CBCT for detecting low BMD were 0.58 (95% CI: 0.50-0.67) and 0.78 (95% CI: 0.69-0.86), respectively. The summary receiver operating characteristic (SROC) curve showed an area under the curve (AUC) of 0.73 (95% CI: 0.69-0.77), indicating moderate diagnostic accuracy. Subgroup analyses revealed heterogeneity influenced by sample size and anatomical site. CBCT demonstrates moderate diagnostic accuracy in detecting low BMD, with better specificity than sensitivity. While promising, standardization of CBCT protocols and larger studies are needed to confirm its role in clinical assessment. PROSPERO REGISTRATION NUMBER: (CRD42024598958).
Clinical relevance
In adult patients, what is the diagnostic accuracy of cone beam computed tomography (CBCT) for detecting low bone mineral density (BMD), using dual-energy X-ray absorptiometry (DEXA) as the reference standard? __SNAPSHOT__ {"studyDesign":"Systematic review and meta-analysis of prospective and retrospective observational studies (diagnostic accuracy design)","setting":"Multi-database literature synthesis (PubMed, Embase, Scopus); individual study settings not reported in abstract","sampleSize":"13 eligible studies included; aggregate patient-level n not reported in abstract","population":"Adult participants aged ≥18 years undergoing CBCT assessment; specific clinical indications (e.g., dental implant planning, osteoporosis screening) and sex distribution not reported in abstract","intervention":"Cone beam computed tomography (CBCT) for assessment of bone mineral density; specific CBCT protocols, anatomical sites, and measurement parameters varied across included studies","comparator":"Dual-energy X-ray absorptiometry (DEXA) as the reference standard for low BMD diagnosis","primaryOutcome":"Diagnostic accuracy of CBCT for detecting low BMD, quantified as pooled sensitivity, specificity, and area under the summary receiver operating characteristic (SROC) curve (AUC)","followUp":"Not reported (cross-sectional diagnostic accuracy studies; no longitudinal follow-up implied)","fundingSource":"Not reported"}
Key findings
Among 13 included studies, the pooled sensitivity of CBCT for detecting low BMD was 0.58 (95% CI: 0.50–0.67), and the pooled specificity was 0.78 (95% CI: 0.69–0.86). The SROC AUC was 0.73 (95% CI: 0.69–0.77), characterizing overall diagnostic performance as moderate. A false-negative rate of approximately 42% (calculated: 1 − pooled sensitivity) means that roughly 4 in 10 individuals with true low BMD would not be identified by CBCT alone. Conversely, CBCT's higher specificity suggests a lower false-positive rate of approximately 22% (calculated). NNT and NNH are not calculable from the reported diagnostic accuracy metrics. Subgroup analyses identified heterogeneity in diagnostic performance attributable to sample size and anatomical site examined; specific subgroup estimates were not reported in the abstract. Clinically important adverse events are not applicable to a diagnostic accuracy review. No p-values for subgroup comparisons were reported in the abstract.
Practice implications
This meta-analysis establishes that CBCT demonstrates only moderate overall diagnostic accuracy (AUC 0.73) for identifying low BMD when benchmarked against DEXA, the accepted clinical reference standard. The asymmetry between sensitivity (0.58) and specificity (0.78) is clinically critical: CBCT is substantially better at ruling in low BMD (when positive) than at ruling it out (when negative), and therefore cannot safely serve as a standalone screening tool to exclude osteoporosis or osteopenia. A negative CBCT result should not be used to reassure clinicians that skeletal BMD is adequate. The findings may be most relevant to opportunistic use — identifying patients who warrant formal DEXA referral when CBCT is already being performed for dental or maxillofacial indications — rather than as a replacement for DEXA in diagnostic pathways. Generalizability is limited because patient-level characteristics (age distribution, sex, menopausal status, ethnicity, comorbidities, indication for imaging) and the range of CBCT protocols and anatomical sites are heterogeneous across included studies and not fully characterized in the abstract. Patients outside the maxillofacial context were likely underrepresented. This summary is based solely on the published abstract; full-text review is required to assess QUADAS-2 domain-specific risk of bias, magnitude of between-study heterogeneity (I²), and threshold effects that may influence the operating point on the SROC curve. __LIMITATIONS__ Only 13 studies included; aggregate patient-level sample size not reported, limiting precision of pooled estimates.,Significant heterogeneity across studies in CBCT protocols, anatomical measurement sites (e.g., mandible, cervical vertebrae), and BMD threshold definitions for 'low BMD,' precluding standardized comparison.,QUADAS-2 was applied, but domain-specific risk of bias results are not reported in the abstract; the degree to which included studies were at risk for spectrum bias, partial verification bias, or incorporation bias is unknown.,Predominantly or exclusively observational study designs introduce potential for selection bias and confounding; randomization was not applicable.,DEXA itself carries measurement variability depending on site (lumbar spine vs. femoral neck vs. total hip) and scanner calibration; inconsistency in the reference standard across studies may artificially inflate heterogeneity.,The abstract is truncated and does not report I² statistics, threshold effect analysis, or Deeks' funnel plot asymmetry test for publication bias.,Subgroup analyses by sample size and anatomical site were performed post hoc; their results and clinical significance are not fully described.,Generalizability is restricted by the likely predominance of dental/maxillofacial patient populations; findings may not apply to general osteoporosis screening populations.,Funding source is not disclosed, precluding assessment of potential conflicts of interest.,The abstract is truncated; conclusions regarding protocol standardization recommendations are incomplete. __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ CBCT demonstrates only moderate diagnostic accuracy for detecting low BMD (pooled sensitivity 0.58, specificity 0.78, AUC 0.73), with a clinically unacceptable false-negative rate of approximately 42%; it cannot replace DEXA as a diagnostic or screening tool. Its highest near-term clinical utility lies in opportunistic flagging of at-risk patients already undergoing CBCT for dental or maxillofacial indications, prompting formal DEXA referral. Standardization of CBCT acquisition protocols and larger, prospectively designed diagnostic accuracy studies are required before broader integration into bone health care pathways.