This systematic review and meta-analysis evaluated the diagnostic accuracy of sonographic optic nerve sheath diameter measurement for detecting intracranial hypertension in emergency and critical care settings, using invasive intracranial pressure monitoring as the reference standard. The analysis, which did not report total sample size, identified a critical diagnostic threshold of ONSD ≥5.1 mm. The summary diagnostic accuracy, expressed as a CDEI (composite diagnostic effectiveness index) root node value, was 0.82 (95% CI: 0.79-0.85), with individual study CDEI values ranging from 0.64 to 0.95.
The review found that diagnostic performance improved with specific techniques and operator experience. Multiplanar assessment significantly enhanced prediction of favorable outcomes (CDEI >0.85, p<0.001), and the use of multiplane techniques by experienced operators was associated with a 50% reduction in missed diagnoses (LR- 0.1). When comparing diagnostic thresholds, using a 25 cmH₂O ICP cutoff showed greater specificity (88.4% vs. 82.9%) and a higher positive likelihood ratio (6.5 vs. 5.0) than a 20 mmHg threshold. Diagnostic accuracy also varied by etiology, with nontraumatic brain injury cases showing superior sensitivity (91.9% vs. 85.2%) and a 107% higher diagnostic odds ratio (63.3 vs. 30.5) compared to traumatic cases.
Safety and tolerability data for the sonographic procedure were not reported. Key limitations include the lack of reported total sample size and follow-up duration, heterogeneity in measurement techniques and population demographics across included studies, and the absence of absolute numbers for many outcomes. The evidence suggests sonographic ONSD measurement, particularly when performed with multiplanar techniques by experienced operators, may provide useful adjunctive information for rapid bedside assessment of suspected intracranial hypertension. However, given the observational nature of the evidence and variability in reported accuracy, it should not replace invasive ICP monitoring when definitive measurement is clinically indicated.
View Original Abstract ↓
OBJECTIVES: In this study, we aimed to assess the clinical efficacy of sonographic optic nerve sheath diameter (ONSD) measurements for detecting intracranial hypertension (ICH). Our objectives included evaluating its diagnostic precision compared with invasive intracranial pressure (ICP) monitoring; exploring potential sources of heterogeneity, such as measurement technique (multi vs. single-planar), population demographics (Asian vs. non-Asian), and major etiologies; and determining validated thresholds for prompt bedside ICP assessment in acute neurological crises.
DATA SOURCES: We comprehensively searched eight databases, including four in English (PubMed, Web of Science, MEDLINE®, and Embase) and four in Chinese (China National Knowledge Infrastructure [CNKI], Wanfang, VIP, and SinoMed). The search covered literature published from the inception of the databases up to May 31, 2025.
STUDY SELECTION: This review includes prospective, retrospective, and case-control studies that assess the efficacy of sonographic ONSD measurement for detecting ICH, defined as an ICP exceeding 20 mm Hg, 20 cmHO, 25 cmHO with invasive ICP measurement as the gold standard.
DATA EXTRACTION: X.T. and Q.T. independently screened the studies and assessed the risk of bias using the QUADAS-2 tool.
STATISTICAL ANALYSIS: Diagnostic accuracy was pooled using a bivariate random-effects model to generate summary receiver operating characteristic (SROC) curves. Subgroup analyses of the measurement technique, study population, and etiology were conducted. Publication bias was assessed using Deeks' test.
DATA SYNTHESIS: The decision tree analysis revealed that an ONSD ≥ 5.1 mm is the critical diagnostic threshold for detecting intracranial hypertension. The regression tree model indicated a range of CDEI values from 0.64 to 0.95, with a root node value of 0.82 (95% CI: 0.79-0.85). The incorporation of a multiplanar assessment significantly enhanced the prediction of favorable outcomes (CDEI > 0.85; p < 0.001). Moreover, compared with a 20 mmHg threshold, a 25 cmH₂O threshold resulted in greater specificity (88.4% vs. 82.9%) and a higher positive likelihood ratio (LR+ 6.5 vs. 5.0). The utilization of multiplane techniques and experienced operators notably improved diagnostic accuracy, leading to a 50% reduction in missed diagnoses (LR- 0.1). Nontraumatic brain injury cases exhibited superior sensitivity (91.9% vs. 85.2%) and a 107% higher diagnostic odds ratio (63.3 vs. 30.5) than traumatic brain injury cases did.
CONCLUSIONS: Findings from this study suggest that an ONSD threshold ≥ 5.1 mm is a promising diagnostic marker for elevated ICP (> 20 cmH₂O/mmHg or acute etiology), although external validation is needed. A multiplane evaluation improves its predictive accuracy for favorable outcomes. The developed decision model supports the use of sonographic ONSD measurement as a supplementary screening tool in emergency and critical care, particularly when invasive monitoring is unavailable. Additionally, an ONSD of approximately 6.0 mm is a potential prognostic indicator for unfavorable outcomes, highlighting a key area for future research. These findings can only be applied after their validation through multicenter studies and the establishment of standardized measurement protocols.
