In an adult healthy volunteer population in Bogota, Colombia, using orbital US we obtained a mean ONSD-TP of 3.96 mm (95% CI: 3.85 mm-4.07 mm) and a mean ONSD-TP of 4.0 mm (95% CI: 3.90 mm-4.11 mm). For the ONSD/ETD ratio, the mean was 0.17 (95% CI: 0.16–0.17). The ranges for ONSD-TP and ONSD-SP were from 2.35 mm to 5.15 mm and from 2.60 mm to 5.20 mm, respectively. The ONSD/ETD ratio ranged from 0.1 to 0.23. We found a significant correlation between ONSD-TP and ONSD-SP (p < 0.0001), and the intraclass correlation between eyes was statistically significant and comparable with previous reports [5, 10]. We did not find any significant association between sex, age, BMI, HC and US measures of ONSD or ONSD/ETD (p > 0.05), which is consistent with the findings of a recent systematic review and meta-analysis and a previous review [9, 11]. Considering that the ONSD did not exceed 5.2 mm in the transversal or sagittal plane, a 30 degrees test to check for undetected intracranial hypertension was not necessary [12, 13].
Previous studies of US ONSD measures in healthy populations (children and adults) reported a range of 2.2 to 5.4 mm. In 67 subjects in the United Kingdom, the mean ONSD ranged between 2.4–4.7 mm (mean 3.2–3.6 mm) [14]. In 26 subjects in Greece, the range was between 2.2–4.9 mm (mean 3.6 mm), and in 136 subjects in Bangladesh, it ranged between 4.24–4.83 mm (mean 4.41 mm) [15, 16]. Higher values of US ONSD have been found in studies in China (range 4.7–5.4 mm, mean 5.1 mm) and Korea (range 4.6–5.2 mm, mean 4.9 mm) [10, 17]. US ONSD cut-off values for the diagnosis of increased ICP ranged from 4.1 to 5.7 mm, with a good correlation with invasive measures. Cut-off measures of US ONSD of 4.7 to 5.7 mm for an increased ICP diagnosis showed a sensitivity of 70–100% and specificity of 31.9–100% [18]. Normative values for ONSD/ETD in healthy subjects are seldom reported. Kim et al. presented the first study establishing a normal ONSD/ETD ratio by ultrasound in 585 healthy volunteers, reporting a mean of 0.18 (range 0.12–0.23), very similar to our results [11]. A US-ONSD/ETD ratio of 0.25 has been proposed as a threshold value with a sensitivity of 90% and specificity of 82.3% [6].
US ONSD measures have shown minimal interobserver and eye-to-eye variations. Interobserver variation ranges between 0.2–0.3 mm, and differences in measures between the axial and sagittal plane of US ONSD are in the range of 0–0.3 mm (mean 0.15 mm) [14]. Variability can also be minimized with appropriate training, and the learning curve for optimal orbital US-ONSD measurement is only between 10 and 25 scans [19]. The studies discussed above also found no association between US-ONSD and age, weight or height, further reducing the variability. A study including HC, did not find a relationship between US-ONSD and HC [16], although ONSD may vary by sex and ethnicity [8].
Although studies have reported on many parameters of US-ONSD in healthy populations, no global consensus exists on normal US-ONSD measures and cut-off values for the diagnosis of increased ICP [20]. There is also no universally accepted standardized protocol for sonographic assessment of the ONSD, even though proposals for protocols have been made [20]. Our study was carried out using the technique performed in most published studies.
Knowledge of the normal range of US ONSD in a healthy local population is essential to interpret the results of this diagnostic test. Based on the results of our study, prior research and expected deviations when measuring US-ONSD, we propose a normal range for US-ONSD of 2.35–5.20 mm in healthy adult Colombians healthy and a cut-off value for diagnosing increased ICP of 5.5 mm.
To the best of our knowledge, this is the first study to report normative values in a sample of exclusively Latin American subjects [9]. Relationships between demographic (sex, age) and anthropometric (weight, height, BMI, and HC) variables with US ONSD were analysed, and no association was found. The sample size was adequate and there was careful exclusion of subjects with potentially abnormal US-ONSD. The subjects were also examined in a systematic and standardized fashion by trained physicians.
This study has some limitations. First, we did not use neuroimaging to confirm the absence of disease/masses. However, patients with ICP typically present with papilledema, which is absent in less than 6% of subjects. In either case, certain symptoms were reported in both groups of patients with ICP (e.g., headache, visual field loss, diplopia), and they were the basis for our exclusion criteria [21]. Second, the mean age of our patients was 26.7 ± 8.3 years, which, considering the possible influence of ageing and brain atrophy on ONSD, would make the ranges found in this study narrower than they would be if older subjects had been included. However, previous studies have shown that ONSD remains similar throughout adult life and does not vary even in healthy adults aged 65 years or older [22]. Third, the interobserver variability was not analysed, as different investigators measured ONSD in different volunteers. A direct measurement of ICP was not performed, and how the upper values of US ONSD correlate with ICP could not be determined. Additionally, we did not compare the ONSD measurements with the results of other noninvasive methods.
We did not associate the standardized A-scan technique with the B-scan ocular ultrasound results. The latter could be affected by the so-called “blooming effect” making measurements unreliable as a recent review suggested [13, 23]. Finally, the lack of previous studies on healthy adult Latin American populations limits the possibility for regional comparisons. Despite the limitations mentioned and the knowledge that further research is needed on US ONSD and its correlation with increased ICP, our study helps build a reference of normative data for Latin American US measurements.
