Twenty-seven healthy participants were enrolled in this prospective study. All participants completed the Xue Mountain trek in March or August 2022. The participants were divided into two groups based on whether they had HM. HM was defined as −6 diopters or more. The Lake Louise AMS scores of the participants, fitness conditions, the use of acetazolamide, IOP, and ONSD were recorded for comparison. All participants lived at an altitude of less than 500 m above sea level, and none of them had been exposed to an altitude above 2,000 m in the 4 months prior to this study. None of the participants had a medical history of hypertension, diabetes, hyperthyroidism, obstructive sleep apnea, glaucoma, ophthalmic surgery, or other ophthalmic diseases. Throughout the entire trek, smoking and the consumption of alcohol were not allowed. All participants provided written informed consent, and the study was approved by the Institutional Review Board of MacKay Memorial Hospital (21MMHIS088e).

This study was designed as a 3-day trek of Xue Mountain (altitude: 2,140–3,886 m), Taiwan (Figure 1). The route began from Taipei City (16 m above sea level) to Wuling Farm (altitude: 1,700 m) by car for 4 h the day before trekking. On the first day of hiking, the participants were driven by car from the lodging location to Xue Mountain trailhead in 20 min (altitude: 2,140 m). Thereafter, they trekked to 369 Cabin (distance: 7.1 km; altitude: 3,150 m) for 8 h. On the second day, they trekked a 3.6 km round trip (summit: 3,886 m, approximately 10 h of walking). On the third day, they descended to the Xue Mountain trailhead.

Diopter and axial lengths were measured in the lowland 1 week before ascent. Diopter measurements were checked using the KR 8900 (Topcon, Tokyo, Japan). IOP and ONSD were obtained at five different time points: day 0 (8–10 a.m.) in Taipei, day 0 (4–6 p.m.) at Wuling Farm (moderate altitude), day 1 and 2 (4–6 p.m.) in 369 Cabin (high altitude), and day 3 (8–10 a.m.) in 369 Cabin (high altitude). Prior to the ophthalmic examination, participants were advised to rest for a minimum of 15 min. Participants were also instructed to avoid taking acetazolamide within the 8 h preceding the examination. An emergency physician, trained in ultrasound skills through the advanced point-of-care ultrasound (POCUS) courses, conducted ophthalmic ultrasound. Each participant was examined while sitting upright and facing forward. An IC100 (Icare Finland Oy, Finland) rebound tonometer was used to measure the IOP. We used the value of the right eye for statistical analysis. An 8–9 MHz linear probe of the Vscan Air (GE Healthcare, Horten, Norway) was used to measure the ONSD. The ONSD was measured 3 mm behind the eye globe. The ONSD value was obtained by averaging the values of the two eyes.

Oxygen saturation (SpO₂) and respiratory rate (RR) were measured using the BeneView T1 (Mindray, China) while the participants quietly sat on a chair after 10 min of rest. The BeneView T1 is a transport monitor approved by the U.S. Food and Drug Administration (size: 10.2 cm × 14.2 cm × 8.1 cm; weight: 0.9 kg). An SpO₂ probe was placed on the right index finger.

AMS was diagnosed based on the Lake Louise Scoring system (LLSS) for AMS. The LLSS is a self-assessment questionnaire that includes four main symptoms (headache, nausea, dizziness, and fatigue), each rated on a scale of 0–3 (0 = no discomfort, 1 = mild, 2 = moderate, and 3 = severe symptoms). LLSS scores ≥ 3 were considered as symptomatic AMS. Furthermore, we defined mild AMS as 3–5 points, moderate AMS as 6–9 points, and severe AMS as 10–12 points (Roach et al., 2018).

All statistical analyses were performed using SPSS version 26 for Windows (Statistical Package for Social Sciences, IBM, Chicago, IL). The Shapiro–Wilk test (p > 0.05) and an inspection of the participants’ histograms, normal Q-Q plots, and box plots were used to test the normality of our data. Descriptive results were reported as means ± standard deviations. Fisher exact test was applied for categorical data. An independent samples t-test (or Mann–Whitney U test, when appropriate) was used to evaluate the association between participants with and without HM. The repeated measured analysis of variance (ANOVA) was used to evaluate the association between five different time points [day 0 (8–10 a.m.); day 0, 1, and 2 (4–6 p.m.); and day 3 (8–10 a.m.)]. The numerical parameters of IOP and ONSD at 1,700 m and 3,150 m altitudes were compared to the baseline values at ground level using the paired samples t-test. The correlations between AMS and fractional changes in IOP and ONSD were calculated using point-biserial correlation. The area under the ROC curve (AUROC), sensitivity, specificity, positive predict value (PPV), and negative predict value (NPV) were applied to obtain the diagnostic performance of ONSD fractional changes. Statistical significance was set at p < 0.05.

The clinical application and normative measurement of ONSD were first established in the 1990s (Hansen and Helmke, 1996). Subsequently, ONSD was widely studied and applied in clinical practice. The measurement of ONSD using ultrasound is preferred because of its low cost, reproducibility, availability, and noninvasiveness. Ultrasound was used to measure ONSD in our study because medical resources and diagnostic capabilities are usually limited in high-altitude environments. A meta-analysis in 2023 pooled six articles on the association between ONSD and AMS symptoms. The majority of studies showed that ONSD was increased during ascent regardless of AMS or not (Lau et al., 2023). In our results, we find that increased ONSD values were observed in both HM and non-HM trekkers from ground to 1,700 m. Our data, although far from conclusive, provide additional information regarding ocular physiological responses at high altitudes.

Research has shown a good correlation between ONSD measured using ultrasound and magnetic resonance imaging (MRI) (Roemer et al., 2023). Nguyen et al. performed MRI to evaluate myopic and emmetropic eyes. They found a smaller cross-sectional area of the optic nerve and subarachnoid space in myopic eyes; however, the overall diameters of the optic nerve sheath complex were similar between the two groups (Nguyen et al., 2021). Our data followed a trend similar to that of Nguyen’s study results. The ONSD values during ascent were not different between HM and non-HM trekkers. Further studies are needed to clarify the association between ONSD and myopia at high altitudes.

During trekking, many physiological and environmental factors can affect the IOP. Exercise contributes to a decreased IOP. Body fluid is shifted to muscle and lost via sweat glands during and after exercise, which results in decreased blood volume and venous pressure and then decreased IOP (Kim and Park, 2019). Bosch et al. (2010a) showed that IOP increased during ascent to 5,533 m and then decreased with further ascent to 6,265 m. They hypothesized that severe hypoxia impairs the function of the non-pigmented ciliary epithelium and thus decreases the production of aqueous humor. Xie et al. (2019) found that hypoxia lowers IOP at high altitudes by decreasing PaCO₂ and bicarbonate levels as a result of hyperventilation, which contribute to decreased production of aqueous humor. Ersanli et al. (2006) revealed a temporary increase in IOP in 34 healthy male pilots, either under normoxia or hypoxia in a simulated hypobaric chamber. They found that IOP changes were not only related to altitude but also to acclimatization.

In contrast, Xin et al. found that individuals with high myopia presented less configuration changes when the ciliary body contracted (Xin et al., 2020). Thus, a smaller change in the configuration of the ciliary body causes slower regulation in response to the IOP. Xu et al. found that IOP increased after ocular compression and that participants with HM had slower IOP decay (Xu and Lam, 2023). In our study, IOP increased in participants with HM on day 1 afternoon at an altitude of 3,150 m. Although the increase was not significant, HM trekkers had greater IOP levels as compared to non-HM ones at that time point. These findings may be due to the impaired regulation of IOP in highly myopic eyes. As the duration at the high altitude increased, the IOP decreased after acclimatization.

The possible factors affecting ONSD during ascending to altitude include craniospinal and spatial compensations. Spatial compensation accommodates ICP by shifting the CSF from the intracranial space to the spinal thecal and dural sacs. Craniospinal compensation refers to elastic properties of the craniospinal system. Craniospinal systems can buffer CSF fluctuations to prevent large changes in ICP (Lawley et al., 2016). Although transient elevation in ICP could activate trigeminal afferents, which triggers headache (Rolan, 2015), the mechanism of compensation in the CSF tends to maintain ICP in the normal range.

Kanaan et al. (2015) performed a cohort study and found that changes in ONSD were higher in participants with AMS. Currently, no physiological parameters detected at moderate altitude can be useful as a sensitive measurement for early prediction of trekkers suffering from AMS while ascending to high altitude. Our study revealed that fractional changes in ONSD on day 0 at an altitude of 1,700 m correlated with LLSS and headache symptoms on day 2. We chose LLSS data from day 2 because previous studies have shown that AMS severity peaks at that time point (Luks et al., 2017). Additionally, in our study, all participants with AMS had mild symptom severity, which might not have the profound effects of increased cerebral blood flow and cerebral edema. According to the above hypothesis, our results imply that fractional changes in ONSD at moderate altitude might be a predictor of mild AMS.

Cerebral edema may result from AMS or HACE (Sagoo et al., 2017). Reliable noninvasive methods for monitoring increased intracranial cerebral pressure (IICP) are desirable for clinical application. IOP has been previously studied for its correlation with ICP. When ICP increases in vivo, it can compress the episcleral veins that drain blood from the eyes, resulting in increased resistance to the outflow of aqueous humor and a subsequent increase in IOP. This is known as the “venous stasis” theory, and it suggests that ICP has a direct effect on IOP (Widakowich, 1976; Han et al., 2008; Ghate et al., 2021). Although ICP is correlated with IOP, the variability between individuals is large (Sheeran et al., 2000; Yavin et al., 2014). In our study, IOP did not seem to correlate with the LLSS. A possible reason why IOP was not associated with LLSS was that the symptoms of AMS in our participants were not severe enough to cause IICP. Sheeran et al. found that an increase of 11 mmHg in ICP caused a 1 mmHg rise in IOP (Sheeran et al., 2000). In our study, the severity of AMS was mild, and no participant had symptoms of HACE. Hence, our study showed that fractional changes in IOP were not correlated with the LLSS. A greater increase in ICP is required to achieve a significant elevation in IOP.

Our study had some limitations. Although the height of the main peak of Xue Mountain is 3,886 m, we only stayed there for a short period. We returned to 369 Cabin (3,150 m) soon after reaching the peak. Thus, the incidence of AMS was not as high as previous research showed (prevalence of 38% at 3,500 m) (Mairer et al., 2009). The incidence of AMS in our study was 14.8% and most of the participants had mild symptoms. Assessing ONSD as a surrogate measure and marker for AMS is challenging at high altitudes due to various adaptive responses. These adaptations include respiratory alkalosis resulting from compensatory tachypnea, changes in sleep patterns, heightened sympathetic tone, and the ambient hypobaric and hypoxic environment (Strapazzon et al., 2014; Dinsmore et al., 2017). Another limitation arose from the lack of control over the use of acetazolamide due to health and safety considerations. Participants were merely advised to refrain from taking acetazolamide within the 8 h leading up to the ophthalmic examination. Acetazolamide is a reversible carbonic anhydrase inhibitor. It not only increases renal excretion of sodium, potassium, bicarbonate, and water, but also decreases ICP and CSF secretion by acting on the choroid plexus (Uldall et al., 2017). Two participants did not take prophylactic acetazolamide. However, they consumed acetazolamide due to symptoms of AMS during the trek. During ascent, central corneal thickness is also affected by hypobaric hypoxia. Hypoxia leads to a reduction in oxidative metabolism, prompting a shift towards anaerobic glycolysis and subsequent lactate accumulation. The elevated lactate levels, in turn, trigger an influx of water, ultimately resulting in corneal edema (Bosch et al., 2010b). As IOP is influenced by corneal thickness, increased corneal thickness may lead to an overestimation of the IOP (Morris et al., 2007).