†These authors contributed equally.
Academic Editor: Jesús Pastor Gómez
This study aims to detect whether the optic nerve sheath diameter (ONSD) can be
used to dynamically monitor intracranial pressure (ICP). Adult patients
undergoing invasive ICP monitoring on the day of admission are included in this
study. For each patient, the ONSD is first measured in the supine position and
then in the 30
Several neurological comorbidities (e.g., head injury, hydrocephalus and
subarachnoid haemorrhage) can result in increased intracranial pressure (ICP) [1, 2]. ICP is typically measured in mmHg or cmH
Various non-invasive ICP estimation methods have been proposed, with different
advantages and limitations; however, none have been sufficiently accurate to
replace invasive ICP measurement [9, 10, 11]. Measurement of the optic nerve sheath
diameter (ONSD) by magnetic resonance imaging, computed tomography or
ultrasonography are promising non-invasive ICP estimation methods [12, 13]. The
diameter of the optic nerve sheath (ONS) is linked to the cerebrospinal fluid
(CSF) pressure [14], and the meningeal envelope around the ONS is a continuation
of the dural and subarachnoid spaces. An increase in ICP can lead to a shift of
CSF into the ONS, resulting in an increase in its diameter, mainly in the
anterior part; it can also cause an increase in the diameter of the nerve itself
due to the accumulation of CSF among the fibres and result in papilledema [15].
Previous studies have demonstrated that the assessment of ONSD is associated with
ICP. Chen et al. [1] studied 84 patients and found that the relative real-time
changes in ICP could be reflected by the ONSD measured via ultrasound.
Hanafi et al. [16] analysed data from 112 patients with traumatic headaches and found
that the ultrasonic measurement of ONSD was sensitive and specific for the
detection of patients with high ICP. A prospective study by Robba et
al. [7] indicated that ONSD was the best non-invasive method to estimate ICP and
suggested the following formula to estimate ICP using ONSD: nICP
This study aims to assess whether the ONSD can be used to dynamically monitor
ICP. Two studies are conducted to achieve this goal. First, the correlation
between the changes in ONSD and ICP values when patients changed from the supine
to the 30
This prospective study was performed in the Department of Neurology of the Third Hospital of Hebei Medical University from May 2019 to June 2021. The study protocol was approved by the Institutional Review Board of the Third Hospital of Hebei Medical University (No. K2019-010-1). All patients enrolled in the study signed informed consent.
The inclusion criteria were as follows: (1)
Two invasive methods were used to detect ICP: an intra-parenchymal fibreoptic transducer (Camino Laboratories, Integra Lifesciences, CA, USA) and a catheter inserted into the brain ventricles that was connected to an external pressure transducer (Codman, Johnson & Johnson Medical Ltd., New Brunswick, NJ, USA). The instruments were operated by neurosurgeons with routine clinical practice in the operating room. ICM+ software (ICM, Cambridge, UK) was used to monitor the data collection and the real-time calculation of ICP values.
The ONSD assessment was performed in B-mode using the
iU22 ultrasound system (Philips, Amsterdam, Netherlands) and a
6–13 MHz linear array transducer (Philips, Amsterdam,
Netherlands) using the lowest possible acoustic power to measure ONSD to avoid
damage to the retina and the lens [18]. The patients were placed in the
supine position or the 30
All patients were carefully managed in accordance with international
guidelines [12, 13]. Mechanical ventilation was used as necessary to
maintain adequate oxygenation (SaO
To further detect whether the ONSD values increased or decreased in response to the ICP changes, a dynamic test was conducted in 16 of the 49 patients. The ONSD was measured in the supine position once a day for three consecutive days, starting on the day of admission. The ICP values were recorded each day (see Fig. 1).
CONSORT patient flow diagram.
The data were analysed using SPSS 22.0 software (IBM Corp., Armonk, NY, USA).
The quantitative data were described as mean
A total of 53 patients were screened for this study, and four patients were excluded because they had histories of ophthalmological disorders or optic nerve trauma. Finally, 49 patients with a median age of 61 (18–75), including 38 men (77.6%) and 11 women (22.5%), requiring ICP monitoring upon admission were enrolled in this study. With respect to the reasons for requiring ICP monitoring, 27 patients (55.1%) had a hypertensive cerebral haemorrhage, six patients (12.2%) had an aneurysmal subarachnoid haemorrhage, 13 patients (26.5%) had a traumatic brain injury and three patients (6.1%) underwent ICP monitoring for other reasons. In terms of the method of ICP monitoring, 27 patients (55.1%) were monitored using an intraparenchymal fibreoptic transducer, and 22 patients (44.9%) were monitored using an external ventricular drainage system. The median GCS score at admission was 6 (3–12), and the median APACHE II score at admission was 19 (17–24). Among the enrolled patients, three (6.1%) were complicated with intracranial infection, eight (16.3%) with cerebral hernia and one (2.0%) with sepsis.
The characteristics of the 49 patients included in the study are shown in Table 1.
Characteristics | N (%) or median (range) | |
Total number | 49 (100) | |
Male/female | 38 (77.6)/11 (22.4) | |
Age (years) | 61 (18–75) | |
Height (cm) | 174 (160–180) | |
Weight (kg) | 70 (50–95) | |
GCS score at admission | 6 (3–12) | |
APACHE II score at admission | 19 (17–24) | |
SBP (mmHg) | 132 (99–202) | |
DBP (mmHg) | 74.5 (34–100) | |
MAP (mmHg) | 93 (59–126) | |
ICP measurements | ||
Intraparenchymal fiberoptic transducer | 27 (55.1) | |
External ventricular drainage system | 22 (44.9) | |
Diagnosis | ||
Hypertension cerebral hemorrhage | 27 (55.1) | |
Aneurysmal subarachnoid hemorrhage | 6 (12.2) | |
Traumatic brain injury | 13 (26.5) | |
Others | 3 (6.1) | |
Complications | ||
Intracranial infection | 3 (6.1) | |
Cerebral hernia | 8 (16.3) | |
Sepsis | 1 (2.0) | |
Note: GCS, Glasgow Coma Scale; APACHE, Acute Physiology and Chronic Health Evaluation; SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial blood pressure. |
Of the 49 patients, 16 underwent dynamic tests, including 12 men (75.0%) and four women (25.0%) with a median age of 62 (34–75). Of the 16 patients, 13 (81.3%) underwent ICP monitoring via the intraparenchymal fibreoptic transducer and three (18.8%) via the external ventricular drainage system. The median GCS score of these 16 patients upon admission was 7 (3–11), and the median APACHE II score was 18.5 (17–22).
In the supine position, the mean ICP of the enrolled patients was 18.30
In the 30
When patients changed from the supine position to the 30
Correlation between change in ICP and change in ONSD when
patients changed the position. The change in ICP was not strongly correlated with
the change in ONSD, with an r of 0.358 (95 % CI, 0.086–0.58; p =
0.012). ICP, intracranial pressure; ONSD, optic nerve sheath diameter. The
correlation among the changes of ONSD and ICP from the supine position to the
30
The dynamic test was performed on 16 patients. As a result of the different units of measurement for the ONSD and ICP, the values were standardised using Z-scores. The Z-scores at D1, D2 and D3 for the ICP and ONSD for each patient are presented in Fig. 3. In terms of the changes in ICP, of the 16 patients, eight (50.0%) demonstrated a decrease from day one to day three, three (18.8%) demonstrated a decrease from day one to day two and an increase from day two to day three, five (31.3%) had an increase from day one to day two and a decrease from day two to day three, one (6.3%) demonstrated a decrease from day one to day two and an increase from day two to day three and one (6.3%) demonstrated an increase from day one to day three.
ICP and ONSD z-scores during dynamic test. In 3 patients (patient 11, 13, and 16), the lines of ICP and ONSD showed completely parallel. Three patients had completely different profiles for ICP and ONSD (patients 1, 2, and 7). ICP, red line; ONSD, blue line; ICP, intracranial pressure; ONSD, optic nerve sheath diameter. * indicated that the patient used intraparenchymal fiberoptic transducer to monitor ICP. The values of ONSD and ICP were standardized using Z-scores (produced in consideration of the intra-patient mean and SD). The Z-scores were calculated at three time points (D1, D2, and D3) and graphically described as points connected by a line. For each patient, if two lines (representing ONSD and ICP) were parallel, the ONSD and ICP would have good agreement. Conversely, if the two lines were divergent, the agreement was incomplete or absent.
In terms of the changes in ONSD, of the 16 patients, five (31.3%) demonstrated a decrease from day one to day three and nine (56.3%) had an increase from day one to day two and a decrease from day two to day three.
The lines on the graph representing the changes in ICP and ONSD were completely parallel in three patients (patients 11, 13 and 16), indicating a good agreement between the ICP and ONSD during the dynamic test for these three patients (18.8%). However, in three patients (patients 1, 2 and 7) the profiles for the ICP and ONSD were completely different.
The results of the present study indicate that in a longitudinal test, the
variation in the ICP was not strongly correlated with the variation in the ONSD
when patients changed from the supine to the 30
Although non-invasive methods for assessing ICP are not accurate enough to substitute for invasive methods, non-invasive ICP estimation may be helpful and could be used as a method to identify patients at high risk for developing intracranial hypertension who require specific monitoring and surveillance or as a diagnostic tool in patients with an unexplained alteration of consciousness outside the ICU [11, 21]. The ONSD measured by ultrasonography is an important non-invasive tool to estimate ICP and can be used to screen patients with high ICP while avoiding the placement of intracranial probes and the associated risks of bleeding and infection [12]. Previous studies have demonstrated the accuracy of the ONSD for measuring ICP [7, 22]. Chen et al. [1] found that the relative real-time changes in ICP could be reflected by the ONSD measured by ultrasound. Hanafi et al. [16] found that ultrasonic measurement of the ONSD was sensitive and specific for the detection of patients with high ICP. In a prospective study, Robba et al. [7] indicated that ONSD was the best non-invasive method to estimate ICP. In a meta-analysis, Robba et al. [13] demonstrated that the ultrasonographic measurement of the ONSD had a high accuracy in the diagnosis of intracranial hypertension. A recently published meta-analysis enrolled 779 patients from 22 studies and suggested that the ONSD had high sensitivity and specificity in diagnosing patients with high ICP [23]. Consistent with these studies, this study revealed that the ONSD and ICP values obtained in the supine position on admission were strongly correlated (r = 0.799), indicating that the ONSD can accurately reflect the ICP.
The head-up position has been used for decades in the treatment of patients with
high ICP, as this position decreases ICP quickly [24]. In this study, when
patients changed from the supine position to the 30
The impairment of the ONS elasticity could be used to explain the results of the
present study. Usually, high ICP results in the transfer of CSF into the ONS,
causing an enlargement of the ONSD [13]. When the ICP decreases, the CSF will
move back into the intracranial subarachnoid space, and the ONS will shrink to
its initial diameter. However, these changes require normal ONS elasticity [12].
Hansen et al. [27] isolated human optic nerve preparations to explore
the elastic properties of the ONS and found that the reversibility of the ONSD
may be impaired after episodes of prolonged intracranial hypertension [27]. In
this study, when the patients changed from the supine position to the
30
This hypothesis was further confirmed by the dynamic test performed in the
present study. From day one to day three, a good agreement between the ICP and
ONSD only existed in three patients (18.8%). In addition, three patients had
completely different profiles for the ICP and ONSD. This result demonstrates that
the ONSD may not be suitable for dynamically monitoring the ICP. However, Wand
et al. [20] indicated that ONSD measurements are a useful tool for
dynamically evaluating ICP and found that changes in ICP and ONSD values were
correlated (r = 0.669) from admission through follow-up. In addition,
Wang et al. [20] had a long follow-up period (one month) and the maximum
ICP value was only 400 mmH
The present study had several limitations. First, it was limited by its small sample size, especially in the dynamic test (16 patients). Larger multicentre studies should be conducted to confirm this study’s findings. Second, the follow-up time was too short, as the dynamic test was only conducted from the day of admission until day three after admission. A study with a longer follow-up time should be performed to determine if ONSD can be used to dynamically monitor ICP. Third, in this study, the reasons for the ONSD being unsuitable for dynamically monitoring ICP were not clarified. The impairment of ONS elasticity and whether it contributed to the findings should be investigated further. Fourth, this study may have been subject to bias (e.g., spectrum bias was likely as most patients had intracranial haemorrhage). Finally, this study did not focus on the difference in the ONSD measurements for each eye or the variance in ONSD measurements in the transverse and sagittal axes. For ONSD measurement, a consensus was not reached regarding the axis to be measured or the eye to be used [28]. These issues should be investigated in future studies.
In the present study, the ONSD and ICP values obtained in the supine position on
admission were strongly correlated. However, the change in ICP was not strongly
correlated with the change in ONSD when patients changed from the supine to the
30
GBW, JT, JYG conceived of the study, and XBL and ZYW participated in its design and coordination and JYG helped to draft the manuscript. All authors read and approved the final manuscript.
This study was conducted in accordance with the Declaration of Helsinki and approved by the ethics committee of The Third Hospital of Hebei Medical University.
Not applicable.
This research was funded by Key R&D program of Hebei Science and Technology, grant number 19277763D.
The authors declare no conflict of interest.