Active distension of ventricular system due to mismatch between CSF production and absorption is called as hydrocephalus (7). Hydrocephalus can cause excessive buildup of CSF leading to increased ICP. Raised ICP can occur in conditions such as traumatic brain injury, meningitis, stroke, post resuscitation syndrome and intracranial tumors(8). Poor intracranial compliance can lead to catastrophic results such as hydrocephalus , uncal, central, transtentorial, or foramen magnum herniation (6). Early detection and intervention of raised ICP results in reduction of morbidity and mortality rates (6). The Brain Trauma Foundation recommends monitoring of raised ICP among patients with Glasgow Coma Scale of 3 to 8 and an abnormal computed tomography scan (6) (9). Patients had poor outcome when ICP exceeded 22 mm of Hg and better outcome when it was lesser than 15 mm of Hg. Serial change in ventricular size alone is not a very good indicator of raised ICP as normal ventricular size or small ventricle can have raised ICP (7).
Invasive methods of monitoring ICP are intraventricular catheter, extraventricular drains, extraparenchymal probes and subarachnoid bolts (6). The noninvasive methods of measuring raised ICP are transcranial Doppler usg, transcranical ultrasound propagation, ONSD, MRI, CT , measurement of dielectric property of cranium , magnetic angiography (6). ONSD is a well-accepted measure for early detection of raised ICP correctly. In recent days it is gaining popularity and is in practice at multiple centers.
Optic nerve sheath is continuation of dura and beneath it subarachnoid space continues along with optic nerve. Therefore when there is a rise in ICP it gets transmitted to optic nerve sheath eventually resulting in swelling of optic disc and papilledema (4) . ONSD as an extension of dura matter enclosing optic nerve was an established fact as early as 1780’s by Zinn and Wisberg (6). Optic nerve a part of central nervous system is covered by a leptomeningeal sheath is an expandable structure present at anterior segment behind the eye globe (6). With the rise in ICP CSF is pushed towards the tiny rim of subarachnoid space between the sheath and nerve causing expansion of dural covering. Previous studies have demonstrated that increase in ICP results in retrobulbar ONSD enlargement within seconds (10) (11). Expansion of sheath is more marked at anterior part of sheath behind the globe (6). The greatest degree of distension occurs 3 mm behind the globe (12). ONSD is usually measured by 3 radiological techniques: ultrasound, CT and MRI.
ONSD measurement by CT scan:
Traditionally, detection of raised ICP was performed during signs such as significant midline shift, effacement of basal cisterns, sulcal effacement, ventricular compression, and cerebral herniation as previously mentioned by Rajajee et al (4). Measurement of ONSD by CT to accurately diagnose raised ICP has been proved by previous studies (6) (13).
A retrospective observational study was conducted by Luyt et al. (8) on 56 patients to study relationship between ONSD measured on CT and raised ICP. Opening pressure manometry was measured during LP for detecting raised ICP (8). 14 patients had elevated ICP > 20 mm Hg and 42 had normal opening pressure. They showed that ONSD cutoff of 4.8mm by CT measurement had sensitivity and specificity of 92.9 % and 97.6%, respectively to detect the raised ICP. Furthermore, authors of this study mentioned that increasing the cut off value to more than 5mm decreased the sensitivity to 85%.
A retrospective cohort study was conducted by Sekhon et al. (13) on 57 TBI patients by measuring ONSD in CT scan and by invasive ICP. They found ONSD as better predictor of raised ICP than other parameters of CT scan such as cisternal effacement, sulcal effacement, ventricular compression and cerebral herniation. Linear and logistic regressions were used in study to see correlation between ONSD and invasive ICP values. Strong correlation between ICP and ONSD r = 0.74 p < 0.001 was found in the study. Liner regression between ICP and ONSD was R2 of 0.56 compared to R2 of 0.21 between CT features and ICP. Authors conclude ONSD has higher predictive value than other traditional signs for detecting raised ICP (13). Legrand et al. conducted an study among 72 patients to evaluate correlation between ONSD by CT scan and prognostic factors in traumatic brain injury patients (11). They found intial ONSD > 7.3mm to be independently associated with ICU mortality rate among TBI patients. Authors of the article suggest ONSD to be included with initial management of TBI patients along with other clinical, radiological and laboratory parameters (11). However they also point out higher ONSD does not necessarily rule out good outcome.
ONSD measurement by MRI:
Kang et al. (1) conducted a study to look for relationship between ONSD in brain MRI and elevated ICP. This study showed that patients with raised ICP have significantly greater diameter of ONSD compared to normal group. Out of 31 patients in the study, 13 patients with mass effect had average ONSD of right and left to be 5.06mm and 5.13mm, respectively. Similarly, patients without features of mass effect had ONSD of 4.545mm (Rt) and 4.35mm (Lt). Kang et al. cited hemodynamic instability and time as a restrictive factor for radiological examinations (1). Kang et al. concluded ONSD measurement by MRI as an efficient tool to predict increased ICP in an unconscious patient.
A study conducted by Geeraerts et al. (10) showed ONSD as an excellent predictor of raised ICP. 38 patients with TBI had undergone MRI in which ONSD was measured and ICP monitoring was done via parenchymal sensor. Results of the study showed ONSD to be 6.31mm +/- 0.5mm among patients with ICP > 20mm of Hg and ONSD of 5.29mm +/- 0.48mm among patients with ICP below 20mm of Hg. Authors say ONSD <5.30mm was unlikely to be associated with raised ICP, whereas an ONSD above 5.82mm was associated with a 90% probability of raised ICP (10). Singhal et al. conducted a study looking for change in ONSD among paediatric population with hydrocephalus who had undergone neurosurgical intervention (7). They assessed ONSD pre- and post-surgery by MRI among 9 patients who had undergone endoscopic third ventriculostomy and 7 patients who had undergone tumour resection (8). Authors of this study found significant decrease in ONSD after the intervention. The mean pre-operative ONSD was 6.1mm compared to mean post-operative ONSD of 5.71mm. Serial change of ventricular size is not a good indicator of uncontrolled hydrocephalus as raised ICP can occur even in normal ventricle size and smaller ventricles (8). They also claim serial measurement of ONSD and its reduction can be taken as a sign of improved hydrocephalus after the surgical intervention (8). Authors further say there was improvement in clinical signs and symptoms after intervention (7). ONSD measurement by ultrasound: A review article by Rosenberg et al. concluded ONSD and Doppler flow as the best noninvasive modalities for determining raised ICP (3). Ultrasonographic measurement of ONSD at a fixed distance from retina has been defined as a standard method of measuring the presence of raised ICP (14). The method of measuring ONSD by sonography using B mode was standardized by Hansel et al. in 1994 (12). It's a bed side procedure which can be easily done by 5 to 10 MHz linear ultrasound probe (3). It takes just 4 minutes to measure ONSD in both eyes (12). Measurement of ONSD requires trained personnel and sound experience to currently predict raised ICP (8). A study conducted by Rajajee et al. (4) found ONSD measurement by USG to be an accurate noninvasive method of detecting raised ICP of more than 20mm of Hg. Rajajee et al.(4) had conducted a prospective blinded observational study in which 536 ONSD measurements were performed on 65 patients with invasive ICP measurement side by side on these patients. Correlation between each ONSD measurement by USG and ICP measurement at the same time period was done in this study (4). This study also concluded that ONSD of 04.8mm as a cut off for raised ICP. Sensitivity with 4.8mm cut off value for ONSD was 96% and specificity was 94% for detecting ICP of more than 20cm of water. However, authors suggested that the ONSD values have to be validated for it to become true. They suggested discarding the risk of invasive monitoring and high cost technique of CT or MRI, and including noninvasive method of ultrasound as an alternative tool for detecting raised ICP. It is more cost-effective and less tedious (4). A study performed by Kimberly et al. (15) identified the correlation of ONSD with direct measurement of raised ICP. They conducted a prospective observational study in which invasive intracranial monitors were placed and ocular ultrasound was performed by 10 to 5 MHz linear probe among 15 individual patients. This study found a significant correlation between ONSD and ICP which was 0.59 calculated by spearman rank correlation coefficient test (15). The authors performed Receiver Operating Characteristic (ROC) curve test to assess the ability of ONSD to detect raised ICP. Authors of this study claim ONSD of greater than 5mm for detecting ICP of more than 20cm of water. Furthermore, the authors demonstrated ROC curve detecting sensitivity of 88% and specificity of 93% (15). They further suggested USG guided ONSD can be used as an noninvasive tool for measuring raised ICP (15). A systematic literature review and meta-analysis conducted by Ohle et al. (12) concluded both USG of ONSD and CT scan were equally effective in detecting raised ICP. The authors performed a full review on 45 articles out of 1214 available article searches until August 2013 and included 12 studies meeting inclusion criteria with a total of 478 participants. Quality assessment of papers was done by Quality assessment of diagnostic accuracy studies tool (12). Their analysis shows sonography of ONSD was 95.6% sensitive with 95% confidence interval (CI) of 87.7-98.5%. Also, it had specificity of 92.3% (95% CI of 77.9%- 98.4%) and positive plus negative likelihood ratio comparable to that of computed tomography. They concluded USG as a good diagnostic test accuracy compared to CT with high sensitivity for ruling out raised ICP especially in low risk group. Similarly, it also had high specificity than CT scan (12). Authors of this paper concluded that USG as a better modality as it is noninvasive, can be performed at bedside and saves time as there is no need to transport the patient to CT machine. This method provides more economic value and centers without CT machines could largely benefit through this technique (12). Newman et al. (16) conducted a study among shunted hydrocephalus children to evaluate the utility of ulrasonographic measurement of ONSD in children suspected of having raised ICP. In their study, 23 children who were shunted for hydrocephalus were included who gave clinical history suggestive of raised intracranial hypertension. ONSD of 102 control patient were also measured. The authors of this study (16) found that children with functioning VP shunt had lesser diameter of ONSD compared to those with non-functional shunt. This study showed the upper limit of ONSD was 4.5mm for children above 1 years of age. Similarly, children with functional VP shunt had mean ONSD of 2.9 (SD 0.5) mm and those with non-functional VP shunt had mean ONSD of 5.6 (SD 0.6mm) with p<0.0001. Newmen et al. (16) conclude USG guided ONSD as a tool to assess and monitor ICP in children with hydrocephalus. Zaidi et al. conducted a study to observe ONSD by USG as an screening tool for measuring raised ICP among shunt malfunctional patients (17). Authors mentioned that the most commonly used method to access the function of VPS is CT scan (17). CT scan causes radiation that has been found to cause harm resulting in lower cognitive test performances (17). Most patients with VPS undergo multiple CT scans. Malfunctioning VPS presents risk with very nonspecific symptoms such as sleepiness, vomiting, headache, and fuzziness among others (17). As MRI is not readily available, slow, expensive and may even require sedation specially for pediatric age group, this is still not the best screening tool for VPS malformation (17). Their study confirms ONSD is greater among the VPS malformation cases. Authors showed clinical correlation with signs and symptoms as equally important to diagnose malfunction and standardization of measurements by USG is important to minimize the inter-observer variation. They also showed ONSD as age and size dependent. A prospective study was conducted by Komut et al. (5) among 100 patients to check for efficacay of ONSD in determining raised ICP among non-trauma patients. Their study found that ONSD among patients with cranial pathology in CT scan was 5.4 +/- 1mm and among patients without cranial pathology in CT scan was 4.1 +/- 0.5mm. This study determined ONSD value of 4.5mm as a cut off value which had sensitivity of 80% and specificity of 60% for detecting raised ICP (5). Authors claimed ONSD can be used for determining midline shift as well. ONSD cut off value for determining midline shift was 5.3mm with sensitivity of 80% and specificity of 60% with area under the curve 0.728, Confidence Interval (95% 0.585 to 0.871). This study also showed ONSD to be larger in size at the corresponding lesion of brain with p<0.5 (5). ONSD was found to be higher among patients with GCS score of less than 13 compared to that having GCS more than 13; p < 0.5. Their study reported ONSD as an important method to detect raised ICP and its severity (5). Authors claim ONSD can be used even for non-traumatic cases in which there is a suspicion of increased ICP (5). It is difficult to estimate raised ICP by history and physical examination specially among patients who are unconscious and intubated (5). A study conducted by Pershad et al. analyzed cost effectiveness between CT, fMRI (fast sequence MRI) and USG measurement of ONSD. They found USG measurement of ONSD as cheaper and easier method (18). Pershad et al. mentioned clinical signs and symptoms of shunt failure are nonspecific and often neuroimaging is used as modality for the diagnosis (18). CT which is often being used for correctly diagnosing shunt failures. This results in exposure to ionizing radiation and has been linked with long term risk of malignancy (18). High negative predictive value of USG to detect raised ICP supports its use as screening tool (18). Preshad et al. suggested a screening test by USG as an intial method for detecting shunt failure and then to proceed for fMRI or CT scan (18). Toscano et al. conducted a retrospective study to analyze ONSD by USG among 21 critically ill patients with neurological disease who developed brain death during the course in ICU (2). Their study showed people with brain death had higher ONSD than normal control group even before having brain death. Authors concluded that daily ONSD measurement by USG among ICU patients could be useful to detect raise ICP earlier. They suggested this too to be very useful setup where invasive monitoring is not available (2).