The assessment of intracranial pressure (ICP) is essential in the management of neurocritical care paediatric patients. The gold standard for invasive ICP is an intraventricular catheter or intraparenchymal microsensor but is invasive and carries some risks. Therefore, a non-invasive method for measuring ICP (nICP) would be desirable especially in the paediatric population. The aim of this study is to assess the relationship between ICP and different ultrasound–based methods in neurocritical care paediatric patients.
Children aged < 16 years with indication for invasive ICP monitoring were prospectively enrolled. The following non-invasive methods were compared with the invasive gold standard: optic nerve sheath diameter ultrasound (ONSD)–derived nICP (nICPONSD); arterial TCD–derived pulsatility index (PIa) and a method based on the diastolic component of the TCD cerebral blood flow velocity and mean arterial blood pressure (nICPFVd).
We analysed 107 measurements from 10 paediatric patients. Results from linear regression demonstrated that, among the nICP methods, ONSD has the best correlation with ICP (r = 0.852 (p < 0.0001)). Results from receiving operator curve analysis demonstrated that using a threshold of 15 mmHg, ONSD has and area under the curve (AUC) of 0.94 (95% CI = 0.892–0.989), with best threshold at 3.85 mm (sensitivity = 0.811; specificity = 0.939).
Our preliminary results suggested that ONSD ultrasonography presents the best accuracy to assess ICP among the methods studied. Given its non-invasiveness, repeatability and safety, this technique has the potential of representing a valid option as non-invasive tool to assess the risk of intracranial hypertension in the paediatric population.
Intracranial hypertension (ICHP) is a devastating complication after brain injury, and its assessment is essential in the management of acute intracranial catastrophe to limit or actively reduce it .
A frequent question in neurocritical care of children relates to the indications and methods for ICHP measurement and treatment. Direct measurement of intracranial pressure (ICP) using an external ventricular drain (EVD) or intraparenchymal ICP microsensor represents the gold standard for confirming the presence of intracranial hypertension . In children, the simplest and longest-standing method of measuring ICP is to perform a lumbar puncture (LP) and to observe the opening and closing pressure. However, intracerebral probes have several risks and complications and LP is an indirect and, at the same time, imprecise procedure [2, 6, 21].
Therefore, invasive measurement of ICP is reserved for the most severely affected children in whom the benefits of direct measurement outweigh the risks of bleeding and infection .
In this setting, a non-invasive method to assess ICP would be crucial for the management of these patients. Despite several authors attempted to find a non-invasive method to measure ICP, there is not currently an absolutely accurate method available . However, ultrasound-based non-invasive methods are gaining interest and have shown promising results, especially in adult populations [22,23,24,25,26,27].
The aim of this study is to assess different ultrasound–based methods to assess ICP non-invasively (nICP) by comparing them with the invasive gold standard in a paediatric population. Using arterial transcranial Doppler ultrasonography (TCD), we assessed the arterial pulsatility index (PIa) and a method based on diastolic cerebral blood flow velocity (FVd), even if it was not primarily intended to estimate ICP but CPP. Using ultrasound, we assessed the optic nerve sheath diameter (ONSD). Also, we aimed to test an ONSD-derived formula for the assessment of ICP (nICPONSD), previously described in the adult population .
The main institutional review board (IRB) of Alessandria Paediatric Hospital, Italy, approved the study (entry code ASO.RianPed.16.04). Detailed written informed consent was provided to the family members (parents or legal tutors) regarding the study protocol, the scope of research and the safety of TCD examination.
Data were collected prospectively from children younger than 16 years with severe brain injury requiring ICP monitoring and admitted to the hospital between 1st July 2015 and 1st January 2018. Exclusion criteria were the absence of an informed consent, a history of ocular pathology or optic nerve trauma, skull base fracture with a cerebrospinal fluid (CSF) leak and inaccessible ultrasound temporal window. Demographic data, Glasgow Coma Scale (GCS) at admission and the paediatric version of Glasgow Outcome Score (GOS-E PEDS)  at discharge from intensive care unit were recorded.
Ultrasound measurement was performed by a selected group of trained operators (CR, FR, SP); the operators used a standardized insonation technique in order to reduce inter-operator variability and were blinded to the patient’s clinical background.
For each patient, we recorded bilateral middle cerebral artery (MCA) flow velocities (systolic [FVs], mean [FVm] and FVd), ONSD and mean arterial pressure (ABPm) twice daily (from day 1 to 5 post ICP probe insertion). Additional measurements were performed in case of acute changes in ICP (higher than 20 mmHg), as previously described . The final FV values were calculated as the average of the right and left MCA measured values.
Transcranial Doppler examinations were performed using a standard TCD machine with a 2-MHz probe as previously described [25, 26]. Simultaneous measurements of ICP were taken at the time of the study. All physiologic parameters were kept stable during the TCD examination.
A linear 7.5-MHz ultrasound probe was used to perform ultrasound examination of the ONSD using a mechanical index (MI) below or at 0.3. Once the image was obtained, it was frozen, the probe removed from the eyelid and all measurements were subsequently performed. Ultrasound gel was applied on the surface of each eyelid. The probe was oriented perpendicularly in the vertical plane and at around 30 degrees in the horizontal plane on the closed eyelids of both eyes of the patient. The measurements were made in both the axial and sagittal planes 3 mm behind the retina in both eyes, and the final ONSD value was calculated as the average of four measured values.
We took an average of 4 measurements per eye and then in the next step the average of both eyes. Non-invasive ICP derived from ONSD was estimated according to the regression analysis between ICP and ONSD obtained from our previous study in a cohort of adult TBI patients :
Statistical analysis of the data was conducted with R Studio software (R version 3.1.2). Data were tested for normal distribution using the Shapiro-Wilk test and are presented as median (interquartile range [IQR]). All parameters assessed were non-parametric in nature. Multiple measurements were considered as independent values.
The correlations between direct ICP and the non-invasive estimators were verified: ONSD, PI, nICPFVd using the Spearman correlation coefficient (R, with the level of significance set at 0.05).
The Bland-Altman method was used to determine the agreement between invasive ICP and nICP estimation methods, with 95% confidence interval for prediction (CI) and bias . The area under the curve (AUC) of the receiver operating characteristic curve (ROC) were performed to determine the ability of the best-performing non-invasive method to detect raised ICP (using a threshold of 15 and 20 mmHg).
A total of 107 measurements from 10 consecutive patients were included in this study. One patient was excluded for the absence of informed consent. The characteristics of the patients are described in Table 1. Table 2 presents median (IQR) for the variables assessed. Results from the regression analysis revealed good correlation between invasive ICP and ONSD, r (Spearman) = 0.852 (p < 0.0001) (Fig. 1). The use of the nICPONSD produced the same correlation coefficient as ONSD itself (r = 0.852, p < 0.0001), although the formula resulted in an underestimation of ICP. A significant but not strong correlation was also found between nICPFVd formula and ICP: (r = 0.441, p < 0.0001) and PIa and ICP (r = 0.321, p < 0.001).
Figure 2 shows the behaviour of ICP and ONSD for each patient during the study. Bland-Altman analysis using nICPONSD formula found a bias of − 5.93 mmHg, with 95% confidence interval (CI) for ICP prediction of ±8.33 mmHg. nICPFVd had a bias of 7.91 mmHg and a 95% CI of ±16.26 mmHg (Fig. 3).
Results from ROC analysis demonstrated that using a threshold of 15 mmHg, ONSD had AUC of 0.94 (95% CI = 0.892–0.989), with best threshold at 3.85 mm (sensitivity = 0.811; specificity = 0.939). Considering a threshold of 20 mmHg, AUC was 0.976 (95% CI = 0.948–1.00), with best threshold at 4.75 mm (sensitivity = 0.956; specificity = 0.938) (Fig. 4).
According to our results, in the paediatric population, ONSD appears as the best estimator of ICP compared to the other methods. At our knowledge, this is the first report comparing different non-invasive ultrasound–based methods with simultaneous direct ICP measurements in a paediatric population. Similarly to our results, in our previous study in an adult cohort of TBI patients, we found that ONSD had a good correlation with invasive ICP, whilst PI and nICPFVd had poor correlation with ICP .
An elevation in intracranial pressure can be a surgical or medical emergency . A recent study  demonstrated that the magnitude and the duration of ICP insult, called ‘dose of ICP’ concept is strongly correlated with the patients’ outcome in both the paediatric and the adult population. In this study, the transition curve of ICP, considering the duration and the intensity of the insult for the paediatric cohort, resembled that of adults. This finding indicated that in children, episodes of lower intensity and shorter duration of intracranial hypertension are associated with worse outcomes compared to adults. Also, in children, the exponential decay transition curve approximates the vertical 10-mmHg line, and above 20 mmHg, the association with worse outcome occurs within 8 min.
There are many possible conditions that can determine elevated ICP in children, including traumatic brain injury, intracerebral haemorrhage, shunt insertion or malfunction, arachnoid cyst and craniosynostosis . However, current guidelines are not clear about indications for invasive ICP measurements. In adults, the latest Brain Trauma Foundation guidelines  have downgraded previous recommendations on ICP monitoring, as evidence did not meet current standards anymore, and the lack of clear indications on ICP monitoring causes significant differences in the clinical practice among centres. For children, guidelines suggest ICP monitoring in severe TBI patients with low GCS (≤ 8), and also highlight the importance of ICP monitoring in conscious children at risk for neurologic deterioration as a result of traumatic mass lesions, or in whom serial neurologic examination is precluded (sedation, neuromuscular blockade or anaesthesia) .
The gold standard for ICP measurement is an external ventricular drain or an intraparenchymal probe [28, 32]. However, invasive monitoring carries several risks including bleeding and infection [2, 21] which preclude their use in some clinical condition (like coagulopathy). Moreover, there are several scenarios where invasive ICP monitoring is not indicated, but a nICP assessment would be useful (e.g. mild or moderate traumatic brain injury, meningitis, hydrocephalus, metabolic coma).
Various methods for nICP monitoring have been described in adult and paediatric populations [23, 32]. Among these, ultrasound-based non-invasive methods are gaining popularity as they are safe, easily available and low cost.
TCD assessment of the cerebral blood flow velocity in the basal cerebral vessels has long been used as a non-invasive surrogate measure of cerebral blood flow and ICP. Pulsatility index has been one of the most used TCD-derived nICP estimation methods. Some authors have reported strong association between PI and ICP ; however, such findings have not been replicated by other studies, which reported poor correlations between these parameters in adult TBI populations [8, 27, 33].
There are few reports on the relationship between PI and ICP in children with TBI. Figaji et al.  in a cohort of 34 children found a weak relationship between mean values of ICP and PI (r = 0.36, p = 0.04). Similarly, our findings demonstrate that PI was not strongly correlated with ICP. In experimental models of intracranial hypertension, nICPFVd has shown promising results , and in a recent pilot study Rasulo et al.  demonstrated in a cohort of 38 adult brain–injured patients that nICPFVd may accurately exclude intracranial hypertension in patients with acute brain injury, showing a sensitivity of 100% and a specificity of 91.2% for a threshold of 20 mmHg. In our group of patients, we found a significant but weak correlation between nICPFVd and ICP; however, at our knowledge, our study is the first one assessing this estimation method in children and needs further validation.
The optic nerve is part of the central nervous system. A rise in ICP causes an increase of the diameter of the perioptic subarachnoid space within the dural optic nerve sheath (ONS), which leads to an increase in optic nerve sheath diameter (ONSD) [24, 29]. Transorbital sonography represents a potentially useful and safe method to measure the ONSD for rapid diagnosis of ICHP in infants.
The upper limit of the normal range for optic nerve sheath diameter is 4.5 mm in patients over 1 year of age, and 4 mm in children. In a study including healthy volunteers, the range of ONSD was 2.1–4.3 mm (mean 3.08 ± 0.36), suggesting that an ONSD greater than 4 mm in infants younger than 1 year old, and ONSD ≥ 4.5 mm in older children, should be considered abnormal . In adults, the threshold of ONSD is between 5 and 6 mm, and this might explain why the use of nICPONSD estimation formula previously described in an adult population results in underestimation of the ICP prediction in children [ 11, 17, 18, 27, 31].
This discrepancy of our ONSD values compared to previously published values  might be the result of a low resolution of ultrasound transducer used in our study (just 7.5 MHz), since most authors more recently have used > 10 MHz.
The application of ONSD in children has been investigated in different clinical scenarios associated with intracranial hypertension, such as acute hydrocephalus [14, 17, 18]. Padayachy et al. [19, 20] reported a good correlation between ONSD and ICP in a cohort of brain-injured paediatric patients (r = 0.66, p < 0.001), with excellent repeatability and intra-observer variability (α = 0.97–0.99). Testing for inter-observer variability revealed also good correlation [14, 19]. However, a limitation of that study was that ONSD measurement was performed prior to invasive measurement of ICP.
In our report, ONSD showed an even better association with ICP, comparable with the results previously described in the adult population . Furthermore, ONSD in the paediatric population also had the best accuracy when compared with TCD-derived nICP methods.
We tried to apply our formula validated in a cohort of adult brain–injured patients for the direct estimation of ICP (ICPONSD). As shown in Fig. 2, there is a large variability of ICP and ONSD within the 10 patients. Also, our results show that there is an underestimation of about 6 mmHg by the formula (bias of − 5.93 mmHg). These preliminary data show that a formula cannot be easily transferred from one cohort to another (especially different populations) and should not be used in the paediatric clinical practice.
Furthermore, it can be inferred from the repeated measurements of ICP/ONSD in 10 patients that the ICP/ONSD relationship is highly individual. Therefore, our understanding is that a formula could be valid for an individual, but not for an entire cohort.
There are several limitations that need to be mentioned. This is an observation pilot study whose major limitation is the very small sample size. Further prospective studies with a larger number of patients will be needed to confirm these preliminary results.
Moreover, cerebral blood flow velocity measurements using TCD were not continuous, precluding the assessment of nICP changes in time domain 23,27.
Finally, most of our measurements were obtained in patients with relatively well-controlled ICP. Also, larger validation studies containing a wider range of intracranial pressure values will be required to assess and validate these results before introduction of ONSD for nICP estimation in paediatric clinical practice.
Non-invasive ICP assessment using ultrasound methods are an attractive option in the paediatric population. These methods are quick, safe, repeatable and allow bedside monitoring in different clinical scenarios (neurocritical and general intensive care, operating room, emergency department).
In this preliminary report, we found that among the studied methods, ONSD presented the best accuracy when compared with other TCD-derived nICP methods and could potentially be useful as non-invasive screening tool when invasive methods are not available or contraindicated. However, ONSD has several limitations and further studies will be needed to confirm our results and assess its role in paediatric clinical practice.
Ballantyne J, Hollman AS, Hamilton R, Bradnam MS, Carachi R, Young DG, Dutton GN (1999) Transorbital optic nerve sheath ultrasonography in normal children. Clin Radiol 54:740–742
Bauer DF, Razdan SN, Bartolucci AA, Markert JM (2011) Meta-analysis of hemorrhagic complications from ventriculostomy placement by neurosurgeons. Neurosurgery 69:255–260
Beers SR, Wisniewski SR, Garcia-Filion P, Tian Y, Hahner T, Berger RP et al (2012) Validity of a pediatric version of the Glasgow Outcome Scale–extended. J Neurotrauma 29:1126–1139 Available: http://online.liebertpub.com/doi/abs/10.1089/neu.2011.2272
Bellner J, Romner B, Reinstrup P, Kristiansson KA, Ryding E, Brandt L (2004) Transcranial Doppler sonography pulsatility index (PI) reflects intracranial pressure (ICP). Surg Neurol 62:45–51
Carney N, Totten AM, O’Reilly C, Ullman JS, Hawryluk GWJ, Bell MJ, et al: Guidelines for the Management of Severe Traumatic Brain Injury, Fourth Edition Neurosurgery 0:1–10, 2016
Czosnyka M (2004) Monitoring and interpretation of intracranial pressure. J Neurol Neurosurg Psychiatry 75:813–821 Available: http://jnnp.bmj.com/cgi/doi/10.1136/jnnp.2003.033126
Czosnyka M, Matta BF, Smielewski P, Kirkpatrick PJ, Pickard JD (1998) Cerebral perfusion pressure in head-injured patients: a noninvasive assessment using transcranial Doppler ultrasonography. J Neurosurg 88:802–808
De Riva N, Budohoski KP, Smielewski P, Kasprowicz M, Zweifel C, Steiner LA et al (2012) Transcranial doppler pulsatility index: what it is and what it isn’t. Neurocrit Care 17:58–66
Figaji AA, Zwane E, Fieggen AG, Siesjo P, Peter JC (2009) Transcranial Doppler pulsatility index is not a reliable indicator of intracranial pressure in children with severe traumatic brain injury. Surg Neurol 72:389–394. https://doi.org/10.1016/j.surneu.2009.02.012
Güiza F, Depreitere B, Piper I, Citerio G, Chambers I, Jones PA, Lo TYM, Enblad P, Nillson P, Feyen B, Jorens P, Maas A, Schuhmann MU, Donald R, Moss L, van den Berghe G, Meyfroidt G (2015) Visualizing the pressure and time burden of intracranial hypertension in adult and paediatric traumatic brain injury. Intensive Care Med 41:1067–1076
Helmke K, Hansen HC (1996) Fundamentals of transorbital sonographic evaluation of optic nerve sheath expansion under intracranial hypertension. I. Experimental study. Pediatr Radiol 26:701–705
Hosmer D, Lameshow S (1989) Applied logistic regression. Wiley, New York
Kochanek PM, N a C, Adelson PD, Ashwal S, Bell M, Bratton SL et al (2012) Guidelines for the acute medical management of severe traumatic brain injury in infants, children, and adolescents - second edition. Pediatr Crit Care Med 13(Suppl):S1–S82 Available: https://www.braintrauma.org/uploads/08/08/guidelines_pediatric2_2.pdf
Le A, Hoehn ME, Smith ME, Spentzas T, Schlappy D, Pershad J (2009) Bedside sonographic measurement of optic nerve sheath diameter as a predictor of increased intracranial pressure in children. Ann Emerg Med 53:785–791
Luerssen TG (1997) Intracranial pressure: current status in monitoring and management. Semin Pediatr Neurol 4:146–155
Michel E, Zernikow B (1998) Gosling’s Doppler pulsatility index revisited. Ultrasound Med Biol 24:597–599
Newman WD (2002) Measurement of optic nerve sheath diameter by ultrasound: a means of detecting acute raised intracranial pressure in hydrocephalus. Br J Ophthalmol 86:1109–1113 Available: http://bjo.bmj.com/cgi/doi/10.1136/bjo.86.10.1109
Oestreich AE (2006) Sonographic evaluation of optic nerve diameter in children with raised intracranial pressure. Yearb Diagnostic Radiol 2006:180–181 Available: http://linkinghub.elsevier.com/retrieve/pii/S0098167208703709
Padayachy LC, Padayachy V, Galal U, Gray R, Fieggen AG (2016) The relationship between transorbital ultrasound measurement of the optic nerve sheath diameter (ONSD) and invasively measured ICP in children: part I: repeatability, observer variability and general analysis. Childs Nerv Syst 32:1769–1778
Padayachy LC, Padayachy V, Galal U, Pollock T, Fieggen AG (2016) The relationship between transorbital ultrasound measurement of the optic nerve sheath diameter (ONSD) and invasively measured ICP in children. Childs Nerv Syst 32:1779–1785
Poca M-A, Sahuquillo J, Arribas M, Báguena M, Amorós S, Rubio E (2002) Fiberoptic intraparenchymal brain pressure monitoring with the Camino V420 monitor: reflections on our experience in 163 severely head-injured patients. J Neurotrauma 19:439–448 Available: http://www.ncbi.nlm.nih.gov/pubmed/11990350
Rasulo FA, Bertuetti R, Robba C, Lusenti F, Cantoni A, Bernini M, Girardini A, Calza S, Piva S, Fagoni N, Latronico N (2017) The accuracy of transcranial Doppler in excluding intracranial hypertension following acute brain injury: a multicenter prospective pilot study. Crit Care 21:44 Available: http://ccforum.biomedcentral.com/articles/10.1186/s13054-017-1632-2
Robba C, Bacigaluppi S, Cardim D, Donnelly J, Bertuccio A, Czosnyka M (2015) Non-invasive assessment of intracranial pressure. Acta Neurol Scand
Robba C, Donnelly J, Bertuetti R, Cardim D, Sekhon MS, Aries M, Smielewski P, Richards H, Czosnyka M (2015) Doppler non-invasive monitoring of ICP in an animal model of acute intracranial hypertension. Neurocrit Care 23:419–426
Robba C, Cardim D, Donnelly J, Bertuccio A, Bacigaluppi S, Bragazzi N et al (2016) Effects of pneumoperitoneum and Trendelenburg position on intracranial pressure assessed using different non-invasive methods. Br J Anaesth 117:783–791 Available: http://bja.oxfordjournals.org/lookup/doi/10.1093/bja/aew356
Robba C, Bragazzi L, Bertuccio A, Cardim D, Donnelly J, Sekhon M et al (2016) Effects of prone position and positive end-expiratory pressure on noninvasive estimators of ICP : a pilot study. J Neurosurg Anesth:1–8
Robba C, Cardim D, Tajsic T, Pietersen J, Bulman M, Donnelly J, Lavinio A, Gupta A, Menon DK, Hutchinson PJA, Czosnyka M (2017) Ultrasound non-invasive measurement of intracranial pressure in neurointensive care: a prospective observational study. PLoS Med 14:e1002356
Robba C, Santori G, Czosnyka M, Corradi F, Bragazzi N, Padayachy L, Taccone FS, Citerio G (2018) Optic nerve sheath diameter measured sonographically as non-invasive estimator of intracranial pressure: a systematic review and meta-analysis. Intensive Care Med 44(8):1284–1294
Robba C, Goffi A, Geeraerts T, Cardim D, Via G, Czosnyka M, Park S, Sarwal A, Padayachy L, Rasulo F, Citerio G (2019) Brain ultrasonography: methodology, basic and advanced principles and clinical applications. A narrative review. Intensive Care Med
Stocchetti N, Maas AIR (2014) Traumatic intracranial hypertension. N Engl J Med 370:2121–2130 Available: http://www.nejm.org/doi/abs/10.1056/NEJMra1208708
Tsung J, Blaivas M, Cooper A, Levick N (2005) A rapid noninvasive method of detecting elevated intracranial pressure using bedside ocular ultrasound: application to 3 cases of head trauma in the pediatric emergency department. Pediatr Emerg Care 21:94–98 Available: http://journals.lww.com/pec-online/Abstract/2005/02000/A_Rapid_Noninvasive_Method_of_Detecting_Elevated.4.aspx
Wiegand C, Richards P (2007) Measurement of intracranial pressure in children: a critical review of current methods. Dev Med Child Neurol 49:935–941
Zweifel C, Czosnyka M, Carrera E, De Riva N, Pickard JD, Smielewski P (2012) Reliability of the blood flow velocity pulsatility index for assessment of intracranial and cerebral perfusion pressures in head-injured patients. Neurosurgery 71:853–861
DC and MC are partially financially supported by NIHR BRC Cambridge, UK, and DC is financially supported by a Cambridge Commonwealth European & International Trust scholarship. For the remaining authors, none was declared.
Conflict of interest
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Robba, C., Cardim, D., Czosnyka, M. et al. Ultrasound non-invasive intracranial pressure assessment in paediatric neurocritical care: a pilot study. Childs Nerv Syst 36, 117–124 (2020). https://doi.org/10.1007/s00381-019-04235-8
- Optic nerve sheath diameter
- Transcranial Doppler
- Pulsatility index