OPEN Research Support

Frantz Rom Poulsen
Department of Neurosurgery, Odense University Hospital

Projekt styring
Projekt status    Sampling ongoing
Data indsamlingsdatoer
Start 01.06.2017  
Slut 01.06.2022  

Estimation of intracranial pressure using non-invasive fundus images

Short summary

Arteriole and venule diameter ratio (A/V-ratio) can be measured using fundus photography. In this pilot study, we will correlate changes in the intracranial pressure with the diameter of vessels of the retina.

We hypothesize a correlation between A/V ratio and ICP and by that will suggest an easily obtainable and usable point-of-care (POC), non-invasive method to estimate the intracranial pressure without the necessity of mydriatic drugs.


It is well-known that intracranial pressure (ICP) ranges from 5-15 mmHg at rest in a healthy adult. The threshold value for intracranial hypertension varies not only throughout the literature but also in clinical practice. In general, ICP > 20 mmHg is considered pathological and often requires treatment. ICP monitoring is considered paramount in diseases like intracranial hemorrhages, traumatic brain injury, subarachnoid hemorrhages, malignant infarction, cerebral edema and infections of the central nervous system to optimize treatment of secondary brain injuries and thereby improve outcome.

Neuromonitoring still heavily relies on invasive methods such as parenchymal or ventricular neuromonitoring (external ventricular drain, EVD), both placed through a cranial burr hole. Invasive procedures, however, carry the risk of bleeding, infection and brain damage.

Furthermore, invasive intracranial pressure tools are prone to device failure due to baseline drift of the probe in the tissue and breakage of equipment, and the invasive nature of reliable ICP-monitoring equipment makes them unsuitable for usage in settings outside neurosurgical intensive care units. Non-invasive techniques have been proved reliable. To date, the most promising non-invasive methods studied are pupillometry and optic nerve sheath diameter. However, they have not been validated for daily clinical use and are prone to a high degree of inter-observer variations. There is therefore a need for further research in other non-invasive intracranial pressure modalities.

In the United States and Europe, the incidence of traumatic brain injury (TBI) is 500/100,000 and more than 200/100,000 patients are admitted due to TBI related incidence in Europe. A quick, non-invasive and easy-to-use screening tool for intracranial pressure would be of great benefit in such and other cases, where elevated intracranial pressure can be a potential hazard.

A relationship between acute elevation in ICP and retinal venule dilation was shown in animal models in the 70's. ICP affects intracranial arteries and venules and thereby influences the draining and feeding vessels to the eye, which in turn under higher ICP can cause papilledema by retrobulbar compression. Guidoboni et al. has published a theoretical work, which describes mathematically the association between vessel diameter, intraocular pressure and optic nerve tissue pressure and by that ICPs involvement.

Furthermore, the intracranial hypertension treatment trial (2015) reported a decreasing retinal A/V-ratio in correlation with the severity of papilledema, which suggests an association between prolonged increased ICP and A/V-ratio.

We here propose a novel, cheap, fast and easy measurement using non-invasive fundoscopy on non-dilated pupils to assess ICP by using retinal A/V-ratio. ( ID NCT03382860).

Description of the cohort

The study is an observational, prospective and single-blinded proof-of-concept study of patients from a neurosurgical department investigated for suspected normal pressure hydrocephalus (NPH) or ventriculoperitoneal (VP) shunt dysfunction. Inclusion criteria are: 18 years or older and competent, informed consent and indication for intraparenchymal, intraventricular or lumbar intracranial pressure (ICP) monitoring.

Data and biological material

We perform fundoscopy with the EpiCam M, which is a handheld retinal fundus camera with a monochromatic image sensor (Class 1 medical device and complies with ISO 10940:2009) and capture images and videos. Furthermore, we collect intracranial pressure via a pressure system, which correlates with the time of the videos.

All demographic and clinical data were either collected from the patient's medical records or bed-side. Data collected included: age, gender, blood pressure at time of data collection, past medical history, medication, smoking and indication for ICP monitoring.

Collaborating researchers and departments

Department of Neurology, Odense University Hospital

  • Consultant Dagmar Beier, Dr.Med.

Department of Neurology, Glostrup, Rigshospitalet

  • Professor, Consultant, Dr. Med. Rigmor Højland Jensen

StatuManu Company

  • Engineer Jakob Find Madsen