ASHOT E. SARGSYAN, MICHAEL BLAIVAS, THOMAS GEERAERTS and DIMITRIOS KARAKITSOS

Overview

Multipurpose ultrasound systems offer a variety of eye and orbit imaging techniques relevant to critical care. Bedside ocular ultrasound can be used for the following purposes: (1) timely identification of threats to ocular health and vision and prevention of vision-related disability in trauma,1–4 (2) assessment of nontraumatic acute change in vision and the need for immediate ophthalmologic consultation,¹ (3) measurement of the diameter of the optic nerve sheath (ONSD) as a surrogate measure for intracranial pressure (ICP),^5–9 (4) assessing the integrity of the bony orbit and periorbital tissues in severe craniofacial trauma,¹⁰ and (5) assessing the pupillary reactivity when visual access to the pupil is impeded.¹¹ However, ocular ultrasound applications are somewhat unique and not standardized at this time, whereas only ONSD measurements have been partially studied in the intensive care unit (ICU) environment. This chapter illustrates specific guidance for ultrasound equipment choice and adjustment, general ocular scanning techniques, and tips to maximize the utility of the ultrasound method for eye and orbit imaging.

Equipment and settings

In principle, any modern ultrasound system is acceptable for eye/orbit imaging if equipped with an appropriate transducer. At the time, multipurpose systems do not have either specialized transducers or factory-programmed “presets” for ocular imaging. To abide by the current safety standards, a machine should be chosen that displays the safety-related meta-data: mechanical index (MI) and thermal index (TI) on the screen. Older-generation machines do not display these values. Ocular scanning with those machines may be done only for vital indications and using the lowest possible power output. Furthermore, some results (e.g., ONSD measurements) may not be accurate because of old technology or wear and tear.

Because of the relatively shallow depths, small volume of the region of interest (ROI) and the very low attenuation in ocular media, the highest available frequencies are used, most commonly, a 12- or 14-MHz linear array transducer. Higher frequencies offer better resolution and smaller, easier-to-handle probes with a small contact surface. Lower frequencies must be used initially in severe edema of the area or a possibility of orbital fractures, globe rupture, or retrobulbar hematoma (e.g., a 5- to 8-MHz microconvex or a 4- to 7-MHz broadband linear transducer). This will ensure confident penetration without applying undue pressure on the transducer, with a commensurate compromise in spatial resolution.

The choice of transducers also depends on the scope of the examination. Some high-frequency transducers offer brilliant images of the soft tissues and anterior segment but do not focus well at depths greater than 2 cm. ONSD measurement with these probes and examination of the orbital apex area is not advisable. Thus frequency is not the only consideration. For these reasons, initial suitability assessment must be conducted by each facility that chooses to implement and use eye/orbit ultrasound in their practice, and for each machine and each associated transducer.

Equipment controls most frequently used in eye/orbit imaging are gain, TGC (time gain compensation), focus, and power output (see Chapter 1). Additional optimization techniques can be invoked as dictated by specific circumstances.

We highly recommend creating a customized eye/orbit preset that could be selected within seconds from the preset menu of the machine. This will save valuable time; ensure standardization; and serve as a safety, training, and quality assurance mechanism. An example is (1) select the factory-made thyroid preset from the preset menu; (2) adjust depth to 4 cm; (3) move the focus marker to 2.5 cm; (4) adjust MI to 0.23^12,13 with the power output control; (5) set TGC rheostats at their midrange; (6) scan a patient with indications for ocular ultrasound (e.g., to measure ONSD), and after identifying the terminal optic nerve, adjust gain appropriately; (7) save current settings under a new custom preset name (e.g., “eye-orbit”); (8) scan several patients to “fine-tune” the preset, and resave it with improved settings every time; (9) when satisfied, copy the preset to all machines in the department (if same model), or reproduce its features manually if machines are different. Make sure that additional setting changes do not increase MI; if they do, reduce power output accordingly. Note that perfect presets do not exist, and machine adjustments (at least, gain and TGC) will still be necessary in most patients.

General scanning technique and primary views

Eye and orbit scanning in the ICU is done with closed eyelids. The scope of the examination will depend on patient’s history, status, and open clinical questions. An ICU patient may need only a limited posterior pole and optic nerve examination. However, in trauma patients, the scope of the examination may be broader (see Chapters 1, 51, and 57), and thus a holistic approach (HOLA) ultrasound ocular protocol would include scanning of the eyelids, globe and orbital contents, as well as investigation of the frontal, temporal, zygomatic and nasal bones and overlying tissues. The eye examination proper, in its turn, may be directed at checking the overall integrity of the globe or at a more thorough scrutiny of its entire volume.

The operator holds the transducer close to the contact surface, touching the patient’s face with the fourth and/or fifth finger for stability. The following views are assuming a forward gaze; a straight gaze forward is highly desirable but may not be possible in unconscious or sedated patients. Directional terms used for ocular scanning are nasotemporal, superoinferior, and anteroposterior. A location within the eyeball (e.g., for a foreign body) is commonly identified in terms of the clock face (e.g., 1 o’clock) and in the anteroposterior (AP) dimension, relative to the following landmarks: macula, optic disk, equator, ora serrata, limbus, and corneal vertex; the distance from the optical axis or the wall is sometimes also used.

The AP views (sagittal and axial) are used for ocular anatomy survey and allow viewing practically all ocular structures to rule out macroscopic pathology (Figure 6-1A-C). The oblique sagittal view of the optic disk and terminal optic nerve is used to assess the shape of the optic disk, the terminal optic nerve with its meninges, and as a backup approach to measure ONSD in a sagittal plane (Figure 6-1E). This view may offer a possibility to visualize a longer span of the optic nerve by using a small rotation of the imaging plane. Note that the transducer is placed laterally to the cornea so that the imaging plane bypasses the anterior structures (cornea, anterior chamber, ciliary body, and the lens). The axial view from lateral approach allows avoiding the edge-shadowing artifact that in AP views obscures the peripheral zones of the vitreous humor and the globe wall near the equator (Figure 6-1D). The oblique axial view with optic disk and optic nerve head is the quickest and the most reproducible way to measure ONSD (Figure 6-2). The transducer position is on the upper eyelid (preferably with a downward gaze) with the marker to the right (for both eyes), and the plane bypasses the anterior structures. The resulting target image is similar to the sagittal view (see Figure 6-1E). The coronal view of the iris is used to assess the circular symmetry of the iris and the pupillary reactivity to light (Figure 6-3).

Trauma

The risks of eye injury are the highest in military combat¹⁴ but always present in both professional and daily human activity. Ultrasound evaluation of the eye is indicated in any case of trauma with the following findings: disturbance of vision of any extent, suspected breech of globe integrity with or without a foreign body, any abnormal finding during visual inspection of the globe, pain or any other persisting symptom, and edema or bruising of periorbital tissues and eyelids. Use of ultrasound with traumatic eye injuries is highly sensitive and specific when performed by an experienced operator using the appropriate transducer and technique.^4,15 We will not describe the specific imaging signs of traumatic eye injuries are not described in this chapter because they are widely available from other sources.¹⁶

Optic nerve sheath diameter measurements and new quality criteria

ONSD measurement is performed as a stand-alone ultrasound examination of eye and orbit in association with known, suspected, or anticipated ICP elevation. The protocol is reasonably simple and straightforward, allowing rapid bedside evaluation in acute conditions with rapidly rising ICP, such as traumatic brain injury and brain infectious disorders, or in emergency nonhospital settings, such as triaging of multiple individuals in mass casualty situations, including assessments in the field. In the ICU, trend monitoring of ONSD alterations is an additional demand, and therefore correct recognition of the imaged anatomic substrates, standardized conduct, high-resolution image acquisition and storage, as well as precision of measurements are expected to be the norm rather than excessive detail. The ICU-based ONSD evaluation can be considered a high-fidelity version of the largely rough ONSD assessment procedures used by a number of emergency medicine centers.

The intraorbital cerebrospinal fluid (CSF) space (i.e., the space between the dural sheath of the optic nerve and the pia-covered nerve proper) is contiguous with the intracranial CSF space.¹⁷ The authors of this chapter, as well as many others, have demonstrated a strong positive correlation between the ONSD and the ICP in animal and cadaveric models as well as in clinical observational trials, the latter primarily in head trauma or cerebrovascular accidents.^5–9,18,19 A leading notion of the published works was to demonstrate the correlation but also to find a “cutoff value” of ONSD that would indicate a pathologic level of ICP with high confidence. The suggested “cutoff,” or “threshold,” values have varied between trials in the range of 5.0 to 5.9 mm. In a meta-analysis of six trials involving 231 patients, Dubourg et al ²⁰ found ONSD to have “a good level of diagnostic accuracy for detecting intracranial hypertension” and suggested that ONSD technique can help make triage or invasive monitoring decisions. Notwithstanding the exhaustive statistical analysis of the data, and the resulting good agreement among the six highest-quality trials, their results remain applicable only to the patient groups involved and do not necessarily discern weaknesses associated with the imaging and measurement technique and the accuracy of recognition of the optic nerve sheath borders (see quality criteria below). A general weakness, besides the lack of standardization and quality control of ONSD measurements, is the lack of normal population data that would provide an alternative approach to ONSD data interpretation, because age, gender, race, anthropometric factors, medical history, and even professional background most likely influence the individual’s baseline (premorbid) ONSD. For these reasons, we strongly recommend exercising caution in using hitherto published threshold values in clinical decisions on the one hand, and making serious efforts in attaining ultrasound views with clearly visible borders of the arachnoid space, on the other. The equipment requirements and the ultrasound views and technique for ONSD measurement are described in previous sections. The first and default view for this purpose is the near-axial (oblique axial) view through the upper eyelid, with a straight-to-downward gaze (see Figure 6-2). In case of gaze deviation in superior (frontal) direction, the lower eyelid should be used. The lateral axial view should be reserved as a backup view if the previous two views fail. As a minimum, one high-quality view of the posterior pole with the terminal optic nerve must be obtained from each side.

The quality criteria for ONSD measurements have not been well described in the literature. Images of quality acceptable for ONSD measurement are provided in Figure 6-2. Our suggested quality criteria are summarized below:

ONSD values, if measured correctly, accurately reflect the anatomic dimension of the sheath at a specific point behind the globe. A high quality of measurements and consistent technique may facilitate monitoring the ICP trend in some patients, although previous evidence has been disappointing in this sense.⁷ This could be attributed to the fact that ONSD reaches a plateau at some point, and further distention may be too small to measure or be due to technical issues because ONSD measurements in early clinical studies have not been performed using refined methodology and according to strict image quality criteria. ONSD still holds promise for trend identification with increased measurement accuracy, at least in situations with rapidly rising ICP, when the difference in ICP between the two measurements is significant. One-time ONSD measurement should be considered with caution, and “cutoff” values reported in literature should be used as “alert levels” to prompt additional attention to ICP. The clinician must also remember that any ICU patient may have had a large ONSD before the current illness. As population distribution data become available and the quality standards are improved, better clinical criteria will likely emerge for both negative and positive prediction, and trend monitoring criteria may also be revised.

Pupillary light reflex assessment

The normal pupillary light reflex evaluation is usually described by the acronym PERRLA (pupils equal, round, reactive to light, and accommodation). The assessment of this fine reflex is part of the routine neurologic examination. The prognostic significance of pupil diameter and constrictive ability in brain-injured patients is well known and has been recommended by the American Association of Neurological Surgeons. Various pupillometric methods have been described, a majority requiring both specialized hardware and visual access to the pupil. The examination of pupil diameter and reactivity by ultrasound is a new alternative method for situations when visual access to the pupil is impossible or fine quantification of the pupillary reactivity is desired.

The method was initially developed for the U.S. Space Program and is not standardized for clinical use; only two successful case studies in trauma ICU patients have been published.^11,21 The method has no real alternatives when visual access to the pupil is impossible, and its results are self-explanatory. Consensual pupillary light reflex is elicited with contralateral transillumination through the eyelids with both eyes closed. The pupillary light reflex ultrasound test can be conducted with a linear-array probe at the highest available frequency (e.g., 12-15 MHz), using the coronal primary view described earlier (see Figure 6-3 and Video 6-1). The method can be combined with ONSD measurements to evaluate a potentially catastrophic intracranial process. The method is difficult in unstable gaze. In cases of severe facial trauma, extreme care should be taken to avoid additional damage to the globe and other tissues.

Additional specialist evaluation

Eye ultrasound data may reveal a need for ophthalmology consultation. Images may also be reviewed by a radiologist experienced in high-resolution ultrasound imaging of the eye and orbit. Because of the substantial differences in scope, methodology, and equipment used in ultrasound B-scans in ophthalmologic practice, imagery acquired by multipurpose ultrasound machines is not commonly referred to specialists in ophthalmic ultrasound.

Safety aspects

Diagnostic ultrasound has an excellent safety record. However, exposure of tissues to intense levels of ultrasound, well above those levels found with diagnostic ultrasound devices, has been shown to have damaging effects (see Chapter 1). Furthermore, delicate structures of the orbit, especially in trauma, can be damaged by mechanical force in case of unstable positioning, incautious movements of personnel, or involuntary movements of the subject. Use of irritant antiseptics for probe disinfection or inappropriate contact gels or lubricants can lead to chemical conjunctivitis. Failure to disinfect the transducer can potentially introduce pathogenic flora to the conjunctival sac. Therefore a set of safety rules is outlined below.

Following the ALARA principle (As Low As Reasonably Achievable), includes (1) controlling energy (having an “ophthalmic” preset with low acoustic output levels: TI less than or equal to 1.0 and MI less than or equal to 0.23; checking TI and MI upon each mode change; using two-dimensional gain, TGC, and other controls before increasing output); (2) controlling exposure time (minimizing data redundancy, minimizing “dwell time” [scanning while not looking]; deferring measurements until after procedure); and (3) controlling scanning technique (achieving gaze control, minimizing movement, eliminating distractions; using good ergonomics; following a plan with prioritization of views; controlling total duration of the procedure; deferring secondary views/objectives to the end of procedure; using sufficient amount of gel).

Observing chemical and biologic safety includes (1) using safe gel with safety data sheet favorable for eye contact and (2) disinfecting the transducer with 70% isopropyl or ethyl alcohol solution and air-drying before gel application. A 0.25% benzalkonium chloride solution may be used, but the residual amount of the disinfectant must be removed by applying and wiping off a small amount of gel. Other disinfectants such as Cydex are not allowed. (3) General skin hygiene is emphasized, and (4) the operator’s hand hygiene and gloves are mandatory.

Mechanical safety expects that (1) operator must assume a stable and comfortable position before commencing the procedure; (2) adequate time for scanning must be available to avoid haste; (3) nonessential personnel must stay at distance; (4) good transducer manipulation technique must be used, including hand anchoring, use of bony prominences of the subject’s periorbital area, and use of both hands as necessary to reliably control transducer position and prevent transducer pressure at all times; and (5) no transducer pressure may be used as a means to improve the image quality. All mechanical safety measures are emphasized in trauma, especially if there is a slightest possibility of globe rupture.

Thus we have presented equipment, technique, quality, and safety aspects of individual ocular imaging modules concerned with ocular as well as intracranial pathology. Because of limited space, we did not discuss other important ultrasound imaging opportunities, such as generic scanning of the orbit and periorbital areas for soft tissue and bone injury, transcranial Doppler (TCD), and a number of related ultrasound-based techniques that are not yet broadly accepted or standardized. Combinations of these and other techniques have been reported as promising in the assessment and monitoring of ICU patients. When following the holistic approach to ultrasound imaging in ICU, a facility may choose to devise an ultrasound protocol oriented toward intracranial pathology, which would combine TCD, optic nerve sheath measurement and possibly characterization, pupil reactivity recording and monitoring, as well as generic scanning of the facial tissues and orbital content in those with history of trauma. This chapter is as much a “how to” manual as an invitation to the reader to contribute to this specific area of critical care medicine by collecting and sharing their experience and moving together toward standardization and optimization of ultrasound use in ICU.

Pearls and highlights

References