How to Zero an External Ventricular Drain (EVD)

Zeroing an external ventricular drain means calibrating the pressure transducer so that intracranial pressure readings reflect true values rather than artifacts of the system’s setup. The process hinges on aligning the drain’s reference point with a specific anatomical landmark, then opening the transducer to atmospheric pressure so the monitor reads zero at baseline. Getting this wrong by even a few centimeters can lead to dangerously inaccurate pressure readings, which in turn can cause too much or too little cerebrospinal fluid to drain. The procedure sounds simple, but the details matter more than most bedside tasks in neurocritical care.

Why the Tragus Is the Landmark That Matters

The entire zeroing process depends on choosing a reference point that corresponds to the foramen of Monro, the small opening between the lateral ventricles where cerebrospinal fluid (CSF) flows into the third ventricle. In a patient lying on their back, the foramen of Monro sits roughly at the level of the tragus, the small cartilage flap in front of the ear canal. Some institutions use the outer canthus of the eye or a point two-thirds of the way up from the ear to the top of the skull, but the tragus is the most widely accepted landmark because it is easy to find and reasonably consistent across patients.

The reference point is not just a suggestion. It defines where zero is for the entire system. If you level the transducer to a point higher than the actual foramen of Monro, the monitor will read lower than true intracranial pressure, and the drain may not open when it should. If you level it too low, the readings will be artificially high, and CSF may drain faster than intended. In practical terms, every centimeter of error translates to roughly 0.74 mmHg of pressure error. That might sound small, but when clinical decisions hinge on whether intracranial pressure is 18 or 22 mmHg, a few centimeters of misalignment can change the treatment plan entirely.

The Zeroing Procedure Step by Step

Before zeroing, the patient should be positioned with the head of the bed at whatever angle the clinical team has ordered, typically 30 degrees for most neurosurgical patients. The steps that follow assume a standard external drainage system with a pressure transducer, collection burette, and drip chamber.

Position the stopcock: Locate the transducer stopcock on the EVD system. This is the point where the system can be opened to atmospheric pressure.
Level to the tragus: Using a laser level or a carpenter’s level taped to an IV pole, align the transducer stopcock with the patient’s tragus. Many units mark the tragus on the patient’s skin with a pen to maintain consistency between nurses.
Turn the stopcock off to the patient: This isolates the transducer from the patient’s CSF column so you are measuring atmospheric pressure alone.
Open the transducer to air: Remove the nonvented cap from the stopcock port so the transducer is exposed to room air. The monitor should now read atmospheric pressure.
Press zero on the monitor: The bedside monitor’s zero function sets the current atmospheric pressure reading as the baseline. Once the display reads zero, the system is calibrated.
Replace the cap and reopen to the patient: Close the stopcock port, turn the stopcock back to its original position so the transducer is again reading the patient’s CSF pressure, and confirm a normal ICP waveform on the monitor.

That final step, confirming the waveform, is not optional. A proper ICP waveform has characteristic pulsations that correspond to the cardiac cycle. If the waveform is flat or dampened after zeroing, something is wrong: the catheter may be kinked, the stopcock may be in the wrong position, or the ventricle may have collapsed around the catheter tip. ICP waveform analysis and close monitoring of CSF drainage can directly affect clinical outcomes. ¹

Open Zeroing Versus Closed Zeroing

The procedure described above is the “open method,” where you physically remove the nonvented cap to expose the transducer to air. There is also a “closed method,” where the transducer is zeroed through the hydrophobic filter at the top of the burette without removing any caps. The closed method is appealing because it avoids opening the system to the environment, which in theory reduces infection risk. The transducer manufacturer’s instructions, however, direct clinicians to use the open method. ²

The accuracy debate between these two methods is more nuanced than you might expect. A blinded prospective comparison found that the closed method actually produced slightly better results: the mean difference from true pressure was 0.29 mmHg for the closed method versus 0.55 mmHg for the open method. More strikingly, the closed method was accurate within 1 mmHg of ground truth in 99% of trials, compared to about 86% for the open method. ³ A separate bedside trial also found equivalence between the methods when comparing ICP values obtained with and without cap removal. ⁴

Despite these findings, most published clinical practice guidelines and the critical appraisal literature still reference the open method as the standard approach for ensuring the most accurate physiological values for treatment decisions. ² The disconnect between the bench data and the guideline recommendations likely reflects how slowly clinical protocols change, along with the fact that most of the existing evidence base was built on the open technique. In practice, your unit’s protocol will dictate which method to use, and following it consistently matters more than picking the theoretically superior option.

How Drain Height Controls CSF Drainage

Zeroing the transducer tells you what the intracranial pressure is. But the height of the drip chamber above the reference point determines how much CSF actually drains. These are related but distinct functions of the system. The drip chamber is typically ordered at a specific height, often 10 to 20 cm above the tragus. CSF will only flow out of the ventricle and into the collection system when the intracranial pressure exceeds the hydrostatic pressure created by that column of fluid.

In engineering terms, the physics are straightforward: CSF flow rate increases with larger catheter diameters and lower system heights. A system set at 15 cm above the foramen of Monro produces the lowest flow rate across all catheter sizes, while a system set below the reference point produces the highest. ⁵ For any given catheter and height, higher intracranial pressure also drives proportionally more CSF out. This is why an incorrectly zeroed transducer is dangerous: if the reference point is set too low, the system behaves as though the drip chamber is lower than it really is relative to the ventricle, and more CSF drains than intended.

An interesting wrinkle in the clinical data, though, is that column height alone does not fully explain real-world drainage volumes. A study of actual patients found no statistically significant association between drain height category and mean daily CSF output, with similar volumes draining at heights ranging from below zero to 15 cm. ⁶ Patient factors like the underlying disease, brain compliance, and CSF production rate appear to play a larger role than simple physics would predict. The takeaway for bedside clinicians is that the ordered drain height is one piece of the puzzle, not the whole picture. Trending output volumes over time and correlating them with ICP readings gives a more complete view.

When You Need to Re-Zero

An EVD does not stay zeroed indefinitely. Transducers can drift, and any change in the patient’s head position relative to the transducer invalidates the calibration. Most protocols call for re-zeroing at the start of every nursing shift and whenever the patient’s position changes. Raising or lowering the head of the bed, turning the patient on their side, or transferring them to a different bed all require a fresh zero.

The reason is purely geometric. If the patient was supine when you zeroed and is now sitting up at 45 degrees, the tragus has moved relative to the transducer. Unless you relevel and re-zero, the ICP reading on the monitor is meaningless. Some experienced nurses develop a habit of checking the level every time they enter the room, which catches subtle shifts like a patient sliding down in bed or a family member adjusting the pillow. These small changes can introduce enough error to alter clinical decisions.

Units that use electronic leveling devices with alarms for positional drift have an advantage here, but the technology is not universal. In most hospitals, it still comes down to a nurse with a level and a consistent workflow.

Moving the Patient Without Losing Control

Intrahospital transport, whether to CT, the operating room, or an angiography suite, is one of the highest-risk periods for a patient with an EVD. The standard practice at many institutions is to clamp the drain during transport, which prevents uncontrolled CSF drainage when the system’s height relative to the patient inevitably changes during movement. However, clamping carries its own risk: intracranial pressure can rise unchecked while the drain is closed.

A study of cerebrovascular patients who underwent transport with clamped EVDs found that roughly 12% of transports resulted in post-transport ICP of 20 mmHg or higher, and about 19% were associated with an escalation in ICP category. All complications occurred in patients whose EVDs had been open and actively draining before transport; patients whose EVDs were already clamped in the ICU had no transport-related complications. ⁷ Pre-transport ICP was a strong predictor of trouble: patients with ICP at or above 20 mmHg before the trip were at substantially higher risk of complications.

The practical implication is that transport planning should include a check of the most recent ICP reading and a conversation with the team about whether the patient can safely tolerate a clamped drain for the expected duration. Once the patient arrives at the destination and is repositioned, the drain needs to be releveled and re-zeroed before being reopened. Skipping the re-zero after transport is one of the more common and consequential errors in EVD management.

Continuous Versus Intermittent Drainage

The drain height and zeroing procedure matter most when the EVD is open and actively draining. But not all patients have their drains open continuously. There are two broad approaches to EVD management, and they have different implications for how often you interact with the system.

In continuous drainage protocols, the EVD is left open at a prescribed height, typically starting around 10 cm H₂O, and CSF drains passively whenever intracranial pressure exceeds that threshold. Weaning in this model is gradual: the drain is raised stepwise, first to 15, then to 20 cm H₂O, and eventually clamped. If the patient tolerates clamping for 24 hours without developing signs of hydrocephalus, the drain is removed. ⁸

In intermittent drainage protocols, the EVD remains clamped by default. It is only opened briefly when the patient shows signs of elevated pressure, such as needing drainage more than twice per hour over a two-hour window, or when decompensated hydrocephalus is evident on imaging. Weaning in this model is faster: the drain is clamped outright rather than raised incrementally. ⁸

For zeroing purposes, the key difference is that a continuously draining EVD requires more vigilance about the reference point, because any drift in leveling directly affects how much CSF leaves the brain. An intermittent drain, being clamped most of the time, is less sensitive to small leveling errors between openings, though accurate ICP readings still depend on a properly zeroed transducer. In either mode, re-zeroing at the start of each shift and with position changes remains the standard.

Keeping the System Sterile

Every time you interact with the EVD system, whether zeroing, sampling CSF, or adjusting the drain height, you are handling a pathway that leads directly into the brain. Ventriculitis, an infection of the ventricular system, is one of the most feared complications of external drainage and can worsen outcomes dramatically. The infection risk is why the debate over open versus closed zeroing matters beyond accuracy: the open method temporarily exposes an access point to the environment, while the closed method keeps the system sealed.

Bundle protocols designed to reduce EVD-related infections emphasize strict sterile technique during any system manipulation. When CSF sampling is required, the port is accessed under aseptic conditions, with the cap discarded and replaced with a sterile one. The Luer fitting is cleaned multiple times with a chlorhexidine swab before and after the draw. ⁹ The same principles apply to zeroing: if you are using the open method and removing the nonvented cap, that cap should not be set down on the bedside table or left dangling. It needs to be handled as a sterile component. Some nurses hold it in their gloved hand during the brief zeroing procedure; others use a sterile field.

The Neurocritical Care Society’s consensus statement on EVD management emphasized the importance of evidence-based protocols that balance the benefits and risks of each aspect of drain care, including zeroing technique and system access. ¹⁰ Institutions that have adopted bundled care approaches, combining standardized zeroing procedures with strict access protocols and routine surveillance cultures, have seen meaningful reductions in infection rates.

Common Mistakes and How to Avoid Them

Certain errors show up repeatedly in EVD management, and most trace back to the zeroing process or its aftermath.

Forgetting to re-zero after repositioning: The most common source of inaccurate ICP readings. If the patient’s head moves and you do not relevel, the numbers on the monitor are wrong. Every position change triggers a re-zero.
Leveling to the wrong landmark: Using the top of the ear, the external auditory meatus, or an arbitrary point on the forehead instead of the tragus introduces systematic error in one direction. Marking the tragus on the patient’s skin helps maintain consistency across caregivers.
Leaving the stopcock in the wrong position: After zeroing, the stopcock must be returned to its original configuration. A stopcock left off to the patient means no ICP reading and no drainage. A stopcock left open to air means uncontrolled CSF loss.
Not confirming the waveform: A zero that looks successful on the monitor but produces a flat waveform indicates a system problem. Always verify pulsatile waveform morphology after completing the zero.
Ignoring transducer drift: Pressure transducers can drift over hours. Shift-to-shift re-zeroing catches this, but a transducer that drifts significantly within a shift may need replacement.

One subtler issue involves the drip chamber’s position on the IV pole. The chamber, not the transducer, determines the drainage threshold. Nurses sometimes confuse the two, carefully leveling the transducer to the tragus but leaving the drip chamber at whatever height it happened to land on. Both need to be set deliberately: the transducer to the tragus for accurate readings, and the drip chamber to the ordered height above the tragus for appropriate drainage.

Pediatric Patients and Anatomical Differences

In adults, the tragus reliably approximates the foramen of Monro. In infants and young children, the anatomy is different enough that the standard approach needs adjustment. The ventricles are proportionally larger relative to the skull, the fontanelles are still open, and head circumference can change meaningfully over days. The tragus remains the most commonly used landmark in pediatric practice, but some centers prefer to use the midpoint of the head measured from ear to ear.

The zeroing procedure itself is identical in principle, but two practical challenges arise. First, small children move unpredictably, which means the reference level changes constantly. Sedated patients are easier to manage, but even minor head turns in a restless toddler can shift the tragus by several centimeters. Second, the clinical significance of small pressure errors is amplified in smaller brains. A 2 mmHg error that would be clinically inconsequential in an adult with robust compensatory mechanisms can matter in a neonate with thin cortical mantle and limited reserve. Frequency of re-zeroing tends to be higher in pediatric units for this reason, with some protocols calling for checks every two to four hours rather than just at shift change.