TBI Guidelines for Mild, Moderate, and Severe Injury

Traumatic brain injury guidelines are a collection of evidence-based protocols that govern nearly every clinical decision after a head injury, from whether a patient needs a CT scan to how aggressively pressure inside the skull should be treated. The most widely referenced set comes from the Brain Trauma Foundation, which has refined its recommendations over roughly 25 years as new clinical evidence has emerged.¹ But “TBI guidelines” is not a single document. It is an ecosystem of protocols covering different severities, age groups, and clinical scenarios, produced by organizations around the world. Understanding what they recommend and where the science is still evolving can make a real difference in outcomes.

How Severity Is Classified

The starting point for almost every TBI guideline is the Glasgow Coma Scale, a 3-to-15 scoring system that rates a patient’s eye opening, verbal response, and motor response. The classic breakdown divides injuries into three tiers: mild (GCS 14–15), moderate (GCS 9–13), and severe (GCS 3–8). Each tier triggers a different management pathway, so getting the classification right matters enormously.

One long-running debate concerns patients who score a 13. The Advanced Trauma Life Support system reclassified GCS 13 as mild TBI, lumping those patients in with the 14–15 group. A study that directly tested this reclassification found that keeping GCS 13 in the moderate category produced better accuracy in predicting mortality. In other words, patients scoring 13 behave more like moderately injured patients than mildly injured ones, and treating them as “mild” risks underestimating their needs.² This is the kind of seemingly small classification shift that can change whether someone gets admitted for observation or sent home.

When Adults Need a CT Scan

One of the most impactful guideline areas involves deciding which patients with mild TBI actually need a head CT. Brain imaging catches bleeding and skull fractures, but CT scans expose patients to radiation, cost money, and clog emergency departments. Several clinical decision rules have been developed to sort out who genuinely needs imaging from who can safely skip it.

The most widely studied tools include the Canadian CT Head Rule, the New Orleans Criteria, and the NEXUS-II rule. A head-to-head comparison of six major decision instruments found that all of them were highly sensitive for detecting surgical bleeding, with sensitivities ranging from about 98% to 100%. Where they differed was in how many unnecessary scans they would generate. The Scandinavian guidelines required imaging for the fewest patients (about 50%), while the National Institute of Clinical Excellence criteria would have scanned roughly 71%. Sensitivity for catching any intracranial lesion ranged from about 96% for the Scandinavian approach to 100% for the NICE criteria.³

The practical trade-off is familiar in medicine: a rule that catches every dangerous bleed will also send a lot of people for scans they do not need, while a more selective rule may miss the occasional injury. Most emergency departments pick one instrument and apply it consistently. The key risk factors that trigger scanning across most rules include loss of consciousness, vomiting, age over 65, evidence of a skull fracture, seizures, and signs of worsening neurological function.

Pediatric Head Injuries and the PECARN Rule

Children present a special challenge because the risks of radiation exposure from CT are higher in younger patients, and kids are often difficult to assess clinically. The PECARN algorithm, developed from a large multicenter study, is the most widely validated tool for deciding which children can safely avoid a scan. It stratifies kids into risk categories based on age (above or below two years), mechanism of injury, symptoms like altered mental status or vomiting, and physical signs like scalp hematoma location.⁴

A French multicenter validation study tested the PECARN rule on nearly 1,500 children and found that its sensitivity for detecting clinically important TBI was 100%, with a specificity of about 70% and a negative predictive value of 100%. None of the nine children who had a significant brain injury were classified as very low risk by the algorithm.⁵ That track record has made PECARN the go-to rule in many pediatric emergency departments worldwide.

Even so, PECARN is not perfect in practice. A separate French study looking at how the rule was actually applied found that if PECARN criteria had been strictly followed, roughly a third of children would have needed a CT, about 29% would have needed clinical monitoring only, and around 38% could have been discharged. Among that last group, though, a small percentage still turned out to have intracranial hemorrhage or fractures, highlighting the gap between algorithmic guidance and real-world judgment calls.⁶

Blood Biomarkers as a CT Alternative

A newer addition to the TBI guideline landscape involves blood tests that measure proteins released when brain cells are damaged. Two biomarkers, GFAP and UCH-L1, have been cleared for clinical use in some settings. The idea is straightforward: if a blood test can reliably rule out brain bleeding, some patients can skip the CT entirely.

Early real-world results from a European emergency department that adopted the GFAP/UCH-L1 test into routine practice found that it spared over 40% of mild TBI patients from CT scanning, with the potential to reduce unnecessary imaging even further as clinicians became more comfortable following the test results.⁷ That is a meaningful reduction in radiation exposure and scanning costs.

The technology is still maturing, though. Other researchers have been more cautious, concluding that further investigation is needed before the blood test can be recommended as a standard complement to CT scanning.⁸ For now, TBI blood biomarkers sit in a transitional space: promising and already in clinical use at some centers, but not yet universally endorsed by major guideline bodies as a replacement for imaging.

Concussion and Mild TBI Recovery

The guidelines around mild TBI recovery have shifted dramatically in recent years. The old advice was prolonged rest in a dark room, sometimes for days or weeks. Current evidence has firmly moved away from that approach. Sustained rest from all activity after a concussion, sometimes called “cocoon therapy,” does not help recovery and may actually delay it.⁹

The current consensus across multiple guideline syntheses recommends relative rest for the first 24 to 48 hours, meaning reduced activity but not complete sensory deprivation. After that brief window, patients are encouraged to gradually return to normal daily activities, including screen time, at a pace that does not significantly worsen existing symptoms or trigger new ones.¹⁰ Physical and cognitive demands get ramped up in stages.

Emerging evidence goes further, suggesting that early active rehabilitation involving light aerobic exercise, kept below the threshold that triggers symptoms, may actually speed up recovery from mild TBI.¹¹ For athletes, return-to-play protocols follow a stepwise structure: symptom-limited activity, light aerobic exercise, sport-specific exercise, non-contact drills, full-contact practice, then game play, with each stage typically requiring at least 24 hours without symptom worsening before progressing.¹²

Critical Care for Severe TBI

When a brain injury is severe (GCS 3–8), the guidelines shift to intensive care protocols focused on preventing secondary damage from swelling, low oxygen, and uncontrolled pressure inside the skull. Intracranial pressure monitoring is a cornerstone. While no top-tier recommendation mandates it, ICP monitoring is considered a standard intervention for severe TBI. Recent studies have also suggested it should be considered for moderate TBI patients who show signs of rising pressure. In severe TBI, sustained ICP above 22 mmHg is associated with increased mortality.¹³

When ICP rises, the two main medical tools are mannitol and hypertonic saline, both of which work by pulling fluid out of brain tissue. Mannitol has long been considered the standard, but hypertonic saline is increasingly used. A review comparing the two in severe TBI found that both effectively reduced ICP across all studies examined, though results were mixed on which one worked better in any given situation.¹⁴

For cases where medical management fails, decompressive craniectomy, a procedure where a section of skull is temporarily removed to give the swelling brain room, is a last-resort option. Updated recommendations integrated findings from two major randomized trials (RESCUEicp and DECRA) and generated several formal recommendations for when this procedure is appropriate.¹⁵ It can be lifesaving, but it carries substantial risks and does not guarantee good functional outcomes.

Multimodal Monitoring and Brain Oxygen

Standard ICP monitoring tells you pressure is high but not always why. Newer approaches add brain oxygen monitoring, which measures how well the injured tissue is actually being perfused. A consensus conference developed treatment algorithms for patients who have both ICP and brain oxygen monitors, creating three separate protocols: one for high ICP with normal brain oxygen, one for low brain oxygen with normal ICP, and one for when both problems occur simultaneously. Each protocol uses a tiered approach, with higher tiers involving progressively riskier therapies.¹⁶

This kind of multimodal monitoring is not yet standard everywhere. It requires specialized equipment and expertise, and is mainly found at major trauma centers. But the logic is sound: treating pressure alone can sometimes push therapies in a direction that actually worsens brain oxygen delivery, so monitoring both lets clinicians make more nuanced decisions.

Seizure Prevention

Seizures after a severe brain injury are common enough that preventive medication has long been part of TBI guidelines. The evidence here divides cleanly into two time windows. Early post-traumatic seizures, those occurring within the first seven days, can be meaningfully reduced with antiseizure medication. A pooled analysis of high-quality trials found that phenytoin cut the risk of early seizures to about a third of what it would otherwise be. But for late seizures, occurring beyond seven days, preventive medication showed no benefit at all.¹⁷

More recent guidelines from the Neurocritical Care Society have shifted the preferred medication from phenytoin to levetiracetam, largely because of a more favorable side-effect profile, though the overall quality of evidence for choosing between them remains low. When seizure prophylaxis is used, guidelines recommend keeping it to seven days or less.¹⁸

Temperature Management

Cooling the body to lower brain temperature, known as therapeutic hypothermia, has been studied extensively in severe TBI. The idea is that lowering metabolic demand in the injured brain might limit secondary damage. Pooled data suggest that prophylactic hypothermia is not clearly associated with reduced mortality overall, although there are hints that maintaining lower temperatures for more than 48 hours may show a greater benefit. Interestingly, hypothermia does appear to be associated with better functional outcome scores compared to patients kept at normal temperature.¹⁹

A systematic review concluded that while the evidence is not yet definitive, hypothermia should be included as a therapeutic option to control ICP in patients with severe TBI, particularly when standard measures are not working.²⁰ The current guideline stance is cautious: hypothermia is on the menu, but not as a first-line treatment.

Managing Patients on Blood Thinners

Patients who are taking anticoagulants or antiplatelet drugs when they sustain a brain injury present one of the trickiest guideline scenarios. These medications increase the risk of intracranial bleeding expanding, but reversing them carries its own dangers, including clotting events like stroke. Dedicated guidelines from the Neurocritical Care Society and the Society of Critical Care Medicine have laid out reversal strategies for each drug class, covering vitamin K antagonists, direct oral anticoagulants, heparin products, and antiplatelet agents.²¹

An intersociety consensus document specifically addressing isolated TBI in patients on antithrombotic drugs proposed 28 statements covering common clinical scenarios. The panel reached agreement on about 71% of those questions, classifying some interventions as appropriate and others as inappropriate. The fact that nearly a third of clinical questions could not reach consensus underscores how thin the evidence still is in this area.²²

Rehabilitation and Persistent Symptoms

Most people with a mild TBI recover fully within weeks, but a subset develops persistent symptoms: headaches, difficulty concentrating, fatigue, sleep problems, mood changes. Guidelines addressing persistent post-concussive symptoms have produced dozens of recommendations spanning everything from vestibular therapy and vision rehabilitation to mental health treatment and return-to-work strategies.²³

Cognitive rehabilitation is one of the better-studied approaches. Treatment typically involves direct attention training, strategies for managing cognitive demands, assistive technology, and psychoeducational support. One study found that roughly 83% of clients achieved self-selected functional goals through these approaches.²⁴ For military veterans and service members, who have high rates of both mild TBI and co-occurring PTSD, integrated approaches that address both conditions simultaneously have shown particular promise.²⁵

One area that often gets missed is the endocrine system. Moderate-to-severe TBI can damage the pituitary gland, leading to hormonal deficiencies that mimic or worsen typical post-TBI symptoms like fatigue, low mood, and cognitive fog. Consensus guidelines recommend systematic screening of pituitary function in all patients with moderate-to-severe TBI who are at risk for these deficits, since appropriate hormone replacement can meaningfully improve rehabilitation prospects.²⁶

Does Following Guidelines Actually Help

This is arguably the most important question. Writing a guideline is one thing; demonstrating that following it improves patient outcomes is another. The evidence here is encouraging. A review of the literature on guideline adherence in severe TBI found that mortality improves with increasing compliance, and the benefits extend to functional outcomes and reduced length of hospital stay, in both adults and children.²⁷ A living systematic review of adult TBI patients came to a similar conclusion: adherence to Brain Trauma Foundation guidelines appeared to be associated with lower mortality.²⁸

In pediatric severe TBI specifically, a scoping review found that all identified studies reported improved survival and functional outcomes in association with guideline adherence, although the strength of the observation was limited by study designs that were observational rather than randomized.²⁹

The Implementation Gap in Lower-Resource Settings

The majority of TBI guidelines have been developed in, and for, well-resourced health systems. This creates a significant problem, because the burden of traumatic brain injury falls disproportionately on low- and middle-income countries. Healthcare workers in those settings often rely on guidelines developed elsewhere, but many recommended interventions may not be practical where infrastructure, equipment, trained personnel, and even basic supplies are limited. Clinical guideline recommendations need to be critically evaluated in the context of what is actually available on the ground.³⁰

ICP monitoring, for example, is central to severe TBI guidelines but requires invasive devices and continuous nursing care that may not be available at a district hospital in sub-Saharan Africa or rural South Asia. Multimodal brain oxygen monitoring is even further out of reach. The tension between evidence-based ideal management and what is feasible in a given setting is one of the largest unsolved problems in global neurotrauma care. Some efforts are underway to develop adapted protocols, but the field has a long way to go before guideline-based TBI care is anything close to universally accessible.