What's New in Critical Illness and Injury Science? The role of a cranial computed tomography scoring tool in the care of older trauma patients.

The increasing longevity of the general population has created a parallel aging of the trauma population. This change in demographics has also lead to a change in injury patterns with falls and traumatic brain injury becoming an increasing proportion of the trauma burden. Because the birth of trauma stems from military experience, much of the lexicon of literature focuses on young and healthy patients – patients who typically are without significant co-morbidities. So the question arises, how do we predict the clinical course of older patients?

In their analysis,[1] the authors are able to correlate initial head CT findings to 30-day mortality, in-hospital mortality, in-hospital morbidity, and the need for neurosurgical intervention. The CT scan findings were scored using pre-specified criteria that many neurosurgeons use to determine of whether or not to perform invasive interventions. The authors did perform a matched pairs analysis between patients who did and did not receive neurosurgical intervention as well. However their matches were based on presentation within 6 months, gender, age within 5 years, and ISS within three points. The matches did not account for GCS on presentation or patients with matched cranial CT scores. While these matches may have accounted for underlying patient characteristics, they did not evaluate how matched patients with similar cranial CT scores did with and without neurosurgical intervention. Without this comparison, it is difficult to assess whether the cranial CT scores themselves are predictive of patient outcomes or whether the trauma burden was overall severe enough to warrant procedural intervention was the more likely correlate. In addition, the ages included in the study are 45 years and older, which many would argue does not fully reflect the patient population often referred to as “older patients.”

The scoring criteria for CT head proposed by the authors include subdural hematoma, midline shift, external trauma, brain edema, soft tissue injury or facial fractures, brain edema, subarachnoid hemorrhage, brain contusion/intraparenchymal hemorrhage, intraventricular hemorrhage, or epidural hemorrhage. When looking at the scoring system, it almost seems to align with the indications for neurosurgical intervention. With that in mind, asserting that both high scores on the initial CT head and neurosurgical intervention are both predictive of similar clinical outcomes is not such a revelatory statement. What is the utility in such a scoring system? Does the scoring system help in prognostication or early identification of patients who require surgical intervention? Would this scoring system obviate the need for a neurosurgery consult? What criteria were used to select the patients in this study for CT scanning? With healthcare currently trending towards algorithm and guideline based medicine and the utilization of physician extenders in the outpatient, emergency, and acute care settings, the development of a scoring system based on CT scan interpretations to triage patients seems like a simple way to enable midlevel providers to appropriately engage the expertise of a neurosurgeon. However, in most hospital settings, these images would be reviewed and interpreted by a neurosurgeon in addition to a radiologist, especially in the setting of a clearly identified intracranial hemorrhage. In some trauma centers, especially those with training programs, any abnormalities on a head CT after trauma mandates a neurologic (medical or surgical) evaluation – however, beyond the 'educational' value, it is unclear if such approaches are either good for patient care or reflect appropriate resource utilization. While the scoring criteria developed in this paper correlates with traumatic brain injury severe enough to require surgical intervention, it does not replace the need for expert evaluation by a neurosurgeon.

In the realm of traumatic brain injury, guidelines and scoring systems do already exist to triage patients that do or do not need head CT scans upon arrival to the hospital, such as the Canadian CT Head Rule guidelines of the UK National Institute for Health and Clinical Excellence (NICE) guidelines. These criteria are inconsistently followed. In younger patients under the age of 65 years, unindicated head CT scans tend to be ordered more often. In addition, physician specialties such as neurology or emergency medicine tend to order more head CTs that do not affect patient management when compared to trauma or neuro surgeons.[2] On the opposite end of the spectrum, some hospitals have found that the implementation of guidelines, such as the NICE guidelines, have lead to an increase in the number of head CT scans performed during night shifts or other shifts that tend to be preferentially staffed with junior Attendings or midlevel providers when compared to day shifts. Without official implementation of these guidelines, many patients were not getting indicated imaging studies. In addition, although the number of head CT scans increased, the overall hospital costs decreased as more patients were cleared for discharge home from the emergency room.[3] Even with these or similar guidelines in place, many patients with minor head trauma and normal Glasgow Coma Scale on presentation that do eventually require neurosurgical intervention may not initially receive a head CT scan. These patients tend to be older than 65 years, male, African American, or have a fall as their mechanism of injury.[4] These findings suggest that the criteria used to order a head CT for younger patients may not be appropriate for older patients with similar presentations, GCS, or physical exam findings. In this setting, the older patient may be better served with universal or near universal performance of CT scans of the head. With universal scanning, the scoring system proposed by the authors becomes more useful to quickly and consistently triage these patients for neurosurgical consult or critical care.

A scoring system for the severity of head CT scans in the trauma setting may also have medico-legal implications to the healthcare system. In legal cases where emergency providers failed to order a head CT in the case of trauma, approximately half of the cases were found to be due to provider neglect and half were settled outside of court. The median settlement amount was $1.5 million while the jury awarded amount was $2.8 million.[5] While these cases are in reference to patients who did not have CT scans ordered, similar questions in situations where an intervention was or was not offered based on a head CT finding could more consistently be defended if based on a radiologically based scoring system. These scoring systems do exist in other disciplines, such as the widely accepted BIRADS system used for mammography interpretation in breast cancer. Having clear criteria to order CT scans and then having clear criteria for triage of patients based on the interpretations of these scans would help provide documentable consistency in clinical decision-making. This consistency could potentially remove some of the liability and repercussions inherent in healthcare delivery. For the trauma patient, where decisions must often be made quickly and often without informed consent, an additional level of transparency may help to alleviate some of the emotional shock of both the trauma and the perceived extremes of operative management.

While the cranial computed tomography scoring tool proposed by the authors may correlate with the need for neurosurgical intervention and subsequent clinical outcomes, the utility of such a scoring system for guiding clinical decision making or triage of patients has yet to be seen. The authors do make an important distinction between the older trauma patient and the classic trauma patient by applying their scoring system to this subset of the trauma population. Additionally, the effects of traumatic brain injury can be devastating and costly, for patients and their families, society, and the healthcare system. In these regards, having a discrete scoring system to increase the transparency of the clinical decision making process for patients, families, and other healthcare providers can help strengthen the multidisciplinary team based approach to care of the trauma patient. However, while such scoring systems and algorithms are important, we must remember that they should never be a definitive substitute for sound clinical judgment based upon multi-disciplinary collaboration, experiences, and a well-rounded understanding of the pathophysiology. In other words, let us all not forget that, at the end of the day, medicine is still an art.