Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Observation is recommended even following a normal CT brain in warfarinised head injuries

Three Part Question

In [adult patients on warfarin with a minor head injury] does a [normal CT brain scan] allow [safe discharge home, or should there be a period of inpatient observation]?

Clinical Scenario

An elderly woman attends your Emergency Department following a mechanical fall. She takes warfarin for atrial fibrillation and has a small occipital haematoma. Her Glasgow Coma Score (GCS) is 15; she has no amnesia and a normal neurological examination but did briefly lose consciousness. The INR (International Normalised Ratio) comes back within the therapeutic range at 2.9 and a computed tomography (CT) scan is requested according to the National Institute of Health and Clinical Excellence (NICE) guidelines.

The scan is reported as normal, and her social circumstances are adequate in that she lives with her husband who can keep an eye on her. You wonder, though, whether it is safe to discharge her or if there is a possibility of delayed intracranial haemorrhage due to her coagulopathy, and therefore she should be admitted for a period of neurological observation so that it can be identified and acted upon at the earliest opportunity.

Search Strategy

MEDLINE 1950 to April week 3 2010 using the OVID interface.
((exp Craniocerebral trauma/ OR head injur$.mp) AND (exp Warfarin/ OR OR exp Coumarins/ OR exp Anticoagulants/ OR anticoagula$.mp OR OR OR OR OR OR OR LIMIT to humans AND English language.

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Kaen et al,
137 Consecutive adult patients admitted within 48 h of minor head injury with normal CT scans and on warfarin. All patients had 24 h observation and a control CT scan before dischargeProspective cohort studyNeurological deterioration during observation periodNo patients developed any neurological deterioration. Small event rate. Larger number of patients required to establish definite conclusions
Any intracranial bleeding on CT scanTwo patients (1.4%) had bleeding on the control CT scan. Neither of these patients required neurosurgical intervention
Itshayek et al,
Selection of four patients who had presented following mild head injury, with a normal initial CT scan, on warfarin (three patients) or enoxaparin and aspirin (one patient) who all subsequently had neurological deterioration due to a delayed SAH Case seriesSelected due to the development of delayed SAHThree patients on warfarin developed a SAH within 24 h. The patient on aspirin and enoxaparin developed a SAH after 3/7. Two patients died and one patient had a GCS of 3 following surgery and the other patient had a GCS of 4 following conservative treatmentSmall case series, cannot extrapolate figures
Cohen et al,
77 Patients in total; 28 from level I trauma centre database and 49 patients who fitted the criteria from a selection of 4000 chart reviews undertaken by the American College of Surgeons review of trauma centres. Inclusion criteria consisted on minor closed head injury (GCS 1315) and concurrent warfarin therapyCohort studyNeurological outcomes in the selected patient group12 Patients were admitted with neurological symptoms hours or days after the injury. 20 patients were seen after head injury and discharged. Seven had a CT scan first, which was normal. Of the 20, two patients died at home and the other 18 were admitted with complications. The mortality for this group was 88.8%. 45 Patients were admitted for observation; 32 had CT scans, 28 were normal. The mortality in this group was 84% Limited information about patient demographics, outcomes and selection processes render this paper more of an extended case series rather than a cohort study
Garra et al,
65 Patients selected from retrospective chart review of electronic records from six community hospital ED including one trauma centre over a 2-year period. Inclusion criteria were patients on anticoagulants who had received a head injury with no LOC, amnesia or new neurological abnormality on examination Cohort studyClinically significant intracranial injury39 Patients had an initial CT scan, which was normal in all cases, and none of these patients had any further investigation or follow-up. 26 Patients had no CT scan but telephone follow-up and no complications were reported Retrospective data collection. No follow-up of patients following CT scan


There is much debate as to how best to manage this group of patients. In the UK, National Institute for Health and Clinical Excellence and SIGN guidelines advise performing a CT scan in such patients only in the presence of loss of consciousness or amnesia. There is no advice regarding the level of coagulopathy or about an observation period, especially in patients who do not meet the criteria for a CT scan. From the available literature, it is clear that there is considerable variation in the management of these patients. Italian guidelines, published in 1996, advise a CT scan in all patients with coagulopathy, observation for 24 h and then a second CT before discharge. Kaen et al suggest that this second CT may not be necessary, but otherwise, this may seem to be a prudent approach.

Evaluation and appropriate correction of the INR is also relevant. Delayed intracranial haemorrhage in this setting, with a normal admission CT and a therapeutic INR, would appear to be a rare occurrence, but does happen. In order to reduce the risk of this possibility, there should be a low threshold for CT scanning, the INR should be checked with consideration given to correction of a high INR, and there should be a period of observation for at least 24 h. The literature demonstrates that normal examination and CT scans do not preclude subsequent rapid deterioration. Delayed brain injury is significantly associated with increased mortality, slower recovery and a poorer outcome. Admission for observation should identify deterioration early, allowing rapid identification and management of problems. Accurate evaluation and treatment of patients who initially appear to be at low risk may be one of the most important factors in the reduction of mortality in head-injured patients

Editor Comment

ED, emergency department; GSC, Glasgow coma scale; LOC, loss of consciousness; SAH, subarachnoid haemorrhage.

Clinical Bottom Line

There is evidence to suggest a risk of delayed intracranial haemorrhage in patients who are receiving anticoagulant therapy who have had a mild head injury, even with a normal CT scan. This suggests the need for a period of observation in these patients. The level of risk is not quantifiable from the available literature.


  1. Kaen A, Jimenez-Roldan L, Arrese I, et al. The value of sequential computed tomography scanning in anticoaguated patients suffering from minor head injury. Journal of Trauma 2010; 68: 895-898.
  2. Itshayek E, Rosenthal G, Fraifeld S, et al.. Delayed posttraumatic acute subdural haematoma in elderly patients on anticoagulation. Neurosurgery 2006; 58: 851-856.
  3. Cohen DB, Rinker C, Wilberger JE, et al. Traumatic brain injury in anticoagulated patients. Journal of Trauma 2006; 60: 553-557.
  4. Garra G, Nashed AH, Capobianco L, et al. Minor head trauma in anticoagulated patients. Academic Emergency Medicine 1999; 6: 121-124.
  5. Study Group on Head Injury of the Italian Society for Neurosurgery. Guidelines for minor head injured patients' management in adult age. J Neurosurg Sci 1996;40:1115.