Emergency computed tomography (CT) brain scans are frequently requested to evaluate elderly patients who present neurological deficits, acute cognitive disorders, or head trauma. Clinical signs of intracranial pathology are sometimes obvious but more often are lacking in this population 1. Moreover, altered sensorium and delirium may wrongly lead the physician to suspect a neurological cause when a drug or a non-neurological disorder is responsible 2,32,3.
In the case of head trauma 4–74–74–74–7, old age is often considered as a risk factor of intracranial hemorrhage 5,65,6.
The aims of our study were to assess the reasons for requesting an urgent CT brain scan, to record the diagnostic yield of cerebral imaging, and to seek out predictive factors of an intracranial pathology in elderly patients admitted to the emergency department (ED).
We conducted a retrospective study of elderly patients who were admitted to the ED and who underwent a CT brain scan. The setting was at the ED of the University Hospital of Nantes. The inclusion criteria were all patients aged 75 years or older who underwent a CT brain scan on admission. Patients were included over a 5-month period. Data were collected from ED charts and hospital discharge reports. The recorded criteria were demographic characteristics, results of CT scans, and reasons for emergency imaging. We then constituted and compared two groups of patients: group ICP, characterized by the presence of intracranial pathology [stroke, transient ischemic attack (TIA), tumors, seizure, and intracranial hemorrhage] and group NICP, characterized by the presence of nonintracranial pathology (drug adverse effects, infections, electrolyte disorders, etc.). Finally, we analyzed the subgroup of patients that presented with an intracranial hemorrhage in order to determine predictive factors of intracranial bleeding.
There was no ethical deliberation because of the retrospective nature of this study.
Data were collected using Excel software (Microsoft Systems, Redmont, Washington, USA) and were analyzed using GraphPad software (GraphPad software Inc., La Jolia, California, USA). Numerical data were expressed as averages, median, and SDs. Categorical data were expressed in percentages. To compare numerical data, the averages were compared using the Student t-test. To compare categorical data, the χ2-test was used. If conditions of application of this test were not met, the Fisher exact test was used. Finally, a logistic regression analysis was carried out to determine factors predicting the occurrence of an intracranial hemorrhage. A univariate logistic regression was first made, followed by a multivariate logistic regression with the backward method. Statistical analysis was carried out bilaterally or unilaterally according to the results of each test, and a P-value less than 0.05 was considered significant.
A total of 291 patients were included, of whom 70% were considered to have a new-onset intracranial pathology (112 ischemic strokes, 34 TIAs, 35 intracranial hemorrhages, five symptomatic intracranial tumors, one pneumocephalus, and 17 seizures unrelated to the previous conditions). Of all the CT scans, 79.7% were abnormal: 33% showed recent or semirecent pathological images explaining the symptoms, and 46.7% revealed ischemic sequelae, or nonspecific abnormalities such as cortical atrophy.
There was no significant statistical difference between groups ICP and NICP in terms of age (P=0.82), sex (P=0.23), place of living (P=0.35), major cardiovascular risk factors [hypertension (P=0.1), diabetes (P=0.4), dyslipidemia (P=1)], history of ischemic stroke or TIA (P=0.5), and consumption of antiplatelet (P=0.89) or anticoagulant (P=0.09) drugs. The only significant difference was in terms of psychotropic drug consumption (P=0.002), with a higher use in the NICP group.
The three main reasons for requesting an urgent CT brain scan were the presence of localizing signs (60%), delirium (21%), and disorders of consciousness with a Glasgow Coma Score of more than 14 (14.5%).
The presence of a localizing sign was more frequently observed in the ICP group (P=0.001), whereas the presence of delirium with no focal neurological deficits was more often found in the NICP group (P=0.03), as reported in Table 1.
In the intracranial hemorrhage subgroup (35 patients), the multivariate logistic regression with the backward method found four predictive factors for an intracranial bleed. They are reported in Table 2. Taking antivitamin K treatment (even if international normalized ratio>3) or antiplatelet drugs, a history of falling without head trauma, or previous intracranial hemorrhage, did not appear to be significantly associated with the risk of discovering an intracranial bleed.
The main limitations of this study are related to its retrospective nature. Indeed, some data could be missing from the ED charts and the hospital discharge reports.
Furthermore, the multivariate logistic regression was made from the intracranial hemorrhage subgroup, which included only 35 patients.
Comparing the two groups ICP and NICP, we found no differences in terms of demographic characteristics or previous pathology, except for a higher use of psychotropic drugs in the NICP group. Our data suggest that in our elderly population there is no typical patient profile when concerned with the risk of having an intracranial pathology. Is the main predictive factor of an intracranial pathology simply old age?
With regard to the reasons for ordering a CT scan, the most frequent was the presence of a focal neurological sign, which was also an excellent predictive factor of an intracranial pathology.
In contrast, delirium without focal neurological signs was more often encountered in the NICP group, and CT scan imaging showed very few specific abnormalities for this indication. Delirium with no focal neurological deficits is a common reason for requesting a CT brain scan, but it has a very low diagnostic yield 2,3,8,92,3,8,92,3,8,92,3,8,9. The French guidelines regarding delirium in elderly patients recommend CT scan imaging only after a full clinical workup, which includes routine biology, electrocardiogram, chest radiograph, and urine dipstick 10. However, cerebral imaging is required immediately if the patient presents with focal neurological signs, head trauma, or sudden-onset headache 10. In our study, logistic regression analysis of the intracranial hemorrhage subgroup (35 patients) found the following predictive factors of hemorrhage: focal neurological signs, disorders of consciousness with a Glasgow Coma Score of less than 14, sudden-onset headache or headache associated with at least two episodes of vomiting, and head trauma.
The above criteria are very similar to the French guidelines regarding the ordering of an urgent CT brain scan in patients with delirium. Their presence increases the diagnostic yield of brain imaging. However, absence of the above predictive factors makes the diagnosis of intracranial hemorrhage unlikely and should lead the physician to preferentially search for an alternative diagnosis, such as adverse drug reaction, infection, urinary retention, fecaloma, or electrolyte disorders.
The presence of a focal neurological sign was the most frequent reason for ordering a CT scan and was also an excellent predictive factor of intracranial pathology.
Delirium with no focal neurological deficits was the second most frequent reason for ordering a CT scan, which had a poor diagnostic yield for this indication.
In our elderly population, we found no typical patient profile when concerned with the risk of having an intracranial pathology.
Altered sensorium and delirium without other clinical findings may wrongly lead the physician to suspect a neurological cause, such as intracranial bleeding, when a drug or a non-neurological disorder is responsible. In these cases, our results suggest that an urgent CT brain scan should be considered only if there is at least one of the following items: localizing signs, head trauma, consciousness disorders, sudden-onset headache, or headache associated with at least two episodes of vomiting. If these criteria are missing, a CT scan should be ordered if a complete workup does not find any explanation for the symptoms.
Further prospective research is needed to determine the place of urgent brain imaging, especially in elderly patients with delirium. It would be particularly interesting to develop a new imaging algorithm to help clinicians.
Conflicts of interest
There are no conflicts of interest.
1. Brown G, Warren M, William JE, Adam EJ, Coles JA. Cranial computed tomography of elderly patients
: an evaluation of its use in acute neurological presentations. Age Ageing. 1993;22:240–243
2. Hardy JE, Brennan N. Computerized tomography of the brain for elderly patients
presenting to the emergency department with acute confusion. Emerg Med Australas. 2008;20:420–424
3. Hufschmidt A, Shabarin V. Diagnostic yield of cerebral imaging in patients with acute confusion. Acta Neurol Scand. 2008;8:245–250
4. Gangavati AS, Kiely DK, Kulchycki LK, Wolfe RE, Mottley JL, Kelly SP, et al. Prevalence and characteristics of traumatic intracranial haemorrhage in elderly fallers presenting to the emergency department without focal findings. J Am Geriatr Soc. 2009;57:1470–1474
5. Stiell IG, Wells GA, Vandemheen K, Clement C, Lesiuk H, Laupacis A, et al. The canadian CT head rule for patients with minor head injury. Lancet. 2001;357:1391–1396
6. Haydel MJ, Preston CA, Mills TJ, Luber S, Blaudeau E, DeBlieux PM. Indications for computed tomography in patients with minor head injury. N Engl J Med. 2000;343:100–105
7. Stiell IG, Clement CM, Rowe BH, Schull MJ, Brison R, Cass D, et al. Comparison of the Canadian CT Head Rule and the New Orleans Criteria in patients with minor head injury. JAMA. 2005;294:1511–1518
8. Bilbault P. Non-trauma CT head scans in the emergency department: are we doing too much? Emerg Med J. 2009;26:151
9. Hirano LA, Bogardus ST Jr, Saluja S, Leo-Summers L, Inouye SK. Clinical yield of computed tomography brain scans in older general medical patients. J Am Geriatr Soc. 2006;54:587–592
10. French Health Authority, recommendations for good practice. Acute delirium
in elderly patients
: management of agitation. May 2009