Keywords
raccoon - battle - BSF - fracture - clinical signs - base of skull
Introduction
A base of skull fracture (BSF) is a break in continuity of one or more of the temporal,
occipital, sphenoid, frontal, and ethmoid bones of the skull. Most BSFs, ∼70%, are
caused by high-velocity blunt trauma such as motor vehicle collisions, motorcycle
crashes, pedestrian injuries, falls, and assaults.[1] It is estimated that nearly 1.5 to 2 million people are injured and 1 million succumb
to death every year in India. Road traffic injuries are the leading cause (60%) of
traumatic brain injuries followed by falls (20–25%) and violence (10%).[2]
As the bones at the base of the skull are very fragile and located close to the dura
mater, meningeal lesions and liquor fistulas are frequent consequences of this type
of fracture. Other possible complications include cranial nerve palsies, cerebrovascular
lesions, and orbital and intracranial lesions.[3]
[4]
Previous studies indicate that BSF can be recognized based on clinical features which
vary depending on the degree of the associated brain and cranial nerve injury. Patients
may present with altered mental status, nausea, and vomiting.[5]
[6] They may cause bruising behind the ears (Battle's sign), bruising around the eyes
(raccoon eyes), or bleeding behind the ear drum (hemotympanum). Oculomotor deficits
due to injuries to cranial nerves III, IV, and VI may be present. Patients may also
present with facial droop due to compression or injury to cranial nerve VII. Hearing
loss or tinnitus suggests damage to cranial nerve VIII.[7]
Advanced trauma care manuals emphasize the presence of these BSF signs as parameters
for decisions about possible treatment strategies.[8] However, the signs do not always manifest early. In addition, the specificity of
these signs is questionable, as their presence does not always confirm BSF and the
absence of these clinical findings does not exclude this lesion. This discrepancy
and the great emphasis of clinical practice recommendations on the relationship among
the presence of clinical signs, diagnosis of BSF, and interventions in the initial
care of trauma patients gave rise to concerns that resulted in the current research
study.[5]
BSFs can be managed conservatively or may require surgery depending on the severity
of the injury. Patients with BSFs require admission for observation. Those with intracranial
hemorrhage require emergent neurosurgical evaluation. Otherwise, skull base fractures
are often managed expectantly.[1]
[9] Surgical management is necessary for cases complicated by intracranial bleeding
requiring decompression, vascular injury, significant cranial nerve injury, or persistent
cerebrospinal fluid (CSF) leak.
This study aims to assess the performance of the clinical signs of BSF for diagnosis
of this injury and to verify the correlation between the presence of these clinical
signs and the severity of head injury and patient outcome.
Materials and Methods
This was an observational cross-sectional study which included patients who presented
to our hospital with a BSF over a 3-year period between 2020 and 2023. Head-injured
patients were observed for clinical signs of BSF. The sample population included all
patients, aged 18 years or older, with blunt head injury who were treated at the hospital
within 24 hours after the head injury. Patients without radiological or surgical evidence
of BSF, who died or who were discharged in less than 24 hours posttrauma were excluded
from the study. The clinical signs of BSF considered in the study were raccoon's sign,
Battle's sign, otorrhea (CSF leak from ear), and rhinorrhea (CSF leak from nose),
as described in the Abbreviated Injury Scale (AIS) 2005, Update 2015. The AIS is used
to describe the injuries resulting from a trauma and to determine their severity.
BSF is diagnosed by direct visualization by a neurosurgeon during surgery or via computed
tomography of the skull (tomographic sections <5 mm) ([Fig. 1]).
Fig. 1 Mechanism of Battle's sign (A) and raccoon's sign (B). (Image courtesy: Wikipedia.)
The presence of fracture lines in the bones of the orbit, ethmoid, sphenoid, and base
of processes of the occipital bone and the temporal bone was considered positive results.[1]
[4]
[9] Head injury severity was estimated using the Glasgow coma scale (GCS) score, calculated
during the neurosurgeon's first evaluation (mild: GCS score = 13–15; moderate: GCS
score = 9–12; severe: GCS score = 3–8). The researchers and assistants were previously
trained to examine the patients and fill out the research data registration tool.
All patients included in the study were monitored from the admission to the emergency
department until the detection of clinical signs of BSF or discharge. We calculated
the prevalence of the clinical signs of BSF and further studied its distribution with
relation to the bone fractured, associated intracranial bleed, and GCS at presentation.
We obtained approval for the study from the institution's research ethics committee.
The participants or legal caregivers, if the patient was unable to express himself/herself,
signed the free and informed consent form.
Descriptive statistics were reported to compare patients with and without clinical
signs of BSF. Mean and standard deviations were reported for continuous variables
and frequencies for categorical variables. Comparison of means and proportions between
groups were done using two-tailed “t” test, chi-square test, or Fisher's exact test as appropriate. An α value <0.05 was
considered statistically significant. All analyses were performed using SPSS.
Results
A total of 292 patients admitted to the neurosurgical intensive care unit were included
in the study after obtaining informing consent. The mean age of the cohort was 36.27 ± 18.68
years; most patients were male, 233 (79.8%). Road traffic accidents (43.5%) were the
most common source of injury, followed by patients with a fall (40.7%). A total of
55 (18.8%) showed at least one of the four signs of BSF.
The characteristics of the study population, along with the presenting symptoms are
summarized in [Table 1]. There was no significant difference in age, sex, GCS, and symptoms at presentation
(all p > 0.05) between those with and those without clinical signs of BSF except patients
with bleeding through the ear or nose (p = 0.022) were more likely to present with at least one of the clinical signs. Mild
head injury was the most frequent injury (51.0%), followed by moderate head injury
(29.1%) according to the GCS, as summarized in [Table 1]. It did not vary between the two groups (p = 0.508).
Table 1
Demographic characteristics and presentation of the study population
|
Characteristics
|
Total, N = 292
|
With clinical signs, N = 55 (18.8%)
|
Without clinical signs
|
p-Value
|
|
Age (y), mean ± SD
|
36.27 ± 18.68
|
35.64 ± 19.03
|
36.42 ± 18.64
|
0.779
|
|
Sex, male, n (%)
|
233 (79.8%)
|
45 (81.8%)
|
188 (80.7%)
|
0.678
|
|
GCS, mean
|
12.37 ± 3.44
|
12.65 ± 3.47
|
12.31 ± 3.43
|
0.504
|
|
Presenting symptoms, n (%)
|
|
Seizures
|
13 (4.4%)
|
4 (7.3%)
|
9 (3.8%)
|
0.260
|
|
Vomiting
|
123 (42.1%)
|
20 (36.4%)
|
103 (43.5%)
|
0.337
|
|
Headache
|
33 (11.3%)
|
5 (15.2%)
|
28 (11.8%)
|
0.565
|
|
Loss of consciousness
|
211 (72.3%)
|
41 (74.5%)
|
170 (71.7%)
|
0.674
|
|
Altered sensorium
|
94 (32.2%)
|
25 (45.5%)
|
69 (29.2%)
|
0.021
|
|
Bleeding through ear and nose
|
156 (53.4%)
|
37 (67.3%)
|
119 (50.2%)
|
0.022
|
|
Severity of brain injury
|
|
Severe
|
43 (14.7%)
|
8 (14.5%)
|
35 (14.8%)
|
0.508
|
|
Moderate
|
87 (29.8%)
|
13 (23.6%)
|
74 (31.2%)
|
|
Minor
|
162 (55.5%)
|
34 (61.8%)
|
128 (54.0%)
|
Abbreviations: GCS, Glasgow coma scale; SD, standard deviation.
The prevalence of clinical signs under study, raccoon's sign, Battle's sign, otorrhea,
rhinorrhea, and cranial nerve palsies is summarized in [Table 2]. Fifty-five patients (18.8%) presented at least one of four investigated signs of
BSF, with maximum presenting with raccoon's sign (9.5%). Cranial nerve palsy was a
common presenting sign, with a prevalence of 15.8%, with the facial nerve being most
affected.
Table 2
Prevalence of clinical signs and cranial nerve palsy
|
Prevalence, N (%)
|
|
Clinical sign
|
|
|
Raccoon's sign
|
28 (9.5%)
|
|
Battle's sign
|
9 (3.1%)
|
|
Otorrhea
|
16 (5.5%)
|
|
Rhinorrhea
|
7 (2.4%)
|
|
Total
|
55 (18.8%)
|
|
Cranial nerve palsy
|
46 (15.8%)
|
|
VII
|
31 (67.4%)
|
|
III
|
12 (26.1%)
|
|
IX
|
1 (2.2%)
|
|
VI
|
2 (4.4%)
|
[Table 3] summarizes the skull base bone fractured in the 292 patients. Patients with frontal
(p = 0.021) or ethmoid (p = 0.049) fractures were significantly more likely to present with at least one of
the four clinical signs of BSF. Among the associated intracranial bleeds, contusion
was the most prevalent, followed by subarachnoid hemorrhage. None of the intracranial
bleeds was associated with the presence of clinical signs of BSF (all p > 0.05).
Table 3
Distribution of the skull base bone fractured and associated intracranial bleeds
|
Characteristic
|
Total, N = 292
|
With clinical signs, N = 55 (18.8%)
|
Without clinical signs
|
p-Value
|
|
Occipital fracture
|
50 (17.1%)
|
5 (9.1%)
|
40 (19.0%)
|
0.079
|
|
Temporal fracture
|
193 (66.1%)
|
37 (67.3%)
|
156 (65.8%)
|
0.838
|
|
Frontal fracture
|
62 (21.2%)
|
18 (32.7%)
|
44 (18.6%)
|
0.021
|
|
Sphenoid fracture
|
30 (10.3%)
|
8 (14.5%)
|
22 (9.3%)
|
0.247
|
|
Ethmoid fracture
|
6 (2.1%)
|
3 (5.5%)
|
3 (1.3%)
|
0.049
|
|
Subarachnoid hemorrhage
|
83 (28.4%)
|
15 (27.3%)
|
68 (28.7%)
|
0.833
|
|
Subdural hematoma
|
94 (32.2%)
|
19 (34.5%)
|
75 (31.6%)
|
0.678
|
|
Intracerebral hemorrhage
|
17 (5.8%)
|
3 (5.5%)
|
14 (5.9%)
|
0.897
|
|
Extradural hemorrhage
|
65 (22.3%)
|
16 (24.6%)
|
49 (20.7%)
|
0.176
|
|
Contusion
|
99 (33.9%)
|
21 (38.2%)
|
78 (32.9%)
|
0.457
|
The correlation between severity of brain injury, based on GCS at presentation and
the presence of clinical signs of BSF was studied. Severity of brain injury was classified
as mild (GCS > 12), moderate (GCS >8 but <13), and severe (GCS <8). There was no significant
correlation between the severity of brain injury and presence of clinical signs of
BSF with a p-value of 0.508. The results are summarized in [Table 1].
Patients with clinical signs of BSF were significantly more likely to develop a complication
in the hospital course (p = 0.024) than those without clinical signs. Cranial nerve palsy (p < 0.001) and CSF leak (p < 0.001) were particularly common complications. The results are summarized in [Table 4].
Table 4
Prevalence of complication developed by the patients during the course of hospital
stay
|
Complication
|
Total, N = 292
|
With clinical signs, N = 55 (18.8%)
|
Without clinical signs
|
p-Value
|
|
Any complication
|
120 (41.1%)
|
30 (54.5%)
|
90 (38.0%)
|
0.024
|
|
Intubation
|
57 (19.5%)
|
9 (16.4%)
|
48 (20.3%)
|
0.512
|
|
Ventilator
|
54 (18.5%)
|
8 (16.4%)
|
45 (19.0%)
|
0.652
|
|
Tracheostomy
|
30 (10.3%)
|
46 (83.6%)
|
216 (91.1%)
|
0.099
|
|
Meningitis
|
1 (0.3%)
|
1 (1.8%)
|
0 (0.0%)
|
0.038
|
|
Cranial nerve palsy
|
43 (14.7%)
|
16 (29.1%)
|
27 (11.4%)
|
<0.001
|
|
Hearing loss
|
6 (2.1%)
|
2 (3.6%)
|
4 (1.7%)
|
0.359
|
|
Cavernous sinus thrombosis
|
1 (0.3%)
|
0 (0.0)
|
1 (0.4%)
|
0.629
|
|
Vertigo
|
11 (3.8%)
|
1 (1.8%)
|
10 (4.2%)
|
0.399
|
|
Higher mental function deficit
|
12 (4.1%)
|
1 (0.3%)
|
11 (4.6%)
|
0.342
|
|
Motor deficits
|
39 (13.4%)
|
9 (16.4)
|
30 (12.7%)
|
0.467
|
|
Sensory deficits
|
3 (1.0%)
|
1 (1.8%)
|
2 (0.8%)
|
0.519
|
|
CSF leak
|
25 (8.6%)
|
22 (40%)
|
3 (1.3%)
|
<0.001
|
Abbreviation: CSF, cerebrospinal fluid.
The correlation between the presence of clinical signs of BSF and the patient outcome,
based on Glasgow outcome score (GOS) was studied ([Table 5]). The two were not found to have a correlation which was statistically significant
with a p-value of 0.926.
Table 5
Outcome of patients based on Glasgow outcome score
|
Glasgow outcome score
|
Total, N = 292
|
With clinical signs, N = 55 (18.8%)
|
Without clinical signs
|
p-Value
|
|
1
|
2 (0.7%)
|
0 (0.0%)
|
2 (0.8%)
|
0.926
|
|
2
|
8 (2.7%)
|
2 (3.6%)
|
6 (2.5%)
|
|
3
|
33 (11.3%)
|
7 (12.7%)
|
26 (11.0%)
|
|
4
|
45 (15.4%)
|
9 (16.4%)
|
36 (15.2%)
|
|
5
|
204 (69.9%)
|
37 (67.3%)
|
167 (70.5%)
|
Discussion
A total of 292 patients with BSF were included in this cross sectional study with
a preponderance of male patients (79.8%) in the fourth decade of life (36.3 ± 18.7
years). Clinical signs of BSF, including Battle's sign (5.5%), raccoon's sign (9.5%),
otorrhea (5.5%), and rhinorrhea (2.4%) were seen in 18.8% of the patients in our study,
which is very low when compared with the existing data.[3]
[4] Previous studies show that the clinical signs of BSF typically manifest within 48 hours
posttrauma, which is similar to our observations in the current study.[6]
[10] In those studies, the prevalence of Battle's and raccoon's sign is as high as 92.8%
patients with BSF which is in stark contrast to our findings,[3] though the prevalence of CSF leak (otorrhea and rhinorrhea), in our study, is similar
to the reported values in the previous studies.[11] This may be explained to some extent by the patient pool and geographic location
of the study. This study was conducted in the southern part of India, which is known
to have a relatively darker skin color as it lies close to the tropics, making it
hard to identify color changes (raccoon's sign and Battle's sign) over the skin. This
finding is further supported by the fact that the prevalence of CSF leak (otorrhea
and rhinorrhea) in our study is similar to the other studies in diverse regions of
the world.
In our study, temporal bone (66.1%) was the most commonly fractured bone and had the
maximum prevalence of clinical signs (67.3%). Patients with frontal (p = 0.021) or ethmoid (0.049) fractures were significantly more likely to present with
at least one of the four clinical signs of BSF. Those with ear or nose bleeding (p = 0.022) were more likely to present with at least one of the clinical signs.
Further, we observed cranial nerve palsies were seen in 15.8% of the cases with the
most common affected nerve being the facial nerve (67.4%), in contrast to the literature,
which states olfactory nerve as the most commonly affected.[7]
[12] This can possibly due to the high prevalence of temporal bone fracture (66.1%) in
our study and its close proximity to the facial nerve anatomically.
In the current study, 21% patients with mild head injury presented with clinical signs,
while they were observed in only 18.6% of patients with severe head injury. Analysis
did not reveal a significant (p = 0.508) correlation between the severity of head injury and the clinical signs.
This is in contrast to the existing data, which show a positive correlation between
the two.[3]
The patients who presented with clinical signs were more likely to have a complication
during the course of the hospital stay (p = 0.024) than those without clinical signs. Cranial nerve palsy (p < 0.001) and CSF leak (p < 0.001) were particularly common complications. Based on the GOS, the outcome of
the patient did not change based on the presence of clinical signs (p = 0.926).
Existing data also indicate that patients with clinical signs of BSF usually have
significant impact on the skull and therefore have a high probability of intracranial
injuries.[6] As seen in our study, 219 (75.0%) patients had associated intracranial injuries
with BSF, with contusions (33.9%) being the most prevalent. The presence of these
intracranial bleeds was not significantly associated with the presence of clinical
signs under study.
Although the clinical signs are not ideal clinical markers of BSF, their presence
does correlate with the base of skull bone fractured and complications that the patient
may develop, and therefore should be closely monitored. Our analysis of the indicators
showed that they are of little clinical relevance in the initial evaluation, as prevalence
of the BSF signs is very low in the South Indian population. Early diagnosis of BSF
is vital as initial care due to the risk of false passage when using the nasal route
to perform procedures at a time when the result of the computed tomography is not
yet available.[13]
[14] This study concluded that the limited visualization of BSF signs in the South Indian
population limit the value of BSF signs in the initial care of trauma patients. In
addition to this observation, 237 (81.2%) displayed no signs of BSF during the course
of their hospital stay. The absence of signs did not indicate a low possibility of
this lesion. Therefore, even without evidence of BSF, health care workers should pay
attention to head-injured patients and consider them as probable cases of BSF until
the outcome of computed tomography excludes the possibility of this lesion.
The present study has several limitations; first, the sample size is known to have
a significant influence on the results. Although our case load is significantly larger
than most available studies, larger studies, primarily in the Southeast Asia region,
are required to assess the dependability of the clinical signs of BSF. Second, different
settings contribute to a different case mix, different quality of care and policies
which can lead to bias and alter the results. Finally, we only observed the patients
for the clinical signs during their stay in the hospital. Although a vast majority
develop clinical signs within 48 hours postinjury, some may develop it later, which
we may have missed due to the lack of follow-up postdischarge.
Our study strongly supports that the presence of clinical signs increases the likelihood
of BSF and is a good marker for severity of head injury and intracranial injuries,
but in the South Indian population, they cannot be relied upon to clinically diagnose
BSF. Thus, these signs should be considered as independent markers of head injury
severity and not necessarily be used to diagnose BSF clinically, especially in the
South Indian population.
Conclusion
These study results indicate a limited diagnostic value of BSF clinical signs in the
South Indian population. Thus, other modalities such as radiological scans should
be considered for the diagnosis when suspected. Further, patients with ENT bleed following
head trauma are more likely to have clinical signs and such patients are eventually
prone to developing complications. These results also discourage the use of the nasal
route in all patients with suspected head injury and emphasize that during the nasal
aspiration procedure, the use of a rigid device is fundamental to avoid false passage
of the aspiration tube from the nasal to the intracranial region, which is possible
in patients presenting BSF in the South Indian population.
