There were 801 patients who had an S.06-diagnosis code between the years of 2012–2022. Patients under the age of 18 years at admission and patients who had no contusions were excluded, generating 272 patients who met the inclusion criteria (Fig. 1). Of these, 35 patients underwent contusion evacuation surgery, 9 patients had a craniectomy, and three of those had both procedures done. There were 31 patients who had an extra-axial hematoma evacuated, and 202 patients were conservatively treated (Table 1). The most common occurring outcome measure was GOSE 7 and 8, together representing 35% of the patient population where outcome was accessible. GOSE was not accessible due to lack of follow up amongst 21% (n = 58) patients (Table 1).
Table 1: Demographics of the study patients.
GCS = Glasgow Coma Scale. SBP = Systolic blood pressure. PK-INR = Prothrombin complex-International ratio. APTT = Activated partial thromboplastin time. NICU = Neurointensive care. LMWH = Low-molecular weight heparin. ASA = American Society of Anesthesiology. IQ = Interquartile range.
*= Could be evacuation of extra-axial hematoma, or a combination of contusionectomy, evacuation of extra-axial hematoma
and decompressive craniectomy.
Parameter (n = 272) | | Median (IQ) / (n. %) | Missing (n. %) |
Age | | 55 (35–68) | |
Sex | Female Male | 87 (32%) 185 (68%) | |
GCS at arrival | | 13 (7–14) | |
Initial treatment | Conservative Contusionectomy Craniectomy Evacuation extra-axial hematoma | 202 (74%) 35 (13%) 9 (3%) 31 (11%) | |
Conservative or Operative treatment | Continous conservative Contusionectomy Craniectomy Other* | 172 (85%) 21 (10%) 3 (1%) 6 (3%) | |
Cause of injury | Fall from same level Fall from height (> 1m) Road traffic accident Sports related Assault Blunt or penetrating trauma | 89 (33%) 58 (21%) 83 (31%) 5 (2%) 15 (6%) 16 (6%) | |
ASA score | | 1 (1–2) | |
SBP admission | | 140 (122–160) | 17 (6%) |
Platelet count > 150x109/l | | 232 (199–288) | |
Platelet count < 150x109/l | | 137 (110–142) | |
Missing platelet information | | | 6 (2%) |
PK-INR | | 1 (1-1.1) | 11 (4%) |
APTT | | 25 (23–27) | 15 (6%) |
Days spent in NICU | | 5 (2–13) | 3 (1%) |
30 days mortality | | 29 (11%) | 4 (1%) |
Thromboembolic event | | 10 (4%) | |
Received LMWH | | 59 (22%) | |
GOSE at 6 months | Dead or vegetative Severely disabled Moderately disabled Good Recovery | 41 (19%) 37 (17%) 62 (29%) 74 (35%) | 58 (21%) |
Contusion expansion
Out of 272, 181 patients experienced contusion expansion (67%). The median contusion volume at admission was 6.8 ml (IQ 2.1–15.9) (Table 2). The median absolute contusion volume increase at follow-up CT scan was 4.0 ml (IQ 0–15) (Table 2). Admission contusion volume had a significant, positive correlation with absolute contusion volume increase (Spearman's rho coefficient 0.19, p = 0.002). The median Marshall score was 2 (IQ 2–3) and the median Rotterdam score was 3 (IQ 2–3). Both Marshall- and Rotterdam score had a significant, positive correlation with absolute contusion volume expansion (Spearman's rho coefficient 0.26, p < 0.001 and 0.24, p < 0.001, respectively)
Table 2
Computer tomography (CT) characteristics of included patients.
Parameter (n = 272) | | Median (IQ) / (n. %) | Missing (n. %) |
Age | | 55 (35–68) | |
Sex | Female Male | 87 (32%) 185 (68%) | |
GCS at arrival | | 13 (7–14) | |
Initial treatment | Conservative Contusionectomy Craniectomy Evacuation extra-axial hematoma | 202 (74%) 35 (13%) 9 (3%) 31 (11%) | |
Conservative or Operative treatment | Continous conservative Contusionectomy Craniectomy Other* | 172 (85%) 21 (10%) 3 (1%) 6 (3%) | |
Cause of injury | Fall from same level Fall from height (> 1m) Road traffic accident Sports related Assault Blunt or penetrating trauma | 89 (33%) 58 (21%) 83 (31%) 5 (2%) 15 (6%) 16 (6%) | |
ASA score | | 1 (1–2) | |
SBP admission | | 140 (122–160) | 17 (6%) |
Platelet count > 150x109/l | | 232 (199–288) | |
Platelet count < 150x109/l | | 137 (110–142) | |
Missing platelet information | | | 6 (2%) |
PK-INR | | 1 (1-1.1) | 11 (4%) |
APTT | | 25 (23–27) | 15 (6%) |
Days spent in NICU | | 5 (2–13) | 3 (1%) |
30 days mortality | | 29 (11%) | 4 (1%) |
Thromboembolic event | | 10 (4%) | |
Received LMWH | | 59 (22%) | |
GOSE at 6 months | Dead or vegetative Severely disabled Moderately disabled Good Recovery | 41 (19%) 37 (17%) 62 (29%) 74 (35%) | 58 (21%) |
IQR = Interquartile range. CT = Computed tomography. MRI = Magnetic resonance imaging. |
DAI = Diffuse agonal injury. |
Table 2: Computer tomography (CT) characteristics of included patients.
IQR = Interquartile range. CT = Computed tomography. MRI = Magnetic resonance imaging.
DAI = Diffuse agonal injury.
Parameter (n = 272) | | Median (IQR) / (n. %) |
Contusion volume at admission (largest contusion) | | 6,8 ml (2.1–15.9) |
Absolute volume expansion at follow up CT scan (within 72 hours) | | 4 ml (4–15) |
Bifrontal contusions | | 85 (31%) |
Location of largest contusion | Frontal Temporal Parietal Occipital | 150 (55%) 112 (41%) 5 (1.5%) 5 (1.5%) |
Location of second largest contusion | None Frontal Temporal Parietal Occipital Cerebellum | 133 (48%) 86 (32%) 46 (17%) 5 (18%) 1 (0.3%) 1 (0.3%) |
Epidural hematoma | | 41 (15%) |
Subdural hematoma | | 94 (35%) |
Basal cistern | Open Compressed Effaced | 181 (67%) 85 (31%) 6 (2%) |
Midline shift | Normal < 5 mm shift > 5 mm shift | 156 (57%) 89 (33%) 27 (10%) |
Marshall Score | | 2 (2–3) |
Rotterdam Score | | 3 (2–3) |
MRI verified DAI | | 30 (11%) |
There was a weak but significant correlation between absolute contusion volume increase and GCS on admission (Spearman's rho coefficient 0.14 p = 0.02, and Spearman's rho coefficient − 0.14 p = 0.018, respectively). A significant difference in CE was found between the group of "Good recovery" (1.4 ml, 0-8.2) and "Dead or vegetative" (14.3 ml, IQ 2-44.1), and "moderately disabled" (8.3 ml, IQ 2–18) (p < 0.001 and p = 0.001, respectively). The presence of other intracranial hematomas did not significantly correlate to contusion expansion, nor did midline shift at admission, systolic blood pressure, known alcohol abuse, blood ethanol levels, age, or sex.
Bifrontal contusions
The median contusion volume amongst patients with contusions at other, non-bifrontal brain regions was 4.9 ml (n = 187, IQ 1.7–13.0), and 11.6 ml amongst patients with bifrontal contusions, measuring the largest (n = 85, IQ 4.2–20,7; p < 0.001). The absolute contusion volume increase amongst patients without bifrontal contusions was 2,1 ml (n = 187, IQ 0-13.2) compared to 10 ml (n = 83, IQ 2-30.6) in the bifrontal group (p < 0.001). Patients with bifrontal contusions had similar GCS scores upon arrival when compared to patients with contusions in other brain regions, including patients with unilateral frontal contusions.
Patients with bifrontal contusions did not undergo contusion evacuation surgery or craniectomy at a greater extent than other contusion patients. The median GOSE in the bifrontal group was equivalent of moderately disabled, which is equivalent of GOSE 5 and 6 (n = 65) which was also the case for the group with contusions in other locations (n = 149). Using Chi-squared test, the bifrontal group had a worse outcome than the group with other-located contusions, χ2 (3, N = 214) = 8.115, p = 0.0043 (Fig. 2, Table 3).
Table 3
Worse outcome in patients with bifrontal contusions when compared to patients with contusions in other locations
Parameter (n = 272) | | Median (IQR) / (n. %) |
Contusion volume at admission (largest contusion) | | 6,8 ml (2.1–15.9) |
Absolute volume expansion at follow up CT scan (within 72 hours) | | 4 ml (4–15) |
Bifrontal contusions | | 85 (31%) |
Location of largest contusion | Frontal Temporal Parietal Occipital | 150 (55%) 112 (41%) 5 (1.5%) 5 (1.5%) |
Location of second largest contusion | None Frontal Temporal Parietal Occipital Cerebellum | 133 (48%) 86 (32%) 46 (17%) 5 (18%) 1 (0.3%) 1 (0.3%) |
Epidural hematoma | | 41 (15%) |
Subdural hematoma | | 94 (35%) |
Basal cistern | Open Compressed Effaced | 181 (67%) 85 (31%) 6 (2%) |
Midline shift | Normal < 5 mm shift > 5 mm shift | 156 (57%) 89 (33%) 27 (10%) |
Marshall Score | | 2 (2–3) |
Rotterdam Score | | 3 (2–3) |
MRI verified DAI | | 30 (11%) |
The difference in outcome between patients with bifrontal contusions and patients with |
contusions in other locations are compared using Chi-square test. A total of 214 patients were included in the study of whom 65 had bifrontal contusions and 149 had contusions in other locations. The difference between the groups was highly significant, p = 0.0044. |
2: Distribution of outcome between patients with bifrontal contusions and
patients with contusions in other locations.
Figure 2 legend: Distribution of outcome between patients with bifrontal contusions and patients with contusions in other locations.
Distribution between patients with bifrontal contusions (n = 65) compared to patients with contusions in other locations (n = 149) with regard to patient outcome at six months post-injury using the Glasgow outcome scale extended (GOSE).
Table 3: Worse outcome in patients with bifrontal contusions when compared to patients with contusions in other locations
The difference in outcome between patients with bifrontal contusions and patients with
contusions in other locations are compared using Chi-square test. A total of 214 patients were included in the study of whom 65 had bifrontal contusions and 149 had contusions in other locations. The difference between the groups was highly significant, p = 0.0044.
Percentage of total within GOSE category (n) |
| Dead or Vegetative | Severely disabled | Moderately disabled | Good recovery |
Contusions, bifrontal (n = 65) | 46% (19) | 19% (7) | 32% (20) | 26% (19) |
Contusions, other locations (n = 149) | 54% (22) | 81% (30) | 68% (42) | 74% (55) |
χ2 = 8.12, p = 0.044 |
Coagulation
The median blood platelet count was 223 x109/L (IQ 183–276) (Table 1). Neither INR nor APTT was significantly associated with outcome, and only 25 patients had an INR above 1.2. median INR was 1 (IQ 1–1,1) (Table 1). For 237 patients, there was no records of any regular intake of anticoagulant drugs. There were 11 patients on Warfarin, and 25 patients were on aspirin. There was no significant difference in outcome or absolute contusion volume increase in the aspirin-treated group compared to patients not treated with aspirin. The number of patients on any other type of antithrombotic drug was too low for further analysis. Out of all patients, 11% (n = 30) received one or more blood platelet transfusions. Using Chi-square test, the patients who received blood platelet transfusion had a higher-than-expected count in the GOSE group of "dead or vegetative", and none in the group of "good recovery", in comparison to the group who did not receive a blood platelet transfusion where the trend was inverted, χ2 (3, n = 214) = 19.78, p < 0.001).
When dichotomized to < 150- and normal (> 150x109/L) (n = 30 and n = 179, respectively), using Chi- square test, the group with a lower platelet count had a significant worse six-month outcome according to GOSE, compared to the group with a higher blood platelet count (χ2 (3, n = 209) = 13.4, p < 0.004) (Fig. 3, Table 4). A difference in absolute contusion volume increase was shown between the patients with a platelet cell count < 150x109/L, and the > 150x109/L group (15.3 ml versus 3.6 ml, p = 0.003).
3: Difference in outcome between patients with a platelet count < 150x109/L, and
patients with a platelet count > 150x109/L.
Figure 3 legend: Difference in outcome between patients with a platelet count < or > than 150x10 9 /L
Distribution in patient outcome according to Glasgow outcome scale extended
(GOSE) dichotomizing according to platelet count less than (n = 30) or above (n = 179) 150x109/L.
Table 4: The difference in outcome between patients with a low versus a
normal platelet count, using Chi-square test.
The difference in outcome between patients with a low platelet count versus a higher platelet
count, using Chi-square test. A total of 209 patients were included in the study, 30 patients had a platelet count < 150x109/L, and 179 had a platelet count > 150x109/L. The difference between the groups was shown significant, p = 0.004.
Percentage of total within GOSE category (n) |
| Dead or Vegetative | Severely disabled | Moderately disabled | Good recovery |
Platelet count < 150x10^9/L (n = 30) | 29% (12) | 14% (5) | 16% (10) | 4% (3) |
Platelet count > 150x10^9/L (n = 179) | 71% (29) | 86% (32) | 84% (51) | 96% (67) |
χ2 = 13.4, p = 0.004 |
Multivariate analysis
A multivariate regression was performed with six-month outcome as outcome variable. This included contusion volume at admission, absolute contusion volume difference, platelet levels < and > 150x109/L, and whether the contusions were bifrontal or not. In the multivariate analysis, 207 patients were included (76.1%) of whom 61 patients had bifrontal contusions, and 29 patients had a platelet count < 150x109/L. The outcome variable was dichotomized into unfavorable (n = 77) and favorable (130), as described above. The remaining variable that had a significant impact on outcome was contusion expansion (OR 0.975, CI 0.955–0.995) (Table 5).
Table 5
Multivariate analysis with regards to patient outcome at six months post-injury. A multivariate regression analysis was performed, including variables who on univariate regression had a significant impact (p < 0.05) on patient outcome according to Glasgow outcome scale extended (GOSE) six months post-injury. Contusion expansion was shown significantly impact outcome measured as either unfavourable (GOSE = 1–4; n = 77) or favourable (GOSE = 5–8; n = 130) outcome. Due to missing data on platelet levels upon arrival, 207 out of 272 patients could be included. OR = Odds ratio; CI = Confidence interval.
Included variables | n = 207 | OR (CI) | p |
Contusion volume at admission | | 0.983 (0.962–1.004) | 0.117 |
Contusion expansion | | 0.975 (0.955–0.995) | 0.013 |
Contusion location (bifrontal & other) | (61 & 146) | 0.778 (0.386–1.567) | 0.482 |
Platelet levels (</>150x109/L) | (29 & 178) | 0.490 (0.208–1.155) | 0.103 |