Minimally-invasive aspiration and drainage for management of traumatic epidural hematoma straddling transverse sinus
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.111111
Source of Support: None, Conflict of Interest: None
Aims: To investigate the therapeutic effect of minimally-invasive aspiration and drainage in traumatic epidural hematoma straddling transverse sinus (TEHSTS). Materials and Methods: Fifty-eight patients (39 males and 19 females) with TEHSTS and initial admission Glasgow Coma Scale (GCS) score of 8-10 (mean = 9) were treated with minimally-invasive aspiration and drainage under computed tomography (CT) guidance. Urokinase was used for irrigation and drainage. Post-operatively CT scan was performed at 3 h, 3 days, and 5 days. The volume of hematoma was calculated, and Glasgow outcome scale (GOS) was evaluated 3 months after the operation. Results: The volume of hematoma at 3 h and 3 days post-operation (20 ± 5 ml and 15 ± 2 ml; respectively) were significantly lower than that of pre-operation (45 ± 10 ml; P < 0.05). The hematoma was totally evacuated on 3-5 days post-operation. The GCS was 12 ± 1 on the 5 th day after the operation, which was significantly higher than that of pre-operation (8 ± 1; P < 0.05). Three months after operation, 45 (77%) patients had good recovery (GOS: 5) and 9 (15%) patients had moderate disability (GOS: 4). Conclusions: Minimally-invasive aspiration and drainage could be potentially effective in the treatment of TEHSTS with GCS score of equal or greater than 8 points.
Keywords: Glasgow coma score, minimally invasive aspiration and drainage, traumatic epidural hematoma straddling transverse sinus
Approximately 60% of the epidural hematomas (EDHs) are predominantly of arterial origin, in 9.7% of the EDHs is from the bleeding of venous vessels or diploic veins in the bone.  Yilmazlar et al.  reported that EDHs with the non-arterial origin account for 25% of the total number of traumatic EDHs. Traumatic EDHs of non-arterial origin are not always associated with skull fractures. The clinical manifestations, image appearance, and surgical management of venous EDHs are different. Our earlier study indicated that the EDHs located adjacent in transverse sigmoid sinus are of non-arterial origin in ~36.7% of cases.  Traumatic epidural hematoma straddling transverse sinus (TEHSTS) also known as mixed posterior cranial fossa EDH is uncommon. The clinical features are not specific, and diagnosis is difficult to make at the early stage. These hematomas progresses very rapidly with a poor prognosis.  Timely evacuation of hematoma through craniectomy could reduce the mortality and disability rates.  This study provides further information regarding clinical and radiological features as well as surgical strategies of 58 patients with TEHSTS.
Inclusion and exclusion criteria
Inclusion criteria included diagnosis of TEHSTS by computed tomography (CT) scan; hematoma volume (HV) of 35-55 ml; age between 10 years and 70 years; and informed consent for operation from the patients and/or their legal representative. Exclusion criteria included, patients with coagulation disorders; non-traumatic intracranial hemorrhage; intracranial or general infection; serious co-morbidities; previous stroke; intracranial hemorrhage secondary to intracranial aneurysm or arteriovenous malformation; post-traumatic Glasgow Coma Scale (GCS) of 3 points or less and bilateral mydriasis; and no consent for inclusion.
Subjects included 58 patients with TEHSTS who met the inclusion criteria admitted to Department of Neurosurgery, First Affiliated Hospital of Henan University of Science and Technology from September 2007 to November 2011. There were 39 males and 19 females with ages range between 13 and 68 (mean age 38.6). Retrospective review of the medical records and radiographs and the data collected included GCS score at admission and in hospital, and Glasgow Outcome Scale (GOS) score at 3 months post-operative. Documented causes of injury were traffic accident (26 cases, 44.8%), falling (23 cases, 39.6%) and assault (9 cases, 15.5%). Acute clinical deterioration (ACD) was defined as a sudden deterioration in the GCS (decrease of the best motor response score by ≥1 and/or decrease of GCS score ≥ 2) within a few hours of admission along with the enlargement of TEHSTS on repeat CT scan.  Operative findings and details were also recorded. GOS scoring: 1 = death; 2 = persistent vegetative state; 3 = severe disability; 4 = moderate disability and 5 = good recovery.
All the 58 patients showed spindle or crescent high density shade in the transverse sinus and posterior fossa on CT. HV was estimated by the formula (A × B × C)/2, where A was the largest diameter of the hematoma in centimeters, B was the diameter of hematoma perpendicular to A and C was the number of CT slices in which the hematoma was visible multiplied by the slice thickness in the centimeters , In admitted patients, the second follow-up CT scan was obtained at least within 2 h from the injury (including CT location scan), then at 3 h, 3 days and 5 days after the operation. Dynamic CT scan was done to measure the volume of hematoma and to evaluate GCS at each time point.
All patients received intravenous antibiotics on the day of surgery and the subsequent 2 days. The scalp was shaved and disinfected with Betadine TM and alcohol. All the patients underwent minimal-invasive surgery under CT guidance and local anesthesia in the emergency room. A puncture point was marked in the center of hematoma cavity, guided by images obtained from CT scan. In lateral prone position, local infiltration anesthesia on the puncture point (supratentorial) was carried out after the head fixation avoiding the transverse sinus and the fracture line. The skilled surgeon placed YL-1 (2.0-2.5 cm) minimally-invasive drainage needle (Beijing Wonderful Medical Apparatus Co., Ltd., China) with an electric drill [Figure 1] and [Figure 2] and extracted immediately as much hematoma as possible under mild negative pressure using a 10 ml syringe. Urokinase (50000-60000 IU) was dissolved in 3.5-5 ml saline and was slowly injected into the hematoma cavity under close observation of pupil size and consciousness by assistant-operator. The drainage was connected to the closed negative pressure hematoma drainage system. The drainage tube was reopened after 1-2 h of closure. Dark bloody fluid discarded immediately after the drainage system opening. Injections were performed 2 times a day and total volume of drainage was recorded. All procedures were completed within 15 min, and a CT scan was performed 3 h after the operation and on each successive day. After operation, dynamic head CT re-examination was carried out. The puncture needle and the drainage system were removed very carefully under aseptic conditions after 3-5 days post-operation.
After operation antibiotic treatment was given routinely to all patients. During the entire procedure attention was paid to two aspects: Dynamic CT re-examination that could find puncture needle position and changes in hematoma size and monitoring of the stabbed needle into the hematoma space not to cross fracture line and the transverse sinus; and following aseptic principles at every step of the operation. The drugs were injected evenly and slowly under close monitoring.
Descriptive statistics were used, and data were expressed as mean ± standard deviation. Statistical analyses were performed using a statistical software program (SPSS 11.0 for Windows, SPSS Inc., Chicago, IL, USA). Mean age and GCS on admission along with the median hematoma thickness and volume were compared among different time-points using the Student's t-test or the Mann-Whitney Test. P < 0.05 were considered to indicate statistical significance.
Of the 58 patients, 44 patients presented with headache, recurrent vomiting, dysphoria, altered mental status and 10 patients presented with frequent vomiting. Thirty-nine patients had occipital scalp hematoma. Admission GSC score ranged between 8 and 10 with of 9. The average time to establish the diagnosis after the trauma was 2-8 h, with an average of 5 h. Moreover, ten patients suffered from delayed hematoma, 24 h after injury and the longest time interval was 3 days. The average HV was 45 ± 10 ml. Forty-two patients suffered from unilateral or bilateral frontal and temporal regional contusion and laceration of brain and/or subdural hematoma. Fifty-five patients had occipital fracture. Three patients had dehiscence of lambdoidal or occipito-mastoid sutures. Ten patients suffered from delayed hematoma and were found following the evacuation of epidural or subdural hematoma in the frontal and/or temporal regions and decompressive craniectomy.
Of the 58 patients, in 54 patients CT scan performed 3 h after operation showed average HV of 20 ± 5 ml, and it was 15 ± 2 ml in the CT scan carried out 3 days after the operation., significantly lower than pre-operative HV (45 ± 10 ml; P < 0.05) [Figure 3]. In six patients, we succeeded in achieving minimal residual EDH (6 ± 2 ml) 3 days after the operation. There was no significant difference between the average GCS 3 h after evacuation (9 ± 1) and initial admission (8 ± 1) (P > 0.05). However, there was a significant difference between the mean GCS 5 days post-evacuation (12 ± 1) and admission GCS score (P < 0.05).
Of 42 patients with brain contusion and laceration and/or subdural hematoma, 32 patients were awake within 24 h after surgical intervention and were ambulatory on day-3. Eight patients suffered from delayed hematoma with a papillary change due to subdural hematoma/edema or intracerebral hematoma enlargement. In these patients, GCS score dropped to 6, the pupillary size returned to normal immediately following the release of the intracranial pressure with the evacuation of hematoma and decompressive craniectomy. However, the GCS score improved with the clearance of TEHSTS.
Two patients with admission GCS score of 8 died after the evacuation of hematoma and decompressive craniectomy secondary to cerebral hernia due to posterior cerebral artery or internal carotid artery infarction. ACD was observed in two patients who presented with pre-operative GCS score of 9 and 10 respectively progressing to vegetative state. One patient suffered from pulmonary embolism and was transferred to the department of respiration. After operation, 6 (10%) patients suffered from deep vein thrombosis of the lower extremity and 24 patients (42%) suffered from stress ulcer.
Urokinase (50000~60000 IU) was dissolved in 3 ml saline, with 2 ml saline flush pipe and was slowly injected into hematoma space twice a day. Total amount of urokinase was 300000-500000 IU. Drainage needle was inserted into hematoma space for 3-5 days (average of 4 days) until drainage fluid became brownish or yellowish and transparent [Figure 4] and [Figure 5]. Total blood loss during surgery ranged from 3 ml to 5 ml (average of 4 ml). The amount of bloody fluid removed through drainage needle ranged from 70 ml to 100 ml.
Three months after operation, good recovery (GOS 5) was achieved in 45 (78%) patients and moderate disability (GOS 4) in 9 (16%) patients.
This is the largest series of TEHSTS reported so far. The lack of specific clinical symptoms and signs is probably related to limited space within posterior fossa.  New appearance of hematomas occurs without any change in the clinical status.  Patients with bleeding from the venous sinus had a higher rate of complications and less often had a favorable outcome.  In this study, there were 2 (3.4%) deaths and 2 (3.4%) patients ended up in a vegetative state similar to the observations in other studies. Vomiting was the most common presenting feature in our study and also in other series.  Some of the clinical features observed in these patients may be related to the associated lesions. Although ACD did not seem relate to poor outcome,  two patients ended up in a vegetative state with ACD in our study.
It is assumed that elevated intracranial pressure associated with other intracranial traumatic lesions in these patient has a "protective" effect and is responsible for the delayed onset of EDH. , Measures such as decompressive surgery and use of osmotic therapy might release the tamponade effect and contribute to the occurrence of delayed EDH. In this study, the mean time for detecting hematoma after trauma was 2-8 h with an average of 5 h. Ten patients suffered from the delayed hematoma on follow-up CT scans, with the detection time of more than 24 h after the trauma. The longest time was 3 days after trauma. Although TEHSTS progressed rapidly, timely CT scan was able to confirm the lesion 4-6 h after trauma. Dynamic plain CT scan was particularly noteworthy in suspected TEHSTS, especially in patients who develop rapid altered mental status after supratentorial craniotomy. We suggest early and repeated CT/MRI scans for identifying venous sinus as the source of bleeding. Early detection and emergency surgery might improve the outcomes.
Asanin  suggested that EDH of posterior fossa are often associated with a skull fracture. TEHSTS result from different reasons: Bleeding from occipital fracture, blood vessels, and the transverse sinus laceration. A CT scan shows spindle or crescent hyperdense lesion within upper and lower part of the transverse sinus. Special attention should be paid in patients with an occipital trauma and scalp hematoma and (or) fracture of occipital condyle with TEHSTS.
There are no standard surgical indications for TEHSTS, mostly because of the location of hematoma. Surgical indications are different from that of supratentorial EDHs. Sullivan et al.  studied 37 patients with hematoma expansion in a series of 160 the patients with EDHs who were initially managed conservatively. Mean time to hematoma expansion was 8 h of injury and 5.3 h of the first CT, with no hematoma enlargement later than 36 h of injury. This data suggest that in patients managed conservatively, follow-up CT scan should be done at least within 6 h of injury. In this study, the second CT scan carried out within 2 h of trauma. The mean time to hematoma expansion was 2-8 h of injury and 5 h of the first CT, with no hematoma expansion later than 72 h of injury.
TEHSTS has distinctive features with respect to source of bleeding, operative approach and post-operative complications which result in poor outcome. An EDH greater than 30 ml should be surgically evacuated regardless of the GCS score.  Emergent surgical evacuation is strongly recommended in patients with an acute EDH and GCS score of <9 and anisocoria. GCS score and HV are the key determinants for treatment of EDH. In our patients, HV was 45 ± 10 ml and with an average of 27.5 ml. Malik et al. A close observation in patients with hematoma in the transverse sinus with a volume less than 20 ml and a dynamic CT scan in case of hematoma expansion has been suggested. In certain emergency situations, an alternative method for rapid and effective hematoma evacuation could be of considerable value to save life and to achieve good functional outcomes. In patients with rapidly enlarging temporal acute EDH of arterial origin which can compromise life,  placement of emergent burr hole over the temporal region can save the life. Noguchi et al.  reported usefulness of emergent needle aspirate under transcranial ultrasound guidance for a massive EDH secondary to vacuum-assisted delivery. Smets and Vanhauwaert  showed the efficacy of aspiration of cephalic hematoma to evacuate a communicating EDH in a newborn infant with food functional recovery. Park et al.  studied urokinase instillation using the closed suction drain in post-craniotomy EDH as feasible method with no complications and better outcomes. Drilling skull with instillation of urokinase into the hematoma had been reported in 22 patients with traumatic EDH. Liu et al. studied 11 of 13 patients with traumatic EDH who were successfully treated by the placement of a flexible tube through burr hole, followed by continuous suction under negative pressure. Only two patients required additional traditional craniotomy. They emphasized on the slow evolving nature of hematoma from non-arterial source as means of selecting the cases for this approach.  In this study, we adopted for the drainage of TEHSTS with a YL-1 needle and the negative pressure drainage system under local anesthesia and CT guidance. The needle was punctured into the center of hematoma cavity and urokinase was used for further irrigation and drainage. The tube was closed 1-2 h before being reopened, and the drainage was performed 2 times a day. The system was removed 3-5 days later with the largest HV of 55 ml. In our patients, no patient required blood transfusion, and none had surgery-related infections. With GCS as the main basis for treatment, this study employed minimally-invasive drainage in patients whose GCS score was more than 8. All the patients were subjected to hematoma irrigation and drainage under emergency CT guidance and local anesthesia with operation duration less than 15 min. Under mild negative pressure, hematoma was evacuated to the large extent. Urokinase was slowly injected, and the tube was reopened to perform continuous external drainage. Three hours after operation, hematoma was significantly less than pre-operative volume, GCS changed non-significantly in the early stage. From 3 days to 5 days after operation, hematoma was totally removed 5 days, after the operation and was associated with a significant increase in GCS score. Three months later, 45 (77%) patients recovered very well while 9 (15%) patients were left with moderate disability. Our experience suggests that 2-2.5 cm YL-1 needle, which is enough to penetrate the diploic layer, is a useful technique to drain these hematomas. If the needle penetrates too deep, it could damage the pia mater and cause subdural effusion and even fresh hemorrhage. It was necessary to install urokinase into the hematoma cavity in order to dissolve blood clots. The installation of urokinase was not more than 3 times a day as too much urokinase injection could increase the risk of intracranial infection and recurrent hemorrhage. The closure duration of the tube was based on clinical manifestation of the patients. If there was no evidence of an increasing intracranial pressure, the closure duration should reasonably extended, not more than 2 h. Such method could not totally remove the hematoma at one go and required drainage several times with urokinase injections. Decompressive craniectomy must be considered as soon as possible in patients with cerebral edema and whose GCS score is not more than 8 accompanied with progressive deterioration of consciousness.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]