Intraoperative Subdural and Postoperative Brainstem Hemorrhage in Glioma Surgery: A Case Report

Introduction

Postoperative hemorrhage is a rare complication following craniotomy, with an incidence ranging from 0.8% [1] to 10.8% [2]. Approximately 20% of patients who undergo reoperation for hematoma evacuation typically do not survive [3]. However, remote intraoperative hemorrhage is exceedingly rare, with only a few cases reported [4], most of which followed posterior fossa craniotomy. The prognosis in these cases is generally poor. Here, we present a case of a 47-year-old woman who experienced continuous intraoperative subdural hematoma and postoperative brainstem hematoma following a supratentorial craniotomy for glioma.

Case Report

A 47-year-old female patient was admitted to our hospital for generalized seizure. Her main clinical manifestation was limb twitching and loss of consciousness, accompanied with mouth foaming and urinary incontinence.

The patient had a history of unilateral thyroidectomy in 2013 and salpingectomy in 2018. The patient also had a history of tuberculosis in 2003, which had been cured. Most importantly, the patient had a 3-year history of hypertension, with highest the systolic blood pressure at 160 mmHg. The patient took telmisartan 40 mg per day, which stabilized her blood pressure to approximately 130/90 mmHg. Physical examination showed no objective findings. Routine blood and urine investigation, electrocardiogram, and chest computed tomography (CT) were all normal. Cranial magnetic resonance imaging revealed a right frontal lobe lesion sized 3.8×5.0×3.0 cm, with inheterogenous enhancement (Figure 1A–1C).

On November 20, 2019, the patient underwent a right frontal lobe tumor resection via a pterional approach. Anesthesia was induced intravenously with midazolam (1 mg), propofol (2 mg/kg), sufentanil (0.8 ug/kg), and cisatracurium besilate (8 mg) to facilitate tracheal intubation. General anesthesia was maintained with continuous intraoperative pumping of remifentanil (0.1 μg/kg/min), propofol (2 mg/kg/h), and sevoflurane inhalation (1% in 50% air-oxygen mixture). No intraoperative complications related to anesthesia were noted. After the anesthesia took effect, the patient was positioned supine, and her head was stabilized using a 3-point Mayfield clamp. With the aid of neuronavigation, a curved incision was made at 9: 45 AM, and 250 mL of 20% mannitol was administered intravenously for intracranial pressure control. Two burr holes were drilled, and a bone flap measuring 6×7 cm was meticulously removed.

During the removal of the bone flap at approximately 10: 15 AM, the dura mater was found to be intact but unusually tense. Concurrently, there was a persistent elevation in the patient’s systolic blood pressure, which rose from 130 mmHg at 10: 00 AM to 180 mmHg at 10: 15 AM. To manage the blood pressure, the anesthesiologists administered 20 mg of furosemide; however, the blood pressure continued to rise. A total of 1500 mL of crystalloids was administered during anesthesia. The estimated blood loss was approximately 100 mL, and urinary output was 1000 mL.

In response to the situation, the scalp was immediately sutured. Upon examination, the patient’s pupils were noted to have a diameter of 2 mm on the left and 5 mm on the right, both unresponsive to light. The patient was then sent for a brain CT scan under general anesthesia, which was performed at 10: 30 AM. The scan revealed the emergence of a right frontotemporal subdural hematoma with mass effect (Figure 2A–2C).

The patient was returned to the operating room for the evacuation of the subdural hematoma. Approximately 60 mL of subdural hematoma was identified in the frontotemporal region, with no obvious vascular sources contributing to the hematoma. After the evacuation, brain tissue exhibited good fluctuation, and there were no significant signs of brain swelling. The dura mater was secured routinely along the margins of the bone window, and her vital signs remained stable. The scalp was then re-sutured following adjustments in positioning. At this time, her pupils measured 2 mm (left) and 4 mm (right).

After the positional adjustment was complete, the original incision was reopened, and the frontal lobe tumor was completely removed with the aid of fluorescein guidance in 60 minutes. A duraplasty was performed using an artificial meningeal patch. Postoperatively, her pupils were noted to be 2 mm (left) and 3 mm (right). A routine CT scan, conducted under general anesthesia, indicated that the right frontotemporal subdural hematoma had been successfully evacuated; however, a new hematoma was found in the brainstem at 4: 20 PM (Figure 3A–3C). Following this, the patient was transferred to the intensive care unit. The patient’s family had opted for conservative treatment instead of pursuing further evacuation of the hematoma. Postoperative pathological examination confirmed the diagnosis of oligodendroglioma, isocitrate dehydrogenase (IDH)–mutant, based on immunohistochemical analysis. Fluorescence in situ hybridization demonstrated 1p/19q chromosomal co-deletion, and the tumor was classified as World Health Organization grade II. Postoperatively, the patient remained in a coma with a Glasgow Coma Scale score of 3 (E1V1M1) and subsequently died of pneumonia in August 2023.

Discussion

Intraoperative subdural hemorrhage and postoperative brainstem hemorrhage are both exceedingly rare complications, and to the best of our knowledge, a case reporting both occurrences has yet to be documented in the literature. While the rate of intraoperative subdural hemorrhage is recognized, the incidence of postoperative hemorrhage varies between 0.8% and 10.8% [5], and nearly all of them were epidural hematoma. For patients requiring surgical evacuation, the prognosis is notably poor [3]. Reports indicate that postoperative hemorrhage contributes to a mortality rate of approximately 30% and is considered one of the most frequent causes of death following craniotomy [6]. Larger tumor sizes were found to correlate with an increased incidence of intracranial postoperative hemorrhage [3]. This association can partly be attributed to several factors: the extensive surgical field involved in larger tumors, the decrease in intracranial pressure that follows the removal of substantial tumors, and the potential brain injury caused by excessive retraction during surgery [7]. In the case of epidural hematomas, it has been suggested that postoperative intracranial hypotension may lead to a collapse of the brain and dura, which in turn could separate the dura from the skull and tear the extradural vessels [2]. According to Seiler et al, changes in intracranial dynamics during the sitting position can result in the tearing of subcortical veins, leading to intracerebral bleeding. In older adult patients after craniotomy with a sudden decrease of intracranial pressure, remote subdural hematoma might occur more often than remote epidural hematoma because of the tight adhesion between dura and skull bone. At that time, the stretch and rupture of the bridging vein of the cortex could easily lead to the remote subdural hematoma [8]. The severity of the hemorrhage can range from mild to severe, depending on the pressure gradient between the cerebral veins and the heart. This gradient can be influenced by patient positioning and the use of positive end-expiratory pressure by anesthesiologists. When patients are repositioned after surgery, this pressure gradient can decrease or become reversed due to actions such as coughing or vomiting, prompting a rerupture of previously torn veins [9].

In the present case, the patient was positioned supine, hypertension was effectively controlled, and the patient was relatively young, at 47 years of age. In addition, intraoperative findings showed no tight adhesion between the dura mater and the skull. Thus, the aforementioned theories do not directly apply to this situation. We propose that a rapid decrease in intracranial pressure following the removal of the bone flap may have contributed to the tearing of bridging veins, resulting in subdural hemorrhage. Additionally, the expedited excision of a large tumor in less than 60 minutes might have generated a substantial disparity between supratentorial and subtentorial pressure, ultimately leading to the tearing of veins in the brainstem, which then caused the occurrence of brainstem hemorrhage.

We hypothesize that placing a lateral ventricle drain prior to craniotomy and releasing cerebrospinal fluid slowly to gradually reduce intracranial pressure could aid in balancing the pressures between the supratentorial and subtentorial areas, which may be beneficial in minimizing relative movement between brain tissue and veins, thereby preventing intraoperative subdural hematoma. In cases in which intraoperative subdural hematoma does occur, it may be prudent to suspend tumor resection temporarily. This pause would allow additional time for cranial cavity pressure to rebalance, potentially enabling the elongated blood vessels to recover. This protocol may help to avert rapid intracranial pressure changes leading to postoperative hemorrhage in the brainstem, thereby reducing the risk of catastrophic hemorrhage.

However, to support these conjectures, further empirical research is necessary to investigate the impact of rapid intracranial pressure changes on the development of intracranial postoperative hemorrhage and to identify effective preventive measures for this complication.

Conclusions

This case underscores the rarity of simultaneous intraoperative subdural hematoma and postoperative brainstem hemorrhage in neurosurgery. The findings suggest a need for further investigation into preventive strategies, such as the use of lateral ventricle drains prior to craniotomy, to mitigate rapid intracranial pressure changes and associated hemorrhagic complications. Comprehensive research is warranted to explore effective interventions aimed at minimizing the risks of postoperative hemorrhage in high-risk surgical patients.