This study is the largest number of cases at present. The main finding is that the incidence of SAH after spinal surgery is 0.15% (23/14526) in patients with spinal surgery and 5.2% (23/441) in patients with CSF leakage. Another important finding is that hypertension, intraoperative CSF leakage and postoperative CSF drainage time are related to the occurrence of SAH after spinal surgery.
At present, the pathophysiological mechanism of postoperative SAH is not clear. Previous studies have reported that hypertension is a risk factor for SAH. Long-term hypertension leads to rupture of deep penetrating vessels, and common sites include basal ganglia, thalamus, brainstem and cerebellum2. Bozkurt suggested that elderly patients with hypertension and brain atrophy were more prone to SAH4. Canao believed that the risk of non-aneurysmal SAH in hypertensive patients was 2.6 times higher than that in control group5. Their report was also confirmed in our study, 73.9% of patients in the SAH group suffered from hypertension before surgery. Binary Logistic regression analysis suggested that cerebrovascular lesions caused by hypertension were the risk factors for SAH after spinal surgery.
Studies have speculated that the cause of SAH after spinal surgery is related to excessive CSF drainage6. A large amount of CSF loss will lead to brain dehydration, further stretching of the bridging vein, and finally rupture of the bridging vein and bleeding. Lower intracranial pressure due to CSF leakage can also cause intracranial hemorrhage, including SAH, subdural hemorrhage, parenchymal hemorrhage, and intraventricular hemorrhage. This phenomenon is most likely the result of a combination of factors, with a decrease in CSF volume causing the brain and spine to sag, thereby tensing the superior cerebellar and bridge veins. At the same time, the compensatory increase in venous volume leads to vascular filling and further expansion of venous structure, which is prone to venous avulsion or other damage7–8. Therefore, many scholars believe that intraoperative CSF leakage is associated with postoperative SAH. However, it is not fully understood which factors are involved in the specific mechanism of SAH after intraoperative CSF leakage. In our case study, from the perspective of the amount of CSF leakage, the mean intraoperative loss of CSF was 118.4 ± 56.9ml in the SAH group and 76.3 ± 23.0ml in control group. Statistical analysis found that intraoperative CSF leakage is a risk factor for postoperative SAH, which is statistically significant. It is suggested that intracranial hypotension caused by massive loss of CSF may induce venous avulsion resulting in SAH.
In addition to low intracranial pressure caused by massive intraoperative CSF leakage, complications such as acute posterior fossa syndrome, infection, SAH, headache after dural puncture, subdural hematoma and other complications have been reported after continuous CSF drainage in the subarachnoid space9–12.For patients with CSF leakage during spinal surgery, continuous large flow of postoperative CSF drainage may be associated with postoperative SAH13. Sasani reported a case of CBH that occurred not immediately after spinal arteriovenous malformation but after lumbar and abdominal shunt14.In the same year, Miglis described a patient whose nervous system was intact after cervical discectomy and fusion, but who developed acute headache, vomiting, and visual impairment 15 hours after insertion of a lumbar drainage tube15.These reports suggest that the placement of the drainage tube may be an important cause of intracranial hemorrhage after spinal surgery by CSF shunt. This trend is also reflected in our cases, in which all patients with CSF leakage were placed with a subfascial drainage tube prior to closing the incision. In our study, we compared the mean postoperative continuous drainage volume of CSF between the two groups, and found that the continuous drainage volume per hour was 15.4 ± 5.8 mL in the SAH group and 9.7 ± 2.1 mL in control group. Statistical analysis showed that continuous flow of CSF was associated with SAH. Continuous low intracranial pressure due to negative pressure drainage tube placement and continuous suction has also been suggested as a possible mechanism for SAH after spinal surgery.
The deterioration of neurological symptoms after surgery may be attributed to intracranial hypotension due to persistent CSF leakage. The most typical symptoms of intracranial hypotension are headache and cerebellar dysfunction. Headache occurred in 37% of patients with lumbar puncture, which usually occurred 24 to 48 hours after dural puncture, aggravated while standing and relieved in supine position16. The headache is most likely caused by movement of the caudal side of the brain due to loss of CSF, which causes tension in the pain-sensitive dural sinuses. This type of headache occurs when there is an excess loss of CSF (about 10mL/ h) and 80% resolves within 5 days. However, headache is also the most common symptom of postoperative SAH and can be easily confused with intracranial hypotension. The same lethargy, altered consciousness, and dysarrhythmia can easily be incorrectly attributed to opioid overdoses. Since head CT scan after spinal surgery is not a routine examination, SAH secondary to spinal surgery may be ignored17. Therefore, we recommend that patients with unexplained severe headache, drowsiness, and altered consciousness should undergo a CT scan immediately after surgery. Patients with mild to moderate headache or confusion should be closely monitored postoperatively, and neurological examinations should be performed more frequently to prevent the disease from worsening.
The treatment of SAH after spinal surgery depends on the site and amount of bleeding and the patient's clinical status. Patients with minor bleeding without significant compression and patients without neurologic status can be managed conservatively. Patients who have massive bleeding and rapid progression of the disease must be surgically removed with or without a ventriculostomy to manage hydrocephalus. Pulled out or intermittent clamping of the drainage tube may be a safer choice to control the drainage flow. Olson et al. found in their study that intermittent drainage compared with continuous drainage can reduce the frequency and severity of cerebral vasospasm in SAH18, but continuous and large amount of CSF leakage may need to be returned to the operating room for dural repair.