Fulminant Cervical Epidural Hematomas: Why Do They Happen, How Can We Minimize Their Occurrence, and What Can We Do When They Do Occur? A Perspective

Background: Epidural hematomas after appropriately performed cervicothoracic interlaminar epidural injections have been associated with the rapid onset of neurological symptoms and devastating outcomes, despite prompt identification and treatment. Anticoagulation issues were initially felt to be the problem, but the occurrence of fulminant hematomas in patients without coagulation forced a reassessment of the causes and responses to this problem. Objectives: To evaluate why fulminant epidural hematomas occur after cervicothoracic epidural injections, with a literature review to survey knowledge about them in the surgical literature, and to offer comments as to what the interventional pain physician can do to minimize their occurrence. Study Design: A perspective piece with a literature review. Settings: Interventional pain management practices. Methods: A perspective on the issue of fulminant cervical hematomas and an associated literature review. Results: Anatomical studies show that there are no meaningful arteries in the posterior epidural spaces which would explain hematomas. There is a dense posterior intravertebral epidural venous plexus at C1 and also at C6-C7 extending caudally to the upper thoracic region. A venous origin has been questioned because venous pressure was felt to be too low to explain the bleeding. The surgical literature, going back 80 years, contains numerous reports of engorged epidural veins causing radiculopathy and myelopathy. These engorged veins can occur in the presence or absence of spinal pathology. There is no known means of reliably identifying these engorged veins; they have been mistaken for disc protrusions. At least one report documents massive bleeding from these veins. Studies done on a feline model of cervical stenosis suggest that the epidural pressure can reach arterial levels. Limitations: No direct documentation of arterialized posterior intravertebral epidural venous pressures has been made. While anatomical anomalies and degeneration contribute to epidural scarring, we do not have a full understanding as to the cause of arterialization of veins, particularly in younger patients with no obvious intraspinal pathology. Conclusion: Fulminant cervicothoracic epidural hematomas after an epidural injection appear to arise from the unintentional and unavoidable puncture of arterialized veins with sharp needles. A technique to open a path out from the foramen so that the blood can escape is described. Alternatively, providers should consider injecting more cephalad, between C2-C3 and C6-C7 in the cervical spine, or an alternative procedure, such as a selective nerve root injection. A cervical transforaminal approach should only be attempted with a blunt needle, which cannot enter an artery. Should symptoms occur, cervical flexion rotation maneuvers should be implemented while awaiting prompt transfer to a facility where an appropriate diagnosis and treatment can be provided.