The spinal dura mater extends from the foramen magnum to the second sacral spinal segment. It contains CSF and encases the spinal nerves after they leave the spinal cord. The pia mater and arachnoid fuse with the connective tissue of the spinal nerves, making up the lateral borders of the dural sac (see Fig. 27-2). The dura is a dense, connective tissue layer created by longitudinal lamella of collagen and elastin fibers. It was believed that these fibers ran in a longitudinal direction and this thought was initially supported by microscopic studies (13). However, more recent light and electron microscopic studies describe the dura mater as consisting of collagen fibers arranged in several layers parallel to the surface of the medulla spinalis. The direction of the fibers in each sub-lamina do not demonstrate any specific orientation (14). The outer layer may be arranged in a longitudinal direction but this is not necessarily repeated through the more interior dural layers.

Over 100 years ago, Bier attributed persistent leakage of CSF through the dural puncture as the cause of the headache, but today the exact mechanism is not still entirely clear. One theory is that when the patient’s position changes
from supine to an upright position, there is a downward traction on pain-sensitive nerves, intracranial veins, and meninges, which causes the pain. This sagging of intracranial structures has been demonstrated on magnetic resonance imaging (MRI) (15). If the leakage of CSF exceeds the rate of production, the cushioning effect on the brain is lost (16). The average production rate of CSF is 0.3 mL/kg/h, 0.3 to 0.4 mL/min, or about 500 mL/day for a 70 kg person. The total volume of CSF is about 150 mL at a given time, distributed half in the intracranial space and the other half cushioning the spinal cord. The choroid plexus secretes the majority of the CSF, and the rest is secreted by the brain (interstitial space of the brain, the ependymal lining of the ventricles, and the dura of the nerve root sleeves), as a byproduct of oxidative metabolism (17). Reabsorption of CSF is primarily by the arachnoid granulations in the venous sinuses.

Headache usually develops when more than 10% of CSF volume is lost (18). In a series of experiments, PDPH was induced in volunteers by draining 15 to 20 mL CSF uniformly and rapidly. Replacement of the CSF volume with a sterile crystalloid relieved the headache completely and in less than a few minutes (18). Another cause of the headache could be from intracranial hypotension after leakage of CSF. This may explain why injection of the crystalloid into the epidural or subarachnoid space, or insertion of blood into the epidural space in an attempt to increase the epidural and subarachnoid pressure, can relieve a headache rapidly. Normal lumbar CSF pressure in the supine position is about 5 to 15 cm H₂O and can increase to about 40 cm H₂O in the sitting position (19).

In addition to the CSF, the arachnoid space contains veins that may dilate as a result of loss of CSF in order to maintain constant intracranial volume. This concept is defined by the Monro–Kellie doctrine, which states that the total intracranial volume is fixed because of the inelastic nature of the skull (20). The intracranial volume is equal to the sum of its components: Brain, CSF, and blood. When there is a decrease in one component, such as loss of CSF, the compensatory action to maintain constant intracranial volume may be vasodilation. Meningeal vessels may be pain sensitive and vasodilation can lead to increased cerebral blood flow. Also, when there is a
sudden decrease in CSF volume, adenosine receptors may be activated to compensate by producing arterial and venous vasodilation. The relief of headache with caffeine supports the vascular theory of PDPH from vasodilation since caffeine inhibits adenosine receptors to act as a cerebral vasoconstrictor (21).

Patients between the ages of 20 and 30 are more likely to develop PDPH after a dural puncture (39,40). Children and patients who are older than 60 years do not commonly report the headache (40,41). This may be due to older people having lower CSF pressures (42,43) and decreased elasticity of the dura, so dural perforations may not stay patent as compared to dural punctures in younger people (16,44). Vandem and Dripps studied over 9,000 patients who had received a spinal anesthetic and found the incidence of PDPH in the 20- to 29-year-old age group to be 16%. This was significantly higher than the 40- to 49-year-old (8%) and 50- to 59-year-old (4%) age group (45).

Most large studies find that there is a significantly higher incidence of PDPH in women compared to men (45,46,47). This may be a reporting error due to differences in pain perception or that women may be more likely to report a headache. Another explanation for this could be due to hormone-mediated cerebrovascular changes, similar to what might occur in migraine headaches. A large study of spinal anesthetics by Lybecker et al. found no significant association between gender and incidence of PDPH (40).

The incidence of PDPH is higher in people with a low BMI (31,48,49,50). One reason could be that there may be increased intra-abdominal pressure from obesity raising CSF pressure, so that even with a CSF leak, these patients do not develop headache. Another is that in attempted epidural placement, the epidural space may be unexpectedly shallow, resulting in an ADP. The incidence of ADP may be higher in the morbidly obese; however, there are not many studies that support this theory (37).

It is unclear whether this is a risk factor by itself, published studies showing the opposite finding (51,52). Laboring patients may be at an increased risk because many of them receive epidural analgesia with a large-bore needle and good positioning may be challenging due to the gravid abdomen as well as influence of contraction pain.

Recent studies have demonstrated that the dura has varying thickness. One reason that some people do not develop a PDPH may be that if the dural puncture occurs in a thicker area, a headache may not develop (12,14).

Many studies have found that patients with a previous history of PDPH or chronic headache are more likely to develop PDPH (40,53,54).

This is a rare, but serious consequence of dural puncture that can occur from intracranial hypotension causing tearing of the bridging veins. The patient can initially present with symptoms of a PDPH, but then the headache becomes severe, non-postural, and associated with neurologic deficits. Zeidan et al. reported a patient who received spinal anesthesia with
a 26 gauge (G) needle for cesarean section who postoperatively developed a severe non-postural headache associated with right eye tearing, V cranial nerve palsy, and left hemiparesis. A cranial subdural hematoma was confirmed with a computed tomography (CT) scan (57). He then performed a literature review and studied 46 patients who received spinal or epidural anesthesia and developed PDPH complicated by subdural hematoma. Most had associated neurologic deficits in addition to the severe headache. Sharma reported a case of a 31-year-old primigravid term woman who had a dural puncture during attempted epidural analgesia for labor and an intrathecal catheter was placed. Twenty-three hours later she had sudden loss of consciousness, her CT scan showed a large subdural hematoma, and craniotomy revealed active bleeding from a ruptured temporal bridging vein (58). Intracranial hypotension occurred while an intrathecal catheter was present demonstrating that the hypothetical plug that a large-bore catheter provides does not necessarily prevent efflux of CSF. Amorin reviewed 35 case reports of intracranial subdural hematoma that occurred after spinal anesthesia. These patients experienced headache (74%), changes in the level of consciousness (40%), vomiting (31%), hemiplegia or hemiparesis (23%), diplopia or VI cranial nerve paresis (14%), and language disorders (11%) (59). After this review, he recommended neuro-imaging evaluation in patients whose headache persisted for more than a week and with disappearance of the postural component. He determined contributing factors to include pregnancy, multiple attempts, use of anticoagulants, intracranial vascular abnormalities, and brain atrophy. In his review of 35 case reports, four patients (11.4%) did not survive (59). Subdural hematomas usually need to be surgically evacuated and time to evacuation is important depending on the size of the hematoma and how rapidly it is expanding (see Fig. 27-3).

If the patient had a headache and high blood pressure prior to or after a neuraxial block was placed, the diagnosis of preeclampsia would have to be considered. New seizure activity in this setting would also lead to increased suspicion of eclampsia or an intracranial vascular injury.

This predominantly occurs in females, the headache is acute in onset, there are focal neurologic signs, there can be a postural component, and it frequently presents during pregnancy and after delivery. Vasodilation that occurs as a result of CSF leakage has a role in the formation of thrombus. A stroke of venous origin is greater in pregnancy because of the prothrombotic state, and hypercoagulability is increased after delivery due to volume depletion and vascular trauma. MRI or angiography may be superior to CT in diagnosis in visualizing venous thrombosis (60). Therapy to prevent seizures and ischemia consists of heparin, oral anticoagulants, and possibly antibiotics for a favorable outcome.

The symptoms of PRES include headache, focal neurologic deficits, seizures, acute changes in blood pressure, and altered mental status. The pathophysiology of PRES is severe hypertension or vasoconstriction leading to brain hypoperfusion that results in altered cerebrovascular regulation and vasogenic edema. PRES is associated with preeclampsia/eclampsia occurring postpartum, and the presence of a coincidental dural puncture may confuse or delay the diagnosis.