Pathophysiology

The pathophysiology of the MPH is not completely understood, but there are several proposed hypotheses. One is based on the Monro–Kellie rule and another is based on mechanical traction. Both hypotheses accept that CSF escapes through a known or probable meningeal puncture at a rate that exceeds CSF production. The Monro–Kellie rule states that in an intact skull, the sum of the volumes of brain, CSF, and intracranial blood are constant; therefore, with CSF volume loss, compensatory vasodilatation and venous hypervolemia occur, which may contribute to the headache. The average CSF production rate is 500 mL/day with an average adult intrathecal volume of 150 mL. CSF leakage from a meningeal puncture, leading to a loss of CSF pressure, contributes to a loss of the buoyancy support of the brain and is one theory that contributes to the headache. A higher level of lumbar puncture reduces the hydrostatic pressure at the meningeal puncture site, which may explain why MPHs are not commonly associated with cervical punctures. The uncompensated loss of CSF leads to a subarachnoid deficit of CSF and often a reduction in the subarachnoid pressure. The normal CSF opening pressure in the horizontal position is 70–180 mm H ₂ O. Although CSF hypotension (CSF pressure < 60 mm H ₂ O) is often noted, the significance of the reduction in subarachnoid pressure is unclear, because it does not consistently correlate with the presentation of headache. The development of an MPH does not seem to be related to the amount of CSF leaked. It is probable that the headache is related to sudden alterations in CSF volume, as proposed by Raskin, who theorized that the sudden loss of CSF volume and the change in pressure differential between the inside and outside of the intracranial venous structures result in venous dilatation. The direct traction hypothesis states that the reduction in CSF total volume, especially in the spinal region, allows the brain to shift caudally, placing traction on the pain-sensitive intracranial structures and causing cerebral vasodilatation that produces the classic headache symptoms. Pain-sensitive intracranial structures include the dura, cranial nerves, and bridging veins. The ophthalmic branch of the trigeminal nerve, which refers pain to the frontal region, innervates the bridging veins and the dura. In addition to causing pain, traction on bridging veins can cause a tear in the dura, thus leading to a potential subdural hemorrhage. The posterior fossa structures are innervated by the glossopharyngeal and vagus nerves that refer pain to the occipital region. Traction of the vagus nerve can also stimulate the chemoreceptor regions of the medulla causing nausea and vomiting. Finally, traction on the upper cervical nerves can present as occipital, cervical, and shoulder pain and stiffness. Schabel et al. reported one case of arm pain following meningeal puncture that resolved with an epidural blood patch (EBP). In addition to generating pain, traction or pressure on the abducens nerve (CN VI) can cause nerve palsy with paralysis of the lateral rectus muscle; this can manifest as diplopia. Diplopia usually occurs 4–10 days after meningeal puncture but can persist for up to 3 weeks. Full recovery can generally be expected in 2 weeks to 8 months, although permanent cases have rarely been reported. Another proposed mechanism for the visual changes is secondary to crowding of the optic chiasm, which is observed on the magnetic resonance imaging (MRI) of patients with intracranial hypotension. Finally, oculomotor nerve (CN III) and trochlear nerve (CN IV) palsies have been attributed to intracranial hypotension due to brainstem compression and ischemia. In contrast, the Monro–Kellie hypothesis proposes that a reduction in intracranial CSF volume is compensated by increased intracranial blood volume, because the sum of the brain, CSF, and blood volumes is constant. In accordance with the Monro–Kellie rule, the increase in blood volume causes cerebral vasodilatation, which activates the trigeminovascular system, similar to what occurs in migraine attacks. The input reaches the thalamus through the quintothalamic tract and refers pain to the ophthalmic branch and the first three cervical roots. This hypothesis is supported by MRI observations of contrast enhancement of the thickened meninges in MPH secondary to dural venous dilation. Decreased intracranial pressure is probably a secondary cause, because not all patients with classic MPH have intracranial hypotension.

The Role of Arachnoid Mater in the Pathogenesis of Cerebrospinal Fluid Leak

Reina et al. demonstrated that the dura mater is composed of 78–82 overlapping layers in multiple orientations, therefore making a hole exclusively parallel or across the fibers impossible. Furthermore, as early as 1938, Weed postulated that the arachnoid might be the barrier between the dura and the CSF. In 1967, Waggener and Beggs, based on electron microscopy observations, labeled the arachnoid membrane as a physiologic barrier impermeable to CSF. However, Nabeshima et al. demonstrated by electron microscopy that tight junctions exist only in the outer layer of the arachnoid, similar to those found in capillary endothelium of the brain. The cells of the dura do not contain tight junctions. Because of these anatomical observations, the concept of exclusive dural puncture to access the CSF and cause a CSF leak is not correct. In a study to differentiate the comparative permeability of the three meningeal layers, Bernards and Hill found the arachnoid mater to be the principal diffusion barrier to CSF. Although anesthesiologists probably intend to include puncture of the arachnoid when they refer to a post-“dural” puncture headache, the importance of arachnoid puncture for CSF access should be emphasized.

Meninges and Response to Trauma

In neurosurgical experience, often-minor meningeal perforations need to be closed either directly or through the application of synthetic or biological meningeal graft material. Failure to close a meningeal perforation may lead to adhesions, a continuous CSF leak, and increased risk of infection. It was initially thought that meningeal closure was facilitated by fibroblastic proliferation from the cut edge of the dura mater. However, in 1959, Keener showed that meningeal repair was facilitated by fibroblastic proliferation from surrounding tissue and blood clots. His study also showed that dura mater repair was promoted by damage to the pia or arachnoid mater and the presence of blood clots. It is therefore possible that a spinal needle carefully placed in the subarachnoid space may not promote substantial dural healing, as trauma to adjacent tissue is minimal.

Diagnosis

The diagnosis of an MPH is described according to the International Classification of Headache Disorders. It is primarily based on the history of a meningeal puncture or possible meningeal puncture with the headache worsening within 15 minutes after sitting or standing; improving within 15 minutes after lying down; and accompanied by the presence of at least one symptom among neck stiffness, tinnitus, hypoacusia, photophobia, and nausea. The time period within which the headache appears is debatable, and it can occur within a day or take as long as 12 days to develop. This differentiates it from pneumocephalus-related headache that is most often present immediately following an inadvertent meningeal puncture and injection of air. These factors will largely establish the diagnosis of MPH that can be accompanied by a multitude of signs and symptoms as illustrated above. Because the diagnosis of MPH is largely based on a thorough history and physical examination, it is important to note that certain critical signs and symptoms may indicate concomitant intracranial pathology. The most important of these signs is a changing pattern of the headache. For example, if the headache is no longer postural, if it becomes constant, if it becomes localized unilaterally, or if there is new-onset nausea and vomiting, then further investigation is warranted. Another critical change is increasing neurological alterations, which include sedation, seizures, and new-onset motor and/or sensory deficits. The presence of these signs and symptoms necessitates neurology consultation and additional diagnostic studies. Based on case reports, the differential diagnosis of MPH with changing symptomatology should include intracerebral hemorrhage, infection, eclampsia, and cerebral venous thrombosis.

Incidence

The incidence of MPH has a very wide range, from 1% to 63%. The determinants of this difference in incidence include the needle size, the design of the needle tip, and the orientation of the needle bevel during meningeal puncture. Smaller needle diameters correlate with a lower incidence of MPH. A study on cadaver dural samples found a sixfold greater mean leak when using 22-gauge needles compared to 25-gauge needles. Another study showed that with a 22-gauge Quincke, the rate of MPH was 11% compared to an incidence of 7% using a 25-gauge pencil point needle, although an obvious flaw in this study was the presence of two independent variables. It is important to realize that most unintentional meningeal punctures during epidural anesthesia occur with a 17- or 18-gauge Tuohy needle, which is a cutting needle. In an in vitro study comparing epidural needle diameter and CSF leakage, Angle et al. found reduced CSF leakage with a 20-gauge versus a 17-gauge Tuohy needle puncture. The effect of needle tip design may be more important for lowering incidence of MPH. Spinal needles are designed with cutting bevels, as in the Quincke-type, or pencil point tips, as in the Whitacre-type. Many studies have shown that noncutting pencil point needles (Whitacre-type) are associated with a lower incidence of MPH. The incidence of MPH with a 20- to 22-gauge cutting needle is estimated to be 36% compared to an incidence of less than 2% with a 22-gauge noncutting needle. There appears to be an even lower incidence of MPH with a larger-diameter blunt tip needle compared to a smaller-diameter cutting needle. One study showed a 2.7% incidence of MPH with a 27-gauge Quincke needle versus 1.2% incidence with a 25-gauge Whitacre needle. The proposed mechanism behind this difference is that a blunt tip needle divides but does not disturb the continuity of the meningeal fibers, whereas a cutting tip needle cuts the meningeal fibers. Electron microscopy studies show that a blunt tip needle produces an irregular hole in the dura versus a clean cut puncture observed with a cutting needle. ( Figs. 21.1 and 21.2 ). As discussed earlier, this may result in an increased inflammatory reaction with the blunt tip needle, which promotes hole closure. The orientation of the bevel to the dura during meningeal puncture has been proposed as a factor affecting the amount of CSF leakage and the incidence of MPH. In an in vitro study, Cruickshank and Hopkinson showed a 21% reduction in the leakage of CSF if the bevel was parallel to the long axis of the spinal cord. In a study by Norris et al. performed in 1558 parturients, the authors compared the incidence of meningeal puncture and MPH between orienting the epidural needle parallel or perpendicular to the long axis of the spinal cord during epidural catheter placement. Although both groups had a similar incidence of meningeal puncture, patients in the parallel orientation group reported significantly less MPH intensity and required fewer EBPs. Richman et al. also found that parallel, rather than perpendicular, insertion of the needle resulted in a statistically significant lower incidence of MPH, 10.9% versus 25.8%, respectively. We know from electron microscopy that the dura mater is a meshwork of collagen, and that these elastic fibers lack a specific orientation. However, the cells of the arachnoid mater are oriented parallel to the long axis of the spinal cord, which may explain the reduction in the rate of MPH when using a parallel insertion technique.

Two types of spinal needle tips: “Atraumatic Sprotte needle” (upper) and “traumatic Quincke needle” (lower) of the same diameter (left). As shown on the right, the atraumatic needle causes a smaller dural defect (upper) than the traumatic needle (lower). The smaller defect should theoretically result in a lower incidence of postlumbar puncture headache.

From Strupp M, Schueler O, Straube A, Von Stuckrad-Barre S, Brandt T: “Atraumatic” Sprotte needle reduces the incidence of postlumbar puncture headaches. Neurology. 57(12):2310-2312, 2001.

Four types of spinal needle tips. Gertie Marx, Sprotte, and Whitacre are noncutting needles, while Quincke is a cutting needle.

From Arendt K, Demaerschalk BM, Wingerchuk DM, Camann W: Atraumatic lumbar puncture needles: after all these years, are we still missing the point? Neurologist. 15(1):17-20, 2009.

Risk Factors

Independent risk factors of MPH include the female gender, pregnancy, falling in the age group of 20–50 years of age, and a having a low body mass index. The decreased incidence in obese patients is due to the increase in intraabdominal pressure, which may act as an abdominal binder that helps in sealing the defect in the meninges and decreasing the loss of CSF. On the other hand, younger women may be at a greater risk because of increased dural fibers elasticity, which tends to maintain a patent meningeal defect when compared to the less elastic dura in older patients. It was previously thought that MPHs were rare in younger children, but it has been demonstrated to occur with equal frequency in children of all ages. Vercauteren et al. reported a higher incidence of MPH after diagnostic meningeal punctures performed by neurologists and neuroradiologists. This is most likely due to the use of larger-gauge needles and less experience performing the procedure. Diagnostic lumbar meningeal punctures require a 20- or 22-gauge needle to facilitate measurement of opening pressure and withdrawal of CSF over a reasonably brief time period. With needles smaller than 22 gauge, collection of 2 mL of CSF may take 6 minutes or longer and measurement of CSF pressure may be less accurate. There is also a higher incidence of MPH in patients with a headache prior to the meningeal puncture and a history of prior MPH. Singh et al. performed a 5-year audit of accidental meningeal punctures and PDPHs in obstetric anesthesia. Of 40,894 consecutive parturients, they found a rate of 0.73% accidental meningeal puncture and a 0.49% incidence of MPH. However, higher rates, ranging between 1% and 2.6%, have been quoted. Choi et al. showed that in parturients who experienced an unintentional meningeal puncture while receiving a labor epidural, more than 50% developed an MPH.

Prevention

Prevention of MPH depends on needle size, needle tip, and bevel orientation during meningeal puncture. The smallest needle with a noncutting tip oriented parallel to the long axis of the spinal cord will reduce the incidence of MPH the most. In the case of accidental meningeal puncture while placing an epidural catheter, placing the catheter intrathecally has been shown to decrease the incidence of MPH. In a comparative outcome study of known subarachnoid punctures with 18-gauge Tuohy needles, Ayad et al. reported that intrathecal catheter placement after accidental meningeal puncture in obstetric patients reduced the MPH incidence from 81% in the group that had an epidural catheter placed to 31% in the group that had an intrathecal catheter placed and removed right after delivery and to 3% in the group that had an intrathecal catheter placed and left in place for 24 hours after delivery. The authors therefore recommended intrathecal catheterization for 24 hours to reduce the rate of MPH. Van de Velde et al. summarized results from a large number of studies and noted that the incidence of MPH was 51% when an intrathecal catheter was used and 66% when an epidural catheter was used and that the incidence of EBP was 33% and 59%, respectively. However, this analysis combined data from both retrospective and prospective studies, many of which were underpowered and not randomized. A survey among obstetric anesthesiologists practicing in North America reported that 75% of respondents place the catheter epidurally after accidental meningeal puncture and that 25% place it intrathecally; however, 76% believed leaving an intrathecal catheter in place for 24 hours would reduce the incidence of headache. Another survey among Australian obstetric anesthetists showed that 64% would remove the Tuohy needle and re-site it in the case of accidental meningeal puncture during the placement of a labor epidural. The most common reason behind this decision was concern regarding the safety of intrathecal catheters, particularly the risk of misuse.

Other proposed preventive procedures include prophylactic EBPs and epidural saline injections and infusions. In a small study, Charsley and Abram found that intrathecal injection of 10 mL normal saline reduced the incidence of MPH. Little evidence exists for prophylactic EBPs, though they are commonly done. Existing studies regarding prophylactic EBP are limited by small patient numbers. Other limitations include the small volumes of blood injected according to some reviews. In a quantitative systemic review, Apfel et al. found that the relative risk (RR) for headache after prophylactic EBP was 0.48 in five nonrandomized controlled trials (non-RCTs) and 0.32 in four randomized controlled trials (RCTs), but due to heterogeneity between the studies and publication bias from small non-RCTs with positive results, a large multi-center RCT to determine the best practice was recommended.

Treatment

Treatment of MPH should be initiated once the diagnosis has been clearly established based on history, physical examination, and appropriate diagnostic tests. Treatment options should be balanced by the understanding that 85% of MPHs last less than 5 days and that MPHs are rarely associated with significant morbidity. Treatment of MPH should start with conservative therapy. Recumbent bed rest relieves the symptoms of MPH but has no therapeutic benefit. A meta-analysis by Thoennissen et al. failed to show that bed rest after meningeal puncture was better than immediate mobilization in reducing the incidence of MPH. Aggressive hydration is a common therapy despite the fact that there are no studies to support its effectiveness. Medications reported to be beneficial in the treatment of MPH include methylxanthines, caffeine, theophylline, sumatriptan, adrenocorticotropic hormone, and corticosteroids. Caffeine, a potent central nervous system stimulant, causes cerebral vasoconstriction and is the most widely used pharmacologic therapy for MPH. Caffeine is administered as an oral dose of 300 mg or an intravenous dose of 500 mg in 500 or 1000 mL of normal saline infused over 2 hours; the intravenous dose can be repeated in 2–4 hours if needed. Although caffeine is safe and effective, there have been reports of seizures, anxiety, and arrhythmias associated with its use. The existing literature does not seem to support caffeine as a therapeutic agent to treat MPH. Caffeine is contraindicated in patients with a history of seizure disorder and in patients with pregnancy-induced hypertension. The effect of caffeine is transient and the dose must be repeated, because it does not address the underlying pathology. Theophylline, another cerebral vasoconstrictor effective in the treatment of MPH, is not widely used or supported by the literature. Other pharmacologic agents such as serotonin agonists (sumatriptan) and corticotrophin are infrequently used and have been found to be ineffective in the treatment of severe MPH. However, Bussone et al. found in a nonrandomized study that frovatriptan was useful in the prevention of MPH in patients undergoing diagnostic lumbar punctures. Recent studies suggest that drugs such as pregabalin and gabapentin might be useful in the management of MPH. One RCT compared the use of pregabalin for 5 days with placebo in 40 patients who developed MPH after spinal anesthesia or lumbar puncture and found that the group that received pregabalin had significantly lower pain scores after the second day of treatment and lower diclofenac requirements. Another RCT comparing the use of gabapentin with the use of ergotamine/caffeine in 42 patients with MPH found that gabapentin significantly reduced pain, nausea, and vomiting compared to ergotamine/caffeine. However, more studies are needed to confirm the usefulness of these drugs.

Once pharmacologic and other noninvasive options have been exhausted without relief and the patient is unable to wait for the natural resolution of the headache, more invasive options may be considered. Epidural treatments for MPH include the administration of saline, colloids, fibrin glue, and blood. The gold standard treatment for MPH is an EBP. There are numerous variations of the EBP; however, the standard and most common treatment remains the epidural autologous blood patch. This treatment offers complete resolution of symptoms in a large proportion of patients. In the remaining patients, it reduces headache severity and allows them to return to their everyday activities. Contraindications to an EBP are similar to those for any spinal or epidural procedure. The first is patient refusal in general or for a specific reason. For a Jehovah’s Witness patient concerned about blood transfusions, alternative patching materials are available. Second, the patient’s coagulation status must be assessed to reduce the risk of an epidural hematoma. Finally, it is not recommended to place an EBP in a septic patient, a febrile patient, or through a localized infection site due to the obvious concern of introducing bacteria into the epidural space. Concerns about an EBP in HIV-positive patients are unfounded, because HIV crosses the blood–brain barrier early in the course of the disease. The mechanism of EBP is controversial; however, it is generally believed to exert its effect via two separate mechanisms. There is an initial early effect, which occurs within minutes, secondary to compression of the dura toward the spinal cord and reduction in intradural volume. The EBP blood spreads both longitudinally and circumferentially, thus enveloping the entire dural sac. The reduction in the spinal intradural volume shifts the CSF cephalad, thus resuspending the brain and reducing traction. In agreement with the Monro–Kellie rule, this intracranial shift in CSF also reduces cerebral vasodilation and intracranial blood volume. Due to this early effect, patients often report rapid relief following an EBP. However, by utilizing post-EBP MRI, Beards et al. demonstrated that the compressive mass effect of a blood patch disappeared by 7 hours post patching. A second more lasting effect is due to sealing of the dural/arachnoid tear with a gelatinous plug. This sealing of the dural/arachnoid hole prevents further loss of CSF and allows for regeneration and restoration of the CSF volume. The plug acts as a bridge until permanent repair of the dural/arachnoid hole occurs. The occurrence of this second effect is more variable and can account for the failure of an EBP despite initial relief. Dubost et al. measured optic nerve sheet diameter (ONSD), which is a marker of intracranial pressure (ICP), in patients with MPH after receiving EBPs. They found that an EBP was followed by ONSD enlargement in subjects with a successful EBP but not in subjects with EBP failure. Since ONSD is a surrogate marker of ICP, this suggests that sustained increase in ICP is associated with successful EBP.

Risk factors for EBP failure include placement earlier than 24 hours after meningeal puncture, using too small of a volume of autologous blood, and performance of the procedure with residual local anesthetic in the epidural space. Ong et al. and Blanche et al. were unable to find any conclusive impact of an EBP on the efficacy of future epidural anesthetics.

The technique of an EBP, first described by Gormley in 1960, is straightforward based on the placement of a single-shot epidural. It usually requires two people, one to locate the epidural space and the other to draw the blood. Sterility is of great importance, both during epidural space localization and during blood collection. The patient can be either sitting or in the lateral decubitus position during the procedure depending on the difficulty in locating the epidural space and the patient’s ability to tolerate the upright position. Selection of the level of placement should be guided by the observation that 15 mL of blood preferentially spreads cephalad six segments and caudad three segments, or one spinal segment per 1.6 mL of blood. It is therefore common to select a site caudad to the suspected meningeal tear. Colonna-Romano and Linton reported that a lumbar EBP was successfully used to treat an MPH due to a C6–C7 cervical meningeal puncture. This may be due to the increase in the subarachnoid pressure and the resultant cerebral vasoconstriction and deactivation of the brain adenosine receptors. More caudad levels are also associated with a reduced risk of direct cord compression. Although historically different volumes of blood have been used, the ideal target volume is 20 mL. This is the most widely accepted and cited volume if the patient tolerates placement. If the patient complains of excessive back or leg pain or pressure during injection, a lower volume can be placed. Chen et al. found that a volume of 7.5 mL of autologous blood in the epidural space was comparable to 15 mL of blood in its analgesic effect on MPH but with less nerve root irritation pain during injection. After the EBP, the patient should remain supine with the legs slightly elevated. Intravenous fluids can be administered during this time. In a small study, Martin et al. found that 2 hours in the supine position, post-EBP provided 100% relief versus 60% relief in patients who remained supine for 30 minutes only. Although initial relief has been reported in the overwhelming majority of patients, the overall long-term relief of MPH from an initial EBP is between 61% and 75%. Alternative meningeal patching materials include epidural fibrin glue and epidural Dextran-40. Although both of these materials have been successfully employed, they are not widely used due to cost and safety concerns, especially when compared to autologous blood. Dextran-40 may be an alternative in Jehovah’s Witnesses. Another alternative to EBP is epidural saline boluses or infusions. Epidural saline boluses or infusions have not been shown to be effective alternatives and often require more interventions with lower success rates, as saline is absorbed from the epidural space faster than blood. In an RCT, Al-metwalli studied the effect of epidural morphine in the prevention of MPH after inadvertent meningeal puncture. One group received an epidural injection of 3 mg of morphine in 10 mL of saline, that was repeated in 24 hours, while the control group received an epidural injection of 10 mL of saline repeated in 24 hours. The incidence of headache was 12% in the morphine group and 48% in the saline group. The need for a therapeutic EBP was also less in the morphine group. As a last resort, surgical repair of a meningeal tear may be considered, although it is a more invasive procedure generally reserved for severe cases of chronic MPHs that have not responded to EBPs.

Complications of an EBP are rare. The most common complication is mild low back and radicular pain following the procedure that resolves spontaneously in a few days and can be treated with nonsteroidal antiinflammatory drugs (NSAIDs). Other possible complications include epidural hematoma, infection, and arachnoiditis due to unintentional subdural or subarachnoid injection of the blood. There are two reported cases of facial nerve paralysis following EBP, which resolved spontaneously. Mokri reported a case of symptomatic intracranial hypertension following an EBP.

Meningeal Puncture Headache Summary

MPH and its management is a well-known and widely accepted entity in the anesthesia community. A postural headache is the cardinal sign that indicates an MPH. In the setting of a meningeal puncture or possible meningeal puncture with pathognomonic symptoms, the diagnosis of MPH should be straightforward. Although it is nonfatal, it can have significant morbidity and should be treated seriously. Treatment should start with conservative therapy, as the headache usually subsides in 5–7 days. An EBP is the gold standard in treating MPH not responding to conservative management. Surgical repair is rarely required in patients with chronic MPH.