Post Lumbar Puncture Headache
Lumbar Puncture and Post-Dural Puncture Headaches: Implications for the Emergency Physician
Robert L. Frank, MD, FAAEM
J Emerg Med. 2008;35(2):149-157. ©2008 Elsevier Science, Inc.
Posted 08/12/2008
Abstract and Introduction
Abstract
Lumbar puncture is a diagnostic procedure commonly performed by emergency physicians. Post-dural puncture headaches occur frequently after this procedure and can be associated with significant morbidity and, occasionally, even death. There is also a lot of variation in how post-dural puncture headaches are treated once they occur. This article seeks to examine the science behind post-dural puncture headaches, their prevention and treatment.
Introduction
Post-dural puncture headache (PDPH) is the most common complication of procedures in which the dura is penetrated, such as diagnostic lumbar punctures, spinal anesthetics, myelograms, and inadvertent dural punctures during epidural injections.[1] PDPH was first described in 1898 by August Bier, who experienced this condition first hand after experimentation with spinal anesthesia with co-pioneer August Hildebrandt.[2] Patients with PDPHs often present to the Emergency Department to seek care for this condition. Additionally, emergency physicians frequently perform diagnostic lumbar punctures, thus putting these patients at risk for PDPH. Because emergency physicians are often faced with treating PDPHs, it is imperative that they are able to accurately diagnose them and are familiar with the available therapies and their effectiveness and risks. Additionally, are there techniques that emergency physicians should utilize in the performance of lumbar punctures to minimize the incidence of PDPHs? This article seeks to examine the science behind these issues.
Anatomy and Physiology
The spinal cord is bathed in cerebrospinal fluid (CSF). The CSF is contained within the dura mater, which extends from the foramen magnum to the second sacral spinal segment. The dura mater is a tubular structure made of dense collagen and elastic fibers. It was originally thought that the fibers were oriented in a direction parallel to the spine, but more recent studies using electron microscopy have shown no particular or consistent pattern of the dural fibers.[3,4] The thickness of the posterior dura is also variable and unpredictable, which may have implications in the occurrence of PDPHs.[3] The normal volume of CSF in the adult is 150 mL, and several times this amount is secreted and reabsorbed by the choroid plexus daily.[5] The arachnoid mater is a thin layer of tissue that is loosely adherent to the under surface of the dura.
Pathophysiology
The headache that ensues after dural puncture is theorized to be primarily due to loss of CSF from a defect made in the dura with resultant intracranial hypotension.[6] A large defect allows for greater loss of CSF and increases the likelihood of intracranial hypotension and PDPH.[7] When the patient then assumes an upright position there is downward traction on pain-sensitive intracranial veins, meninges, and cranial nerves caused by gravity and loss of buoyancy from the reduced CSF pressure.[8,9,10] This "sagging" of intracranial structures has been demonstrated on magnetic resonance imaging.[11] CSF leak after dural puncture has been well documented.[12] Manometric studies have demonstrated that adult subarachnoid pressure is reduced from the normal 5-15 cm H2O to < 4 cm H2O, and the rate of CSF leak is generally greater than CSF production.[12] Support for this theory of PDPH also lies in the fact that intrathecal or epidural injections of saline restore the CSF volume, increase epidural and subarachnoid pressure, and alleviate the headache.[13,14] Other proposed mechanisms for the pain of PDPH include a resultant dilatation of the cerebral blood vessels as a compensatory mechanism to restore intracranial volume, which may further exacerbate the symptoms.[6] Clark also found that patients with low CSF levels of the neurotransmitter substance P, which is released with dural puncture, were three times more likely to have PDPHs than those with higher levels.[15] Hypersensitivity to substance P with upregulation of receptors is thought to be a causative factor in post-dural puncture as well as other types of headache.
Incidence
PDPHs occur with a wide range of reported frequency, from <1% in some studies to as much as 70% in others.[16,17] This wide range is due to the fact that certain patient populations and dural puncture procedures carry quite variable degrees of risk for PDPH. Patients who receive dural punctures with large-bore needles (diagnostic myelography and inadvertent dural puncture during labor epidural anesthesia) are very likely to sustain PDPHs. Patients who receive spinal anesthesia with small (24-30-gauge) non-cutting needles generally have significantly reduced risk of PDPH, with rates as low as 2% or less.[7] PDPH after diagnostic lumbar puncture (excluding myelography, pneumoencephalography, and cisternal puncture) has an incidence of 6-40%.[9,18,19,20,21] This number is reduced to about 5% when special measures are taken to reduce PDPH.[22,23]
Symptoms
Although PDPHs can occur from immediately to months after dural puncture, 90% occur within the first 72 h and most within 48 h.[24,25] Headaches that occur beyond this time frame should raise suspicion for other causes. PDPH is typically frontal or occipital, "burning," and radiates into the neck and shoulders.[5] Pain, though, may be in any location in the head or neck.[26,27] Isolated back pain may infrequently be the presenting complaint.[28] The most important diagnostic feature is that the headache is minimal or absent in the supine position and is exacerbated with upright posture.[5] If this component is not present, diagnosis of PDPH should be reconsidered. PDPH is often associated with nausea and vomiting, photophobia, diplopia, low back pain, neck stiffness, dizziness, tinnitus, hearing changes, and cranial nerve palsies.[27,28,29,30] PDPHs can be quite variable in their severity and duration. Seventy-two percent of PDPHs will resolve spontaneously in 7 days.[31] Case reports have documented a PDPH lasting as long as 8 years.[32] The headache may be mild and resolve spontaneously in a few days, or severe, lasting a week or longer and causing significant disability. A study by Tohmo et al. found that 39% of patients with PDPH had 1 week of impaired ability to perform activities of daily living.[33] Additionally, PDPHs that persist untreated can predispose to subdural hematoma, herniation, and death.[34,35,36]
Diagnosis
Diagnosis of PDPH is usually a clinical diagnosis, generally aided by the history that there has been some recent procedure with actual or potential risk of dural puncture. This may include diagnostic lumbar puncture, myelogram, spinal anesthetic, epidural anesthetic or steroid injection, and spine surgery. A characteristic positional headache is also essential. Various imaging techniques to aid diagnosis have been described but are rarely needed except to rule out other causes of headache.[37,38] In the postpartum period, other causes of headache such as preeclampsia, posterior leukoencephalopathy, caffeine withdrawal, migraine, and cocaine abuse should be considered. A procedure described by Gutsche also may be helpful in patients in whom the diagnosis of PDPH is uncertain.[39] Firm continuous abdominal pressure is applied with the examiner's hand while the headache is present. A PDPH will usually be significantly relieved within 30 s and return when pressure is released. This is thought to be due to an increase in pressure relieving CSF hypotension.
Risk Factors for PDPH
Demographic and Other Risk Factors
There are certain demographic factors that seem to be associated with risk of PDPH for reasons that are not well understood. Patient age is a risk factor, with ages between 18 and 40 years the highest risk range.[26,32,40,41,42] The risk of PDPH at age 25 years is 3-4 times that at age 65 years.[31,42,43] Children younger than 13 years rarely get PDPH.[44,45] This is thought to be due to lower CSF pressure in children.[46,47] PDPHs do occur with increasing frequency in adolescents and are similar to those seen in adults.[48,49] There is also significantly decreased frequency after age 60 years, which also may be related to reduced CSF pressure.[50] Female sex, regardless of age, is also a risk factor for PDPH for unknown reasons. Women have approximately twice the likelihood compared to men.[18,31,51] Race does not seem to be a factor in development of PDPH.[25] A history of chronic or recurrent headache has been found in nearly 60% of those with PDPH.[15] Previous history of PDPH is also a risk factor for development of future headaches.[43] Although there is a high incidence of pregnancy-related PDPHs, a meta-analysis by Morewood did not show that pregnancy in and of itself was likely to be the cause.[9] The increased frequency may be more likely related to patient age and sex. Inadvertent dural puncture during the performance of labor epidural anesthesia significantly increases the risk of PDPH, but even in pregnant women who receive only spinal anesthesia, the risk of PDPH is considerably elevated. Volume of CSF removed and its role in causing PDPH is unclear. Removal of 15-20 mL of CSF reliably caused headaches in one study.[26] Volume of CSF removed did not seem to be a factor in a study by Kuntz et al..[18] The small volume usually removed during diagnostic lumbar punctures performed by emergency physicians is not likely to be a significant factor. Experience of the practitioner performing the dural puncture does not seem to be a factor in development of PDPH, nor does multiple attempts at lumbar puncture.[52] Low body mass index is another a risk factor for PDPH.[18] Thus, young, thin women seem to be at highest risk for PDPH. These results are summarized in Table 1 .
Spinal Needles and Their Relationship to PDPH
Size of Spinal Needle
There are several characteristics related to the spinal needle used that are critical determinants in development of PDPH. Although these issues have long been part of anesthesia practice, they are rarely considered by other specialists who perform dural puncture procedures.[53,54,55] Thoughtful consideration of these issues is one area in which emergency physicians can have a significant impact on reducing the occurrence of PDPHs. The first issue relates to size of the needle. Larger needles leave larger holes, allow for greater CSF leak and, thus, cause more PDPHs.[7,43,50,53,56,57] The incidence of PDPH with the standard 20- or 22-gauge Quincke cutting beveled needle commonly used by non-anesthetists for diagnostic lumbar puncture is as high as 40%.[9,18,19,20,21] This could be reduced to as low as 5% using a similar 24-27-gauge needle.[58] It has been a long-held belief that the small needles used for performance of spinal anesthetics allow for too-slow fluid collection and unacceptable difficulty for use in the performance of diagnostic lumbar punctures.[7] Carson and Serpell found that use of needles smaller than 22-gauge required >6 min to collect 2 mL of CSF and measurement of opening pressure was similarly slow and potentially inaccurate.[7] Thus, it is felt by some that needles smaller than 22-gauge are inadequate for diagnostic lumbar punctures.[59] Strachan et al., however, showed that 2 mL of CSF could be obtained by gentle aspiration through a 24-gauge needle in <1 min.[60] In addition to using a smaller-gauge needle, using a needle with an atraumatic tip can further reduce the incidence of PDPH, as will be discussed below.
Needle Shape
An additional factor important in reducing PDPH is the shape of the tip of the spinal needle. Green, in 1926, showed that blunt-tipped spinal needles that separate dural fibers and allow recoil with minimal tearing significantly reduce CSF leak and PDPH.[61] Holst et al. showed with electron microscopy that atraumatic needles leave smaller holes in the dura that tend not to remain open and have three times less CSF leakage than Quincke (BD, Franklin Lakes, NJ) needles.[62] In recent years, several atraumatic (also known as pencil point or non-cutting) spinal needles have been introduced, with the Sprotte (B. Braun Medical Inc., Bethlehem, PA) and Whitacre (BD, Franklin Lakes, NJ) brands being most commonly used (Figure 1). The anesthesia literature has shown conclusively that atraumatic needles significantly reduce incidence of PDPH compared to cutting Quincke-type needles typically used by non-anesthesia practitioners.[53] Although less studied in diagnostic lumbar puncture, there is evidence to support the use of atraumatic needles for PDPH reduction. Thomas et al. showed that PDPH incidence could be reduced from 54% to 29% if a 20-gauge atraumatic needle was used rather than a Quincke cutting beveled needle.[63] The incidence could be further reduced to 4% when 22-gauge atraumatic spinal needles are used.[23] Strupp et al. found that patients who received a diagnostic lumbar puncture with a 22-gauge atraumatic needle had a PDPH rate of 12.2% vs. 24.4% in those who received lumbar puncture with a 22-gauge Quincke needle.[19] Other investigators have shown similar reductions in PDPH with the use of 22-gauge atraumatic needles.[54,64,65] Other studies have not shown a reduction in PDPH when atraumatic needles were compared to similar-gauge cutting beveled needles. Lenaerts et al. found no benefit when the 20-gauge atraumatic Sprotte needle was compared to the 20-gauge cutting Yale needle (BD Madrid, Spain).[66] These findings are likely related to the fact that even though an atraumatic needle was used, it was a large-bore needle that would cause a sizeable dural rent despite its non-cutting tip. Other investigators also have found no reduction in PDPH after diagnostic lumbar puncture when atraumatic 22-gauge needles were compared with cutting Quincke needles.[67,68] Disadvantages of the atraumatic needles includes increased cost, different "feel" and lack of "pop" that is often felt upon piercing the dura, occasional failure to obtain CSF, and difficulty penetrating the skin due to the dull tip. This last issue can be overcome by insertion of the atraumatic spinal needle through a traditional cutting 18-gauge needle acting as an introducer placed beyond the epidermis. Taking into consideration the desirable characteristics of a spinal needle for use in diagnostic lumbar puncture and the goal of reducing the risk of PDPH, a 24-gauge atraumatic needle may be the needle of choice, especially for those at high risk for PDPH. Atraumatic needles are slightly more expensive than similar Quincke needles (approximately $12 versus $4, respectively), although this cost could be easily offset by the reduced numbers of spinal headaches.[20]
Figure 1.
Drawing showing three types of spinal needle tips: the Quincke, the Whitacre, and the Sprotte.
Bevel Orientation
Another factor that significantly reduces likelihood of PDPH when using a cutting beveled Quincke needle is the orientation of the needle tip to the dura. Although the dural fibers in general have no consistent orientation, branching elastic fibers do tend to lie in an orientation that minimizes the size of the opening when the cutting bevel of the needle is inserted parallel to the long axis of the spine.[3,4] This causes fewer fibers to be cut than if the bevel were inserted in a perpendicular orientation. Thus, a smaller hole in the dura is created, less CSF leakage occurs, and there is reduced likelihood of PDPH.[43,52,55,69,70,71] Therefore, insertion of the needle with the cutting bevel parallel to the long axis of the spine is indicated. Tearing of the dura may also occur upon removal of the needle if it is rotated to a perpendicular orientation after insertion.[72] Bevel orientation is not an issue with atraumatic needles, as they tend to separate dura fibers rather than cutting them, allowing them to return to their original position with decreased CSF leakage.[61]
Stylet Replacement
Replacement of the spinal needle stylet before removal of the needle also has been shown to reduce the incidence of PDPH. Strupp et al. found that stylet reinsertion reduced PDPH from 16% to 5% with 21-gauge Sprotte atraumatic needles.[19] It was postulated that a strand of arachnoid mater may be pulled by the open needle on its removal and enter the dural rent, maintaining an opening that allows greater CSF leakage.[19] A similar study has not been performed using cutting Quincke needles. A similar effect may occur if the spinal needle strikes bone upon insertion. This has been reported to cause a burr to develop at the needle tip that will drag the arachnoid mater out as it is withdrawn, thus creating a CSF leak.[73] This concept has not been formally studied.
Bed Rest
Use of bed rest to prevent PDPH was first advocated by Bier in 1899.[20] It is still advocated by many that a patient must have some period of bed rest after a lumbar puncture to prevent PDPH. A meta-analysis by Thoennisen et al. and a Cochrane review by Sudlow and Warlow evaluated all studies on this topic and found no difference in incidence of PDPH for those with immediate mobilization vs. bed rest for up to 24 h.[29,74] Thus, bed rest after lumbar puncture has no role in PDPH prevention.
Hydration
It is advocated by some that increased hydration after dural puncture will minimize PDPH occurrence, possibly by increasing the rate of production of CSF to replace the fluid lost from leakage. This was not found to be beneficial in the only study that has examined this issue.[75] Thus, it would seem that intake of additional fluids after a lumbar puncture does not assist in PDPH prevention.[76]
Patient Positioning During Lumbar Puncture
The position that the patient is placed in during LP has been postulated to play a role in the development of PDPH.[77] A controlled study did not show the position used during performance of LP to have a relationship to development of PDPH.[78] Strategies the emergency physician can employ to reduce the risk of PDPH are summarized in Table 2 .
Treatment of PDPH
Treatment options for dural puncture headache depend on many factors such as severity of headache, degree of interference with activities of daily living, associated symptoms, response to conservative care, and presence of potential contraindications to more definitive therapies. Multiple treatment regimens have been advocated with varying degrees of success and risk, including medications and epidural blood patches ( Table 3 ).
Methylxanthine Derivatives
Methylxanthine derivatives such as caffeine and aminophylline have been recommended for the treatment of PDPH. It has been postulated that at least part of the pain from PDPH is due to cerebral vasodilatation as a compensatory attempt to restore intracranial volume.[6] It is thought that methylxanthine medications cause vasoconstriction of these vessels, thus decreasing pain. It has also been theorized that these drugs antagonize purine receptors and relieve headache by this mechanism.[25] Caffeine is usually given as 500-mg caffeine sodium benzoate in 1 L of intravenous fluid over 1 h, although some have given it as a rapid intravenous bolus.[79,80] A second dose usually can be repeated in 1-2 h if needed. If aminophylline is used, it is given 5-6 mg/kg over 20 min or given orally as theophylline 300 mg every 6-8 h.[25] Side effects include central nervous system stimulation, seizures, gastric irritation, and provocation of cardiac dysrhythmias.[80] Methylxanthine medications have been reported to be effective in alleviating PDPH in up to 90% of patients.[79,80,81,82,83,84,85,86] Unfortunately, the data used to support the efficacy of caffeine are limited to one oft-cited small, methodologically flawed study and several case reports.[73,76,77,80,82,84,86,87,88,89] The effects of caffeine seem to be temporary at best, with headache recurrence rates of up to 60%.[9,83,86,89] It has no effect on CSF leakage nor does it restore normal CSF dynamics, which are thought to be the primary causes of PDPH.[5] Camann et al. found oral caffeine to provide significantly better pain relief than placebo for PDPH initially, but there was no significant difference in pain scores at 24 h or in the number of epidural blood patches performed between the two groups.[89] A recent North American hospital survey reports that most practitioners have abandoned the use of caffeine for treatment of PDPH due to its perceived ineffectiveness. Additionally, although it is sometimes recommended that patients consume caffeine after a lumbar puncture to prevent PDPH, it seems that this strategy, too, is ineffective.[90] Further quality investigation is needed to determine the effectiveness of caffeine and theophylline in the prevention and treatment of PDPH.
Sumatriptan
Sumatriptan is a serotonin agonist occasionally used in the treatment of migraines. It has been reported successful for the treatment of PDPH.[43,91,92] Others have found it to be ineffective.[93] A controlled trial found no evidence of benefit when using sumatriptan in the treatment of PDPH.[94]
Epidural Blood Patch
Despite a recent Cochrane review that was unable to draw unequivocal conclusions about its efficacy, the epidural blood patch (EBP) is generally considered the definitive treatment for PDPH, especially for those that are severe or debilitating.[5,16,29,43,44,81,95,96,97,98,99] It was first described in 1960 by Gormley after it was noted that "bloody taps" were associated with decreased incidence of PDPH.[100] EBP is performed by injecting 15-30 mL of the patient's blood into the epidural space through a Tuohy epidural needle.[16] This is ideally done at the site of the previous dural puncture. The injected blood spreads in both a cranial and caudal direction. The thecal sac is compressed and displaced and is thought to elevate and thus restore the subarachnoid pressure.[5] It is also thought that the blood clot that forms seals off the rent in the dura and prevents further leakage of CSF. Success rates have generally been reported in 72-98% of patients, although less favorable results have been noted.[1,16,39,81,97,98,101,102,103,104,105] Usually, there is immediate relief of the headache. EBP is most effective when performed at least 24 h after the initial puncture.[25] Complications are rare but include radicular pain from nerve root irritation or displacement, cranial nerve palsies, meningeal irritation, elevated intracranial pressure, paraparesis, cauda equina syndrome, infection, and subdural hematoma.[5,106,107,108] Complications are usually either rare or self-limited. Contraindications to EBP include patient refusal, fever or suspected bacteremia and anticoagulation.[5,108] EBP has been successfully used in adolescents.[48] Patients who get no or incomplete relief after a first EBP have equivalent or higher rates of success on a second attempt.[25,102] Failure of EBP is seen most often in patients with dural puncture from large bore needles, such as with inadvertent dural puncture with 16-gauge Tuohy needles during performance of labor epidural anesthesia.[1,102]
Prophylactic Epidural Blood Patch
Some practitioners advocate use of a "prophylactic" EBP to prevent PDPH, especially in those at high risk (partiuent with inadvertent dural puncture with a large bore epidural needle). The datasupporting this practice are contradictory. Several studies have shown benefit, others have not.[109,110,111,112] A recent Cochrane review found there were insufficient quality data to reach a definitive conclusion on the effectiveness of this practice.[29] This practice is not likely to have much impact on the prevention of PDPH from diagnostic lumbar punctures performed by emergency physicians.
Conclusions
Dural puncture procedures are commonly performed by various medical practitioners, including emergency physicians, for many reasons. PDPH is a relatively common complication. The degree of pain and disability caused by PDPHs is considerable. Appropriate use of smaller-gauge, atraumatic needles is likely the single most important factor in reducing PDPH. Hydration and bed rest do little to prevent PDPH. Treatment of mild, non-debilitating headaches usually can be accomplished with limited activity, analgesics, and methylxanthine therapy until they resolve spontaneously. For more severe headaches, methylxanthine derivatives may be tried with EBP as a first-line alternative or to be used if methylxanthines fail. It is helpful to use 24-gauge Sprotte spinal needles in all patients at risk for PDPH. For patients who have moderate to severe pain, cannot or do not want to remain on bed rest, and those in whom narcotic pain medication is not a practical treatment option, EBP is an appropriate first-line therapy. In general, EBP is safe, well tolerated and effective. Though the technique for epidural blood patch is not significantly more difficult than lumbar puncture, it is a procedure most commonly performed by anesthesiologists.
