P&S Medical Review: Nov 1993, Vol.1, No.1
Post-Lumbar Puncture Headache

William Fearon
Class of 1994
Columbia University
College of Physicians and Surgeons, New York, NY

Introduction

HEADACHE following lumbar puncture was first described in 1898 by Dr. August Bier who stated, after his assistant Dr. Hildebrandt attempted to deliver a spinal anesthetic to him, `All these symptoms [pressure in the head and dizziness] disappeared as soon as I lay down horizontally, but they returned when I arose.'[1] Post-lumbar puncture headache (PLPH) is easily diagnosed by its postural nature and distinct history. Its exact pathophysiology is unclear but stems from persistent leakage of cerebrospinal fluid (CSF) from the puncture site. Typically self-limited in nature, the occasional persistence of PLPH has encouraged development of a variety of prophylaxes and treatments. This review will describe clinical characteristics of the syndrome, discuss theories of pathogenesis and detail common management.

Clinical Characteristics

PLPH can arise after any invasion of the subarachnoid space via dural puncture, including spinal anesthesia, lumbar myelography, diagnostic lumbar puncture (LP), and coincidental dural tap during epidural anesthesia. The reported incidence of PLPH ranges from 15 percent to 30 percent.[2,3] A review of the literature from 1891 to 1964 revealed a 32 percent incidence after diagnostic LP and an incidence of 18 percent in obstetric and 13 percent in non- obstetric patients after spinal anesthesia.[4] The incidence was less in the latter two groups when smaller needles were used, when anesthesiologists performed the procedure, and when fluid was injected, not removed.[4] PLPH occurs more often in young adults, who are affected four times more often than patients aged 60 to 69 years.[5] The lower incidence among older people has been suggested to arise from a decrease in elasticity of pain-sensitive structures secondary to atherosclerosis, or from age-related mechanical changes in the epidural space.[6] Young women experience PLPH about twice as often as young men, but several papers reporting this finding have not addressed age and sex as independent risk factors. [4, 5]

The onset of PLPH is usually within 24 to 48 hours after dural puncture but may be delayed as long as 12 days. Symptoms are usually self-limited. With conservative management 50 percent of headaches resolve spontaneously within four days, 75 percent within seven days, and 95 percent within six weeks.[6] The longest reported PLPH lasted 19 months.[8] Early onset of headache after LP correlates with a worse prognosis.

PLPH is typically described as dull or throbbing, frontal or occipital in origin and radiating to other areas of the head. Its most unique, often diagnostic characteristic is its postural nature; standing exacerbates the headache, while lying down partly or totally alleviates it. Headshaking, coughing and jugular compression worsen the symptoms. Associated complaints may include nuchal pain and stiffness, low back pain, diplopia, tinnitus, nausea and vomiting.

Pathogenesis

The exact pathophysiology of PLPH remains unclear. Bier first suggested that persistent leakage of CSF greater than its production by the choroid plexus causes a net decrease in CSF volume and intracranial pressure. When a patient stands up after LP, the resulting downward shift of the brain stretches pain- sensitive structures and results in a headache.[1] The work of Sciarra and Carter in 1952 at the Neurological Institute supported Bier's hypothesis. They performed LPs on 102 patients, half of whom had 10 to 12 mL of CSF removed, half of whom had no fluid removed. The investigators found no significant difference in incidence of PLPH between the two groups and concluded that whether or not CSF was removed during LP did not influence the subsequent development of PLPH.[9] Persistence of the dural hole and loss of CSF through the defect has been verified by direct observation at autopsies[10] and laminectomies,[11] by myeloscopy[12] and by radioisotope myelography.[13] The postural nature of PLPH lends support to this theory; in an upright position, CSF leakage and CSF hypotension are increased by gravity, further decreasing the brain's supportive cushion and exacerbating symptoms. More recently, Grant, et al. used magnetic resonance imaging to demonstrate that 19 of 20 patients had decreased CSF volume 24 hours post LP; 11 of the 20 complained of headache. These results suggest a relationship between intracranial CSF volume loss and development of PLPH.[14]

Other evidence points to different factors. Marshall performed LPs on 43 patients and twenty-four hours later repeated LPs on all 43 patients, whether or not a patient had developed PLPH.[15] In the 6 patients who had developed PLPH, repeat LP demonstrated opening pressures of 150, 105, 60, 35, 0, and 0 mm H₂0. Seven patients who had not developed PLPH all had opening pressures less than 50 mm H₂0. Marshall's results demonstrated no significant association between CSF pressure at repeat LP and presence of PLPH.

Some have postulated that development of PLPH may be more dependent upon creation of an intra- and extra-luminal pressure differential in intracranial vessels and not upon CSF hypotension per se.[16] In fact, jugular compression has been noted to exacerbate PLPH, despite increasing CSF pressure.[16] The common factor in both jugular compression and CSF removal appears to be the differential that results between intravascular pressure and surrounding intracranial pressure. Pressure changes are speculated to cause traction on pain-sensitive anchoring structures and consequently PLPH.[5]

The possible importance of vascular status is demonstrated by the fact that caffeine administration has been shown to relieve PLPH.[18] Caffeine causes cerebral vasoconstriction by blocking adenosine receptors, which dilate vessels when activated. Less cerebral blood flow decreases the amount of blood in the brain and may lessen the traction on pain sensitive supporting structures, relieving PLPH. Adenosine receptors may therefore play an important role in the syndrome's pathogenesis. The decrease in CSF volume secondary to persistent CSF leakage after LP may activate the adenosine receptors directly, causing cerebral vasodilatation, stretching pain sensitive cerebral structures and resulting in PLPH.

Psychological factors have been suggested as important in the pathogenesis of PLPH. Daniels and Sallie explicitly warned a subgroup of their patients about the possibility of PLPH and found a 46 percent incidence of PLPH in this subgroup compared to a 6 percent incidence in those who did not receive such notification.[19] Vandam and Dripps, however, delivered spinal anesthesia to 100 patients already under general anesthesia without their knowledge.[6] These investigators found the same incidence of PLPH in these patients as in those with previous knowledge of the procedure, suggesting that psychological factors play little role in development of this syndrome. Thus, the role of psychological factors in the development of PLPH is unclear and requires further study.

In an attempt to explain why not all patients develop PLPH after dural puncture, Gobel, et al. used mechanical pressure stimuli to induce pain prior to LP.[20] He classified patients into three groups with varying pain thresholds: sensitive, medium- sensitive, and insensitive. The percentage of those patients `sensitive' to the mechanical pressure stimuli developed PLPH at a significantly higher rate than those categorized as `insensitive'. These findings suggest that some patients may be at greater inherent risk of developing PLPH.

In another study Gobel, et al. used transcranial dopplers to demonstrate that patients who develop PLPH have a significantly higher systolic flow velocity in the right middle cerebral artery prior to LP.[21] There was significant decrease in flow velocity in the right middle cerebral artery after LP only in those patients who had developed PLPH. Although no inverse relation between flow velocity and vessel diameter was proved, one may be inferred. Therefore, Gobel's findings not only offer a predictor for which patients are most susceptible to PLPH, but also support the theory that dilatation of intracranial vessels plays a role in onset of PLPH.

Persistent leakage of CSF is the initiating abnormality in the pathogenesis of PLPH. Whether and exactly how CSF hypotension, cerebral vasodilatation or both cause traction on cerebral pain- sensitive supporting structures and contribute to PLPH development remains to be further clarified, as does any role of psychological factors.

Management

Prevention of this syndrome is controversial. Bed rest for several hours after LP has been the standard prophylactic measure since Sicard first suggested it in 1902.[4] Brocker added support to this recommendation in 1958 when he found a less than 0.5 percent incidence of PLPH in patients who remained supine for three hours after LP, compared to a 36 percent incidence in those who did not.[22]

More recently, however, some have suggested that bed rest after LP may only delay symptoms or decrease the severity of PLPH but not prevent it.[3] In a study of 300 patients, Vilming, et al. found no significant difference in incidence of PLPH among patients who ambulated immediately after LP and those who remained supine.[23] In addition, patients who ambulated and developed PLPH were found to have fewer associated symptoms, such as nausea, compared to those who remained supine and developed PLPH. Further evidence against prophylactic bedrest comes from Thornberry, et al. who, in a prospective randomized trial of 80 obstetric patients, found that early ambulation actually decreases the incidence and intensity of PLPH.[22] Prophylactic bedrest after LP is now less favored and early ambulation is recommended by some.[5, 22]

Introduction of the needle with its bevel parallel to the dura's longitudinal fibers is another prophylactic measure that may reduce the incidence of PLPH.[25] This method of insertion separates the fibers rather than cutting them, facilitating closure of the hole upon needle withdrawal. Dittman, et al. studied the effect of the bevel's position relative to dural fibers and found that although the shape of the hole is different, the size is unaffected.[26] The authors also discovered that dural thickness ranges between 0.5 mm to 2.0 mm and that holes in thicker areas tend to close more rapidly than holes in thinner areas. Fortuitous puncture of a thicker area might be expected to decrease the incidence of PLPH, which may explain the wide range reported.

During insertion the angle of the needle in relation to the thecal sac may also affect the incidence of PLPH. By inserting the needle at a narrow angle (more tangentially) the dura and arachnoid are punctured at different points along the course of the thecal sac. This leaves an unmolested area between the two holes which may become apposed upon needle withdrawal and prevent CSF leakage. Other prophylaxes such as removing the needle when the patient is prone with the patient's head down, providing adequate hydration (orally or intravenously), and using antidiuretics have been suggested, but never proved to decrease incidence of PLPH.

The use of an epidural blood patch (EBP) is well recognized for treatment of PLPH and has been suggested for prophylactic use. An EBP is performed by removing 10 to 20 mL of the patient's blood from a peripheral vein then slowly injecting it into the lumbar epidural space. It is not necessary to inject the blood at the exact spot of the LP because it will migrate a few spinal segments superiorly and inferiorly. After injection the patient remains supine for 6 hours. Berrettini, et al., however, found prophylactic EBP ineffective.[27] In their study the small amount of blood injected (only 3 mL) and the large needle used (20 gauge) may have skewed the results. In addition, CSF leakage is greatest during the first 24 hours post-LP and interferes with blood clotting. Heide and Deiner in 1990 demonstrated a 22 percent incidence of PLPH in patients with a prophylactic EBP versus a 45 percent incidence in those without.[28] They used a 22 gauge needle, injected 4 to 8 mL of blood and more carefully evaluated the time course and severity of patients' complaints. EBP may be an effective means of prophylaxis against PLPH, depending on the exact protocol, however, it is much less effective than therapeutic EBP (discussed below) and may be unnecessary in a typically mild, self-limited syndrome.

Recently, a device which delivers absorbable collagen through an introducer sheath to promote hemostasis at an arterial puncture site has been studied.[29] This method of sealing a puncture has not been reported as a means of prophylaxis against PLPH.

The only measure clearly and repeatedly shown to decrease the incidence of PLPH is the use of smaller gauge needles; incidence of PLPH decreases as needle size decreases.[30] The smaller needle creates a smaller hole which closes more rapidly resulting in less leakage of CSF. Many investigators have recently found that use of a blunt ended needle results in a significantly lower incidence of PLPH compared to a sharp ended needle.[31] A sharp needle cuts the dural fibers while a blunt needle merely pushes them aside, allowing the hole to reseal more easily upon needle withdrawal.

Although the PLPH syndrome typically resolves spontaneously, multiple methods of treatment have been tested with a range of outcomes. Initially, bed rest in the horizontal position and adequate hydration, are recommended. The former may decrease the amount of CSF leakage by eliminating the effects of gravity, while the later is thought to maintain a ready supply of fluid for the production of CSF. Several authors disagree with the later hypothesis, feeling that forced fluid intake is useless and has no effect on CSF production.[1, 5]

Simple analgesics and caffeine may ameliorate symptoms.[5] Intravenous administration of 500 mg of caffeine sodium benzoate has been shown to relieve 75 percent of PLPHs. A second dose of 500 mg can be given to those 25 percent unaffected by the first, and 10 percent more will be relieved of PLPH.[18] Presumably caffeine has an inhibitory interaction with the adenosine receptors.

In the past abdominal binders and epidural saline infusions (20 mL per hour for 48 to 72 hours) were suggested for persistent PLPH.[32] Both methods purport to slow or stop CSF leakage by building a favorable pressure differential between the epidural and subarachnoid spaces. It has been shown, however, that epidural saline injections actually increase both epidural and subarachnoid pressures so that the subarachnoid pressure remains greater than epidural pressure.[33] If true, another mechanism must be responsible for relief provided by this treatment. The increase in epidural pressure, although not leading to a pressure differential, may be enough to invert the edges of the dural puncture, making the hole smaller, lessening CSF leakage and aiding in ultimate closure.[4] It is also possible that the increase in epidural and subarachnoid pressures causes inactivation of adenosine receptors, leading to vasoconstriction and relief of PLPH.[4] Abdominal binders and epidural saline infusions are infrequently utilized modes of treatment because of their uncomfortable, unwieldy natures.

The role of an EBP in treatment of a PLPH was first suggested by Gormley who observed a decreased incidence of PLPH after traumatic LPs and hypothesized that the blood must somehow prevent PLPH.[34] He performed the first EBP in 1960; subsequently it has been shown to have a 90 to 95 percent success rate.[32] Headache typically improves within a few hours after EBP and resolves completely by the next day. EBP most likely leads to clot formation and plugging of the dural hole thereby stopping CSF leakage. However this mechanism explains only the longterm effects of the EBP. The mechanism responsible for the immediate relief provided by an EBP might be similar to one suggested for saline infusions, namely, inhibition of adenosine receptors. The side effects of EBPs include transient paresthesias, residual back stiffness in approximately 15 percent which usually resolves within days, but may last longer; acute aseptic meningitis after unintentional subarachnoid injection; and rarely, subdural hematoma.[4] If the first EBP fails, a second one may be attempted, often with success.

Alternatives to EBP in persistent cases include epidural Dextran 40 patching and surgical closure of the dural rent. Use of a Dextran 40 patch instead of blood has not been extensively studied but was shown to have a greater than 70 percent success rate in a series of 56 patients.[35] It may prove to be an important option for specific patients such as Jehovah's Witnesses who refuse blood products, but it does have the added risk of causing an allergic reaction. Surgical closure of the dural rent is rarely attempted but has relieved PLPH refractory to standard treatment.

Conclusion

After LP, headache associated with sitting or standing and relieved by reclining is a common occurrence in young adults. The pathophysiology of PLPH is related to persistent leakage of CSF, causing downward shift of the brain and traction on its pain- sensitive structures. Dilatation of intracranial vessels has also been implicated in its pathogenesis. Use of smaller gauge needles has been demonstrated to decrease the incidence of PLPH. Since PLPH is usually self-limited, conservative treatment such as bedrest, hydration, and analgesia are tried first. If persistent, PLPH may be alleviated with caffeine or EBP. The pathophysiology and treatment of PLPH remain controversial. Recent advances in imaging techniques and rigorous prospective studies should clarify much of this controversy.

References

Brownridge P. The management of headache in accidental dural puncture in obstetric patients. Anesthesia and Intensive Care 1983;11:4-15.

Flaatten H, Rodt SA.,Vannes J, et al. Postdural puncture headache using 26- or 29-gauge needles in young patients (abstract). Reg Anaesth 1988;11:5.

Carbaat PAT, van Crevel H. Lumbar puncture headache: Controlled study on the preventive effect of 24 hour bed rest. Lancet 1981;1:1133-5.

Fernandez E. Headaches associated with low spinal fluid pressure. Headache 1990;30:122-8.

Fishman, R.`Examination of the cerebrospinal fluid: techniques and complications.' in Cerebrospinal fluid in the diseases of the nervous system, 2d ed. Philadelphia: W.B. Saunders Company, 1992.

Drippps RD, Vandam LD. Long-term follow-up of patients who received 10,098 spinal anesthetics. JAMA 1954;156:1486-91.

Raymond JR, Raymond PA. Post-lumbar puncture headache. Western Journal of Medicine 1988;148:551-4.

Niall CT, Globerson JA, de Rosayro MA. Epidural blood patch for postdural puncture headache: It's never too late. Anesthesia and Analgesia 1986;65:895-6.

Sciarra, D, Carter C. Lumbar puncture headache. JAMA 1952;148:841- 42.

Thorson G. Neurologic complications after spinal anesthesia. Acta Chir Scand Suppl. 1947;121:34.

Brown BA, Jones OW. Prolonged headache following spinal puncture. Response to surgical treatment. J Neurosurgery 1962;19:349-50.

Pool JL. Myeloscopy: intraspinal endoscopy. Surgery 1942;11:169-82.

Lieberman LM, Tourtellotte WW, Newkirk TA. Prolonged post lumbar puncture cerebrospinal fluid leakage from the lumbar subarachnoid space demonstrated by radioisotope myelography. Neurology 1971;21:925-29.

Grant R, Condon B, Hart I, Teasdale GM. Changes in intracranial CSF volume after lumbar puncture and their relationship to post-L.P. headache. J Neurol Neurosurg Psychiatry 1991;54:440-2.

Marshall J. Lumbar puncture headache. J Neurol Neurosurg Psychiatry 1950;13:71-4.

Kunkle EC, Ray BS, Wolff HG. Experimental studies on headache: analysis of the headache associated with changes in intracranial pressure. Arch Neurol Psychiatry1943;49:323-58.

Forbes HS, Nason GI. The cerebral circulation. Arch Neurol Psychiatry 1935;34:533-47.

Sechzer PH, Abel L. Post-spinal anesthesia headache treated with caffeine. Part II: Intracranial vascular distention, a key factor. Curr Ther Res 1979;26:440-8.

Daniels AM, Sallie R. Headache, lumbar puncture, and expectation. Lancet 1981;i:1003.

Gobel H, Schenkl S. Post-lumbar puncture headache: the relation between experimental suprathreshold pain sensitivity and a quasi- experimental clinical pain syndrome. Pain 1990;40:267-78.

Gobel H, Klostermann H, Lindner V, Schenkl S. Changes in cerebral haemodynamics in cases of post-lumbar puncture headache: a prospective transcranial doppler ultrasound study. Cephalgia 1990;10:117-22.

Brocker RG. Technique to avoid spinal-tap headache. JAMA 1958;168:261-3.

Vilming ST, Schrader H, Monstad I. Post-lumbar puncture headache: the significance of body posture. A controlled study of 300 patients. Cephalgia 1988;8:75-8.

Thornberry EA, Thomas TA. Posture and post-spinal headache. A controlled trial in 80 obstetric patients. BR J Anaesth 1988;60:195-7.

Mihic Dn. Post-spinal headache and relationship of needle bevel to longitudinal dural fibers. Reg Anaesth 1985;10:76-81.

Dittman M, Schafer HG, Ulrich J, Bond-Taylor W. Anatomical re- evaluation of lumbar dura mater with regard to post-spinal headache. Effect of dural puncture. Anaesthesia 1988;43:635-7.

Berrettini WH, Nurnberger JI. Epidural blood patch does not prevent headache after lumbar puncture. Lancet 1987;i:856-7.

Heide W, Deiner HC. Epidural blood patch reduces the incidence of post lumbar puncture headache. Headache 1990;30:280-1.

Sanborn TA, Gibbs HH, Brinker JA, Knopf WD, Kosinski EJ, Roubin GS. A multicenter randomized trial comparing a percutaneous collagen hemostasis device with conventional manual compression after diagnostic angiography and angioplasty. J Am Coll Cardiol 1993;22:1273-9.

Tourtellotte WW. A randomized, double-blind clinical trial comparing the 22 versus 26 gauge needle in the production of the post-lumbar puncture syndrome in normal individuals. Headache 1972;12:73-8.

Braune HJ, Huffman G. A prospective double-blind clinical trial, comparing the sharp Quincke needle (22G) with an `atraumatic' needle (22G) in the induction of PLPH. Acta Neurol Scand 1992;86:50-4.

Raskin NH. Lumbar puncture headache: a review. Headache 1990;30:197-200.

Usubiaga JE, Usubiaga LE, Brea LM, Goyena R. Effect of saline injections on epidural and subarachnoid space pressures and relation to post-spinal anesthesia headache. Anesth Analg 1967;46:293-6.

Gormley JB. Treatment of postspinal headache. Anesthesiology 1960;21:565-6.

Aldrete JA, Barrios-Alarcon J, Tapia D, Penas M, Lavine L. Treatment of post-spinal headache with epidural Dextran 40. Anesthesiology 1987;67:A221.