INTRODUCTION — Post dural puncture headache (PDPH), also known as post lumbar puncture (LP) headache, is a common complication of diagnostic LP. It also can occur following spinal anesthesia or, more commonly, inadvertent dural puncture during attempted epidural catheter placement. The headache is usually positional (worse when upright, better when lying flat) and is often accompanied by neck stiffness, photophobia, nausea, or subjective hearing symptoms.
This topic will review PDPH. Techniques for LP, spinal, epidural, and combined spinal-epidural (CSE) anesthesia are discussed separately. (See "Lumbar puncture: Technique, contraindications, and complications in adults" and "Spinal anesthesia: Technique" and "Epidural and combined spinal-epidural anesthesia: Techniques".)
PATHOPHYSIOLOGY — The precise etiology of headache after dural puncture is unclear, but is thought to relate to leakage of cerebrospinal fluid (CSF) through the dural hole created by the needle. If CSF leaks at a rate greater than the rate of CSF production, low CSF pressure can result, accentuated at the level of the brain in the upright position. However, not all patients with PDPH have low CSF pressure, and not all patients with significant CSF leak develop a headache.
Three pathophysiologic mechanisms have been proposed:
●CSF hypotension results in compensatory meningeal venodilation and blood volume expansion, with headache caused by acute venous distention. This mechanism is consistent with magnetic resonance imaging (MRI) in several reported cases of PDPH [1,2].
●Intracranial hypotension related to CSF leak may cause sagging of intracranial structures and stretch of sensory intracranial nerves, causing pain and cranial nerve palsies. In one study of seven patients with intracranial hypotension, MRI findings of downward displacement of the brain were associated with headache, and resolved along with the headache symptoms [3]. Traction of the upper cervical nerves may cause PDPH associated neck, back and shoulder pain [4].
●Altered craniospinal elasticity after lumbar puncture (LP) results in increased caudal compliance relative to intracranial compliance and acute intracranial venodilation in the upright position [5].
INCIDENCE — The incidence of PDPH varies widely, depending on patient (eg, age, sex, pregnancy, body mass index [BMI]) and procedural (eg, needle size and type, bevel orientation for cutting needles) risk factors. The incidence of PDPH after spinal anesthesia is generally <3 percent, but may occur in up to 9 percent of cases, depending on the type and size of needle used [6-8]. PDPH after lumbar puncture (LP) occurs in approximately 11 percent of cases when a standard, traumatic needle is used [9].
PDPH related to neuraxial anesthesia is most common in obstetric patients. The reported incidence of PDPH after unintentional dural puncture (UDP) with an epidural needle (which is larger in diameter than a spinal needle) for labor analgesia is between 81 and 88 percent [10,11]. UDP occurs in <1 to as high as 6 percent of epidural placements [7,12-19], with 11 to 33 percent of UDP going unrecognized until the patient presents with PDPH [13,14,19-22].
RISK FACTORS
Patient risk factors — In various studies, common patient risk factors for PDPH have included female sex, pregnancy, age 18 to 50 years compared with older or younger ages, and a prior history of headache:
●Female sex – Several studies have found that women have a two to three times increased risk for PDPH compared with men [23-25].
●Pregnancy – Pregnancy confers an additional risk for PDPH, possibly due to increased cerebral vasodilation in response to cerebrospinal fluid (CSF) hypotension, related to high levels of circulating estrogen [24,26]. It is also thought that increased CSF pressure during labor leads to a larger CSF leak and higher risk of PDPH; however, definitive evidence is lacking [27,28]. Second stage pushing in labor [28,29] and multiparity [16] have been associated with an increased PDPH risk in the postpartum period [29,30].
●Prior headaches – History of prior headaches (both PDPH and chronic headaches) may be a risk factor for PDPH [4,23,31-33]; although, this has not been observed in all studies [34-36].
●Age – In most studies, extremes of age are associated with lower incidence of PDPH, with young adults (18 to 50 years of age) having the highest risk [23,24,31,37-39]. As an example, in one prospective study of patients who had spinal anesthesia with 25 or 27 gauge Quincke (ie, traumatic) needles, PDPH occurred in 11 percent of patients aged 31 to 50 years, compared with 4 percent of others [24]. The incidence in very young children may be under-reported due to their inability to report headache and lack of behavioral documentation by parents/caregivers and clinicians [40,41].
●Low opening pressure – Low opening pressure during lumbar puncture (LP) may also predict an increased risk of PDPH [42]. However, a 2019 case-control retrospective study and systematic literature review found no association between either opening pressure or closing pressure and PDPH risk [43].
●Volume of CSF removed – PDPH may have distinct temporal and prognostic profiles depending on the volume of CSF removed, particularly for high volume CSF removal. A systematic literature review found that high volume CSF removal (ie, 20 to 30 mL) may be associated with increased risk of immediate PDPH, but decreased risk of PDPH at 24 hours, compared with lower volume CSF removal [43]. In one retrospective study of over 300 LPs performed as part of an Alzheimer disease research trial, patients who had >30 mL CSF removed were more likely to have an immediate onset PDPH, but were less likely to need an epidural blood patch (EBP) [44]. High volume CSF removal with lowered closing pressure may help reduce further CSF leakage to heal the dural hole.
●Low body mass index (BMI) – The literature on the effect of BMI on the risk of PDPH is inconclusive. Some studies have reported a higher risk of PDPH in patients with low BMI (≤25 kg/m2) after diagnostic LP [45] and lower risk of PDPH after unintentional dural puncture (UDP) in obese parturients (BMI ≥31.5 kg/m2) [30,46]. In contrast, other studies have reported no effect of BMI on the incidence of PDPH [23,47-51].
Procedural risk factors — The choice of spinal needle and procedural factors can affect the risk of PDPH. We suggest the use of pencil point (atraumatic) needles for spinal anesthesia and diagnostic LP, in agreement with clinical practice guidelines [52,53].
●Needle tip – For both diagnostic LP and spinal anesthesia, we recommend the use of spinal needles with a pencil point tip, rather than needles with a sharp cutting tip. The use of pencil point spinal needles reduces the risk of PDPH compared with cutting needles of the same size [8,54-56].
Whitacre and Sprotte needles, which are the most commonly used pencil point needles, have a closed tip shaped like that of a pencil, with the hole on the side of the needle near the tip (figure 1). The Quincke needle, which is the most commonly used conventional, or cutting, spinal needle, has a sharp, cutting tip, with the hole at the end of the needle.
The lower incidence of PDPH with pencil point (atraumatic) needles was illustrated by a 2017 meta-analysis of randomized controlled trials that compared pencil point and conventional (cutting) needles for LP [9]. The review identified 110 trials from 29 countries with over 31,000 subjects. The incidence of PDPH was significantly lower in the atraumatic group compared with the conventional group (4.2 versus 11 percent, relative risk [RR] 0.40, 95% CI 0.34-0.47, absolute risk reduction 6.8 percent, number needed to treat to avoid one headache 15). In addition, use of atraumatic needles was associated with a significant reduction in the need for EBP (RR 0.5, 95% CI 0.33-0.75). There was no significant difference between groups for the success rate of LP on first attempt, LP failure rate, incidence of traumatic tap, or backache.
●Needle size – Larger conventional needle size (ie, lower needle gauge) is correlated with an increased incidence of PDPH, as demonstrated in a 2016 meta-analysis that included 12 studies with over 3100 patients who had neuraxial anesthesia with cutting needles [57]. As examples from the meta-analysis, the incidence of PDPH with a 22 or 23 gauge Quincke needle ranged from 8 to 25 percent, whereas the incidence of PDPH with a 27 gauge Quincke needle ranged from 0 to 14 percent.
In obstetric patients who undergo spinal anesthesia, there is little difference in the rate of PDPH or the need for EBP with the commonly used sizes of pencil point needles. The 2016 meta-analysis identified 10 studies with over 2400 patients who had neuraxial anesthesia with pencil point needles and found no correlation between the incidence of PDPH and pencil point needle size [57]. For patients who undergo diagnostic LP, the relationship between pencil point needle size and the risk of PDPH is less clear, as there are few data.
Smaller spinal needles may be technically more difficult to use, as they tend to bend during insertion. Thus, they are typically used with an introducer needle, which is inserted through the skin and along the desired needle path. Flow rate is slower than with larger bore needles; confirmatory CSF appears more slowly in the needle hub, and collection of CSF is slower. In our practice, spinal anesthesia is performed using a small gauge noncutting needle (ie, 25 gauge Whitacre). (See "Spinal anesthesia: Technique", section on 'Choice of spinal needle'.)
The optimal size of spinal needle may be different for spinal anesthesia versus diagnostic LP in which opening pressure and/or CSF sampling is required. In one laboratory study that compared flow rates and the rate of accurate pressure transduction for various cutting and pencil point spinal needles, the 20 gauge Sprotte needles provided rapid pressure transduction (>90 percent of true pressure within one minute) and an optimal flow rate, defined as 2 mL per minute. [58]. The technique of diagnostic LP is discussed in detail elsewhere. (See "Lumbar puncture: Technique, contraindications, and complications in adults".)
●Needle orientation – PDPH is more than twice as likely if a cutting spinal or epidural needle is inserted with the bevel perpendicular, rather than parallel, to the long axis of the spine [31,59-61]. As an example, in one study, 218 patients who underwent nonobstetric surgery with spinal anesthesia using a 27 gauge Quincke needle were randomly assigned to have the needle inserted parallel or transverse to the long axis of the spine [61]. PDPH occurred in 4 percent of patients in the parallel group, and 23 percent of patients in the transverse group.
Since pencil point needles have no bevels, needle orientation is not a relevant issue when they are used.
●Reinsertion of stylet – Whether reinserting the stylet before removing the spinal needle reduces the risk of PDPH is unclear. Available data suggest that reinsertion does not reduce PDPH when a cutting needle is used, and the data regarding the use of pencil point needles is equivocal. Authors to this topic do not routinely reinsert the stylet before removing a spinal needle.
•Pencil point needles – Evidence of benefit for stylet reinsertion comes from a randomized trial of 600 patients undergoing diagnostic LP using a 21 gauge Sprotte needle, which found that reinsertion of the stylet prior to removing the needle (versus no reinsertion) resulted in a decreased incidence of PDPH (absolute risk reduction 11 percent, 95% CI 6.5-16.2 percent) [62].
In contrast, a randomized trial of 952 patients undergoing diagnostic LP (using 22 gauge atraumatic, 25 gauge atraumatic, or 25 gauge cutting needles) found similar rates of PDPH in patients who had the stylet reinserted versus removal of the needle without reinsertion [63].
•Cutting needles – In one study including 630 patients who underwent spinal anesthesia with a 25 gauge Quincke needle, the PDPH rate was similar in patients randomly assigned to have the stylet reinserted before needle removal or needle removal without reinsertion of the stylet [64].
●Combined spinal-epidural – The incidences of UDP and PDPH after combined spinal-epidural (CSE) anesthesia/analgesia are similar to the incidences after an epidural technique [65-67]. CSE is discussed separately. (See "Epidural and combined spinal-epidural anesthesia: Techniques".)
●Others – Placement of spinal needle in the sitting as opposed to lateral position [68] and a higher number of needle passes may be associated with increased risk for PDPH in spinal anesthetics, but the evidence is conflicting [31,69].
Paramedian approach to the spinal space and decreased operator experience [70,71] may confer additional increased risk for PDPH, but results are limited and conflicting between the orthopedic and obstetric populations [72,73].
In epidural placements, studies have failed to show a difference in PDPH risk with the use of air versus saline for the loss of resistance technique to identify the epidural space across all patient populations [74,75]. The onset of headache may be sooner in obstetric patients when air is used for loss of resistance, likely related to pneumocephalus rather than low pressure headache [20].
PREVENTION OF PDPH AFTER DURAL PUNCTURE — In addition to procedure modification, a number of strategies have been used to attempt to prevent PDPH after dural puncture.
●Bed rest – Despite common recommendations for bed rest following dural puncture, this remedy has not been shown to significantly decrease the risk of PDPH [4,76-79]. A meta-analysis of 16 randomized controlled trials of LP performed for anesthesia, myelography, or diagnostic purposes found no evidence in any trial that longer bed rest was superior to immediate patient mobilization or shorter bed rest [78]. The relative risk (RR) of headache among patients undergoing a diagnostic LP was 0.97 with longer bed rest. A similar lack of benefit was noted in a later meta-analysis [76]. While of no use for prevention of PDPH, bed rest does decrease the intensity of the headache. (See 'Treatment' below.)
●Abdominal binder – Given the low risk of this intervention, some of the authors of this topic routinely offer abdominal binders to patients who have unintentional dural puncture (UDP) during epidural anesthesia. In their experience, some patients enjoy wearing the binder in the immediate postpartum period. However, abdominal binders are not widely recommended [80], and there is little literature on their use [81].
There is a theoretical role for abdominal compression with a tight abdominal binder for prevention and/or treatment of PDPH. The rationale for such treatment is that increased intra-abdominal pressure may be transmitted to the epidural space, may help to seal the dural puncture and decrease cerebrospinal fluid (CSF) leak, and may improve an existing headache.
●Prophylactic drug therapy – Based on the limited evidence available, we do not recommend administration of prophylactic medications after UDP or LP to prevent PDPH. Small trials have reported that prophylactic administration of epidural morphine [82] and intravenous (IV) cosyntropin [83] may reduce the incidence of PDPH after UDP during obstetric anesthesia [84]. In a case series of 31 obstetric patients with UDP, triple prophylactic treatment with epidural saline, intravenous cosyntropin and epidural morphine was associated with decreased incidence of PDPH and need for epidural blood patch, compared with other prophylactic treatment or conservative management [85]. In addition, ondansetron may reduce the risk of PDPH. In one randomized trial, 210 women were randomly assigned to receive ondansetron 0.15 mg/kg IV or saline during spinal anesthesia with a 25 gauge Quincke needle for cesarean delivery [86]. PDPH occurred in 5 percent of patients who received ondansetron, compared with 21 percent of patients who did not. Ondansetron may trigger migraine headache in susceptible patients [87]. Further research is needed to elucidate the role of these drugs in PDPH prevention.
In obstetric and general surgical populations, oral caffeine has not been shown to prevent PDPH after dural puncture [88,89].
●Prophylactic epidural blood patch – Epidural blood patch (EBP) is an effective treatment for PDPH, and may also be performed prophylactically before a headache occurs for patients in whom an epidural catheter is placed after an inadvertent dural puncture. For prophylactic EBP, blood is injected into the epidural catheter prior to its removal after anesthesia. (See 'Epidural blood patch' below.)
We do not routinely perform prophylactic EBP, since the efficacy is unclear [90-95]. A review of the literature of prophylactic EBP in obstetric patients found that it does not appear to decrease the incidence of PDPH, but may decrease the intensity and/or duration of symptoms [94].
●Intrathecal catheter placement – When UDP occurs during an attempt at epidural catheter placement, some experts have advocated threading an epidural catheter into the intrathecal space at the time of UDP, in an attempt to reduce the incidence of PDPH. However, the efficacy of intrathecal catheter placement has not been established in randomized controlled trials, and most studies have reported no benefit of intrathecal catheter placement. Furthermore, intrathecal catheters after UDP may have a higher rate of failed labor analgesia compared with re-sited epidural catheters [96]. For these reasons, we do not routinely place an intrathecal catheter after UDP, but we do place intrathecal catheters selectively (eg, after a difficult epidural procedure). Spinal catheters may not prevent PDPH, but they reduce the need for multiple epidural attempts and the risk of subsequent, additional UDP.
For patients in whom we place a spinal catheter, we remove the catheter as soon as possible after delivery or surgery to reduce the risk of accidental injection of unintended medication into the intrathecal space and to reduce the risk of infection (ie, meningitis). There is no strong data in support of prolonged catheterization with the goal of reducing headache [97].
Prophylactic intrathecal catheters have been studied most extensively in obstetric anesthesia patients. Examples of those studies are as follows:
•In a 2013 meta-analysis of nonrandomized studies, intrathecal catheter placement after UDP for obstetric anesthesia did not change the incidence of PDPH compared with no catheter, but the need for EBP was reduced (RR 0.64, 95% CI 0.49-0.84) [97].
•In contrast, two retrospective studies completed after that meta-analysis found that intrathecal catheter placement after UDP was associated with both a reduced incidence of headache and a reduced need for blood patch, although the latter did not reach statistical significance [98,99].
•A 2016 retrospective review of 235 parturients with UDP compared outcomes for those who were managed with either placement of an intrathecal catheter or replacement of an epidural catheter [96]. There was no difference in the incidence of PDPH between the two groups.
CLINICAL FEATURES — Patients with PDPH typically present with frontal or occipital headache that is worse with sitting or standing, and relieved by lying flat. The headache tends to be worse if it occurs in the first 24 hours, and associated symptoms are more common with severe headaches. In some patients, the symptoms of PDPH are similar to their previously experienced migraine headaches, except that there is an added postural component.
Associated symptoms occur in up to 70 percent of patients [24], and may include nausea, neck stiffness, low back pain, vertigo, vision changes (diplopia, blurred vision, or photophobia), dizziness, or auditory disturbances (tinnitus, hearing loss) [42]. In one study of 133 patients who developed PDPH after diagnostic lumbar puncture (LP), neck stiffness was reported in 55.6 percent, shoulder stiffness in 46.6 percent, nausea and vomiting in 33 percent, tinnitus in 22 percent, and photophobia in 23 percent [4].
Approximately 90 percent of PDPHs occur within 72 hours after a dural puncture, though onset has rarely been reported up to two weeks later [42,100]. Without treatment, most headaches resolve within one week, and one-half resolve by four to five days after dural puncture [24,31,101]. However, some retrospective studies and small prospective studies suggest that unintended dural puncture is associated with an increased risk of longer-term or persistent headaches [102-107]. Longer duration of PDPH has been associated with the use of a cutting needle for dural puncture compared with a pencil point needle [108], dural puncture with the patient in the sitting position compared with lateral decubitus, multiple attempts at dural puncture, and patient history of depression [36].
DIAGNOSIS OF PDPH — The diagnosis of PDPH is made clinically by identifying the typical positional headache within 72 hours after a dural puncture. Other causes may need to be excluded if symptoms are atypical.
Neuroimaging with computed tomography (CT) or magnetic resonance imaging (MRI) is not indicated unless required to exclude alternative diagnoses. If imaging is performed, findings consistent with PDPH include small ventricles, downward displacement (sagging) of the brain, engorged cerebral venous sinuses, subdural fluid collections, pituitary enlargement, and diffuse meningeal enhancement [2,109,110]. These findings are analogous to those reported in patients with spontaneous intracranial hypotension. (See "Spontaneous intracranial hypotension: Pathophysiology, clinical features, and diagnosis", section on 'Initial diagnostic neuroimaging'.)
Diagnostic lumbar puncture (LP) should be avoided if possible, due to the risk of worsening an existing PDPH. If performed, however, a low cerebrospinal fluid (CSF) opening pressure or dry tap (indicative of intracranial hypotension) with increased CSF protein and lymphocyte count may be present [111].
DIFFERENTIAL DIAGNOSIS — The differential diagnosis of headache after dural puncture is broad, as outlined in the table (table 1), but a postural headache in this setting is most likely related to the neuraxial procedure. Serious and/or life-threatening etiologies (eg, hemorrhage, thrombosis, vasculopathy, meningitis), which may or may not be related to the dural puncture, must be ruled out in the presence of focal or worsening neurologic deficits. Findings that should prompt further investigation are shown in a table (table 2).
Subdural hematoma and other forms of intracranial hemorrhage are rare but potentially lethal complications of dural puncture. Intracranial hemorrhage should be suspected in patients with prolonged headache (ie, beyond five to seven days), whose headaches become non-positional, who do not improve or worsen after an epidural blood patch (EBP), or who develop focal neurologic symptoms [112,113].
Other possible causes of postural headache (eg, migraine, headache associated with postural orthostatic tachycardia) may occur coincidentally after dural puncture, and may be excluded based on history, symptoms, and physical examination. Some headaches may improve but are not completely abolished with rest, including in the supine position (eg, migraine).
PDPH related to neuraxial anesthesia is most common in obstetric patients, compared with non-obstetric patients. Headache in the postpartum period is remarkably common, with a reported incidence as high as 39 percent in the first week [114]. The majority of these are primary headache (tension or migraine type headache), with other common causes being preeclampsia/eclampsia, lactation headache, and PDPH [115]. Rare causes include reversible cerebral vasoconstriction syndrome (RCVS) and posterior reversible encephalopathy syndrome (PRES). The evaluation of postpartum headache is reviewed separately. (See "Headache during pregnancy and postpartum", section on 'Evaluation of postpartum patients'.)
TREATMENT — The treatment of PDPH depends upon the severity of headache and its impact on the patient's ability to function.
Mild PDPH — Patients with PDPH who can tolerate an upright position and perform activities of daily living are considered to have mild PDPH; obstetric patients with mild PDPH are able to care for their baby. Patients with mild PDPH may benefit from an initial trial of conservative treatment including bed rest as needed, and a brief course of oral analgesics (see 'Drug therapy' below) and antiemetics as necessary. Vigorous oral and/or intravenous (IV) hydration are also routinely encouraged in postpartum patients as a noninvasive, low-risk therapy [116].
Debilitating PDPH — Patients with PDPH who are unable to tolerate sitting or standing, or are unable to perform activities of daily living (including obstetric patients unable to care for their baby) and whose headache is refractory to a brief trial of conservative measures are considered to have moderate to severe PDPH. These patients should be offered an epidural blood patch (EBP), which may provide permanent symptomatic relief.
Epidural blood patch — EBP is considered the definitive treatment for PDPH. A systematic review published in 2010 noted that EBP reduced the duration and intensity of post-dural puncture headache compared with both conservative treatment and sham procedure [117] based upon the results of three randomized controlled trials with a total of 86 subjects [118-120]. EBP usually provides immediate relief. The success rate after the first EBP is between 65 and 98 percent, with a similar success rate for a second EBP, if it is required [10,118,121-124]. The success rate for EBP may be lower if dural puncture occurs with a larger diameter needle (ie, ≥20 gauge) [125].
Contraindications — Similar to epidural anesthesia, EBP is contraindicated in patients with coagulopathy or on anticoagulation, and in patients with systemic infection or infection at the site for epidural needle placement. Patients with HIV present a theoretical risk for developing central nervous system (CNS) infection after EBP. However, HIV is known to be a neurotropic virus, infecting the CNS in its earliest stages, and no adverse effects have been reported after either neuraxial anesthesia or EBP in HIV-infected patients [126].
Patients with PDPH after an initially challenging neuraxial anesthetic may benefit from fluoroscopically-guided EBP placement. The decision to place EBP must always be weighed against the risks of repeated dural trauma or failed EBP due to technical difficulty. In general, we recommend EBP be performed only by experienced providers.
EBP technique — Epidural blood patch (EBP) is performed by injecting the patient's blood through an epidural needle into the epidural space.
Standard monitors are placed and a free-flowing IV line is secured in case emergency medications need to be administered. Two sterile fields are created for two operators: one for the blood draw and the other for the epidural procedure. The epidural procedure is initiated first, unless difficult blood draw is anticipated. An epidural needle is inserted as it would be for epidural anesthesia, using loss of resistance to saline to identify the epidural space, without inserting a catheter (see "Epidural and combined spinal-epidural anesthesia: Techniques", section on 'Epidural anesthesia technique'). Once the epidural space has been identified in this manner, a syringe of autologous venous blood is drawn by a second operator using aseptic technique; the blood is slowly injected through the epidural needle. Typically, headache symptoms will improve within seconds to minutes of the completion of the procedure, although a transient sensation of "fullness" in the back is common.
Upon completion of the EBP, the patient is instructed to lie flat (or at a maximum 30 degree angle) for one to two hours with minimal movement. After this, the patient may stand and resume normal activities. It is reasonable to avoid heavy lifting and strenuous activity within the first 24 hours of EBP, although evidence is lacking to support this recommendation. Patients should also be instructed to taper analgesic medications (eg, butalbital-acetaminophen-caffeine) to avoid the possibility of a rebound headache from sudden medication withdrawal.
●Optimal volume of blood for EBP – We aim to inject 20 mL of blood for blood patch, and stop injection if the patient complains of significant pain or pressure. A randomized trial that compared injection of 15, 20, or 30 mL of epidural blood for EBP in 120 obstetric patients found that the optimal volume of blood for complete relief of headache was 20 mL [122]. Complete relief of headache was greater in patients who received 20 mL than 15 mL, but there was no difference between the 20 mL and 30 mL group. Similarly, a retrospective review of 466 EBPs in obstetric patients reported that the average volume of injected blood was 20.5 mL (with the volume injected determined by pain on injection), and increasing the volume to 30 mL did not reduce the need for repeat EBP [127].
●Timing of EBP – Optimal timing for EBP placement has not been determined. Several retrospective studies have reported increased effectiveness if EBP is performed more than 24 to 48 hours after dural puncture [125,127,128]. However, the timing of EBP was not controlled, and early EBP may have been a marker for a more severe, difficult to treat PDPH. We perform EBP when indicated by patient symptoms and request, and do not delay the procedure specifically to increase efficacy.
●Spinal level for EBP – We generally perform EBP at the same spinal interspace as the initial dural puncture, if it is known. If there were needle insertions at multiple levels, we insert the epidural needle at the lowest interspace that was entered, because the injected blood usually spreads more cephalad than caudad [129-131]. In one study, magnetic resonance imaging (MRI) was performed approximately 45 minutes after EBP with 20 mL of blood, performed with the patient in the lateral decubitus position [131]. The blood spread a mean of 4.6 interspaces, approximately 3.5 interspaces cephalad and 1 interspace caudad to the injection.
Mechanism of action — The mechanism by which EBP relieves PDPH is unclear, and may be multifactorial. It is thought that the injection of blood directly compresses the thecal sac, thereby increasing lumbar and intracranial cerebrospinal fluid (CSF) pressure. Once the injected blood clots, it may plug the CSF leak and/or initiate an inflammatory reaction that facilitates healing of the puncture site. Regardless of the mechanism, clotting appears to be important to the success of injectates used for EBP, as whole blood, fibrin glue, and possibly platelet-rich plasma more effectively maintain CSF pressure than saline, anticoagulated blood, or dextran solutions in animal models [91,132,133].
Complications — The most common complication of EBP is back pain, which occurs in 25 to 35 percent of patients [121,134,135]. Back pain usually resolves within 48 hours of EBP. However, in one prospective observational study of women who underwent EBP for PDPH, back pain lasted from 3 to 100 days (mean 28 days) in 16 percent of patients [121]. Rare complications include misplacement of blood resulting in spinal subdural hematoma [136] or intrathecal injection and arachnoiditis [134,137]. Other rare complications include infection, subdural abscess [138], facial nerve paralysis [121,139-141], spastic paraparesis, and cauda equina syndrome [142,143].
Drug therapy — A number of drugs have been investigated for the treatment of PDPH in small trials and studies, but none have been proven beneficial for this specific indication. Nevertheless, oral caffeine is a low-risk option for most patients, and caffeine and oral analgesics are options for the symptomatic treatment of PDPH.
●Caffeine – Caffeine has commonly been used for treatment of PDPH (sometimes in combination with butalbital and/or acetaminophen), without high-quality supporting evidence [88]. We encourage self-administered oral caffeine in patients who normally drink caffeinated beverages on a daily basis, in order to avoid headache and other symptoms of caffeine withdrawal. A 2015 systematic review of the literature on drug therapy for PDPH found two small randomized trials (approximately 40 patients in each) that compared oral and IV caffeine with placebo for PDPH after neuraxial anesthesia [144]. The trials had methodologic flaws and meta-analysis was not performed because of heterogeneity. One of these trials reported lower pain scores at four hours in patients who received a single dose of 300 mg of caffeine orally, with no difference in pain scores at 24 hours or the need for EBP [145]. In the other trial, PDPH was relieved in 75 percent of patients who received IV caffeine 500 mg IV, but 24 hours after treatment there was no difference in pain scores between patients who received caffeine and placebo, and no difference in the number of patients who required EBP [146].
Neither of these trials reported significant complications of caffeine administration. However, there are case reports of grand mal seizures after IV administration of caffeine for treatment of PDPH [147,148].
●Simple analgesics – Some experts prefer oral analgesics for symptomatic treatment of PDPH, typically with acetaminophen or a nonsteroidal antiinflammatory drug (NSAID). However, there is a lack of high-quality evidence to support these interventions, specifically for PDPH.
●Combination analgesics – Some authors of this topic use the combination butalbital-acetaminophen-caffeine for postpartum patients presenting with PDPH, while other authors use the combination acetaminophen-caffeine and avoid barbiturates and opioids. As with other drug treatments for PDPH, supportive evidence is lacking.
●Other drugs – Limited evidence suggests that gabapentin, hydrocortisone, theophylline, and neostigmine/atropine are modestly effective for decreasing the severity of PDPH [144,149]. Sumatriptan [150] and adrenocorticotrophic hormone [151,152] have not been shown to be beneficial for treatment of PDPH [144]. We do not recommend the routine use of these medications in the absence of high-quality supportive evidence.
Alternative treatments for PDPH — Alternatives to EBP include bilateral transnasal sphenopalatine block and greater occipital nerve block, as discussed below, and some experts recommend mindfulness meditation for patients with headache and other painful conditions. However, data are limited, and prospective randomized trials are needed to fully evaluate the efficacy of these interventions.
Transnasal sphenopalatine block — Bilateral transnasal sphenopalatine ganglion block (SPGB), using topical intranasal local anesthetic, is emerging as a promising therapy for the PDPH. SPGB blocks sympathetic, parasympathetic, and somatic sensory nerves, and may treat PDPH via multiple mechanisms [153,154]. The topical SPGB is easily performed, noninvasive, low risk, and has been used to treat PDPH in post-anesthesia and emergency department settings [153,155-158].
For the transnasal SPGB, the patient is positioned supine or semi-sitting, and a cotton tip applicator saturated with local anesthetic (typically 2 to 5% lidocaine) is inserted through each nostril until it contacts the posterior pharynx. The applicators are left in place for at least 10 minutes before removal. Most patients receive almost immediate relief of headache. In our experience, the pain relief from SPGB with 4% lidocaine typically lasts only four to six hours, limiting its widespread utility as a first-line treatment for PDPH. However, in some reported cases, transnasal SPGB has relieved headache for up to 18 hours or more and may benefit select patients in whom EBP is relatively contraindicated [157].
Among the small numbers of cases reported, the majority of patients reported at least temporary relief of headache, and many avoided the need for EBP. In a retrospective single center review of treatment of PDPH in 81 postpartum patients, approximately one-half of whom received EBP and one-half SPGB, headache was relieved more quickly after SPGB [159]. Twenty three percent of the patients in the EBP group returned to the emergency department for further treatment of the headache, whereas no patient in the SPGB group returned. All patients were headache-free at one week. A prospective observational study of 20 patients with PDPH comparing SPGB with traditional oral or parenteral analgesics suggested that SPGB could be used as an initial treatment for PDPH given its rapid onset of effect within 30 minutes [160]. However, pain scores subsequently increased and were similar to those in the group receiving traditional analgesics after eight hours. Similarly, in the clinical experience of one of the authors of this topic, the pain relief after SPGB is often temporary. Of note, a randomized trial investigating the effect of transnasal SPGB with local anesthetic versus saline found no significant difference in pain intensity between the two groups after 30 minutes [161]. Pain scores were reduced and EBP was avoided in half the patients in both groups, suggesting that the effect or mechanism of the SPGB may not relate primarily to local anesthetics.
Greater occipital nerve block — Greater occipital nerve block has also been reported in the treatment of PDPH with positive results. The mechanism is via interruption of pain transmission to the trigeminal and greater occipital nerves, which are activated by dural stretch in PDPH [162,163].
OTHER COMPLICATIONS OF DURAL PUNCTURE — PDPH is the most common adverse outcome of dural puncture. Other complications are less common, longer lasting, and potentially serious. These include:
●Infection (eg, meningitis)
●Bleeding, including spinal epidural hematoma
●Intracranial hypotension syndromes (eg, cranial neuropathies, subdural hematoma, cerebral venous thrombosis)
●Cerebral herniation
●Cerebral vasoconstriction and encephalopathy syndromes
●Pneumocephalus
●Radicular or back pain
●Late onset of epidermoid tumors of the thecal sac
These complications of dural puncture are reviewed in detail separately. (See "Lumbar puncture: Technique, contraindications, and complications in adults", section on 'Complications' and "Overview of neuraxial anesthesia", section on 'Adverse effects and complications'.)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Post dural puncture headache".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Spinal headache (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Definition and incidence – Post dural puncture headache (PDPH), also known as post lumbar puncture (LP) headache may occur after diagnostic or therapeutic LP, spinal anesthesia, or unintentional dural puncture (UDP) during epidural anesthesia. PDPH is caused by leakage of cerebrospinal fluid through the dural hole created by the needle. (See 'Introduction' above.)
The incidence of PDPH varies significantly depending on the size and type of needle used and patient risk factors. PDPH occurs in up to approximately 10 percent of patients after dural puncture with a spinal needle, and up to approximately 85 percent of patients who have an UDP with an epidural needle. (See 'Incidence' above.)
●Risk factors – Risk factors for PDPH include patient age (highest among adults 18 to 50 years of age), female sex, pregnancy, possibly prior history of headache, the use of cutting versus pencil point spinal needles, and the use of larger needles. For patients undergoing LP or spinal anesthesia, we recommend performing the procedure with spinal needles that have a pencil point (atraumatic) tip, rather than needles with a sharp cutting tip (Grade 1A). (See 'Risk factors' above.)
●Prevention – Measures to prevent PDPH after dural puncture (eg, bed rest, placement of a spinal catheter after UDP, medications) have not been proven effective. (See 'Prevention of PDPH after dural puncture' above.)
●Clinical features – PDPH is a postural headache (ie, worse when upright, improved when supine) that usually occurs within 72 hours of dural puncture. In most cases, PDPH resolves within one week, even without treatment. Other causes of headache must be ruled out (table 1). (See 'Clinical Features' above and 'Differential diagnosis' above.)
●Diagnosis – The diagnosis of PDPH is made clinically by identifying the typical positional headache within 72 hours after a dural puncture. In patients with atypical features, other causes may need to be excluded (table 1 and table 2). (See 'Diagnosis of PDPH' above and 'Differential diagnosis' above.)
●Treatment – Conservative treatment for patients with mild PDPH includes bed rest and a brief course of oral caffeine and/or oral analgesics. (See 'Mild PDPH' above and 'Drug therapy' above.)
For patients with moderate to severe PDPH that is prolonged (>24 hours) and refractory to conservative measures, we recommend treatment with epidural blood patch (EBP) (Grade 1B). EBP is performed by injecting autologous blood through an epidural needle immediately after drawing the blood under sterile conditions. (See 'Debilitating PDPH' above and 'Epidural blood patch' above.)
Do you want to add Medilib to your home screen?