A 26-year-old woman presents to the emergency department with complaints of intermittent diplopia, pulsatile tinnitus and a dull headache that has waxed and waned for 1 month. She reports no constitutional symptoms and has no other complaints.
A thorough history and physical are completed, but the cause of the young woman's symptoms is not apparent. Vital signs and blood work, including a complete blood count, complete metabolic panel, erythrocyte sedimentation rate and C-reactive protein, are within normal limits. Magnetic resonance imaging and computed tomography of the brain and skull with and without contrast show no evidence of acute changes including mass hemorrhage or midline shift.
Post-Lumbar Puncture Headache
At this point, the patient is considered a candidate for diagnostic lumbar puncture (LP). Pseudotumor cerebri is suspected, but the differential diagnoses include aseptic meningitis, migraine headache and space-occupying lesions. You are aware of post-lumbar puncture headache syndrome (PLPH) and wish to take measures to prevent this event.
It has been hypothesized that PLPH occurs as a result of cerebrospinal fluid (CSF) leakage due to delayed closure of the dural fibers after LP.1-4 PLPH presents as bilateral headache that is occasionally accompanied by tinnitus, neck stiffness, hypacusia, photophobia and nausea. It typically occurs within 5 days of the procedure. Because the headache results from a decrease in CSF pressure, it is aggravated by ambulation and relieved by recumbence. Symptoms generally resolve spontaneously within 1 week. Blood patching, a procedure in which the patient's blood is injected into the epidural space to form a clot that occludes the dural puncture site, usually leads to resolution of symptoms within 48 hours.1-3
The American Academy of Neurology (AAN) hypothesizes that a decrease in the incidence and intensity of PLPH may be related to both the direction of the needle bevel in relation to dural fibers and replacement of the stylet prior to needle withdrawal.1,2 Additionally, the AAN notes that needle type (traumatic vs. atraumatic) may contribute to the likelihood of developing PLPH.1,2 Although distinct AAN practice guidelines about the use of a small-gauge atraumatic needle in spinal anesthesia exist, practice guidelines for diagnostic LP have been predominantly correlated from anesthesiology-based studies.2,5
Two spinal needle types are currently available: the traditional cutting spinal needle (Quincke) and the atraumatic (Whitacre or Sprotte) needle. Optimally, the needle type least likely to result in development of PLPH should be used to perform the procedure.
This article explores the clinical question: Do atraumatic spinal needles decrease the incidence and intensity of PLPH in lumbar puncture in comparison to traditional traumatic cutting spinal needles?
A PubMed search of peer-reviewed literature utilizing the clinical query systematic reviews generator with a narrow scope and the terms "post-lumbar puncture headache" and "needle" yielded several results. The search was limited to systematic reviews, meta-analyses and randomized, controlled trials conducted between 2000 and 2012. Eighteen articles fit these criteria. A secondary search of the same terms in the Cochrane Library and ISI Web of Knowledge produced no additional related articles.
A tertiary search of bibliographies returned one randomized, controlled trial meeting inclusion criteria. Of the 19 articles, those that focused on epidural procedures, spinal anesthesia and PLPH treatments were excluded, as were studies that did not compare traumatic vs. atraumatic needles. Of the remaining articles, two randomized, controlled trials met all inclusion criteria and were of the highest levels of evidence available.
No systematic review or meta-analysis examining the effect of diagnostic LP needle type on incidence and intensity of PLPH has been published. The authors noted no conflicts of interest.
The first study, titled "'Atraumatic' Sprotte needle reduces the incidence of post-lumbar puncture headaches," was conducted by Strupp et al. This randomized, double-blind controlled trial took place between November 2000 and March 2001. Patients reporting recent headache, abnormal CT/MRI and/or recent LP (within preceding 4 weeks) were excluded, resulting in a total study sample size of 230 patients (62% women).
Patients were randomized to the 22-gauge (g) Quincke (traumatic; T) (n = 115) or the 22-gauge Sprotte (atraumatic; AT) (n = 115) groups. PLPH was operationally defined as bilateral headache occurring within 7 days of LP, lasting no longer than 14 days, and demonstrating both aggravation with ambulation and relief with recumbence. Adjustments were made for age, caffeine intake, previous LP or previous headache.
PLPH intensity was reported independently by each participant and graded via a 10-point visual analogue scale (0 = no pain, 10 = severe pain). Multiple neurologists, blinded to needle type, performed all procedures. Regardless of the neurologist used, all patients underwent LP in the seated position. Inducers were used in all patients, with depth determined by body mass index (BMI). Each neurologist removed 10 mL of CSF from each patient.
The investigators noted no statistically significant difference (p < 0.05) between the demographics of the two groups with respect to age, gender, BMI, previous headache and use of caffeine before or after procedure. They concluded that, in comparison to traditional cutting needles, use of atraumatic spinal needles for diagnostic lumbar puncture resulted in a 50% reduction in the incidence of moderate to severely intense PLPH (24.4% vs. 12.2%, p < 0.05). No statistically significant difference in the intensity of symptoms was noted (6.4 ± 2.3 T vs. 7.5 ± 2.2 AT, p > 0.05). In addition, needle type did not affect time to onset of symptoms following LP (1.0 ± 0.9 days T vs. 1.3 ± 1.5 days AT).1
The other study that met our search criteria, titled "Standard vs. atraumatic Whitacre needle for diagnostic lumbar puncture," was conducted by Lavi et al. This prospective, randomized, controlled trial enrolled 58 adults (57% women) older than 18. Exclusion criteria included low platelet count (< 80 x 109/L), previous abnormal CT/MRI and recent LP (within preceding 7 days).
Patients were randomized to the 22-gauge Quincke (T)(n = 29) or 22-gauge Whitacre (AT) (n = 29) groups. The study was conducted over 6 months (July through December 2004). PLPH was defined as positional headache, and it was assessed by an independent physician (blinded) via questionnaire at 2 and 7 days post-LP. All patients underwent LP via sterile technique, performed by a single anesthesiologist who was unblinded to needle type. All patients were placed in the lateral recumbent position during the procedure. In all patients, the needle bevel was oriented parallel to dural fibers and the stylet replaced prior to needle withdrawal.
The volume of CSF removed varied between patients in the traumatic and atraumatic groups, but this variation was not statistically significant (11 ± 0.8 mL vs. 9.2 ± 0.8 mL, p = 0.09). Lavi et al reported no statistically significant differences in demographic variables such as age, gender, BMI and caffeine consumption between the two groups. The authors concluded that the incidence of PLPH was significantly greater in the Quincke group than in the Whitacre group (36% vs. 3%; p = 0.002). Logistic regression analyses indicated that needle type was the only recorded variable associated with PLPH. In addition, analysis indicated that needle type was the most significant risk factor for PLPH (HR = 24.3, p = 0.004).3
Conclusion and Bias
The evidence presented suggests that the use of an atraumatic spinal needle, 22-gauge Sprotte or 22-gauge Whitacre, results in statistically significant reduction in PLPH incidence and intensity following diagnostic LP in comparison to traditional traumatic 22-gauge Quincke needles.
Although the reviewed studies utilized different patient positions (sitting vs. lateral recumbent) during the procedure, no difference in overall incidence of PLPH was found between the two studies (see table). Other studies have suggested that placing patients in the seated vs. the lateral recumbent position may result in higher incidence of PLPH.2,4 The effect of patient positioning on the incidence of PLPH is currently being studied because research to date has been inconclusive.
Several biases exist within these two studies. Neither study had a large patient population (230 and 58), nor did they report the calculation of power analyses for determining appropriate sample sizes. Neither study was conducted for longer than 6 months, presumably limiting the sample population. Nor did either study report results with 95% confidence intervals. In addition, Strupp et al reported that LPs were performed (via the same technique) by multiple trained neurologists, whereas Lavi et al utilized one anesthesiologist who performed all procedures. It can be inferred that use of more than one clinician might allow for variability in outcomes due to variation in technique.
Strupp et al made no mention of bevel orientation with respect to dural fibers. This is in contrast to Lavi et al, who reported bevel orientation parallel to dural fibers throughout all procedures. Needle orientation may be a confounding variable, since it has been theorized that insertion of the needle bevel parallel to the dural fibers (vs. perpendicular to the fibers) may reduce the incidence of PLPH.4,5
Both studies utilized stylet replacement prior to withdrawal of the needle, which may lead to decreased incidence of PLPH.2,5,6 Diagnostic lumbar puncture is a common procedure used in anesthesiology, neurology, interventional radiology and emergency medicine. It is clear that additional research using larger sample sizes, valid methodology and consistent procedures and outcome measures are warranted.
Clinical Bottom Line
The literature reviewed is of the highest level of evidence and supports the use of a 22-gauge atraumatic spinal needle (Sprotte or Whitacre) to reduce the incidence of PLPH among adults undergoing diagnostic LP. This conclusion also corresponds with current practice guidelines published by the American Academy of Neurology.7
Although there is a vast, well-documented and accepted body of knowledge about the use of atraumatic needles to reduce PLPH associated with the delivery of spinal anesthesia, the instrumentation requirements for diagnostic lumbar puncture are different.2 Specifically, diagnostic lumbar puncture requires a larger needle diameter.1,2,6 Therefore, the need for further research specifically evaluating needle type, gauge, bevel orientation and patient positioning is imperative to advance the current body of knowledge and ultimately reduce the incidence of PLPH.
Chad McCollum is a physician assistant at Southwest Diagnostic Imaging in Phoenix, where he specializes in interventional radiology. Danielle L. Kempton is a physician assistant who is the assistant director of clinical education for the PA program at Midwestern University in Glendale, Ariz. She is also an assistant professor in the program. The authors have completed disclosure statements and report no relationships related to this article.
1. Strupp M, et al. "Atraumatic" Sprotte needle reduces the incidence of post-lumbar puncture headaches. Neurology. 2001;26;57(12):2310-2312.
2. Evans RW, et al. Assessment: prevention of post-lumbar puncture headaches: report of the therapeutics and technology assessment subcommittee of the American Academy of Neurology. Neurology. 2000;10;55(7):909-914.
3. Lavi R, et al. Standard vs atraumatic Whitacre needle for diagnostic lumbar puncture: a randomized trial. Neurology. 2006;24;67(8):1492-1494. Erratum in: Neurology. 2007;68(9):711. Yernitzky D [corrected to Yarnitsky D].
4. Majd SA, et al. Evaluation of pre lumbar puncture position on post lumbar puncture headache. J Res Med Sci. 2011;16(3):282-286.
5. Deibel M, et al. Best evidence topic report: reinsertion of the stylet before needle removal in diagnostic lumbar puncture. Emerg Med J. 2005;22(1):46.
6. Richman JM, et al. Bevel direction and postdural puncture headache: a meta-analysis. Neurologist. 2006;12(4):224-228.
7. Evans RW, Armon C. Addendum to assessment: prevention of post-lumbar puncture headaches: report of the therapeutics and technology assessment subcommittee of the American Academy of Neurology. Neurology. 2005;65:510-512.