I have a liquor leak

Guideline day: Diagnosis and therapy of post-puncture and spontaneous CSF negative pressure syndrome

What's new?

Post-puncture CSF negative pressure syndrome

  • According to a recent Cochrane analysis, there is “moderate-quality evidence” that atraumatic needles reduce the risk of post-puncture syndrome without increasing the risk of adverse effects (back pain or paresthesia); this assessment is supported by a current meta-analysis.
  • Cochrane analyzes show the effectiveness of caffeine, gabapentin and theophylline in the treatment of post-puncture syndrome.
  • The effectiveness of the epidural blood patch for the treatment of post-puncture syndrome was also supported by a Cochrane analysis; the prophylactic blood patch is not recommended.

Spontaneous intracranial hypotension (SIH)

  • Spinal liquor fistulas are causal and can be specifically divided into 1. ventral dura tears, usually caused by microspurs at the level of the intervertebral discs, 2. meningeal diverticula and weak points in the area of ​​the nerve roots and 3. direct fistulas between the liquor and epidural veins.
  • In therapy-resistant cases, microsurgical closure of the dural corner can achieve a high healing rate (> 90%).

The most important recommendations at a glance

  • Atraumatic needles reduce the likelihood of post-puncture headaches.
  • Needles with a smaller diameter are less likely to cause post-puncture headaches.
  • The incidence of post-puncture syndrome is lower if the bevel of the puncture needle is rotated 90 ° and the stylet is reinserted before removing the puncture needle.
  • A diffuse pachymeningeal gadolinium accumulation in magnetic resonance imaging is almost indicative of a CSF negative pressure syndrome.
  • A very reliable method for detecting spinal CSF leaks is CT or, alternatively, MRI myelography.
  • Radioisotope cisternography with intrathecally applied indium 111 is suitable for diagnosing a CSF leak.
  • Both intravenous or oral administration of caffeine and oral administration of gabapentin and theophylline are effective for symptomatic treatment.
  • The therapy of choice is an epidural blood patch after the above-mentioned conservative measures have failed, which may be repeated several times.
  • Microsurgical closure of spinal CSF fistulas is indicated in treatment-refractory cases and in the exact localization of the spinal fistula.

Introduction: Scope and purpose of the guideline (German-speaking countries Germany, Austria, Switzerland)

Justification of the need for the guideline

A CSF negative pressure syndrome is predestined for a guideline due to the various causes (postpuncture, spontaneous or fistula) with good therapeutic options. In particular, a CSF negative pressure syndrome after diagnostic lumbar puncture or spinal anesthesia is common in everyday clinical practice.

Objectives of the guideline

The aim of this evidence-based guideline is to optimize the prevention and therapy of the CSF negative pressure syndrome on the basis of the most precise diagnostic criteria of the IHS as well as possible methods of diagnosis including therapy options depending on the genesis of the CSF negative pressure syndrome. The guideline is evidence-based and a further development of the guideline of the DGN (Diener and Commission guidelines of the German Society for Neurology, 2008).

Patient target group

Patients with position-dependent headache

Supply area

Outpatient and inpatient diagnostics such as therapy


Neurologists, neurosurgeons, neuroradiologists


CSF negative pressure syndrome, postpuncture, CSF fistula, CSF puncture, blood patch, spontaneous intracranial hypotension

Definition and classification

Definition of terms

In the case of negative liquor pressure syndromes (LUS), a distinction must be made between negative liquor pressure associated headaches after diagnostic CSF puncture (postpuncture syndrome) or as a complication of spinal anesthesia and idiopathic spontaneous CSF leaks as well as spinal CSF fistulas.

The “Post-dural puncture headache” (ICHD-3 2013: 7.2.1) or the “Post-puncture syndrome” (PPS) is classified according to the ICHD-3 criteria from 2013 (ICHD-3) as a “headache that occurs within five Days after a lumbar puncture (LP) occurs and is caused by the leakage of CSF due to a LP "defined. This position-dependent / orthostatic headache is usually associated with stiff neck and / or subjective hearing impairment and improves spontaneously within two weeks. The diagnostic criteria are:

  1. Any type of headache that meets criterion C.
  2. A dural puncture has been made.
  3. Headache develops within 5 days after the dural puncture.
  4. Not better explained by another ICHD-3 diagnosis.

Another form of position-dependent / orthostatic headache is the "spontaneous CSF negative pressure syndrome" or "spontaneous intracranial hypotension" (SIH). In the ICHD-3 criteria (ICHD-3 2013), the “headache attributed to spontaneous intracranial hypotension” is characterized as follows: headache caused by spontaneous intracranial negative pressure. This is usually associated with stiff neck and / or subjective hearing impairment. It regresses after the intracranial pressure has normalized. This diagnosis cannot be made in a patient who received LP within the previous month.

The causes of the spontaneous CSF negative pressure syndrome are assumed to be spontaneous or traumatic dural defects (root pocket ruptures, especially thoracic) (overview in Mokri, 2015). According to recent studies, spontaneous intracranial hypotension is a spinal disease with cerebrospinal fluid leaks from the dura. These spinal liquor fistulas can be divided into three types: 1. Detectable ventral (or very rarely a posterolateral) dural corner, mostly caused by microspurs or calcifications (depending on the series in 27–70% of cases).
2. Meningeal diverticula / ectasias / dural weak points in the area of ​​the spinal nerve roots (20–42%). 3. Direct fistulas between the CSF space and the epidural vein (<1–3%) (Beck et al., 2016; Schievink et al., 2016a; Kranz et al., 2017). When taking anamnesis, you should explicitly ask about the orthostatic component at the beginning of the symptoms, as this can decrease in the course of the disease. In the history, there are sometimes indications of an increase in intracranial pressure, such as violent coughing. The diagnostic criteria are also increasingly being expanded from orthostatic “headaches” to orthostatic “symptoms” that can be explained by loss of CSF.


The most common cause of a CSF negative pressure syndrome is a CSF leak after diagnostic CSF puncture, which previously developed in 65% of patients within 24 hours and in approx. 90% within 48 hours (Lybecker et al., 1995; Vilming & Kloster, 1997). After lumbar punctures with thin (22–24 gauge) and atraumatic cannulas, it still occurs in 5–10% today (Armon & Evans, 2005; Lavi et al., 2006; Bezov et al., 2010b). Symptoms usually begin within 24 to 72 hours after the CSF puncture (within 48 hours in 80% and within 72 hours in 90% of those affected). Patients often complain of additional back pain, but the frequency does not correlate with the incidence of PPS (Halpern & Preston, 1994). Temporary cranial nerve lesions are rare (most common abdominal paresis). Subdural hematomas and hygromas or mostly reversible inner ear low-level hearing loss (2%) are also rare; The latter are explained by an endolymphatic hydrops as a result of the CSF hypotension.

The rate of spontaneous remissions is 50–80% in the first four to seven days (Dripps & Vandam, 1954; Lybecker et al., 1995). Symptoms that persist for weeks to months due to protracted epidural cerebrospinal fluid drainage are rare. If the headache persists for a longer period of time, the character of the headache may change. An initially position-dependent headache can turn into a dull, pressing, permanent headache with little modulation by the body position (Mokri, 2015).

The frequency of the PPS depends on

  1. the cannula thickness (Bezov et al., 2010a),
  2. the shape of their tip, e.g. atraumatic spinal cannulas according to Sprotte with a conical rounded tip and lateral opening or Whitacre needle are favorable (Braune & Huffmann, 1992; Müller et al., 1994; Strupp et al., 2001; Armon & Evans, 2005; current overviews: Arevalo-Rodriguez et al., 2017; Xu et al., 2017),
  3. the orientation of the cannula bevel in traumatic needles during insertion (parallel to the dura fibers, pushing apart, not cutting through the longitudinal dura fibers) (Lybecker et al., 1990; Evans et al., 2000; Bezov et al., 2010a). As a meta-analysis has shown, this reduces the incidence of PPS from an average of 25.8% to 10.9% (Richman et al., 2006).
  4. The reinsertion of the stylet after CSF removal in order to then remove both - stylet and cannula - seems to reduce the incidence (5 vs. 16% (Strupp et al., 1998; Strupp & Brandt, 1997)).

With a volume of 10-25 ml, the amount of CSF removed does not play a role in the development of post-puncture headaches (Alpers, 1925; Kuntz et al., 1992).

Spontaneous CSF leak (spontaneous intracranial hypotension, SIH)

The clinical appearance is much more variable with regard to the phenomenology of the headache and the accompanying symptoms in spontaneous CSF leaks, which are mainly found in the thoracic spine (Mokri, 2015). Usually there is also an initial headache in an upright position (Marcelis & Silberstein, 1990). Occasionally, pain between the shoulder blades and cervically precede the actual headache by days or weeks. Some patients only have a chronic dull, pressing headache and other headaches that increase with physical exertion, but are otherwise not position-dependent. In other cases there is no symptom in the first half of the day and an increase in headache in the afternoon and evening and, in very rare cases, paradoxical orthostatic headache that is present when lying down and improves when standing and walking. In most cases, the etiology of a spontaneous CSF leak cannot be clarified (Mokri, 2015). Spontaneous or traumatic dural defects, such as root pocket ruptures (especially thoracic), are assumed to be the cause, but these can only be detected in a few cases despite extensive imaging diagnostics (Mokri, 2015). Sometimes there is a constitutional structural weakness of the dural sac, e.g. in the context of Marfan's syndrome (Davenport et al., 1995). Sometimes trivial trauma while practicing the Valsalva maneuver, such as coughing, lifting heavy objects, or exercising, can be asked for (Mokri, 2015). More recent literature has shown that the spinal dural corner can be caused by microspurs or calcifications, whereby three types of CSF fistulas can be distinguished (Schievink et al., 2016a; Beck et al., 2016; Kranz et al., 2017). A reduced opening pressure (<6 cm H2O, lying down) could not be measured frequently during the lumbar puncture and is therefore not a conditio sine qua non. The opening pressure at the lumbar puncture is higher in 30–60% of patients with SIH (Kranz et al., 2015; Beck et al., 2017). Therefore, CSF hypovolemia was postulated as an alternative to explain the clinical symptoms and the signs in the imaging (Mokri, 1999; Kranz et al., 2017).

Previously expressed pathophysiological ideas that a CSF negative pressure syndrome can also be caused by decreased CSF production or increased CSF absorption could not be proven (Fishman, 1992). Symptoms of PPS can also result from overdrainage of a ventriculo-peritoneal or atrial shunt to treat hydrocephalus or through a CSF fistula.

A multitude of accompanying symptoms can occur with post-puncture headache as well as with spontaneous CSF leak. The most common are nausea, vomiting, dizziness, double vision (due to paresis of the eye muscles), blurred vision, photophobia, neck stiffness, tinnitus and hearing disorders (Mokri, 2003, 2004).