Indications

LP is used as both a diagnostic and therapeutic measure for

• 1.
  

  Temporizing: In emergency relief of CSF pressure, particularly in communicating hydrocephalus. The duration of benefit is typically short-lived, especially in high-pressure hydrocephalus, due to the CSF production rate of approximately 20 mL/h, leading to quick reaccumulation unless a lumbar drain is used.

  
  
• 2.
  

  Diagnosis: For measuring opening pressure, performing CSF infusion studies, and conducting tap tests, especially in suspected iNPH. It can also assist in the diagnosis of shunt infection or failure, in communicating hydrocephalus, through opening pressure measurement and CSF sampling.

Precautions and contraindications

• 1.
  

  Precautions : Recent guidelines strongly support the use of neuroimaging before LP in patients with focal neurologic deficits, papilledema, or altered mental status.

  
  
• 2.
  

  Contraindications : Caution is advised in cases of intraventricular outflow obstruction or basal meningitis. LP in patients with obstructive hydrocephalus, with a functioning ventricular CSF shunt, can precipitate acute low pressure hydrocephalus and should also be avoided.

Technique

• 1.
  

  Positioning: LP can be performed in the prone, lateral decubitus, or sitting positions. However, sitting LP has no advantage in terms of likelihood of success compared to lateral decubitus LP, which has a lower incidence of post-LP headache. Prone positioning has been used when using fluoroscopic guidance but a recent study shows no difference in success rates when compared to fluoroscopic-guided lateral decubitus LP, which has the additional advantage of facilitating opening pressure measurement with a manometer and CSF drainage.

  
  
• 2.
  

  Preparation: Proper aseptic technique, including skin preparation and sterile draping is required. Lidocaine is typically used for local anesthesia.

  
  
• 3.
  

  Procedure: Use anatomic landmarks for needle insertion. The needle is typically inserted either between the L3 and L4 or between L4 and L5 interspinous spaces, identified by palpating the iliac crests and following an imaginary line (Tuffier’s line [corresponding to L4/5]) to the spine. The needle should be advanced slowly, with the bevel parallel to the longitudinal dural fibers to minimize postprocedure CSF leakage. Fluoroscopy or ultrasound guidance may be used in difficult cases, particularly in patients with obesity, scoliosis, or previous spinal surgery. ^, This imaging guidance can help reduce the number of attempts and improve success rates in challenging LP cases.

  
  
• 4.
  

  Pressure measurement: Use a manometer or fluid-coupled transducer with a 3 way tap to facilitate infusion studies if needed.

Needle choice

The choice of needle for LP in hydrocephalus management depends on the indication. For diagnostic purposes, fine 25G pencil-point (Whitacre or Sprotte) or needles are often preferred to reduce the risk of postdural puncture headache. For therapeutic purposes or tap tests, larger gauge cutting, bevel-tip (Quincke) needles (eg, 20G) may be beneficial to encourage persistent CSF dural leakage and augment the therapeutic effect. In patients with obesity, extralong needles may be necessary.

Complications and risk mitigation

• 1.
  

  Post-LP headache : It is the most common complication (4.2%–11.0%). Use of atraumatic needles can reduce the risk by over 50%. Risk can be reduced by reinsertion of the stylet before removal of the needle, and orienting the bevel edge parallel to the long axis of the dural fibers when using a cutting-type needle. However, in hydrocephalus patients, the incidence of headache is generally lower as CSF leakage at the dural puncture site often has a therapeutic effect and the risk of iatrogenic intracranial hypotension is lower in this patient group, even with large gauge needles and extended CSF drainage. There is no evidence that bed rest after dural puncture reduces headache risk. Oral pregabalin and intravenous aminophylline may be helpful in both treatment and prevention of post-LP headache. For persistent headache, an epidural blood patch is usually effective.

  
  
• 2.
  

  Infection : It is rare, but serious. Strict aseptic technique is required.

  
  
• 3.
  

  Bleeding : It is rare in patients with normal coagulation. Caution is advised in anticoagulated patients. Current guidelines suggest LP is safe with an International Normalized Ratio (INR) of less than 1.5 and a platelet count of greater than 50,000/μL. ^,

  
  
• 4.
  

  Cerebral herniation : It is rare, with proper patient selection and pre-LP imaging. ^,

Tap test technique

A standard LP is performed and a large volume, typically around 40 mL, of CSF at a rate of 2 mL/min. However higher CSF volumes (40–60 mL), and larger gauge needles are associated with a better diagnostic yield. ^, Prior to, and following, the high-volume LP an objective measure of gait such as a 10 m gait test (recording steps, time and whether assistance was required) and/or the Timed Up and Go (TUG) test are performed, 3 trials are recommended, taking the best score as the response. Maximum walk speed is a more accurate measure than comfortable walk speed, with gait velocity being the most responsive parameter. Small improvements in the TUG test, by as little as 13%, or more than a 10% improvement in gait velocity are likely to represent meaningful improvements and be clinically significant. Additional objective measures of postural stability (Tinetti balance score, Berg Balance Score, 360 turn [steps/time], and 6 m tandem walk test) can be added. An objective measure of cognition (such as a montreal cognitive assessment [MoCA] or neuropsychological evaluation) can also be performed, but a cognitive response to a single tap test lacks sensitivity in identifying patients likely to experience cognitive improvement with CSF shunting. Timing of the repeat test is usually done 2 to 4 hours following the CSF drainage; however, gait measures remain valid for at least 24 hours. Gait responses may start to attenuate after 24 hours and day 4 testing is less sensitive, although a repeat test at day 7 may identify delayed responders. Measures of cognition such as the MOCA may improve in the days following the tap test and repeat testing at day 7 may also improve diagnostic yield. The presence of significant parkinsonism may attenuate the tap test response and worsen the negative predictive value (NPV) of the test.

Augmented tap test techniques

• 1.
  

  Repeated daily LP: Some centers perform repeated daily LPs with gait testing for up to 5 days. However, this approach can be uncomfortable for patients and may not offer significant advantages over an extended lumbar drainage (ELD) test.

  
  
• 2.
  

  Various techniques have been proposed to enhance the predictive value of the tap test, including combining it with a variety of imaging techniques or other bedside clinical assessments such as optic nerve sheath diameter measurement. None have been shown to increase the predictive value of the test sufficient to rule out a diagnosis of iNPH.

  
  
• 3.
  

  Supplementary biomarker assay may facilitate interpretation of results in patients with comorbid dementia, a low CSF Abeta (42/40) ratio is associated with a poorer cognitive response, but not gait and urinary response, to a tap test in iNPH.

The key flaw with all of the earlier studies is that none have consistently shunted negative responders, severely limiting the ability of these tests’ utility to “rule out” shunt-responsiveness. While the positive predictive value (PPV) of these tests (both standard tap tests and “augmented” techniques) is good, the evidence supporting their routine diagnostic use (over and above the use of clinical and radiological criteria) remains weak and they should not be used to exclude patients from shunting. The persistence of the tap test in clinical practice, despite these limitations, may be attributed to its ease of performance, low cost, and minimal invasiveness compared to more extensive procedures like ELD. Additionally, in resource-limited settings, it may serve as an initial screening tool before proceeding to more involved diagnostic measures.

Cerebrospinal fluid infusion study

Concept of R _out and Davson’s Equation

The resistance to CSF outflow (R _out ) is a key parameter derived from CSF infusion studies, based on Davson’s equation:

ICP = (R _out × formation rate) + Pss

where ICP is intracranial pressure, formation rate is the CSF production rate, and Pss is the sagittal sinus pressure. R _out is, therefore, a physiologic parameter that reflects the ease with which CSF can exit the intracranial compartment and enter the venous system. Increased resistance to CSF outflow can cause disruptions in CSF flow, leading to its accumulation, which could potentially result in conditions like hydrocephalus. Elevation of CSF outflow resistance in probable normal pressure hydrocephalus is a predictor of a positive clinical response to CSF shunt insertion. A number of thresholds used to predict a likely response to CSF shunt placement have been suggested, ranging from 12 to 18 mm Hg/mL/min.

Technique

In patients with communicating hydrocephalus, the test is usually performed via a LP, with the CSF compartment accessed via a single (usually larger gauge, needle [connected to a 3 way connector, 1 arm for transducing the ICP and 1 arm to perform the infusion]) or 2 separate needles (1 for transducing ICP and 1 for the infusion). In obstructive hydrocephalus, the test can be conducted via an external ventricular drain (again using a 3 way connector) or via an implanted ventricular access device (Ommaya reservoir; Fig. 1 A, B ) or CSF shunt reservoir (this must be proximal to the valve mechanism) accessed via two 25F butterfly needles (1 for ICP recording and 1 for the infusion). The patient is in the lateral decubitus position (for the lumbar infusion test). Lumbar CSF pressure is monitored for 5 minutes to establish a baseline level for ICP (ICP baseline [ICPb]). Normal saline or Ringer’s lactate solution is then infused, via a syringe driver, into the CSF compartment (subarachnoid space or cerebral ventricle) at a constant rate, typically 1.5 mL/min and the change in CSF pressure is recorded. Eventually a new “steady state” ICP is reached, known as the plateau pressure (ICP plateau [ICPp]; see Fig. 1 A, B). When the plateau pressure is reached, the R _out can be calculated from the equation: ICPp minus ICPb, divided by the infusion rate (If):

R _out = (ICPp − ICPb)/If

( A ) Infusion test via ventricular access device in patient with headaches but unchanged ventricular size following suspected failed ETV showing elevated R _out . ( B ) Same patient following insertion of ventriculo-peritoneal (VP) shunt with recurrent headaches, repeat infusion test confirms low R _out in keeping with functioning shunt system.

A “normal” CSF infusion test result would be a combination of baseline pressure less than 10 mm Hg and R _out less than 11 mm Hg/mL/min although normal thresholds are still subject to debate.

Use in Shunt Assessment

In vivo CSF infusion studies can be valuable in assessing CSF shunt function. A 2009 study by Petrella and colleagues demonstrated that infusion studies performed via VADs or shunt reservoirs could accurately detect shunt malfunction. CSF shunt infusion studies reduce radiation exposure and are cost-effective in the evaluation of suspected CSF shunt malfunction. The ICM+ software (Cambridge Enterprise Ltd, Cambridge, UK) has become a standard tool in many centers for automated analysis of infusion study data. The software defines a “shunt critical ICP,” specific to a shunt valve type that has a high PPV for shunt malfunction. Test interpretation in the setting of partial proximal catheter occlusion or slit ventricle syndrome can be more challenging. High infusion rates (>1.5 mL/min) can trigger antisiphon devices like the Codman SiphonGuard (Integra LifeSciences Corporation, Mansfield, MA), leading to false-positive results for shunt malfunction. Lower infusion rates (0.5–1.0 mL/min) are recommended when testing shunts with these devices. The technique has also been used (via an implanted VAD) to evaluate possible treatment failure after endoscopic third ventriculostomy (see Fig. 1 A, B).

Evidence and Limitations of the Tap Test and Lumbar Infusion Test (R _out )

The European iNPH Multicentre Study remains a key study for both the lumbar infusion test and the tap test’s predictive value in the assessment of shunt responsiveness in probable iNPH. This well powered, prospective study, in which all patients underwent both tests, and were blinded to the results and all underwent CSF shunt insertion, demonstrated that neither the tap test nor the lumbar infusion test can be used to rule out hydrocephalus due to their poor NPV. In this study, if a negative tap test and/or CSF infusion study had been used to exclude possible patients with iNPH from a shunt, 85% of those with less than a 10% improvement in gait score following a tap test and 81% of those with an R _out less than 18 mm Hg/mL/min, who still improved with a shunt despite a “negative test,” would have been denied beneficial treatment. They may still have value as a “rule-in” test, and both tests are associated with lower risks compared to extended lumbar drainage (ELD). Patients who fail the tap test but for whom there is a strong clinical suspicion of iNPH should still be considered for ELD, as it may reveal cases missed by the tap test alone.

Surgical technique

The catheters are inserted via a Tuohy needle. While the technique is similar to that for LP, for ease of passage of the catheter into the subarachnoid space, a different entry point and a needle trajectory are preferred. The entry point is 1 cm paramedian and 1 to 2 levels below the target interspace facilitating a more oblique needle trajectory into the target interspace. This avoids a more acute angle for passage of the catheter making it easier for the semirigid catheter to pass into the spinal subarachnoid space.

Choice of drain

• 1.
  

  Polyamide epidural catheter. The authors prefer to use these catheters for lumbar CSF drainage. Typically, a 16G epidural catheter is used, placed in the lumbar epidural space, typically at the L3/4 level and inserted to a length of 20 cm. These catheters are more rigid than traditional silicone catheters and do not require a guidewire.

  
  
• 2.
  

  Silicone drain: These drains typically are inserted with a guidewire and are also placed via a Tuohy needle. They may be more prone to catheter kinking than more rigid polyamide catheters. They are barium sulfate impregnated (radio-opaque), facilitating fluoroscopic guidance.

Extended lumbar drainage (ELD) test in the evaluation of probable Normal Pressure Hydrocephalus

ELD can be used as an alternative to the CSF tap test in the assessment of iNPH. A lumbar drain is typically inserted for between 2 and 5 days, although there is no evidence that CSF drainage beyond 2 days increases diagnostic yield and may increase the risk of CSF infection. CSF is drained at a rate of 10 mL/h, although the rate is slowed if low-pressure symptoms develop. As with the CSF tap test (see earlier discussion), an objective measure of gait such as a timed 10 m gait test or TUG test, is performed before and after the drainage period. Cognitive testing predrainage and postdrainage can also be performed. The test has a higher PPV than the CSF tap test, but the NPV, which is dependent on the prevalence of the disease in the test population, is less certain. A recent systematic review calculated that, with an estimated prevalence of shunt responsive iNPH of 60%, both PPV and NPV would be estimated at 90%, making it a more robust method of identifying shunt responders than either the tap test or CSF infusion study. However, the evidence-base for the use of ELD as a test to identify patients unlikely to respond to shunting is still limited and further clinical trials are needed. The predictive value of the ELD may be further enhanced using a post-drain diary evaluation and offering a CSF shunt to patients reporting a subjective improvement.

Temporizing acute communicating hydrocephalus

External lumbar drainage is an effective means of temporizing acute communicating hydrocephalus, particularly in settings where insertion of a CSF shunt device would be contraindicated such as in the presence of infection or following subarachnoid hemorrhage. The use of lumbar CSF drains to treat hydrocephalus associated with subarachnoid hemorrhage is associated with reduced rates of vasospasm, secondary cerebral infarctions, and mortality, without an increased risk of adverse events.

Pitfalls and considerations

• 1.
  

  External lumbar drainage is contraindicated in obstructive hydrocephalus and in patients with cerebellar tonsillar descent.

  
  
• 2.
  

  Lumbar stenosis is a relative contraindication. Symptoms of lumbar claudication should prompt MRI to rule out stenosis.

  
  
• 3.
  

  Careful patient supervision and monitoring is required with gravity drainage to ensure drainage height is adjusted in response to patient position to prevent CSF overdrainage, particularly in confused/demented patients. Novel devices such as the LiquoGuard (MÖLLER Medical GmbH, Fulda, Germany) can automate drainage to a specific hourly volume and monitor pressure without a risk of inadvertent overdrainage.

  
  
• 4.
  

  Caution is required when using a lumbar drain to treat hydrocephalus in patients with large decompressive craniectomy defects as this may precipitate paradoxical brain herniation.