Medicine is a personal and ever changing science where providers of medical care have a responsibility to use the most recent and reliable information to guide their patients. Curbside allows medical providers to create and distribute clinical pathways to improve the care provided at their institutions. However, there is the constant potential for human error, changes in medical sciences or patient specific variables that may not be taken into account in these pathways. Therefore, Curbside cannot warrant that the information contained in these resources is accurate or complete and are not responsible for any errors or omissions or for the results obtained from the use of this information. All users of Curbside should confirm the information provided here with other trusted sources and consult with expert health providers before making any health care decision.
The lumbar puncture is a common and safe procedure for multiple diagnostic and therapeutic interventions. Patients who are clinically unstable may have their LPs deferred until such time as they can safely be performed. In the case of presumed bacterial meningitis, if the LP cannot be performed quickly or the patient is acutely ill, giving antibiotics prior to LP may be warranted.
Certain patients with cardiopulmonary disease (including newborns with congenital heart disease) may be at increased risk of cardiovascular compromise if they are placed in the sitting or lateral recumbent position for LP. It is rare that these patients cannot still undergo an LP with attenuated positioning to ensure venous return.
The provider should evaluate all patients for anatomical issues which may make LP more difficult. Severe obesity, orthopedic complications, spine surgery and other issues may prompt the provider to consult with intervential radiology for fluoroscopic guidance for LP.
The most feared complication of a lumbar puncture is the risk of cerebral herniation secondary to the effect on the brain anatomy from an occupying mass or other cause in increased intracranial pressure. Multiple studies have tried to identify patients who are at high risk of herniation. Classically bacterial meningitis was a concern, however more recent studies have shown that meningitis is of little risk and that lumbar puncture can be safely preformed in these patients without the need of antecedent brain imaging.
Any patient with altered mental status, papilledema, seizures in the past week, impaired cellular immunity, history of intracranial pathology or other focal neurologic signs of increased intracranial pressure should undergo brain imaging (CT or MRI) prior to LP.
It should be noted that a normal CT does not completely remove the risk of herniation and that providers should proceed with caution in high risk patients.
In a large study of greater than 5,000 ALL patients of which 941 patients had platelet counts less than 50k and 29 patients had platelet counts less than 10k there were no significant complications from LP (of note, this is only a small number of patients less than 10k platelets and is difficult to interoperate). Also, in a study looking at hemophilia patients who received factor replacement prior to LP there were no major complications. However, patients with uncorrected bleeding risks, LP should only be undergone after a complete review of the risks and may be further clarified by consultation with a hematologist.
Anticoagulation has also been associated with an increased risk of spinal hematoma and bleeding complications, especially with INR greater than 4. Aspirin has not been associated with these risks. In any patient considered at risk of bleeding the provider should make a throughout risk assessment prior to proceeding. Spinal hematomas should be high on the differential of any patient with a bleeding diathesis and prompt intervention should proceed given the risk of neurologic compromise.
Topical anesthetics can provide pain relief for the skin puncture associated with lumbar puncture and should be used in most cases where time allows and there are no contraindications.
ELMA should only be placed on intact skin as it can be absorbed systemically through wounded tissue. It also carries the risk of methemoglobinemia in patients with underlying risk factors. The key to efficacy in use of EMLA appears to be length of time the medication is left on intact skin. Anesthesia measured by needle insertion was 3 mm one hour and 5 mm after 1.5 to 2 hours.
Liposomal lidocaine (ELA-max) has shown efficacy vs. placebo and in some cases vs. EMLA (although there are contradicting studies). It does have a faster onset of action than EMLA.
Similar to EMLA and Liposomal lidocaine, the self-heating lidocaine and tetracaine patch (Synera) has shown improvement over placebo. It has the shortest onset of action of the three.
Sedation has been shown to improve provider’s ability to adequately preform procedures in patients who otherwise would not be able to tolerate the procedure. There are no hard and fast indications for the use of procedural sedation. The providers should assess the patient's level of anxiety, behavioral status and ability to tolerate the procedure and weight this against the risks inherent in sedation.
Lumbar puncture can be a scary event, especially for young children. In appropriate cases, the use of anxiolysis can improve the safety of the procedure and the provider’s ability to perform the procedure. If initial non-pharmalogic methods are ineffective the use of nitrous oxide or midazolam can be helpful.
Evidence suggests that the use of nitrous oxide may be more effective and have a shorter recovery time than midazolam for short, minimally painful procedures (digital blocks, laceration repair, skin biopsy, etc.). Pediatric patients below 4 years of age are less likely to have successful anxiolysis with nitrious oxide.
Midazolam’s onset is 20 – 30 minutes, so the provider should wait this length of time minimum before starting the procedure (duration 30 – 60 min).
Oral sucrose (and other sweet tasting liquids) has been shown to have an anesthetic effect on infants up to 3 months of age (including preterm infants). Studies noted: reduced crying, reduced changes in physiologic responses (HR, O2sat, and vagal tone), low pain score from facial expression and improved composite pain scores. Sucrose likely works by activating the infant’s opioid receptors and non-opioid systems. Naloxone has been shown to block the effects of giving sucrose as well.
Therefore, for mildly painful procedures sucrose is likely a helpful adjunct to other anesthetics or used on its own.
The patient may be placed in the lateral recumbent or sitting position for lumbar puncture. The sitting position may be useful for ill patients or those with respiratory distress. It is often easier to monitor patients in this position and may not cause as much respiratory compromise as there is less neck hyperflexion. The lateral recumbent position is necessary for opening pressure measurement. However, an ultrasound study has shown that hyperflexion of the neck in lateral recumbent positioning did not increase the interspinous spaces an may not be useful (unless the child is uncooperative).
In order to safely perform a lumbar puncture the needle must enter the subarachnoid space below the terminating level of the spinal cord within the cauda equina. One should note that the level of spinal cord termination changes with age. At birth the tip is at the L3 vertebral body and as the child grows to adulthood the vertebral column grows faster than the cord so that by adulthood the tip is at the inferior border of the L1 vertebral body. Therefore in children less than 1 year the LP must be performed below the L2-L3 interspace.
To locate the L4 vertebral body the provider can follow a perpendicular line between the iliac crests. Ultrasound guidance has been shown helpful in identifying the correct lumbar interspaces and positioning for spinal needle insertion.
The use of local anesthetic (1% lidocaine) has been shown to decrease pain in children and adults and may improve the likelihood of successful procedure. Even in young children the use of local anesthetic should be used unless there are contraindications to administration.
The angle of the needle should be with the bevel parallel to the fibers of the ligamentum flavum (upwards in the lateral decubitus - sideways for sitting). This spreads the fibers of the ligamentum flavum instead of cutting them and may lead to a decreased incidence of post-LP headache.
Using higher gauge needles also has decreased the incidence of post-LP headaches as has the use of spinal (atraumatic) needles vs. conventional (cutting) needles.
A rare complication of lumbar puncture is the transfer of epidermal tissue into the spinal column causing an epidermal tumor. The lack of a stylet or loose stylet can catch epidermal tissue as it goes through the skin. Therefore a stylet should always be used during the skin puncture. However, once through the skin the stylet can be removed. Several studies have shown improved rates for successful and nontraumatic lumbar punctures with the removal of the stylet as it allows the provider to immediately know when CSF is obtained through visualization of fluid at the hub. After fluid is collected and the needle is to be removed, the stylet should always be reinserted into the needle as this decreases leakage through the dura and the incidence of post-LP headache.
Even with flexion of the lumbar spine, the needle often will need to be angled cephelad to transverse the interspinous space. Classically the provider was instructed to direct the needle towards the patient's umbilicus. In a study by Bruccoleri et al using ultrasound measurements the angle was more acute in infants less than one year than older children and using this angle may theoretically improve the provider’s success rate.
Normal opening pressure 50 - 200 mmH20 in relaxed patient & 100 - 280 mmH20 in patients with neck and legs flexed. Opening pressures are only valid if patient in the lateral recumbent position.
Previously patients were told to lay flat after a lumbar puncture to prevent post-LP headaches. However, there is no evidence that this helps and is not necessary for this procedure.