Intended for healthcare professionals


Taking the sting out of lumbar puncture

BMJ 2013; 346 doi: (Published 26 March 2013) Cite this as: BMJ 2013;346:f1734
  1. Paul Rizzoli, clinical director
  1. 1JR Graham Headache Center, Brigham and Women’s Faulkner Hospital, Boston, MA 02130, USA
  1. prizzoli{at}

Ultrasound guided procedures seem less likely to fail

Lumbar puncture remains an important and commonly performed diagnostic procedure,1 but training for its performance is not standardized.2 Although most diagnostic lumbar punctures are performed by neurologists, hospitalists, emergency department physicians, and pediatricians, physicians in many different specialties should have some experience with lumbar punctures and may on occasion need to perform one.

A well designed linked systematic review and meta-analysis by Shaikh and colleagues (doi:10.1136/bmj.f1720) investigates the benefit of using ultrasound guidance when performing lumbar puncture in routine diagnostic and therapeutic settings and in the performance of epidural catheterizations, mainly for giving anesthesia.3 The meta-analysis looked at 14 randomized studies with results from more than 1300 participants. It found a significant risk reduction for the primary outcome measure of failed procedures for ultrasound guided compared with the traditional anatomic approach to lumbar puncture. Failed procedures were defined conservatively as any failure to achieve the goals intended for the procedure. Epidural catheterizations were judged equivalent to subarachnoid punctures for assessing ultrasound guidance, and studies of either when combined achieved statistical significance. Six of 624 procedures failed in the ultrasound group compared with 44 of 610 in the control group (risk ratio 0.21, 95% confidence interval 0.10 to 0.43). Summary estimates for secondary outcomes of traumatic procedures, needle reinsertions, and needle redirections all supported the primary outcome finding. Time considerations in performance of the procedures could not be assessed owing to variability of reporting in the component studies.

Strengths of this meta-analysis include its comprehensive search for relevant studies and the high quality and low (modest) heterogeneity of the included studies. Methodological limitations involved variability in reporting of outcomes in the included studies. Complete blinding was logistically difficult. Most studies included young women receiving obstetric anesthesia administered by highly experienced practitioners, so generalizability to non-obstetric populations is limited. However, ultrasound guidance for lumbar puncture might offer even more benefits in non-obstetric populations. In these groups, lumbar puncture is more likely to be performed by practitioners with less procedural experience than obstetric anesthetists. The benefits shown may underestimate the potential benefits of a more general application of ultrasound guidance.

The authors point out that ultrasound guidance is now used at the bedside in the performance of many medical and surgical procedures, so its extension to lumbar puncture seems an inevitable trend towards improving procedural outcomes. Lumbar puncture is probably underused in the investigation of many problems, including chronic headache disorders, where identification of low or high pressure headaches with the measurement of opening pressure (which should almost always be obtained) may strongly affect treatment. Though the baseline failure rate for lumbar puncture was low in the studies even without ultrasound guidance, the same may not be true for less experienced operators. Furthermore, this analysis cannot provide information about lumbar punctures that were indicated but not performed. Because lumbar punctures may be refused by patients out of fear, or deferred by reluctant providers, ultrasound guidance may improve patient acceptance and reduce failure rate in this wider population.

This analysis provides no data on the impact of ultrasound guided lumbar puncture on the common complication of postdural puncture headache. This is a question of great clinical interest that merits further research. Unconfirmed clinical impressions suggest that cleaner less traumatic taps may paradoxically increase the risk of such headaches. This might be due to lower levels of clotting factors in the area of the tap that could help prevent a spinal fluid leak. Though this matter should be investigated in future research, other factors such as needle type may be more important determinants of this complication.

Identification of anatomic landmarks before lumbar puncture does not seem to be as accurate as ultrasound guidance, and it does not provide adequate information about optimal angle of needle insertion or required depth for the procedure. Pre-procedural static ultrasound can help by showing the midline, optimal vertebral level, and target depth. Dynamic ultrasound scanning allows the operator to follow progression of needle insertion. The use of ultrasound guidance does mean that the performance of lumbar punctures will become the province of specialized clinicians. Ultrasound guided lumbar puncture is not difficult to master and does not greatly increase the time needed to perform the procedure.4

The results of this analysis suggest one way to modernize and standardize the performance of lumbar puncture. Further research should investigate potential barriers to its implementation, confirm and quantify benefit, identify appropriate settings and patient populations, and investigate appropriate protocols and possible amendments to practice standards. Taken as a whole, the findings of this meta-analysis are compelling and support further investigation of the routine use of ultrasound to aid the performance of lumbar punctures. Ultrasound guidance shows promise as a way to “take the sting” out of lumbar punctures for patients and clinicians.


Cite this as: BMJ 2013;346:f1734


  • Research, doi:10.1136/bmj.f1720
  • Competing interests: I have read and understood the BMJ Group policy on declaration of interests and declare the following interests: none.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract