Abstract
Background
Approximately 400,000 diagnostic lumbar punctures (LPs) are performed by neurologists yearly in the United States. There has been a transition from neurologists performing >40% of LPs 2 decades ago to now <15% of the time. High body mass index (BMI) is one of the key obstacles of LP for neurology residents, and the success rates drop to 58% when BMI is >35 kg/m2.
Methods
A prospective intervention study was conducted from March 2017 to March 2018 at an academic medical center. Patients were divided into ultrasound-guided LP or conventional LP. The primary outcome was the success rate of LP. The secondary outcome was the parameters regarding LP performance, the patients' feedback after procedures, and the residents' feedback about the learning modules.
Results
Eighty-two patients were enrolled with inclusion criteria of age >18 years, BMI ≥25 kg/m2, and able to give written informed consent. The ultrasound-guided LP group showed higher success rate (92.7% vs 68.3%, p < 0.05) and less duration of time (17.02 vs 37.63 minutes, p < 0.05) compared with the conventional LP group. It also showed advantages of decreased trials of needle insertions and redirections and reduced pain levels perceived by patients.
Conclusions
This study suggests that using ultrasound to localize before LP could increase the LP success rate and improve other related LP outcomes in the overweight populations. Our study also provides evidence that brief, targeted teaching modules are feasible and effective for ultrasound-guided LP training.
Classification of evidence
This study provides Class III evidence that for patients with BMI ≥25 kg/m2 undergoing LP, ultrasound guidance increases the LP success rate.
Lumbar puncture (LP) is a common diagnostic and therapeutic procedure for various neurologic diseases, and approximately 400,000 LPs are performed yearly by neurologists in the United States.1 However, LPs performed nationwide by neurologists and neurosurgeons dropped significantly overtime from 49.4% in 1991 to <15% in 2011.2
Recent studies have investigated the major obstacles of LP for neurology residents. Patients' body mass index (BMI) is one of the vital factors: the success rate dropped dramatically from >90% to 58% when BMI increased from <25 to >35 kg/m2.3 It is not surprising to see that LP providers have shifted toward radiologists2 because the prevalence of the overweight population currently is as high as 70.9% in adult males and 61.9% in adult females in United States.4 There is unmet need of developing new approaches to train neurology residents performing LPs to accommodate this dynamic change.
A recent meta-analysis of assistance with ultrasound for LP and epidural catheterization showed that ultrasound can help reduce the risk of number of needle redirections and traumatic taps.5 However, there are mixed results regarding the benefit when comparing routine use of ultrasound-guided vs conventional LP.6–8 Furthermore, although few randomized studies showed that ultrasound-guided LP could increase the success rate,6 they failed to show difference in obese patients when stratified by BMI.
To address this issue, this study sought to determine whether the guidance of ultrasound could improve the outcome of LPs performed by neurology residents in the overweight population after a targeted training.
Methods
Study design
This is a prospective intervention study in an academic medical center's LP clinic and neurology ward service from March 28, 2017, to March 28, 2018.
The inclusion criteria include (1) older than 18 years, needing a routine nonurgent LP (intensive care unit patients were not included) and (2) overweight by NIH definition, with BMI ≥25 kg/m2. The exclusion criteria include (1) not able to give consent, (2) pregnant women, and (3) prisoners.
Recruited patients in the LP clinic received ultrasound-guided LP, and patients in the neurology ward service received conventional LP. The recruited patients were informed that it was a randomized study, and they did not know whether they would be placed into either conventional LP or ultrasound-guided LP group. These procedures were performed by the same group of PGY-2 to PGY-4 neurology residents (n = 16) who had been certified with this procedure and completed the ultrasound-guided LP teaching module.
Parameters between these 2 groups were compared, including the LP time, which was defined by the time from palpitating the insertion site with or without the help of ultrasound until the beginning of successful CSF withdrawal or the stopping point of LP if the procedure failed. In addition, other LP-related parameters were also measured: the levels of needle insertions, which were defined by needle insertions with penetration of different skin areas within the same or different vertebral levels; the attempts of needle redirections, which were defined by redirection of needles without changing the penetration point of skin; whether radicular pain was felt during the procedure by the patients, and whether a traumatic tap, which was defined by using a cutoff of greater than 400 red blood cells, which is a common parameter of visual threshold for bloody fluid.9 The procedure performer reported the above numbers after LP. In addition, whether the patients have concomitant back disease (including lumbar scoliosis, L3-S1 spinal stenosis with or without surgery, lumbar arthritis, and lumbar spondylosis), which might influence the LP success rate, was also documented for outcome-stratified analysis. Post-procedure surveys were given to all patient participants, which contained questions including evaluation of the pain level from 0 to 10, ease of the procedure from 0 to 5, and satisfaction with the providers from 0 to 5. All patients were instructed to inform the providers either by phone call or email if there is persistent headache up to 3 days after the procedure to assess the need for blood patch, and these were also compared.
Standard protocol approvals, registrations, and patient consents
This investigation was approved by our institutional review board (#H00011379), and all recruited patients gave written informed consent. We adhere to the STrengthening the Reporting of OBservational studies in Epidemiology guidelines (strobe-statement.org).
Residents training module development
We developed a standardized 60-minute teaching module for the technique of ultrasound-guided LP. All levels of residents participated in this teaching module before the current study. This teaching module consists of a 15-minute lecture describing the use of the ultrasound machine, 7-minute video about ultrasound-guided LP (youtube.com/watch?v=DbqETxLurS0&t=249s), 30-minute simulation session of ultrasound guidance to localize insertion points for LP, and lastly 5–10 minutes for questions. Resident surveys were completed during the end of training module.
Ultrasound-guided LP
Patients are placed in the lateral decubitus position. Before using the ultrasound, visual and manual palpation of the spine is performed, the same as conventional LP, to approximate localization of the spinous process. Ultrasound-guided localization using a LOGIQ e ultrasound machine (General Electrics, Milwaukee, WI) with the 12L-RS probe is then performed10 (for details, please see figure 1), and the insertion point is marked. Next, the site is sterilized, and local anesthesia is given before routine LP is performed.
Figure 1. Ultrasound-guided lumbar puncture.
(A) Transverse view of lumbar spine ultrasound image. An ultrasound probe was first placed on the lumbar area perpendicular to the axis of spinous process, with slow movements of the transducer identification of the spinous processes; it showed typical hypoechoic crescent shape of the spinous process with posterior acoustic shadowing. (B) Longitudinal view of lumbar spine ultrasound image. After locating the spinous process in the center and a mark is drawn, the transducer was rotated 90° to align the ultrasound beam longitudinally over the midline and was moved along to find the widest interspinous space: there are crescent-shaped structures identified as “eye brow” shape, indicating spinous process. Centered in between these spinous process is the intervertebral space with visible superior and inferior borders.
Statistics
Unless otherwise stated, continuous variables are reported as mean (SD) or median (interquartile range). Categorical variables are reported as proportions. Between-group comparisons for continuous and ordinal variables were made with the Student t test. Categorical variables were compared using the χ2 test or Fisher exact test when expected cell sizes were less than 5. Non-normally distributed continuous variables were compared using the Wilcoxon rank-sum test. A multivariable logistic regression model was performed to investigate whether ultrasound guidance is independently associated with success of LP. The Hosmer-Lemeshow goodness-of-fit statistic was used to assess model fit. p < 0.05 was considered statistically significant with two-sided significance tests. All statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC).
Data availability
Further anonymized data used in developing the results presented in this article can be made available to qualified investigators upon reasonable request who have received ethical clearance from the host institution.
Results
Study cohort characteristics
One hundred nineteen consecutive adult patients who required nonurgent LPs were screened. Thirty-two patients were excluded either because of BMI <25 kg/m2 (n = 31) or because the patient was a prisoner (n = 1). Of 87 qualified patients, 5 patients chose not to consent to participate in the study. A total of 82 patients were enrolled, with 41 patients in the ultrasound-guided LP group and 41 patients in the conventional LP group.
Overall, the patients enrolled in the ultrasound-guided LP group had similar BMI (33.07 vs 33.35 kg/m2), mean value of weight (90.60 vs 92.25 kg), age (45.12 vs 44.56 years), and indication for LP compared with patients from the conventional LP group (p > 0.05, table 1).
Table 1.
Demographic and clinical features of the study participants
Feedback from residents after the learning module
In total, 19 junior and senior residents participated in the learning module and finished survey afterward giving feedback on the learning session (of these, 16 participated in performing LP, and the other 3 who were PGY-4 residents turned out not having any LP responsibility during the enrollment period). Of note, 73.7% of residents strongly agreed that ultrasound is easy to use for LP guidance vs 5.3% did not feel so after the training. Of note, 73.7% thought that it would be helpful for obese patients, and 94.7% felt more confident when using ultrasound guidance (figure e-1, links.lww.com/CPJ/A129).
Outcome comparison of ultrasound-guided LP vs conventional LP
The ultrasound-guided LP group showed higher success rates (92.7% vs 68.3%) compared with the conventional LP group (p < 0.05, table 2). When broken down to different BMI subgroups, the success rate of ultrasound-guided LP was 100% in patients with the BMI range of 25–35 kg/m2, 81.8% in patients with the BMI range of 35–40 kg/m2, and 75% in patients with the BMI range of more than 40 kg/m2. Consistently, the success rate is higher than that in the corresponding conventional ultrasound group because the success rate dropped to 57.1% when BMI >35 kg/m2. This low success rate in the conventional LP group in our cohort is similar to the previously reported literature3,8 (table e-1, links.lww.com/CPJ/A130).
Table 2.
Comparison of outcome of LP with and without ultrasound guidance
In addition, ultrasound-guided LP showed advantages of less duration of time (17.02 vs 37.63 minutes, p < 0.05), decreased level of needle insertions, and decreased trials of needle redirections (p < 0.05). There were also fewer traumatic taps in the ultrasound-guided LP group (5.3% vs 11.1%, p = 0.34), although not statistically different (table 2). For the significant complications of positional headache for potential blood patch, there was 1 patient in each group who contacted the providers and was advised to go to the emergency room for evaluation of blood patch despite conservative treatment. Neither patient required blood patch, as symptoms improved after receiving IV fluids.
To further understand the predictors associated with the primary outcome of LP success, we constructed multivariable logistic regression models to determine whether different BMI, ultrasound guidance, and history of back disease affect the predictive ability of LP success. It showed that ultrasound guidance is independently associated with the success of LP after adjustment of BMI and history of back diseases (OR: 9.36, 95% CI: 1.93–45.17, table 3).
Table 3.
Multivariable logistic regression analysis for factors that are associated with the success rate of LP in the entire cohort
Eighty-two LPs were performed in the study conducted by 16 neurology residents including a mixture of PGY-2 (22 LP), PGY-3 (32 LP), and PGY-4 (28 LP) trainees. There is no major difference in years of residents training who performed the LP procedures between the ultrasound-guided LP group and the conventional LP group (p = 0.26), nor is there considerable difference in the LP success rate among the different PGY levels of resident trainees (p = 0.56).
Patients' feedback for the procedure
Eighty-one patients (98.7%) in our cohort finished the survey after the LP, regardless of the success of the procedure. Noticeably, utilization of ultrasound guidance reduced the pain level associated with the procedure, with only 14.6% of patients reporting significant pain (pain level between 7 and 10/10) compared with those in the conventional group (58.5%, figure 2). Patients in the ultrasound-guided LP group also reported fewer incidences of radicular pain during the procedure (9.8% vs 29.3%, p < 0.05).
Figure 2. Post-LP survey from patients in the ultrasound-guided LP group and the conventional LP group.
(A) Feedback of pain level of the LP procedure. It showed that fewer patients who received ultrasound-guided LP complained about severe pain compared with those in the conventional LP group. (B) Feedback of the overall evaluation of the LP procedure. It showed that more patients in the ultrasound-guided LP group felt that it is an “easy procedure” compared with those in the conventional LP group. LP = lumbar puncture.
Discussion
Our study provides evidence that ultrasound-guided LPs, performed by neurology residents, could significantly increase the success rate, reduce LP time, and limit trials of the level of needle insertions and needle redirections in the overweight population. Utilization of ultrasound guidance may also decrease the pain level for patients. In addition, our study provides evidence that brief, targeted teaching modules are feasible and effective for ultrasound-guided LP training.
There has been subjective perceptions among neurologists that the proportion of failed LPs has increased over the last decade,3 and certainly, radiologists have indeed become one of the major providers covering more than 40% of LP procedures nationwide.2 After neurologists' failure of the bedside LP, patients are typically referred to radiologists for fluoroscopy-guided LP. Based on recently published data, the cost of fluoroscopy-guided LP is almost 1.5 times higher than conventional LP (559$ vs 393$),11 and there is always a time delay to fluoroscopy-guided LP due to multiple reasons such as availability of radiologists and insurance authorization, which at times extends hospital stays as well. Failure of bedside LPs increases patients' frustration, delays treatment or diagnosis, and increases the financial burden. Ultrasound machines are often easily accessible, noninvasive, have minimal requirement of additional staff, and are radiation-free. In the current study, it generally took less than 2 minutes to localize the potential intervertebral space, and one can help choose the optimal needle insertion level by observing different widths of intervertebral spaces. It is crucial to have a way to obtain LPs in a time-efficient and cost-effective manner.
A recent randomized controlled study that enrolled 100 patients showed no advantage of routine use of ultrasound localization for LP insertion, with similar success rate between the ultrasound-guided group (76%) and the conventional LP group (78%),8 which surprisingly drew opposite conclusion to previous studies with a smaller sample size (n = 46).6 This discrepancy could be due to the patients' enrollment with different mean value of BMI, skills of performers, and limitations of the sample size in the cohorts. In addition, both of these studies performed subgroup analysis for patients who are difficult to palpate, or obese patients with BMI > 30 kg/m2, and showed a slightly improved procedural success rate in the ultrasound-guided group of 77% vs 62% (n = 51)8 and 100% vs 42.8% (n = 12)6 separately; however, neither of these results showed difference statistically between the ultrasound-guided group vs conventional LP group because of the limitation of the small sample size and not including only typically challenging LP subsets as our study did. Our results provide an evidence that ultrasound-guided LP could be an optimal alternative method to improve the outcome of LP for overweight population, as an easily accessible, cost-effective method.
Success of LP depends on not only the skills of performers, patients' body habitus such as the amount of subcutaneous soft tissue, but may also be affected by the existence of altered anatomy of the lumbar spine such as lumbar spondylosis, previous lumbar surgery, or ligamentous calcifications. In our cohort, 3 of 41 patients failed ultrasound-guided LP, all of them have BMI > 35 kg/m2, and 33.3% (1/3) had lumbar spondylosis. Hence, we conducted a multivariable regression analysis to see whether the history of back disease could be a confounding factor. Our results show that ultrasound guidance is independently associated with outcome of LP success when considering BMI and history of back disease. Whether ultrasound-guided LP could help increase the success rate of patients with altered anatomy of lumbar spine is worthwhile for further investigation and is beyond the scope of the current study because of the limited sample size for patients who have back disease.
Our study developed a 1-hour-long targeted training module including both lecture and simulation session, which this combination has been demonstrated to be an effective teaching method for neurology residents.12 This teaching module is well received by ultrasound-naive trainees. Previous teaching methods used for anesthesiologist have a design of even shorter session, where they were taught with a cascade method using 5 minutes of instruction and 2 scans13; and it has been reproduced in other study with good performance response.6 It supports the notion that ultrasound-guided LP is an easy procedure to learn and the learning goal can be achieved with short and focused training.
Strengths of the current study are related to the investigation in a prospective fashion of a well-characterized patient population with moderate sample size for testing a specific hypothesis targeting one of the most common neurologist-performed procedures. By focusing on the overweight population, whose prevalence is now up to 60%–70% of adults, ultrasound-guided LP has been shown in our study as an effective option for this commonly encountered patient population.
Study limitations come from the different patient selection in groups by location of inpatient or outpatient, which was due to consideration of practical issues relating to the ready availability of ultrasound machine in our outpatient neurology LP clinic. This assignment method may introduce potential selection bias, although all the baseline characteristics of participants in these 2 groups have no major differences. It may also influence the pain level as patients who are hospitalized may have a higher pain level to begin with. It might also change why the procedure time was longer with the hospitalized (conventional LP) patients if there were perhaps more challenges with positioning in a hospital bed for example. The location where the procedure was performed introduces confounding that we cannot adjust for.
Our study demonstrated that ultrasound-guided LPs significantly increases the success rate, reduces the LP time, and limits the number of needle insertions and needle redirections in the overweight patient population. In addition, it also improves patients' satisfaction by decreasing their pain level.
Acknowledgments
The authors are grateful to Dr. David Paydarfar for his keen insights about this project. They are also grateful to all the patients who participated in the study and all their neurology residents listed as follows for their time and expertise in performing the procedure. Resident collaborators: Kelly Ramirez, MD, Mehdi Ghasemi, MD, Edouard Kamel, MD, Navid Valizadeh, MBBCh, Oleg Yrokhin, MD, Joshua Claunch, MD, Muhammad Azeem, MBBS, Tafline Kao, MBBChB, Abdul Mikati, MD, Malgorzata Miller, MD, Kathy Niu, MD, Andrew Slater, DO, Raffaella Umeton, MD, Dalia Abou Zeki, MD, Janette Abramowitz, MD, Justin Donaghy, MD, Andrew Galica, MD, Sylvia Josephy, MD, Daniella Pimental Maldonado, MD, and Phillippe Rosenberg, MD.
Appendix. Authors
Footnotes
Class of Evidence: NPub.org/coe
Study funding
This project received the 70th Annual American Academy of Neurology meeting safety and quality award in 2018 (Y.L.).
Disclosure
The authors report no disclosures relevant to the manuscript. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.
TAKE-HOME POINTS
→ The success rate of LP drops in overweight and obese patients.
→ Using ultrasound to localize intervertebral space before LP could increase the success rate in overweight and obese populations.
→ Brief, targeted teaching modules are effective for ultrasound-guided LP training.
References
- 1.Tung CE, So YT, Lansberg MG. Cost comparison between the atraumatic and cutting lumbar puncture needles. Neurology 2012;78:109–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kroll H, Duszak R Jr., Nsiah E, Hughes DR, Sumer S, Wintermark M. Trends in lumbar puncture over 2 decades: a dramatic shift to radiology. AJR Am J Roentgenol 2015;204:15–19. [DOI] [PubMed] [Google Scholar]
- 3.Edwards C, Leira EC, Gonzalez-Alegre P. Residency training: a failed lumbar puncture is more about obesity than lack of ability. Neurology 2015;84:e69–e72. [DOI] [PubMed] [Google Scholar]
- 4.Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2014;384:766–781. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Shaikh F, Brzezinski J, Alexander S, et al. Ultrasound imaging for lumbar punctures and epidural catheterisations: systematic review and meta-analysis. BMJ 2013;346:f1720. [DOI] [PubMed] [Google Scholar]
- 6.Nomura JT, Leech SJ, Shenbagamurthi S, et al. A randomized controlled trial of ultrasound-assisted lumbar puncture. J Ultrasound Med 2007;26:1341–1348. [DOI] [PubMed] [Google Scholar]
- 7.Mofidi M, Mohammadi M, Saidi H, et al. Ultrasound guided lumbar puncture in emergency department: time saving and less complications. J Res Med Sci 2013;18:303–307. [PMC free article] [PubMed] [Google Scholar]
- 8.Peterson MA, Pisupati D, Heyming TW, Abele JA, Lewis RJ. Ultrasound for routine lumbar puncture. Acad Emerg Med 2014;21:130–136. [DOI] [PubMed] [Google Scholar]
- 9.Shah KH, Richard KM, Nicholas S, Edlow JA. Incidence of traumatic lumbar puncture. Acad Emerg Med 2003;10:151–154. [DOI] [PubMed] [Google Scholar]
- 10.Soni NJ, Franco-Sadud R, Schnobrich D, et al. Ultrasound guidance for lumbar puncture. Neurol Clin Pract 2016;6:358–368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Eskey CJ, Ogilvy CS. Fluoroscopy-guided lumbar puncture: decreased frequency of traumatic tap and implications for the assessment of CT-negative acute subarachnoid hemorrhage. AJNR Am J Neuroradiol 2001;22:571–576. [PMC free article] [PubMed] [Google Scholar]
- 12.Gupta DK, Khandker N, Stacy K, Tatsuoka CM, Preston DC. Utility of combining a simulation-based method with a lecture-based method for fundoscopy training in neurology residency. JAMA Neurol 2017;74:1223–1227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Watson MJ, Evans S, Thorp JM. Could ultrasonography be used by an anaesthetist to identify a specified lumbar interspace before spinal anaesthesia? Br J Anaesth 2003;90:509–511. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Further anonymized data used in developing the results presented in this article can be made available to qualified investigators upon reasonable request who have received ethical clearance from the host institution.