Facebook Twitter YouTube


A Prospective Randomized Trial of the Effectiveness of Lumbar Puncture Simulation Training In Pediatric Residents

Sunday, October 21, 2012
Room 281-282 (Morial Convention Center)
Ryan S. Bode, M.D.1, Kelly Kelleher1, Dominic Moore1, Clarke Daxa1, Jeffrey Foti, M.D.2 and Caputo Grace1, (1)Department of Hospital Medicine, Phoenix Children's Hospital, Phoenix, AZ, (2)Department of Hospital Medicine, Seattle Children's Hospital, Seattle, WA


There is a clear emphasis on “real and/or simulated training” of procedures in the proposed ACGME program requirements for pediatrics. Recent evidence has shown a formal curriculum of patient simulation training could improve resident competency in core procedural skills. However, studies are needed to determine if pediatric procedural simulation training results in improved real patient performance. Objectives: The objectives of our study are to measure: 1.The difference in success rates, defined as successfully obtaining cerebral spinal fluid (CSF) in performing lumbar punctures (LP) between the control and study group 2. The difference in number of LP attempts prior to obtaining CSF between the control and study group 3. The proportion of patients with traumatic LPs, defined as CSF red blood cells (RBC) >400 , between the control and study group 4. The perceptions of the study group on LP simulation education.


PL-1 residents were prospectively randomized. The control group received the historical gold standard of LP procedural education, which is the "see one, do one, teach one" model. The study group received a LP tutorial and patient simulation in addition to standard training. The study group completed a survey before and after simulation education. Residents’ procedure logs and patient charts were reviewed to obtain patient age, successful completion of LP, number of attempts, and number of CSF RBC.



Control Group                        

Study Group              

Fisher exact test

 2 tailed p value

Participants with ≥ 1 LP attempt recorded




LP’s attempted (individual patient encounters)




Total number of LP attempts

45 (1.5 attempts/LP)

68 (1.7 attempts/LP)


CSF successfully obtained

16 (53.3%)

23 (59.0%)

.8068 = NS

CSF obtained on 1st attempt

10 (33.3%)

10 (25.6%)

.3332 = NS

Number of traumatic LP’s (rbc >400)

5 (31.3%)

5 (21.7%)

.7110 = NS


Although data trended towards better success rates and decreased traumatic taps with the simulation exposed study group, none of the data showed statistical significance.  The study will continue with tracking of the first 5 LPs of intern year for both groups and will continue with an additional recruitment of 31 new interns this summer.  Survey data of the study group did reveal significant improvement in confidence level following simulation.   Further subject recruitment and data collection will help determine if pediatric LP procedural simulation training results in improved real patient performance and outcomes.