Effect of ultrasonography localization of spinal landmarks on lumbar puncture in the emergency department

Abstract

Study objectives: We determine whether the use of ultrasonographic localization of spinal landmarks performed by emergency physicians affects the performance of lumbar puncture. Methods: This study was performed in an urban county teaching hospital. Adults 18 years or older who were to receive a lumbar puncture for routine clinical care in the emergency department were included. Target patient enrollment was set at 100. Patients were randomized to either undergo preprocedural ultrasonographic localization of the puncture site by the treating emergency physician or to have the puncture site determined by the usual landmark palpation method. Patients were assessed by the treating physician with respect to ease with which puncture-site landmarks could be palpated: easily palpable, difficult to palpate, or unable to palpate. The primary endpoint was success of the procedure as defined by return of cerebrospinal fluid; secondary endpoints were the number of needle passes, pain associated with the procedure, time to perform the procedure, and patient satisfaction with the procedure. The primary analysis compared use of ultrasonography versus palpation alone for all endpoints using the χ 2 test, Fisher's exact test, and the Wilcoxon rank sum test, as appropriate. A subgroup analysis was performed by dividing patients into subgroups according to ease of landmark palpation. No consideration was given to terminating this study according to this interim analysis. Results: Sixty-six patients have been enrolled to date. Median age of the patients was 39 years, and 48% were men. Thirty-three (50%) of the patients were considered easy to palpate, 21 (32%) were difficult to palpate, and 12 (18%) were designated as having landmarks that were not palpable. Thirty-three (50%) patients were randomized into each of the 2 study groups. There were no significant differences between the 2 groups in terms of age, sex, body mass index, or ease of palpation of landmarks. For the primary outcome, procedural success, there were no significant differences between those undergoing ultrasonographic localization and those with palpation alone. The only secondary outcome that was significantly different was the median time to perform the procedure, with palpation at 2.8 minutes (interquartile ratio [IQR] 4.8 to 14.2) versus ultrasonography at 7.9 minutes (IQR 0.9 to 14.0; P =.03). Subgroup analysis showed that in the group of patients whose landmarks were either difficult to palpate or not palpable, there was a trend toward improved success rate using ultrasonography. Success rate for ultrasonography group was 100% (14 of 14) versus 84% for the palpation-only group (16 of 19; P =.24). In this subgroup, there were no significant differences between groups for number of needle attempts, pain scores, patient satisfaction, or time of procedure. Conclusion: Preliminary data do not suggest any advantage to the routine use of ultrasonography in patients who require lumbar puncture. There may be an improved success rate in those patients whose spinal landmarks are either difficult to palpate or not palpable. Use of ultrasonography for lumbar puncture localization adds 5.1 minutes to the median time required to perform the procedure.