Study design
This trial was approved by the institutional review board of Seoul National University Hospital (No. 2005–149-1125) and written informed consent was obtained from all subjects participating in the trial. The trial was registered prior to patient enrollment at ClinicalTrials.gov (NCT04426916, Principal investigator: Jin-Tae Kim, Date of registration: 11 June 2020.)
From June to August 2020, we enrolled adult patients with an American Society of Anesthesiologists physical status of I–III who were scheduled for elective surgery at Seoul National University Hospital, Seoul, Republic of Korea. The first patient was enrolled in June, 23, 2020. Total of 40 participants, 20 patients with single-level spondylolisthesis (group S) and 20 patients with normal spinal anatomy (group N) were recruited. The sample size was set empirically by referring to previous studies comparing the distance on ultrasound images [11,12,13]. Lumbosacral spinal X-ray image were taken in all patients scheduled for surgery under spinal anesthesia for preoperative evaluation. Based on the preoperative X-ray images, patients with spondylolisthesis at only one lumbar level were screened and enrolled in group S. Patients without spondylolisthesis were enrolled in group N, and patients in group N were selected in a way that they were of the same gender, within 10% of the age, 10% of the height, and 15% of the weight of patients with spondylolisthesis. We excluded patients who had difficulty achieving the lateral decubitus position for spinal anesthesia, a history of lumbar spine surgery, serious spinal anatomical deformities other than spondylolisthesis, or spondylolisthesis at more than one level. All patients provided written informed consent prior to participation.
Ultrasonography procedure
All participants underwent preoperative spinal ultrasonography in a curtained waiting room or an operating room. Ultrasonography was performed by an anesthesiologist (YK) with experience of performing ultrasound-assisted spinal anesthesia in more than 50 cases. The C5-2 s convex array (frequency range: 2–5 MHz) of a TE7 Touch Enabled Ultrasound System (Mindray, Shenzhen, China) or the C1-5 convex array of a Venue Go™ (GE Healthcare, Chicago, IL) was used for scanning. A pillow was placed under the patient’s head in the lateral decubitus position to align their spine and their knees were bent toward their chest with neck flexion to attain the best position for spinal anesthesia. Ultrasonography was performed in the midline transverse and paramedian sagittal oblique views at the level of the spondylolisthesis (e.g., L4 on L5, L4/5) and the adjacent upper level (e.g., L3/4); thus, four images were obtained for each patient.
First, we used the transverse midline view to identify and mark the tips of the spinous processes on the skin and drew the neuraxial midline by connecting the tips and determined the interspinous spaces. The probe was then placed at the middle of the interspinous space, where the anterior complex (AC) and LFD were visible in a plane perpendicular to the back. At the point, the probe was tilted cephalad or caudad with respect to perpendicular plane to determine the angle at which the LFD was the longest. This point and angle were regarded as the “expected optimal needle insertion point and angle” of transverse midline approach for successful spinal anesthesia. The angle between the central axis of the lateral face of ultrasound probe and the patient's back at the expected needle insertion point was measured using a protractor (Fig. 1). If the ultrasound probe was tilted cephalad at the longest observed LFD, the angle was marked as positive, and if it was tilted caudad, the angle was marked as negative. The ultrasonography image in the paramedian sagittal oblique view was obtained on the dependent side by placing the ultrasound probe lateral to the midline and tilting it medially. The level of each interlaminar space was confirmed by counting up from the sacrum [11]. The expected optimal needle insertion angle was measured using the same method that was used for the transverse midline view (Fig. 1). The probe was placed at where the AC and LFD were visible and was tilted cephalad or caudad to determine the angle at which the interlaminar height of the LFD, defined as the length of the hyperechoic line visible through the interlaminar space, was the longest. The angle between the central axis of the frontal face of probe and patient’s back was measured. In the patients in group N, the angle was measured at a level corresponding to the spondylolisthesis level of the patient in group S with whom they were matched. Using the obtained images, the interlaminar height of the LFD, depth from the skin to the LFD, depth from the skin to the AC, and width of the intrathecal space (distance between LFD and AC) were measured. Another anesthesiologist who was blinded to group allocation performed these measurements. Representative ultrasound images of the transverse midline and parasagittal oblique view with measurements are shown in Fig. 2.
Outcome assessment
The primary outcome was the angle between the central axis of the ultrasound probe and the patient's back at the point at which the interlaminar space was widest in the midline transverse and paramedian sagittal oblique views at the level of the spondylolisthesis. Secondary outcomes included the angles measured using same method at the adjacent upper level of the spondylolisthesis and the interlaminar height of the LFD, depth from the skin to the LFD, depth from the skin to the AC, and width of the intrathecal space (distance between LFD and AC) at each level and view.
All methods were carried out in accordance with relevant guidelines and regulations.
Statistical analysis
Continuous data were tested for normality using the Kolmogorov–Smirnov and Shapiro–Wilk tests. Normally distributed data were compared using the Student t-test (mean \(\pm\) SD), and non-normally distributed data were compared using the Mann–Whitney test (median [interquartile range]). A two-tailed P < 0.05 was considered statistically significant. Categorical data were collated as numbers and percentages and compared using χ2 test or Fisher’s exact test. Data were analyzed using SPSS Statistics version 25.0 (IBM Corp., NY, USA).