- Email: [email protected]
Lumbar intersegmental spacing and angulation in the modified lateral decubitus position versus variants of prone positioning
The Spine Journal 9 (2009) 580–584
Lumbar intersegmental spacing and angulation in the modified lateral decubitus position versus variants of prone positioning Vijay Agarwal, BSa, Michael Wildstein, MDa, John B. Tillman, PhDb, William L. Pelkey, PhDb, Todd F. Alamin, MDa,* a
Department of Orthopaedic Surgery, Stanford University School of Medicine, 300 Pasteur Drive, Stanford, CA 94305, USA b Medtronic Spine, LLC, 1221 Crossman Avenue, Sunnyvale, CA 94089, USA Received 10 July 2008; received in revised form 1 December 2008; accepted 6 April 2009
Abstract
BACKGROUND CONTEXT: Interspinous process devices represent an emerging treatment for neurogenic intermittent claudication resulting from lumbar spinal stenosis. Most published descriptions of the operative technique involve treatment of patients in the modified lateral decubitus knee-chest position (modified lateral decubitus), and yet many surgeons have begun to perform the procedure in various prone positions. The patient’s positioning on the operating room table seems likely to influence resting interspinous distance, and thus implant sizing and possibly the risk of intraoperative spinous process fracture. The intersegmental lumbar effect of variants on operative prone positioning compared with the modified lateral decubitus position has not been studied. PURPOSE: We performed this study to determine the comparative differences in interspinous distance and intersegmental angulation effected by the lateral decubitus knee-chest position and the variants on prone positioning used in practice. STUDY DESIGN/SETTING: Experimental human radiographic study. PATIENT SAMPLE: Twenty healthy male volunteers with a mean age of 43.6610.8 years (range, 24–63), without chronic back pain, symptoms of neurogenic claudication, or history of lumbar surgery were enrolled. OUTCOME MEASURES: Interspinous distance, anterior and posterior disc heights, disc angulation were measured on PACS monitor. METHODS: Lateral X-rays were taken of the lower lumbar spine in each of four different surgical positions (modified lateral decubitus, Andrews frame, Wilson frame, and Jackson frame). Statistical analysis was performed on the resultant data points to assess the significance of the effect of the position of the subject on intersegmental spacing and angulation. RESULTS: The 20 enrollees had a mean age of 43.6610.8 years (range, 24–63). The mean interspinous distance at the L4–L5 level was greatest on the Andrews table (23.568.3 mm) followed by the modified lateral decubitus position (19.665.1 mm), the Wilson frame (15.664.6 mm), and then the Jackson frame (10.164.7 mm; significantly less than all other positions p#.036). Mean segmental extension at the L4–L5 level was least in the modified lateral decubitus position ( 0.1 62.9 ); this was statistically similar to extension on the Andrews table (1.5 64.7 , p51.0), but significantly less than that recorded on the Wilson frame (4.6 63.1 , p!.001), and also significantly less than that recorded on the Jackson frame (p#.001). Similar differences in segmental measurements were observed at L3–L4. CONCLUSIONS: Prone positioning of patients in flexion on the operating table using the Andrews table or Wilson frame resulted in similar lumbar interspinous distance compared with the modified lateral decubitus position. Prone positioning on the Jackson frame resulted in statistically less interspinous distance than all other positions. Positioning on the Andrews table resulted in
FDA device/drug status: not applicable. Author disclosures: JBT (stockholder, manager for medical affairs, Medtronic Spine LLC); WLP (stockholder, data and statistics director at Medtronic Spine LLC); TA (royalties, Medtronic Spine; stockholder at Simpirica Spine; consultant for Medtronic and Simpirica; member of board 1529-9430/09/$ – see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2009.04.002
of directors and scientific advisory board of Simpirica Spine; research support from SPO #38201). * Corresponding author. Department of Orthopaedic Surgery, Stanford University School of Medicine, 300 Pasteur Drive, room R-171, Stanford, CA 94305, USA. Tel.: (650) 725-6797; fax: (650) 723-9805. E-mail address: [email protected] (V. Agarwal)
V. Agarwal et al. / The Spine Journal 9 (2009) 580–584
581
similar segmental angulation to the modified lateral decubitus position. Extrapolation from these data, obtained in healthy males younger than the typical age of patients treated with interspinous distraction devices, should clearly be done with caution. However, it seems reasonable to suggest that performing these procedures in the prone position using the Andrews table (greatest interspinous distance) is unlikely to result in the placement of significantly undersized implants, or significantly increase the force required to insert an implant. Ó 2009 Elsevier Inc. All rights reserved. Keywords:
Interspinous process decompression; Lumbar spine; Lateral film; Spinous process distance; Vertebral disc angulation; Operative position
Introduction
Materials and methods
Direct decompression is the standard surgical treatment for patients with symptomatic lumbar spinal stenosis [1–6]. A new alternative treatment method for symptomatic lumbar spinal canal stenosis is the placement of an interspinous spacer device, which was developed to kinematically decouple the symptomatic stenotic segment from the rest of the lumbar spine during extension. In preventing the symptomatic segment from extending, the hope is that patients will experience the relief of their symptoms in all positions that they note pain in the sitting or flexed position. The force needed to distract the spinous processes enough to place an interspinous device, is likely related to the resting relationship of the spinous processes, which is the net result of the patient’s muscular activity, the competence of the static restraints (ie, the inter/supraspinous ligamentous complex, ligamentum flavum, and facet capsules), and the patient’s position on the operative table [7–9]. This insertional force required is likely inversely correlated to the interspinous distance and directly correlated with the implant size. The extent of decompression in any given patient is likely directly correlated with the size of the interspinous process device that is placed. The method generally used to determine implant size is to find the ‘‘elastic limit’’ of the interspinous interval (the point at which the force needed to distract the interval further begins to dramatically increase) via manual distractive force exerted by the sizing device, and then to insert an implant that is 2 mm smaller (Zucherman J., personal communication, 2007). It is the surgeon’s assessment of the ‘‘feel’’ of the interspinous distraction, then, which determines the choice of implant size. In the original Food and Drug Administration study for the approval of the first interspinous device available in the United States, insertion was performed under local anesthetic in the modified lateral decubitus (MLD) knee-chest position [10,11]. In the broader postapproval use of the device, this method has not been universally adopted by surgeons. Many spine surgeons find the more familiar prone positioning used for posterior lumbar procedures an easier position in which to operate than the recommended MLD position. The current study evaluates the comparative differences in interspinous distance, as well as measures of disc height and angulation, between the MLD position and the variants on prone positioning used commonly in practice.
Study subjects Twenty healthy male volunteers, with a mean age of 43.6610.8 years (range, 24–63), consented to participate in the study. Exclusion criteria for the study included female gender (to avoid potential radiation to a fetus in an unknowingly pregnant subject), chronic back pain, symptoms of neurogenic claudication (more specifically leg pain aggravated by standing and walking and relieved by sitting), and/or history of lumbar surgery. The study protocol and informed consent form were approved by the Stanford University Institutional Review Board before the start of the study. Procedures Lateral lumbar spine radiographs of each subject were obtained in each of four positions: 1) left MLD with knees
Fig. 1. Measurements taken for each subject in each position at each lumbar level were spinous process distance, anterior and posterior disc heights, and segmental angulation. ADH, anterior disc height; PDH, posterior disc height; ID, interspinous distance.
582
V. Agarwal et al. / The Spine Journal 9 (2009) 580–584
Fig. 2. (Top) Mean and 95% confidence interval values are shown for interspinous distance and (Bottom) segmental extension for each position (labeled under each bar) for each lumbar level (labeled at the x-axis). p Values for each Tukey’s pairwise comparison within each level are shown below each graph.
and hips in approximately 120 flexion; 2) prone on the Andrews frame in the knee-chest position, with chest pad and adjustable tibial support adjusted to obtain 90 hip and knee flexion; 3) prone on a fully elevated Wilson frame (Mizuho OSI, Union City, CA, USA); and 4) prone on a Jackson frame (Mizuho OSI, Union City, CA, USA). Segmental disc height, angulation, and interspinous distance were measured as in Fig. 1 by the senior author. To eliminate the effect of source to film variation and subsequent magnification errors on measurements, the midline L4 vertebral body height was measured for all subjects and used to normalize measurements. Statistical methods To assess radiographic variables, we used analysis of variance. Comparisons were made to assess the effects of position, lumbar level, and any position by level
interaction. Tukey’s honestly significantly different post hoc pairwise comparisons were calculated with an alpha less than or equal to 5% considered statistically significant (p#.05). For spinous process distance and disc heights, pairwise comparisons were performed by level because of inherent differences in anatomy and biomechanical properties. Statistical analysis was performed using R version 2.6.0 for Macintosh.
Results At the L4–L5 level, the mean interspinous distance in the MLD position was 19.665.1 mm. The mean interspinous distances measured for L4–L5 using the Wilson frame (15.664.6 mm) and the Andrews table (23.568.3 mm) were not statistically different from that measured in the MLD position. However, positioning on the Jackson table
V. Agarwal et al. / The Spine Journal 9 (2009) 580–584
583
Table 1 Positional posterior and anterior disc heights by level Statistical comparison Level PDH (mm) L1–L2
L2–L3
L3–L4
L4–L5
L5–S1
ADH (mm) L1–L2
L2–L3
L3–L4
L4–L5
L5–S1
Position
Mean6SD
95% CI
Andrews
Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson
9.762.1 8.161.3 7.961.6 7.461.5 11.662.8 9.861.7 10.062.0 8.861.6 13.162.9 10.461.7 10.562.0 9.361.7 12.763.2 10.262.2 9.562.1 8.562.0 9.962.7 8.361.4 7.961.6 7.961.6
8.5–10.8 7.4–8.7 7.0–8.7 6.6–8.1 10.3–12.9 9.0–10.6 9.0–10.9 8.1–9.6 11.7–14.4 9.6–11.2 9.6–11.4 8.5–10.1 11.2–14.2 9.1–11.2 8.5–10.4 7.4–9.5 8.6–11.2 7.6–8.9 7.2–8.7 7.1–8.6
d NS NS NS d NS NS p5.004 d p5.008 p5.014 p!.001 d p5.015 p5.001 p!.001 d NS NS NS
Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson
10.463.3 8.061.9 8.661.7 9.662.5 12.363.3 9.361.9 10.962.3 12.662.1 13.763.5 10.461.7 12.361.9 12.962.1 14.163.8 10.762.0 13.262.6 13.561.6 15.061.9 11.961.7 13.762.2 13.662.2
8.7–12.2 7.1–8.9 7.8–9.5 8.3–10.8 10.7–13.8 8.4–10.2 9.8–12.1 11.6–13.6 12.1–15.4 9.6–11.2 11.4–13.2 11.8–13.9 12.3–15.9 9.8–11.7 12.0–14.4 12.6–14.4 14.1–15.9 11.0–12.7 12.6–14.7 12.5–14.7
d NS NS NS d p5.019 NS NS d p5.003 NS NS d p5.003 NS NS d p5.012 NS NS
Lateral
Wilson
Jackson
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS p5.004
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
SD, standard deviation; 95% CI, 95% confidence interval; PDH, posterior disc height; ADH, anterior disc height; NS, not significant.
resulted in significantly less L4–L5 distance than all other positions (10.164.7 mm; p#.036). Similar findings were noted at the L3–L4 level (Fig. 2). The interspinous distance for L5–S1 was not summarized as only four patients had spinous processes at S1. Positioning on the Andrews table resulted in similar mean segmental angulation compared with the MLD position for all levels (p51.0). The degree of segmental extension on the Jackson frame was statistically significantly greater when compared with the MLD position at all levels except L1–L2. At the L4–L5 level the Wilson frame resulted in relatively more segmental extension (4.6 63.1 ) compared with the MLD position ( 0.1 62.9 ; p!.001)
but was not statistically different from that obtained using the Andrews table (1.5 64.7 ; p5.09). Similar results were obtained at the L3–L4 level (Fig. 2). On average, positioning on the Andrews table resulted in significantly greater posterior disc height (PDH) than the other positions at both the L4–L5 (p#.015) and L3–L4 (p#.014) levels; all other PDH comparisons were not statistically significantly different. Mean anterior disc height (ADH) at the L4–L5 spinal level was also greater when subjects were positioned on the Andrews table (14.163.8 mm) compared with the MLD position (10.762.0 mm; p5.003); all other comparisons were not statistically significant. Similar results were found at the L3–L4 level (Table 1).
584
V. Agarwal et al. / The Spine Journal 9 (2009) 580–584
Both lumbar level and positioning had a statistically significant effect on disc angulation and interspinous distance as well as ADH and PDH (p!.001 for each); additionally, there was a statistically significant lumbar level by position interaction (p5.005) for disc angulation but not for interspinous distance or ADH and PDH (pO.5 for each).
Discussion We conducted this study to determine the comparative differences in interspinous process distance, ADH and PDH, and disc angulation effected by the modified lateral decubitus knee-chest position to variants on prone positioning commonly used in spinal surgical practice. Because the patient’s positioning on the operating room table may have a major impact on the difficulty of distracting the interspinous interval, these data may be helpful in avoiding problems, such as improper implant sizing and intraoperative spinous process fracture related to the insertional force required to implant the device. To our knowledge, no previous studies have reported on specific measures of intersegmental geometry, such as ADH and PDH, disc angulation, and interspinous distance in different spinal operative positions. Of the data collected in this study, the interspinous distance may be the most relevant measurement to the sizing and placement of an interspinous device as it is a direct measurement of the space available for the implant. No statistically significant differences were noted in the interspinous distance at any level on the Wilson frame or the Andrews table when compared with the MLD position. Between L2 and L5, prone positioning on the Jackson table was noted to result in significantly less interspinous distance when compared with all other positions. Similar segmental flexion results were obtained using the Wilson frame in segments between L1 and L3; however, at the L4–L5 level segmental flexion achieved was statistically significantly less compared with the MLD position. Angulation measurements achieved in the MLD position were comparable to the Andrews table in segments between L1 and L5. A trend was noted toward increasing interspinous distance on the Andrews table, and less on the Wilson frame, when compared with the MLD position. In addition to increasing interspinous distance, a trend was also noted for decreased segmental flexion on the Andrews table when compared with the MLD position. Some may find the trend of increased interspinous distance and decreased segmental flexion on the Andrew table intriguing. The likely explanation for this observation is the effect of the contraction of the lumbar flexors required to maintain the MLD positiondcontraction of these large muscles, such as the psoas and rectus, axially loads the spine and in doing so compresses the segment, narrowing the interspinous distance at a given angulation. This explanation is
supported by the finding of decreased ADH and PDH noted in the MLD position. This study has several limitations. The participants in this study were healthy volunteers and not patients with lumbar spinal stenosis; thus, the results may not be generalizable. The study population and design precludes the ability to use these data to predict clinical outcomes in patients treated with interspinous process devices. The sample size of the study was relatively small and may have limited our ability to detect clinically relevant differences between the positions. Each post hoc calculation performed increased the probability that one comparison would be statistically significant solely by chance. Extrapolation from these data, obtained in healthy males younger than the typical age of patients treated with interspinous distraction devices, should clearly be done with caution. However, it seems reasonable to suggest that performing these procedures in the prone position using the Andrews table is unlikely to result in the placement of significantly undersized implants, or significantly increase the force required to insert the implant. Based on our data, the only position which resulted in significantly less interspinous distance than that in the MLD position is prone positioning on the Jackson frame. References [1] Atlas SJ, Deyo RA, Keller RB, et al. The Maine Lumbar Spine Study, Part III. 1-year outcomes of surgical and nonsurgical management of lumbar spinal stenosis. Spine 1996;21:1787–94; discussion 94–5. [2] Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med 2008;358: 794–810. [3] Chang Y, Singer DE, Wu YA, et al. The effect of surgical and nonsurgical treatment on longitudinal outcomes of lumbar spinal stenosis over 10 years. J Am Geriatr Soc 2005;53:785–92. [4] Costa F, Sassi M, Cardia A, et al. Degenerative lumbar spinal stenosis: analysis of results in a series of 374 patients treated with unilateral laminotomy for bilateral microdecompression. J Neurosurg Spine 2007;7:579–86. [5] Ausman JI. Spinal stenosisdwhat is the best treatment? Surg Neurol 2007;68:486. [6] Chad DA. Lumbar spinal stenosis. Neurol Clin 2007;25:407–18. [7] Peterson MD, Nelson LM, McManus AC, Jackson RP. The effect of operative position on lumbar lordosis. A radiographic study of patients under anesthesia in the prone and 90-90 positions. Spine 1995;20:1419–24. [8] Benfanti PL, Geissele AE. The effect of intraoperative hip position on maintenance of lumbar lordosis: a radiographic study of anesthetized patients and unanesthetized volunteers on the Wilson frame. Spine 1997;22:2299–303. [9] Stephens GC, Yoo JU, Wilbur G. Comparison of lumbar sagittal alignment produced by different operative positions. Spine 1996;21: 1802–7. [10] Zucherman JF, Hsu KY, Hartjen CA. A prospective randomized multi-center study for the treatment of lumbar spinal stenosis with the X STOP interspinous implant: 1-year results. Euro Spine J 2004;13:22–31. [11] Zucherman JF, Hsu KY, Hartjen CA, et al. A multicenter, prospective, randomized trial evaluating the X STOP interspinous process decompression system for the treatment of neurogenic intermittent claudication: two-year follow-up results. Spine 2005;30: 1351–8.
Lumbar intersegmental spacing and angulation in the modified lateral decubitus position versus variants of prone positioning Vijay Agarwal, BSa, Michael Wildstein, MDa, John B. Tillman, PhDb, William L. Pelkey, PhDb, Todd F. Alamin, MDa,* a
Department of Orthopaedic Surgery, Stanford University School of Medicine, 300 Pasteur Drive, Stanford, CA 94305, USA b Medtronic Spine, LLC, 1221 Crossman Avenue, Sunnyvale, CA 94089, USA Received 10 July 2008; received in revised form 1 December 2008; accepted 6 April 2009
Abstract
BACKGROUND CONTEXT: Interspinous process devices represent an emerging treatment for neurogenic intermittent claudication resulting from lumbar spinal stenosis. Most published descriptions of the operative technique involve treatment of patients in the modified lateral decubitus knee-chest position (modified lateral decubitus), and yet many surgeons have begun to perform the procedure in various prone positions. The patient’s positioning on the operating room table seems likely to influence resting interspinous distance, and thus implant sizing and possibly the risk of intraoperative spinous process fracture. The intersegmental lumbar effect of variants on operative prone positioning compared with the modified lateral decubitus position has not been studied. PURPOSE: We performed this study to determine the comparative differences in interspinous distance and intersegmental angulation effected by the lateral decubitus knee-chest position and the variants on prone positioning used in practice. STUDY DESIGN/SETTING: Experimental human radiographic study. PATIENT SAMPLE: Twenty healthy male volunteers with a mean age of 43.6610.8 years (range, 24–63), without chronic back pain, symptoms of neurogenic claudication, or history of lumbar surgery were enrolled. OUTCOME MEASURES: Interspinous distance, anterior and posterior disc heights, disc angulation were measured on PACS monitor. METHODS: Lateral X-rays were taken of the lower lumbar spine in each of four different surgical positions (modified lateral decubitus, Andrews frame, Wilson frame, and Jackson frame). Statistical analysis was performed on the resultant data points to assess the significance of the effect of the position of the subject on intersegmental spacing and angulation. RESULTS: The 20 enrollees had a mean age of 43.6610.8 years (range, 24–63). The mean interspinous distance at the L4–L5 level was greatest on the Andrews table (23.568.3 mm) followed by the modified lateral decubitus position (19.665.1 mm), the Wilson frame (15.664.6 mm), and then the Jackson frame (10.164.7 mm; significantly less than all other positions p#.036). Mean segmental extension at the L4–L5 level was least in the modified lateral decubitus position ( 0.1 62.9 ); this was statistically similar to extension on the Andrews table (1.5 64.7 , p51.0), but significantly less than that recorded on the Wilson frame (4.6 63.1 , p!.001), and also significantly less than that recorded on the Jackson frame (p#.001). Similar differences in segmental measurements were observed at L3–L4. CONCLUSIONS: Prone positioning of patients in flexion on the operating table using the Andrews table or Wilson frame resulted in similar lumbar interspinous distance compared with the modified lateral decubitus position. Prone positioning on the Jackson frame resulted in statistically less interspinous distance than all other positions. Positioning on the Andrews table resulted in
FDA device/drug status: not applicable. Author disclosures: JBT (stockholder, manager for medical affairs, Medtronic Spine LLC); WLP (stockholder, data and statistics director at Medtronic Spine LLC); TA (royalties, Medtronic Spine; stockholder at Simpirica Spine; consultant for Medtronic and Simpirica; member of board 1529-9430/09/$ – see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2009.04.002
of directors and scientific advisory board of Simpirica Spine; research support from SPO #38201). * Corresponding author. Department of Orthopaedic Surgery, Stanford University School of Medicine, 300 Pasteur Drive, room R-171, Stanford, CA 94305, USA. Tel.: (650) 725-6797; fax: (650) 723-9805. E-mail address: [email protected] (V. Agarwal)
V. Agarwal et al. / The Spine Journal 9 (2009) 580–584
581
similar segmental angulation to the modified lateral decubitus position. Extrapolation from these data, obtained in healthy males younger than the typical age of patients treated with interspinous distraction devices, should clearly be done with caution. However, it seems reasonable to suggest that performing these procedures in the prone position using the Andrews table (greatest interspinous distance) is unlikely to result in the placement of significantly undersized implants, or significantly increase the force required to insert an implant. Ó 2009 Elsevier Inc. All rights reserved. Keywords:
Interspinous process decompression; Lumbar spine; Lateral film; Spinous process distance; Vertebral disc angulation; Operative position
Introduction
Materials and methods
Direct decompression is the standard surgical treatment for patients with symptomatic lumbar spinal stenosis [1–6]. A new alternative treatment method for symptomatic lumbar spinal canal stenosis is the placement of an interspinous spacer device, which was developed to kinematically decouple the symptomatic stenotic segment from the rest of the lumbar spine during extension. In preventing the symptomatic segment from extending, the hope is that patients will experience the relief of their symptoms in all positions that they note pain in the sitting or flexed position. The force needed to distract the spinous processes enough to place an interspinous device, is likely related to the resting relationship of the spinous processes, which is the net result of the patient’s muscular activity, the competence of the static restraints (ie, the inter/supraspinous ligamentous complex, ligamentum flavum, and facet capsules), and the patient’s position on the operative table [7–9]. This insertional force required is likely inversely correlated to the interspinous distance and directly correlated with the implant size. The extent of decompression in any given patient is likely directly correlated with the size of the interspinous process device that is placed. The method generally used to determine implant size is to find the ‘‘elastic limit’’ of the interspinous interval (the point at which the force needed to distract the interval further begins to dramatically increase) via manual distractive force exerted by the sizing device, and then to insert an implant that is 2 mm smaller (Zucherman J., personal communication, 2007). It is the surgeon’s assessment of the ‘‘feel’’ of the interspinous distraction, then, which determines the choice of implant size. In the original Food and Drug Administration study for the approval of the first interspinous device available in the United States, insertion was performed under local anesthetic in the modified lateral decubitus (MLD) knee-chest position [10,11]. In the broader postapproval use of the device, this method has not been universally adopted by surgeons. Many spine surgeons find the more familiar prone positioning used for posterior lumbar procedures an easier position in which to operate than the recommended MLD position. The current study evaluates the comparative differences in interspinous distance, as well as measures of disc height and angulation, between the MLD position and the variants on prone positioning used commonly in practice.
Study subjects Twenty healthy male volunteers, with a mean age of 43.6610.8 years (range, 24–63), consented to participate in the study. Exclusion criteria for the study included female gender (to avoid potential radiation to a fetus in an unknowingly pregnant subject), chronic back pain, symptoms of neurogenic claudication (more specifically leg pain aggravated by standing and walking and relieved by sitting), and/or history of lumbar surgery. The study protocol and informed consent form were approved by the Stanford University Institutional Review Board before the start of the study. Procedures Lateral lumbar spine radiographs of each subject were obtained in each of four positions: 1) left MLD with knees
Fig. 1. Measurements taken for each subject in each position at each lumbar level were spinous process distance, anterior and posterior disc heights, and segmental angulation. ADH, anterior disc height; PDH, posterior disc height; ID, interspinous distance.
582
V. Agarwal et al. / The Spine Journal 9 (2009) 580–584
Fig. 2. (Top) Mean and 95% confidence interval values are shown for interspinous distance and (Bottom) segmental extension for each position (labeled under each bar) for each lumbar level (labeled at the x-axis). p Values for each Tukey’s pairwise comparison within each level are shown below each graph.
and hips in approximately 120 flexion; 2) prone on the Andrews frame in the knee-chest position, with chest pad and adjustable tibial support adjusted to obtain 90 hip and knee flexion; 3) prone on a fully elevated Wilson frame (Mizuho OSI, Union City, CA, USA); and 4) prone on a Jackson frame (Mizuho OSI, Union City, CA, USA). Segmental disc height, angulation, and interspinous distance were measured as in Fig. 1 by the senior author. To eliminate the effect of source to film variation and subsequent magnification errors on measurements, the midline L4 vertebral body height was measured for all subjects and used to normalize measurements. Statistical methods To assess radiographic variables, we used analysis of variance. Comparisons were made to assess the effects of position, lumbar level, and any position by level
interaction. Tukey’s honestly significantly different post hoc pairwise comparisons were calculated with an alpha less than or equal to 5% considered statistically significant (p#.05). For spinous process distance and disc heights, pairwise comparisons were performed by level because of inherent differences in anatomy and biomechanical properties. Statistical analysis was performed using R version 2.6.0 for Macintosh.
Results At the L4–L5 level, the mean interspinous distance in the MLD position was 19.665.1 mm. The mean interspinous distances measured for L4–L5 using the Wilson frame (15.664.6 mm) and the Andrews table (23.568.3 mm) were not statistically different from that measured in the MLD position. However, positioning on the Jackson table
V. Agarwal et al. / The Spine Journal 9 (2009) 580–584
583
Table 1 Positional posterior and anterior disc heights by level Statistical comparison Level PDH (mm) L1–L2
L2–L3
L3–L4
L4–L5
L5–S1
ADH (mm) L1–L2
L2–L3
L3–L4
L4–L5
L5–S1
Position
Mean6SD
95% CI
Andrews
Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson
9.762.1 8.161.3 7.961.6 7.461.5 11.662.8 9.861.7 10.062.0 8.861.6 13.162.9 10.461.7 10.562.0 9.361.7 12.763.2 10.262.2 9.562.1 8.562.0 9.962.7 8.361.4 7.961.6 7.961.6
8.5–10.8 7.4–8.7 7.0–8.7 6.6–8.1 10.3–12.9 9.0–10.6 9.0–10.9 8.1–9.6 11.7–14.4 9.6–11.2 9.6–11.4 8.5–10.1 11.2–14.2 9.1–11.2 8.5–10.4 7.4–9.5 8.6–11.2 7.6–8.9 7.2–8.7 7.1–8.6
d NS NS NS d NS NS p5.004 d p5.008 p5.014 p!.001 d p5.015 p5.001 p!.001 d NS NS NS
Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson Andrews Lateral Wilson Jackson
10.463.3 8.061.9 8.661.7 9.662.5 12.363.3 9.361.9 10.962.3 12.662.1 13.763.5 10.461.7 12.361.9 12.962.1 14.163.8 10.762.0 13.262.6 13.561.6 15.061.9 11.961.7 13.762.2 13.662.2
8.7–12.2 7.1–8.9 7.8–9.5 8.3–10.8 10.7–13.8 8.4–10.2 9.8–12.1 11.6–13.6 12.1–15.4 9.6–11.2 11.4–13.2 11.8–13.9 12.3–15.9 9.8–11.7 12.0–14.4 12.6–14.4 14.1–15.9 11.0–12.7 12.6–14.7 12.5–14.7
d NS NS NS d p5.019 NS NS d p5.003 NS NS d p5.003 NS NS d p5.012 NS NS
Lateral
Wilson
Jackson
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS p5.004
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
d NS NS
d NS
d
SD, standard deviation; 95% CI, 95% confidence interval; PDH, posterior disc height; ADH, anterior disc height; NS, not significant.
resulted in significantly less L4–L5 distance than all other positions (10.164.7 mm; p#.036). Similar findings were noted at the L3–L4 level (Fig. 2). The interspinous distance for L5–S1 was not summarized as only four patients had spinous processes at S1. Positioning on the Andrews table resulted in similar mean segmental angulation compared with the MLD position for all levels (p51.0). The degree of segmental extension on the Jackson frame was statistically significantly greater when compared with the MLD position at all levels except L1–L2. At the L4–L5 level the Wilson frame resulted in relatively more segmental extension (4.6 63.1 ) compared with the MLD position ( 0.1 62.9 ; p!.001)
but was not statistically different from that obtained using the Andrews table (1.5 64.7 ; p5.09). Similar results were obtained at the L3–L4 level (Fig. 2). On average, positioning on the Andrews table resulted in significantly greater posterior disc height (PDH) than the other positions at both the L4–L5 (p#.015) and L3–L4 (p#.014) levels; all other PDH comparisons were not statistically significantly different. Mean anterior disc height (ADH) at the L4–L5 spinal level was also greater when subjects were positioned on the Andrews table (14.163.8 mm) compared with the MLD position (10.762.0 mm; p5.003); all other comparisons were not statistically significant. Similar results were found at the L3–L4 level (Table 1).
584
V. Agarwal et al. / The Spine Journal 9 (2009) 580–584
Both lumbar level and positioning had a statistically significant effect on disc angulation and interspinous distance as well as ADH and PDH (p!.001 for each); additionally, there was a statistically significant lumbar level by position interaction (p5.005) for disc angulation but not for interspinous distance or ADH and PDH (pO.5 for each).
Discussion We conducted this study to determine the comparative differences in interspinous process distance, ADH and PDH, and disc angulation effected by the modified lateral decubitus knee-chest position to variants on prone positioning commonly used in spinal surgical practice. Because the patient’s positioning on the operating room table may have a major impact on the difficulty of distracting the interspinous interval, these data may be helpful in avoiding problems, such as improper implant sizing and intraoperative spinous process fracture related to the insertional force required to implant the device. To our knowledge, no previous studies have reported on specific measures of intersegmental geometry, such as ADH and PDH, disc angulation, and interspinous distance in different spinal operative positions. Of the data collected in this study, the interspinous distance may be the most relevant measurement to the sizing and placement of an interspinous device as it is a direct measurement of the space available for the implant. No statistically significant differences were noted in the interspinous distance at any level on the Wilson frame or the Andrews table when compared with the MLD position. Between L2 and L5, prone positioning on the Jackson table was noted to result in significantly less interspinous distance when compared with all other positions. Similar segmental flexion results were obtained using the Wilson frame in segments between L1 and L3; however, at the L4–L5 level segmental flexion achieved was statistically significantly less compared with the MLD position. Angulation measurements achieved in the MLD position were comparable to the Andrews table in segments between L1 and L5. A trend was noted toward increasing interspinous distance on the Andrews table, and less on the Wilson frame, when compared with the MLD position. In addition to increasing interspinous distance, a trend was also noted for decreased segmental flexion on the Andrews table when compared with the MLD position. Some may find the trend of increased interspinous distance and decreased segmental flexion on the Andrew table intriguing. The likely explanation for this observation is the effect of the contraction of the lumbar flexors required to maintain the MLD positiondcontraction of these large muscles, such as the psoas and rectus, axially loads the spine and in doing so compresses the segment, narrowing the interspinous distance at a given angulation. This explanation is
supported by the finding of decreased ADH and PDH noted in the MLD position. This study has several limitations. The participants in this study were healthy volunteers and not patients with lumbar spinal stenosis; thus, the results may not be generalizable. The study population and design precludes the ability to use these data to predict clinical outcomes in patients treated with interspinous process devices. The sample size of the study was relatively small and may have limited our ability to detect clinically relevant differences between the positions. Each post hoc calculation performed increased the probability that one comparison would be statistically significant solely by chance. Extrapolation from these data, obtained in healthy males younger than the typical age of patients treated with interspinous distraction devices, should clearly be done with caution. However, it seems reasonable to suggest that performing these procedures in the prone position using the Andrews table is unlikely to result in the placement of significantly undersized implants, or significantly increase the force required to insert the implant. Based on our data, the only position which resulted in significantly less interspinous distance than that in the MLD position is prone positioning on the Jackson frame. References [1] Atlas SJ, Deyo RA, Keller RB, et al. The Maine Lumbar Spine Study, Part III. 1-year outcomes of surgical and nonsurgical management of lumbar spinal stenosis. Spine 1996;21:1787–94; discussion 94–5. [2] Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med 2008;358: 794–810. [3] Chang Y, Singer DE, Wu YA, et al. The effect of surgical and nonsurgical treatment on longitudinal outcomes of lumbar spinal stenosis over 10 years. J Am Geriatr Soc 2005;53:785–92. [4] Costa F, Sassi M, Cardia A, et al. Degenerative lumbar spinal stenosis: analysis of results in a series of 374 patients treated with unilateral laminotomy for bilateral microdecompression. J Neurosurg Spine 2007;7:579–86. [5] Ausman JI. Spinal stenosisdwhat is the best treatment? Surg Neurol 2007;68:486. [6] Chad DA. Lumbar spinal stenosis. Neurol Clin 2007;25:407–18. [7] Peterson MD, Nelson LM, McManus AC, Jackson RP. The effect of operative position on lumbar lordosis. A radiographic study of patients under anesthesia in the prone and 90-90 positions. Spine 1995;20:1419–24. [8] Benfanti PL, Geissele AE. The effect of intraoperative hip position on maintenance of lumbar lordosis: a radiographic study of anesthetized patients and unanesthetized volunteers on the Wilson frame. Spine 1997;22:2299–303. [9] Stephens GC, Yoo JU, Wilbur G. Comparison of lumbar sagittal alignment produced by different operative positions. Spine 1996;21: 1802–7. [10] Zucherman JF, Hsu KY, Hartjen CA. A prospective randomized multi-center study for the treatment of lumbar spinal stenosis with the X STOP interspinous implant: 1-year results. Euro Spine J 2004;13:22–31. [11] Zucherman JF, Hsu KY, Hartjen CA, et al. A multicenter, prospective, randomized trial evaluating the X STOP interspinous process decompression system for the treatment of neurogenic intermittent claudication: two-year follow-up results. Spine 2005;30: 1351–8.