e
PRIMARY RESEARCH
Expansion Open-door Laminoplasty With Foraminotomy Versus Anterior Cervical Discectomy and Fusion for Coexisting Multilevel Cervical Myelopathy and Unilateral Radiculopathy Zhao Fang, MD,*w Rong Tian, MD,* Tian-wei Sun, MD,* Sandip K. Yadav, MD,* Wei Hu, MD,* and Shui-qing Xie, MD*w
Study Design: This was a clinical prospective study. Objective: To assess whether clinical and radiologic outcomes differ between expansion open-door laminoplasty with foraminotomy (EOLF) and anterior cervical discectomy and fusion (ACDF) in the treatment of coexisting multilevel cervical myelopathy and unilateral radiculopathy (CMUR). Summary of Background Data: No reports to date have compared clinical outcomes between anterior and posterior decompression for CMUR. Materials and Methods: We prospectively performed ACDF (n = 59) in 2004, 2006, and 2008 and EOLF (n = 62) in 2005, 2007, and 2009. The Japanese Orthopedic Association (JOA) score and recovery rate were evaluated. For radiographic evaluation, the lordotic angle and range of motion at C2–C7 were investigated. Results: Only 110 patients could be followed for >3 years (EOLF/ACDF: 56/54; follow-up rate, 90.9%). Demographics were similar between the 2 groups. Compared with ACDF, in EOLF group there were shorter operating time (144 vs. 178 min), less bleeding (175 vs. 192 mL), and fewer complications (P < 0.05). Results of JOA score and recovery rate, at 3-year postoperative follow-up, showed no statistical difference for the 2 groups. Cervical lordosis of ACDF increased from 13.7 to 16.2 degrees, whereas that of EOLF group decreased from 14.6 to 13.3 degrees (P < 0.05). The percentage of range of motion declined in the 2 groups (ACDF/EOLF), 57.4% versus 74.7% (P < 0.05).
Received for publication September 3, 2013; accepted December 9, 2013. From the *Department of Spine Surgery, Tianjin Union Medicine Center, Tianjin Institute of Spine Surgery; and wNankai Medical School, Nankai University, Tianjin, China. Z.F. and R.T. contributed equally and were considered co-first authors. The authors declare no conflict of interest. Reprints: Tian-wei Sun, MD, Department of Spine Surgery, Tianjin Union Medicine Center, Tianjin Institute of Spine Surgery, 190 JieYuan Rd, Hongqiao District, Tianjin 300121, China (e-mail: [email protected]). Copyright r 2013 Wolters Kluwer Health, Inc. All rights reserved.
Clin Spine Surg
Volume 29, Number 1, February 2016
Conclusions: The 2 surgical procedures have similar clinical effects in treating multisegmental CMUR. However, the EOLF group demonstrated shorter operative time, less blood loss, and fewer complications; therefore, it proved to be a more effective and safer method. Key Words: cervical radiculomyelopathy, foraminotomy, laminoplasty, anterior decompression, prospective comparative study (Clin Spine Surg 2016;29:E21–E27)
C
oexisting multilevel cervical myelopathy and unilateral radiculopathy (CMUR) is due to the cervical spine degeneration and some other factors causing cervical and nerve root canal stenosis, resulting in the compression of spinal cord and nerve tissues. Surgical decompression is the treatment choice. At present, there are 2 surgical procedures for the disease caused by disk herniation: anterior cervical discectomy and fusion (ACDF) and laminoplasty. The traditional posterior laminoplasty does not relieve foraminal stenosis causing nerve root compression, whereas in case of ACDF, the adjacent disk degeneration is accelerated.1 As Hilibrand et al2 reported, of all the 374 patients who underwent anterior cervical surgery, symptomatic adjacent segment disease occurred at a relatively constant incidence of 2.9% per year and that 25.6% of the patients had new disease at an adjacent level within 10 years after the operation. When we applied expansion open-door laminoplasty with foraminotomy (EOLF) to treat multisegment CMUR, we found satisfactory clinical outcomes and less complication.
MATERIALS AND METHODS Patient Series This was a single-centered prospective comparative study of 2 surgical procedures in treating CMUR. The study was approved by the hospital medical ethics committee. All patients were well explained about the study, and an informed consent was obtained from each one. www.clinicalspinesurgery.com |
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
E21
Clin Spine Surg
Fang et al
Inclusion criteria were as follows: (1) cervical disk herniation causing myelopathy; (2) compression of spinal cord at Z3 levels; (3) unilateral nerve root compression symptoms; and (4) >3 years of follow-up. Exclusion criteria were as follows: (1) patients with ossification of posterior longitudinal ligament, hypertrophy of flavum ligament, excessive cervical kyphosis above 15 degrees, and segmental instability; (2) previous history of cervical trauma or surgery; and (3) patients of bilateral nerve root compression symptoms. In this study, we also included the patients with small central calcification of disk. After a proper evaluation, all those cases were made sure that operative procedures could be performed. A total of 121 patients (62 cases of EOLF and 59 of ACDF) were included in the study with the mean age of 58.6 years. According to the imaging studies, spinal cord compression levels include C3/4–C5/6 in 37 cases, C4/5– C6/7 in 41 cases, and C3/4–C6/7 in 33 cases. The nerve root compression level include: C3/4 in 6 cases, C4/5 in 23 cases, C5/6 in 39 cases, C6/7 in 16 cases, C4/5–C5/6 in 19 cases, and C5/6–C6/7 in 18 cases.
Surgical Procedures After assessing preoperative signs and symptoms, and the imaging studies, the segment responsible for the compression of spinal cord was determined. On the basis of imaging findings, and patients radicular signs and symptoms, the foraminotomy level and side were decided. Every alternate year, 59 patients were enrolled in the ACDF group in 2004, 2006, and 2008 and 62 patients were enrolled in the EOLF group in 2005, 2007 and 2009. All procedures throughout this study were performed by a senior spinal surgeon and his team.
Anterior Cervical Discectomy Decompression and Cage Insertion With Titanium Plate and Screw Fixation After giving the general anesthesia, patient was kept in supine position with neck slightly extended. A right transverse paramedian incision was made at the level of anterior transverse stripe of the neck. After careful dissection of subcutaneous tissue and maintaining the hemostasis, the platysma muscle was exposed. Thereafter, the platysma was split to expose the inner edge of the right sternocleidomastoid. Sternocleidomastoid muscle was retracted laterally and then a blunt dissection was made. After this supraomohyoid muscle along with the carotid sheath pulled to the lateral side, and the trachea, esophagus with its sheath pulled medial side. Then prevertebral fascia was separated layer by layer under direct vision and prevertebral space exposed. The intervertebral space was located with the help of a needle under the guidance of image intensifier (C-arm). Then after cutting the anterior longitudinal ligament, intervertebral discectomy was performed at corresponding levels. With the help of Kerrison punch and nucleus rongeur, the margins of the gap cleared and the posterior longitudinal ligament (PLL) was excised. With the excision of PLL, epidural space was exposed and the decompression was confirmed. Then spaces were irrigated
E22 | www.clinicalspinesurgery.com
Volume 29, Number 1, February 2016
with sterile normal saline solution and appropriate-sized cages incorporated with bone allograft were placed, respectively. Small pieces of bone allograft were made and filled in the intervertebral spaces for the fusion. The vertebrae above and below the fusion level were fixed with the appropriate-sized titanium plate and screw. The positions of cage, plate, and screws were confirmed with the C-arm. At last, drainage tube was placed and the wound was closed in layers (Figs. 1A–D).
EOLF After administering general anesthesia, patient was placed in prone position with neck slightly flexed. The cervical spine was gently fixed with a Mayfield collar. Through a midline vertical incision, the spinous process, laminae, and facet joints were exposed from C3–C7. Then, after cutting the roots of spinous process, a hole was made in the spinous process. Using a high-speed drill, a trough was created on the “open” side at the junction between the laminae and the facets. The trough was drilled down, and a thin remnant of laminae was left. This thin rim and associated ligament were then removed with a 1-mm or 2-mm Kerrison punch. A second trough was then drilled in the opposite or “closed” side with a highspeed drill. Care was taken to only thin the lamina on this side and not cut all the way through. The opening on the open side was then gently expanded, thus lifting the lamina off the spinal cord and expanding the canal. The surgeon then gently expanded the opening with a Penfield dissector, while the assistant gently rotated the laminae toward the closed side using some appropriate instruments. Care was taken to not allow the block of laminae to slip and rapidly snap back into original position as this may result in spinal cord injury. The distance of opened door was 10 mm or the lamina angle was maintained at 45–60 degrees. Thereafter, the door was kept open by placing a suture through the facet capsule on the closed side and through the spinous process. After the side opening procedure, the stenosed foramen on the open side was selected. While cutting the posterior wall of the intervertebral foramen, medial edge of the upper and lower facet resection should be r50%. This resection removed the posterior part of the foramen, thus making nerve root decompressed. Then, small pieces of bone graft were embedded in the hinge side of the door to increase the stability. At last, drainage tube was placed after thorough irrigation and wound was closed in layers (Figs. 2A–D).
Postoperative Rehabilitation Postoperative care was almost same in both the groups. The drainage tube was routinely placed in all patients for 24–72 hours. Postoperative pain management, nutrition therapy, and some adjuvant therapy were equally emphasized. Cervical collar was provided for 4–6 weeks. From the second postoperative day, isometric muscle strengthening exercises were started under the guidance of the surgeon. Patients were usually discharged at postoperative day 7. Copyright
r
2013 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Clin Spine Surg
Volume 29, Number 1, February 2016
EOLF Versus ACDF in the Treatment of CMUR
FIGURE 1. A 48-year-old woman with B/L upper and lower extremities weakness for 3 years with increased numbness of right upper extremity since 1 week. A, Magnetic resonance imaging showing posterior intervertebral disk herniation at C3/4–C6/7 level compressing the spinal cord. B, Left oblique radiograph showing C5/6 foraminal narrowing. C, Postoperative radiograph showing C3/4, C4/5, and C5/6 intervertebral cage in situ with titanium plate and screw fixation. D, Postoperative magnetic resonance imaging showing the decompression of cord and restoring the spinal canal.
Evaluation Clinical Outcomes
The surgical complications that occurred, blood loss, and operation time were also recorded.
The Japanese Orthopedic Association (JOA) scoring system was used for the evaluation of cervical radiculomyelopathy before and after surgery. The recovery rate was calculated to compare preoperative and postoperative JOA scores. Each item of the recovery rate about each function was calculated as follows: recovery rate (%) = (postoperative individual JOA scorepreoperative individual JOA score) 100/(total individual JOA score preoperative individual JOA score). Copyright
r
2013 Wolters Kluwer Health, Inc. All rights reserved.
Radiologic Evaluation C2–C7 lordotic angle (a): cervical sagittal alignment was measured by tangential lines on the posterior edge of C2 and C7 bodies on lateral radiograph in a neutral position according to the Harrison posterior tangent method.3 Total range of motion (ROM) of the cervical spine was obtained by summation of the angles on flexion and www.clinicalspinesurgery.com |
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
E23
Clin Spine Surg
Fang et al
Volume 29, Number 1, February 2016
FIGURE 2. A 62-year-old man with B/L lower extremities numbness and weakness for 3 years with increased left upper extremity weakness and pain for 15 days. A, Magnetic resonance image scans showing multisegment (C2–C7) posterior disk herniation causing spinal cord compression and spinal canal narrowing. B, Postoperative computed tomography scan showing single opendoor (left to right) laminoplasty with left foraminotomy at C5/6 level. C, Postoperative magnetic resonance imaging showing decompression of spinal cord and restoration of spinal canal. D, Postoperative 3-dimensional computed tomography scan showing left C5/6 foraminotomy.
extension lateral radiographs before surgery and at the time of annual visit. Angles were created by way of cervical sagittal alignment and tangential lines on the posterior edge of C2 and C7 bodies on lateral radiograph in a neutral position (a), and on flexion (g) and extension lateral radiographs (b). Percentage of ROM (ROM%) was defined as 3 y postoperative ROM/preoperative ROM 100. The radiographic measurements were performed by 2 independent surgeons who were blinded about the clinical status of the patients (Fig. 3).
E24 | www.clinicalspinesurgery.com
Statistical Analysis The t test was used in the statistical analysis with applying SPSS19.0 statistical software. All P values
PRIMARY RESEARCH
Expansion Open-door Laminoplasty With Foraminotomy Versus Anterior Cervical Discectomy and Fusion for Coexisting Multilevel Cervical Myelopathy and Unilateral Radiculopathy Zhao Fang, MD,*w Rong Tian, MD,* Tian-wei Sun, MD,* Sandip K. Yadav, MD,* Wei Hu, MD,* and Shui-qing Xie, MD*w
Study Design: This was a clinical prospective study. Objective: To assess whether clinical and radiologic outcomes differ between expansion open-door laminoplasty with foraminotomy (EOLF) and anterior cervical discectomy and fusion (ACDF) in the treatment of coexisting multilevel cervical myelopathy and unilateral radiculopathy (CMUR). Summary of Background Data: No reports to date have compared clinical outcomes between anterior and posterior decompression for CMUR. Materials and Methods: We prospectively performed ACDF (n = 59) in 2004, 2006, and 2008 and EOLF (n = 62) in 2005, 2007, and 2009. The Japanese Orthopedic Association (JOA) score and recovery rate were evaluated. For radiographic evaluation, the lordotic angle and range of motion at C2–C7 were investigated. Results: Only 110 patients could be followed for >3 years (EOLF/ACDF: 56/54; follow-up rate, 90.9%). Demographics were similar between the 2 groups. Compared with ACDF, in EOLF group there were shorter operating time (144 vs. 178 min), less bleeding (175 vs. 192 mL), and fewer complications (P < 0.05). Results of JOA score and recovery rate, at 3-year postoperative follow-up, showed no statistical difference for the 2 groups. Cervical lordosis of ACDF increased from 13.7 to 16.2 degrees, whereas that of EOLF group decreased from 14.6 to 13.3 degrees (P < 0.05). The percentage of range of motion declined in the 2 groups (ACDF/EOLF), 57.4% versus 74.7% (P < 0.05).
Received for publication September 3, 2013; accepted December 9, 2013. From the *Department of Spine Surgery, Tianjin Union Medicine Center, Tianjin Institute of Spine Surgery; and wNankai Medical School, Nankai University, Tianjin, China. Z.F. and R.T. contributed equally and were considered co-first authors. The authors declare no conflict of interest. Reprints: Tian-wei Sun, MD, Department of Spine Surgery, Tianjin Union Medicine Center, Tianjin Institute of Spine Surgery, 190 JieYuan Rd, Hongqiao District, Tianjin 300121, China (e-mail: [email protected]). Copyright r 2013 Wolters Kluwer Health, Inc. All rights reserved.
Clin Spine Surg
Volume 29, Number 1, February 2016
Conclusions: The 2 surgical procedures have similar clinical effects in treating multisegmental CMUR. However, the EOLF group demonstrated shorter operative time, less blood loss, and fewer complications; therefore, it proved to be a more effective and safer method. Key Words: cervical radiculomyelopathy, foraminotomy, laminoplasty, anterior decompression, prospective comparative study (Clin Spine Surg 2016;29:E21–E27)
C
oexisting multilevel cervical myelopathy and unilateral radiculopathy (CMUR) is due to the cervical spine degeneration and some other factors causing cervical and nerve root canal stenosis, resulting in the compression of spinal cord and nerve tissues. Surgical decompression is the treatment choice. At present, there are 2 surgical procedures for the disease caused by disk herniation: anterior cervical discectomy and fusion (ACDF) and laminoplasty. The traditional posterior laminoplasty does not relieve foraminal stenosis causing nerve root compression, whereas in case of ACDF, the adjacent disk degeneration is accelerated.1 As Hilibrand et al2 reported, of all the 374 patients who underwent anterior cervical surgery, symptomatic adjacent segment disease occurred at a relatively constant incidence of 2.9% per year and that 25.6% of the patients had new disease at an adjacent level within 10 years after the operation. When we applied expansion open-door laminoplasty with foraminotomy (EOLF) to treat multisegment CMUR, we found satisfactory clinical outcomes and less complication.
MATERIALS AND METHODS Patient Series This was a single-centered prospective comparative study of 2 surgical procedures in treating CMUR. The study was approved by the hospital medical ethics committee. All patients were well explained about the study, and an informed consent was obtained from each one. www.clinicalspinesurgery.com |
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
E21
Clin Spine Surg
Fang et al
Inclusion criteria were as follows: (1) cervical disk herniation causing myelopathy; (2) compression of spinal cord at Z3 levels; (3) unilateral nerve root compression symptoms; and (4) >3 years of follow-up. Exclusion criteria were as follows: (1) patients with ossification of posterior longitudinal ligament, hypertrophy of flavum ligament, excessive cervical kyphosis above 15 degrees, and segmental instability; (2) previous history of cervical trauma or surgery; and (3) patients of bilateral nerve root compression symptoms. In this study, we also included the patients with small central calcification of disk. After a proper evaluation, all those cases were made sure that operative procedures could be performed. A total of 121 patients (62 cases of EOLF and 59 of ACDF) were included in the study with the mean age of 58.6 years. According to the imaging studies, spinal cord compression levels include C3/4–C5/6 in 37 cases, C4/5– C6/7 in 41 cases, and C3/4–C6/7 in 33 cases. The nerve root compression level include: C3/4 in 6 cases, C4/5 in 23 cases, C5/6 in 39 cases, C6/7 in 16 cases, C4/5–C5/6 in 19 cases, and C5/6–C6/7 in 18 cases.
Surgical Procedures After assessing preoperative signs and symptoms, and the imaging studies, the segment responsible for the compression of spinal cord was determined. On the basis of imaging findings, and patients radicular signs and symptoms, the foraminotomy level and side were decided. Every alternate year, 59 patients were enrolled in the ACDF group in 2004, 2006, and 2008 and 62 patients were enrolled in the EOLF group in 2005, 2007 and 2009. All procedures throughout this study were performed by a senior spinal surgeon and his team.
Anterior Cervical Discectomy Decompression and Cage Insertion With Titanium Plate and Screw Fixation After giving the general anesthesia, patient was kept in supine position with neck slightly extended. A right transverse paramedian incision was made at the level of anterior transverse stripe of the neck. After careful dissection of subcutaneous tissue and maintaining the hemostasis, the platysma muscle was exposed. Thereafter, the platysma was split to expose the inner edge of the right sternocleidomastoid. Sternocleidomastoid muscle was retracted laterally and then a blunt dissection was made. After this supraomohyoid muscle along with the carotid sheath pulled to the lateral side, and the trachea, esophagus with its sheath pulled medial side. Then prevertebral fascia was separated layer by layer under direct vision and prevertebral space exposed. The intervertebral space was located with the help of a needle under the guidance of image intensifier (C-arm). Then after cutting the anterior longitudinal ligament, intervertebral discectomy was performed at corresponding levels. With the help of Kerrison punch and nucleus rongeur, the margins of the gap cleared and the posterior longitudinal ligament (PLL) was excised. With the excision of PLL, epidural space was exposed and the decompression was confirmed. Then spaces were irrigated
E22 | www.clinicalspinesurgery.com
Volume 29, Number 1, February 2016
with sterile normal saline solution and appropriate-sized cages incorporated with bone allograft were placed, respectively. Small pieces of bone allograft were made and filled in the intervertebral spaces for the fusion. The vertebrae above and below the fusion level were fixed with the appropriate-sized titanium plate and screw. The positions of cage, plate, and screws were confirmed with the C-arm. At last, drainage tube was placed and the wound was closed in layers (Figs. 1A–D).
EOLF After administering general anesthesia, patient was placed in prone position with neck slightly flexed. The cervical spine was gently fixed with a Mayfield collar. Through a midline vertical incision, the spinous process, laminae, and facet joints were exposed from C3–C7. Then, after cutting the roots of spinous process, a hole was made in the spinous process. Using a high-speed drill, a trough was created on the “open” side at the junction between the laminae and the facets. The trough was drilled down, and a thin remnant of laminae was left. This thin rim and associated ligament were then removed with a 1-mm or 2-mm Kerrison punch. A second trough was then drilled in the opposite or “closed” side with a highspeed drill. Care was taken to only thin the lamina on this side and not cut all the way through. The opening on the open side was then gently expanded, thus lifting the lamina off the spinal cord and expanding the canal. The surgeon then gently expanded the opening with a Penfield dissector, while the assistant gently rotated the laminae toward the closed side using some appropriate instruments. Care was taken to not allow the block of laminae to slip and rapidly snap back into original position as this may result in spinal cord injury. The distance of opened door was 10 mm or the lamina angle was maintained at 45–60 degrees. Thereafter, the door was kept open by placing a suture through the facet capsule on the closed side and through the spinous process. After the side opening procedure, the stenosed foramen on the open side was selected. While cutting the posterior wall of the intervertebral foramen, medial edge of the upper and lower facet resection should be r50%. This resection removed the posterior part of the foramen, thus making nerve root decompressed. Then, small pieces of bone graft were embedded in the hinge side of the door to increase the stability. At last, drainage tube was placed after thorough irrigation and wound was closed in layers (Figs. 2A–D).
Postoperative Rehabilitation Postoperative care was almost same in both the groups. The drainage tube was routinely placed in all patients for 24–72 hours. Postoperative pain management, nutrition therapy, and some adjuvant therapy were equally emphasized. Cervical collar was provided for 4–6 weeks. From the second postoperative day, isometric muscle strengthening exercises were started under the guidance of the surgeon. Patients were usually discharged at postoperative day 7. Copyright
r
2013 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Clin Spine Surg
Volume 29, Number 1, February 2016
EOLF Versus ACDF in the Treatment of CMUR
FIGURE 1. A 48-year-old woman with B/L upper and lower extremities weakness for 3 years with increased numbness of right upper extremity since 1 week. A, Magnetic resonance imaging showing posterior intervertebral disk herniation at C3/4–C6/7 level compressing the spinal cord. B, Left oblique radiograph showing C5/6 foraminal narrowing. C, Postoperative radiograph showing C3/4, C4/5, and C5/6 intervertebral cage in situ with titanium plate and screw fixation. D, Postoperative magnetic resonance imaging showing the decompression of cord and restoring the spinal canal.
Evaluation Clinical Outcomes
The surgical complications that occurred, blood loss, and operation time were also recorded.
The Japanese Orthopedic Association (JOA) scoring system was used for the evaluation of cervical radiculomyelopathy before and after surgery. The recovery rate was calculated to compare preoperative and postoperative JOA scores. Each item of the recovery rate about each function was calculated as follows: recovery rate (%) = (postoperative individual JOA scorepreoperative individual JOA score) 100/(total individual JOA score preoperative individual JOA score). Copyright
r
2013 Wolters Kluwer Health, Inc. All rights reserved.
Radiologic Evaluation C2–C7 lordotic angle (a): cervical sagittal alignment was measured by tangential lines on the posterior edge of C2 and C7 bodies on lateral radiograph in a neutral position according to the Harrison posterior tangent method.3 Total range of motion (ROM) of the cervical spine was obtained by summation of the angles on flexion and www.clinicalspinesurgery.com |
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
E23
Clin Spine Surg
Fang et al
Volume 29, Number 1, February 2016
FIGURE 2. A 62-year-old man with B/L lower extremities numbness and weakness for 3 years with increased left upper extremity weakness and pain for 15 days. A, Magnetic resonance image scans showing multisegment (C2–C7) posterior disk herniation causing spinal cord compression and spinal canal narrowing. B, Postoperative computed tomography scan showing single opendoor (left to right) laminoplasty with left foraminotomy at C5/6 level. C, Postoperative magnetic resonance imaging showing decompression of spinal cord and restoration of spinal canal. D, Postoperative 3-dimensional computed tomography scan showing left C5/6 foraminotomy.
extension lateral radiographs before surgery and at the time of annual visit. Angles were created by way of cervical sagittal alignment and tangential lines on the posterior edge of C2 and C7 bodies on lateral radiograph in a neutral position (a), and on flexion (g) and extension lateral radiographs (b). Percentage of ROM (ROM%) was defined as 3 y postoperative ROM/preoperative ROM 100. The radiographic measurements were performed by 2 independent surgeons who were blinded about the clinical status of the patients (Fig. 3).
E24 | www.clinicalspinesurgery.com
Statistical Analysis The t test was used in the statistical analysis with applying SPSS19.0 statistical software. All P values
