- Email: [email protected]
Durotomy is associated with pseudoarthrosis following lumbar fusion
Journal of Clinical Neuroscience 22 (2015) 544–548
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Durotomy is associated with pseudoarthrosis following lumbar fusion Mohamad Bydon a,b,1, Rafael De la Garza-Ramos a,b,1, Nicholas B. Abt a,b, Mohamed Macki a,b, Daniel M. Sciubba a,b, Jean-Paul Wolinsky a,b, Ali Bydon a,b, Ziya L. Gokaslan a,b, Timothy F. Witham a,b,⇑ a b
The Spinal Column Biomechanics and Surgical Outcomes Laboratory, Johns Hopkins University School of Medicine, Baltimore, MD, USA Department of Neurosurgery, Johns Hopkins University School of Medicine, 600 North Wolfe Street, Meyer 5-109, Baltimore, MD 21287, USA
a r t i c l e
i n f o
Article history: Received 2 July 2014 Accepted 1 August 2014
Keywords: Cerebrospinal fluid leak Complications Degenerative spine disease Fusion Instrumentation Lumbar Pseudoarthrosis
a b s t r a c t Pseudoarthrosis is a known complication following lumbar fusion, and although several risk factors have been established, the association of durotomy and pseudoarthrosis has not been studied to our knowledge. A retrospective review was performed to identify all adult patients who underwent lumbar posterolateral fusion (without interbody fusion) for degenerative spine disease over a 20 year period at a single institution. Patients were divided into durotomy and no durotomy cohorts. Patients were included if they had at least 1 year of follow-up. The main outcome variable was development of pseudoarthrosis. A total of 327 patients were identified, of whom 17 (5.19%) had a durotomy. Pseudoarthrosis rates were significantly higher in the durotomy group (35.29%) when compared to the no durotomy group (13.87%), with the difference being statistically significant (p = 0.016). Univariate analysis revealed that durotomy (p = 0.003) and the number of levels fused (p = 0.015) were the only two significant risk factors for pseudoarthrosis. After controlling for the number of levels fused, the adjusted relative risk (RR) revealed that patients with a durotomy were 2.23 times more likely to develop pseudoarthrosis (RR 2.23; 95% confidence interval 1.05–4.75) when compared to patients without durotomy. The findings in the present study suggest an association between durotomy and pseudoarthrosis development. Patients with a durotomy were 2.2 times more likely to develop pseudoarthrosis compared to patients without a durotomy. Future and larger studies are required to corroborate our findings. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Pseudoarthrosis is a known complication following instrumentation of the lumbar spine. Although the use of instrumentation has increased fusion rates, approximately 23.6% of lumbar revision surgeries are due to pseudoarthrosis [1]. Numerous risk factors have been implicated in the development of pseudoarthrosis, including excessive motion at the fusion site, infection and smoking [2]. The purpose of this study is to evaluate whether a durotomy following lumbar posterolateral fusion (PLF) is associated with development of pseudoarthrosis. 2. Methods Medical records were retrospectively reviewed from 1990– 2011 to identify all adult patients who underwent instrumented PLF of the lumbar spine for the treatment of degenerative spine ⇑ Corresponding author. Tel.: +1 443 287 4934; fax: +1 410 502 3399 1
E-mail address: [email protected] (T.F. Witham). These authors have contributed equally to the manuscript.
http://dx.doi.org/10.1016/j.jocn.2014.08.023 0967-5868/Ó 2014 Elsevier Ltd. All rights reserved.
disease at our institution. Patients who had operations for traumatic, infectious or neoplastic causes were excluded, as were patients with less than 1 year of follow-up. Patients who underwent interbody fusion were also excluded. Demographic variables such as age, comorbidities (coronary artery disease, diabetes, hypertension, osteoporosis, obesity, chronic obstructive pulmonary disease, depression and smoking), pre-operative diagnoses and initial presenting symptoms were documented for all patients. Intra- and peri-operative variables, such as number of levels fused, use of bone morphogenetic protein (BMP), use of autograft and/or allograft, and estimated blood loss were collected from operative notes. Length of stay, complications, development of pseudoarthrosis and symptoms at last follow-up were assessed via follow-up clinical notes. All patients were instrumented with transpedicular titanium screws and rods. Patients were divided into two cohorts: durotomy and no durotomy. The durotomy cohort included both patients with (1) an incidental durotomy with intra-operative closure and no postoperative cerebrospinal fluid (CSF) leak, and (2) post-operative CSF leak out of the surgical wound. The main outcome variable in the present study was development of pseudoarthrosis requiring
M. Bydon et al. / Journal of Clinical Neuroscience 22 (2015) 544–548
re-operation, which was defined as any of the following: (1) absence of bridging bone across the fusion sites; (2) radiolucent areas around any screw site on CT scans; or (3) abnormal movement between vertebral bodies on dynamic (flexion–extension) radiographs. These were assessed via a combination of radiographs and CT scans and were independently assessed by a radiologist.
545
A descriptive statistical analysis was performed to compare the two groups via Student’s t-test and chi-squared test for continuous and non-continuous data, respectively. Data are presented as means ± standard deviation when applicable. A univariate analysis was performed to identify any significant factors associated with pseudoarthrosis (including patient age, sex, comorbidities, intraoperative variables such as BMP/allograft/autograft use and presence of laminectomy). Afterwards, factors with a p value < 0.05 were included in a multivariate regression model to analyze the outcome in the form of an adjusted relative risk (RR) and 95% confidence interval (CI). Statistical analyses were performed using STATA 12 SE (StataCorp LP, College Station, TX, USA). Statistical significance was set at p < 0.05.
proportion of patients in the durotomy group were males (70.59%; p = 0.043). The most common comorbidities were hypertension in 114 (34.86%) patients, followed by smoking in 55 (16.82%) patients; the rates of comorbidities between sub-groups were not significantly different except for chronic obstructive pulmonary disease, which was more common in the durotomy group. In the no durotomy group, 33.5% of patients had history of previous spinal surgery, compared to 52.9% of patients in the durotomy cohort (p = 0.102). However, a significantly higher proportion of patients in the durotomy group had history of previous laminectomy (47.06%) compared to only 21.3% in the no durotomy group (p = 0.013). Presenting symptoms were not significantly different between groups, with the most common symptom being back pain in 292 (89.30%) of patients, followed by radiculopathy in 231 (70.64%). Intra-operatively, the mean number of levels fused was 1.9 ± 0.9 in patients without a durotomy (Table 2), compared to 2.7 ± 1.1 in the durotomy group (p = 0.002). A similar proportion of patients in both groups underwent laminectomy (87.4% in patients without a durotomy versus 94.12% in the durotomy group; p = 0.412). The rate of BMP use between cohorts was not significantly different, and neither was use of local/iliac crest autograft or bone allograft.
3. Results
3.2. Outcomes
3.1. Demographics and surgical variables
Estimated blood loss was not significantly different between groups, but patients in the durotomy group stayed for a significantly longer time in the hospital, with an average of 8.4 days (Table 3). Overall post-operative complications were also similar between the groups, with the most common being wound infection in 13 (4.19%) patients without a durotomy and in two (11.76%) patients in the durotomy group. A significantly higher number of patients in the durotomy group developed a post-operative hematoma (11.76%; p < 0.001). Of the 17 patients with a durotomy, 11 had primary closure of the durotomy with no post-operative sequelae. The remaining six continued to have CSF leakage post-operatively: two (33%) had a post-operative pseudomeningocele that required re-operation, and four (66%) had CSF leaking from the wound (requiring stitches and strict bed rest). Among the six patients with post-operative CSF leak, three (50%) developed pseudoarthrosis. Of the remaining 11 patients, three (27%) developed pseudoarthrosis. Within the total durotomy cohort of 17 patients, six (35.29%) developed pseudoarthrosis, compared to 13.87% of patients without a durotomy (p = 0.016). Pseudoarthrosis rates were higher in the durotomy group (35.29%) when compared to the no durotomy group (13.87%), with the difference being statistically significant (p = 0.016). Univariate analysis revealed that durotomy (p = 0.003) and the number of levels fused (p = 0.015) were the only two significant risk factors for pseudoarthrosis. General characteristics of the six patients who developed pseudoarthrosis are summarized in Table 4. The unadjusted RR of developing pseudoarthrosis if a patient had a durotomy was 2.54 (95% CI 1.26–5.13; p = 0.009). Following this, after controlling for the number of levels fused, the adjusted RR revealed that patients with a durotomy were 2.23 times more likely to develop pseudoarthrosis (RR 2.23; 95% CI 1.05–4.75; p = 0.036) when compared to patients without a durotomy. Symptom resolution at last follow-up was similar between groups, and the average follow-up time for all patients was 57.25 ± 48.79 months (range, 12–240 months). Demographic and intra-operative variables, as well as outcomes of patients with and without pseudoarthrosis are summarized in Table 5–7. Demographic variables were not significantly different between these cohorts (Table 5), except for preoperative diagnoses. Intra-operative variables such as BMP use or bone graft material were also not significantly different (Table 6).
2.1. Statistical analysis
A total of 327 patients were identified with 17 (5.20%) in the durotomy group (Table 1). Over the 21 year period, 13 different surgeons performed these operations; the 17 patients with a durotomy were operated on by nine different surgeons and all 13 surgeons had at least one case of pseudoarthrosis. The average age for all patients was 59.79 ± 12.34 years and a significantly higher
Table 1 Demographic characteristics of all patients undergoing posterolateral fusion with no interbody fusion based on durotomy Characteristic
No durotomy
Durotomy
Patients, n Age, mean ± SD Sex (male, %)
310 59.73 ± 12.35 141 (45.48)
17 60.84 ± 12.59 12 (70.59)
0.639 0.043
38 (12.26) 38 (12.26) 110 (35.48) 12 (3.87) 25 (8.06) 1 (0.32) 33 (10.65) 52 (16.77)
4 (23.53) 3 (17.65) 4 (23.53) 1 (5.88) 0 (0.00) 1 (5.88) 1 (5.88) 3 (17.65)
0.176 0.514 0.314 0.679 0.223 0.004 0.531 0.925
104 (33.55) 66 (21.29) 38 (12.26)
9 (52.94) 8 (47.06) 1 (5.88)
0.102 0.013 0.430
Diagnosis Spinal stenosis (%) Spondylolisthesis (%) DDD (%)
206 (66.45) 70 (22.58) 34 (10.97)
5 (29.41) 8 (47.06) 4 (23.53)
0.002 0.021 0.116
Presenting symptom Back pain (%) Radiculopathy (%) Motor deficit (%) Sensory deficit (%) Bowel/bladder dysfunction (%)
276 (89.03) 220 (70.97) 36 (11.61) 38 (12.26) 17 (5.48)
16 (94.12) 11 (64.71) 4 (23.53) 2 (11.76) 1 (5.88)
0.509 0.581 0.144 0.952 0.944
Comorbidities CAD (%) Diabetes (%) Hypertension (%) Osteoporosis (%) Obesity (%) COPD (%) Depression (%) Smoking (%) Previous spinal surgery (%) Previous decompression (%) Previous decompression + fusion (%)
p value
CAD = coronary artery disease, COPD = chronic obstructive pulmonary disease, DDD = degenerative disc disease, SD = standard deviation. Bold denotes statistically significant results.
546
M. Bydon et al. / Journal of Clinical Neuroscience 22 (2015) 544–548
Table 2 Intra-operative characteristics of all patients undergoing posterolateral fusion with no interbody fusion based on occurrence of durotomy Characteristic
No durotomy
Durotomy
310 1.9 ± 0.9 271 (87.42) 144 (46.45) 264 (85.16) 237 (76.56) 49 (15.81) 169 (54.52)
17 2.7 ± 1.1 16 (94.12) 6 (35.29) 14 (82.35) 13 (76.47) 5 (29.41) 12 (66.67)
Patients, n Levels fused (mean ± SD) Laminectomy (%) BMP (%) Bone autograft (%) Local autograft (%) Iliac crest autograft (%) Bone allograft (%)
p value 0.002 0.412 0.369 0.752
0.314
BMP = bone morphogenetic protein, SD = standard deviation. Bold denotes statistically significant results.
Table 3 Outcomes of all patients undergoing posterolateral fusion based on occurrence of durotomy Characteristic
No durotomy
Patients, n Estimated blood loss, ml (mean ± SD) Length of stay (mean ± SD)
Durotomy
310 17 764.01 ± 770.36 983.33 ± 652.43
0.751
5.76 ± 4.41
8.37 ± 5.64
0.024
20 (6.45) 13 (4.19) 1 (0.32) 2 (0.65) 1 (0.32) 3 (0.97) 1 (0.32) 0 (0.00) 43 (13.87)
2 (11.76) 2 (11.76) 0 (0.00) 0 (0.00) 2 (11.76) 0 (0.00) 0 (0.00) 0 (0.00) 6 (35.29)
0.395 0.146 0.815 0.740 <0.001 0.684 0.815 0.016
155 (50.00) 101 (32.58) 19 (6.13) 18 (5.81) 8 (2.58) 101 (33.22) 58.11 ± 49.30
7 (41.18) 8 (47.06) 1 (5.88) 2 (11.76) 0 (0.00) 6 (35.29) 41.50 ± 35.94
0.479 0.218 0.967 0.318 0.502 0.860 0.086
Post-operative complication Wound infection (%) DVT (%) Pneumonia (%) Hematoma (%) Wound dehiscence (%) Myocardial infarction (%) Death (%) Pseudoarthrosis (%) Symptoms at last follow-up Back pain (%) Radiculopathy (%) Motor deficit (%) Sensory deficit (%) Bowel/bladder dysfunction (%) Re-operation Follow-up, months (mean ± SD)
p value
Table 5 Demographics of all patients undergoing posterolateral fusion with no interbody fusion based on pseudoarthrosis development Characteristic
No Pseudoarthrosis pseudoarthrosis
Patients, n Age (mean ± SD) Sex (male, %)
p value
278 60.33 ± 12.29 124 (44.60)
49 56.71 ± 12.34 29 (59.18)
0.101 0.059
32 (11.51) 34 (12.23) 99 (35.61) 12 (4.32) 21 (7.55) 2 (0.72) 28 (10.07) 43 (15.47)
10 (20.41) 7 (14.29) 15 (30.61) 1 (2.04) 4 (8.16) 0 (0.00) 6 (12.24) 12 (24.49)
0.086 0.689 0.498 0.452 0.882 0.551 0.646 0.120
94 (33.81) 63 (22.66) 31 (11.15)
19 (38.78) 11 (22.45) 8 (16.33)
0.501 0.974 0.303
Diagnosis Spinal stenosis (%) Spondylolisthesis (%) DDD (%)
190 (68.35) 65 (23.38) 23 (8.27)
21 (42.86) 13 (26.53) 15 (30.61)
0.001 0.633 <0.001
Presenting symptom Back pain (%) Radiculopathy (%) Motor deficit (%) Sensory deficit (%) Bowel/bladder dysfunction (%)
251 (90.29) 204 (73.38) 33 (11.87) 37 (13.31) 16 (5.76)
41 (83.67) 27 (55.10) 7 (14.29) 3 (6.12) 2 (4.08)
0.167 0.010 0.634 0.157 0.636
Comorbidities CAD (%) Diabetes (%) Hypertension (%) Osteoporosis (%) Obesity (%) COPD (%) Depression (%) Smoking (%) Previous spinal surgery (%) Previous decompression (%) Previous decompression + fusion (%)
CAD = coronary artery disease, COPD = chronic obstructive pulmonary disease, DDD = degenerative disc disease, SD = standard deviation. Bold denotes statistically significant results.
4. Discussion
DVT = deep vein thrombosis, SD = standard deviation. Bold denotes statistically significant results.
Regarding outcomes, patients in the pseudoarthrosis cohort stayed a significantly longer time in hospital (7.5 days versus 5.7 days for patients without pseudoarthrosis; p = 0.048) and also had higher rates of wound dehiscence (4.08 versus 0.36; p = 0.012) (Table 7). Overall, patients with pseudoarthrosis had higher rates of re-operation (85.1%) compared to patients without pseudoarthrosis (23.81%; p < 0.001).
The findings in the present study suggest an association between durotomy and the development of pseudoarthrosis following PLF of the lumbar spine. To the best of our knowledge, this association has not been studied before, so results must be interpreted with caution. Pseudoarthrosis rates vary depending on fusion techniques and site, but following PLF of the lumbar spine, it is estimated to occur in 16% of patients [3]. There exist several well-established known risk factors for development of pseudoarthrosis, among them tobacco smoking, infection, length of fusion and type of fusion [2,4]. Brown et al. reported a 40% rate of pseudoarthrosis in smokers versus 8% in non-smokers undergoing lumbar fusion [5]. Similarly, Wetzel et al. found that smoking decreased the likelihood of achieving solid fusions [6]. When studying the effect of post-operative wound infections on pseudoarthrosis rates, Weiss et al. found an arthrodesis rate of 62.1% following lumbar fusion [7]. Increasing fusion lengths are also associated with higher pseudoarthrosis
Table 4 Characteristics of patients with a durotomy who developed pseudoarthrosis Patient 1 2 3 4 5 6
Age, sex 62, 35, 40, 66, 59, 83,
Male Male Male Male Female Male
CAD = coronary artery disease, DM = diabetes mellitus.
Comorbidities
Number of levels fused
CAD, smoking – – DM, smoking, hypertension – Hypertension
4 2 3 2 3 1
Assessed by CT CT CT CT CT CT
scan scan scan, radiograph scan, radiograph scan scan, radiograph
M. Bydon et al. / Journal of Clinical Neuroscience 22 (2015) 544–548 Table 6 Intra-operative characteristics of all patients undergoing posterolateral fusion with no interbody fusion based on pseudoarthrosis development Characteristic Patients, n Levels fused (mean ± SD) Laminectomy (%) BMP (%) Bone autograft (%) Local autograft (%) Iliac crest autograft (%) Bone allograft (%)
No pseudoarthrosis
Pseudoarthrosis
278 1.9 ± 0.9 244 (87.77) 138 (49.64) 242 (87.05) 220 (79.14) 40 (14.39) 152 (54.68)
49 2.2 ± 0.9 43 (87.76) 20 (40.81) 36 (73.46) 30 (61.22) 14 (28.57) 29 (59.18)
p value 0.084 0.998 0.218 0.310
4.1. Limitations
Table 7 Outcomes of all patients undergoing posterolateral fusion based on pseudoarthrosis development
Patients, n Estimated blood loss, ml (mean ± SD) Length of stay (mean ± SD) Post-operative complication Wound infection (%) DVT (%) Pneumonia (%) Hematoma (%) Wound dehiscence (%) Myocardial infarction (%) Death (%) Symptoms at last follow-up Back pain (%) Radiculopathy (%) Motor deficit (%) Sensory deficit (%) Bowel/bladder dysfunction (%) Re-operation (%) Follow-up, months (mean ± SD)
No Pseudoarthrosis pseudoarthrosis
p value
278 783.18 ± 501
49 693.75 ± 200.7
0.375
5.72 ± 4.47
7.48 ± 4.66
0.048
11 (3.96) 1 (0.36) 2 (0.72) 2 (0.72) 1 (0.36) 1 (0.36) 0 (0.00) 127 (45.68) 87 (31.29) 13 (4.68) 15 (5.40) 7 (2.52) 65 (23.81) 53.5 ± 43.3
4 0 0 1 2 0 0
(8.16) (0.00) (0.00) (2.04) (4.08) (0.00) (0.00)
35 (71.43) 22 (44.90) 7 (14.29) 5 (10.20) 1 (2.04) 40 (85.11) 78.7 ± 60.1
osteoinductive mediators or wash away the bone graft from the fusion site. There may also be elements in CSF that inhibit bone formation. For example, the acidic pH of CSF (mean pH of 7.3) [13,14] may negatively impact osteogenesis by stimulating osteoclasts [15] and inhibiting osteoblasts [16]. Additionally, one must consider the possibility that an intra-operative durotomy adds a challenge to the placement of bone graft due to the presence of CSF or excessive epidural bleeding washing out the bone graft.
0.558
BMP = bone morphogenetic protein, SD = standard deviation.
Characteristic
547
0.194 0.674 0.551 0.371 0.012 0.674 0.001 0.063 0.010 0.195 0.842 <0.001 <0.001
DVT = deep vein thrombosis, SD = standard deviation. Bold denotes statistically significant results.
rates; in a study of long-segment fusions to the sacrum (average of 11 vertebrae), the pseudoarthrosis rate was 24% [8]. Additionally, circumferential fusions achieve fusion rates of 91%, compared to posterolateral fusion alone (85%) or posterior interbody fusions (89%) [9]. CSF leaks are estimated to occur in 9.0% of patients undergoing open lumbar foraminotomy, discectomy or laminectomy [10]. Patients with CSF leak have increased risk of post-operative postural headaches, development of meningitis or a pseudomeningocele. Additionally, these patients may require additional surgeries [11,12], leading to prolonged flat bed rest, placement of a lumbar drain, or revision surgery for dural repair [10]. In the present study, patients with a durotomy had a significantly higher risk of developing pseudoarthrosis. Although not a directly proven causality, the association still persisted after controlling for possible confounding variables, such as the number of levels fused. Other potential confounders may exist, but in the present study variables such as comorbidities, previous surgery, and bone graft material were not found to be significantly associated with pseudoarthrosis on univariate analysis, and were therefore not included in the stepwise multivariate model. While the association between durotomy and pseudoarthrosis has not been previously reported to our knowledge, there are possible pathophysiological mechanisms that may explain this phenomenon. For example, CSF may dilute the necessary
The foremost limitation of this study is the retrospective nature and the likelihood of confounding by both measured and unmeasured factors. Given the small sample size of patients with a durotomy and the smaller group of patients who had both a durotomy and pseudoarthrosis, there was a limited ability to evaluate potential confounders (for example, smoking, age, infection) in the multivariate analysis. Another limitation is that incidental durotomies may be under-reported by surgeons, which may explain the lower number of patients in this cohort. In the present study, the variation in surgical techniques among surgeons might have also influenced the development of a durotomy. This cannot be excluded as a possible confounder. An additional confounder may be that suboptimal technique led to both the durotomy and improper bone graft placement. We encourage further study of this association with larger prospective cohort studies. 5. Conclusion The findings in the present study suggest an association between durotomy and pseudoarthrosis following posterolateral fusion of the lumbar spine. Pseudoarthrosis rates were significantly higher in the durotomy group (35.29%) when compared to the no durotomy group (13.87%), with the difference being statistically significant (p = 0.016). Patients with a durotomy were 2.23 times more likely to develop pseudoarthrosis compared to patients without a durotomy, a statistically significant finding. To the best of our knowledge, this association has not been previously reported. Larger prospective studies are required before establishing a definitive conclusion. Conflicts of Interest/Disclosures Timothy F. Witham is the recipient of a research grant from Eli Lilly and Co and The Gordon and Marylin Macklin Foundation. Ali Bydon is the recipient of a research grant from Depuy Spine. He serves on the clinical advisory board of MedImmune, LLC. Ziya Gokaslan is the recipient of research grants from Depuy Spine, AO Spine North America, Medtronic, NREF, Integra Life Sciences, and K2M. He receives fellowship support from AO Spine North America. He holds stock in Spinal Kinetics and US Spine. Daniel M. Sciubba is the recipient of a research grant from DePuy Spine. He has consulting relationships with Medtronic, Nuvasive, Globus and DePuy. The other authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. References [1] Martin BI, Mirza SK, Comstock BA, et al. Reoperation rates following lumbar spine surgery and the influence of spinal fusion procedures. Spine (Phila Pa 1976) 2007;32:382–7. [2] Raizman NM, O’Brien JR, Poehling-Monaghan KL, et al. Pseudarthrosis of the spine. J Am Acad Orthop Surg 2009;17:494–503. [3] Dimar 2nd JR, Glassman SD, Burkus JK, et al. Two-year fusion and clinical outcomes in 224 patients treated with a single-level instrumented posterolateral fusion with iliac crest bone graft. Spine J 2009;9:880–5.
548
M. Bydon et al. / Journal of Clinical Neuroscience 22 (2015) 544–548
[4] Kim YJ, Bridwell KH, Lenke LG, et al. Pseudarthrosis in primary fusions for adult idiopathic scoliosis: incidence, risk factors, and outcome analysis. Spine (Phila Pa 1976) 2005;30:468–74. [5] Brown CW, Orme TJ, Richardson HD. The rate of pseudarthrosis (surgical nonunion) in patients who are smokers and patients who are nonsmokers: a comparison study. Spine (Phila Pa 1976) 1986;11:942–3. [6] Wetzel FT, Hoffman MA, Arcieri RR. Freeze-dried fibular allograft in anterior spinal surgery: cervical and lumbar applications. Yale J Biol Med 1993;66:263–75. [7] Weiss LE, Vaccaro AR, Scuderi G, et al. Pseudarthrosis after postoperative wound infection in the lumbar spine. J Spinal Disord 1997;10:482–7. [8] Kim YJ, Bridwell KH, Lenke LG, et al. Pseudarthrosis in long adult spinal deformity instrumentation and fusion to the sacrum: prevalence and risk factor analysis of 144 cases. Spine (Phila Pa 1976) 2006;31:2329–36. [9] Bono CM, Lee CK. Critical analysis of trends in fusion for degenerative disc disease over the past 20 years: influence of technique on fusion rate and clinical outcome. Spine (Phila Pa 1976) 2004;29:455–4563 [discussion Z5]. [10] Wong AP, Shih P, Smith TR, et al. Comparison of symptomatic cerebral spinal fluid leak between patients undergoing minimally invasive versus open
[11]
[12]
[13] [14] [15]
[16]
lumbar foraminotomy, discectomy, or laminectomy. World Neurosurg 2014;81:634–40. Jankowitz BT, Atteberry DS, Gerszten PC, et al. Effect of fibrin glue on the prevention of persistent cerebral spinal fluid leakage after incidental durotomy during lumbar spinal surgery. Eur Spine J 2009;18:1169–74. Khan MH, Rihn J, Steele G, et al. Postoperative management protocol for incidental dural tears during degenerative lumbar spine surgery: a review of 3183 consecutive degenerative lumbar cases. Spine (Phila Pa 1976) 2006;31:2609–13. Felgenhauer K. Protein size and cerebrospinal fluid composition. Klin Wochenschr 1974;52:1158–64. Greenberg MS. CSF constituents. Handbook of neurosurgery. New York, New York: Thieme; 2010. p. 298. Barzel US. The skeleton as an ion exchange system: implications for the role of acid-base imbalance in the genesis of osteoporosis. J Bone Miner Res 1995;10:1431–6. Arnett TR. Extracellular pH regulates bone cell function. J Nutr 2008;138:415S–8S.
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Durotomy is associated with pseudoarthrosis following lumbar fusion Mohamad Bydon a,b,1, Rafael De la Garza-Ramos a,b,1, Nicholas B. Abt a,b, Mohamed Macki a,b, Daniel M. Sciubba a,b, Jean-Paul Wolinsky a,b, Ali Bydon a,b, Ziya L. Gokaslan a,b, Timothy F. Witham a,b,⇑ a b
The Spinal Column Biomechanics and Surgical Outcomes Laboratory, Johns Hopkins University School of Medicine, Baltimore, MD, USA Department of Neurosurgery, Johns Hopkins University School of Medicine, 600 North Wolfe Street, Meyer 5-109, Baltimore, MD 21287, USA
a r t i c l e
i n f o
Article history: Received 2 July 2014 Accepted 1 August 2014
Keywords: Cerebrospinal fluid leak Complications Degenerative spine disease Fusion Instrumentation Lumbar Pseudoarthrosis
a b s t r a c t Pseudoarthrosis is a known complication following lumbar fusion, and although several risk factors have been established, the association of durotomy and pseudoarthrosis has not been studied to our knowledge. A retrospective review was performed to identify all adult patients who underwent lumbar posterolateral fusion (without interbody fusion) for degenerative spine disease over a 20 year period at a single institution. Patients were divided into durotomy and no durotomy cohorts. Patients were included if they had at least 1 year of follow-up. The main outcome variable was development of pseudoarthrosis. A total of 327 patients were identified, of whom 17 (5.19%) had a durotomy. Pseudoarthrosis rates were significantly higher in the durotomy group (35.29%) when compared to the no durotomy group (13.87%), with the difference being statistically significant (p = 0.016). Univariate analysis revealed that durotomy (p = 0.003) and the number of levels fused (p = 0.015) were the only two significant risk factors for pseudoarthrosis. After controlling for the number of levels fused, the adjusted relative risk (RR) revealed that patients with a durotomy were 2.23 times more likely to develop pseudoarthrosis (RR 2.23; 95% confidence interval 1.05–4.75) when compared to patients without durotomy. The findings in the present study suggest an association between durotomy and pseudoarthrosis development. Patients with a durotomy were 2.2 times more likely to develop pseudoarthrosis compared to patients without a durotomy. Future and larger studies are required to corroborate our findings. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Pseudoarthrosis is a known complication following instrumentation of the lumbar spine. Although the use of instrumentation has increased fusion rates, approximately 23.6% of lumbar revision surgeries are due to pseudoarthrosis [1]. Numerous risk factors have been implicated in the development of pseudoarthrosis, including excessive motion at the fusion site, infection and smoking [2]. The purpose of this study is to evaluate whether a durotomy following lumbar posterolateral fusion (PLF) is associated with development of pseudoarthrosis. 2. Methods Medical records were retrospectively reviewed from 1990– 2011 to identify all adult patients who underwent instrumented PLF of the lumbar spine for the treatment of degenerative spine ⇑ Corresponding author. Tel.: +1 443 287 4934; fax: +1 410 502 3399 1
E-mail address: [email protected] (T.F. Witham). These authors have contributed equally to the manuscript.
http://dx.doi.org/10.1016/j.jocn.2014.08.023 0967-5868/Ó 2014 Elsevier Ltd. All rights reserved.
disease at our institution. Patients who had operations for traumatic, infectious or neoplastic causes were excluded, as were patients with less than 1 year of follow-up. Patients who underwent interbody fusion were also excluded. Demographic variables such as age, comorbidities (coronary artery disease, diabetes, hypertension, osteoporosis, obesity, chronic obstructive pulmonary disease, depression and smoking), pre-operative diagnoses and initial presenting symptoms were documented for all patients. Intra- and peri-operative variables, such as number of levels fused, use of bone morphogenetic protein (BMP), use of autograft and/or allograft, and estimated blood loss were collected from operative notes. Length of stay, complications, development of pseudoarthrosis and symptoms at last follow-up were assessed via follow-up clinical notes. All patients were instrumented with transpedicular titanium screws and rods. Patients were divided into two cohorts: durotomy and no durotomy. The durotomy cohort included both patients with (1) an incidental durotomy with intra-operative closure and no postoperative cerebrospinal fluid (CSF) leak, and (2) post-operative CSF leak out of the surgical wound. The main outcome variable in the present study was development of pseudoarthrosis requiring
M. Bydon et al. / Journal of Clinical Neuroscience 22 (2015) 544–548
re-operation, which was defined as any of the following: (1) absence of bridging bone across the fusion sites; (2) radiolucent areas around any screw site on CT scans; or (3) abnormal movement between vertebral bodies on dynamic (flexion–extension) radiographs. These were assessed via a combination of radiographs and CT scans and were independently assessed by a radiologist.
545
A descriptive statistical analysis was performed to compare the two groups via Student’s t-test and chi-squared test for continuous and non-continuous data, respectively. Data are presented as means ± standard deviation when applicable. A univariate analysis was performed to identify any significant factors associated with pseudoarthrosis (including patient age, sex, comorbidities, intraoperative variables such as BMP/allograft/autograft use and presence of laminectomy). Afterwards, factors with a p value < 0.05 were included in a multivariate regression model to analyze the outcome in the form of an adjusted relative risk (RR) and 95% confidence interval (CI). Statistical analyses were performed using STATA 12 SE (StataCorp LP, College Station, TX, USA). Statistical significance was set at p < 0.05.
proportion of patients in the durotomy group were males (70.59%; p = 0.043). The most common comorbidities were hypertension in 114 (34.86%) patients, followed by smoking in 55 (16.82%) patients; the rates of comorbidities between sub-groups were not significantly different except for chronic obstructive pulmonary disease, which was more common in the durotomy group. In the no durotomy group, 33.5% of patients had history of previous spinal surgery, compared to 52.9% of patients in the durotomy cohort (p = 0.102). However, a significantly higher proportion of patients in the durotomy group had history of previous laminectomy (47.06%) compared to only 21.3% in the no durotomy group (p = 0.013). Presenting symptoms were not significantly different between groups, with the most common symptom being back pain in 292 (89.30%) of patients, followed by radiculopathy in 231 (70.64%). Intra-operatively, the mean number of levels fused was 1.9 ± 0.9 in patients without a durotomy (Table 2), compared to 2.7 ± 1.1 in the durotomy group (p = 0.002). A similar proportion of patients in both groups underwent laminectomy (87.4% in patients without a durotomy versus 94.12% in the durotomy group; p = 0.412). The rate of BMP use between cohorts was not significantly different, and neither was use of local/iliac crest autograft or bone allograft.
3. Results
3.2. Outcomes
3.1. Demographics and surgical variables
Estimated blood loss was not significantly different between groups, but patients in the durotomy group stayed for a significantly longer time in the hospital, with an average of 8.4 days (Table 3). Overall post-operative complications were also similar between the groups, with the most common being wound infection in 13 (4.19%) patients without a durotomy and in two (11.76%) patients in the durotomy group. A significantly higher number of patients in the durotomy group developed a post-operative hematoma (11.76%; p < 0.001). Of the 17 patients with a durotomy, 11 had primary closure of the durotomy with no post-operative sequelae. The remaining six continued to have CSF leakage post-operatively: two (33%) had a post-operative pseudomeningocele that required re-operation, and four (66%) had CSF leaking from the wound (requiring stitches and strict bed rest). Among the six patients with post-operative CSF leak, three (50%) developed pseudoarthrosis. Of the remaining 11 patients, three (27%) developed pseudoarthrosis. Within the total durotomy cohort of 17 patients, six (35.29%) developed pseudoarthrosis, compared to 13.87% of patients without a durotomy (p = 0.016). Pseudoarthrosis rates were higher in the durotomy group (35.29%) when compared to the no durotomy group (13.87%), with the difference being statistically significant (p = 0.016). Univariate analysis revealed that durotomy (p = 0.003) and the number of levels fused (p = 0.015) were the only two significant risk factors for pseudoarthrosis. General characteristics of the six patients who developed pseudoarthrosis are summarized in Table 4. The unadjusted RR of developing pseudoarthrosis if a patient had a durotomy was 2.54 (95% CI 1.26–5.13; p = 0.009). Following this, after controlling for the number of levels fused, the adjusted RR revealed that patients with a durotomy were 2.23 times more likely to develop pseudoarthrosis (RR 2.23; 95% CI 1.05–4.75; p = 0.036) when compared to patients without a durotomy. Symptom resolution at last follow-up was similar between groups, and the average follow-up time for all patients was 57.25 ± 48.79 months (range, 12–240 months). Demographic and intra-operative variables, as well as outcomes of patients with and without pseudoarthrosis are summarized in Table 5–7. Demographic variables were not significantly different between these cohorts (Table 5), except for preoperative diagnoses. Intra-operative variables such as BMP use or bone graft material were also not significantly different (Table 6).
2.1. Statistical analysis
A total of 327 patients were identified with 17 (5.20%) in the durotomy group (Table 1). Over the 21 year period, 13 different surgeons performed these operations; the 17 patients with a durotomy were operated on by nine different surgeons and all 13 surgeons had at least one case of pseudoarthrosis. The average age for all patients was 59.79 ± 12.34 years and a significantly higher
Table 1 Demographic characteristics of all patients undergoing posterolateral fusion with no interbody fusion based on durotomy Characteristic
No durotomy
Durotomy
Patients, n Age, mean ± SD Sex (male, %)
310 59.73 ± 12.35 141 (45.48)
17 60.84 ± 12.59 12 (70.59)
0.639 0.043
38 (12.26) 38 (12.26) 110 (35.48) 12 (3.87) 25 (8.06) 1 (0.32) 33 (10.65) 52 (16.77)
4 (23.53) 3 (17.65) 4 (23.53) 1 (5.88) 0 (0.00) 1 (5.88) 1 (5.88) 3 (17.65)
0.176 0.514 0.314 0.679 0.223 0.004 0.531 0.925
104 (33.55) 66 (21.29) 38 (12.26)
9 (52.94) 8 (47.06) 1 (5.88)
0.102 0.013 0.430
Diagnosis Spinal stenosis (%) Spondylolisthesis (%) DDD (%)
206 (66.45) 70 (22.58) 34 (10.97)
5 (29.41) 8 (47.06) 4 (23.53)
0.002 0.021 0.116
Presenting symptom Back pain (%) Radiculopathy (%) Motor deficit (%) Sensory deficit (%) Bowel/bladder dysfunction (%)
276 (89.03) 220 (70.97) 36 (11.61) 38 (12.26) 17 (5.48)
16 (94.12) 11 (64.71) 4 (23.53) 2 (11.76) 1 (5.88)
0.509 0.581 0.144 0.952 0.944
Comorbidities CAD (%) Diabetes (%) Hypertension (%) Osteoporosis (%) Obesity (%) COPD (%) Depression (%) Smoking (%) Previous spinal surgery (%) Previous decompression (%) Previous decompression + fusion (%)
p value
CAD = coronary artery disease, COPD = chronic obstructive pulmonary disease, DDD = degenerative disc disease, SD = standard deviation. Bold denotes statistically significant results.
546
M. Bydon et al. / Journal of Clinical Neuroscience 22 (2015) 544–548
Table 2 Intra-operative characteristics of all patients undergoing posterolateral fusion with no interbody fusion based on occurrence of durotomy Characteristic
No durotomy
Durotomy
310 1.9 ± 0.9 271 (87.42) 144 (46.45) 264 (85.16) 237 (76.56) 49 (15.81) 169 (54.52)
17 2.7 ± 1.1 16 (94.12) 6 (35.29) 14 (82.35) 13 (76.47) 5 (29.41) 12 (66.67)
Patients, n Levels fused (mean ± SD) Laminectomy (%) BMP (%) Bone autograft (%) Local autograft (%) Iliac crest autograft (%) Bone allograft (%)
p value 0.002 0.412 0.369 0.752
0.314
BMP = bone morphogenetic protein, SD = standard deviation. Bold denotes statistically significant results.
Table 3 Outcomes of all patients undergoing posterolateral fusion based on occurrence of durotomy Characteristic
No durotomy
Patients, n Estimated blood loss, ml (mean ± SD) Length of stay (mean ± SD)
Durotomy
310 17 764.01 ± 770.36 983.33 ± 652.43
0.751
5.76 ± 4.41
8.37 ± 5.64
0.024
20 (6.45) 13 (4.19) 1 (0.32) 2 (0.65) 1 (0.32) 3 (0.97) 1 (0.32) 0 (0.00) 43 (13.87)
2 (11.76) 2 (11.76) 0 (0.00) 0 (0.00) 2 (11.76) 0 (0.00) 0 (0.00) 0 (0.00) 6 (35.29)
0.395 0.146 0.815 0.740 <0.001 0.684 0.815 0.016
155 (50.00) 101 (32.58) 19 (6.13) 18 (5.81) 8 (2.58) 101 (33.22) 58.11 ± 49.30
7 (41.18) 8 (47.06) 1 (5.88) 2 (11.76) 0 (0.00) 6 (35.29) 41.50 ± 35.94
0.479 0.218 0.967 0.318 0.502 0.860 0.086
Post-operative complication Wound infection (%) DVT (%) Pneumonia (%) Hematoma (%) Wound dehiscence (%) Myocardial infarction (%) Death (%) Pseudoarthrosis (%) Symptoms at last follow-up Back pain (%) Radiculopathy (%) Motor deficit (%) Sensory deficit (%) Bowel/bladder dysfunction (%) Re-operation Follow-up, months (mean ± SD)
p value
Table 5 Demographics of all patients undergoing posterolateral fusion with no interbody fusion based on pseudoarthrosis development Characteristic
No Pseudoarthrosis pseudoarthrosis
Patients, n Age (mean ± SD) Sex (male, %)
p value
278 60.33 ± 12.29 124 (44.60)
49 56.71 ± 12.34 29 (59.18)
0.101 0.059
32 (11.51) 34 (12.23) 99 (35.61) 12 (4.32) 21 (7.55) 2 (0.72) 28 (10.07) 43 (15.47)
10 (20.41) 7 (14.29) 15 (30.61) 1 (2.04) 4 (8.16) 0 (0.00) 6 (12.24) 12 (24.49)
0.086 0.689 0.498 0.452 0.882 0.551 0.646 0.120
94 (33.81) 63 (22.66) 31 (11.15)
19 (38.78) 11 (22.45) 8 (16.33)
0.501 0.974 0.303
Diagnosis Spinal stenosis (%) Spondylolisthesis (%) DDD (%)
190 (68.35) 65 (23.38) 23 (8.27)
21 (42.86) 13 (26.53) 15 (30.61)
0.001 0.633 <0.001
Presenting symptom Back pain (%) Radiculopathy (%) Motor deficit (%) Sensory deficit (%) Bowel/bladder dysfunction (%)
251 (90.29) 204 (73.38) 33 (11.87) 37 (13.31) 16 (5.76)
41 (83.67) 27 (55.10) 7 (14.29) 3 (6.12) 2 (4.08)
0.167 0.010 0.634 0.157 0.636
Comorbidities CAD (%) Diabetes (%) Hypertension (%) Osteoporosis (%) Obesity (%) COPD (%) Depression (%) Smoking (%) Previous spinal surgery (%) Previous decompression (%) Previous decompression + fusion (%)
CAD = coronary artery disease, COPD = chronic obstructive pulmonary disease, DDD = degenerative disc disease, SD = standard deviation. Bold denotes statistically significant results.
4. Discussion
DVT = deep vein thrombosis, SD = standard deviation. Bold denotes statistically significant results.
Regarding outcomes, patients in the pseudoarthrosis cohort stayed a significantly longer time in hospital (7.5 days versus 5.7 days for patients without pseudoarthrosis; p = 0.048) and also had higher rates of wound dehiscence (4.08 versus 0.36; p = 0.012) (Table 7). Overall, patients with pseudoarthrosis had higher rates of re-operation (85.1%) compared to patients without pseudoarthrosis (23.81%; p < 0.001).
The findings in the present study suggest an association between durotomy and the development of pseudoarthrosis following PLF of the lumbar spine. To the best of our knowledge, this association has not been studied before, so results must be interpreted with caution. Pseudoarthrosis rates vary depending on fusion techniques and site, but following PLF of the lumbar spine, it is estimated to occur in 16% of patients [3]. There exist several well-established known risk factors for development of pseudoarthrosis, among them tobacco smoking, infection, length of fusion and type of fusion [2,4]. Brown et al. reported a 40% rate of pseudoarthrosis in smokers versus 8% in non-smokers undergoing lumbar fusion [5]. Similarly, Wetzel et al. found that smoking decreased the likelihood of achieving solid fusions [6]. When studying the effect of post-operative wound infections on pseudoarthrosis rates, Weiss et al. found an arthrodesis rate of 62.1% following lumbar fusion [7]. Increasing fusion lengths are also associated with higher pseudoarthrosis
Table 4 Characteristics of patients with a durotomy who developed pseudoarthrosis Patient 1 2 3 4 5 6
Age, sex 62, 35, 40, 66, 59, 83,
Male Male Male Male Female Male
CAD = coronary artery disease, DM = diabetes mellitus.
Comorbidities
Number of levels fused
CAD, smoking – – DM, smoking, hypertension – Hypertension
4 2 3 2 3 1
Assessed by CT CT CT CT CT CT
scan scan scan, radiograph scan, radiograph scan scan, radiograph
M. Bydon et al. / Journal of Clinical Neuroscience 22 (2015) 544–548 Table 6 Intra-operative characteristics of all patients undergoing posterolateral fusion with no interbody fusion based on pseudoarthrosis development Characteristic Patients, n Levels fused (mean ± SD) Laminectomy (%) BMP (%) Bone autograft (%) Local autograft (%) Iliac crest autograft (%) Bone allograft (%)
No pseudoarthrosis
Pseudoarthrosis
278 1.9 ± 0.9 244 (87.77) 138 (49.64) 242 (87.05) 220 (79.14) 40 (14.39) 152 (54.68)
49 2.2 ± 0.9 43 (87.76) 20 (40.81) 36 (73.46) 30 (61.22) 14 (28.57) 29 (59.18)
p value 0.084 0.998 0.218 0.310
4.1. Limitations
Table 7 Outcomes of all patients undergoing posterolateral fusion based on pseudoarthrosis development
Patients, n Estimated blood loss, ml (mean ± SD) Length of stay (mean ± SD) Post-operative complication Wound infection (%) DVT (%) Pneumonia (%) Hematoma (%) Wound dehiscence (%) Myocardial infarction (%) Death (%) Symptoms at last follow-up Back pain (%) Radiculopathy (%) Motor deficit (%) Sensory deficit (%) Bowel/bladder dysfunction (%) Re-operation (%) Follow-up, months (mean ± SD)
No Pseudoarthrosis pseudoarthrosis
p value
278 783.18 ± 501
49 693.75 ± 200.7
0.375
5.72 ± 4.47
7.48 ± 4.66
0.048
11 (3.96) 1 (0.36) 2 (0.72) 2 (0.72) 1 (0.36) 1 (0.36) 0 (0.00) 127 (45.68) 87 (31.29) 13 (4.68) 15 (5.40) 7 (2.52) 65 (23.81) 53.5 ± 43.3
4 0 0 1 2 0 0
(8.16) (0.00) (0.00) (2.04) (4.08) (0.00) (0.00)
35 (71.43) 22 (44.90) 7 (14.29) 5 (10.20) 1 (2.04) 40 (85.11) 78.7 ± 60.1
osteoinductive mediators or wash away the bone graft from the fusion site. There may also be elements in CSF that inhibit bone formation. For example, the acidic pH of CSF (mean pH of 7.3) [13,14] may negatively impact osteogenesis by stimulating osteoclasts [15] and inhibiting osteoblasts [16]. Additionally, one must consider the possibility that an intra-operative durotomy adds a challenge to the placement of bone graft due to the presence of CSF or excessive epidural bleeding washing out the bone graft.
0.558
BMP = bone morphogenetic protein, SD = standard deviation.
Characteristic
547
0.194 0.674 0.551 0.371 0.012 0.674 0.001 0.063 0.010 0.195 0.842 <0.001 <0.001
DVT = deep vein thrombosis, SD = standard deviation. Bold denotes statistically significant results.
rates; in a study of long-segment fusions to the sacrum (average of 11 vertebrae), the pseudoarthrosis rate was 24% [8]. Additionally, circumferential fusions achieve fusion rates of 91%, compared to posterolateral fusion alone (85%) or posterior interbody fusions (89%) [9]. CSF leaks are estimated to occur in 9.0% of patients undergoing open lumbar foraminotomy, discectomy or laminectomy [10]. Patients with CSF leak have increased risk of post-operative postural headaches, development of meningitis or a pseudomeningocele. Additionally, these patients may require additional surgeries [11,12], leading to prolonged flat bed rest, placement of a lumbar drain, or revision surgery for dural repair [10]. In the present study, patients with a durotomy had a significantly higher risk of developing pseudoarthrosis. Although not a directly proven causality, the association still persisted after controlling for possible confounding variables, such as the number of levels fused. Other potential confounders may exist, but in the present study variables such as comorbidities, previous surgery, and bone graft material were not found to be significantly associated with pseudoarthrosis on univariate analysis, and were therefore not included in the stepwise multivariate model. While the association between durotomy and pseudoarthrosis has not been previously reported to our knowledge, there are possible pathophysiological mechanisms that may explain this phenomenon. For example, CSF may dilute the necessary
The foremost limitation of this study is the retrospective nature and the likelihood of confounding by both measured and unmeasured factors. Given the small sample size of patients with a durotomy and the smaller group of patients who had both a durotomy and pseudoarthrosis, there was a limited ability to evaluate potential confounders (for example, smoking, age, infection) in the multivariate analysis. Another limitation is that incidental durotomies may be under-reported by surgeons, which may explain the lower number of patients in this cohort. In the present study, the variation in surgical techniques among surgeons might have also influenced the development of a durotomy. This cannot be excluded as a possible confounder. An additional confounder may be that suboptimal technique led to both the durotomy and improper bone graft placement. We encourage further study of this association with larger prospective cohort studies. 5. Conclusion The findings in the present study suggest an association between durotomy and pseudoarthrosis following posterolateral fusion of the lumbar spine. Pseudoarthrosis rates were significantly higher in the durotomy group (35.29%) when compared to the no durotomy group (13.87%), with the difference being statistically significant (p = 0.016). Patients with a durotomy were 2.23 times more likely to develop pseudoarthrosis compared to patients without a durotomy, a statistically significant finding. To the best of our knowledge, this association has not been previously reported. Larger prospective studies are required before establishing a definitive conclusion. Conflicts of Interest/Disclosures Timothy F. Witham is the recipient of a research grant from Eli Lilly and Co and The Gordon and Marylin Macklin Foundation. Ali Bydon is the recipient of a research grant from Depuy Spine. He serves on the clinical advisory board of MedImmune, LLC. Ziya Gokaslan is the recipient of research grants from Depuy Spine, AO Spine North America, Medtronic, NREF, Integra Life Sciences, and K2M. He receives fellowship support from AO Spine North America. He holds stock in Spinal Kinetics and US Spine. Daniel M. Sciubba is the recipient of a research grant from DePuy Spine. He has consulting relationships with Medtronic, Nuvasive, Globus and DePuy. The other authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. References [1] Martin BI, Mirza SK, Comstock BA, et al. Reoperation rates following lumbar spine surgery and the influence of spinal fusion procedures. Spine (Phila Pa 1976) 2007;32:382–7. [2] Raizman NM, O’Brien JR, Poehling-Monaghan KL, et al. Pseudarthrosis of the spine. J Am Acad Orthop Surg 2009;17:494–503. [3] Dimar 2nd JR, Glassman SD, Burkus JK, et al. Two-year fusion and clinical outcomes in 224 patients treated with a single-level instrumented posterolateral fusion with iliac crest bone graft. Spine J 2009;9:880–5.
548
M. Bydon et al. / Journal of Clinical Neuroscience 22 (2015) 544–548
[4] Kim YJ, Bridwell KH, Lenke LG, et al. Pseudarthrosis in primary fusions for adult idiopathic scoliosis: incidence, risk factors, and outcome analysis. Spine (Phila Pa 1976) 2005;30:468–74. [5] Brown CW, Orme TJ, Richardson HD. The rate of pseudarthrosis (surgical nonunion) in patients who are smokers and patients who are nonsmokers: a comparison study. Spine (Phila Pa 1976) 1986;11:942–3. [6] Wetzel FT, Hoffman MA, Arcieri RR. Freeze-dried fibular allograft in anterior spinal surgery: cervical and lumbar applications. Yale J Biol Med 1993;66:263–75. [7] Weiss LE, Vaccaro AR, Scuderi G, et al. Pseudarthrosis after postoperative wound infection in the lumbar spine. J Spinal Disord 1997;10:482–7. [8] Kim YJ, Bridwell KH, Lenke LG, et al. Pseudarthrosis in long adult spinal deformity instrumentation and fusion to the sacrum: prevalence and risk factor analysis of 144 cases. Spine (Phila Pa 1976) 2006;31:2329–36. [9] Bono CM, Lee CK. Critical analysis of trends in fusion for degenerative disc disease over the past 20 years: influence of technique on fusion rate and clinical outcome. Spine (Phila Pa 1976) 2004;29:455–4563 [discussion Z5]. [10] Wong AP, Shih P, Smith TR, et al. Comparison of symptomatic cerebral spinal fluid leak between patients undergoing minimally invasive versus open
[11]
[12]
[13] [14] [15]
[16]
lumbar foraminotomy, discectomy, or laminectomy. World Neurosurg 2014;81:634–40. Jankowitz BT, Atteberry DS, Gerszten PC, et al. Effect of fibrin glue on the prevention of persistent cerebral spinal fluid leakage after incidental durotomy during lumbar spinal surgery. Eur Spine J 2009;18:1169–74. Khan MH, Rihn J, Steele G, et al. Postoperative management protocol for incidental dural tears during degenerative lumbar spine surgery: a review of 3183 consecutive degenerative lumbar cases. Spine (Phila Pa 1976) 2006;31:2609–13. Felgenhauer K. Protein size and cerebrospinal fluid composition. Klin Wochenschr 1974;52:1158–64. Greenberg MS. CSF constituents. Handbook of neurosurgery. New York, New York: Thieme; 2010. p. 298. Barzel US. The skeleton as an ion exchange system: implications for the role of acid-base imbalance in the genesis of osteoporosis. J Bone Miner Res 1995;10:1431–6. Arnett TR. Extracellular pH regulates bone cell function. J Nutr 2008;138:415S–8S.