Factors influencing complications after 3-columns spinal osteotomies for fixed sagittal imbalance from multiple etiologies: a multicentric cohort study about 286 cases in 273 patients

doi:10.21203/rs.3.rs-1730895/v1

Download PDF

Research Article

Factors influencing complications after 3-columns spinal osteotomies for fixed sagittal imbalance from multiple etiologies: a multicentric cohort study about 286 cases in 273 patients

https://doi.org/10.21203/rs.3.rs-1730895/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Purpose: Spinal osteotomies performed to treat fixed spinal deformities are technically demanding and associated with a high complications rate. The main purpose of this study was to analyze complications and their risk factors in spinal osteotomies performed for fixed sagittal imbalance from multiple etiologies.

Methods: The study consisted of a blinded retrospective analysis of prospectively collected data from a large multicentre cohort of patients who underwent 3-columns (3C) spinal osteotomy, between january 2010 and january 2017. Clinical and radiological datas were compared pre and post operatively. Complications and their risks factors were analyzed.

Results: Two hundred eighty-six 3C osteotomies were performed in 273 patients. At 1year follow-up, both clinical (VAS pain, ODI and SRS-22 scores) and radiological (SVA, SSA, loss of lordosis and pelvic version) parameters were significantly improved (p<0.001). A total of 164 patients (59.2%) experienced at least 1 complication (277 complications). Complications-free survival rates was only 30% at 5 years. Most of those were mechanical (35.2%), followed by general (17.6%), surgical site infection (17.2%) and neurological (10.9%). Pre-operative neurological status (RR=2.3 [1.32-4.00]), operative time (+19% of risk each additional hour) and combined surgery (RR=1.76 [1.08-2.04]) were assessed as risk factors for overall complication (p<0.05). The use of Patient-specific rods appeared to be significantly associated with less overall complications (RR=0.5 [0.29-0.89]) (p=0.02).

Conclusion: Spinal 3C osteotomies were efficient to improve both clinical and radiological parameters despite high rates of complication. Efforts should be made to reduce operative time which appears to be the strongest predictive risk factor for complication.

Transpedicular osteotomy

fixed sagittal imbalance

outcome

complication

corrective surgery

spinal deformity.

Under normal conditions, the differents segments of the spine allow the head and the trunk to be positionned above the pelvis. Disturbance of the normal alignement of the spine usually refers to sagittal imbalance. This may be the result of iatrogenic flat-back syndrome, post traumatic kyphosis, ankylosing spondylitis or severe lumbar degeneration [1–3]. To stand erect, patients with rigid sagittal imbalance use compensatory knee flexion, hip extension and overwork of the extensor muscles of the spine [4]. Chronic back pain, muscular fatigue, walking disability and inability to look straight ahead in severe cases lead to significantly loss of overall quality of life, such as chronic diseases like diabetes, cancers or cardio-vascular diseases [5].

To correct fixed sagittal deformities, Thomasen described in 1985 the technique of Pedicle Subtraction Osteotomy (PSO) consisting of a posterior wedge shaped resection of the vertebral body, through the pedicles, with the hinge located anteriorly close to the anterior cortex [6]. This procedure results into a shortening of the posterior column, allowing up to 30–40° of sagittal correction per level. Furthermore, not lengthening the anterior column avoids stretching on the main vessels and viscera anterior to the spine and all the correction is done posteriorly. Since Thomasen's original description, many variants have been reported in the literature, and osteotomies are nowadays often classified into 6 grades of resection according to the extent of bone resection [7]. Finally, in the most severe spinal deformities requiring major correction in the coronal and/or sagittal plane, Vertebral Column Resection (VCR) may be the treatment of choice [8].

Despites their many advantages, PSO and VCR are technically demanding and are classically associated with a high risk of intra-operative (MEP/SSEP alterations, dural tears, massive bleeding,), peri operative (deep vein thrombosis, myocardial infarction, visual defect, pulmonary embolism) and post-operative (neurological deficits, wound infection, pseudarthrosis, rod breakage, junctional kyphosis) complications [3, 8–17]. The rate of perioperative morbidity can affect up to 70% of patients [18]. However, most of the studies focusing on spinal osteotomies and their complications are limited by several factors, including relatively small numbers of patients [1, 3, 11, 13, 19–21], single-center series [1, 9, 11, 12, 19, 21, 22], and the fact that they were conducted more than 10 years ago [1, 9–11, 20, 21]. Recent advances in understanding the sagittal balance, improvement of methods to stabilize the spine following an osteotomy and establishment of experienced teams should reduce the rate of complications associated with PSO at the present time and may also change their type.

In this context, the main purpose of this study was to analyse the complication rates associated with 3C spinal osteotomies for fixed sagittal imbalance through a retrospective analysis of a large prospective cohort follow up based on a multicenter study design and a standardized data collection form.

The present study is a BLINDED retrospective analysis of prospectively collected data from a multicenter cohort of patients who underwent 3-columns spinal osteotomy between january 2010 and january 2017. Seven BLINDED hospitals were involved in patient’s recruitment.

Inclusion criteria

Patients could be enrolled in our study if they were adults (≥ 18 years old), had undergone a 3-columns spinal osteotomy of grade ≥ 3 (according to Shwab et al [7]) between T1 and S1, had a follow-up of at least 1 year, and had performed pre operative full spine radiographies.

Investigators were well-trained in the use of the software KEOPS Spine® (Lyon, Smaio, France).

Exclusion criteria

Patients were exluded from the study if they suffered neurological degenerative disorders (Parkinson’s disease), were treated for an acute spine injury or an active infectious process, or underwent grades 1 or 2 osteotomy.

Clinical data

Demographic characteristics (sex, age, Body Mass Index (BMI)), American Society of Anesthesiologists (ASA) score, smoking status and etiologies of the fixed sagittal imbalance were recorded at baseline. Patients were clinically evaluated before surgery and during follow-up. Motor deficit was evaluated by the surgeon who performed the procedure and was classified using the American Spinal cord Injury Association (ASIA) score. Patients were also asked to fill questionnaires (Visual Analogue Scale (VAS) for lumbar and radiculair pain, Oswestry Disability Index (ODI) and Scoliosis Research Society-22 (SRS-22) score) and to assess for their walking distance.

Radiological data

The software KEOPS Spine® was used to collect radiographies and analyze sagittal parameters. All patients underwent full spine radiographies preoperatively, 12 months after surgery, and at last follow up. Pelvic incidence, pelvic tilt, sacral slope, sagittal vertical axis (SVA), spino-sacral angle (SSA), C7 ratio and T1 pelvic agle (TPA) were recorded. L1-S1 lordosis, lumbar lordosis and inflexion point were also evaluated. Lack of lordosis was calcuated using the following formula : theorical L1-S1 lordosis - mesured L1-S1 lordosis, with theorical L1-S1 lordosis being calculated according to the work by Barrey et al [23]. The osteotomy correction angle was calculated by measuring the difference between the pre and post-operative angle between the upper and lower endplates of the osteotomized vertebrae.

Surgical data and instrumentation

Mean operative time and mean operative blood loss were recorded. Cell-saver autotransfusion and somatosensory evoked potential monitoring were used in every posterior procedure. All surgical procedure were classified into grade 3, 4, 5 or 6 according to the work of Schwab et al [7]. Site of osteotomy was analyzed. Levels of instrumentation, material and rod diameter, utilization of additional rod(s) (3/4 rods construct), implantation of patient-specific pre-bended and pre-sized rod, use/nature of grafting and technique for osteotomy closure were collected. If an anterior procedure was performed, approach and levels fused were also noted.

Post operative management

The patient was generally raised up the first post-operative day and a brace was wearing for 4 months. Average length of hospital stay was also recorded.

Complications and mortality

Complications were defined as an unexpected event that appear once the surgical procedure was over. Data collected via the KEOPS software were classified as "yes" if a complication occurred or "no" if no complication occurred or if there was no response.

Complications were divided into one of the 4 following categories: general, mechanical, neurological or infectious. Regarding mechanical complications, proximal junction kyphosis (PJK) was defined as an angle > 10° between the last instrumented vertebrae and the first vertebre above the instrumentation.

The rates of mortality and reoperations were also recorded.

Statistical analysis

Categorical variables were expressed as number (n) and percentage.

Quantitative variables were expressed as means ± standard deviation when the distribution was normal or median with first and third quartile when the distribution was not normal.

The hypothesis of normal distribution of quantitative variables was tested using the ShapiroWilk test and graphically confirmed with a histogram.

Categorical variables were compared using the Chi 2 test or Fisher's exact test when the conditions of application of Chi square test were not met.

Quantitative variables were compared between independent groups using Student's t test after verification of equality of variances when data were normally distributed, and with the nonparametric test of Wilcoxon when the hypothesis of normality of distribution was not verified.

The comparison of functional and clinical data before and after surgery were compared using a Paired Student test when data were normally distributed, and with the nonparametric test of Rank signed Wilcoxon test when the hypothesis of normality of distribution was not verified.

The Complication free survival was defined by the delay between date of surgery and the date of first complication of any type or date of point fixed at March 26th, 2018.

Description of overall and complication-free survival was estimated by the Kaplan–Meier product limit method, and the effect of different parameters was assessed using the log rank test. Prognostic factors were assed using the semi-parametric Cox model after verification of the proportional hazard hypothesis with a univariate analysis. Due to methodological limitations and the study design, multivariate analyses were not performed.

The statistical tests are bilateral and the level of significance was set to 5% (p < 0.05).

Statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc, NC).

Ethical and Legal Considerations

The study was approved by the ethic comitee of the Hospices Civils de Lyon.

As data were collected in the standard of care, no consents were required.

A total of 286 spinal osteotomies were performed in the 273 patients enrolled in the study. Patients’ characteristics and etiologies for fixed sagital imbalance are given in table 1.

Table 1: Demographical characteristics and etiology of deformity

Patients characteristics	Total population, n=273
Sex (n, %)
Men	182 (66.7%)
Women	91 (33.3%)
Mean age at the time of surgery (years old)	63.1± 11.3 [23-81]
ASA Score (n, %)
ASA 1	36 (15.7%)^a
ASA 2	145 (63.0%)^a
ASA 3	48 (20.9%)^a
ASA 4	1 (0.4%)^a
Mean BMI (kg.m²)	26.4 ±4.9 [16-49].
Diabetus (n, %)
No	154 (92,2%)^b
Yes	13 (7,8%)^b
History of spinal surgery with instrumentation (n, %)
No	257 (94.1%)^c
Yes	16 (5.9%)^c
Etiologies of fixed sagittal imbalance (n, %)
Lumbar degenerative kyphosis	102 (43.8%)^d
Post operative iatrogenic flat-back	73 (31.3%)^d
Multifactorial	21 (9.0%)^d
Post traumatic kyphosis	20 (8.6%)^d
Others (Ankylosis spondylis, Proximal junction kyphosis)	17 (7.3%)^d

ASA : American Society of Anesthesiologists, BMI : Body Mass Index

^a percentages among n=230 ; ^b among n=167; ^c among n=273; ^d among n=233

Clinical parameters at baseline

Most patients (63%) suffered from neurogenic claudication, with most of them being unable to walk more than 200 meters (69.2%). No neurological deficit (ASIA score E) were found in most pateients (79.1%), whereas 0.9% presented with mild or moderate deficit (ASIA score C or D). Mean pre-operative scores are given in Table 2.

Table 2

Comparison of clinical and radiological characteristics at baseline, 6 months, 1 year, 2 years and at last follow-up
Clinical characteristics (mean ± SD)	Pre-operative	6 months	1 year	2 years	last follow up	p*
VAS for lumbar pain	7.2 ± 1.9	3.6 ± 2.3	3,3 ± 2.2	4.4 ± 2.6	3.8 ± 2.5	< 0.0001
VAS for radicular pain	4.2 ± 3.6	2.4 ± 2.7	2.2 ± 2.7	2.3 ± 2.9	2.1 ± 2.7	< 0.0001
ODI	55.6 ± 16.6	38.4 ± 18.8	34.9 ± 17.6	36.3 ± 19.2	39.3 ± 20.5	< 0.0001
SRS-22	2.3 ± 0.5	3.2 ± 0.7	3.4 ± 0.7	3.3 ± 0.6	3.2 ± 0.7	< 0.001
Radiological characteristics (mean ± SD)
SVA	121.6 ± 129.3	65.9 ± 85.3	58.2 ± 66.3	75.6 ± 82.5	68.7 ± 82.0	< 0.0001
Loss of lordosis	-32.2 ± 20.9	-15.3 ± 13.1	-16.6 ± 21.6	-16.7 ± 13.3	-16.7 ± 18.5	< 0.0001
Pelvic tilt	29.9 ± 11.4	24.6 ± 9.8	25.5 ± 10.4	27.3 ± 9.5	25.7 ± 10.2	< 0.0001
SSA	109.2 ± 15.0	119.6 ± 11.0	119.2 ± 13.6	118.9 ± 12.4	118.2 ± 13.2	< 0.0001
TPA	38.0 ± 16.6	27.1 ± 12.4	29.0 ± 13.3	30.0 ± 12.3	29.6 ± 12.7	< 0.0001
L1-S1 lordosis	33.0 ± 22.1	50.4 ± 13.1	49.3 ± 25.1	50.4 ± 15.6	50.1 ± 22.3	< 0.0001
T4-T12 kyphosis	-35.0 ± 20.9	-42.7 ± 16.4	-44.2 ± 16.5	-40 ± 16	-43.9 ± 16.8	< 0.001
*Comparison between baseline and time at follow up

Radiographical parameters at baseline

At baseline, mean pelvic incidence and mean pelvic tilt were respectively 58.8 ± 15° and 29.9 ± 11°. L1S1 lordosis was measured to -33.0 ± 22° and lordosis to 39.6 ± 20°, with an inflexion point being located from T7 to L5. Lack of lordosis was calculated pre-operatively to -32 ± 21°. SVA was calculated to 121.6 ± 129 mm, SSA to 109.2 ± 15°, and TPA to 38 ± 17° (Table 2).

Operative data

General informations on surgical treatment are presented in Table 3

Table 3

General informations on surgical treatment
	Total population, n = 273
Number of surgical procedure per patient, n (%)
1	123 (45.1%)
2	91 (33.3%)
3	37 (13.6%)
4	17 (6.2%)
5	3 (1.1%)
6	2 (0.7%)
Surgical management, n (%)
Anterior + posterior	117 (42.9%)
Posterior	156 (57.1%)
Operative time in min (mean ± SD)
Anterior	81.9 ± 50.1
Posterior	260.0 ± 146.2
Anterior + posterior	327.5 ± 53.9

Mean operative time was 269.5 ± 75.8 min with a mean operative blood loss estimated to 1.8 ± 1.3L with additional 0.8L in the drains the first 24hours. Average length of hospital stay was 16.2 ± 12.1 days.

Posterior instrumentation was classified as upper thoracic (T1-T5 to L4-iliac) in 83 patients (30.4%), thoraco-lumbar (T9-L2 to L4-iliac) in 103 patients (37.7%) and other (between T9 and L4) in 87 patients (31.9%).

Bone grafting was performed in 95% of cases, with the use of autologous bone in 97% of patients and additional bone substitutes in 21.2% of patients. BMP was used in only 1.9% of patients.

A total of 199 anterior surgeries were performed in 117 patients (42.9%) to fuse adjacent disks, using extreme Lateral Interbody Fusion (XLIF) or Oblique Lumbar Interbody Fusion (OLIF) in 73.8% of cases and Anterior Lumbar Interbody Fusion (ALIF) in 26.2% of cases. Among these 117 patients, 50.4% of them underwent at least 2 anterior surgical procedures (for example ALIF at L5S1 and OLIF at L3L4 and L4L5). The site of anterior fusion was situated between L2-L3 and L5-S1 in 85% of cases, with L3-L4 being the most performed level (32%).

Among the 286 osteotomies, the majority (n = 261) consisted of a pedicle subtraction osteotomy (PSO) with 211 grade 3 (73.8%) and 50 grade 4 (17.5%). There were 25 VCR regrouping 18 grade 5 (6.3%) and 7 grade 6(2.4%). L4 was the level most frequently osteotomized (53.6%), followed by L3 (21.7%) (Fig. 1).

Closing of the posterior osteotomy was achieved by compression on the rod in 44.8%, in situ bending in 16.8%, using a domino in 16.1%, mobilization of the operative table in 14.7%, or others (combination of these techniques) in 7.6% of cases.

Instrumentation (rods) was made of titanium in 74.4% and Chrome Cobalt en 24.8%. In 69.3% and 28.4% of cases, 5.5 and 6 mm diameter rods were used, respectively. Patient-specific personalized pre-bended and pre-sized rods were used for 18.5% patients. Additional rods were implanted in 14.6% patients.

Post-operative clinical and radiographical parameters

At 1year follow-up, mean values of VAS for lumbar and radicular pain, ODI and SRS-22 scores were all significantly improved (p < 0.001) (Table 2).

Regarding radiographical measurements, SVA, SSA, loss of lordosis, pelvic tilt, TPA, L1-S1 lordosis and T4-T12 kyphosis were significantly improved during follow-up, p < 0.001 (Table 2). Osteotomy correction angle was calculated to 20.6 ± 8.3° and 30.4 ± 5° for patients who underwent grade 3 and 4 osteotomy, respectively, p < 0.05.

Overall complications and reoperations

A total of 164 patients (59.2%) experienced ≥ 1 complication during the follow-up period, with a total of 277 complications. Most of the patients suffered 1 complication (56.3%), whereas 8.9% had ≥ 4 complications. Cumulative percentage of complication rate and complication-free survival rate during the follow-up period are respectively given in Figs. 2 and 3. Survival rates without complications were 57.4%, 47.4% and 30%, respectively at 1, 2 and 5 years. The main complication were mechanicals (35.2%), general (17.6%), infectious (17.2%), neurological (10.9%) (Table 4).

Table 4

Mortality, reoperation rates and overall complications
Total population	n = 273
Mortality, n (%)	1.8%
Reoperation, n (%)	27.5%
General complications, n (%)	17.2%^a
Number of complications per patient, n (%)^b
1	72.3%^b
2	12.8%^b
3	4.3%^b
4	8.5%^b
5	2.1%^b
Mechanical complications, n (%)	34.4%^c
Number of complications per patient, n(%)^d
1	77.7%^d
2	21.3%^d
3	1.1%^d
Rod breakage, n (%)	63.8%^d
Pseudarthrosis, n (%)	31.9%^d
Proximal junction kyphosis, n (%)	17.0%^d
Instrumentation failure, n (%)	5.3%^d
Neurological complications, n (%)	10.6%^e
Number of complications per patient, n (%)^f
1	93.1%^f
2	6.9%^f
Cauda equina syndrome	17.2%^f
Monoradicular palsy	75.9%^f
Pluriradicular palsy	6.9%f
Central paraparesis	6.9%^f
Surgical site infection, n (%)	17.2%^g
Number of complications per patient, n (%)^h
1	87.2%^h
2	10.6%^h
3	2.1%^h
^a patients with at least ≥ 1 general complication; ^b percentages among n = 47; ^c patients with at least ≥ 1 general complication; ^d percentages among n = 94; ^e patients with at least ≥ 1 neurological complication; ^f percentages among n = 29; ^g patients with at least ≥ 1 SSI; ^h percentages among n = 47

A total of 85 reoperations were performed in 75 patients, with a reoperation rate of 27.5%. After statistical analysis, risks factor for overall complications were pre-operative neurological status (Score ASIA D, RR = 2.3 [1.32-4.00], p = 0.03), operative time (+ 19% of risk per additional hour) and combined surgery (RR = 1.76 [1.08–2.04]) (p < 0.05)). Significant risk factors for complications are given in Table 5.

Table 5

Significant risk factor for specific complication (univariate analysis)
	Overall complication	General complications	Mechanical complications	Neurological complications	Surgical site infection
Age > 60	RR: 1.26 (0.88–1.79), p: 0.21	RR: 1.57 (0.80–3.07), p: 0.19	RR: 1.36 (0.85–2.16), p: 0.20	RR: 1.31 (0.58–2.97), p: 0.51	RR: 1.25 (0.66–2.38), p: 0.49
Sex (men)	RR: 0.87 (0.62–1.22), p:0.41	RR : 0.63 (0.33–1.22), p: 0.17	RR: 0.76 (0.49–1.20), p:0.24	RR: 0.79 (0.36–1.74), p: 0.56	RR : 1.04 (0.57–1.89), p: 0.91
BMI > 35 kg.m²	RR: 1.86 (0.85–4.07), p:0.12	RR: 2.18 (0.64–7.36), p:0.21	RR: 0.60 (0.08–4.38) p:0.62	RR: 3.58 (0.99–12.85), p: 0.05	RR: 3.10 (1.05–9.17), p:0.04
Diabetus	RR:1.17 (0.59–2.33), p:0.66	RR: 1.46 (0.52–4.10), p: 0.47	RR:0.63 (0.19–2.01), p:0.43	RR: 3.75 (1.19–11.81), p: 0.02	RR : 0.52 (0.07–3.82), p:0.52
ASIA score*	ASIA D: RR:2.30 (1.32-4.00), p:0.03	ASIA score D : RR: 4.54 (1.90-10.88), p: 0.0007	Non significant for every ASIA score	ASIA E : RR: 0.39 (0.16–0.99) , p: 0.04	ASIA score D : RR: 2.78 (1.19–6.46), p: 0.01
Etiology of spinal deformity*	Non significant for every etiologyNon significant for every etiology	flat back : RR : 0.32 (0.11–0.93), p: 0.03	Non significant for every etiology	Non significant for every etiology	Non significant for every etiology
Grade of osteotomy*	Non significant for every grade	Non significant for every grade	Non significant for every grade	Non significant for every grade	Non significant for every grade
Reoperation	RR:1.22 (0.80–1.87), p:0.36	RR: 3.26 (0.94–11.36), p: 0.06	RR:0.94 (0.55–1.60), p: 0.82	RR: 0.89 (0.31–2.54), p: 0.83	RR: 1.33 (0.60–2.93), p: 0.46
Operative time	RR: 1.19 (1.05–1.36), p:0.007	RR: 1.30 (1.05–1.61), p: 0.02	RR: 1.24 (1.05–1.46), p: 0.009	RR : 1.006 (1.00-1.011), p: 0.03	RR: 1.42 (1.31–1.78), p: 0.02
Type of surgery (combined surgery)	RR: 0.57 (0.49–0.93), p:0.01	RR: 0.56 (0.31–0.99), p:0.02p = 0.05	RR: 0.49 (0.32–0.74), p:0.0006	RR: 0.63 (0.30–1.31), p: 0.22	RR: 0.88 (0.49–1.59), p: 0.67
Bone graft	RR: 1.28 (0.67–2.43), p:0.45	RR: 1.85 (0.45–7.63), p:0.39	RR: 1.03 (0.50–2.13), p: 0.94	missing data	RR: 1.27 (0.39–4.10), p: 0.69
Patient specific rod	RR: 0.51 (0.29–0.89), p:0.02	RR: 0.38 (0.17–0.85), p: 0.02	R: 0.59 (0.27–1.29), p: 0.19	RR: 0.47 (0.14–1.61), p: 0.23	RR: 0.46 (0.17–1.27), p: 0.13
Pre operative loss of lordosis > -40°	RR:0.77 (0.50–1.21), p: 0.20	RR: 0.69 (0.34–1.40), p: 0.30	RR: 0.65 (0.37–1.16), p: 0.14	RR: 1.30 (0.41–4.07), p: 0.66	RR: 0.81 (0.32–2.06), p: 0.60
Pre operative SVA < 120mm	RR:1.02 (0.66–1.56, p: 0.93	RR: 0.70 (0.31–1.55), p: 0.37	RR: 1.28 (0.74–2.18), p: 0.37	RR: 0.88 (0.34–2.29), p: 0.79	RR: 1.18 (0.50–2.78), p: 0.70
History of spinal surgery with instrumentation	RR:1.22 (0.69–2.16), p: 0.49	No complication occured in patients with history of spinal surgery	RR: 1.71 (0.91–3.22), p: 0.09	RR: 1.06 (0.25–4.46), p:0.94	RR: 1.56 (0.62–3.95), p: 0;35
Pre operative ODI	Non significant for every ODI score	Non significant for every ODI score	Non significant for every ODI score	ODI 30–60 : RR2.46 (1.05–5.77), p: 0.04	Non significant for every ODI score
Importance of surgical correction	Non significant for SVA and loss of lordosis	Non significant for SVA and loss of lordosis	Non significant for SVA and loss of lordosis	Non significant for SVA and loss of lordosis	Non significant for SVA and loss of lordosis
*Due to the multiple sub-categories, only significant results are presented in these rows
Significant results are marked in bold

Patient specific rods appeared to be significantly associated with less overall complications (RR = 0.51 [0.29–0.89], p = 0.02), being so a protective factor. The following parameters were not associated with an increase of complications rates (p > 0.05): age, aetiology of the spinal deformity, severity of preoperative impbalance, importance of the surgical correction, and the type of osteotomy.

General complications

In the series, we found that mortality rates were 1.8% (5 patients). Main causes of perioperative mortality were blood loss (2 patients), cardiovascular (1 patient), pulmonay embolism (1 patient), and post operative infection (1 patient). Age, sex and BMI were not considered as risk factor, whereas pre-operative ASIA score (score of D), post operative flat-back as etiology of the spinal deformity, posterior surgery only, long operative time and absence of use of patient specific rods were significantly associated with general complications (Table 5) (p < 0.05)

Mechanical complications

A total of 94 mechanicals were observed in our series, experienced by 73 patients.

Pseudarthrosis and rod breakage were found in 60 patients (22.0%). The delay of this complication was available for 57 patients and was estimated to 15 months after surgery, with 17 (29.8%) occuring within the first post operative year and 40 (70.2%) after 1 year. We observed disparities between the different inclusion centers for this complication (min: 8%, max 32%). No mechanical failure occured in patients who underwent VCR. No significant difference according to sex or BMI > 25 kg/m² was observed for pseudarthrosis and rod breakage. Short operative time and combined approaches were significantly associated with less mechanical complications (Table 5).

Proximal junction kyphosis was found in 16 patients (5.9%). 9 patients hade undergone a long-segment construct (T3-T4 to iliac) and 7 a short-segment construct (thoraco-lumbar to iliac). 5 revision surgeries were performed because of neurological injury.

Neurological complications

A total of 31 neurological complications were observed in 29 patients (10.6%), with 2 patients experienced 2 neurological complications. Most complications (n = 29, 93.5%) concerned the peripheral nervous system, including 22 monoradicular paresis (75.8%), 5 cauda equina syndrome (17.2%) and 2 pluri-radicular paresis (6.8%), and 2 complications (6.5%) consisted of spinal cord injury and resulted in central paraparesis. Dural tears occured in 10% of the procedure. After statistical analysis, diabetus, ASIA score, operative time and pre operative ODI were significantly associated with a higher risk for neurological complications (Table 5) (p < 0.05)

Surgical site infection

Surgical site infection was found in 47 patients (17.2%), with a predominance of gram-negative bacteria (32.7%), staphylococcus (30.6%) and skin bacteria (28.6%). Reoperation was performed once in 41 patients (87.2%), twice in 5 patients (10.6%) and three times in 1 patient (2.1%). After statistical analysis, worse pre-operative neurological status, BMI > 35kg/m² and operative time were considered as risk factor for surgical site infection (Table 5) (p < 0.05).

Due to population ageing, patients suffering from disability due to fixed sagittal imbalance are becoming more and more numerous, and more and more in need for a surgical deformity correction. However, this kind of surgery, still under controversy, represents a very heavy procedure and it is crucial to determine risk factors associated with per/postoperative complications to both inform the patients and help the clinicians in the selection of the patients supposed to really benefit from the surgery.

The purpose of the current study was, through a large multicentric cohort, to provide valuable data and updates on complication rates associated with 3-columns spinal osteotomies. To our knowledge, the present report is one of the largest series of patient who underwent spinal osteotomies for fixed sagittal imbalance of multiple etiologies.

Overall complications and reoperations

A total of 277 complications were reported in this series, with 59.2% of patients experiencing at least 1 complication during the follow-up period. This is quite consistent with the series of Amzallag et al, O'Neill et al, Eskillson et al and Bianco et al finding respectively 45%, 36%, 57% and 42% of complications [14–17]. Our reoperation rate (27.5%) was also consistent with other report, estimating this one between 25 and 31% [14–16].

Considering our results, operating time was found to represent a significant factor for complications (+ 19% of risk for each additional hour). We therefore assume that 3C spinal osteotomies gained to be planned rigorouslypre-operatively, with establishment of a clear surgical strategy and identification of potential intra-operative difficulties, in order to reduce the operating time. Indeed, surgeons performing spinal osteotomies should make efforts to reduce operative time as much as possible. In that objective, recents series underscore the importance of the learning curve in spinal osteotomies [10, 25, 26], and emphasize the role of meticulous preparation of the surgery and the use of 2 experienced surgeons [22]. Furthermore, even if it requires more pre-operative preparation of the procedure, patient-specific rods help to restore a more suitable sagittal balance in patients and their use might reduce the risk of overall complications.

General complications

Conversely, our number of peri operative deaths (1.8%) seems quite low, since recent large series tend to estimate this rate between 2.1% [2] and 4.6% [12] after spinal osteotomy. This highlights the complexity in management of patients undergoing spinal surgery due to fixed sagittal imbalance, and therefore the need for assessment of complications risk factors. Among those ones, operating time appears to be the most predictive of experiencing a post operative complication in our series with an overrisk of 19% for every additional operative hour.

Blood loss can be associated with complications such as organ damages, coagulopathy, hypotension, or transfusion related injuries. Our mean values for operative duration and blood loss are nervertheless consistent with other report (Ikenaga et al : 277min, 1988mL, Kim et al : 275min, 1515mL, Eskillson et al 336min, 2000mL and Smith et al 8 hours, 3300mL) [2, 10, 15, 24], and operative time or blood loss are not always significantly associated with complication rate in spinal osteotomies.

Mechanicals complications

Mechanical complications were the most frequent (> 35% of the complications), and typically occurred more during the follow-up period than immediately after the procedure. Among mechanical complications, pseudarthrosis and rod breakage were the most frequent. The same percentage of events was found in a study evaluating risk factors for rod fracture following PSO [27]. In an other recent series, Smith et al also reported rod breakage as the most common complication after osteotomy for adult spinal deformity, but with a complication rate being higher than ours (31.7% versus 21.9% in our serie) [24]. To our mind, this difference could be explained by several factors. First, it is well known that anterior column support decreases the rate of pseudarthrosis associated with long-segment posterior fusion [28], and only 17% of the patients in their study underwent an anterior procedure (versus 42.9% herein). Secondly, patients tend to have a higher BMI (29.0) than the ones in our serie, with BMI being often associated with an over-risk of rod breakage [27]. Finally, as highlighted by our work and the work of Smith et al [27], there is some discrepancies regarding the rate of rod breakage across centers. Recent contributions from the literature emphasise that the use of chrome cobalt rod is associated with an over-risk of mechanical failures [27], and that multiple-rod constructs provide increased stability across spinal osteotomies and significantly prevent rod breakage and pseudarthrosis in comparison to a standard dual-rod construct [29, 30].

Therefore, in order to minimize the risk of mechanical failure, we recommend to use a ≥ 3-rods construct if possible with patient specific rods and to perform an additional anterior procedure when necessary (unfused discs) in patient undergoing spinal osteotomy.

Proximal junction kyphosis (PJK) was found in 16 patients (5.9%), with 5 of them undergoing a reoperation. Although frequently not mentionned in the literature, Smith et al reported a somewhat higher rate of PJK in their serie (9.8%), accounting for 44.4% of the radiographic complications and requiring reoperation in 6 cases (7.3%) [24]. Many risk factors for PJK have been described in the literrature, with most of them being independent of the surgical procedure (osteoporosis, overweight, older age, spinal muscular disease, high pre operative SVA) [31–33]. Herein, we found a non-significant trend to a failure to restore a suitable L4-S1 lordosis in patients developping PJK. Therefore, in order to minimize the risk of post operative PJK in patients with spinal osteotomies, we recommend to perfectly correct the sagittal balance with a harmonious distribution of the spinal curves and a sufficient restoration of the L4-S1 lordosis.

Neurological complications

Postoperative neurologic deficits are relatively common following spinal osteotomy, as shown by our neurologic complication rate around 11%. Although consistent with the work of Buchowski et al [9], 2 recent literature reviews have found relatively lower rates of neurological complications, estimated around 9% [34, 35]. No differences between grade 3/4 and grade 5/6 osteotomies were observed, as expected by the recent work of Sciubba who find same results in PSO and VCR [35]. In spite of the fact that corrective osteotomy in regard of the spinal cord is believed to be more likely to cause neurological deficits because of the shortening of the spine [36], we did not found more neurological complications in patients who underwent osteotomies over L1. In fact, no levels was found to be more associated with neurological deficits, as also shown by Buchowski et al [9]. Causes were found to be nerve root damage, compression of dural sheath during osteotomy closure, or instability when performing VCR. Even if operative time was found to be significantly associated with neurological complications in our series, we believe that great caution regarding neurological elements is absolutely necessary throughout the whole procedure. Attention should also be paid to intraoperative neuromonitoring, even though it may not detect a potential deficits [9]. Prevention of those complications might be done while enlarging the canal centrally, reducting fibrosis in revision surgery, search for posterior bony contact in osteotomy closure and using temporary rod in VCR to control any translation.

Surgical site infection

Surgical site infection was found in 47 patients (17.2%). This rate is higher than reported in recent series (mean value of 7.5%) [17, 24], which may be explained by our long follow-up and prospective collection of data. Worse pre-operative neurological status, BMI > 35kg/m² and operating time were considered as risk factor for SSI in our study, but in fact spinal osteotomy surgery combines many risk factors. Those can be both inherent to the patient (57.6% of patients with BMI > 25, 27.1% smokers, 21.3% of patients classified as ASA score ≥ 3 and 7.8% of patients suffering from diabetus) or to the surgical procedure (mean operative time > 4h, important blood loss, posterior approach). The outcome of patient who experienced SSI in our serie has not been studied yet but recent litterature data show a trend develop pseudarthrosis [37] after SSI, therefore reinforceing the fact that care must be taken to avoid such a complication.

Clinical and radiological outcome

In spite of the non-negligible complication rate in our series, outcome and patient satisfaction were fortunately found to be significantly improved post operatively and during the whole follow-up period. Significant improvement in VAS for lumbar pain and SVA were associated with a decrease in disability experienced by patients. Our reduction of SVA (-63mm) is somewhat equivalent to the one expected considering the recent literrature (O'neill − 93, Eskillson − 25, Bourghil − 83) [15, 16, 26].

Osteotomy correction angles were respectively 20.6 ± 8.3° and 30.4 ± 5° for grade 3 and grade 4 osteotomies, which is close to reported values (Kim et al 36.2°, Barrey et al 27°, Eskillsson et al 27.2°) [2, 13, 15]. Regarding pain and functional scores, ODI score improved by + 21% post operatively, which was consistent with previous data reported by Kim et al (+ 26.7%), O'neill et al (+ 22%) and Eskillson et al (+ 16%) [2, 15, 16]. Considering all these data, spinal osteotomy as correction surgery provides clinical benefit to patients. This highlights the importance of publishing new large series of patients in order to regularly update knowledges about complication risk factor in this type of surgery.

Strengths and limitations of our series

Strenghts of the current study include the prospective multicenter design for data collection with the use of an on-site study coordinator and a standardized data collection form, as long as the high number of patients included.

We admit some limitations in our study. The extrapolability of our results to the general population may be affected by the fact that the patients selected were operated on in expert centre for spinal deformity surgery (thus underestimating complication rate compared to non-expert centres). Although we have made every effort to be comprehensive about the occurrence of complications in every patient included, it is possible that patients who were lost to follow-up may have experienced additional complications after their last follow-up visit. Furthermore, the design of the study limited the ability to detect some complications, such as asymptomatic pseudoarthrosis that may not be seen on full spine radiographs.

The current sudy represents, to our knowledge one of the largest series of 3-columns spinal osteotomies for fixed sagittal imbalance of multiple etiologies and therefore provide valuable data on the frequency and the nature of complications associated with this surgical procedure. Despite a non-negligible complication rate, a significant improvement in both clinical and radiological parameters was found post operatively, reinforceing the fact that spinal osteotomies are helpful to treat spinal deformity and efficient to improve clinically these patients regarding both pain and disability. Operative time was found to be consistenly associated with an increased risk for complications (+ 19% each additional hour) and and surgeons should always keep in mind to reduce operative time as much as possible. Finally, patient-specific rods appeared through this study to act as a protective factor for the ocurrence of complciations and its use should thus be promoted. Finally, the use of patient-specific rods appeared to be a protective factor for complications after spinal osteotomies and may represent a promising innovative technology to treat adult spinal deformities.

3 columns

ALIF

Anterior Lumbar Interbody Fusion

ASA

American Society of Anesthesiologists

ASIA

American Spinal cord Injury Association

BMI

Body Mass Index

ODI

Oswestry Disability Index

OLIF

Oblique Lumbar Interbody Fusion

PJK

Proximal Junction Kyphosis

PSO

Pedicle Substraction Osteotomy

SFCR

Société Française de Chirurgie Rachidienne)

SPO

Smith-Peterson osteotomies

SRS-22

Scoliosis Research Society-22

SSA

Spino Sacral Angle

SSI

Surgical Site Infection

SVA

Sagittal Vertical Axis

TPA

T1 Pelvic Angle

VAS

Visual Analogue Scale

VCR

Vertebral Column Resection

XLIF

eXtreme Lateral Interbody Fusion

ACKNOWLEDGEMENTS

The authors sincerely thank all the persons who participated in the data collection.

Funding: None

Conflicts of interest/Competing interests: All authors declare: no support from any organization for this study; no financial relationships with any organizations that might have a benefit in the submitted work.

Permission: None

Ethical approval Given the documents at its disposal, the Research Ethics Committee of the Hospices Civils de Lyon issued a favorable opinion for the publication of this research

Bridwell KH, Lewis SJ, Rinella A, Lenke LG, Baldus C, Blanke K (2004) Pedicle subtraction osteotomy for the treatment of fixed sagittal imbalance. Surgical technique. J Bone Joint Surg Am 86–A(Suppl 1):44–50
Kim KT, Lee SH, Suk KS, Lee JH, Jeong BO (2012) Outcome of pedicle subtraction osteotomies for fixed sagittal imbalance of multiple etiologies: a retrospective review of 140 patients. Spine (Phila Pa 1976) 37(19):1667–1675
Kim YJ, Bridwell KH, Lenke LG et al (2007) Results of lumbar pedicle subtraction osteotomies for fixed sagittal imbalance: A minimum 5-year follow-up study. Spine (Phila Pa 1976) 32:2189e97
Barrey C, Roussouly P, Le Huec JC, D’Acunzi G, Perrin G (2013) Compensatory mechanisms contributing to keep the sagittal balance of the spine. Eur Spine J. 2013 Nov;22 Suppl 6(Suppl 6):S834-41
Kim KT, Suk KS, Cho YJ, Hong GP, Park BJ (2002) Clinical outcome results of pedicle subtraction osteotomy in ankylosing spondylitis with kyphotic deformity. Spine 27(6):612–618
Thomasen E (1985) Vertebral osteotomy for correction of kyphosis in ankylosing spondylitis. Clin Orthop Relat Res 142 – 52
Schwab F, Blondel B, Chay E, Demakakos J, Lenke L, Tropiano P, Ames C, Smith JS, Shaffrey C, Glassman S, Farcy JP, Lafage V (2014) The comprehensive anatomical spinal osteotomy classification. Neurosurgery 74:112–120
Suk SI, Kim JH, Kim WJ, Lee SM, Chung ER, Nah KH (2002) Posterior vertebral column resection for severe spinal deformities. Spine (Phila Pa 1976) 1(21):2374–2382
Buchowski JM, Bridwell KH, Lenke LG, Kuhns CA, Lehman RA Jr, Kim YJ, Stewart D, Baldus C (2007) Neurologic complications of lumbar pedicle subtraction osteotomy: a 10-year assessment. Spine (Phila Pa 1976) 15(20):2245–2252
Ikenaga M, Shikata J, Takemoto M, Tanaka C (2007) Clinical outcomes and complications after pedicle subtraction osteotomy for correction of thoracolumbar kyphosis. J Neurosurg Spine 6(4):330–336
Bridwell KH, Lewis SJ, Edwards C, Lenke LG, Iffrig TM, Berra A, Baldus C, Blanke K (2003) Complications and outcomes of pedicle subtraction osteotomies for fixed sagittal imbalance. Spine (Phila Pa 1976) 15(18):2093–2101
Daubs MD, Brodke DS, Annis P, Lawrence BD (2016) Perioperative Complications of Pedicle Subtraction Osteotomy. Global Spine J 6(7):630–635
Barrey C, Perrin G, Michel F, Vital JM, Obeid I (2014) Pedicle subtraction osteotomy in the lumbar spine: indications, technical aspects, results and complications. Eur J Orthop Surg Traumatol 24(1, Suppl 1):S21–S30
Amzallag J (2008) Complications of spinal osteotomies: multicenter study of 402 cases [in French]. Paris Val-de-Marne University, Créteil
Eskilsson K, Sharma D, Johansson C, Hedlund R (2017) Pedicle subtraction osteotomy: a comprehensive analysis in 104 patients. Does the cause of deformity influence the outcome? J Neurosurg Spine 27(1):56–62
Oʼneill KR, Lenke LG, Bridwell KH, Hyun SJ, Neuman B, Dorward I, Koester L (2014) Clinical and radiographic outcomes after 3-column osteotomies with 5-year follow-up. Spine (Phila Pa 1976) 39(1):424–432
Bianco K et al (2014) Complications and intercenter variability of three-column osteotomies for spinal deformity surgery: a retrospective review of 423 patients. NeuroSurg Focus 36(5):E18
Mummaneni PV, Dhall SS, Ondra SL et al (2008) Pedicle subtraction osteotomy. Neurosurgery 63(suppl):171e6
Kiaer T, Gehrchen M (2010) Transpedicular closed wedge osteotomy in ankylosing spondylitis: Results of surgical treatment and prospective outcome analysis. Eur Spine J 19:57e64
Berven SH, Deviren V, Smith JA, Emami A, Hu SS, Bradford DS (2001) Management of fixed sagittal plane deformity: results of the transpedicular wedge resection osteotomy. Spine (Phila Pa 1976).15;26(18):2036-43
Yang BP, Ondra SL, Chen LA et al (2006) Clinical and radiographic outcomes of thoracic and lumbar pedicle subtraction osteotomy for fixed sagittal imbalance. J Neurosurg Spine 5:9–17
Ames CP, Barry JJ, Keshavarzi S, Dede O, Weber MH, Deviren V (2013) Perioperative Outcomes and Complications of Pedicle Subtraction Osteotomy in Cases With Single Versus Two Attending Surgeons. Spine Deform 1(1):51–58
Barrey C, Jund J, Noseda O, Roussouly P (2007) Sagittal balance of the pelvis-spine complex and lumbar degenerative diseases. A comparative study about 85 cases. Eur Spine J 16(9):1459–1467
Smith JS, Shaffrey CI, Klineberg E, Lafage V, Schwab F, Lafage R, Kim HJ, Hostin R, Mundis GM Jr, Gupta M, Liabaud B, Scheer JK, Diebo BG, Protopsaltis TS, Kelly MP, Deviren V, Hart R, Burton D, Bess S, Ames CP, on behalf of the International Spine Study Group (2017) Complication rates associated with 3-column osteotomy in 82 adult spinal deformity patients: retrospective review of a prospectively collected multicenter consecutive series with 2-year follow-up. J Neurosurg Spine 27(4):444–457
Choi HY, Hyun SJ, Kim KJ, Jahng TA, Kim HJ (2017) Surgical and Radiographic Outcomes After Pedicle Subtraction Osteotomy According to Surgeon’s Experience. Spine (Phila Pa 1976) 42(13):1
Bourghli A, Cawley D, Novoa F, Rey M, Alzakri A, Larrieu D, Vital JM, Gille O, Boissiere L, Obeid I (2018) 102 lumbar pedicle subtraction osteotomies: one surgeon’s learning curve. Eur Spine J 27(3):652–660
Smith JS, Shaffrey E, Klineberg E, Shaffrey CI, Lafage V, Schwab FJ, Protopsaltis T, Scheer JK, Mundis GM Jr, Fu KM, Gupta MC, Hostin R, Deviren V, Kebaish K, Hart R, Burton DC, Line B, Bess S, Ames CP, International Spine Study Group (2014) Prospective multicenter assessment of risk factors for rod fracture following surgery for adult spinal deformity. J Neurosurg Spine 21(6):994–1003
Saer EH III, Winter RB, Lonstein JE (1990) Long scoliosis fusion to the sacrum in adults with nonparalytic scoliosis. An improved method. Spine 15:650–653
Merrill RK, Kim JS, Leven DM, Kim JH, Cho SK (2017) Multi-Rod Constructs Can Prevent Rod Breakage and Pseudarthrosis at the Lumbosacral Junction in Adult Spinal Deformity. Global Spine J 7(6):514–520
Hyun SJ, Lenke LG, Kim YC, Koester LA, Blanke KM (2014) Comparison of standard 2-rod constructs to multiple-rod constructs for fixation across 3-column spinal osteotomies. Spine (Phila Pa 1976) 15(22):1899–1904
Liu FY, Wang T, Yang SD, Wang H, Yang DL, Ding WY (2016) Incidence and risk factors for proximal junctional kyphosis: a meta-analysis. Eur Spine J 25(8):2376–2383
Zhao J, Chen K, Zhai X, Chen K, Li M, Lu Y (2021) Incidence and risk factors of proximal junctional kyphosis after internal fixation for adult spinal deformity: a systematic evaluation and meta-analysis. Neurosurg Rev 44(2):855–866
Sebaaly A et al (2018) Incidence and Risk Factors for Proximal Junctional Kyphosis. Clinical Spine Surgery E178-E183
Zanirato A, Damilano M, Formica M, Piazzolla A, Lovi A, Villafañe JH, Berjano P (2018) Complications in adult spine deformity surgery: a systematic review of the recent literature with reporting of aggregated incidences. Eur Spine J 27(9):2272–2284
Sciubba DM, Yurter A, Smith JS, Kelly MP, Scheer JK, Goodwin CR, Lafage V, Hart RA, Bess S, Kebaish K, Schwab F, Shaffrey CI, Ames CP, International Spine Study Group (ISSG) (2015) A Comprehensive Review of Complication Rates After Surgery for Adult Deformity: A Reference for Informed Consent. Spine Deform 3(6):575–594
Kawahara N, Tomita K, Kobayashi T, Abdel-Wanis ME, Murakami H, Akamaru T (2005) Influence of acute shortening on the spinal cord: an experimental study. Spine 30:613–620
Wang TY, Back AG, Hompe E, Wall K, Gottfried ON (2017) Impact of surgical site infection and surgical debridement on lumbar arthrodesis: A single-institution analysis of incidence and risk factors. J Clin Neurosci 39:164–169

Download PDF

Version 1

posted

You are reading this latest preprint version

Factors influencing complications after 3-columns spinal osteotomies for fixed sagittal imbalance from multiple etiologies: a multicentric cohort study about 286 cases in 273 patients

Status:

Version 1

Abstract

Figures

Introduction

Material And Methods

Inclusion criteria

Exclusion criteria

Clinical data

Radiological data

Surgical data and instrumentation

Post operative management

Complications and mortality

Statistical analysis

Ethical and Legal Considerations

Results

Clinical parameters at baseline

Radiographical parameters at baseline

Operative data

Post-operative clinical and radiographical parameters

Overall complications and reoperations

General complications

Mechanical complications

Neurological complications

Surgical site infection

Discussion

Overall complications and reoperations

General complications

Mechanicals complications

Neurological complications

Surgical site infection

Clinical and radiological outcome

Strengths and limitations of our series

Conclusion

Abbreviations

Declarations

References

Status:

Version 1