- Email: [email protected]
The benefit of nonoperative treatment for adult spinal deformity: identifying predictors for reaching a minimal clinically important difference
Accepted Manuscript Title: The benefit of non-operative treatment for adult spinal deformity: identifying predictors for reaching a minimal clinically important difference Author: Shian Liu, Bassel Diebo, Jensen K. Henry, Justin S. Smith, Richard Hostin, Matthew E. Cunningham, Gregory Mundis, Christopher P. Ames, Douglas Burton, Shay Bess, Behrooz Akbarnia, Robert Hart, Peter G. Passias, Frank J. Schwab, Virginie Lafage, on behalf of the International Spine Study Group (ISSG) PII: DOI: Reference:
S1529-9430(15)01628-9 http://dx.doi.org/doi: 10.1016/j.spinee.2015.10.043 SPINEE 56673
To appear in:
The Spine Journal
Received date: Revised date: Accepted date:
21-6-2015 18-9-2015 22-10-2015
Please cite this article as: Shian Liu, Bassel Diebo, Jensen K. Henry, Justin S. Smith, Richard Hostin, Matthew E. Cunningham, Gregory Mundis, Christopher P. Ames, Douglas Burton, Shay Bess, Behrooz Akbarnia, Robert Hart, Peter G. Passias, Frank J. Schwab, Virginie Lafage, on behalf of the International Spine Study Group (ISSG), The benefit of non-operative treatment for adult spinal deformity: identifying predictors for reaching a minimal clinically important difference, The Spine Journal (2015), http://dx.doi.org/doi: 10.1016/j.spinee.2015.10.043. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Title: The Benefit of Non-Operative Treatment for Adult Spinal Deformity: Identifying
2
Predictors for Reaching a Minimal Clinically Important Difference
3
Authors:
4
Shian Liu, MD1; Bassel Diebo, MD1; Jensen K. Henry, BA1; Justin S. Smith, MD PhD2; Richard
5
Hostin, MD3; Matthew E. Cunningham, MD4; Gregory Mundis, MD5; Christopher P. Ames,
6
MD6; Douglas Burton, MD7; Shay Bess, MD8; Behrooz Akbarnia, MD9; Robert Hart, MD10;
7
Peter G. Passias, MD1; Frank J. Schwab, MD1; Virginie Lafage, PhD1; on behalf of the
8
International Spine Study Group (ISSG)
9
Affiliations:
10
1
Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, 301 East 17th St., New
11
York, NY, United States, 10003
12
2
13
United States, PO Box: 800212
14
3
15
Alliance Blvd, #810, Plano TX, USA
16
4
17
States. 535 East 70th St., New York, NY 10021, USA
18
5
Scripps Clinic Torrey Pines, 10666 N Torrey Pines Rd, La Jolla, CA 92037
19
6
Department of Neurosurgery, University of California, San Francisco Medical Center, San
20
Francisco, CA, United States. 400 Parnassus Street, San Francisco, CA, USA.
Department of Neurosurgery, University of Virginia Medical Center, Charlottesville, VA,
Department of Orthopaedic Surgery, Baylor Scoliosis Center, Plano, TX, United States 4708
Department of Orthopaedic Surgery, Hospital for Special Surgery, New York, NY, United
1 Page 1 of 28
1
7
Department of Orthopaedic Surgery, University of Kansas Medical Center, Kansas City, KS
2
8
Rocky Mountain Scoliosis and Spine Center, Denver, CO 80205, USA
3
9
Department of Orthopaedic Surgery, San Diego Center for Spinal Disorders, La Jolla, CA
4
10
5
Corresponding Author:
6
Virginie Lafage, PhD
7
NYU Hospital for Joint Diseases
8
306 East 15th St., Suite 1F
9
New York, NY 10003
Department of Orthopaedic Surgery, Oregon Health Sciences University, Portland, OR
10
Email: [email protected]
11
Phone: 646-794-9640
12
Fax: 646-602-6927
13 14
Acknowledgements: Data collection for this study was supported by a grant from Depuy to the
15
International Spine Study Group Foundation. No specific sources of funding were linked to this
16
work in particular.
17
ABSTRACT
18
Background Context: Adult spinal deformity (ASD) patients may gain a MCID in one or more
19
of the HRQOL instruments without surgical intervention. This study identifies baseline
20
characteristics of this subset of non-operative patients and proposes predictors of those most
21
likely to benefit.
2 Page 2 of 28
1
Purpose: Determine factors that affect likelihood of non-operative patients to reach minimum
2
clinically important difference (MCID).
3
Study Design/Setting: Retrospective review of prospective, multi-center database.
4
Patient Sample: Non-operative ASD patients.
5
Outcome Measures: Health-related quality of life measures (HRQOL), including the Scoliosis
6
Research Society (SRS)-22 questionnaire.
7
Methods: Multicenter database of 215 non-operative patients with ASD and minimum 2-year
8
follow-up. Using a multivariate analysis, two groups were compared to identify possible
9
predictors: those that reached an MCID in SRS Pain or Activity (n=86) at 2 years, and those who
10
did not reach MCID (n=129). Subgroup multivariate analysis of patients with a deficit (potential
11
improvement) in both SRS Pain and Activity (n=84) was performed. Data collection was
12
supported by a grant from Depuy for the International Spine Study Group Foundation.
13
Results: At baseline, the non-operative patients that reached MCID had a significantly lower
14
SRS Pain score (3.0 vs 3.6), smaller thoracolumbar (TL) Cobb angle (29.6 vs. 36.5; 87 patients
15
with SRS-Schwab classification Lumbar or Double), sacral slope (33.1 vs. 36.4), and less
16
lumbar lordosis (46.5 vs. 52.8) (all P<0.05). SRS Pain and TL Cobb were significant predictors
17
of reaching MCID. PI-LL was significant on univariate analysis but not by multivariate (7.5 vs.
18
2.6; P=0.14). In the subset of severely disabled patients, worse vertebral obliquity was a
19
predictor for not achieving MCID (P<0.05).
20
Conclusions: Non-operative ASD patients who achieved an MCID in SRS Activity or Pain had
21
a lower baseline SRS Pain Score and less coronal deformity in the TL region. Greater baseline 3 Page 3 of 28
1
pain offers significant room for potential improvement, which may be important in identifying
2
ASD patients who have the potential to reach an MCID non-operatively. Coronal deformities in
3
the TL region, and associated vertebral obliquity may negatively impact improvement potential
4
with non-operative care.
5
6
Key words: spinal deformity; non-operative; treatment; outcomes; HRQOL; health-related
7
quality of life
8
4 Page 4 of 28
1
MANUSCRIPT
2
INTRODUCTION
3
Spinal deformity in the skeletally mature patient, with an incidence up to 32% in adults
4
and 60% in the elderly, is becoming a more commonly recognized condition among both spine
5
surgeons and general healthcare providers [1–4]. Physicians across all specialties are
6
increasingly recognizing the impact of adult spinal deformity (ASD) on patients and the
7
healthcare system [5]. Healthcare costs for treating spinal deformity are rising [5]. While large
8
gains have been made in the evolution of surgical treatment, non-operative management in
9
certain patients is prudent and necessary in everyday practice. The economic issues of delivering
10
care involve a balance between surgery with evident clinical improvement and non-operative
11
management which may be effective for some patients.
12
The treatment of ASD can be assessed quantitatively and tracked over time using
13
validated patient-reported outcomes such as the Scoliosis Research Society questionnaire (SRS-
14
22r), the Oswestry Disability Index (ODI), and the Short Form 36 questionnaire (SF-36) [3,5–
15
10]. Since the incorporation of these health-related quality of life (HRQOL) measures into
16
medical practice and research, multiple reports have shown the significant impact of surgical
17
intervention for ASD [11–16].
18
With the inception of HRQOLs arose the concern that improvements in these scores do
19
not necessarily translate into a clinically discernible difference that is of significance to the
20
patient. Hence, the concept of a minimal clinically important difference (MCID) has been
21
introduced in the spine literature to quantify the absolute minimum change that can be
22
considered a success [17–19]. Many studies have demonstrated a clear advantage in gaining
5 Page 5 of 28
1
MCID after surgical treatment for spinal pathologies such as spondylolisthesis, disc pathology,
2
spinal stenosis and ASD, with comparatively poor improvement with non-operative care [17,20–
3
23].
4
While surgical treatment can improve both pain and disability [11–13,24,25], there are
5
risks involved [26,27].Thus, it is important to thoroughly assess all treatment options, including
6
non-operative care, which has not been as well studied in the literature. Often non-operative
7
patients are grouped together in cohort analyses, but it is possible that non-operative
8
management may actually offer acceptable quality of life maintenance or improvements for a
9
certain subset of
patients[23,28]. A recent study on 464 patients with ASD found that a
10
subgroup of non-operatively treated patients actually improved in MCID: up to 52% (n=117)
11
reached MCID in at least 1 HRQOL, with 20% reaching MCID in SRS Activity and 24%
12
reaching MCID in SRS Pain[28].
13
The purpose of the present study was to characterize a subset of patients with ASD that
14
demonstrated improvement with non-operative management and assess for baseline
15
characteristics that may distinguish them from those that failed to significantly improve with
16
non-operative treatment.
17
18
METHODS
19
Study Design
20
This was a retrospective analysis of a prospective, multicenter consecutive case series of
21
215 patients with ASD. Internal review board approval was obtained from all 11 high volume 6 Page 6 of 28
1
centers that care for patients with ASD prior to data collection and informed consent was
2
obtained from each patient to be included in the study. Inclusion criteria for the database were
3
age over 18 years and radiographic evidence of spinal deformity, defined as one or more of the
4
following: coronal Cobb ≥ 20, sagittal vertical axis (SVA) ≥5cm, pelvic tilt ≥25, thoracic
5
kyphosis (TK) ≥ 60. For the purposes of this study, only non-operative patients who completed
6
2-year clinical and radiographic follow up were analyzed. Treatment was not randomized, but
7
rather was based on a combination of patient input and physician counseling on the complexities
8
and outcomes of care. The specific type of non-operative care was not specified and some of the
9
patients had observational care only. Patients with malignancy, active infection or rheumatologic
10
disease, and those with neuromuscular or congenital scoliosis were excluded. Patients who
11
crossed over to the operative arm were excluded in the analysis, as this would falsely elevate the
12
number of non-operative patients who would reach an MCID.
13
The MCID values used for each HRQOL were established by prior studies using the
14
external anchor methodology with the SF-36 health transition item[17,18,20–22,28]. Established
15
MCID values were -12.8 for ODI, +0.587 for SRS Pain, +0.375 for SRS Activity, +0.800 for
16
SRS appearance, +0.420 for SRS mental, and +4.9 for the SF-36 physical component score (SF-
17
36 PCS)[18,20,29–31].
18
Patients with ASD characteristically present with pain and disability [7,32,33]. To
19
quantify the need for improvement at baseline, MCID deficit in SRS Activity or Pain was
20
determined by being at least 1 or more MCID below a normative, age and gender matched
21
population[34]:
7 Page 7 of 28
1
Non-operative patients were also grouped into two cohorts: those that reached MCID in SRS
2
Activity or Pain at 2 years (rMCID) and those that missed MCID (mMCID) at 2 years:
3
MCID change greater than +1 was considered an improvement (reached MCID); less than -1 was
4
considered deterioration (missed MCID); and between -1 and +1 was considered unchanged
5
from baseline.
6
Data Collection
7
Patient demographic data included age, gender, race, body mass index (BMI), medical
8
co-morbidities, prior surgery, smoking status, employment status, and Charlson Comorbidity
9
Index (CCI) [35]. Radiographic measurements based on coronal and sagittal full-length standing
10
films were obtained. Patient-reported outcomes, including SRS-22r, SF-36, and ODI were
11
recorded at the initial baseline visit and at 2-year follow-up. The SRS-22r is a disease-specific
12
questionnaire with scores ranging from 0 (worst) to 5 (best). SF-36 is a generic measure of health
13
that is not age, treatment, or disease-specific; it ranges from 0 to 100, with higher scores
14
signifying a better health status [3,9,36]. The SF-36 can be divided into Physical Component
15
Scores (PCS) and Mental Component Scores (MCS). ODI is specific to disability in the lumbar
16
spine and ranges from 0 to 100, with higher numbers denoting more disability [37,38]. Data were
17
uploaded to a central electronic data capture system and queried by professional clinical research
18
monitors to ensure accuracy and reliability.
19
Radiographic Analysis Radiographic parameters at baseline were measured using SpineView software
20 21
(Laboratory of Biomechanics, ENSAM, Paris, France).[39]
Sagittal and coronal spinal
8 Page 8 of 28
1
parameters included sacral slope (SS), pelvic tilt (PT), pelvic incidence (PI), lumbar lordosis
2
(LL), sagittal vertical axis (C7-S1 SVA), maximum thoracic kyphosis (TK), thoraco-lumbar
3
Cobb angle in the coronal plane (TL Cobb), and maximum vertebral obliquity for
4
Thoracolumbar vertebrae (T10 – L5) (angle measured from the vertebral endplate to the
5
horizontal[40,41]). Patients were also categorized according to the SRS-Schwab classification by
6
coronal curve type and sagittal modifiers[42].
7
Statistical analysis
8
Descriptive statistics were used to characterize mean age, gender distribution, BMI,
9
baseline radiographic parameters, and baseline HRQOLs in the overall group. Baseline
10
differences between the patients that reached MCID (rMCID) or missed MCID (mMCID) at 2
11
years in SRS Activity or SRS Pain were identified. MCID analysis was performed to determine
12
how many patients reached an MCID in HRQOL other than SRS Activity and SRS Pain at 2
13
years. For these HRQOL, rMCID and mMCID cohorts were compared utilizing risk ratios (RR),
14
calculated by contingency tables with 95% confidence intervals. Univariate analysis was used to
15
identify possible predictors of reaching MCID, and were considered significant if the P-value
16
was <0.05. For subsequent multivariate analysis, variables with a P-value <0.2 but >0.05 were
17
also included. Backward stepwise and forward stepwise multivariate logistic regression analyses
18
were subsequently performed for modeling and confirmation of identified predictors and a P-
19
value of <0.05 was considered statistically significant. All data were analyzed using SPSS 20.0
20
(SPSS Inc., Chicago, IL).
21
9 Page 9 of 28
1
RESULTS
2
Patient Characteristics
3
There were 447 patients in the database, of whom 371 were eligible for 2 year follow up.
4
Of those, there were 5 patients who crossed over to the surgical arm, 2 deaths, 149 who were lost
5
to follow up, and 215 with 2-year data who were included for analysis. The non-operative
6
patients who were excluded were younger (47.6 years vs. 52.6 years; P=0.003) and had poorer
7
scores for baseline ODI (26.2 vs. 22.5; P= 0.035), SF-36 PCS (40.5 vs. 43.5; P=0.007), and
8
SRS-22 Activity (3.7 vs. 3.9; P=0.012). However, these differences between HRQOL were not
9
clinically significant, and there were no differences in SRS-22 Pain (3.2 vs. 3.4; P=0.142). The
10
cohort of 215 patients included had mean age of 52.6 ± 16.0 years, height of 164.2 ± 9.0 cm, and
11
BMI of 25.3 ± 5.7 kg/m2. Eighty-seven percent were women (n=187) and the overall cohort
12
(90%) was predominantly Caucasian (10% Asian, Black or Hispanic). Additional baseline
13
characteristics including prior spine surgery (18%, n =39); years of spine problems; prior hip
14
replacements (3%, n=6), CCI, and SRS-Schwab classification were also included (Figure 1).
15
Baseline HRQOL scores demonstrated moderate disability and severity of clinical
16
symptoms. Mean values for baseline HRQOL were ODI 22.5 ± 15.8; SF-36 PCS 43.3 ± 9.7; SRS
17
Activity 3.9 ± 0.8; SRS Pain 3.4 ± 0.9; SRS Appearance 3.3 ± 0.8; and SRS Mental 3.8 ± 0.8.
18
Baseline spino-pelvic parameters included SS 35.3 ± 10.2; PT 19.0 ± 10.2; PI 54.3 ± 12.7; LL
19
50.1 ± 15.9; PI-LL 4.3 ± 16.8; SVA 20.52 ± 52.4mm; and coronal TL Cobb 33.8 ± 16.0.
20
Baseline SRS Pain score significantly correlated with SRS Activity score (R=0.737, P<0.001).
21 22
In terms of baseline MCID deficit, 43% (n=92) of the 215 patients had an MCID deficit in SRS
23
Activity, 64% (n=137) had an MCID deficit in SRS Pain, and 39% (n=84) had an MCID deficit 10 Page 10 of 28
1
in both SRS Activity and Pain. Fourteen percent of patients (n=30) did not have MCID deficit
2
below that of the normative reference population.
3
In the overall cohort of 215 patients, at 2 years there were no significant differences from
4
baseline in any HRQOL. Additionally, there were no significant radiographic changes from
5
baseline to 2 years in TL Cobb or any spino-pelvic parameter except for SVA, which worsened
6
slightly from 20.5 to 25.9 mm (P<0.001). Moreover, there was no significant change in PI-LL.
7
At 2 years, SRS Pain still significantly correlated with SRS Activity (R=0.688, P<0.001).
8
MCID Analysis
9
Among the entire group of 215 non-operative patients, 40% (n=86) reached MCID in
10
SRS Activity or Pain (rMCID), while 60% (n=129) missed MCID (mMCID) in SRS Activity or
11
Pain. Only 13% (n=29) reached MCID in both SRS Activity and Pain. Further analysis of the
12
entire cohort revealed that a select number of patients reached MCID in ODI (n=16), SF-36 PCS
13
(n=36), SRS Appearance (n=20), and SRS Mental (n=44). Among the 84 patients who had a baseline MCID deficit in both SRS Activity and SRS
14 15
Pain, 54% (n=45) reached MCID in SRS Activity or Pain while 46% (n=39) missed MCID.
16
The rMCID group differed significantly from the mMCID group in ODI, SF-36 PCS,
17
SRS Appearance, and SRS Mental (P<0.05). The rMCID group was also more likely to reach
18
MCID at 2 years in these HRQOL (Table 1).
19
20
Multivariate Analysis
21
There were no statistically significant differences in age, BMI, or baseline SVA between
22
rMCID and mMCID cohorts. Significant predictors of reaching MCID included a worse SRS 11 Page 11 of 28
1
Pain score, a smaller coronal TL Cobb, lower SS, and less LL (Table 2). The strictest prediction
2
model (backward and subsequent forward stepwise regression) demonstrated that lower baseline
3
SRS Pain scores and smaller TL Cobb angles were significant predictors of reaching MCID.
4
While univariate analysis demonstrated significant differences in baseline curve type and
5
2-year PI-LL values between rMCID and mMCID, neither SRS-Schwab classification nor PI-LL
6
were predictors in the multivariate models. Multivariate analysis of the 84 patients who had both
7
baseline SRS-Activity and Pain deficits revealed that the 45 patients who did reach MCID at 2
8
years had less vertebral obliquity than the 39 patients who missed an MCID (Table 3). This
9
analysis was also confirmed with forward logistic regression.
10
SRS Pain Trend
11
Trends in SRS Pain score were recorded from baseline to 2 years. There were no
12
significant differences in SRS Pain in the overall cohort. However, the rMCID group underwent
13
a significant improvement in SRS Pain (3.0 to 3.7; P<0.001), whereas the mMCID group
14
underwent a significant decrease in SRS Pain (3.6 to 3.3; P<0.001) (Figure 2). There were no
15
significant changes in TL Cobb in the overall group or in individual cohorts.
16
DISCUSSION
17
Summary of Key Findings
18
Patients that noted a clinically important improvement from non-operative treatment and
19
reached an MCID in SRS Activity or Pain had greater baseline pain and smaller coronal
20
deformity in the thoracolumbar region. In the subset of non-operative patients who were below
21
the normative population and needed improvement, less vertebral obliquity was a predictor of
12 Page 12 of 28
1
reaching an MCID, which is consistent with a previous study by Schwab et al[40]. Pain and
2
thoracolumbar deformity are predictors for reaching or missing MCID in a non-surgical setting.
3
MCID helps identify patients that exhibited a clinically noticeable improvement in pain
4
despite the fact that the pain trend appeared to show no change over 2 years. Analyzing the non-
5
operative group as a whole suggested that non-operative patients improve minimally in pain, but
6
a closer MCID analysis shows that the mean was actually masking a subset of patients who
7
improved. 84 patients had an MCID deficit in both SRS Activity and SRS Pain and that up to
8
54% (n=45) of these severely symptomatic patients improve in pain or activity with a clinically
9
noticeable benefit from non-operative treatment at 2 years.
10
Non-operative treatment is beneficial if applied to selected patients with moderate
11
disability. Moreover, when a subset of patients had clinically noticeable benefit from non-
12
operative treatment, the final clinical outcome was limited to improvement in 1 MCID. In coarse
13
comparison with the literature, patients underwent surgical treatment gained better final clinical
14
status in comparison to non-operative treatment [11,12,23,43]. Further studies should compare
15
both treatments in historically and deformity controlled cohorts.
16
Pain and Disability
17
The clinical presentations of ASD cover a broad spectrum of pain and disability [44]. It is
18
of importance to determine the drivers of medical care sought by the patient, and match them
19
with the treatment offered by the physician [7,32,33,44]. Pekmezci et al noted that the functional
20
status of the patients appeared to be a more important determinant than pain, in pursuing
21
operative treatment. Pain is complex and one of the significant components of disability, such as
22
functional, emotional, social and mental components.
13 Page 13 of 28
1
Moreover, understanding of the pathomechanism and identifying the pattern of pain is
2
crucial because it can profoundly impact the choice of treatment[32,45]. In ASD, back pain can
3
be secondary to instability of spinal segments, dynamic stretching of nerve roots, or axial pain
4
[32,45–49]. Thus, it is fundamental to distinguish the main driver of disability in ASD patients
5
undergoing decision making for treatment.
6
Thoracolumbar Deformity
7
The significant impact of sagittal plane deformity on quality of life [33,50] does not
8
necessarily mean focal coronal plane deformities are clinically irrelevant. Magnitude of TL
9
deformity in the current study proved to be an important radiographic parameter. While there
10
was no direct correlation between TL Cobb and SRS Pain, less coronal deformity in the TL
11
region and greater baseline pain were significant predictors for reaching an MCID. Schwab et al
12
demonstrated that of all the coronal curve types, TL curves had the highest percentage of patients
13
who received surgical treatment[51]. Prior studies have also shown that thoracolumbar and
14
lumbar curves generate less favorable HRQOL scores than thoracic curves[50].
15
Because operative treatment can provide significant improvement in HRQOLs for
16
sagittally malaligned patients, this subsequent shift of operative focus in ASD treatment towards
17
sagittal plane correction has resulted in more pure coronal deformities receiving non-operative
18
care. The findings in this study suggest that in cases of moderate sagittal malalignment, larger
19
TL coronal deformities negatively impact a patient's ability to improve with non-operative care,
20
supporting the notion that there is a complex relationship of TL deformity, disability, and
21
pain[50].
14 Page 14 of 28
1
Limitations
2
A primary limitation was study design. Although the database was prospectively
3
collected, the analysis is still limited by the retrospective design of the present study. In addition,
4
all data are limited by the capabilities of the HRQOL; SRS Pain scores may be poor due to
5
radiographic findings, radicular symptoms, neurogenic claudication, or other factors. Moreover,
6
patients in this cohort were enrolled in this study as an observational ASD cohort, rather than
7
enrolled specifically to evaluate non-operative treatment. There was also insufficient data
8
available regarding treatment modality. Thus, there is a need for high-level prospective studies
9
specifically focusing on non-operative patients with ASD and the various forms of non-operative
10
treatment. Future work including randomized non-operative treatment options would
11
undoubtedly be valuable. Additionally, longer-term follow-up would be important to validate
12
these findings, as non-operative outcomes for ASD represent a long-term concern for patients
13
and physicians. Finally, by design, this study did not include patients who crossed over from the
14
non-operative to operative groups. There were only 5 patients who crossed over, thus limiting the
15
potential for any high-powered analysis. Future research dedicated to this subset of patients
16
would also be instrumental in understanding outcomes in operative and non-operative patients.
17
18
CONCLUSIONS
19
MCID analysis is a useful tool to unmask the subset of patients that might improve to a
20
clinically relevant degree from non-operative treatment. A higher baseline level of pain and
21
smaller thoracolumbar curve are significant predictors for identifying patients that may reach an
22
MCID when treated non-operatively. A subset of severely symptomatic patients may be more
23
likely to reach an MCID if they have less vertebral obliquity. Pain is quite often the presenting 15 Page 15 of 28
1
complaint in patients with ASD and the focus of non-operative ASD treatment. Improvement in
2
function or pain at the 2-year time point in a subset of patients is possible. While it is still
3
uncertain if this group will continue to function reasonably in the upcoming years, HRQOL
4
scores remain important parameters in quantifying patient improvement/deterioration.
5
REFERENCES
6
[1]
Schwab FJ, Dubey A, Gamez L, El Fegoun AB, Hwang K, Pagala M, et al. Adult
7
scoliosis: prevalence, SF-36, and nutritional parameters in an elderly volunteer population.
8
Spine (Phila Pa 1976) 2005;30:1082–5.
9
[2]
Schwab FJ, Lafage V, Farcy JP, Bridwell KH, Glassman SSD, Shainline MR. Predicting
10
outcome and complications in the surgical treatment of adult scoliosis. Spine (Phila Pa
11
1976) 2008;33:2243–7. doi:10.1097/BRS.0b013e31817d1d4e.
12
[3]
Schwab FJ, Dubey A, Pagala M, Gamez L, Farcy JP. Adult scoliosis: a health assessment
13
analysis
14
doi:10.1097/01.BRS.0000049924.94414.BB.
15
[4]
SF-36.
Spine
(Phila
Pa
1976)
2003;28:602–6.
Robin GC, Span Y, Steinberg R, Makin M, Menczel J. Scoliosis in the elderly: a followup study. Spine (Phila Pa 1976) 1982;7:355–9.
16
17
by
[5]
McCarthy I, Hostin R, O’Brien M, Saigal R, Ames CP. Health economic analysis of adult
18
deformity
surgery.
Neurosurg
19
doi:10.1016/j.nec.2012.12.005.
Clin
N
Am
2013;24:293–304.
16 Page 16 of 28
1
[6]
2007;18:403–5. doi:10.1016/j.nec.2007.03.001.
2
3
Burton DC, Glattes RC. Measuring outcomes in spinal deformity. Neurosurg Clin N Am
[7]
Bridwell KH, Cats-baril W, Harrast J, Berven S, Glassman S, Farcy J, et al. The validity
4
of the SRS-22 instrument in an adult spinal deformity population compared with the
5
Oswestry and SF-12: a study of response distribution, concurrent validity, internal
6
consistency, and reliability. Spine (Phila Pa 1976) 2005;30:455–61.
7
[8]
Bridwell KH, Berven S, Glassman SD, Hamill C, Horton W, Lenke LG, et al. Is the SRS-
8
22 instrument responsive to change in adult scoliosis patients having primary spinal
9
deformity
Spine
(Phila
Pa
1976)
2007;32:2220–5.
doi:10.1097/BRS.0b013e31814cf120.
10
11
surgery?
[9]
Berven S, Deviren V, Demir-Deviren S, Hu SS, Bradford DS. Studies in the modified
12
Scoliosis Research Society Outcomes Instrument in adults: validation, reliability, and
13
discriminatory capacity. Spine (Phila Pa 1976) 2003;28:2164–9; discussion 2169.
14
doi:10.1097/01.BRS.0000084666.53553.D6.
15
[10] Asher MA, Lai SM, Glattes RC, Burton DC, Alanay A, Bago J. Refinement of the SRS-22
16
Health-Related Quality of Life questionnaire Function domain. Spine (Phila Pa 1976)
17
2006;31:593–7. doi:10.1097/01.brs.0000201331.50597.ea.
18
[11] Bridwell KH, Glassman S, Horton W, Shaffrey C, Schwab FJ, Zebala LP, et al. Does
19
treatment (nonoperative and operative) improve the two-year quality of life in patients
20
with adult symptomatic lumbar scoliosis: a prospective multicenter evidence-based
17 Page 17 of 28
1
medicine
study.
Spine
2
doi:10.1097/BRS.0b013e3181a8fdc8.
(Phila
Pa
1976)
2009;34:2171–8.
3
[12] Smith JS, Shaffrey CI, Berven S, Glassman S, Hamill C, Horton W, et al. Operative
4
versus nonoperative treatment of leg pain in adults with scoliosis: a retrospective review
5
of a prospective multicenter database with two-year follow-up. Spine (Phila Pa 1976)
6
2009;34:1693–8. doi:10.1097/BRS.0b013e3181ac5fcd.
7
[13] Smith JS, Shaffrey CI, Berven S, Glassman SD, Hamill C, Horton W, et al. Improvement
8
of back pain with operative and nonoperative treatment in adults with scoliosis.
9
Neurosurgery
2009;65:86–93;
discussion
93–4.
doi:10.1227/01.NEU.0000347005.35282.6C.
10
11
[14] Albert T, Purtill J, Mesa J, Mclntosh T, Balderston R. Health outcome assessment before
12
and after adult deformity surgery: a prospective study. Spine (Phila Pa 1976)
13
1995;20:2002–4; discussion p2005.
14
[15] Glassman SD, Hamill CL, Bridwell KH, Schwab F, Dimar JR, Lowe TG. The impact of
15
perioperative complications on clinical outcome in adult deformity surgery. Spine (Phila
16
Pa 1976) 2007;32:2764–70. doi:10.1097/BRS.0b013e31815a7644.
17
[16] Cho SK, Bridwell KH, Lenke LG, Cho W, Zebala LP, Pahys JM, et al. Comparative
18
analysis of clinical outcome and complications in primary versus revision adult scoliosis
19
surgery. Spine (Phila Pa 1976) 2012;37:393–401. doi:10.1097/BRS.0b013e31821f0126.
18 Page 18 of 28
1
[17] Carragee EJ, Cheng I. Minimum acceptable outcomes after lumbar spinal fusion. Spine J 2010;10:313–20. doi:10.1016/j.spinee.2010.02.001.
2
3
[18] Copay AG, Subach BR, Glassman SD, Polly DW, Schuler TC. Understanding the
4
minimum clinically important difference: a review of concepts and methods. Spine J
5
2007;7:541–6. doi:10.1016/j.spinee.2007.01.008.
6
[19] Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989;10:407–15.
7
8
[20] Copay AG, Glassman SD, Subach BR, Berven S, Schuler TC, Carreon LY. Minimum
9
clinically important difference in lumbar spine surgery patients: a choice of methods using
10
the Oswestry Disability Index, Medical Outcomes Study questionnaire Short Form 36, and
11
pain scales. Spine J 2008;8:968–74. doi:10.1016/j.spinee.2007.11.006.
12
[21] Parker SL, Mendenhall SK, Shau DN, Adogwa O, Anderson WN, Devin CJ, et al.
13
Minimum clinically important difference in pain, disability, and quality of life after neural
14
decompression and fusion for same-level recurrent lumbar stenosis: understanding clinical
15
versus
16
doi:10.3171/2012.1.SPINE11842.
statistical
significance.
J
Neurosurg
Spine
2012;16:471–8.
17
[22] Parker SL, Adogwa O, Mendenhall SK, Shau DN, Anderson WN, Cheng JS, et al.
18
Determination of minimum clinically important difference (MCID) in pain, disability, and
19
quality of life after revision fusion for symptomatic pseudoarthrosis. Spine J
20
2012;12:1122–8. doi:10.1016/j.spinee.2012.10.006.
19 Page 19 of 28
1
[23] Youssef J a, Orndorff DO, Patty C a, Scott M a, Price HL, Hamlin LF, et al. Current
2
Status of Adult Spinal Deformity. Glob Spine J 2013;3:51–62. doi:10.1055/s-0032-
3
1326950.
4
[24] Smith JS, Fu K-M, Urban P, Shaffrey CI. Neurological symptoms and deficits in adults
5
with scoliosis who present to a surgical clinic: incidence and association with the choice
6
of operative versus nonoperative management. J Neurosurg Spine 2008;9:326–31.
7
doi:10.3171/SPI.2008.9.10.326.
8
[25] Smith JS, Shaffrey CI, Glassman SD, Berven SH, Schwab FJ, Hamill CL, et al. Risk-
9
benefit assessment of surgery for adult scoliosis: an analysis based on patient age. Spine (Phila Pa 1976) 2011;36:817–24. doi:10.1097/BRS.0b013e3181e21783.
10
11
[26] Smith JS, Shaffrey CI, Glassman SD, Carreon LY, Schwab FJ, Lafage V, et al. Clinical
12
and radiographic parameters that distinguish between the best and worst outcomes of
13
scoliosis surgery for adults. Eur Spine J 2013;22:402–10. doi:10.1007/s00586-012-2547-
14
x.
15
[27] Moal B, Lafage V, Smith JS, Ames CP, Mundis GM, Terran JS, et al. Clinical
16
Improvement Through Surgery for Adult Spinal Deformity (ASD): What Can Be
17
Expected and Who is Likely to Benefit Most? Proc. NASS 27th Annu. Meet., vol. 12, The
18
Spine Journal; 2012, p. 99S – 165S. doi:10.1016/j.spinee.2012.08.395.
19
[28] Liu S, Schwab FJ, Smith JS, Klineberg E, Ames CP, Mundis G, et al. Likelihood of
20
reaching minimal clinically important difference in adult spinal deformity: a comparison
21
of operative and nonoperative treatment. Ochsner J 2014;14:67–77. 20 Page 20 of 28
1
[29] Berven S, D.V., Demir-Deviren S, Hu S BD. Minimal Clinically Important Difference in
2
Adult Spinal Deformity: How much change is significant? IMAST Present. Int. Meet.
3
Adv. Surg. Tech. IMAST, Banff, Canada: 2005.
4
[30] Blondel B, Schwab FJ, Ungar B, Smith JS, Bridwell KH, Glassman SD, et al. Impact of
5
magnitude and percentage of global sagittal plane correction on health-related quality of
6
life
7
doi:10.1227/NEU.0b013e31825d20c0.
at
2-years
follow-up.
Neurosurgery
2012;71:341–8;
discussion
348.
8
[31] Smith JS, Klineberg E, Schwab FJ, Shaffrey CI, Moal B, Ames CP, et al. Change in
9
Classification Grade by the SRS-Schwab Adult Spinal Deformity Classification Predicts
10
Impact on Health-Related Quality of Life Measures: Prospective Analysis of Operative
11
and
12
doi:10.1097/BRS.0b013e31829ec563.
13
Non-operative
Treatment.
Spine
(Phila
Pa
1976)
2013;38:1663–71.
[32] Kotwal S, Pumberger M, Hughes A, Girardi F. Degenerative scoliosis: a review. HSS J 2011;7:257–64. doi:10.1007/s11420-011-9204-5.
14
15
[33] Glassman SD, Bridwell K, Dimar JR, Horton W, Berven S, Schwab FJ. The impact of
16
positive sagittal balance in adult spinal deformity. Spine (Phila Pa 1976) 2005;30:2024–9.
17
[34] Baldus C, Bridwell K, Harrast J, Shaffrey C, Ondra S, Lenke L, et al. The Scoliosis
18
Research Society Health-Related Quality of Life (SRS-30) age-gender normative data: an
19
analysis of 1346 adult subjects unaffected by scoliosis. Spine (Phila Pa 1976)
20
2011;36:1154–62. doi:10.1097/BRS.0b013e3181fc8f98.
21 Page 21 of 28
1
[35] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying
2
prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis
3
1987;40:373–83.
4
[36] Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:473–83.
5
6
[37] Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy 1980;66:271–3.
7
8
[38] Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine (Phila Pa 1976) 2000;25:2940–52; discussion 2952.
9
10
[39] Champain S, Benchikh K, Nogier A, Mazel C, Guise J De, Skalli W. Validation of new
11
clinical quantitative analysis software applicable in spine orthopaedic studies. Eur Spine J
12
2006;15:982–91. doi:10.1007/s00586-005-0927-1.
13
[40] Schwab FJ, Smith V a, Biserni M, Gamez L, Farcy J-PC, Pagala M. Adult scoliosis: a quantitative radiographic and clinical analysis. Spine (Phila Pa 1976) 2002;27:387–92.
14
15
[41] Schwab FJ, el-Fegoun AB, Gamez L, Goodman H, Farcy J. A lumbar classification of
16
scoliosis in the adult patient: preliminary approach. Spine (Phila Pa 1976) 2005;30:1670–
17
3.
18
[42] Schwab F, Ungar B, Blondel B, Buchowski J, Coe J, Deinlein D, et al. Scoliosis Research
19
Society-Schwab adult spinal deformity classification: a validation study. Spine (Phila Pa
20
1976) 2012;37:1077–82. doi:10.1097/BRS.0b013e31823e15e2. 22 Page 22 of 28
1
[43] Bridwell KH, Baldus C, Berven S, Edwards C, Glassman S, Hamill C, et al. Changes in
2
radiographic and clinical outcomes with primary treatment adult spinal deformity
3
surgeries from two years to three- to five-years follow-up. Spine (Phila Pa 1976)
4
2010;35:1849–54. doi:10.1097/BRS.0b013e3181efa06a.
5
[44] Schwab FJ, Lafage V, Farcy J-P, Bridwell KH, Glassman SD, Ondra S, et al. Surgical
6
rates and operative outcome analysis in thoracolumbar and lumbar major adult scoliosis:
7
application of the new adult deformity classification. Spine (Phila Pa 1976)
8
2007;32:2723–30. doi:10.1097/BRS.0b013e31815a58f2.
9
[45] Bradford DS, Tay BK, Hu SS. Adult scoliosis: surgical indications, operative
10
management, complications, and outcomes. Spine (Phila Pa 1976) 1999;24:2617–29.
11
[46] Van Dam BE. Nonoperative Treatment of Adult Scoliosis. Orthop Clin North Am 1988;19:347–51.
12
13
[47] Jackson RP, Simmons EH, Stripinis D. Coronal and sagittal plane spinal deformities
14
correlating with back pain and pulmonary function in adult idiopathic scoliosis. Spine
15
(Phila Pa 1976) 1989;14:1391–7.
16
[48] Kostuik JP, Bentivoglio J. The incidence of low-back pain in adult scoliosis. Spine (Phila Pa 1976) 1981;6:268–73.
17
18
[49] Vanderpool DW, James JI, Wynne-Davies R. Scoliosis in the elderly. J Bone Joint Surg Am 1969;51:446–55.
19
23 Page 23 of 28
1
[50] Glassman SD, Berven S, Bridwell K, Horton W, Dimar JR. Correlation of radiographic
2
parameters and clinical symptoms in adult scoliosis. Spine (Phila Pa 1976) 2005;30:682–
3
8.
4
[51] Schwab FJ, Farcy J, Bridwell K, Berven S, Glassman S, Harrast J, et al. A clinical impact
5
classification of scoliosis in the adult. Spine (Phila Pa 1976) 2006;31:2109–14.
6
doi:10.1097/01.brs.0000231725.38943.ab.
7
24 Page 24 of 28
1
FIGURE LEGENDS
2 3
Figure 1: Baseline characteristics of the non-operative cohort included history of spine problems
4
and Charlson Comorbidity Index, and Scoliosis Research Society (SRS)-Schwab classification.
5
SRS-Schwab
6
thoracolumbar/lumbar curves >30), T (thoracic curve only, with lumbar curve <30), L
7
(thoracolumbar/lumbar curve only, with thoracic curve <30), or N (no major coronal deformity
8
with all coronal curves <30). SRS-Schwab Sagittal Modifiers include pelvic incidence minus
9
lumbar lordosis mismatch (PI-LL), sagittal vertical axis (global alignment, GA), and pelvic tilt
10
(PT). Grade 0 denotes normal alignment, Grade 1 denotes moderate deformity, and Grade ++
11
denotes severe deformity.
curve
types
are
denoted
by
D
(double
curve
with
thoracic
and
12
13
Figure 2. Pain trends demonstrated no change in the mean pain level of the overall non-operative
14
group. However, the rMCID sub-group had a significant decrease in pain level at 2 years (SRS
15
Pain score 3.0 to 3.7; P<0.001), while the mMCID subgroup had a significant increase in pain
16
level at 2 years (SRS Pain score 3.6 to 3.3; P<0.001). The rMCID subgroup had a significantly
17
lower SRS Pain score at baseline than the mMCID subgroup (P<0.001).
18
25 Page 25 of 28
1
Table 1: The number of patients in rMCID (n=86) and mMCID groups (n=129) who reached MCID in
2
other HRQOL (aside from SRS-Pain or SRS-Activity) at 2 years. The rMCID cohort was more likely to
3
reach MCID in other HRQOL, such as ODI, SF-36 PCS, SRS appearance, and SRS mental. Confidence
4
intervals set at 95%.
5 MCID in Additional HRQOL Domains (N, %) rMCID
mMCID
n=86
n=129
ODI
14 (16.3%)
SF-36 PCS
Risk Ratio
Confidence Interval
2 (1.6%)
10.3
(9.3 – 11.3)
29 (33.7%)
7 (5.4%)
5.6
(4.7 – 6.7)
SRS appearance
15 (17.4%)
5 (3.9%)
4.5
(4.1 – 5.0)
SRS mental
26 (30.2%)
18 (14.0%)
2.1
(1.9 – 2.5)
6 7
26 Page 26 of 28
1 2 3 4 5
Table 2: Comparisons of predictors for rMCID (reached minimal clinically important difference) and mMCID (missed MCID). Predictors shown were modeled using a backwards stepwise logistic regression. *Denotes predictors that were further confirmed using a forwards stepwise logistic regression. SRS-Pain = Scoliosis Research Society Pain Score. TL Cobb = Thoracolumbar Cobb angle. SS = Sacral Slope. LL = Lumbar Lordosis.
6 Predictor
rMCID
mMCID
p-value
(multivariate)
n=86
n=129
(multivariate)
*SRS Pain
3.0
3.6
0.001
*TL Cobb (coronal)
29.6
36.5
0.007
SS
33.1
36.4
0.847
LL
46.5
52.8
0.875
7 8
27 Page 27 of 28
1
Table 3. Predictors shown were modeled using a backwards and forwards stepwise logistic regression.
2 Predictor
rMCID
mMCID
p-value
(multivariate)
n=45
n=39
(multivariate)
Maximum vertebral obliquity
24.3
27.2
0.035
3 4 5
28 Page 28 of 28
S1529-9430(15)01628-9 http://dx.doi.org/doi: 10.1016/j.spinee.2015.10.043 SPINEE 56673
To appear in:
The Spine Journal
Received date: Revised date: Accepted date:
21-6-2015 18-9-2015 22-10-2015
Please cite this article as: Shian Liu, Bassel Diebo, Jensen K. Henry, Justin S. Smith, Richard Hostin, Matthew E. Cunningham, Gregory Mundis, Christopher P. Ames, Douglas Burton, Shay Bess, Behrooz Akbarnia, Robert Hart, Peter G. Passias, Frank J. Schwab, Virginie Lafage, on behalf of the International Spine Study Group (ISSG), The benefit of non-operative treatment for adult spinal deformity: identifying predictors for reaching a minimal clinically important difference, The Spine Journal (2015), http://dx.doi.org/doi: 10.1016/j.spinee.2015.10.043. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Title: The Benefit of Non-Operative Treatment for Adult Spinal Deformity: Identifying
2
Predictors for Reaching a Minimal Clinically Important Difference
3
Authors:
4
Shian Liu, MD1; Bassel Diebo, MD1; Jensen K. Henry, BA1; Justin S. Smith, MD PhD2; Richard
5
Hostin, MD3; Matthew E. Cunningham, MD4; Gregory Mundis, MD5; Christopher P. Ames,
6
MD6; Douglas Burton, MD7; Shay Bess, MD8; Behrooz Akbarnia, MD9; Robert Hart, MD10;
7
Peter G. Passias, MD1; Frank J. Schwab, MD1; Virginie Lafage, PhD1; on behalf of the
8
International Spine Study Group (ISSG)
9
Affiliations:
10
1
Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, 301 East 17th St., New
11
York, NY, United States, 10003
12
2
13
United States, PO Box: 800212
14
3
15
Alliance Blvd, #810, Plano TX, USA
16
4
17
States. 535 East 70th St., New York, NY 10021, USA
18
5
Scripps Clinic Torrey Pines, 10666 N Torrey Pines Rd, La Jolla, CA 92037
19
6
Department of Neurosurgery, University of California, San Francisco Medical Center, San
20
Francisco, CA, United States. 400 Parnassus Street, San Francisco, CA, USA.
Department of Neurosurgery, University of Virginia Medical Center, Charlottesville, VA,
Department of Orthopaedic Surgery, Baylor Scoliosis Center, Plano, TX, United States 4708
Department of Orthopaedic Surgery, Hospital for Special Surgery, New York, NY, United
1 Page 1 of 28
1
7
Department of Orthopaedic Surgery, University of Kansas Medical Center, Kansas City, KS
2
8
Rocky Mountain Scoliosis and Spine Center, Denver, CO 80205, USA
3
9
Department of Orthopaedic Surgery, San Diego Center for Spinal Disorders, La Jolla, CA
4
10
5
Corresponding Author:
6
Virginie Lafage, PhD
7
NYU Hospital for Joint Diseases
8
306 East 15th St., Suite 1F
9
New York, NY 10003
Department of Orthopaedic Surgery, Oregon Health Sciences University, Portland, OR
10
Email: [email protected]
11
Phone: 646-794-9640
12
Fax: 646-602-6927
13 14
Acknowledgements: Data collection for this study was supported by a grant from Depuy to the
15
International Spine Study Group Foundation. No specific sources of funding were linked to this
16
work in particular.
17
ABSTRACT
18
Background Context: Adult spinal deformity (ASD) patients may gain a MCID in one or more
19
of the HRQOL instruments without surgical intervention. This study identifies baseline
20
characteristics of this subset of non-operative patients and proposes predictors of those most
21
likely to benefit.
2 Page 2 of 28
1
Purpose: Determine factors that affect likelihood of non-operative patients to reach minimum
2
clinically important difference (MCID).
3
Study Design/Setting: Retrospective review of prospective, multi-center database.
4
Patient Sample: Non-operative ASD patients.
5
Outcome Measures: Health-related quality of life measures (HRQOL), including the Scoliosis
6
Research Society (SRS)-22 questionnaire.
7
Methods: Multicenter database of 215 non-operative patients with ASD and minimum 2-year
8
follow-up. Using a multivariate analysis, two groups were compared to identify possible
9
predictors: those that reached an MCID in SRS Pain or Activity (n=86) at 2 years, and those who
10
did not reach MCID (n=129). Subgroup multivariate analysis of patients with a deficit (potential
11
improvement) in both SRS Pain and Activity (n=84) was performed. Data collection was
12
supported by a grant from Depuy for the International Spine Study Group Foundation.
13
Results: At baseline, the non-operative patients that reached MCID had a significantly lower
14
SRS Pain score (3.0 vs 3.6), smaller thoracolumbar (TL) Cobb angle (29.6 vs. 36.5; 87 patients
15
with SRS-Schwab classification Lumbar or Double), sacral slope (33.1 vs. 36.4), and less
16
lumbar lordosis (46.5 vs. 52.8) (all P<0.05). SRS Pain and TL Cobb were significant predictors
17
of reaching MCID. PI-LL was significant on univariate analysis but not by multivariate (7.5 vs.
18
2.6; P=0.14). In the subset of severely disabled patients, worse vertebral obliquity was a
19
predictor for not achieving MCID (P<0.05).
20
Conclusions: Non-operative ASD patients who achieved an MCID in SRS Activity or Pain had
21
a lower baseline SRS Pain Score and less coronal deformity in the TL region. Greater baseline 3 Page 3 of 28
1
pain offers significant room for potential improvement, which may be important in identifying
2
ASD patients who have the potential to reach an MCID non-operatively. Coronal deformities in
3
the TL region, and associated vertebral obliquity may negatively impact improvement potential
4
with non-operative care.
5
6
Key words: spinal deformity; non-operative; treatment; outcomes; HRQOL; health-related
7
quality of life
8
4 Page 4 of 28
1
MANUSCRIPT
2
INTRODUCTION
3
Spinal deformity in the skeletally mature patient, with an incidence up to 32% in adults
4
and 60% in the elderly, is becoming a more commonly recognized condition among both spine
5
surgeons and general healthcare providers [1–4]. Physicians across all specialties are
6
increasingly recognizing the impact of adult spinal deformity (ASD) on patients and the
7
healthcare system [5]. Healthcare costs for treating spinal deformity are rising [5]. While large
8
gains have been made in the evolution of surgical treatment, non-operative management in
9
certain patients is prudent and necessary in everyday practice. The economic issues of delivering
10
care involve a balance between surgery with evident clinical improvement and non-operative
11
management which may be effective for some patients.
12
The treatment of ASD can be assessed quantitatively and tracked over time using
13
validated patient-reported outcomes such as the Scoliosis Research Society questionnaire (SRS-
14
22r), the Oswestry Disability Index (ODI), and the Short Form 36 questionnaire (SF-36) [3,5–
15
10]. Since the incorporation of these health-related quality of life (HRQOL) measures into
16
medical practice and research, multiple reports have shown the significant impact of surgical
17
intervention for ASD [11–16].
18
With the inception of HRQOLs arose the concern that improvements in these scores do
19
not necessarily translate into a clinically discernible difference that is of significance to the
20
patient. Hence, the concept of a minimal clinically important difference (MCID) has been
21
introduced in the spine literature to quantify the absolute minimum change that can be
22
considered a success [17–19]. Many studies have demonstrated a clear advantage in gaining
5 Page 5 of 28
1
MCID after surgical treatment for spinal pathologies such as spondylolisthesis, disc pathology,
2
spinal stenosis and ASD, with comparatively poor improvement with non-operative care [17,20–
3
23].
4
While surgical treatment can improve both pain and disability [11–13,24,25], there are
5
risks involved [26,27].Thus, it is important to thoroughly assess all treatment options, including
6
non-operative care, which has not been as well studied in the literature. Often non-operative
7
patients are grouped together in cohort analyses, but it is possible that non-operative
8
management may actually offer acceptable quality of life maintenance or improvements for a
9
certain subset of
patients[23,28]. A recent study on 464 patients with ASD found that a
10
subgroup of non-operatively treated patients actually improved in MCID: up to 52% (n=117)
11
reached MCID in at least 1 HRQOL, with 20% reaching MCID in SRS Activity and 24%
12
reaching MCID in SRS Pain[28].
13
The purpose of the present study was to characterize a subset of patients with ASD that
14
demonstrated improvement with non-operative management and assess for baseline
15
characteristics that may distinguish them from those that failed to significantly improve with
16
non-operative treatment.
17
18
METHODS
19
Study Design
20
This was a retrospective analysis of a prospective, multicenter consecutive case series of
21
215 patients with ASD. Internal review board approval was obtained from all 11 high volume 6 Page 6 of 28
1
centers that care for patients with ASD prior to data collection and informed consent was
2
obtained from each patient to be included in the study. Inclusion criteria for the database were
3
age over 18 years and radiographic evidence of spinal deformity, defined as one or more of the
4
following: coronal Cobb ≥ 20, sagittal vertical axis (SVA) ≥5cm, pelvic tilt ≥25, thoracic
5
kyphosis (TK) ≥ 60. For the purposes of this study, only non-operative patients who completed
6
2-year clinical and radiographic follow up were analyzed. Treatment was not randomized, but
7
rather was based on a combination of patient input and physician counseling on the complexities
8
and outcomes of care. The specific type of non-operative care was not specified and some of the
9
patients had observational care only. Patients with malignancy, active infection or rheumatologic
10
disease, and those with neuromuscular or congenital scoliosis were excluded. Patients who
11
crossed over to the operative arm were excluded in the analysis, as this would falsely elevate the
12
number of non-operative patients who would reach an MCID.
13
The MCID values used for each HRQOL were established by prior studies using the
14
external anchor methodology with the SF-36 health transition item[17,18,20–22,28]. Established
15
MCID values were -12.8 for ODI, +0.587 for SRS Pain, +0.375 for SRS Activity, +0.800 for
16
SRS appearance, +0.420 for SRS mental, and +4.9 for the SF-36 physical component score (SF-
17
36 PCS)[18,20,29–31].
18
Patients with ASD characteristically present with pain and disability [7,32,33]. To
19
quantify the need for improvement at baseline, MCID deficit in SRS Activity or Pain was
20
determined by being at least 1 or more MCID below a normative, age and gender matched
21
population[34]:
7 Page 7 of 28
1
Non-operative patients were also grouped into two cohorts: those that reached MCID in SRS
2
Activity or Pain at 2 years (rMCID) and those that missed MCID (mMCID) at 2 years:
3
MCID change greater than +1 was considered an improvement (reached MCID); less than -1 was
4
considered deterioration (missed MCID); and between -1 and +1 was considered unchanged
5
from baseline.
6
Data Collection
7
Patient demographic data included age, gender, race, body mass index (BMI), medical
8
co-morbidities, prior surgery, smoking status, employment status, and Charlson Comorbidity
9
Index (CCI) [35]. Radiographic measurements based on coronal and sagittal full-length standing
10
films were obtained. Patient-reported outcomes, including SRS-22r, SF-36, and ODI were
11
recorded at the initial baseline visit and at 2-year follow-up. The SRS-22r is a disease-specific
12
questionnaire with scores ranging from 0 (worst) to 5 (best). SF-36 is a generic measure of health
13
that is not age, treatment, or disease-specific; it ranges from 0 to 100, with higher scores
14
signifying a better health status [3,9,36]. The SF-36 can be divided into Physical Component
15
Scores (PCS) and Mental Component Scores (MCS). ODI is specific to disability in the lumbar
16
spine and ranges from 0 to 100, with higher numbers denoting more disability [37,38]. Data were
17
uploaded to a central electronic data capture system and queried by professional clinical research
18
monitors to ensure accuracy and reliability.
19
Radiographic Analysis Radiographic parameters at baseline were measured using SpineView software
20 21
(Laboratory of Biomechanics, ENSAM, Paris, France).[39]
Sagittal and coronal spinal
8 Page 8 of 28
1
parameters included sacral slope (SS), pelvic tilt (PT), pelvic incidence (PI), lumbar lordosis
2
(LL), sagittal vertical axis (C7-S1 SVA), maximum thoracic kyphosis (TK), thoraco-lumbar
3
Cobb angle in the coronal plane (TL Cobb), and maximum vertebral obliquity for
4
Thoracolumbar vertebrae (T10 – L5) (angle measured from the vertebral endplate to the
5
horizontal[40,41]). Patients were also categorized according to the SRS-Schwab classification by
6
coronal curve type and sagittal modifiers[42].
7
Statistical analysis
8
Descriptive statistics were used to characterize mean age, gender distribution, BMI,
9
baseline radiographic parameters, and baseline HRQOLs in the overall group. Baseline
10
differences between the patients that reached MCID (rMCID) or missed MCID (mMCID) at 2
11
years in SRS Activity or SRS Pain were identified. MCID analysis was performed to determine
12
how many patients reached an MCID in HRQOL other than SRS Activity and SRS Pain at 2
13
years. For these HRQOL, rMCID and mMCID cohorts were compared utilizing risk ratios (RR),
14
calculated by contingency tables with 95% confidence intervals. Univariate analysis was used to
15
identify possible predictors of reaching MCID, and were considered significant if the P-value
16
was <0.05. For subsequent multivariate analysis, variables with a P-value <0.2 but >0.05 were
17
also included. Backward stepwise and forward stepwise multivariate logistic regression analyses
18
were subsequently performed for modeling and confirmation of identified predictors and a P-
19
value of <0.05 was considered statistically significant. All data were analyzed using SPSS 20.0
20
(SPSS Inc., Chicago, IL).
21
9 Page 9 of 28
1
RESULTS
2
Patient Characteristics
3
There were 447 patients in the database, of whom 371 were eligible for 2 year follow up.
4
Of those, there were 5 patients who crossed over to the surgical arm, 2 deaths, 149 who were lost
5
to follow up, and 215 with 2-year data who were included for analysis. The non-operative
6
patients who were excluded were younger (47.6 years vs. 52.6 years; P=0.003) and had poorer
7
scores for baseline ODI (26.2 vs. 22.5; P= 0.035), SF-36 PCS (40.5 vs. 43.5; P=0.007), and
8
SRS-22 Activity (3.7 vs. 3.9; P=0.012). However, these differences between HRQOL were not
9
clinically significant, and there were no differences in SRS-22 Pain (3.2 vs. 3.4; P=0.142). The
10
cohort of 215 patients included had mean age of 52.6 ± 16.0 years, height of 164.2 ± 9.0 cm, and
11
BMI of 25.3 ± 5.7 kg/m2. Eighty-seven percent were women (n=187) and the overall cohort
12
(90%) was predominantly Caucasian (10% Asian, Black or Hispanic). Additional baseline
13
characteristics including prior spine surgery (18%, n =39); years of spine problems; prior hip
14
replacements (3%, n=6), CCI, and SRS-Schwab classification were also included (Figure 1).
15
Baseline HRQOL scores demonstrated moderate disability and severity of clinical
16
symptoms. Mean values for baseline HRQOL were ODI 22.5 ± 15.8; SF-36 PCS 43.3 ± 9.7; SRS
17
Activity 3.9 ± 0.8; SRS Pain 3.4 ± 0.9; SRS Appearance 3.3 ± 0.8; and SRS Mental 3.8 ± 0.8.
18
Baseline spino-pelvic parameters included SS 35.3 ± 10.2; PT 19.0 ± 10.2; PI 54.3 ± 12.7; LL
19
50.1 ± 15.9; PI-LL 4.3 ± 16.8; SVA 20.52 ± 52.4mm; and coronal TL Cobb 33.8 ± 16.0.
20
Baseline SRS Pain score significantly correlated with SRS Activity score (R=0.737, P<0.001).
21 22
In terms of baseline MCID deficit, 43% (n=92) of the 215 patients had an MCID deficit in SRS
23
Activity, 64% (n=137) had an MCID deficit in SRS Pain, and 39% (n=84) had an MCID deficit 10 Page 10 of 28
1
in both SRS Activity and Pain. Fourteen percent of patients (n=30) did not have MCID deficit
2
below that of the normative reference population.
3
In the overall cohort of 215 patients, at 2 years there were no significant differences from
4
baseline in any HRQOL. Additionally, there were no significant radiographic changes from
5
baseline to 2 years in TL Cobb or any spino-pelvic parameter except for SVA, which worsened
6
slightly from 20.5 to 25.9 mm (P<0.001). Moreover, there was no significant change in PI-LL.
7
At 2 years, SRS Pain still significantly correlated with SRS Activity (R=0.688, P<0.001).
8
MCID Analysis
9
Among the entire group of 215 non-operative patients, 40% (n=86) reached MCID in
10
SRS Activity or Pain (rMCID), while 60% (n=129) missed MCID (mMCID) in SRS Activity or
11
Pain. Only 13% (n=29) reached MCID in both SRS Activity and Pain. Further analysis of the
12
entire cohort revealed that a select number of patients reached MCID in ODI (n=16), SF-36 PCS
13
(n=36), SRS Appearance (n=20), and SRS Mental (n=44). Among the 84 patients who had a baseline MCID deficit in both SRS Activity and SRS
14 15
Pain, 54% (n=45) reached MCID in SRS Activity or Pain while 46% (n=39) missed MCID.
16
The rMCID group differed significantly from the mMCID group in ODI, SF-36 PCS,
17
SRS Appearance, and SRS Mental (P<0.05). The rMCID group was also more likely to reach
18
MCID at 2 years in these HRQOL (Table 1).
19
20
Multivariate Analysis
21
There were no statistically significant differences in age, BMI, or baseline SVA between
22
rMCID and mMCID cohorts. Significant predictors of reaching MCID included a worse SRS 11 Page 11 of 28
1
Pain score, a smaller coronal TL Cobb, lower SS, and less LL (Table 2). The strictest prediction
2
model (backward and subsequent forward stepwise regression) demonstrated that lower baseline
3
SRS Pain scores and smaller TL Cobb angles were significant predictors of reaching MCID.
4
While univariate analysis demonstrated significant differences in baseline curve type and
5
2-year PI-LL values between rMCID and mMCID, neither SRS-Schwab classification nor PI-LL
6
were predictors in the multivariate models. Multivariate analysis of the 84 patients who had both
7
baseline SRS-Activity and Pain deficits revealed that the 45 patients who did reach MCID at 2
8
years had less vertebral obliquity than the 39 patients who missed an MCID (Table 3). This
9
analysis was also confirmed with forward logistic regression.
10
SRS Pain Trend
11
Trends in SRS Pain score were recorded from baseline to 2 years. There were no
12
significant differences in SRS Pain in the overall cohort. However, the rMCID group underwent
13
a significant improvement in SRS Pain (3.0 to 3.7; P<0.001), whereas the mMCID group
14
underwent a significant decrease in SRS Pain (3.6 to 3.3; P<0.001) (Figure 2). There were no
15
significant changes in TL Cobb in the overall group or in individual cohorts.
16
DISCUSSION
17
Summary of Key Findings
18
Patients that noted a clinically important improvement from non-operative treatment and
19
reached an MCID in SRS Activity or Pain had greater baseline pain and smaller coronal
20
deformity in the thoracolumbar region. In the subset of non-operative patients who were below
21
the normative population and needed improvement, less vertebral obliquity was a predictor of
12 Page 12 of 28
1
reaching an MCID, which is consistent with a previous study by Schwab et al[40]. Pain and
2
thoracolumbar deformity are predictors for reaching or missing MCID in a non-surgical setting.
3
MCID helps identify patients that exhibited a clinically noticeable improvement in pain
4
despite the fact that the pain trend appeared to show no change over 2 years. Analyzing the non-
5
operative group as a whole suggested that non-operative patients improve minimally in pain, but
6
a closer MCID analysis shows that the mean was actually masking a subset of patients who
7
improved. 84 patients had an MCID deficit in both SRS Activity and SRS Pain and that up to
8
54% (n=45) of these severely symptomatic patients improve in pain or activity with a clinically
9
noticeable benefit from non-operative treatment at 2 years.
10
Non-operative treatment is beneficial if applied to selected patients with moderate
11
disability. Moreover, when a subset of patients had clinically noticeable benefit from non-
12
operative treatment, the final clinical outcome was limited to improvement in 1 MCID. In coarse
13
comparison with the literature, patients underwent surgical treatment gained better final clinical
14
status in comparison to non-operative treatment [11,12,23,43]. Further studies should compare
15
both treatments in historically and deformity controlled cohorts.
16
Pain and Disability
17
The clinical presentations of ASD cover a broad spectrum of pain and disability [44]. It is
18
of importance to determine the drivers of medical care sought by the patient, and match them
19
with the treatment offered by the physician [7,32,33,44]. Pekmezci et al noted that the functional
20
status of the patients appeared to be a more important determinant than pain, in pursuing
21
operative treatment. Pain is complex and one of the significant components of disability, such as
22
functional, emotional, social and mental components.
13 Page 13 of 28
1
Moreover, understanding of the pathomechanism and identifying the pattern of pain is
2
crucial because it can profoundly impact the choice of treatment[32,45]. In ASD, back pain can
3
be secondary to instability of spinal segments, dynamic stretching of nerve roots, or axial pain
4
[32,45–49]. Thus, it is fundamental to distinguish the main driver of disability in ASD patients
5
undergoing decision making for treatment.
6
Thoracolumbar Deformity
7
The significant impact of sagittal plane deformity on quality of life [33,50] does not
8
necessarily mean focal coronal plane deformities are clinically irrelevant. Magnitude of TL
9
deformity in the current study proved to be an important radiographic parameter. While there
10
was no direct correlation between TL Cobb and SRS Pain, less coronal deformity in the TL
11
region and greater baseline pain were significant predictors for reaching an MCID. Schwab et al
12
demonstrated that of all the coronal curve types, TL curves had the highest percentage of patients
13
who received surgical treatment[51]. Prior studies have also shown that thoracolumbar and
14
lumbar curves generate less favorable HRQOL scores than thoracic curves[50].
15
Because operative treatment can provide significant improvement in HRQOLs for
16
sagittally malaligned patients, this subsequent shift of operative focus in ASD treatment towards
17
sagittal plane correction has resulted in more pure coronal deformities receiving non-operative
18
care. The findings in this study suggest that in cases of moderate sagittal malalignment, larger
19
TL coronal deformities negatively impact a patient's ability to improve with non-operative care,
20
supporting the notion that there is a complex relationship of TL deformity, disability, and
21
pain[50].
14 Page 14 of 28
1
Limitations
2
A primary limitation was study design. Although the database was prospectively
3
collected, the analysis is still limited by the retrospective design of the present study. In addition,
4
all data are limited by the capabilities of the HRQOL; SRS Pain scores may be poor due to
5
radiographic findings, radicular symptoms, neurogenic claudication, or other factors. Moreover,
6
patients in this cohort were enrolled in this study as an observational ASD cohort, rather than
7
enrolled specifically to evaluate non-operative treatment. There was also insufficient data
8
available regarding treatment modality. Thus, there is a need for high-level prospective studies
9
specifically focusing on non-operative patients with ASD and the various forms of non-operative
10
treatment. Future work including randomized non-operative treatment options would
11
undoubtedly be valuable. Additionally, longer-term follow-up would be important to validate
12
these findings, as non-operative outcomes for ASD represent a long-term concern for patients
13
and physicians. Finally, by design, this study did not include patients who crossed over from the
14
non-operative to operative groups. There were only 5 patients who crossed over, thus limiting the
15
potential for any high-powered analysis. Future research dedicated to this subset of patients
16
would also be instrumental in understanding outcomes in operative and non-operative patients.
17
18
CONCLUSIONS
19
MCID analysis is a useful tool to unmask the subset of patients that might improve to a
20
clinically relevant degree from non-operative treatment. A higher baseline level of pain and
21
smaller thoracolumbar curve are significant predictors for identifying patients that may reach an
22
MCID when treated non-operatively. A subset of severely symptomatic patients may be more
23
likely to reach an MCID if they have less vertebral obliquity. Pain is quite often the presenting 15 Page 15 of 28
1
complaint in patients with ASD and the focus of non-operative ASD treatment. Improvement in
2
function or pain at the 2-year time point in a subset of patients is possible. While it is still
3
uncertain if this group will continue to function reasonably in the upcoming years, HRQOL
4
scores remain important parameters in quantifying patient improvement/deterioration.
5
REFERENCES
6
[1]
Schwab FJ, Dubey A, Gamez L, El Fegoun AB, Hwang K, Pagala M, et al. Adult
7
scoliosis: prevalence, SF-36, and nutritional parameters in an elderly volunteer population.
8
Spine (Phila Pa 1976) 2005;30:1082–5.
9
[2]
Schwab FJ, Lafage V, Farcy JP, Bridwell KH, Glassman SSD, Shainline MR. Predicting
10
outcome and complications in the surgical treatment of adult scoliosis. Spine (Phila Pa
11
1976) 2008;33:2243–7. doi:10.1097/BRS.0b013e31817d1d4e.
12
[3]
Schwab FJ, Dubey A, Pagala M, Gamez L, Farcy JP. Adult scoliosis: a health assessment
13
analysis
14
doi:10.1097/01.BRS.0000049924.94414.BB.
15
[4]
SF-36.
Spine
(Phila
Pa
1976)
2003;28:602–6.
Robin GC, Span Y, Steinberg R, Makin M, Menczel J. Scoliosis in the elderly: a followup study. Spine (Phila Pa 1976) 1982;7:355–9.
16
17
by
[5]
McCarthy I, Hostin R, O’Brien M, Saigal R, Ames CP. Health economic analysis of adult
18
deformity
surgery.
Neurosurg
19
doi:10.1016/j.nec.2012.12.005.
Clin
N
Am
2013;24:293–304.
16 Page 16 of 28
1
[6]
2007;18:403–5. doi:10.1016/j.nec.2007.03.001.
2
3
Burton DC, Glattes RC. Measuring outcomes in spinal deformity. Neurosurg Clin N Am
[7]
Bridwell KH, Cats-baril W, Harrast J, Berven S, Glassman S, Farcy J, et al. The validity
4
of the SRS-22 instrument in an adult spinal deformity population compared with the
5
Oswestry and SF-12: a study of response distribution, concurrent validity, internal
6
consistency, and reliability. Spine (Phila Pa 1976) 2005;30:455–61.
7
[8]
Bridwell KH, Berven S, Glassman SD, Hamill C, Horton W, Lenke LG, et al. Is the SRS-
8
22 instrument responsive to change in adult scoliosis patients having primary spinal
9
deformity
Spine
(Phila
Pa
1976)
2007;32:2220–5.
doi:10.1097/BRS.0b013e31814cf120.
10
11
surgery?
[9]
Berven S, Deviren V, Demir-Deviren S, Hu SS, Bradford DS. Studies in the modified
12
Scoliosis Research Society Outcomes Instrument in adults: validation, reliability, and
13
discriminatory capacity. Spine (Phila Pa 1976) 2003;28:2164–9; discussion 2169.
14
doi:10.1097/01.BRS.0000084666.53553.D6.
15
[10] Asher MA, Lai SM, Glattes RC, Burton DC, Alanay A, Bago J. Refinement of the SRS-22
16
Health-Related Quality of Life questionnaire Function domain. Spine (Phila Pa 1976)
17
2006;31:593–7. doi:10.1097/01.brs.0000201331.50597.ea.
18
[11] Bridwell KH, Glassman S, Horton W, Shaffrey C, Schwab FJ, Zebala LP, et al. Does
19
treatment (nonoperative and operative) improve the two-year quality of life in patients
20
with adult symptomatic lumbar scoliosis: a prospective multicenter evidence-based
17 Page 17 of 28
1
medicine
study.
Spine
2
doi:10.1097/BRS.0b013e3181a8fdc8.
(Phila
Pa
1976)
2009;34:2171–8.
3
[12] Smith JS, Shaffrey CI, Berven S, Glassman S, Hamill C, Horton W, et al. Operative
4
versus nonoperative treatment of leg pain in adults with scoliosis: a retrospective review
5
of a prospective multicenter database with two-year follow-up. Spine (Phila Pa 1976)
6
2009;34:1693–8. doi:10.1097/BRS.0b013e3181ac5fcd.
7
[13] Smith JS, Shaffrey CI, Berven S, Glassman SD, Hamill C, Horton W, et al. Improvement
8
of back pain with operative and nonoperative treatment in adults with scoliosis.
9
Neurosurgery
2009;65:86–93;
discussion
93–4.
doi:10.1227/01.NEU.0000347005.35282.6C.
10
11
[14] Albert T, Purtill J, Mesa J, Mclntosh T, Balderston R. Health outcome assessment before
12
and after adult deformity surgery: a prospective study. Spine (Phila Pa 1976)
13
1995;20:2002–4; discussion p2005.
14
[15] Glassman SD, Hamill CL, Bridwell KH, Schwab F, Dimar JR, Lowe TG. The impact of
15
perioperative complications on clinical outcome in adult deformity surgery. Spine (Phila
16
Pa 1976) 2007;32:2764–70. doi:10.1097/BRS.0b013e31815a7644.
17
[16] Cho SK, Bridwell KH, Lenke LG, Cho W, Zebala LP, Pahys JM, et al. Comparative
18
analysis of clinical outcome and complications in primary versus revision adult scoliosis
19
surgery. Spine (Phila Pa 1976) 2012;37:393–401. doi:10.1097/BRS.0b013e31821f0126.
18 Page 18 of 28
1
[17] Carragee EJ, Cheng I. Minimum acceptable outcomes after lumbar spinal fusion. Spine J 2010;10:313–20. doi:10.1016/j.spinee.2010.02.001.
2
3
[18] Copay AG, Subach BR, Glassman SD, Polly DW, Schuler TC. Understanding the
4
minimum clinically important difference: a review of concepts and methods. Spine J
5
2007;7:541–6. doi:10.1016/j.spinee.2007.01.008.
6
[19] Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989;10:407–15.
7
8
[20] Copay AG, Glassman SD, Subach BR, Berven S, Schuler TC, Carreon LY. Minimum
9
clinically important difference in lumbar spine surgery patients: a choice of methods using
10
the Oswestry Disability Index, Medical Outcomes Study questionnaire Short Form 36, and
11
pain scales. Spine J 2008;8:968–74. doi:10.1016/j.spinee.2007.11.006.
12
[21] Parker SL, Mendenhall SK, Shau DN, Adogwa O, Anderson WN, Devin CJ, et al.
13
Minimum clinically important difference in pain, disability, and quality of life after neural
14
decompression and fusion for same-level recurrent lumbar stenosis: understanding clinical
15
versus
16
doi:10.3171/2012.1.SPINE11842.
statistical
significance.
J
Neurosurg
Spine
2012;16:471–8.
17
[22] Parker SL, Adogwa O, Mendenhall SK, Shau DN, Anderson WN, Cheng JS, et al.
18
Determination of minimum clinically important difference (MCID) in pain, disability, and
19
quality of life after revision fusion for symptomatic pseudoarthrosis. Spine J
20
2012;12:1122–8. doi:10.1016/j.spinee.2012.10.006.
19 Page 19 of 28
1
[23] Youssef J a, Orndorff DO, Patty C a, Scott M a, Price HL, Hamlin LF, et al. Current
2
Status of Adult Spinal Deformity. Glob Spine J 2013;3:51–62. doi:10.1055/s-0032-
3
1326950.
4
[24] Smith JS, Fu K-M, Urban P, Shaffrey CI. Neurological symptoms and deficits in adults
5
with scoliosis who present to a surgical clinic: incidence and association with the choice
6
of operative versus nonoperative management. J Neurosurg Spine 2008;9:326–31.
7
doi:10.3171/SPI.2008.9.10.326.
8
[25] Smith JS, Shaffrey CI, Glassman SD, Berven SH, Schwab FJ, Hamill CL, et al. Risk-
9
benefit assessment of surgery for adult scoliosis: an analysis based on patient age. Spine (Phila Pa 1976) 2011;36:817–24. doi:10.1097/BRS.0b013e3181e21783.
10
11
[26] Smith JS, Shaffrey CI, Glassman SD, Carreon LY, Schwab FJ, Lafage V, et al. Clinical
12
and radiographic parameters that distinguish between the best and worst outcomes of
13
scoliosis surgery for adults. Eur Spine J 2013;22:402–10. doi:10.1007/s00586-012-2547-
14
x.
15
[27] Moal B, Lafage V, Smith JS, Ames CP, Mundis GM, Terran JS, et al. Clinical
16
Improvement Through Surgery for Adult Spinal Deformity (ASD): What Can Be
17
Expected and Who is Likely to Benefit Most? Proc. NASS 27th Annu. Meet., vol. 12, The
18
Spine Journal; 2012, p. 99S – 165S. doi:10.1016/j.spinee.2012.08.395.
19
[28] Liu S, Schwab FJ, Smith JS, Klineberg E, Ames CP, Mundis G, et al. Likelihood of
20
reaching minimal clinically important difference in adult spinal deformity: a comparison
21
of operative and nonoperative treatment. Ochsner J 2014;14:67–77. 20 Page 20 of 28
1
[29] Berven S, D.V., Demir-Deviren S, Hu S BD. Minimal Clinically Important Difference in
2
Adult Spinal Deformity: How much change is significant? IMAST Present. Int. Meet.
3
Adv. Surg. Tech. IMAST, Banff, Canada: 2005.
4
[30] Blondel B, Schwab FJ, Ungar B, Smith JS, Bridwell KH, Glassman SD, et al. Impact of
5
magnitude and percentage of global sagittal plane correction on health-related quality of
6
life
7
doi:10.1227/NEU.0b013e31825d20c0.
at
2-years
follow-up.
Neurosurgery
2012;71:341–8;
discussion
348.
8
[31] Smith JS, Klineberg E, Schwab FJ, Shaffrey CI, Moal B, Ames CP, et al. Change in
9
Classification Grade by the SRS-Schwab Adult Spinal Deformity Classification Predicts
10
Impact on Health-Related Quality of Life Measures: Prospective Analysis of Operative
11
and
12
doi:10.1097/BRS.0b013e31829ec563.
13
Non-operative
Treatment.
Spine
(Phila
Pa
1976)
2013;38:1663–71.
[32] Kotwal S, Pumberger M, Hughes A, Girardi F. Degenerative scoliosis: a review. HSS J 2011;7:257–64. doi:10.1007/s11420-011-9204-5.
14
15
[33] Glassman SD, Bridwell K, Dimar JR, Horton W, Berven S, Schwab FJ. The impact of
16
positive sagittal balance in adult spinal deformity. Spine (Phila Pa 1976) 2005;30:2024–9.
17
[34] Baldus C, Bridwell K, Harrast J, Shaffrey C, Ondra S, Lenke L, et al. The Scoliosis
18
Research Society Health-Related Quality of Life (SRS-30) age-gender normative data: an
19
analysis of 1346 adult subjects unaffected by scoliosis. Spine (Phila Pa 1976)
20
2011;36:1154–62. doi:10.1097/BRS.0b013e3181fc8f98.
21 Page 21 of 28
1
[35] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying
2
prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis
3
1987;40:373–83.
4
[36] Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:473–83.
5
6
[37] Fairbank JC, Couper J, Davies JB, O’Brien JP. The Oswestry low back pain disability questionnaire. Physiotherapy 1980;66:271–3.
7
8
[38] Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine (Phila Pa 1976) 2000;25:2940–52; discussion 2952.
9
10
[39] Champain S, Benchikh K, Nogier A, Mazel C, Guise J De, Skalli W. Validation of new
11
clinical quantitative analysis software applicable in spine orthopaedic studies. Eur Spine J
12
2006;15:982–91. doi:10.1007/s00586-005-0927-1.
13
[40] Schwab FJ, Smith V a, Biserni M, Gamez L, Farcy J-PC, Pagala M. Adult scoliosis: a quantitative radiographic and clinical analysis. Spine (Phila Pa 1976) 2002;27:387–92.
14
15
[41] Schwab FJ, el-Fegoun AB, Gamez L, Goodman H, Farcy J. A lumbar classification of
16
scoliosis in the adult patient: preliminary approach. Spine (Phila Pa 1976) 2005;30:1670–
17
3.
18
[42] Schwab F, Ungar B, Blondel B, Buchowski J, Coe J, Deinlein D, et al. Scoliosis Research
19
Society-Schwab adult spinal deformity classification: a validation study. Spine (Phila Pa
20
1976) 2012;37:1077–82. doi:10.1097/BRS.0b013e31823e15e2. 22 Page 22 of 28
1
[43] Bridwell KH, Baldus C, Berven S, Edwards C, Glassman S, Hamill C, et al. Changes in
2
radiographic and clinical outcomes with primary treatment adult spinal deformity
3
surgeries from two years to three- to five-years follow-up. Spine (Phila Pa 1976)
4
2010;35:1849–54. doi:10.1097/BRS.0b013e3181efa06a.
5
[44] Schwab FJ, Lafage V, Farcy J-P, Bridwell KH, Glassman SD, Ondra S, et al. Surgical
6
rates and operative outcome analysis in thoracolumbar and lumbar major adult scoliosis:
7
application of the new adult deformity classification. Spine (Phila Pa 1976)
8
2007;32:2723–30. doi:10.1097/BRS.0b013e31815a58f2.
9
[45] Bradford DS, Tay BK, Hu SS. Adult scoliosis: surgical indications, operative
10
management, complications, and outcomes. Spine (Phila Pa 1976) 1999;24:2617–29.
11
[46] Van Dam BE. Nonoperative Treatment of Adult Scoliosis. Orthop Clin North Am 1988;19:347–51.
12
13
[47] Jackson RP, Simmons EH, Stripinis D. Coronal and sagittal plane spinal deformities
14
correlating with back pain and pulmonary function in adult idiopathic scoliosis. Spine
15
(Phila Pa 1976) 1989;14:1391–7.
16
[48] Kostuik JP, Bentivoglio J. The incidence of low-back pain in adult scoliosis. Spine (Phila Pa 1976) 1981;6:268–73.
17
18
[49] Vanderpool DW, James JI, Wynne-Davies R. Scoliosis in the elderly. J Bone Joint Surg Am 1969;51:446–55.
19
23 Page 23 of 28
1
[50] Glassman SD, Berven S, Bridwell K, Horton W, Dimar JR. Correlation of radiographic
2
parameters and clinical symptoms in adult scoliosis. Spine (Phila Pa 1976) 2005;30:682–
3
8.
4
[51] Schwab FJ, Farcy J, Bridwell K, Berven S, Glassman S, Harrast J, et al. A clinical impact
5
classification of scoliosis in the adult. Spine (Phila Pa 1976) 2006;31:2109–14.
6
doi:10.1097/01.brs.0000231725.38943.ab.
7
24 Page 24 of 28
1
FIGURE LEGENDS
2 3
Figure 1: Baseline characteristics of the non-operative cohort included history of spine problems
4
and Charlson Comorbidity Index, and Scoliosis Research Society (SRS)-Schwab classification.
5
SRS-Schwab
6
thoracolumbar/lumbar curves >30), T (thoracic curve only, with lumbar curve <30), L
7
(thoracolumbar/lumbar curve only, with thoracic curve <30), or N (no major coronal deformity
8
with all coronal curves <30). SRS-Schwab Sagittal Modifiers include pelvic incidence minus
9
lumbar lordosis mismatch (PI-LL), sagittal vertical axis (global alignment, GA), and pelvic tilt
10
(PT). Grade 0 denotes normal alignment, Grade 1 denotes moderate deformity, and Grade ++
11
denotes severe deformity.
curve
types
are
denoted
by
D
(double
curve
with
thoracic
and
12
13
Figure 2. Pain trends demonstrated no change in the mean pain level of the overall non-operative
14
group. However, the rMCID sub-group had a significant decrease in pain level at 2 years (SRS
15
Pain score 3.0 to 3.7; P<0.001), while the mMCID subgroup had a significant increase in pain
16
level at 2 years (SRS Pain score 3.6 to 3.3; P<0.001). The rMCID subgroup had a significantly
17
lower SRS Pain score at baseline than the mMCID subgroup (P<0.001).
18
25 Page 25 of 28
1
Table 1: The number of patients in rMCID (n=86) and mMCID groups (n=129) who reached MCID in
2
other HRQOL (aside from SRS-Pain or SRS-Activity) at 2 years. The rMCID cohort was more likely to
3
reach MCID in other HRQOL, such as ODI, SF-36 PCS, SRS appearance, and SRS mental. Confidence
4
intervals set at 95%.
5 MCID in Additional HRQOL Domains (N, %) rMCID
mMCID
n=86
n=129
ODI
14 (16.3%)
SF-36 PCS
Risk Ratio
Confidence Interval
2 (1.6%)
10.3
(9.3 – 11.3)
29 (33.7%)
7 (5.4%)
5.6
(4.7 – 6.7)
SRS appearance
15 (17.4%)
5 (3.9%)
4.5
(4.1 – 5.0)
SRS mental
26 (30.2%)
18 (14.0%)
2.1
(1.9 – 2.5)
6 7
26 Page 26 of 28
1 2 3 4 5
Table 2: Comparisons of predictors for rMCID (reached minimal clinically important difference) and mMCID (missed MCID). Predictors shown were modeled using a backwards stepwise logistic regression. *Denotes predictors that were further confirmed using a forwards stepwise logistic regression. SRS-Pain = Scoliosis Research Society Pain Score. TL Cobb = Thoracolumbar Cobb angle. SS = Sacral Slope. LL = Lumbar Lordosis.
6 Predictor
rMCID
mMCID
p-value
(multivariate)
n=86
n=129
(multivariate)
*SRS Pain
3.0
3.6
0.001
*TL Cobb (coronal)
29.6
36.5
0.007
SS
33.1
36.4
0.847
LL
46.5
52.8
0.875
7 8
27 Page 27 of 28
1
Table 3. Predictors shown were modeled using a backwards and forwards stepwise logistic regression.
2 Predictor
rMCID
mMCID
p-value
(multivariate)
n=45
n=39
(multivariate)
Maximum vertebral obliquity
24.3
27.2
0.035
3 4 5
28 Page 28 of 28