Abstract
Study Design
This is an analysis of a prospective 2-year study on nonoperative patients enrolled in the Adult Symptomatic Lumbar Scoliosis (ASLS) National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) trial.
Objective
The purpose was to evaluate the impact of serious adverse events (SAEs) on patient-reported outcomes (PROs) in nonoperative management of ASLS as measured by Scoliosis Research Society-22r (SRS-22r), Oswestry Disability Index (ODI) and Short Form-12 (SF-12) at 2-year follow-up.
Summary of Background Data
Little is known about PROs in the nonoperative management of ASLS or the prevalence and impact of SAEs on PROs.
Methods
The ASLS trial dataset was analyzed to identify adult lumbar scoliosis patients electively choosing or randomly assigned to nonoperative treatment with minimum 2-year follow-up. Patient data was collected prospectively from 2010–2015 as part of NIAMS R01-AR055176-01A2 “A Multi-Centered Prospective Study of Quality of Life in Adult Scoliosis”. SAEs were defined as life threatening medical events, new significant or permanent disability, new or prolonged hospitalization or death.
Results
105 nonoperative patients were studied to 2-year follow-up. Twenty-seven patients (25.7%) had 42 SAEs; 15 (14.3%) had a SAE during the first year. The SAE group had higher body mass index (29.4 vs 25.2; p=0.008) and reported worse SRS-22r Function scores than the non-SAE group at baseline (3.3 vs 3.6; p=0.024). At 2-year follow-up, SAE patients experienced less improvement (change) in SRS-22r Self-Image (−0.07 vs 0.26; p=0.018) and Mental Health domains (−0.19 vs 0.25; p=0.002) than non-SAE patients and had lower SRS-22r Function, Self-Image, Subscore and SF-12 Mental and Physical component scores (MCS/PCS). Fewer SAE patients reached Minimal Clinically Important Difference (MCID) threshold in SRS-22r Mental Health (14.8% vs 43.6%; p=0.01).
Conclusions
A high percentage (25.7%) of ASLS patients managed nonoperatively experienced SAEs. Those patients who sustained a SAE had less improvement in reported outcomes.
Keywords: adult symptomatic lumbar scoliosis, degenerative scoliosis, complications, adverse events, nonoperative treatment, NIH
INTRODUCTION
Adult spinal deformity (ASD) remains a challenging and costly condition to treat. Adult scoliosis rates have been estimated to exceed 50% in the population of those over 60 years.1 The natural history of these patients is typically one of gradual functional decline, continued pain and deterioration in health status.2 Previous studies have suggested that nonoperative modalities are not effective in improving Patient Reported Outcomes (PROs) in Adult Symptomatic Lumbar Scoliosis (ASLS).3, 4
Operative management in the older adult population, however, is not benign. Post-operative minor and major complication rates have been reported to exceed 60% and 30%, respectively.5–10 The magnitude of the surgery, patient comorbidities, advanced age, and lower physiologic reserve have all been implicated as possible reasons for high complication rates.8, 11 In spite of these issues, several studies have demonstrated short- and mid-term clinical benefits to operative intervention,12–14 but often compare heterogeneous patient cohorts.
While multiple studies have demonstrated high complication and adverse events rates after operative intervention for ASLS, to our knowledge none have examined the adverse event rate in nonoperative ASLS patients. Previous studies comparing operative to nonoperative treatment lack the follow-up3 to capture adverse events of nonoperative patients for years after initial presentation.
A multicenter, dual arm study examining operative and nonoperative treatment of ASLS has had a primary aim of investigating the effectiveness of operative versus nonoperative treatment for ASLS. Funding was provided by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) for “A Multi-Center Prospective Study of Quality of Life in Adult Scoliosis” - R01AR055176-01A2.15 Patients participating in the clinical trial for a minimum of 2 years, both operative and nonoperative, were evaluated for the occurrence of all serious adverse events (SAEs). SAE occurrence was defined by this trial as any death, life-threatening event, event that caused significant or permanent disability or event that led to prolonged or new hospitalization. The purpose of this study was to evaluate the prevalence of SAEs in the nonoperative cohort while evaluating their impact on patient-reported outcomes (PROs). We hypothesized that the occurrence of a SAE would negatively influence the effectiveness of nonoperative treatment for ASLS at 2 years follow-up.
METHODS
Study Design
A prospective, multicenter series of ASLS patients was evaluated from 2010 to 2015. Patients were enrolled throughout nine centers across the United Sates and Canada. Patients that met inclusion criteria and agreed to participate chose between a randomized and observational arm. Those desiring to choose their own treatment remained in the observational arm. Both study arms contained an operative and nonoperative cohort. For this study, all (n=105) nonoperative patients (regardless of randomized or observational arm) with 2-year follow-up were included in the analysis. Thirty-two patients crossing over to operative treatment prior to 2-year follow-up in the nonoperative arm were not studied. Patients were monitored at each enrollment site for the occurrence of SAEs. Monitoring was performed prospectively by each site during follow-up clinical visits at 3, 12 and 24 months. In addition, PROs were completed by mail at 6, 9, 15 18 and 21 months. Any changes in PRO scores (drop in SRS domain scores of 0.5 or more or increase in ODI of 10 points or greater) resulted in a follow-up call to the patient to determine if an adverse event had occurred. Site monitoring visits were performed annually to review research and clinical charts. SAE details (date of onset, diagnosis, expectedness, relationship to treatment, severity, outcome) were recorded by the enrollment site and reported to the coordinating center for review within 24 hours of discovery. The coordinating center reviewed data and then forwarded to the Data Safety Monitoring Board (DSMB) for review. Institutional Review Board approval was obtained at each participating center.15
Patient population
Patients between the ages of 40 and 80 years and diagnosed with ASLS were eligible for study enrollment. ASLS was defined as an idiopathic or de novo lumbar scoliosis with a Cobb measurement ≥30°. Symptomatic was defined as a Scoliosis Research Society (SRS)-22r score ≤4.0 in the domains of Pain, Function and/or Self-Image and/or Oswestry Disability Index (ODI) score of ≥20. Age and diagnosis categories were defined as such because the majority of patients presenting with symptomatic adult spinal deformity have idiopathic or de novo scoliosis. Idiopathic implies progression of pre-existent teenage scoliosis and de novo represents patients who have no history of scoliosis as adolescents, but then develop a deformity as an adult. In general, patients younger than 40 years do not typically have the degenerative changes and comorbidities that make the decision to operate more challenging. Also, patients 81 years or older were not eligible for inclusion as most providers would not recommend surgical procedures for this age group due to increased risk of perioperative morbidity and mortality. All patients, regardless of treatment arm, were considered reasonable operative candidates at the time of enrollment. Patients with excessive medical comorbidities, pregnancy, osteoporosis (defined by femoral neck DEXA t-score <−3.0), previous thoracolumbar fusion, multilevel thoracolumbar decompression, high-grade spondylolisthesis, congenital spine anomalies, neuromuscular scoliosis and a high risk of operative failure or morbidity were not enrolled in the study.
At enrollment, data regarding patient demographics, comorbidities and disease severity were captured. Our analysis included basic patient demographic information (age, gender, race), job/working status (defined as full or part-time work outside the home), health wellness characteristics (body mass index (BMI)/obesity, alcohol use, smoking status), medical comorbidities (such as cardiac, lung, circulatory, endocrine, renal, gastrointestinal) and other disorders. A history of medically diagnosed psychiatric disorders, including depression and anxiety, was collected from physician and patient questionnaires, and grouped for dichotomous analysis. Standard baseline radiographic measurements were recorded. Consideration was given to stratification of continuous variables such as patient Age, but background analysis revealed that no differences existed among cohorts.
Outcomes
PROs, as measured by the SRS-22r, ODI and SF-12, at 2-year follow-up were the primary outcomes. The reproducibility of patient reported outcomes has been previously validated, and determined to not require additional separate internal validation3, 16, 17. SAEs were considered for analysis regardless of their relatedness to the spine pathology or treatment modality. All SAEs within the 2-year follow-up period were reviewed and sub-categorized by the most significant related intervention or sequela and grouped according to diagnostic category.
Analysis
Patients meeting inclusion criteria were compared based on the occurrence of any SAE during 2 years following enrollment. Inferential statistics were used to compare baseline patient characteristics between those with and without a SAE. Radiographic measurements underwent independent evaluation by two experienced readers: a spinal surgeon (not otherwise involved in the study clinically or academically) and the clinical trial research nurse. Each reader performed two independent reads (separated by several weeks). The four reads were compared using Intraclass Correlation Coefficients (ICC) for both intra- and inter-observer reliability and the results demonstrated very high-reliability18. With the exception of T2-T5 coronal measurements (kappa=0.71), all ICCs ranged from k=0.90-0.99.
Standard statistical tests, including chi-square for categorical variables and Student’s t-test for continuous variables and ANOVA, were performed. Baseline and 2-year PRO scores and changes in PRO scores were compared between cohorts. Multivariate logistic regression analysis was used to test the influence of baseline characteristics on the occurrence of SAEs. All variables with p < 0.05 from the univariate analysis were included in the multivariate model. Model discrimination was measured as the c-index/c-statistic. The c-index is a measure of goodness of fit for binary outcomes in a logistic regression model; values over 0.717 indicate a good model, where those above 0.8 indicate a strong model. The number of patients from each cohort achieving a Minimal Clinically Important Difference (MCID) for SRS-22r, ODI and SF-12 scores were compared. MCID PRO thresholds were determined based on previous work: SRS-22r (0.4), ODI (10), SF-12 (5).19–21,20 All statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC). The level of significance for all univariate and multivariate analysis was set at p < 0.05.
RESULTS
One hundred five patients were studied to 2-year follow-up. Eight patients who withdrew before completing 2-year follow-up were not studied. During the 2 years, 27 patients (25.7%) had 42 SAEs (Table 1). Ten (9.5%) of the 27 patients experienced multiple SAEs. Fifteen (14.3%) patients had a SAE during the first year. Over one-third of 27 SAE patients experienced at least two events during the follow-up period. Nearly 90% of these patients had associated hospitalization. The most common reason for an event was operative intervention related to a progressive medical condition not related to the patient’s spinal deformity. Other musculoskeletal (major joint arthritis, cervical myelopathy/radiculopathy, rotator cuff disease; 25.9%), cardiac (14.8%), gastrointestinal (18.5%) and genitourinary (14.8%) events were the top diagnostic categories of SAEs (Table 1). Four out of the 42 SAEs were directly related to the patient’s spinal deformity, namely side effects of medications (NSAIDs).
Table 1.
Summary of Nonoperative Cohort Experiencing a Serious Adverse Event over 2 Years
| Number of Serious Adverse Events | (N) Patients | Percent |
|---|---|---|
| 1 Event | 17 | 63.0% |
| 2 Events | 7 | 25.9% |
| 3 Events | 2 | 7.4% |
| 4 Events | 1 | 3.7% |
| TOTAL | 27 | 100.0% |
| Highest Event Intervention/Sequele | ||
| Death | 0 | 0.0% |
| Emergency Room Visit | 2 | 7.4% |
| Hospitalization | 7 | 25.9% |
| Surgery (urgent) | 5 | 18.5% |
| Surgery (elective) | 12 | 44.4% |
| Unknown | 1 | 3.7% |
| TOTAL | 27 | 100.0% |
| Diagnostic Category | ||
| Cardiac | 4 | 14.8% |
| Gastrointestinal | 5 | 18.5% |
| Genitourinary | 4 | 14.8% |
| Musculoskeletal | 7 | 25.9% |
| Oncologic | 2 | 7.4% |
| Neurologic | 3 | 11.1% |
| Respiratory | 1 | 3.7% |
| Unknown | 1 | 3.7% |
| TOTAL | 27 | 100.0% |
Demographics were similar between SAE and non-SAE cohorts based on age (60.1 vs 63.5 years, p=0.24) and gender (14.8% vs 6.4% males, p=0.24). Slightly more non-white patients (p=0.022) and non-working patients (p=0.015) had a SAE (Table 2). The prevalence of wellness habits (alcohol use, smoking history, p>0.5) and comorbidities, such as cardiac, respiratory, vascular, renal, psychiatric and oncologic disease (all p>0.25) did not differ between cohorts. Those experiencing a SAE had higher BMI (29.4 vs 25.2, p=0.008) (Table 2). When BMI was stratified by range (< 30, 30–40, and > 40), this statistical association persisted (p=.014). Both cohorts had the same number of obese (9) and morbidly obese (2) patients, but there was a higher percentage of patients with BMI 30–40 and > 40 in the SAE group. In the multivariate logistic regression model, BMI remained a significant predictor of SAE, with a c-index of 0.702. A higher BMI was associated with the occurrence of SAEs (OR 3.408 (95% CI: 1.301-8.930)).
Table 2.
Comparison of Baseline Case Characteristics of Nonoperative ASLS Patients With and Without a Serious Adverse Event
| Overall (n=105) | No SAE (n=78) | SAE (n=27) | P value | |
|---|---|---|---|---|
| Patient Demographics | Mean (sd) | Mean (sd) | Mean (sd) | |
| Age (years) | 61.5 (10.0) | 60.8 (10.5) | 63.5 (8.0) | 0.236 |
| BMI | 26.3 (6.2) | 25.2 (5.4) | 29.4 (7.3) | 0.008 |
| % | % | % | ||
| Gender: Males | 8.6 | 6.4 | 14.8 | 0.231 |
| Race | 0.022 | |||
| White | 89.4 | 92.2 | 81.5 | |
| Black | 7.7 | 7.8 | 7.4 | |
| Other | 2.9 | 0.0 | 11.1 | |
| Work Status (Yes %) | 63.8 | 70.5 | 44.4 | 0.015 |
| Baseline Comorbidities | ||||
| Alcohol / drugs | 1.0 | 1.3 | 0.0 | 1.000 |
| Autoimmune | 3.8 | 2.6 | 7.4 | 0.272 |
| Cancer | 22.9 | 23.1 | 22.2 | 0.927 |
| Cardiac Disease | 7.6 | 7.7 | 7.4 | 1.000 |
| Circulatory disorders, arterial | 1.0 | 0.0 | 3.7 | 0.257 |
| Circulatory disorders, venous | 2.9 | 2.6 | 3.7 | 1.000 |
| Diabetes Mellitus | 2.9 | 2.6 | 3.7 | 1.000 |
| Gastrointestinal (ulcer, stomach) | 11.4 | 12.8 | 7.4 | 0.727 |
| Hypertension | 37.1 | 35.9 | 40.7 | 0.654 |
| Infection history | 4.8 | 3.9 | 7.4 | 0.601 |
| Lung disease / Asthma | 13.3 | 12.8 | 14.8 | 0.752 |
| Nervous System Disorders | 0.0 | 0.0 | 0.0 | 1.000 |
| Obesity | 46.7 | 39.7 | 66.7 | 0.016 |
| Psychiatric | 21.9 | 19.2 | 29.6 | 0.260 |
| Renal disease | 1.0 | 1.3 | 0.0 | 1.000 |
| Smoking | 14.3 | 14.1 | 14.8 | 1.000 |
| Coronal Plane | Mean (sd) | Mean (sd) | Mean (sd) | |
| Thoracic Cobb (°) | 47.4 (13.0) | 47.3 (13.9) | 47.5 (9.2) | 0.978 |
| Lumbar Cobb (°) | 49.2 (11.8) | 48.2 (10.3) | 52.0 (15.2) | 0.236 |
| Fractional Cobb (°) | 21.1 (8.8) | 20.6 (8.1) | 22.6 (10.5) | 0.298 |
| Coronal Balance (mm) | 20.4 (16.5) | 19.9 (15.7) | 21.8 (18.7) | 0.605 |
| Thoracic Curve >30° (%) | 50.5 | 53.9 | 40.7 | 0.240 |
| Sagittal Plane | ||||
| Sagittal Balance (mm) | 37.8 (30.3) | 35.3 (31.0) | 44.9 (27.6) | 0.157 |
| Pelvic Incidence (°) | 55.7 (13.2) | 54.4 (12.3) | 59.2 (15.0) | 0.110 |
| Sacral Slope (°) | 33.7 (11.3) | 33.5 (11.2) | 34.2 (11.7) | 0.780 |
| Pelvic Tilt (°) | 22.9 (9.6) | 22.0 (9.1) | 25.2 (10.7) | 0.153 |
| PI minus LL mismatch | 17.5 (13.9) | 16.9 (13.7) | 19.1 (14.6) | 0.484 |
| ASLS Treatment History | % | % | % | |
| Pharmacological | 91.4 | 93.6 | 85.2 | 0.231 |
| Physical Therapy | 81.0 | 82.1 | 77.8 | 0.626 |
| Spine Injection(s) | 16.2 | 15.4 | 18.5 | 0.703 |
ASLS=Adult Symptomatic Lumbar Scolisois, SAE=Serious Adverse Event, sd=Standard Deviation, BMI=Body Mass Index; PI=Pelvic Incidence, LL=Lumbar Lordosis
Overall, patients with and without a SAE were similar in terms of radiographic parameters and previous treatment (Table 2). Coronal thoracic, lumbar (48.2° vs 52.0°, p=0.24) and fractional lumbar Cobb measurements were similar between non-SAE and SAE cohorts. Likewise, in the sagittal plane, overall sagittal alignment (C7 sagittal vertical axis), pelvic incidence (PI), lumbar lordosis (LL), pelvic tilt and PI-LL mismatch did not differ (all p>0.23) (Table 2). Both SAE and non-SAE patients had similar types and rates of nonoperative treatment, including medications, physical therapy and spine injections (Table 2).
The SAE group reported worse SRS Function than the non-SAE group at baseline (3.25 vs 3.55, p=0.024). There were no statistical differences in the other baseline SRS-22r, ODI or SF-12 scores (Table 3).
Table 3.
Comparison of Patient-Reported Outcomes in Nonoperative ASD Patients With and Without a Serious Adverse Event in 2 Years
| Patient-Reported Outcomes | |||||
|---|---|---|---|---|---|
| All (n=105) | No SAE (n=78) | SAE (n=27) | P value |
||
| Mean (sd) | Mean (sd) | Mean (sd) | |||
| Index Visit | SRS-22r Pain | 3.05 (0.63) | 3.12 (0.59) | 2.86 (0.72) | 0.062 |
| SRS-22r Function | 3.47 (0.59) | 3.55 (0.58) | 3.25 (0.57) | 0.024 | |
| SRS-22r Self Image | 3.15 (0.66) | 3.18 (0.69) | 3.08 (0.57) | 0.516 | |
| SRS-22r Mental Health | 3.74 (0.72) | 3.70 (0.71) | 3.85 (0.73) | 0.332 | |
| SRS-22r Subscore | 3.35 (0.46) | 3.39 (0.45) | 3.26 (0.48) | 0.226 | |
| ODI | 30.00 (13.78) | 28.85 (13.10) | 33.33 (15.37) | 0.146 | |
| SF-12 MCS | 51.16 (10.81) | 51.52(10.76) | 50.13 (11.07) | 0.566 | |
| SF-12 PCS | 37.87 (10.30) | 38.93 (10.63) | 34.82 (8.75) | 0.074 | |
| 2-Year Follow-up | SRS-22r Pain 2-Year | 3.46 (0.62) | 3.47 (0.59) | 3.42 (0.67) | 0.734 |
| SRS-22r Function 2-Year | 3.55 (0.66) | 3.65 (0.61) | 3.27 (0.71) | 0.009 | |
| SRS-22r Self Image 2-Year | 3.33 (0.74) | 3.44 (0.61) | 3.01 (0.96) | 0.034 | |
| SRS-22r Mental Health 1-Year | 3.88 (0.68) | 3.95 (0.64) | 3.66 (0.76) | 0.054 | |
| SRS-22r Subscore 2-Year | 3.55 (0.54) | 3.63 (0.48) | 3.34 (0.66) | 0.042 | |
| ODI 2-Year | 27.28 (14.87) | 26.43 (14.33) | 29.70 (16.37) | 0.327 | |
| SF-12 MCS 2-Year | 51.97 (10.67) | 53.22 (10.36) | 48.41 (10.93) | 0.044 | |
| SF-12 PCS 2-Year | 39.56 (11.35) | 40.88 (10.59) | 35.79 (12.73) | 0.044 | |
| 2-Year PRO Change | Change in SRS-22r Pain | 0.40 (0.58) | 0.35 (0.55) | 0.56 (0.65) | 0.098 |
| Change in SRS-22r Function | 0.08 (0.43) | 0.10 (0.40) | 0.01 (0.51) | 0.378 | |
| Change in SRS-22r Self Image | 0.18 (0.64) | 0.26 (0.60) | (−)0.07 (0.71)* | 0.018 | |
| Change in SRS-22r Mental Health | 0.14 (0.66) | 0.25 (0.65) | (−)0.19 (0.56)* | 0.002 | |
| Change in SRS-22r Subscore | 0.20 (0.38) | 0.24 (0.33) | 0.08 (0.47) | 0.104 | |
| Change in ODI | (−)2.72 (9.46) | (−)2.41 (9.52) | (−)3.62 (9.37) | 0.569 | |
| Change in SF-12 MCS | 0.78 (11.84) | 1.66 (12.07) | (−)1.71 (10.97)* | 0.205 | |
| Change in SF-12 PCS | 1.52 (9.65) | 1.72 (9.35) | 0.97 (10.63) | 0.731 | |
Denotes a deterioration from baseline
ASD=Adult Spinal Deformity; SAE=Serious Adverse Event; sd=Standard Deviation; SRS=Scoliosis Research Society; ODI=Oswsetry Disability Index; MCS=Mental Component Score; PCS=Physical Component Score
At 2-year follow-up, patients experiencing a SAE had less improvement (change) in SRS-22r Self-Image (mean difference=0.34, 95% CI: 0.22-0.46, p=0.018) and Mental Health (0.45, 95% CI: 0.33 vs 0.57, p=0.002) domain scores than the non-SAE group. Furthermore, at 2 years SAE patients had lower SRS Function (mean diff=0.38, 95% CI: 0.26-0.50, p=0.009) and Self-Image (0.43, 95% CI: 0.30-0.57, p=0.034), SRS Subscore (0.29, 95% CI: 0.19-0.39 p=0.042), SF-12 MCS (4.81, 95% CI: 2.80-6.82, p=0.044) and SF-12 PCS (5.01, 95% CI: 2.95-7.23, p=0.044) compared to those without a SAE at 2-year follow-up (Table 3). Finally, significantly fewer patients with a SAE (14.8% vs 43.6%, p=0.01) reached MCID for the SRS Mental Health domain (Table 4).
Table 4.
Percentage of Patients that Reached A Minimal Clinically Important Difference (MCID) from Baseline to 2 Years
| All (n=105) |
No SAE (n=78) |
SAE (n=27) |
P value | |
|---|---|---|---|---|
| SRS-22r Pain (%) | 54.3 | 50.0 | 66.7 | 0.134 |
| SRS-22r Function (%) | 26.7 | 25.6 | 29.6 | 0.686 |
| SRS-22r Self Image (%) | 32.4 | 35.9 | 22.2 | 0.191 |
| SRS-22r Mental Health (%) | 36.2 | 43.6 | 14.8 | 0.010 |
| SRS-22r Subscore (%) | 29.5 | 30.8 | 25.9 | 0.634 |
| Oswestry Disability Index (%) | 9.5 | 10.3 | 7.4 | 1.000 |
| SF-12 Mental Component Score (%) | 31.4 | 34.6 | 22.2 | 0.217 |
| SF-12 Physical Component Score (%) | 34.3 | 32.1 | 40.7 | 0.437 |
SAE=Serious Adverse Event; SRS=Scoliosis Research Society
MCID thresholds: SRS-22r (0.4), ODI (10), SF-12 (5)17
DISCUSSION
This study examines the influence adverse events have on PROs over 2 years using the nonoperative cohort in our NIAMS-funded study on Adult Symptomatic Lumbar Scoliosis. Overall, we found over one-quarter of patients experienced a SAE by 2 years. Most of these SAEs were related to interventions for progressive medical conditions not related to the spinal deformity and were associated with lower PROs at 2-year follow-up than for those patients without a SAE.
The first objective of this study was to identify the prevalence of SAEs in the nonoperative arm of the NIH ASLS study. Nearly 15% of patients had a SAE at 1 year and 25% at 2 years following study enrollment. Even if not choosing surgery, patients should realize that concomitant conditions may arise and these conditions may have a negative effect on patient-reported outcomes.
Several progressive arthritic conditions treated operatively met study criteria for SAEs, though were unrelated to the nonoperative management of ASLS. Within the nonoperative cohort, major joint arthritis and cervical myelopathy treated operatively accounted for nearly one-third of the SAEs during the 2-year follow-up period. In ASLS patients, both spinal stenosis and lower extremity joint arthritis may be an additional source of pain and dysfunction, which negatively affect PROs regardless of ASLS treatment choice.
From the list of patient demographics, comorbidities, radiographic and treatment factors, few had any association with the development of a SAE. In both the univariate and multivariate logistic regression analysis, obesity was associated with SAE occurrence. The influence of obesity has been widely reported as a risk factor for the development of other medical conditions, such as cardiac disease and major joint arthritis.22 Obesity is also a well accepted risk factor for operative complications.11, 23 Given the necessity for all patients to be considered operative candidates, there was a low prevalence of more severe comorbidities, such as diabetes (3%).
The occurrence of a SAE was associated with a negative change in SRS-22r Self-Image and Mental Health domains from baseline to 2-year follow-up. We assume the SRS-22r Self-Image result is not related to the patient’s spinal deformity as this parameter did not change over the 2-year follow-up period. The non-SAE cohort had positive and statistically significant improvements in many SRS PROs.
When analyzing the absolute values of 2-year PRO scores, patients not experiencing a SAE had significantly higher SRS Subscore and Function and Self-Image domains and SF-12 MCS/PCS scores than their counterparts with no SAE. Scheer et al7 reviewed the effect of complications (minor, major, reoperation) on 2-year outcomes following ASD surgery and found any complication negatively influenced mental recovery and reoperation also impacted overall satisfaction.
Most patients failed to meet PRO-specific MCID thresholds. The only PRO for which more than 50% of patients reached a MCID threshold was SRS Pain (Table 4). For SRS Mental Health, 43.6% without a SAE reached MCID, while only 14.8% of patients with a SAE reached the MCID threshold of 0.4.19 These findings are consistent with a multitude of studies.3, 24, 3 Slobodyanyuk et al showed only 24% of patients treated nonoperatively had clinical improvement at 1 year.4
The data collected throughout the NIH ASLS trial represents a very complete series of patients with ASLS considered for operative treatment. Our 2-year follow-up rate in the nonoperative cohort was 93%, the highest reported. A previous multicenter study reported the outcomes of nonoperative ASLS patients with a 2-year follow-up rate of 45%.3 A more recent study from Liu et al used a registry to identify patient factors associated with clinical improvement in nonoperative treatment.25 Complete 2-year follow-up was only available in 215 of 371 patients (58%).
Our study, however, is not without limitation. The study inclusion criteria were restricted to patients with ASLS. Patients with other scoliosis etiologies or previous fusion surgeries were excluded, and thus our results cannot necessarily be applied to all pathologies evaluated by adult spinal deformity surgeons. Furthermore, this study does not include analysis of the 32 patients that crossed over into the operative arm before 2-year follow-up. Presumably, these would be patients with the lowest baseline PROs. Thus, the reported changes in the nonoperative cohort PROs at 2 years may be somewhat positively skewed. Given the number of PROs analyzed, further statistical work, such as a regression analysis, could not be reasonably performed for each scenario. Finally, patients from both the randomized and observational cohorts were included in the analysis, potentially introducing selection bias in the observational patients.
Conclusion
The prevalence of Serious Adverse Events in the nonoperative cohort of the NIAMS-funded study on Adult Symptomatic Lumbar Scoliosis was over 25% (27/105) at 2 years. The occurrence of SAEs impacted many patient reported outcomes at 2 years, most noticeably patient mental health. In the context of an operative treatment option with significant risks and resource utilization, these findings highlight the importance of understanding the high baseline rate of adverse events within the aging, adult deformity population. Surgeons should set expectations to patients, hospitals and policymakers that, regardless of treatment chosen, the natural history of patients suffering from ASLS includes SAEs unrelated to their spine and detrimental to Patient Reported Outcomes.
Supplementary Material
Acknowledgments
The manuscript submitted does not contain information about medical device(s)/drug(s).
The National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) funds were received in support of this work.
We would like to acknowledge Jacob Buchowski, MD; Charles H. Crawford, III, MD; Han Jo Kim, MD; Lawrence G. Lenke, MD; Justin Smith, MD for contributing patients
Footnotes
Relevant financial activities outside the submitted work: board membership, consultancy, grants, royalties, employment, patents, payment for lecture, stocks.
References
- 1.Schwab F, Dubey A, Gamez L, El Fegoun AB, Hwang K, Pagala M, Farcy JP. Adult scoliosis: prevalence, SF-36, and nutritional parameters in an elderly volunteer population. Spine. 2005;30(9):1082–1085. doi: 10.1097/01.brs.0000160842.43482.cd. [DOI] [PubMed] [Google Scholar]
- 2.Aebi M. The adult scoliosis. Eur Spine J. 2005;14(10):925–948. doi: 10.1007/s00586-005-1053-9. [DOI] [PubMed] [Google Scholar]
- 3.Bridwell KH, Glassman S, Horton W, Shaffrey C, Schwab F, Zebala LP, Lenke LG, Hilton JF, Shainline M, Baldus C, Wootten D. Does treatment (nonoperative and operative) improve the two-year quality of life in patients with adult symptomatic lumbar scoliosis: a prospective multicenter evidence-based medicine study. Spine. 2009;34(20):2171–2178. doi: 10.1097/BRS.0b013e3181a8fdc8. [DOI] [PubMed] [Google Scholar]
- 4.Slobodyanyuk K, Poorman CE, Smith JS, Protopsaltis TS, Hostin R, Bess S, Mundis GM, Jr, Schwab FJ, Lafage V International Spine Study Group. Clinical improvement through nonoperative treatment of adult spinal deformity: who is likely to benefit? Neurosurg Focus. 2014;36(5):E2. doi: 10.3171/2014.3.FOCUS1426. [DOI] [PubMed] [Google Scholar]
- 5.Charosky S, Guigui P, Blamoutier A, Roussouly P, Chopin D Study Group on S. Complications and risk factors of primary adult scoliosis surgery: a multicenter study of 306 patients. Spine. 2012;37(8):693–700. doi: 10.1097/BRS.0b013e31822ff5c1. [DOI] [PubMed] [Google Scholar]
- 6.Daubs MD, Lenke LG, Cheh G, Stobbs G, Bridwell KH. Adult spinal deformity surgery: complications and outcomes in patients over age 60. Spine. 2007;32(20):2238–2244. doi: 10.1097/BRS.0b013e31814cf24a. [DOI] [PubMed] [Google Scholar]
- 7.Scheer JK, Mundis GM, Klineberg E, Hart RA, Deviren V, Burton DC, Protopsaltis TS, Gupta M, Rolston JD, Bess S, Shaffrey CI, Schwab F, Lafage V, Smith JS, Ames CP International Spine Study Group. Recovery following adult spinal deformity surgery: the effect of complications and reoperation in 149 patients with 2-year follow-up. Eur Spine J. 2016;25(8):2612–2621. doi: 10.1007/s00586-015-3787-3. [DOI] [PubMed] [Google Scholar]
- 8.Smith JS, Klineberg E, Lafage V, Shaffrey CI, Schwab F, Lafage R, Hostin R, Mundis GM, Jr, Errico TJ, Kim HJ, Protopsaltis TS, Hamilton DK, Scheer JK, Soroceanu A, Kelly MP, Line B, Gupta M, Deviren V, Hart R, Burton DC, Bess S, Ames CP International Spine Study Group. Prospective multicenter assessment of perioperative and minimum 2-year postoperative complication rates associated with adult spinal deformity surgery. J Neurosurg Spine. 2016;25(1):1–14. doi: 10.3171/2015.11.SPINE151036. [DOI] [PubMed] [Google Scholar]
- 9.Soroceanu A, Burton DC, Oren JH, Smith JS, Hostin R, Shaffrey CI, Akbarnia BA, Ames CP, Errico TJ, Bess S, Gupta MC, Deviren V, Schwab FJ, Lafage V International Spine Study Group. Medical complications after adult spinal deformity surgery: incidence, risk factors, and clinical impact. Spine. 2016;41(22):1718–1723. doi: 10.1097/BRS.0000000000001636. [DOI] [PubMed] [Google Scholar]
- 10.Soroceanu A, Diebo BG, Burton D, Smith JS, Deviren V, Shaffrey C, Kim HJ, Mundis G, Ames C, Errico T, Bess S, Hostin R, Hart R, Schwab F, Lafage V International Spine Study Group. Radiographical and implant-related complications in adult spinal deformity surgery: incidence, patient risk factors, and impact on health-related quality of life. Spine. 2015;40(18):1414–1421. doi: 10.1097/BRS.0000000000001020. [DOI] [PubMed] [Google Scholar]
- 11.Soroceanu A, Burton DC, Diebo BG, Smith JS, Hostin R, Shaffrey CI, Boachie-Adjei O, Mundis GM, Jr, Ames C, Errico TJ, Bess S, Gupta MC, Hart RA, Schwab FJ, Lafage V International Spine Study Group. Impact of obesity on complications, infection, and patient-reported outcomes in adult spinal deformity surgery. J Neurosurg Spine. 2015;23(5):656–664. doi: 10.3171/2015.3.SPINE14743. [DOI] [PubMed] [Google Scholar]
- 12.Bridwell KH, Baldus C, Berven S, Edwards C, II, Glassman S, Hamill C, Horton W, Lenke LG, Ondra S, Schwab F, Shaffrey C, Wootten D. Changes in radiographic and clinical outcomes with primary treatment adult spinal deformity surgeries from two years to three- to five-years follow-up. Spine. 2010;35(20):1849–1854. doi: 10.1097/BRS.0b013e3181efa06a. [DOI] [PubMed] [Google Scholar]
- 13.Scheer JK, Smith JS, Clark AJ, Lafage V, Kim HJ, Rolston JD, Eastlack R, Hart RA, Protopsaltis TS, Kelly MP, Kebaish K, Gupta M, Klineberg E, Hostin R, Shaffrey CI, Schwab F, Ames CP International Spine Study Group. Comprehensive study of back and leg pain improvements after adult spinal deformity surgery: analysis of 421 patients with 2-year follow-up and of the impact of the surgery on treatment satisfaction. J Neurosurg Spine. 2015;22(5):540–553. doi: 10.3171/2014.10.SPINE14475. [DOI] [PubMed] [Google Scholar]
- 14.Smith JS, Shaffrey CI, Berven S, Glassman S, Hamill C, Horton W, Ondra S, Schwab F, Shainline M, Fu KM, Bridwell K Spinal Deformity Study Group. Improvement of back pain with operative and nonoperative treatment in adults with scoliosis. Neurosurgery. 2009;65(1):86–93. doi: 10.1227/01.NEU.0000347005.35282.6C. discussion 93–94. [DOI] [PubMed] [Google Scholar]
- 15.Bridwell K. A Multi-Center Prospective Study of Quality of Life in Adult Scoliosis (R01AR055176-01A2). National Institude of Health. ClinicalTrials.Gov [Internet] 2016. [Google Scholar]
- 16.Berven S, Deviren V, Demir-Deviren S, et al. Studies in the modified Scoliosis Research Society Outcomes Instrument in adults: validation, reliability, and discriminatory capacity. Spine. 2003;28:2164–9. doi: 10.1097/01.BRS.0000084666.53553.D6. discussion 9. [DOI] [PubMed] [Google Scholar]
- 17.Burton DC, Glattes RC. Measuring outcomes in spinal deformity. Neurosurg Clin N Am. 2007;18:403–405. doi: 10.1016/j.nec.2007.03.001. [DOI] [PubMed] [Google Scholar]
- 18.Chmura Kraemer H, Periyakoil VS, Noda A. Kappa coefficients in medical research. Statistics in medicine. 2002;21:2109–2129. doi: 10.1002/sim.1180. [DOI] [PubMed] [Google Scholar]
- 19.Crawford CH, III, Glassman SD, Bridwell KH, Berven SH, Carreon LY. The minimum clinically important difference in SRS-22R total score, appearance, activity and pain domains after surgical treatment of adult spinal deformity. Spine. 2015;40(6):377–381. doi: 10.1097/BRS.0000000000000761. [DOI] [PubMed] [Google Scholar]
- 20.Copay AG, Subach BR, Glassman SD, Polly DW, Jr, Schuler TC. Understanding the minimum clinically important difference: a review of concepts and methods. Spine J. 2007;7(5):541–546. doi: 10.1016/j.spinee.2007.01.008. [DOI] [PubMed] [Google Scholar]
- 21.Copay AG, Glassman SD, Subach BR, Berven S, Schuler TC, Carreon LY. Minimum clinically important difference in lumbar spine surgery patients: a choice of methods using the Oswestry Disability Index, Medical Outcomes Study questionnaire Short Form 36, and pain scales. Spine J. 2008;8(6):968–974. doi: 10.1016/j.spinee.2007.11.006. [DOI] [PubMed] [Google Scholar]
- 22.Lievense AM, Bierma-Zeinstra SM, Verhagen AP, van Baar ME, Verhaar JA, Koes BW. Influence of obesity on the development of osteoarthritis of the hip: a systematic review. Rheumatology (Oxford) 2002;41(10):1155–1162. doi: 10.1093/rheumatology/41.10.1155. [DOI] [PubMed] [Google Scholar]
- 23.D'Apuzzo MR, Novicoff WM, Browne JA. The John Insall Award: Morbid obesity independently impacts complications, mortality, and resource use after TKA. Clin Orthop Relat Res. 2015;473(1):57–63. doi: 10.1007/s11999-014-3668-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Glassman SD, Carreon LY, Shaffrey CI, Polly DW, Ondra SL, Berven SH, Bridwell KH. The costs and benefits of nonoperative management for adult scoliosis. Spine. 2010;35(5):578–582. doi: 10.1097/BRS.0b013e3181b0f2f8. [DOI] [PubMed] [Google Scholar]
- 25.Liu S, Diebo BG, Henry JK, Smith JS, Hostin R, Cunningham ME, Mundis G, Ames CP, Burton D, Bess S, Akbarnia B, Hart R, Passias PG, Schwab FJ, Lafage V International Spine Study Group. The benefit of nonoperative treatment for adult spinal deformity: identifying predictors for reaching a minimal clinically important difference. Spine J. 2016;16(2):210–218. doi: 10.1016/j.spinee.2015.10.043. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.