Efficacy of the dynamic radiographs for diagnosing acute osteoporotic vertebral fractures

Development and validation of a machine learning model to predict imminent new vertebral fractures after vertebral augmentation

BACKGROUND

Accurately predicting the occurrence of imminent new vertebral fractures (NVFs) in patients with osteoporotic vertebral compression fractures (OVCFs) undergoing vertebral augmentation (VA) is challenging with yet no effective approach. This study aim to examine a machine learning model based on radiomics signature and clinical factors in predicting imminent new vertebral fractures after vertebral augmentation.

METHODS

A total of 235 eligible patients with OVCFs who underwent VA procedures were recruited from two independent institutions and categorized into three groups, including training set (n = 138), internal validation set (n = 59), and external validation set (n = 38). In the training set, radiomics features were computationally retrieved from L1 or adjacent vertebral body (T12 or L2) on T1-w MRI images, and a radiomics signature was constructed using the least absolute shrinkage and selection operator algorithm (LASSO). Predictive radiomics signature and clinical factors were fitted into two final prediction models using the random survival forest (RSF) algorithm or COX proportional hazard (CPH) analysis. Independent internal and external validation sets were used to validate the prediction models.

RESULTS

The two prediction models were integrated with radiomics signature and intravertebral cleft (IVC). The RSF model with C-indices of 0.763, 0.773, and 0.731 and time-dependent AUC (2 years) of 0.855, 0.907, and 0.839 (p < 0.001 for all) was found to be better predictive than the CPH model in training, internal and external validation sets. The RSF model provided better calibration, larger net benefits (determined by decision curve analysis), and lower prediction error (time-dependent brier score of 0.156, 0.151, and 0.146, respectively) than the CPH model.

CONCLUSIONS

The integrated RSF model showed the potential to predict imminent NVFs following vertebral augmentation, which will aid in postoperative follow-up and treatment.

Equivalent values between anterior vertebral height, wedge ratio, and wedge angle for evaluating vertebral mobility and deformity in osteoporotic vertebral fractures: a conventional observational study

BACKGROUND

Vertebral mobility (V-mobility) has been used to diagnose fresh osteoporotic vertebral fractures (OVFs) and determine bone union by setting cutoff values for these purposes. V-mobility is the difference in vertebral height on dynamic radiographs taken in the sitting and lateral decubitus or supine positions. The dimensions for V-mobility were presented as anterior vertebral height (Ha; mm), wedge ratio (WR; %), and wedge angle (WA; °) in previous reports. This study was performed to obtain WR and WA values equivalent to V-mobility of 1.0 mm in Ha.

METHODS

Lateral radiographs of 284 OVFs (grade 1-3 deformed vertebrae) from T11 to L2 were obtained from 77 patients with OVF. V-mobility presented as Ha, posterior vertebral height, and WA was obtained by the difference in these dimensions on dynamic radiographs. The WR and WA values equivalent to 1.0 mm in Ha were obtained by dividing the V-mobility values for WR and WA by that for Ha.

RESULTS

The mean WR values corresponding to 1.0 mm in Ha for grade 1, 2, and 3 vertebrae were 3.2% ± 1.4%, 3.2% ± 0.9%, and 3.4% ± 1.0%, respectively, and the corresponding value for grade 1-3 vertebrae was 3.3% ± 1.0%. The mean WA values corresponding to 1.0 mm in Ha for grade 1, 2, and 3 vertebrae were 1.5° ± 0.8°, 1.5° ± 0.6°, and 1.5° ± 0.8°, respectively, and the corresponding value for grade 1-3 vertebrae was 1.5° ± 0.7°.

CONCLUSIONS

The WR and WA values equivalent to V-mobility of 1.0 mm in Ha were 3.3% and 1.5°, respectively, in grade 1-3 vertebrae. These findings may be useful to secure a reliable value of V-mobility of OVFs using simultaneous measurements in three dimensions (Ha, WR, and WA) in clinical practice and to establish cutoff values for V-mobility to determine bone union.

A novel radiological assessment to identify acute vertebral compression fractures: A pilot observational study

Aim

The diagnosis of acute vertebral compression fractures (AVCFs) is often challenging. An alternative to magnetic resonance imaging, which may not always be available, includes a comparison of supine and sitting/standing position radiographs. However, this cannot be accomplished in patients with acute vertebral compression fractures who require emergency transport and are in severe pain. In this study, aimed to assess the diagnostic accuracy of comparing lateral-view radiographs of the thoracolumbar spine in supine and 30° head-elevated positions, which are less painful.

Methods

We retrospectively examined 30 patients with AVCFs who were transported by ambulance to our emergency department between June 2018 and May 2019. All underwent 30° head-elevated lateral-view thoracolumbar spine radiography and magnetic resonance imaging. We evaluated vertebral fractures by examining changes in vertebral wedging ratio (WR) from supine to 30° head-elevated position (Δ WR) using the following equation: Δ WR = WR (30° head-elevated) - WR (supine). We compared Δ WR to that of unfractured vertebrae as control.

Results

A total of 176 vertebrae were included (fractured, 32 and non-fractured, 144). Δ WR of fractured vertebrae ranged between 5.1% and 24.4%, whereas non-fractured vertebrae ranged between -6.7% and 4.3%. Median Δ WR of fractured vertebrae was significantly higher than non-fractured vertebrae (12.6% versus -0.5%, p < 0.001). No patients reported pain during 30° head-elevated positioning.

Conclusions

Lateral radiographs in supine and 30° head-elevated positions can accurately diagnose of AVCF, without worsening pain. This study showed a Δ WR value of ≥5.1% for AVCFs.

Therapeutic Effects of Conservative Treatment with 2-Week Bed Rest for Osteoporotic Vertebral Fractures: A Prospective Cohort Study

UPDATE

This article was updated on October 19, 2022, because of previous errors, which were discovered after the preliminary version of the article was posted online. On page 1787, in the legend for Figure 2, the sentence that had read "The vertebral collapse rate (in %) was defined as 1 - (A/P) × 100, and vertebral instability (in %) was defined as the difference in vertebral collapse rate between the loaded and non-loaded images." now reads "The vertebral collapse rate (in %) was defined as (1 - [A/P]) × 100, and vertebral instability (in %) was defined as the difference in vertebral collapse rate between the loaded and non-loaded images." On page 1788, in the section entitled "Data Collection," the sentence that had read "The vertebral collapse rate (in %) was defined as 1 - (anterior vertebral wall height/posterior vertebral wall height) × 100, and vertebral instability (in %) was defined as the difference in vertebral collapse rate between the loaded and non-loaded images 9 ." now reads "The vertebral collapse rate (in %) was defined as (1 - [anterior vertebral wall height/posterior vertebral wall height]) × 100, and vertebral instability (in %) was defined as the difference in vertebral collapse rate between the loaded and non-loaded images 9 ." Finally, on page 1791, in Table IV, the footnote for the "Primary outcome" row that had read "N = 113 in the rest group and 99 in the no-rest group." now reads "N = 116 in the rest group and 108 in the no-rest group."

Equivalent values between anterior vertebral height, wedge ratio, and wedge angle in osteoporotic vertebral fractures

INTRODUCTION

Vertebral mobility (V-mobility) has been used to diagnose fresh osteoporotic vertebral fractures (OVFs), and determine or predict bone union by setting cutoff values for these purposes. V-mobility is defined as the difference in shape of vertebral bodies between lateral radiographs taken in weight-bearing and non-weight-bearing positions. The parameters for V-mobility have varied in previous reports among anterior vertebral height (Ha, mm), wedge ratio (WR, %), and wedge angle (WA, degrees). The present study aimed to clarify WR and WA equivalent to Ha of 1.0 mm, and to compare the reported cutoff values for V-mobility presented as Ha, WR, or WA.

MATERIALS AND METHODS

Lateral radiographs of 446 normal vertebrae (grade 0) and 146 deformed vertebrae (grade 1-3) from T11 to L2 were obtained from 183 female patients aged > 60 years. WR (%) values equivalent to Ha of 1.0 mm were calculated by Ha (1.0 mm)/Hp × 100 (Hp: posterior vertebral height). Corresponding WA values were calculated by trigonometric function using vertebral dimensions.

RESULTS

The mean WR values equivalent to Ha of 1.0 mm in the vertebrae from T11 to L2 were 3.2%, 3.2%, 3.5%, and 3.7% for grades 0, 1, 2, and 3, respectively, and the corresponding WA values were 1.6°, 1.6°, 1.5°, and 1.4°.

CONCLUSION

The equivalent values for V-mobility presented as Ha, WR, and WA were obtained. The mean WR and WA values equivalent to Ha of 1.0 mm in grade 1-3 vertebrae were 3.5% and 1.5°, respectively.

Effects of conservative treatment of 2-week rigorous bed rest on muscle disuse atrophy in osteoporotic vertebral fracture patients

Objective: Osteoporotic vertebral fracture (OVF) is conventionally treated with conservative management such as bed rest, but a relatively prolonged bed rest has the potential risk of muscle disuse atrophy. This study aimed to examine whether the 2-week of rigorous bed rest affects muscle disuse atrophy in OVF patients.

Patients and Methods: A total of 54 OVF patients (16 males; 38 females; mean age, 80.2 ± 9.2 years) were treated with an initial 2-week rigorous bed rest by hospitalization with persistent rehabilitation. Cognitive function, swallowing function, grip strength, and lower extremity circumference were evaluated at three-time points (admission, end of bed rest, and discharge).

Results: Of the 51 patients who were able to walk independently before the injury, one patient (2.0%) had to use a wheelchair after the injury. During hospitalization, cognitive function decline was observed in 33.3% of patients, but not in patients with Revised Hasegawa’s Dementia Scale score ≥25 at admission. Swallowing function decline was observed in one patient, and none of the patients developed aspiration pneumonia during hospitalization. The grip strength significantly improved both at the end of bed rest (P=0.04) and discharge (P=0.02). Although the lower extremity circumference significantly decreased at the end of bed rest (P<0.01), it was recovered afterward. The lower extremity circumference did not significantly differ between the admission and discharge (P=0.17).

Conclusion: Our results suggested that conservative treatment of OVF through an initial 2-week rigorous bed rest with persistent hospital rehabilitation poses a low risk of muscle disuse atrophy. If cognitive dysfunction is observed on admission, close monitoring for exacerbation should be performed during the hospital stay.

Opportunistic diagnosis of osteoporotic vertebral fractures on standard imaging performed for alternative indications

Osteoporotic vertebral fractures (VFs) are the most common type of osteoporotic fracture. Patients with VF are at increased risk of hip fractures or additional VFs, both of which contribute to patient morbidity and mortality. Early diagnosis of VFs is essential so patients can be prescribed appropriate medical therapy.

Most patients with clinical suspicion for VF have an X-ray of the spine. Many VFs are invisible on X-ray and require further imaging. CT can provide excellent bony detail but uses high doses of ionising radiation. MRI provides excellent soft tissue detail and can distinguish old from new fractures in addition to differentiating osteoporotic VFs from other causes of back pain. Bone scans have a limited role due to poor specificity.

The literature suggests that radiologists frequently miss or do not report VFs when imaging is requested for an alternative clinical indication and when there is no clinical suspicion of VF. Common examples include failure to identify VFs on lateral chest X-rays, sagittal reformats of CT thorax and abdomen, lateral localizers on MRI and scout views on CT.

Failure to diagnose a VF is a missed opportunity to improve management of osteoporosis and reduce risk of further fractures. This article discusses the role of radiographs, CT, MRI and bone scintigraphy in the assessment and recognition of osteoporotic fractures. This article focuses on opportunistic diagnosis of VFs on imaging studies that are performed for other clinical indications. It does not discuss use of dual energy X-ray absorptiometry which is a specific imaging modality for osteoporosis.

Vertebral mobility is a valuable indicator for predicting and determining bone union in osteoporotic vertebral fractures: a conventional observation study

Background

Conservative treatments for osteoporotic vertebral fractures (OVFs) have not been standardized, and criteria for determining bone union have not been established. To determine bone union, we have adopted a cutoff value of 1.0 mm for vertebral mobility (V-mobility), defined as the difference in anterior vertebral height (Ha) between lateral radiographs taken in weight-bearing and non-weight-bearing positions. The present study aimed to investigate the usefulness of V-mobility for determining bone union and predicting bone union at 6 months after OVF onset.

Methods

The study included 54 acute OVFs from T11 to L3 in 53 patients (12 males, 41 females; mean age 82 years; age range 55–97 years) who were hospitalized at ≤ 3 weeks after OVF onset. Vertebral deformity (V-deformity) and V-mobility were evaluated in accordance with Ha on lateral radiographs taken in the sitting position (SIT), lateral decubitus position (DEC), and supine position (SUP). OVFs showing V-mobility of ≤ 1.0 mm between SIT and DEC radiographs and no intravertebral cleft on DEC radiograph were defined as semi-union, while those showing V-mobility of ≤ 1.0 mm between SIT and SUP radiographs and no intravertebral cleft on SUP radiograph were defined as bone union. We calculated the bone union rates including semi-unions associated with V-mobility cutoff values of 1.0 mm, 1.5 mm, and 2.0 mm and estimated cutoff values for V-mobility at 5 weeks after OVF onset to predict bone union at 6 months after OVF onset.

Results

The cumulative number of bone unions including semi-unions was more influenced by the different V-mobility cutoff values in Ha for determining bone union in the earlier period compared with the later period in the time course of OVF. Receiver-operating characteristic curve analyses revealed that V-mobility cutoff value of 2.1 mm in Ha between SIT and DEC radiographs at 5 weeks after OVF had moderate accuracy for predicting bone union including semi-union at 6 months after OVF. The mean V-deformity value on SIT radiographs did not progress significantly.

Conclusion

V-mobility in the early stage after OVF can predict bone union at 6 months after OVF and is a useful quantitative indicator for determining bone union.

Effects of Weekly Teriparatide Administration for Vertebral Stability and Bony Union in Patients with Acute Osteoporotic Vertebral Fractures

Study Design

An open-label, non-randomized prospective study.

Purpose

Teriparatide (TPTD) is known to be an antiosteoporotic agent that may accelerate the healing of fractures. This study was designed to investigate the effect of once-weekly TPTD administration on vertebral stability and bony union after acute osteoporotic vertebral fracture (OVF).

Overview of Literature

Once-weekly TPTD administration can lead to early vertebral stability and promote bony union of fractured vertebrae in patients with severe osteoporosis.

Methods

Forty-eight subjects with acute OVF were assigned to receive activated vitamin D3 and calcium supplementation or onceweekly subcutaneous injection of TPTD (56.5 μg) in combination with activated vitamin D3 and calcium supplementation for 12 weeks. Vertebral stability was assessed using lateral plain radiography. Vertebral height at the anterior location (VHa) and the difference in VHa {ΔVHa=VHa (supine position)−VHa (weight-bearing position)} were measured at baseline and 12 weeks after starting treatment. Bony union was defined as the absence of a vertebral cleft or abnormal motion (ΔVHa >2 mm).

Results

Although not significant, ΔVHa, indicating vertebral stability, tended to be lower in the TPTD group at 12 weeks (p =0.17). As for subjects with severe osteoporosis, ΔVHa at 12 weeks was significantly lower in the TPTD group than in the control group (mean ΔVHa: control group, 3.1 mm (n=15); TPTD group, 1.4 mm (n=16); p =0.02). The rate of bony union was significantly higher in the TPTD group than in the control group (control group, 40%; TPTD group, 81%; p =0.03).

Conclusions

Once-weekly TPTD administration may facilitate early bony union after acute OVF accompanied by severe osteoporosis.

Initial hospitalization with rigorous bed rest followed by bracing and rehabilitation as an option of conservative treatment for osteoporotic vertebral fractures in elderly patients: a pilot one arm safety and feasibility study

We assessed the safety and feasibility of a unified conservative treatment protocol for osteoporotic vertebral fractures in the elderly patients with a 24-week follow-up. Our results showed that initial hospitalization with rigorous bed rest followed by a rehabilitation program using a Jewett brace was safe and feasible in managing patients.

PURPOSE

The purpose of this study was to prove the safety and feasibility of a unified conservative treatment protocol, which included initial hospitalization with rigorous bed rest followed by a rehabilitation program with Jewett brace for osteoporotic vertebral fractures (OVFs) in the elderly patients with a 24-week follow-up.

METHODS

Between April 2012 and Mach 2015, one hundred fifty-four patients met the eligibility for this study. Radiological findings at the 3-week, 6~8-week, 24-week assessment were evaluated. Among these, 11 patients underwent early surgery within the first 2 weeks after admission and 19 patients lost follow-up. Therefore, 124 patients were assessed at the final follow-up visit.

RESULTS

The average vertebral instability in all the present series was 4.9 ± 4.8° at 3-week, 2.9 ± 3.5° at 6~8-week, and 1.8 ± 3.0° at 24-week follow-up visit. Delayed union was observed in 16 patients on the 24-week follow-up visit. Therefore, the present conservative treatment protocol resulted in bony union in 98 out of 124 patients (79.0%, per protocol set analysis) and 98 out of 154 patients including drop-out (63.6%, intention-to-treat analysis). There was no severe adverse event related to initial bed rest. The vertebral instability at 3-week assessment was significantly higher in the delayed union group when compared with that in the union group. Univariate analyses followed by multivariate logistic regression analysis revealed that T2-weighted image of confined high intensity on MRI and having more than 5° of vertebral instability on dynamic X-ray at 3-week assessment are the independent risk factors for delayed union of conservative treatment in the present series.

CONCLUSIONS

Our results showed that initial hospitalization with rigorous bed rest followed by a rehabilitation program using a Jewett brace was safe and feasible. Therefore, the present conservative treatment protocol can be one of the acceptable treatment options in managing OVF patients.

A retrospective analysis of nonresponse to daily teriparatide treatment

Some patients with osteoporosis do not respond to teriparatide treatment. Prior bisphosphonate use, lower bone turnover marker (BTMs) concentrations, and lower early increases in BTMs were significantly associated with a blunted lumbar spine (LS) bone mineral density (BMD) response to daily treatment with teriparatide, although the impact was limited.

INTRODUCTION

Some osteoporosis patients do not respond to teriparatide treatment. To better understand the factors underlying treatment nonresponses, we compared nonresponders' and responders' characteristics.

METHODS

We retrospectively analyzed 354 male and female patients with osteoporosis who were administered teriparatide (20 μg/day) for 24 months. The patients were categorized as responders (≥3 % lumber spine (LS) bone mineral density (BMD) increase) or nonresponders (<3 % LS BMD increase), and the groups were compared.

RESULTS

The univariate analyses determined that prior bisphosphonate use, a lower baseline procollagen type I N-terminal propeptide (PINP) concentration and a lower urinary N-telopeptide of type I collagen (uNTX) concentration at baseline were significantly associated with teriparatide nonresponses, but these factors were not significant following multivariate analysis. Diminished early increases in the bone turnover markers (BTMs) were also related to nonresponses after teriparatide treatment began. In the nonresponders, the mean (standard deviation (SD)) absolute LS and femoral neck (FN) BMD changes were -0.002 g/cm(2) (0.032) and -0.010 g/cm(2) (0.045), respectively. In the responders, the mean (SD) absolute LS and FN BMD changes were 0.118 g/cm(2) (0.056) and 0.021 g/cm(2) (0.046), respectively. The serum PINP and uNTX levels increased rapidly in both groups, but the responders showed higher early absolute serum PINP and uNTX increases.

CONCLUSIONS

The factors associated with nonresponses were prior bisphosphonate use, lower baseline BTM levels, and lower early increases in the BTMs after starting teriparatide treatment, but the impact of these factors on achieving a ≥3 % LS BMD increase at 24 months was limited.

Violet Fox: A Clinical View of Vertebral Fractures

Had Violet's abdominal MR not been performed, or its findings not appreciated, the cause of her clinical event might never have been known because our current concept of osteoporotic vertebral fracture (VF) is substantially predicated on a change in either vertebral height or shape on lateral or sagittal spine imaging. The intention of this commentary is to stimulate a multidisciplinary conversation of osteoporotic VFs from an integrated clinical, physiological, and imaging perspective. For research and epidemiological purposes, osteoporotic VFs have been defined as a reduction in anterior, middle, or posterior vertebral height although the required minimum height reduction (e.g., 15% or 20%) varies among definition schemes. We further classify osteoporotic VFs to be "clinical" when they are accompanied by back pain and "morphometric" when they are not, and we have generally accepted the assertion that most of the osteoporotic VFs are painless, that is, morphometric. This dichotomous VF definition scheme has been the foundation of osteoporosis epidemiology and the primary endpoint in most pivotal osteoporosis pharmaceutical trials. Although, having served the osteoporosis community well, our clinical experience, refined by recent insights into vertebral anatomy and spinal biomechanics, advances in vertebral imaging, and 2 decades of vertebral augmentation suggest that the spectrum of osteoporotic VFs is more complicated than this scheme suggests.