External validation of a deep learning model for predicting bone mineral density on chest radiographs.

We developed a new model for predicting bone mineral density on chest radiographs and externally validated it using images captured at facilities other than the development environment. The model performed well and showed potential for clinical use. PURPOSE: In this study, we performed external validation (EV) of a developed deep learning model for predicting bone mineral density (BMD) of femoral neck on chest radiographs to verify the usefulness of this model in clinical practice. METHODS: This study included patients who visited any of the collaborating facilities from 2010 to 2020 and underwent chest radiography and dual-energy X-ray absorptiometry (DXA) at the femoral neck in the year before and after their visit. A total of 50,114 chest radiographs were obtained, and BMD was measured using DXA. We developed the model with 47,150 images from 17 facilities and performed EV with 2914 images from three other facilities (EV dataset). We trained the deep learning model via ensemble learning based on chest radiographs, age, and sex to predict BMD using regression. The outcomes were the correlation of the predicted BMD and measured BMD with diagnoses of osteoporosis and osteopenia using the T-score estimated from the predicted BMD. RESULTS: The mean BMD was 0.64±0.14 g/cm2 in the EV dataset. The BMD predicted by the model averaged 0.61±0.08 g/cm2, with a correlation coefficient of 0.68 (p<0.01) when compared with the BMD measured using DXA. The accuracy, sensitivity, and specificity of the model were 79.0%, 96.6%, and 34.1% for T-score < -1 and 79.7%, 77.1%, and 80.4% for T-score ≤ -2.5, respectively. CONCLUSION: Our model, which was externally validated using data obtained at facilities other than the development environment, predicted BMD of femoral neck on chest radiographs. The model performed well and showed potential for clinical use.

Inter-observer Agreement and Reproducibility of Pertrochanteric Fracture Classification Using Plain Radiograph Versus Computed Tomogram Images: A Study of 523 Patients.

Background A precise preoperative imaging classification system for pertrochanteric fractures is imperative due to the reported unreliability of the current classification system, which relies solely on plain radiographs. This study aims to achieve two primary objectives: (i) elucidate the reproducibility of pertrochanteric fracture evaluation based on the Revised Arbeitsgemeinschaft für Osteosynthesefragen/Orthopedic Trauma Association (AO/OTA) Classification, comparing plain radiographs and computed tomography (CT) scan images, and (ii) investigate the consistency of fracture classification between both imaging modalities. Methods A total of 523 patients (112 males and 411 females, mean age 85 years) who had both preoperative plain radiographic and three-dimensional CT images were enrolled in this study. Following the Revised AO/OTA Classification, three individual observers initially classified the fractures in plain radiograph images as either Stable (A1) or Unstable (A2). Subsequently, they further categorized them into five sub-categories (A1.1, A1.2, A1.3, A2.2, and A2.3). The same classification system was applied to the CT scan images. Inter-observer agreement and consistency of fracture classification between plain radiographs and CT scan images were assessed. Results The inter-observer agreement for fractures classified as stable or unstable using only plain radiographs was found to be fair among the three observers, with a mean κ of 0.397 (95% CI: 0.316-0.478). However, inter-observer agreement improved significantly when using CT scans, with a mean κ of 0.590 (95% CI: 0.518-0.662). Our results demonstrated a consistency level between two graphical modalities ranging from fair to moderate, with κ values of 0.581, 0.383, and 0.335, respectively. It's worth noting that plain radiographic classification occasionally resulted in underestimations, with each observer identifying 16.1%, 34.0%, and 37.9%, respectively, of cases as A1 in plain radiographs that were classified as A2 in CT scans. Conclusions This study reveals a moderate to substantial level of inter-observer agreement for fracture classification when using CT scan images, in contrast to plain radiographs. Fracture evaluation relying solely on plain radiographs sometimes underestimates fracture classification and exhibits less consistency compared to using CT scan images.

A Scientific Proposal for Surgical Decision-Making in Occult Intertrochanteric Fractures Based on Finite Element Analysis.

Background In the treatment of femoral intertrochanteric fractures, there is still a lack of consensus on the optimal approach for isolated greater trochanteric fractures and insufficient intertrochanteric fractures. The limited number of patients and restricted access to accurate assessment of fracture extension using magnetic resonance imaging contribute to the unclear treatment strategy. This study aims to utilize finite element (FE) analysis to analyze stress values at the fracture line and investigate their influence on intertrochanteric fracture extension under different loading conditions. The hypothesis is that fracture extension occurs following certain conditions, supporting the need for surgery based on scientific evidence. Methodology Osseous data from a computed tomography (CT) scan was used to create a proximal femur FE model using FEA software. CT scan data were converted to Digital Imaging and Communications in Medicine format and used to generate the FE model. Trabecular bone and cortex were meshed into tetrahedral elements. The model consisted of 1,592,642 elements and 282,530 nodes. Two models were created, namely, healthy proximal femur (HF) and femoral insufficient intertrochanteric fracture (FIF). Material properties were assigned based on CT values and conversion equations. The distal end of the femur was constrained. Stress analysis using the dynamic explicit approach was performed. Von Mises stresses were calculated for the proximal femur. The number of elements exceeding yield stress was counted to predict fracture risk by focusing on fracture line spots. In this study, the distribution of von Mises stress was compared between the HF and the FIF models. Six loading combinations were considered, namely, two weight-bearing conditions (3 W loading simulating for walking and 1/3 W for touch-down standing) and three hip flexion angles (0°, 15°, and 23°). Results Under 3 W loading, no significant stress elevations were observed in the HF model at any flexion angles. However, the FIF model exhibited increased stress at the site of the posterior fracture line extension. This stress-induced element destruction was observed in both cortical and cancellous bone. For the 1/3 W loading condition, only minimal stress elevation was observed in both HF and FIF models. To assess the influence on fracture extension, the number of yielded elements was evaluated along the fracture line edges (greater trochanter and middle of the intertrochanteric ridge). Under 3 W loading, the HF model had only one yielded element, indicating minimal fracture risk. In contrast, the FIF model exhibited a notable presence of yield elements in various regions (total/greater trochanter/shaft) at different flexion angles: 0° (115/16/28), 15° (265/158/23), and 23° (446/233/34). Under the 1/3 W loading condition, neither the HF nor the FIF models showed any yielding elements, regardless of the direction of external force. Conclusions The results demonstrated elevated stress levels at the fracture line in the FIF model, particularly during walking, indicating a higher risk of fracture extension at the flex position. However, under reduced weight-bearing conditions, the stress at the fracture site remained within the yield stress range, suggesting a relatively low risk of fracture extension. These findings hold significant clinical implications for developing surgical protocols that consider patients' compliance with weight-bearing restrictions.

Intra-soft tissue and intramedullary antibiotic perfusion in combination with negative pressure wound therapy.

OBJECTIVE: Intra-soft tissue and intramedullary antibiotic perfusion (iSAP and iMAP), which combine continuous administration of antibiotic solution to the wound and negative pressure wound therapy (NPWT), have been reported to be a useful management approach for hard-to-heal ulcers in the field of orthopaedic surgery. We report the efficacy of this treatment and discuss the key points. METHOD: The recipients of this treatment had contaminated fresh severe trauma with a high risk of infection, or hard-to-heal ulcers which were expected to be difficult to manage with conventional NPWT alone. Continuous administration of 1200µg/ml of gentamicin (GM) solution to the wound was performed along with NPWT. The GM solution was administered subcutaneously using a small catheter for iSAP, while intramedullary administration used a bone marrow needle for iMAP. RESULTS: iSAP was employed in all 10 patients who took part, and iMAP in three of these patients. The average treatment time was 13.6 days with iSAP and 9.3 days with iMAP. The mean serum GM level during the therapy was 1.02µg/ml. Moderate GM-induced acute kidney injury was suspected in one case, but resolved spontaneously after GM administration was stopped. Favourable wound bed preparation was achieved in all cases without recurrence of infection. CONCLUSION: Combination with continuous suction by NPWT is able to keep the local concentration of antibiotic above the minimum inhibitory concentration of biofilm-coated bacteria within the wound. We have referred to this treatment as continuous local antibiotic perfusion. Further investigation of local pharmacodynamics in the wound and side-effects of this treatment are warranted.

Multicenter study on atypical femoral fractures in patients with bone metastases taking bone- modifying agents.

Bone-modifying agents (BMAs), with bone-resorptive inhibitory effects, such as zoledronic acid and denosumab, are widely used at higher doses for bone-related events caused by bone metastasis of malignant tumors. These drugs have been suggested to be associated with atypical femoral fractures (AFFs), and the relationship between BMAs and AFFs has attracted attention. To investigate the clinical features including bone union time of AFFs in patients administered BMA for bone metastasis, we conducted a retrospective multicenter study. Thirty AFFs from 19 patients were enrolled in this study. Thirteen patients had bilateral AFFs, and nineteen AFFs had prodromal symptoms. Eighteen AFFs underwent surgery after complete fracture, three failed to achieve bone union and required nonunion surgery, and 11 AFFs that achieved bone union had an average period until bone union of 16.2 months, which was much longer than that previously reported for ordinary AFFs. Seven patients discontinued the BMAs, but not due to AFFs. Stopping BMAs in patients with bone metastasis would make it difficult to secure their performance of activities of daily living, and AFF with BMA administration might require a longer time for union. Therefore, it would be important to prevent incomplete AFF from becoming complete AFF via prophylactic internal fixation.

Comparative Study of Different Entry Spots on Postoperative Gluteus Medius Muscle Cross-Sectional Area in Patients With Intertrochanteric Fractures Nailing.

Introduction In a preliminary study of cephalo-medullary (CM) nailing in patients with femoral intertrochanteric fractures, the authors of this study found a 25% to 30% decrease in muscle strength, especially abduction force, during the postoperative follow-up period. This decline was partially attributed to the entry point for the nail insertion causing damage to the gluteus medius tendon at the junction of the greater trochanter after reaming. Therefore, we assumed that changing the position of nail insertion to a "bald spot (BS)" could mitigate postoperative functional impairment. Automated computed tomography (CT) imaging of skeletal muscle cross-sectional area (CSA) and adipose tissue ratio (ATR) can show pathological changes on the operated side compared with the non-operated side. In this study, the authors quantified the difference in postoperative CSA and ATR of the gluteus medius muscle after bald spot nailing versus nail insertion through the conventional tip of the greater trochanter. It was hypothesized that bald spot nailing could avoid significant injury to the gluteus medius muscle. Materials and methods Patients with femoral intertrochanteric fractures were grouped according to the site of cephalo-medullary nailing: greater trochanteric tip (TIP) in 27 patients (8 men and 19 women, mean age 84.9±5.1 years) and BS in 16 patients (3 men and 13 women, mean age 86.9±6.2 years). The CSA and ATR of the gluteus medius muscles were assessed in three slices (A, B, and C from proximal to distal). Each slice was manually traced and automatically calculated based on its contour. Adipose tissue (-100 to -50 in Hounsfield units) in the designated area was distinguished by a bimodal image histogram resulting from the distribution of CT numbers of adipose tissue and muscle. The body mass index (BMI) was used to correct the CSA in each patient. Results In the TIP group, the mean CSA values (mm2) from the non-operated/operated sides were as follows: slice A, 2180.2 ± 616.5/1976.3 ± 421.2; slice B, 2112.3 ± 535.7/1857.7 ± 386.7; and slice C: 1671.8 ± 460.0/1404.1 ± 404.3 (p<0.01 in slices A, B, and C). In the BS group, slice A was 2044.1 ± 473.0/2016.9 ± 388.4; slice B was 2073.2 ± 540.7/1848.3 ± 411.1; and slice C was 1659.1 ± 477.2/1468.5 ± 341.7 (p=0.34 in slice A, and p<0.05 in slices B and C, respectively). The mean CSA values (mm2) of the non-operated minus operated side between the TIP/BS groups were as follows: slice A, 241.3 ± 424.3/-11.8 ± 285.6; slice B, 290.3 ± 313.0/211.8 ± 333.2; and slice C, 276.4 ± 270.4/162.8 ± 319.3 (p < 0.05 in slice A, 0.45, 0.24 in slices B, C, respectively). The mean adjusted CSA per BMI values (mm2) of the non-operated minus the operated side between the TIP/BS groups were slice A, 10.6 ± 19.7/-0.4 ± 14.8; slice B, 13.3 ± 15.0/10.1 ± 16.3; and slice C, 13.1 ± 13.4/ 8.7 ± 15.3 (p < 0.05 in slice A and 0.54 and 0.36 in slices B and C, respectively). Conclusion Nail insertion at the bald spot resulted in a significantly smaller decrease in the CSA of the gluteus medius muscle compared with the conventional tip entry. In addition, an examination of BMI-adjusted CSA showed that CSA was maintained in some image slices. These results suggest that nailing from the BS of the greater trochanter can reduce damage to the gluteus medius muscle and highlight the importance of imaging beyond the usual assessment of skeletal changes.

A Demographic Survey of Pertrochanteric Fractures Based on the Revised Arbeitsgemeinschaft für Osteosynthesefragen/Orthopedic Trauma Association (AO/OTA) Classification Using 3D CT Scan Images.

Introduction A demographic survey of femoral pertrochanteric fractures provides several important information for the healthcare system of a country since this fracture is commonly seen in the elderly and has a poor postoperative functional prognosis that is a burden on society. The importance of accurately classifying pertrochanteric fractures as stable or unstable cannot be understated. However, the use of plain radiograph images alone is known to underestimate fracture severity with low inter- or intra-observer agreement. Computed tomography (CT) images offer information for a more accurate classification of pertrochanteric fractures. With this three-dimensional (3D) CT-based study using the revised Arbeitsgemeinschaft für Osteosynthesefragen/Orthopedic Trauma Association (AO/OTA) classification, the purpose of this study is to elucidate the epidemiological demography of patients with pertrochanteric fractures. Material and methods We retrospectively collected 808 patients from five hospitals, classified into two groups: stable (A1) or unstable (A2). Age, gender, fracture laterality, and surgery timing were identified as epidemiological variables. Patients with both preoperative plain radiographs and 3D CT scans were included in the study. The exclusion criteria were AO/OTA A3 type fractures, pathological fractures, previous ipsilateral surgery, 60 years old or younger, and conservatively treated patients. The primary outcome involved detailing the total number of fractures based on classification (A1 or A2) and variables. The secondary outcome involved a comparison between the A1 and A2 groups. Results The mean age of patients at the time of surgery was 85 years (range: 61-103 years). There were 637 female and 171 male patients. There were 463 left-sided fractures and 345 right-sided fractures. Of the 808 patients, 371 (45.9%) were classified to have A1 fractures, and 437 (54.1%) had A2 fractures. The age at surgery, gender, fracture laterality, and surgery timing between the A1 and A2 groups were compared. The mean and standard deviation of the age at surgery for patients in the A1 and A2 groups were 84.9±7.7 and 86.9±6.8, respectively. The number of patients for each age distribution of 61-69, 70-74, 75-79, 80-84, 85-89, 90-94, and 95 or older for the A1 and A2 groups was 18 and 7, 18 and 12, 43 and 44, 76 and 82, 107 and 132, 79 and 110, and 30 and 50, respectively, showing that the difference in categorial distribution was statistically significant (p=0.002). Overall, 278 females and 93 males were classified to have A1 fractures compared with 359 females and 78 males with A2 fractures (p=0.01). There were 166 right-sided and 205 left-sided stable A1 fractures and 179 right-sided and 258 left-sided A2 fractures (not significant (NS)). Among the total number of A1 and A2 surgeries by month, the most were in December with 77 surgeries (37 and 40, respectively), and the least was in June with 37 (18 and 19, respectively). The seasonal classification for A1 and A2 surgeries is as follows: spring with 172 (74 and 98, respectively), summer with 150 (70 and 80, respectively), autumn with 193 (90 and 103, respectively), and winter with 208 (97 and 111, respectively) (NS). Conclusion In this demographic study of 808 patients with pertrochanteric fractures classified by 3D CT images, 371 had A1 fractures and 437 had A2 fractures. A2 fractures were significantly more in females with an age peak of 85-89 years.

Posterior Protrusion Measures (PPM) for Three-Dimensional (3D) CT Classification of Pertrochanteric Fractures.

Introduction We introduced a novel numerical index known as posterior protrusion measures (PPM), derived from lateral plain radiograph images, which effectively serves to distinguish stable from unstable pertrochanteric fractures. The present study aims to scrutinize PPM values among two classified fracture patterns, stable and unstable, within the three-dimensional (3D) CT classification system, establishing a numeric threshold for PPM to differentiate between these groups; explore the potential relationship between the PPM index and unclassified categories; investigate how groups divided by the PPM threshold value can predict fracture stability based on 3D CT. Materials and methods In this study, three observers were tasked with measuring PPM on a single occasion. The chi-square test assessed the association between each demographic parameter on a categorical scale and stable/unstable groups. Continuous variables were also subject to examination. Receiver operating characteristic (ROC) analysis was employed to determine optimal cut-off points of PPM for predicting the presence of stable versus unstable groups. Additionally, the chi-square test examined the linear relation between separated groups based on the defined threshold PPM value and the stable/unstable groups. Results A total of 106 pertrochanteric fractures were identified using CT scan images and plain radiographs in the 3D CT classification system, revealing the stable group of 35 patients and the unstable group of 71 patients. The PPM values for stable/unstable fractures were, on average (± standard deviation), 0.34±0.25/0.50±0.29 for observer 1, 0.31±0.23/0.57±0.31 for observer 2, and 0.41±0.29/0.57±0.26 for observer 3, respectively (p<0.01). We established 0.3 as the cut-off value for PPM. The average PPM value among three observers represented each patient to assess fracture stability. The group with PPM <0.3 included 27 patients (16 stable and 11 unstable), and the group with PPM ≥0.3 group comprised 79 patients (19 stable and 60 unstable; p<0.005). Conclusion The present study revealed a significant difference in PPM values among stable and unstable 3D CT classification groups. Additionally, a threshold PPM value of 0.3 suggests a pivotal point for differentiating fracture stability. This innovative methodology makes a substantial contribution to clinical endeavors, potentially circumventing the necessity for 3D CT scanning.

Deep Learning for Bone Mineral Density and T-Score Prediction from Chest X-rays: A Multicenter Study.

Although the number of patients with osteoporosis is increasing worldwide, diagnosis and treatment are presently inadequate. In this study, we developed a deep learning model to predict bone mineral density (BMD) and T-score from chest X-rays, which are one of the most common, easily accessible, and low-cost medical imaging examination methods. The dataset used in this study contained patients who underwent dual-energy X-ray absorptiometry (DXA) and chest radiography at six hospitals between 2010 and 2021. We trained the deep learning model through ensemble learning of chest X-rays, age, and sex to predict BMD using regression and T-score for multiclass classification. We assessed the following two metrics to evaluate the performance of the deep learning model: (1) correlation between the predicted and true BMDs and (2) consistency in the T-score between the predicted class and true class. The correlation coefficients for BMD prediction were hip = 0.75 and lumbar spine = 0.63. The areas under the curves for the T-score predictions of normal, osteopenia, and osteoporosis diagnoses were 0.89, 0.70, and 0.84, respectively. These results suggest that the proposed deep learning model may be suitable for screening patients with osteoporosis by predicting BMD and T-score from chest X-rays.

Ampicillin/sulbactam versus cefazolin plus aminoglycosides for antimicrobial prophylaxis in management of Gustilo type IIIA open fractures: A retrospective cohort study.

BACKGROUND: The antibiotic regimens for prophylaxis in the management of open fractures remain controversial. Although the use of aminoglycosides is widely accepted for treatment of Gustilo type III open fractures, aminoglycosides are often avoided in patients with risk factors. This study aimed to compare efficacy and safety of two regimens, cephazolin plus aminoglycoside (amikacin or gentamicin) and ampicillin/sulbactam (ABPC/SBT), in patients with Gustilo type IIIA open fractures. METHODS: A total of 95 Gustilo type IIIA fractures in 90 patients were retrospectively reviewed in this study. The cohort was categorized into two groups that were treated in accordance with the institutional prescribed regimen in different periods: (1) cefazolin plus aminoglycoside (January 1, 2014-September 30, 2017) and (2) ABPC/SBT monotherapy (October 1, 2017-September 30, 2020). Cefazolin was used at 1-2 g every 8 h, aminoglycoside (amikacin or gentamicin) was used daily depending on body weight, and ABPC/SBT was used at 3 g every 8 h The antibiotic administration was continued within 3 days or until successful soft tissue coverage was achieved. The infection rate and the incidence of acute kidney injury (AKI) in both groups were assessed. RESULTS: ABPC/SBT was used in 34 patients (36 fractures), and 56 patients (59 fractures) received cefazolin plus aminoglycoside for antibiotic prophylaxis. Infection developed in 2 of 36 fractures in ABPC/SBT group and 4 of 59 fractures in the cefazolin plus aminoglycoside group (p > 0.99). The average serum creatinine levels on admission, baseline, and peak during the hospital stay were not significantly different between the two groups. One case of AKI was identified in each group, indicating that incidence rate of AKI was not significantly different between the two groups. CONCLUSION: We demonstrated the non-inferiority of ABPC/SBT therapy over cefazolin plus aminoglycoside regimen for type IIIA open fractures. The ABPC/SBT regimen may be an alternative option for managing Gustilo type IIIA open fractures. Further prospective studies with larger samples are needed to verify these results.

Continuous Local Antibiotics Perfusion Therapy for Acute Deep Infections after Open Fractures.

Unresolved bone and soft tissue infections remain a great hindrance to fracture management worldwide, both economically and functionally for the patient. For this purpose, the benefits of local antibiotic administration besides systemic therapy have been elucidated. We present a retrospective descriptive analysis of six patients (4 males and 2 females) with acute deep infections after open fractures managed using the continuous local antibiotic perfusion (CLAP) therapy. After sufficient debridement, gentamicin solution concentrated at 1,200 µg/mL was continuously infused (2 mL/h) for 7-12 days by syringe pump through an inlet tube placed on the infected area. The antibiotics injected into the infected area were both collected and perfused by negative pressure using a negative-pressure wound therapy system. After an average of 9.5 days of CLAP therapy, symptoms of infection disappeared, and the bacterial culture was negative. There were no cases of recurrence during the follow-up period, and no complications, such as acute renal failure, ototoxicity, allergic or hypersensitivity reactions, and impaired fracture healing, were observed. All six cases were successfully managed with the CLAP therapy without any serious side effects. CLAP therapy may be a potential treatment option for acute deep infections after open fractures.

Endochondral Bone Tissue Engineering Using Human Induced Pluripotent Stem Cells.

The use of induced pluripotent stem cells (iPSCs) shows potential in bone regenerative strategies. In this study, we investigated whether implantation of chondrogenically differentiated iPSC-derived mesenchymal stem cells (iMSCs) can lead to successful bone regeneration in nude mice with bone defects. Two human iPSC clones (201B7 and 454E2) were used. After generating iMSCs, chondrogenic differentiation was achieved by three-dimensional pellet culture. Thereafter, a 2-mm defect was created in the radius of nude mice, and chondrogenically differentiated iMSC pellets were transplanted in the defect. Microcomputed tomography imaging was performed 8 weeks posttransplantation to assess bone regeneration. All (100%) radii in the 201B7 cell-derived pellet transplantation group and 7 of 10 (70%) radii in the 454E2 cell-derived pellet transplantation group showed bone union. In contrast, 2 of 11 radii (18%) in the control group showed bone union. Thus, the experimental groups showed significantly higher bone union rates than the control group (p < 0.05). Histological analysis 2 weeks postimplantation in the experimental groups revealed hypertrophic chondrocytes within grafted iMSC pellets and the formation of woven bone around them. This hypertrophic chondrocyte transitioning to newly formed bone suggests that the cartilaginous template can trigger endochondral bone ossification (ECO). Four weeks postimplantation, the cartilage template was reduced in size; newly formed woven bone was predominant in the defect site. New vessels were surrounded by a matrix of woven bone, and hypertrophic chondrocytes transitioning to newly formed bone indicated the progression of ECO. Eight weeks postimplantation, the pellets were completely resorbed and replaced by bone; complete bone union was observed. Dense mature bone developed with evidence of lamellar-like bone formation. Collectively, our results suggest that using iMSC-based cartilage grafts recapitulating the morphogenetic process of ECO in the context of embryonic skeletogenesis is a promising strategy for repairing large bone defects. Impact statement We investigated whether implantation of chondrogenically differentiated iPSC-derived mesenchymal stem cells (iMSCs) could lead to the successful regeneration of bone defects in vivo. We implanted two different clones of human induced pluripotent stem cells into a radial bone defect model. Eleven of 11 (100%) and 7 of 10 (70%) radii in the 201B7 and 454E2 cell-derived pellet transplantation groups, respectively, showed bone union, which were significantly higher than those in the control group [only 2 of 11 radii (18%)]. Overall, our results support the use of iMSC-based cartilage grafts recapitulating the morphogenetic process of endochondral bone ossification for repairing large bone defects.

Posterior Protrusion Measures (PPM) as an Innovative Index in Classifying Plain Lateral Radiograph Images of Pertrochanteric Fracture Using the Revised AO Foundation/Orthopaedic Trauma Association (AO/OTA) Classification.

Introduction The absence of a precise fracture classification system that classifies pertrochanteric fractures into either stable or unstable contributes to a burden on healthcare and has several major implications. We propose an innovative graphical index, which we refer to as posterior protrusion measures (PPM), using plain lateral view radiograph images for the revised AO Foundation (Arbeitsgemeinschaft für Osteosynthesesfragen)/Orthopedic Trauma Association (AO/OTA) classification system. This study aims to: (i) introduce the use of PPM for classifying fractures into stable or unstable under the revised AO/OTA classification system and set the threshold numeric value, (ii) elucidate the reproducibility of inter and intra-observer agreement, and investigate the consistency of fracture classification using PPM versus computed tomography (CT) scan images. Materials and methods Out of 146 patients identified from the database, a total of 126 patients were enrolled in the study. Pertrochanteric fractures were classified as either stable or unstable. Three surgeons were assigned for PPM determination. Regarding the demographical data, the chi-square test was used to assess the significance of each parameter on a categorical scale between the two groups. The independent sample t-test or the Mann-Whitney U test was used to compare the two independent groups. Interclass correlation coefficient (ICC) values for continuous variables and kappa values (κ) for categorical variables were calculated to assess inter-observer and intra-observer agreement. Receiver-operating characteristic (ROC) analysis was used to determine optimal cut-off points of PPM to predict consistency between separate fracture classification groups, one using PPM values with a threshold derived from plain radiograph images, and the other using CT scan images. Results Among a total of 126 pertrochanteric fractures, the A1 (stable) group consisted of 39 patients (10 males, 29 females), whereas the A2 (unstable) group consisted of 87 patients (14 males, 73 females) (not significant, NS). Intraclass correlation coefficient (ICC) values of PPM for the inter-observer agreement were 0.796 (0.723-0.852), 0.664 (0.554-0.751), and 0.702 (0.601-0.781) at first examination and 0.729 (0.635-0.801) at the second. The intra-observer agreement was 0.869 (0.819-0.906) and 0.603 (0.480-0.703). We examined for consistency of fracture classification group of PPM values with a threshold of 0.4 (A1<0.4, A2=0.4 or more) and CT-based group. For the first examination, there was mostly "moderate" agreement in fracture classification (stable or unstable) between plain radiograph and CT scan images, κ (95%CI): 0.427 (0.266-0.588), 0.493 (0.335-0.651), and 0.359 (0.176-0.544), and for the second, 0.418 (0.251-0.585), and 0.451 (0.284-0.620), respectively. Conclusion We propose a supplementary tool, namely PPM that allows for possible alternative classification of pertrochanteric fractures into A1 (stable) and A2 (unstable) using plain radiograph images under the revised AO/OTA classification system. In this study, a PPM threshold value of 0.4 demonstrated a moderate inter- and intra-observer agreement. It is noteworthy to mention that there was a satisfactory consistency of fracture classification using PPM derived from plain radiograph images when compared to classification using CT scan images. In addition, the PPM method provides a numerical score.

Percutaneous CO2 Treatment Accelerates Bone Generation During Distraction Osteogenesis in Rabbits.

BACKGROUND: Distraction osteogenesis has been broadly used to treat various structural bone deformities and defects. However, prolonged healing time remains a major problem. Various approaches including the use of low-intensity pulsed ultrasound, parathyroid hormone, and bone morphogenetic proteins (BMPs) have been studied to shorten the treatment period with limited success. Our previous studies of rats have reported that the transcutaneous application of CO2 accelerates fracture repair and bone-defect healing in rats by promoting angiogenesis, blood flow, and endochondral ossification. This therapy may also accelerate bone generation during distraction osteogenesis, but, to our knowledge, no study investigating CO2 therapy on distraction osteogenesis has been reported. QUESTIONS/PURPOSES: We aimed to investigate the effect of transcutaneous CO2 during distraction osteogenesis in rabbits, which are the most suitable animal as a distraction osteogenesis model for a lengthener in terms of limb size. We asked: Does transcutaneous CO2 during distraction osteogenesis alter (1) radiographic bone density in the distraction gap during healing; (2) callus parameters, including callus bone mineral content, volumetric bone mineral density, and bone volume fraction; (3) the newly formed bone area, cartilage area, and angiogenesis, as well as the expression of interleukin-6 (IL-6), BMP-2, BMP-7, hypoxia-inducible factor (HIF) -1α, and vascular endothelial growth factor (VEGF); and (4) three-point bend biomechanical strength, stiffness, and energy? METHODS: Forty 24-week-old female New Zealand white rabbits were used according to a research protocol approved by our institutional ethical committee. A distraction osteogenesis rabbit tibia model was created as previously described. Briefly, an external lengthener was applied to the right tibia, and a transverse osteotomy was performed at the mid-shaft. The osteotomy stumps were connected by adjusting the fixator to make no gap. After a 7-day latency phase, distraction was continued at 1 mm per day for 10 days. Beginning the day after the osteotomy, a 20-minute transcutaneous application of CO2 on the operated leg using a CO2 absorption-enhancing hydrogel was performed five times per week in the CO2 group (n = 20). Sham treatment with air was administered in the control group (n = 20). Animals were euthanized immediately after the distraction period (n = 10), 2 weeks (n = 10), and 4 weeks (n = 20) after completion of distraction. We performed bone density quantification on the plain radiographs to evaluate consolidation in the distraction gap with image analyzing software. Callus parameters were measured with micro-CT to assess callus microstructure. The newly formed bone area and cartilage area were measured histologically with safranin O/fast green staining to assess the progress of ossification. We also performed immunohistochemical staining of endothelial cells with fluorescein-labeled isolectin B4 and examined capillary density to evaluate angiogenesis. Gene expressions in newly generated callus were analyzed by real-time polymerase chain reaction. Biomechanical strength, stiffness, and energy were determined from a three-point bend test to assess the mechanical strength of the callus. RESULTS: Radiographs showed higher pixel values in the distracted area in the CO2 group than the control group at Week 4 of the consolidation phase (0.98 ± 0.11 [95% confidence interval 0.89 to 1.06] versus 1.19 ± 0.23 [95% CI 1.05 to 1.34]; p = 0.013). Micro-CT demonstrated that bone volume fraction in the CO2 group was higher than that in the control group at Week 4 (5.56 ± 3.21 % [95% CI 4.32 to 6.12 %] versus 11.90 ± 3.33 % [95% CI 9.63 to 14.25 %]; p = 0.035). There were no differences in any other parameters (that is, callus bone mineral content at Weeks 2 and 4; volumetric bone mineral density at Weeks 2 and 4; bone volume fraction at Week 2). At Week 2, rabbits in the CO2 group had a larger cartilage area compared with those in the control group (2.09 ± 1.34 mm [95% CI 1.26 to 2.92 mm] versus 5.10 ± 3.91 mm [95% CI 2.68 to 7.52 mm]; p = 0.011). More newly formed bone was observed in the CO2 group than the control group at Week 4 (68.31 ± 16.32 mm [95% CI 58.19 to 78.44 mm] versus 96.26 ± 19.37 mm [95% CI 84.25 to 108.26 mm]; p < 0.001). There were no differences in any other parameters (cartilage area at Weeks 0 and 4; newly formed bone area at Weeks 0 and 2). Immunohistochemical isolectin B4 staining showed greater capillary densities in rabbits in the CO2 group than the control group in the distraction area at Week 0 and surrounding tissue at Weeks 0 and 2 (distraction area at Week 0, 286.54 ± 61.55 /mm [95% CI 232.58 to 340.49] versus 410.24 ± 55.29 /mm [95% CI 361.78 to 458.71]; p < 0.001; surrounding tissue at Week 0 395.09 ± 68.16/mm [95% CI 335.34 to 454.83] versus 589.75 ± 174.42/mm [95% CI 436.86 to 742.64]; p = 0.003; at Week 2 271.22 ± 169.42 /mm [95% CI 122.71 to 419.73] versus 508.46 ± 49.06/mm [95% CI 465.45 to 551.47]; p < 0.001 respectively). There was no difference in the distraction area at Week 2. The expressions of BMP -2 at Week 2, HIF1-α at Week 2 and VEGF at Week 0 and 2 were greater in the CO2 group than in the control group (BMP -2 at Week 2 3.84 ± 0.83 fold [95% CI 3.11 to 4.58] versus 7.32 ± 1.63 fold [95% CI 5.88 to 8.75]; p < 0.001; HIF1-α at Week 2, 10.49 ± 2.93 fold [95% CI 7.91 to 13.06] versus 20.74 ± 11.01 fold [95% CI 11.09 to 30.40]; p < 0.001; VEGF at Week 0 4.80 ± 1.56 fold [95% CI 3.43 to 6.18] versus 11.36 ± 4.82 fold [95% CI 7.13 to 15.59]; p < 0.001; at Week 2 31.52 ± 8.26 fold [95% CI 24.27 to 38.76] versus 51.05 ± 15.52 fold [95% CI 37.44 to 64.66]; p = 0.034, respectively). There were no differences in any other parameters (BMP-2 at Week 0 and 4; BMP -7 at Weeks 0, 2 and 4; HIF-1α at Weeks 0 and 4; IL-6 at Weeks 0, 2 and 4; VEGF at Week 4). In the biomechanical assessment, ultimate stress and failure energy were greater in the CO2 group than in the control group at Week 4 (ultimate stress 259.96 ± 74.33 N [95% CI 167.66 to 352.25] versus 422.45 ± 99.32 N [95% CI 299.13 to 545.77]; p < 0.001, failure energy 311.32 ± 99.01 Nmm [95% CI 188.37 to 434.25] versus 954.97 ± 484.39 Nmm [95% CI 353.51 to 1556.42]; p = 0.003, respectively). There was no difference in stiffness (216.77 ± 143.39 N/mm [95% CI 38.73 to 394.81] versus 223.68 ± 122.17 N/mm [95% CI 71.99 to 375.37]; p = 0.92). CONCLUSION: Transcutaneous application of CO2 accelerated bone generation in a distraction osteogenesis model of rabbit tibias. As demonstrated in previous studies, CO2 treatment might affect bone regeneration in distraction osteogenesis by promoting angiogenesis, blood flow, and endochondral ossification. CLINICAL RELEVANCE: The use of the transcutaneous application of CO2 may open new possibilities for shortening healing time in patients with distraction osteogenesis. However, a deeper insight into the mechanism of CO2 in the local tissue is required before it can be used in future clinical practice.

Altered microRNA profile during fracture healing in rats with diabetes.

BACKGROUND: MicroRNAs (miRNAs) are a class of small non-coding RNA molecules that regulate gene expression. There is increasing evidence that some miRNAs are involved in the pathology of diabetes mellitus (DM) and its complications. We hypothesized that the functions of certain miRNAs and the changes in their patterns of expression may contribute to the pathogenesis of impaired fractures due to DM. METHODS: In this study, 108 male Sprague-Dawley rats were divided into DM and control groups. DM rats were created by a single intravenous injection of streptozotocin. Closed transverse femoral shaft fractures were created in both groups. On post-fracture days 5, 7, 11, 14, 21, and 28, miRNA was extracted from the newly generated tissue at the fracture site. Microarray analysis was conducted with miRNA samples from each group on post-fracture days 5 and 11. The microarray findings were validated by real-time polymerase chain reaction (PCR) analysis at each time point. RESULTS: Microarray analysis revealed that, on days 5 and 11, 368 and 207 miRNAs, respectively, were upregulated in the DM group, compared with the control group. The top four miRNAs on day 5 were miR-339-3p, miR451-5p, miR-532-5p, and miR-551b-3p. The top four miRNAs on day 11 were miR-221-3p, miR376a-3p, miR-379-3p, and miR-379-5p. Among these miRNAs, miR-221-3p, miR-339-3p, miR-376a-3p, miR-379-5p, and miR-451-5p were validated by real-time PCR analysis. Furthermore, PCR analysis revealed that these five miRNAs were differentially expressed with dynamic expression patterns during fracture healing in the DM group, compared with the control group. CONCLUSIONS: Our findings will aid in understanding the pathology of impaired fracture healing in DM and may support the development of molecular therapies using miRNAs for the treatment of impaired fracture healing in patients with DM.

Topical cutaneous application of CO2 accelerates bone healing in a rat femoral defect model.

BACKGROUND: Bone defects may occur because of severe trauma, nonunion, infection, or tumor resection. However, treatments for bone defects are often difficult and have not been fully established yet. We previously designed an efficient system of topical cutaneous application of carbon dioxide (CO2) using a novel hydrogel, which facilitates CO2 absorption through the skin into the deep area within a limb. In this study, the effect of topical cutaneous application of CO2 on bone healing was investigated using a rat femoral defect model. METHODS: In this basic research study, an in vivo bone defect model, fixed with an external fixator, was created using a rat femur. The affected limb was shaved, and CO2 was applied for 20 min/day, 5 days/week. In the control animals, CO2 gas was replaced with air. Radiographic, histological, biomechanical, and genetic assessments were performed to evaluate bone healing. RESULTS: Radiographically, bone healing rate was significantly higher in the CO2 group than in the control group at 4 weeks (18.2% vs. 72.7%). The degree of bone healing scored using the histopathological Allen grading system was significantly higher in the CO2 group than in the control group at 2 weeks (1.389 ± 0.334 vs. 1.944 ± 0.375). The ultimate stress, extrinsic stiffness, and failure energy were significantly greater in the CO2 group than in the control group at 4 weeks (3.2 ± 0.8% vs. 38.1 ± 4.8%, 0.6 ± 0.3% vs. 41.5 ± 12.2%, 2.6 ± 0.8% vs. 24.7 ± 5.9%, respectively.). The volumetric bone mineral density of the callus in micro-computed tomography analysis was significantly higher in the CO2 group than in the control group at 4 weeks (180.9 ± 43.0 mg/cm3 vs. 247.9 ± 49.9 mg/cm3). Gene expression of vascular endothelial growth factor in the CO2 group was significantly greater than that in the control group at 3 weeks (0.617 ± 0.240 vs. 2.213 ± 0.387). CONCLUSIONS: Topical cutaneous application of CO2 accelerated bone healing in a rat femoral defect model. CO2 application can be a novel and useful therapy for accelerating bone healing in bone defects; further research on its efficacy in humans is warranted.

Triweekly administration of parathyroid hormone (1-34) accelerates bone healing in a rat refractory fracture model.

BACKGROUND: Some reports have shown that intermittent parathyroid hormone (PTH) (1-34) treatment for patients with delayed union or nonunion have led to successful healing. In this study, we investigated whether systemic intermittent administration of PTH (1-34) has a beneficial effect on bone healing in a rat refractory fracture model. METHODS: We created a refractory femoral fracture model in 32 rats with periosteal cauterization that leads to atrophic nonunion at 8 weeks after surgery. Half the rats received subcutaneous intermittent human PTH (1-34) injections at a dosage of 100 µg/kg, thrice a week for 8 weeks. The other half received the vehicle only. At 8 weeks after fracture, radiographic, histological and mechanical assessments were performed. RESULTS: Radiographic assessments showed that the union rate was significantly higher in the PTH group than in the control group (P < 0.05). The degree of fracture repair as scored using the Allen grading system in histological assessment was significantly greater in the PTH group than in the control group (P < 0.05). The ultimate stress and stiffness measurements were significantly greater in the PTH group than in the control group (p < 0.05). CONCLUSIONS: We demonstrated that triweekly administration of PTH (1-34) increased union rate and accelerated bone healing in a rat refractory fracture model, suggesting that systemic administration of PTH (1-34) could become a novel and useful therapy for accelerating fracture healing in patients at high risk of delayed union or nonunion.

Altered expression of SDF-1 and CXCR4 during fracture healing in diabetes mellitus.

PURPOSE: Diabetes mellitus (DM) is known to impair fracture healing. The purpose of this study was to elucidate and compare the gene expression patterns and localization of stromal cell-derived factor 1 (SDF-1) and CXC chemokine receptor 4 (CXCR4) during fracture healing of the femur in rats with and without DM. METHODS: Closed transverse fractures were created in the femurs of rats equally divided into a DM group and control group; DM was induced by streptozotocin. At post-fracture days five, seven, 11, 14, 21 and 28, total RNA was extracted from the fracture callus and mRNA expression levels of SDF-1 and CXCR4 were measured by real-time polymerase chain reaction. Localization of SDF-1 and CXCR4 proteins at the fracture site was determined by immunohistochemistry at days 21 and 28. RESULTS: SDF-1 expression was significantly lower in the DM group than in the healthy group on days 21 and 28, and showed a significant difference between days 14 and 21 in the healthy group. There was no significant difference in CXCR4 expression levels between the healthy and DM groups at any time point. On day 21 immunoreactivity of SDF-1 and CXCR4 was detected at the fracture site of the healthy group but no immunoreactivity was observed in the DM group. On day 28, immunoreactivity of SDF-1 and CXCR4 was detected at the fracture site in both groups. CONCLUSION: Gene expression and localization of SDF-1 and CXCR4 was altered during fracture healing, which may contribute to the impaired fracture healing in DM.