THE INTERNATIONAL JOURNAL OF MEDICAL ROBOTICS AND COMPUTER ASSISTED SURGERY ORIGINAL Int J Med Robotics Comput Assist Surg 2015; 11: 109–119. Published online 22 August 2014 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/rcs.1604
ARTICLE
Novel computer-assisted preoperative planning system for humeral shaft fractures: report of 43 cases
Yanxi Chen* Minfei Qiang Kun Zhang Haobo Li Hao Dai Department of Orthopaedic Trauma, East Hospital, Tongji University School of Medicine, Shanghai, China *Correspondence to: Yanxi Chen, MD, PhD, Department of Orthopaedic Trauma, East Hospital, Tongji University School of Medicine, 150 Jimo Road, 200120 Shanghai, China. Email: [email protected]
Abstract Background Treatment for humeral shaft fractures is restoration of a painless shoulder and elbow with satisfactory function, but achieving good clinical outcomes for fracture healing is difficult, particularly in elderly patients. The aim of the study was to determine if computer-assisted preoperative planning (CAPP) improves clinical outcomes of humeral shaft fractures. Methods In total, 43 patients were treated for humeral shaft fractures using locking plates with CAPP. The total time needed and total cost for fracture fragments’ virtual segmentation/reduction/fixation were recorded. Intra- and inter-observer reliability was analyzed with intraclass correlation coefficients (ICCs). Clinical function was analyzed with Constant Score, Mayo Elbow Performance Score (MEPS), visual analogue scale (VAS) for pain, short-form health survey (SF-36), and radiology. Results Mean total CAPP time for 12-A, 12-B, and 12-C fractures were 12.78 ±1.19, 22.07±1.12, and 38.56±2.11 min, respectively. Observer reliability was high (ICC 0.766–0.995). Mean operation time was 76.8±9.2 min. Follow-up (39/43 cases) averaged 36.5 months. Mean Constant Score and MEPS were 85.2±8.1 and 95.7±3.2, respectively. Average VAS was 1.3 points. Mean physical and mental component summary SF-36 scores were 74.3±5.1 and 76.9±5.9, respectively. Two patients had delayed union. Conclusions The novel CAPP system was efficient and reliable, providing excellent clinical outcomes for treatment of humeral shaft fractures using locking plates. Copyright © 2014 John Wiley & Sons, Ltd. Keywords humeral shaft fractures; computer-aided design; preoperative planning; 3D imaging
Introduction
Accepted: 7 July 2014
Copyright © 2014 John Wiley & Sons, Ltd.
Fractures of the humeral shaft account for 1–3% of all fractures and approximately 20% of all humeral fractures (1). The mid-shaft is fractured in 43.2%, the proximal end in 40.8%, and the distal end in 16% (2). For the past few years, the incidence of humeral shaft fractures seen in clinical practice among aging populations has been increasing. The goal of treatment for humeral shaft fractures is restoration of a painless shoulder and elbow that
Y. Chen et al.
110
have satisfactory function. A number of techniques have been described for fixation of humeral shaft fractures, such as plates and screws, intramedullary (IM) devices, and external fixation (3–5). Despite the numerous surgical techniques available, plate fixation remains the gold standard for humeral shaft fractures (6). Currently, combined open reduction and internal fixation (ORIF) of a plate has become a widely accepted treatment. However, ORIF is associated with extensive soft tissue dissection and bears the risk of iatrogenic injury to the radial nerve (7–10). Also, minimally invasive plate osteosynthesis (MIPO) with locking plate systems has become practicable, combining the benefits of plate fixation and biological osteosynthesis and allowing early function after treatment (11–15). Minimally invasive surgical procedures minimize local soft tissue trauma, reduce postoperative pain, create fewer periarticular adhesions, and result in better joint function. However, MIPO techniques are generally considered difficult to perform, and the reduction is often less than perfect. The most common complications associated with MIPO techniques include malunion, iatrogenic radial nerve injury (IRNI), infection, and fixation material problems (6,11–17). Therefore, achieving a good clinical outcome with fracture healing is difficult in patients with humeral shaft fractures, particularly elderly patients. Preoperative planning has an important role in orthopedic surgery. Accurate, precise preoperative planning can provide information instrumental in performing less invasive articular fracture surgery (18). Over the years, many scholars have tried numerous methods of preoperative planning based on radiology and computer technology. They have reported favorable results in limb, pelvic, and acetabular fracture reduction and osteosynthesis (19–25). However, these techniques applied to orthopedic trauma had poor repeatability and took more time because the process was complex, requiring several sets of software and too much manual manipulation. Fornaro et al. (26) tested the feasibility of preoperative surgical planning for acetabular fractures, including fracture reduction, fixation, and measurement. The median time consumed was 130 minutes. Suero et al. (27) reported using virtual three-dimensional (3D) software for planning tibial plateau fracture reconstruction. The methods included segmentation, which had to be performed manually by the operator in all slices in all three planes. Also, type C fractures required close to 4 h of planning. Hence, the time requirement and reliability may be the current deterrents to using computer-assisted preoperative planning (CAPP) in the clinical setting (28). The aims of this study were to: (1) determine the time needed for a novel CAPP system for humeral shaft fractures; (2) assess repeatability using this virtual surgical technique; and (3) determine the clinical outcomes Copyright © 2014 John Wiley & Sons, Ltd.
after treating humeral shaft fractures using a locking plate based on CAPP.
Methods Study population This study was based on a series of 58 consecutive patients with 60 humeral shaft fractures who were operated on from February 2009 to March 2012. According to inclusion and exclusion criteria (Table 1), the study excluded 15 patients with 17 humeral shaft fractures. The final study group then comprised 43 patients with humeral shaft fractures who were to be treated using locking plates with CAPP. There were 28 women and 15 men, with an average age of 71.5 years (range 52–88 years). The left side was involved in 12 cases and the right side in 31 cases. In all, 13 patients had a high-energy injury, and 30 had a low-energy injury. The AO/OTA fracture classification systems (29) were used, and the classifications were confirmed intraoperatively. Six patients had 12-A, 12 patients had 12-B, and 25 patients had 12-C fractures.
Radiology technique Anteroposterior (AP) and axillary view radiographs for humeral shaft fractures were obtained. Computed tomography (CT) scans from the proximal humerus to the elbow were performed using a 16-detector spiral CT scanner (GE LightSpeed CT; GE Healthcare, Milwaukee, WI, USA). CT imaging parameters for plain scanning images were as follows: section thickness 0.625–1.25 mm; tube voltage 120 kV; pitch 1.375; matrix 512×512. All of the data were reviewed retrospectively by a musculoskeletal radiologist Table 1. Study criteria Inclusion criteria 1. Closed humeral shaft fractures that were displaced 12-A, 12-B, or 12-C according to the AO/OTA classification systems 2. Fracture less than 3 weeks old 3. Without neurologic injuries 4. Age ≥ 50 yr 5. Single side injured 6. Informed consent before preoperative planning Exclusion criteria Undisplaced fractures Multiple trauma Severe dementia (unable to follow postoperative recommendations) Pathological fractures Open fractures Ipsilateral fractures of the forearm
Int J Med Robotics Comput Assist Surg 2015; 11: 109–119. DOI: 10.1002/rcs
111
Computer-assisted preoperative planning for humeral shaft fractures
and a trauma surgeon and were saved in Digital Imaging and Communication in Medicine (DICOM 3.0) format (.dcm).
Computer-assisted preoperative planning Thin-slice CT axial images of all subjects were input into the CAPP system (SuperImage Orthopedics Edition 1.0; Cybermed Ltd, Shanghai, China) (30,31). The software was developed using Java language on NetBeans (Sun Microsystems, Inc., Santa Clara, CA) and OpenInventor (Mercury Computer Systems/TGS Unit, San Diego, CA) platforms. Two-dimensional (2D) and 3D images of the fracture zone of the humeral shaft were reconstructed by multiple planar reconstruction (MPR) and volume rendering technology (VRT), respectively. For analyzing injury details, the CAPP steps were as follows (Figure 1). I. Fracture fragments segmentation: 3D images of the humeral shaft fractures were reconstructed by a surface shaded display (SSD) algorithm with a reconstruction interval of 0.625–1.25 mm (31–33). The density threshold was 150H, and automatic removal of the image size was
ARTICLE
Novel computer-assisted preoperative planning system for humeral shaft fractures: report of 43 cases
Yanxi Chen* Minfei Qiang Kun Zhang Haobo Li Hao Dai Department of Orthopaedic Trauma, East Hospital, Tongji University School of Medicine, Shanghai, China *Correspondence to: Yanxi Chen, MD, PhD, Department of Orthopaedic Trauma, East Hospital, Tongji University School of Medicine, 150 Jimo Road, 200120 Shanghai, China. Email: [email protected]
Abstract Background Treatment for humeral shaft fractures is restoration of a painless shoulder and elbow with satisfactory function, but achieving good clinical outcomes for fracture healing is difficult, particularly in elderly patients. The aim of the study was to determine if computer-assisted preoperative planning (CAPP) improves clinical outcomes of humeral shaft fractures. Methods In total, 43 patients were treated for humeral shaft fractures using locking plates with CAPP. The total time needed and total cost for fracture fragments’ virtual segmentation/reduction/fixation were recorded. Intra- and inter-observer reliability was analyzed with intraclass correlation coefficients (ICCs). Clinical function was analyzed with Constant Score, Mayo Elbow Performance Score (MEPS), visual analogue scale (VAS) for pain, short-form health survey (SF-36), and radiology. Results Mean total CAPP time for 12-A, 12-B, and 12-C fractures were 12.78 ±1.19, 22.07±1.12, and 38.56±2.11 min, respectively. Observer reliability was high (ICC 0.766–0.995). Mean operation time was 76.8±9.2 min. Follow-up (39/43 cases) averaged 36.5 months. Mean Constant Score and MEPS were 85.2±8.1 and 95.7±3.2, respectively. Average VAS was 1.3 points. Mean physical and mental component summary SF-36 scores were 74.3±5.1 and 76.9±5.9, respectively. Two patients had delayed union. Conclusions The novel CAPP system was efficient and reliable, providing excellent clinical outcomes for treatment of humeral shaft fractures using locking plates. Copyright © 2014 John Wiley & Sons, Ltd. Keywords humeral shaft fractures; computer-aided design; preoperative planning; 3D imaging
Introduction
Accepted: 7 July 2014
Copyright © 2014 John Wiley & Sons, Ltd.
Fractures of the humeral shaft account for 1–3% of all fractures and approximately 20% of all humeral fractures (1). The mid-shaft is fractured in 43.2%, the proximal end in 40.8%, and the distal end in 16% (2). For the past few years, the incidence of humeral shaft fractures seen in clinical practice among aging populations has been increasing. The goal of treatment for humeral shaft fractures is restoration of a painless shoulder and elbow that
Y. Chen et al.
110
have satisfactory function. A number of techniques have been described for fixation of humeral shaft fractures, such as plates and screws, intramedullary (IM) devices, and external fixation (3–5). Despite the numerous surgical techniques available, plate fixation remains the gold standard for humeral shaft fractures (6). Currently, combined open reduction and internal fixation (ORIF) of a plate has become a widely accepted treatment. However, ORIF is associated with extensive soft tissue dissection and bears the risk of iatrogenic injury to the radial nerve (7–10). Also, minimally invasive plate osteosynthesis (MIPO) with locking plate systems has become practicable, combining the benefits of plate fixation and biological osteosynthesis and allowing early function after treatment (11–15). Minimally invasive surgical procedures minimize local soft tissue trauma, reduce postoperative pain, create fewer periarticular adhesions, and result in better joint function. However, MIPO techniques are generally considered difficult to perform, and the reduction is often less than perfect. The most common complications associated with MIPO techniques include malunion, iatrogenic radial nerve injury (IRNI), infection, and fixation material problems (6,11–17). Therefore, achieving a good clinical outcome with fracture healing is difficult in patients with humeral shaft fractures, particularly elderly patients. Preoperative planning has an important role in orthopedic surgery. Accurate, precise preoperative planning can provide information instrumental in performing less invasive articular fracture surgery (18). Over the years, many scholars have tried numerous methods of preoperative planning based on radiology and computer technology. They have reported favorable results in limb, pelvic, and acetabular fracture reduction and osteosynthesis (19–25). However, these techniques applied to orthopedic trauma had poor repeatability and took more time because the process was complex, requiring several sets of software and too much manual manipulation. Fornaro et al. (26) tested the feasibility of preoperative surgical planning for acetabular fractures, including fracture reduction, fixation, and measurement. The median time consumed was 130 minutes. Suero et al. (27) reported using virtual three-dimensional (3D) software for planning tibial plateau fracture reconstruction. The methods included segmentation, which had to be performed manually by the operator in all slices in all three planes. Also, type C fractures required close to 4 h of planning. Hence, the time requirement and reliability may be the current deterrents to using computer-assisted preoperative planning (CAPP) in the clinical setting (28). The aims of this study were to: (1) determine the time needed for a novel CAPP system for humeral shaft fractures; (2) assess repeatability using this virtual surgical technique; and (3) determine the clinical outcomes Copyright © 2014 John Wiley & Sons, Ltd.
after treating humeral shaft fractures using a locking plate based on CAPP.
Methods Study population This study was based on a series of 58 consecutive patients with 60 humeral shaft fractures who were operated on from February 2009 to March 2012. According to inclusion and exclusion criteria (Table 1), the study excluded 15 patients with 17 humeral shaft fractures. The final study group then comprised 43 patients with humeral shaft fractures who were to be treated using locking plates with CAPP. There were 28 women and 15 men, with an average age of 71.5 years (range 52–88 years). The left side was involved in 12 cases and the right side in 31 cases. In all, 13 patients had a high-energy injury, and 30 had a low-energy injury. The AO/OTA fracture classification systems (29) were used, and the classifications were confirmed intraoperatively. Six patients had 12-A, 12 patients had 12-B, and 25 patients had 12-C fractures.
Radiology technique Anteroposterior (AP) and axillary view radiographs for humeral shaft fractures were obtained. Computed tomography (CT) scans from the proximal humerus to the elbow were performed using a 16-detector spiral CT scanner (GE LightSpeed CT; GE Healthcare, Milwaukee, WI, USA). CT imaging parameters for plain scanning images were as follows: section thickness 0.625–1.25 mm; tube voltage 120 kV; pitch 1.375; matrix 512×512. All of the data were reviewed retrospectively by a musculoskeletal radiologist Table 1. Study criteria Inclusion criteria 1. Closed humeral shaft fractures that were displaced 12-A, 12-B, or 12-C according to the AO/OTA classification systems 2. Fracture less than 3 weeks old 3. Without neurologic injuries 4. Age ≥ 50 yr 5. Single side injured 6. Informed consent before preoperative planning Exclusion criteria Undisplaced fractures Multiple trauma Severe dementia (unable to follow postoperative recommendations) Pathological fractures Open fractures Ipsilateral fractures of the forearm
Int J Med Robotics Comput Assist Surg 2015; 11: 109–119. DOI: 10.1002/rcs
111
Computer-assisted preoperative planning for humeral shaft fractures
and a trauma surgeon and were saved in Digital Imaging and Communication in Medicine (DICOM 3.0) format (.dcm).
Computer-assisted preoperative planning Thin-slice CT axial images of all subjects were input into the CAPP system (SuperImage Orthopedics Edition 1.0; Cybermed Ltd, Shanghai, China) (30,31). The software was developed using Java language on NetBeans (Sun Microsystems, Inc., Santa Clara, CA) and OpenInventor (Mercury Computer Systems/TGS Unit, San Diego, CA) platforms. Two-dimensional (2D) and 3D images of the fracture zone of the humeral shaft were reconstructed by multiple planar reconstruction (MPR) and volume rendering technology (VRT), respectively. For analyzing injury details, the CAPP steps were as follows (Figure 1). I. Fracture fragments segmentation: 3D images of the humeral shaft fractures were reconstructed by a surface shaded display (SSD) algorithm with a reconstruction interval of 0.625–1.25 mm (31–33). The density threshold was 150H, and automatic removal of the image size was
