Aims: The purpose of the study is to analyze the most frequent errors in the densitometric diagnosis of osteoporosis and to evaluate their influence on the final results of examination. Methods: On the basis of their own experience, the authors have presented here the most common errors encountered in the densitometric technique. A DPX-L densitometric apparatus (Lunar) was used in this research. Errors are divided into three groups: those dependent on the object investigated, those dependent on data analysis, and others. In the first group we took into account factors which can lead to either overestimating (+) or underestimating (−) the final result: degenerative changes (+), scoliosis (+), foreign bodies, such as metal (−), status post fracture (+), pathological structures (±), osteoporotic fractures (+), and incorrect arrangement of the investigated object (±). Errors in data analysis included erroneous data entry regarding the patient’s age, height, body mass and sex, and incorrect settings regarding the measurement field. Other errors included failure to calibrate on improper calibration of the measurement apparatus and errors in computer programs. Results: Individual errors were responsible for falsifying results from 1 to 37%. If several of these errors are accumulated, the accuracy of examination may change even more than 100%. Densitometric examinations of the lumbar spine are the most subject to error. The repeatability of the results generated by the DEXA apparatus ranges for particular skeletal regions from 0.9 to 2.5% of the CV, while the precision ranges from 3 to 5%. We also cannot exclude the impact of errors of the three types listed above on the final results. Conclusions: We are convinced that the DEXA method is an excellent instrument for the diagnosis of osteoporosis (in static bone evaluation). However; its valve diminishes in monitoring dynamic changes in bone tissue, even at 1–2-year intervals

The minimisation of errors incurred during the learning process is thought to enhance motor learning and improve performance under pressure or in multitasking situations. If this is proven in surgical skills learning, it has the potential to enhance the delivery of surgical education. We aimed to compare errorless and errorful learning using the high-speed burr.

Medical students (n=30) were recruited and allocated randomly to an errorless or errorful group. The errorless learning group progressively learnt tasks from easy to difficult on cedar boards simulating bone. The errorful learning group also progressed through the same tasks but not in order of difficulty.

Transfer tasks assessed students’ performance of cervical laminoplasty on saw bone models to assess their level of learning from previous stages. During transfer task 2, students completed the procedure under time pressure and in the presence of distractors, in order to simulate real-life stressors in theatre. Accuracy, precision and safety of the procedure were scored by expert opinions from spine surgeons blinded to the grouping of the participants.

Both errorless and errorful learners demonstrated improvements in performance with increasing amounts of practice (demonstrated by the decreased time taken for the task as well as improvement in accuracy of the cuts (depth, width and smoothness). The performance of both groups was not impaired by the incorporation of a secondary task which required participants to multitask. No statistically significant difference in performance was noted between the two groups.

In contrast to previous research, there was no significant difference between errorless or errorful learning to develop skills with a high-speed, side-cutting burr. In both groups, practical learning during the session has led to improvement in overall performance with the burr relevant to cervical laminoplasty.

Currently implemented accuracy metrics in open-source libraries for segmentation by supervised machine learning are typically one-dimensional scores [1]. While extremely relevant to evaluate applicability in clinics, anatomical location of segmentation errors is often neglected.

This study aims to include the three-dimensional (3D) spatial information in the development of a novel framework for segmentation accuracy evaluation and comparison between different methods.

Predicted and ground truth (manually segmented) segmentation masks are meshed into 3D surfaces. A template mesh of the same anatomical structure is then registered to all ground truth 3D surfaces. This ensures all surface points on the ground truth meshes to be in the same anatomically homologous order. Next, point-wise surface deviations between the registered ground truth mesh and the meshed segmentation prediction are calculated and allow for color plotting of point-wise descriptive statistics. Statistical parametric mapping includes point-wise false discovery rate (FDR) adjusted p-values (also referred to as q-values).

The framework reads volumetric image data containing the segmentation masks of both ground truth and segmentation prediction. 3D color plots containing descriptive statistics (mean absolute value, maximal value,…) on point-wise segmentation errors are rendered. As an example, we compared segmentation results of nnUNet [2], UNet++ [3] and UNETR [4] by visualizing the mean absolute error (surface deviation from ground truth) as a color plot on the 3D model of bone and cartilage of the mean distal femur.

A novel framework to evaluate segmentation accuracy is presented. Output includes anatomical information on the segmentation errors, as well as point-wise comparative statistics on different segmentation algorithms. Clearly, this allows for a better informed decision-making process when selecting the best algorithm for a specific clinical application.

Abstract

Wrong-level surgery is a unique pitfall in spinal surgery and is part of the wider field of wrong-site surgery. Wrong-site surgery affects both patients and surgeons and has received much media attention. We performed this systematic review to determine the incidence and prevalence of wrong-level procedures in spinal surgery and to identify effective prevention strategies. We retrieved 12 studies reporting the incidence or prevalence of wrong-site surgery and that provided information about prevention strategies. Of these, ten studies were performed on patients undergoing lumbar spine surgery and two on patients undergoing lumbar, thoracic or cervical spine procedures. A higher frequency of wrong-level surgery in lumbar procedures than in cervical procedures was found. Only one study assessed preventative strategies for wrong-site surgery, demonstrating that current site-verification protocols did not prevent about one-third of the cases. The current literature does not provide a definitive estimate of the occurrence of wrong-site spinal surgery, and there is no published evidence to support the effectiveness of site-verification protocols. Further prevention strategies need to be developed to reduce the risk of wrong-site surgery.

We have investigated the errors in the identification of the transepicondylar axis and the anteroposterior axis between a minimally-invasive and a conventional approach in four fresh-frozen cadaver knees. The errors in aligning the femoral prosthesis were compared with the reference transepicondylar axis as established by CT.

The error in the identification of the transepicondylar axis was significantly higher in the minimal approach (4.5° of internal rotation, sd 4) than in the conventional approach (3° of internal rotation, sd 4; p < 0.001). The errors in identifying the anteroposterior axis in the two approaches were 0° (sd 5) and 1.8° (sd 5) of internal rotation, respectively (p < 0.001).

Conventional marker based optical motion capture (mocap) methods for estimating the position and orientation (pose) of anatomical segments use assumptions that anatomical segments are rigid bodies and the position of tracking markers is invariant relative to bones. Soft tissue artefact (STA) is the error in pose estimation due to markers secured to soft tissue that moves relative to bones. STA is a major source of pose estimation error and is most prevalent when markers are placed over joints. Mocap and bi-plane videoradiography data were recorded synchronously while three individuals walked on a treadmill. For all three, pose of the thigh and shank, and movement of markers relative to the bones, were determined from the videoradiography data (DSX, C-Motion). Independently, pose of thighs and shanks was estimated using mocap data (Visual3D, C-Motion). Our measures of error in the mocap pose estimation were the relative thigh and shank translations. X-ray data from two subjects were used to generate a regression model for the antero/posterior movement of the lateral knee marker against internal/external hip rotation. The mocap translation errors of the third subject, attributed to STA of the knee marker, were 15.6mm and 32.0mm respectively. The pose of the third subject was then estimated using a probabilistic algorithm incorporating our regression model. Mocap translation errors were reduced to 10.6mm (thigh) and 4.4mm (shank). The results from these data suggest that errors in pose estimation due to STA may possibly be reduced via the application of algorithms based on probabilistic inference to mocap data.

The treatment of orthopedic implant infections is often difficult and complex, although the chances of successful treatment with a properly selected diagnostic, surgical and antibiotic treatment protocol have recently increased significantly. Surgical treatment is a key factor in the treatment of infections of orthopedic implants, and any errors in this respect often lead to worse clinical outcomes.

Surgical errors. The most important and frequent surgical errors include:

- conservative treatment of periprosthetic infections with antibiotics alone: successful treatment requires adequate surgical procedure combined with long-term antimicrobial Th that is active against biofilm microorganism. Without adequate surgical procedure just the suppression of symptoms is usually achieved, rather than eradication of the infection.

We present illustrative cases with each common surcical error combined with proper solution.

Treatment of PJI is a demanding procedure, the goal is a long-term pain-free functional joint, that can be achieved by eradication of the infection. For a successful clinical outcome an appropriate diagnostic, surgical and antimicrobial procedure for the individual patient has to be selected.

1. In the course of 2,668 operations for suspected cartilage lesions 119 errors of diagnosis were revealed.

2. These errors are analysed and certain conclusions are drawn.

Purpose

Femoral component malrotation is a common cause for persisting symptoms and revision following total knee arthroplasty (TKA). There is ongoing debate about the most appropriate use of femoral landmarks to determine rotation. The Sulcus Line (SL, See Figure 1) is a three-dimensional curve produced from multiple points along the trochlear groove. Whiteside's Line, also known as the anteroposterior axis (APA), is derived from single anterior and posterior points. The purposes of the three studies presented are to i) assess the SL in a large clinical series, ii) demonstrate the effect of parallax error on rotational landmarks, and iii) assess the accuracy of a device which transfers a geometrically corrected SL onto the distal cut surface of the femur.

Achieving precise component alignment of total knee arthroplasty produces good clinical outcome. However, the cutting errors between planed and final bone resection planes during the procedure of total knee arthroplasty were less evaluated. The aim of this study was to evaluate the cutting errors during total knee arthroplasty using the navigation system.

In a prospective series of 60 total knee replacements with image-free navigation system, the planed resection plane and final resection plane in frontal and sagittal planes were evaluated. The cutting errors standard deviations ranged from 1.01° to 1.21° in final frontal femoral and tibia plane and 1.23° in final sagittal femoral and tibia plane. The cutting errors showed only significant difference in the sagittal plane of femoral resection and only 9 cuts (4%) 3 of all plane and the maximal error was 4 in only 2 cases (0.8%).

Our results support to use the navigation system to adjust the cutting block and correct the cutting errors. This would lead to a more precise cut and result in better leg alignment and component orientation than the conventional TKR technique.

Introduction The effect of hip rotation on the measurement of femoral offset is determined firstly using artificial bones in an anatomical study and then in a patient population. Its effect on the choice of femoral component in total hip arthroplasty is discussed

Methods Radiographs were taken of a series of saw bone models rotated through a range of angles. The resultant offset was then measured Standardised and Control (unstandardised) radiographs of the pelvis were taken of patients presenting to orthopaedic outpatients. Femoral offset was measured from each radiograph

Results In the anatomical study angles of rotation differed significantly with respect to measurement of offset (p< 0.0001 Friedman 2-way analysis of variance by ranks). The greatest measurement of offset was at 15 degrees internal rotation. Offset decreased with external rotation. The clinical study had power of 80%. Femoral offset was increased in all the standardised x-rays compared with their controls (n=108, mean=7.64, SD=5.55, 95% CI (6.58,8.70)). A one-sample t-test was performed to see if the standardised and control films were greater than 5mm different (t=14.30 (107df), p< 0.01).

Conclusions The clinical study confirmed the findings of the anatomical study. A standardised AP radiograph of the pelvis improves the measurement of femoral offset. For surgeons using the Exeter hip system failure to account for offset could lead to the selection of a stem two sizes too small with regards to offset. Lesser degrees of rotation, not readily identified by looking at the radiograph, could still lead to the selection of an incorrectly sized stem. Offset has been shown to increase the range of movement, abductor strength and stability of the hip joint whilst decreasing the rate of wear. It therefore benefits patients to account for offset, ensuring a correctly sized hip replacement.

Summary Statement

The current study introduced the effects of projection errors on ankle morphological measurements using CT-based simulated radiographs by correlation analysis between 2D/3D dimensions and reliability analysis with randomised perturbations while measuring planar parameters on radiographs.

Heterotopic ossification (HO) is a frequent pathological phenomenon after total hip arthroplasty. Incidence of HO after the total hip arthroplasty is in average 43%. Most often classification from Brooker and co-authors has been used to assess the HO. Overall 47% of all studies published until the 1999 have used Brooker’s system, but there has been reported a “fair” reproducibility (Cohen’s kappa 0.5) of this system.

The aim of the present study was to estimate the influence of used classification system to the evidence of HO, to assess the reliability of the more often used classification systems, to determine the sources of errors in the assessment-process of HO and, to compose a new classification system with higher reliability.

Four investigators assessed HO in 111 patients applying most often used classification systems and the new system. Six investigators measured dimensionality of HO in 28 patients applying method of digital planimetry. Kappa statistics of all the compared classification systems were calculated. Main sources of errors were detected by dispersion model.

Average evidence of HO differed up to 1.4 times if the different systems were applied to assess HO. Therefore, using literature data, always the criteria of a system used should be taken into consideration.

Most important source of errors in the HO assessment was the error of diagnosing process. This source consisted of two components: inter-observer variation that formed 25.5% (+/−8.0%; p=0.0015) of total error and intra-observer variation that formed 60.9% (+/−7.3%; p< 0.0001). Technical performing error had less contribution in total error, namely 8,0% (+/−0.6%; p< 0.0001) and subspecialisation of the investigators did not cause any systematic bias having a proportion of 5.7% (+/−4.9%; p=0.2457).

Reliability of the Brooker’s system was lowest among the compared systems. Combining the Della Valle’s system which had high reliability with the Brooker’s system, we composed a new classification preserving high reliability. The results obtained using the new classification allow comparison with the results obtained using each of the parent classifications.

Background

There has been widespread interest in medical errors since the publication of ‘To Err is Human’ by the Institute of Medicine in 1999. The Patient Safety Committee of the American Academy of Orthopaedic Surgeons has compiled results of a member survey to identify trends in orthopaedic errors that would help direct quality assurance efforts.

This study investigated the intra-observer errors in obtaining visually selected anatomic landmarks that were used in registration process in a non-image based computer assisted TKR system.

The landmarks studied were centre of distal femur, medial and lateral femoral epicondyle, centre of proximal tibia, medial malleolus and lateral malleolus. Repeated registration in the above sequence was done for one hundred times by one single surgeon.

The maximum combined errors in the mechanical axis of the lower limb were only 1.32 degrees (varus/valgus) in the coronal plane and 4.17 degrees (flexion/extension) in the sagittal plane. The maximum error in transepicondylar axis was 8.2 degrees.

The errors using the visual selection of anatomic landmarks for the registration technique of bony landmarks in non-image based navigated TKR did not introduce significant error in the mechanical axis of the lower limb in the coronal plane. However, the error in the transepicondylar axis was significant in the “worst case scenario”.

Aim: To identify the incidence of regular medication not being prescribed on initial admission of emergency surgical cases

Material and Method: The data for this audit was taken from all surgical admissions over an approximately one month period between December 2004 and January 2005. Prescribed drugs for each patient were checked the morning after admission. If the patient was unable to provide this information, we obtained their regular medication list from the GP.

A total of 71 admissions were studied, 58% were referred from A& E, whereas 34% from the GP or primary care and a small percentage came from referrals from other wards within the hospital.

Results: Out of the 71 admissions, 46 patients were receiving at least one regular medication. Obviously, some patients were on medications for more than one disease and these were looked at individually.

From all 71 admissions, 21% of them had at least one error, i.e. at least one regular medication was not prescribed on admission.

If there was a documented reason for the omission of a particular drug then this was not counted as an error.

Analysing each co morbidity individually, 42% of IHD medication were not prescribed despite being taken on a regular basis by the patient, 33% for hypertensive and diabetic medication, 50% for asthmatic and psychiatric medication and 29% for medication for other less serious conditions.

81% of the errors made were on patients referred from A& E, while 15% where from patients received from the GP/primary care. Only 4% of the errors was made on patients referred from other wards. However, A& E referrals were almost double those of GPs. Hence, in a total of 41 A& E referrals 21 errors were made, while in a total of 24 GP referrals only 4 errors were made. Only 1 error was made in the total of 6 ward referrals.

Conclusion and Discussion: The results of this audit are surprising and alarming. 21% of admissions had at least one regular medication not prescribed by the admitting doctors.

Missing out on regular medication can have potentially life-threatening effects on patients as well as severe medico-legal implications.

Most of the mistakes were being made with patients that are referred from the accident and emergency department. These patients are generally more unwell than the ones referred from the GP or primary care, and quite often are elderly patients on a multitude of drugs that are unable to remember some or all of their tablets. Patients admitted out of hours present an added difficulty in that GPs are not available for confirmation of the patient’s regular medication.

So, more care and emphasis need to be given on drug history when admitting a patient.

Background: Procedures performed at the incorrect anatomical site are perceived as rare events. However, they can be devastating for patients and doctors. In 2007, the Chief Medical Officer for England highlighted surgical site errors as an area of concern. Evidence from the UK and North America suggests that these events are more common than we think. We present our findings on the rates of surgical errors in England and Wales.

Methods: We obtained information from the National Patient Safety Agency (NPSA) and the NHS Litigation Authority (NHSLA) detailing reported cases of surgical error in England and Wales in the time period 1995–2007. This information was evaluated in conjunction with data detailing the total number of operations performed in England and Wales for the same time periods.

Results: Since 1995 there have been 292 cases of errors recorded on the NHSLA database. Orthopaedic surgery was the worst offending speciality with 87 cases (29.8%). There were 49 cases (16.8%) in dental surgery and 40 cases (13.7%) in general surgery. The most commonly affected sub-specialty area of orthopaedic surgery was the hand (21 cases), followed by foot and ankle (18 cases) and spine (16 cases). The total cost of the cases which have been settled was £2,252,752.58. For the year 2006–2007 the risk of an event in orthopaedic surgery was calculated to be 1 in 105,712 cases. The risk in general surgery and obstetrics and gynaecology during this period was 1 in 353,511 and 1 in 133,371 respectively.

Discussion: Orthopaedic procedures featured more commonly on the NHSLA database than any other specialty. Rates were also higher after adjustment for case load. We would like to draw attention to policies such as the WHO surgical safety checklist which aim to reduce the occurrence of such adverse incidents.

Introduction: Inaccuracies in kinematic data recording due to skin movement artefact are inherent with motion analysis. Image registration techniques have been used extensively to measure joint kinematics more accurately. The aim of this study was to assess the feasibility of using MRI for creating 3D models and to quantify errors in data collection methods by comparing kinematics computed from motion analysis and image registration.

Methodology : 5 healthy and 5 TKR knees were examined for a step up/down task using dynamic fluoroscopy and motion capture. MRI scans of the knee, femur and tibia were performed on the healthy subjects and were subsequently segmented using ScanIP(Simpleware) to produce 3D bone models. Registration of the models produced from fine and coarse scan data was used to produce bony axes for the femoral and tibial models. Tibial and femoral component CAD models were obtained for the TKR patients. The 3D knee solid models and the TKR CAD models were then registered to a series of frames from the 2D fluoroscopic image data (Figure 1) obtained for the 10 subjects, using KneeTrack(S. Banks, Florida) to produce kinematic waveforms. The same subjects were also recorded whilst performing the same action, using a Qualisys (Sweden) motion capture system with a pointer and marker cluster-based technique developed to quantify the knee kinematics.

Results: The motion analysis method measured significantly larger frontal and transverse knee rotations and significantly larger translations than the image registration method.

Conclusion: The study demonstrated that MRI, rather than CT scan, can be used as a non-invasive tool for developing segmented 3D bone models, thus avoiding highly invasive CT scanning on healthy volunteers. It describes an application of combining fine and coarse scan models to establish anatomical or mechanical axes within the bones for use with kinematic modeling software. It also demonstrates a method to investigate errors associated with measuring knee kinematics.