Interpretation of Post-operative Distal Humerus Radiographs After Internal Fixation: Prediction of Later Loss of Fixation

PURPOSE

Stable fixation of distal humerus fracture fragments is necessary for adequate healing and maintenance of reduction. The purpose of this study was to measure the reliability and accuracy of interpretation of postoperative radiographs to predict which implants will loosen or break after operative treatment of bicolumnar distal humerus fractures. We also addressed agreement among surgeons regarding which fracture fixation will loosen or break and the influence of years in independent practice, location of practice, and so forth.

METHODS

A total of 232 orthopedic residents and surgeons from around the world evaluated 24 anteroposterior and lateral radiographs of distal humerus fractures on a Web-based platform to predict which implants would loosen or break. Agreement among observers was measured using the multi-rater kappa measure.

RESULTS

The sensitivity of prediction of failure of fixation of distal humerus fracture on radiographs was 63%, specificity was 53%, positive predictive value was 36%, the negative predictive value was 78%, and accuracy was 56%. There was fair interobserver agreement (κ = 0.27) regarding predictions of failure of fixation of distal humerus fracture on radiographs. Interobserver variability did not change when assessed for the various subgroups.

CONCLUSIONS

When experienced and skilled surgeons perform fixation of type C distal humerus fracture, the immediate postoperative radiograph is not predictive of fixation failure. Reoperation based on the probability of failure might not be advisable.

TYPE OF STUDY/LEVEL OF EVIDENCE

Diagnostic III.

Survey of domestic food handling practices in New Zealand

The purpose of this survey was to obtain information on the domestic meat and poultry handling practices of New Zealanders in order to support the development of quantitative risk models, as well as providing data to underpin food safety campaigns to consumers. A sample of 1000 New Zealand residents, over 18 years of age, were randomly selected from the electoral roll and asked to participate in a national postal food safety study during 2005. Three hundred and twenty six respondents completed and returned questionnaires containing usable answers, and most of these respondents 'always' prepared the main meal within the household. The majority of meat (84.6%) and poultry (62.9%) purchased by New Zealanders was fresh (rather than frozen), and most consumers (94.4%) claimed that the time taken from food selection to reaching their home was 1 h or less. The majority (approximately 64%) of fresh meat and poultry was frozen in the home and the most favoured method of thawing was at room temperature for up to 12 h. The most common time period for storing cooked or raw meat and poultry in domestic refrigerators was up to 2 days. Most survey respondents preferred their meat and poultry to be cooked either medium or well done. The most popular cooking method for chicken was roasting or baking, while most respondents preferred to pan-fry steak/beef cuts, minced beef or sausages/hamburgers. The potential for undercooking was greatest with pan-fried steak with 19.8% of respondents preferring to consume this meat raw or rare. In answer to questions relating to food handling hygiene practices, 52.2% of respondents selected a hand washing sequence that would help prevent cross contamination. However, it was estimated that 41% and 28% of respondents would use knives and kitchen surfaces respectively in a manner that could allow cross contamination. The data in this survey are self-reported and, particularly for the hygiene questions, respondents may report an answer that they perceive as being correct rather than reflecting their actual behaviour. Nevertheless, the data on food processing, transport, storage and cooking preferences represent useful inputs into the assessment of food safety along the meat and poultry food chains.

Radiation repair in human endothelial cells

Damage to blood vessels caused by ionizing radiation is considered to be an important dose limiting factor for late effects in many organs. However, the radiation response of the endothelial cells which line the vasculature has not been well-documented, particularly for human endothelial cells. In the present study, human endothelial cells were obtained from fresh umbilical cords and cultured in monolayer. Immunofluorescent staining of factor VIII antigen was used to verify the endothelial nature of the cultured cells. The cells were irradiated with Cs-137 rays (1.35 Gy/min) in plateau phase to determine radiation sensitivity and ability to repair potentially lethal damage (PLD) and sub-lethal damage (SLD). The endothelial cells demonstrated moderate radiosensitivity that varied slightly between different cords. As demonstrated by delayed plating experiments, PLD repair ability was substantial, with repair factors of 0.70-0.80. SLD repair capability, as demonstrated by split dose experiments, was relatively modest. A survival enhancement of 2.0-2.2, for example, was observed when 8 Gy single dose survival was compared to two 4 Gy doses. In terms of PLD and SLD repair as well as initial slope (alpha) the endothelial cells were similar to normal human lung and skin fibroblasts previously studied. Compared to malignant cell lines, PLD repair was generally larger whereas the initial slope (alpha) was intermediate-steeper than the radioresistant tumor types but shallower than the more sensitive tumor derived cells.