Publikationsdatenbank

Scintigraphic image evaluation of thoracic processus spinosi in horses (2019)

Art

Hochschulschrift

Autor

Zadelhoff, Claudia van (WE 17)

Quelle

Berlin: Mensch und Buch Verlag Berlin, 2019 — 45 Seiten

Sprache

Englisch

Verweise

URL (Volltext): https://refubium.fu-berlin.de/handle/fub188/24640

Kontakt

Pferdeklinik

Oertzenweg 19 b
14163 Berlin
+49 30 838 62299 / 62300
pferdeklinik@vetmed.fu-berlin.de

Abstract / Zusammenfassung

Scintigraphy of the processus spinosi is a standard procedure for the evaluation of horses with back pain. The scintigraphic evaluation is based on a subjective or semi-quantitative analysis of areas with increased radiopharmaceutical uptake. The first part of this retrospective, analytical thesis aimed to compare subjective and semi-quantitative methods for the evaluation of scintigraphic images of the processus spinosi in the equine thoracic spine. Additionally, a new modified semi-quantitative evaluation method was developed and analysed for inter- and intraobserver agreement. The second aim was to investigate the agreement between the modified semi-quantitative scintigraphic evaluation technique and other established scintigraphic evaluation methods. Subsequently the agreement of scintigraphic evaluation techniques with clinical and radiological findings was analysed. Scintigraphic images of the thoracic processus spinosi of 20 Warmblood horses were subjectively assessed by eleven veterinary surgeons using red-green-blue (RGB) scale and greyscales for entire scintigraphic images or masked images. For semi-quantitative assessment, the observers placed "regions of interest" (ROIs) across each processus spinosus of the 13th to 18th thoracic vertebrae. The radiopharmaceutical uptake of each processus spinosus was determined in comparison with a reference ROI. Subsequently, a modified semi-quantitative calculation was developed in which only the highest counts per pixel for a given number of pixels were included in the calculation. The inter- and intraobserver agreement was determined using intraclass correlation coefficients (ICCs). Inter- and intraobserver ICCs were 41.65% and 71.39%, respectively, for the subjective image scoring. A correlation between intraobserver agreement, experience of the operator and scintigraphic greyscale images was identified. Inter- and intraobserver agreement was significantly increased using the semi-quantitative analysis (97.35%, 98.36%) or modified semi-quantitative calculation (98.61%, 98.82%). For the second part of the study, 223 Warmblood horses underwent scintigraphic, radiographic and clinical examination of the thoracic spine and were included in a retrospective study. Scintigraphic images were assessed using subjective, semi-quantitative and the newly developed modified semi-quantitative techniques. Radiographs were evaluated subjectively (grade 0-7). The analysed horses were assigned to one of two groups, including horses with or without thoracolumbar pain. The total radiographic and total (subjective) scintigraphic grades were significantly higher in horses with thoracolumbar pain than in horses without thoracolumbar pain (p < 0.05). Semi-quantitative and modified semi-quantitative radiopharmaceutical uptake ratios did not differ significantly in horses with or without thoracolumbar pain. The calculation of kappa agreement between scintigraphic evaluation methods revealed a substantial agreement between the modified semi-quantitative and the semi-quantitative scintigraphic evaluation technique. The agreement between subjective scintigraphic and modified semi-quantitative scintigraphic image analysis was fair. There was slight agreement between all scintigraphic techniques and the radiographic findings. In summary, the modified semi-quantitative scintigraphic image scoring has provided results that are comparable with the established scintigraphic evaluation methods but with a significant increase in inter- and intraobserver agreement. However, an association between results obtained with semi-quantitative scintigraphic image evaluation techniques and clinical signs of thoracolumbar pain could not be observed. More factors may need to be added to semi-quantitative radiopharmaceutical uptake ratio calculations to improve the clinical significance of these methods.