Article
Risk of cement leakage during vertebroplasty depending onAO- and OF-Fracture classification, as well as Hounsfield units in perioperative computed tomography – a retrospective monocentric analysis
Eine retrospektive monozentrische Risikoanalyse für die Zementleckage bei perkutaner Augmentation, basierend auf Frakturklassifikation, Hounsfield-Skala und Wirbelkörperkonfiguration
Search Medline for
Authors
-
Andreas Baumann
- Krankenhaus Köln-Merheim, Klinikum der Universität Witten/Herdecke, Klinik für Neurochirurgie, Köln, Deutschland
-
Ghassan Abuharbid
- Krankenhaus Köln-Merheim, Klinikum der Universität Witten/Herdecke, Klinik für Neurochirurgie, Köln, Deutschland
-
Kazimierz Sadowy
- Krankenhaus Köln-Merheim, Klinikum der Universität Witten/Herdecke, Klinik für Neurochirurgie, Köln, Deutschland
-
Paiman Shalchian-Tehran
- Krankenhaus Köln-Merheim, Klinikum der Universität Witten/Herdecke, Klinik für Neurochirurgie, Köln, Deutschland
-
Makoto Nakamura
- Krankenhaus Köln-Merheim, Klinikum der Universität Witten/Herdecke, Klinik für Neurochirurgie, Köln, Deutschland
| Published: | May 25, 2022 |
|---|
Outline
Text
Objective: Osteoporotic vertebral fracture is a frequent incident in the growing older population. Percutaneous cement augmentation is an established procedure reducing pain in the acute phase. There is varying evidence regarding the frequency of cement leakage (20 to 70%). The revision and simplification of the AO-fracture classification in 2013 and the available OF-classification could give an opportunity for a standardized approach to fractures safely suitable for cement augmentation. We wanted to stratify the risk of cement leakage by these scores and Hounsfield units in computed tomography scans of the spine (HU).
Methods: A retrospective analysis was conducted including 136 cases of cement augmented thoracolumbar vertebral bodies between January 2017 and December 2019. The patients were followed based on medical records, operative reports and imaging studies. The major depended variable was cement leakage. Independent variables have been tested univariately, as well as with logistic regression. Data analysis has been performed with SPSS version 21 (Armonk, NY: SPSS IBM Corp.).
Results: The mean age of our cohort has been 75+/-10 years. Postoperative imaging revealed a total of 70 leakages (51%), mainly not within the spinal canal (54/70=77%). Three patients underwemt urgent revision surgery. None of these patients suffered a lasting neurologic deficit. Fracture classification has shown as following: A1 – 61, A2 – 42, >A3 – 33. The mean HU have been 79 (+/- 45). There was significant correlation of cement leakage and a fracture more than A1 (p=0,045), a vertebral body collapse of more than 25% (p=0,002) and at least a tendency concerning HU less than 50 (p=0,086). The sole analysis of transcortical leakage in particular, other than intraspinal leakage revealed an even better correlation with AO classification and the vertebral body height, confirmed by logistic regression.
Conclusion: The AO fracture classification system is a possible way to stratify the risk of cement leakage during percutaneous cement augmentation in osteoporotic compression fracture. Nevertheless simple measuring of vertebral body height seams to allow an even better prediction of this common complication. Pain reduction and the low frequency of cement outflow within the spinal canal itself stays an argument for the safeness of the procedure.
