The impact of vasoactive substances on intraosseous pressure and blood flow alterations in the femoral head: a study based on magnetic resonance imaging

Abstract

In beagle dogs, the alterations of intraosseous pressure and blood supply in the femoral head that result from the administration of vasoactive substances were examined, and the changes were documented by magnetic resonance imaging (MRI). Vasoactive substances were infused into the medial and lateral circumflex femoral arteries of 12 beagle dogs. All infusions were done under standardised conditions with simultaneous measurements of venous blood flow and intraosseous pressure distribution in the proximal femur. The drugs were infused in three cycles of 30 min each separated by 30 min recovery periods, followed by MRI examination at the end of each experiment. At an intraosseous pressure of 14.3 (± 4.2) mm Hg in the femoral head epiphysis (I), 11.6 (± 2.7) mm Hg in the greater trochanter (II) and 9.3 (± 3.2) mm Hg in the femoral shaft (III), a baseline flow of 96.2 (± 18.8, n = 12) ml/min was measured in the femoral vein. After infusing bradykinin at a concentration of 10^–6 moles, which is commonly known to lead to cerebral and subcutaneous oedema formation by vessel dilatation, the intraosseous pressure increased to (I): 49.1 (± 6.2) mm Hg, (II): 42.5 (± 5.8) mm Hg and (III): 38.3 (± 7.1) mm Hg in the three measured femoral areas (n = 3). After the bradykinin injection, femoral vein flow increased to a peak value of 238.4 (± 43.4) ml/min and then dropped to 62.3 (± 14.2) ml/min after discontinuation of the bradykinin infusion. In a second and third series of tests, hyperosmolar (20% NaCl) and hypoosmolar (distilled water) solutions were applied, also resulting in increased but lower mean intraosseous pressure values (17.3 ± 4.1 and 25.7 ± 5.1 ml/min, respectively) in all regions. When administering bradykinin, MRI scans taken immediately after completion of the experiment showed substantial oedema in the femoral muscular system, but without any changes of osseous signals in T₁- or short time inversion recovery (STIR)-weighted images, nor did any changes occur when solutions of 20% NaCl or distilled H₂O were injected. The results of our experiments demonstrate that acute increases of intraosseous pressure do not cause MRI signal alterations. We therefore conclude that in addition to the described pressure increase, other intraosseous alterations must occur to lead to the detectable signal changes found among patients with diagnosed femoral head necrosis. Finally, the short time period between the rise in intraosseous pressure and performing a conventional MRI may be one reason for missing the development of an intraosseous oedema. On the other hand, conventional MRI might have additional disadvantages for detecting intraosseous fluid compared with a dynamic imaging modality.