Background
Intraosseous pressure (IOP) and bone blood flow have been studied by authors interested in bone circulation and physiology for more than 50 years. Varying techniques using different needles, flushing and recording methods have produced differing results, and there has been limited progress in understanding IOP physiology since Azuma reported IOP fluctuation in a rabbit model in 1964 [
1‐
7]. Measurement of IOP in man has also proved variable [
8]. IOP has generally been found to be raised in bone diseases such as osteonecrosis and after steroid use [
9‐
13]. A raised IOP has been associated with pain in osteoarthritic joints, chondromalacia patellae and with cartilage degeneration. IOP may also be important in driving fluid through canaliculi, hence governing osteocyte activity and bone turnover [
14,
15]. It has usually been thought that IOP had a static or fixed value which was due to venous back pressure, intramedullary pressure or interstitial pressure [
2]. IOP has been measured experimentally in animals and in man. Steroid-induced models of avascular necrosis have been developed in order to study IOP and its treatment [
16,
17]. Ficat and others developed a technique for the ‘functional exploration’ of bone in patients with early osteonecrosis [
10]. However, the factors that control IOP at rest and during activity and the physiology of subchondral bone circulation remain largely unknown [
18].
In a preliminary unpublished clinical work, we measured IOP prior to forage, osteotomy or decompression. The findings were variable. The rationale for our study was to explore the physiology of IOP in healthy subchondral cancellous bone at rest in an animal model.
Discussion
This study of the physiology of resting intraosseous pressure measurement in a healthy animal model gives insights which contradict some current views. IOP has usually been thought to have a fixed or static value which reflects an interstitial pressure, tissue turgor or venous back pressure [
2,
10]. We found variation in IOP between and within normal subjects even when using a standardised approach. Although IOP was not related to gender, weight, size or site of needle, it was proportional to blood pressure. In addition, pulse pressure was proportional to IOP. The demonstration of variability in IOP and proportionality between IOP and PP shown by our 44 different sites in otherwise normal bone strongly suggests that the needle tip strikes different vessels within the cancellous bone by chance and that the IOP and PP reflect conditions in the blood pool at the needle tip rather than IOP being of a fixed value for any whole bone or subject. In the blood pool at the needle tip, we assume that IOP from the highest pressure vessel will be recorded. Larger arterioles will give a higher IOP and PP as they are nearer the arterial side of the vascular tree. Smaller vessels or capillaries give a lower IOP and lower PP. Venules, fat and trabeculae may return virtually no pressure.
IOP not only correlated with pulse pressure, but IOP also correlated with blood pressure, and was seen to correspond to respiratory waves and systemic or drug circulatory effects further suggesting that IOP is governed mainly by the arterial circulation. Moreover, there was a greater fall in IOP with arterial occlusion than a rise with venous occlusion. This would also indicate that the majority of the recorded IOP is due to the arterial supply side. The disappearance from the trace of the pulse pressure with arterial occlusion also demonstrates that the IOP and both the associated pulse and respiratory waves are mediated through the arterial supply. Both are preserved when the proximal femoral vein is clamped. Once again this suggests that IOP is therefore a ‘supply’ side phenomenon rather than a venous back pressure or measure of drainage. We found no previous references to this analysis of the physiology of IOP [
8].
We have shown that IOP is not a constant but varies between and within subjects and also depends on the method and timing of measurement. Within that variability, there are recognisable wave forms with arterial, respiratory and drug or circulatory time patterns.
Whilst trying to optimise the method of measuring IOP, we compared the usual standard method, which involves injecting a small clearance bolus of saline (Ficat technique), with a novel method in which the needle was aspirated prior to IOP measurement. In healthy bone, both techniques resulted in an initial decrease in IOP which recovered but not necessarily to exactly the same starting point (Fig.
2). The recovery after aspiration was quicker and was usually stable within 1 min. After saline injection up to 10 min was required for recovery. Saline bolus injection appears to be harmful to the local microcirculation [
19]. The delay in return to normal IOP may reflect a washout or recovery period. The subsequent slightly higher IOP may reflect renewed contact with larger arterial supply vessels following local destruction. The use of forced saline injection should probably be avoided. The aspiration method appears preferable. Previous workers assessing IOP after using the bolus injection method may in fact have been measuring an IOP which was temporarily lowered as a result of their saline injection [
2,
7,
9,
10]. It is equally possible that in osteonecrotic bone the IOP recorded would be high because of the pressure of the injection itself into poorly drained or ischaemic bone. A third possibility is that the flushing of heparinised saline, blood, fat and bone fragments backwards into the delicate subchondral vascular tree is physically damaging or toxic [
19]. The slower recovery after injection may represent a prolonged washout or recovery phase which is not required after less damaging aspiration.
The residual pressure after arterial clamping, for example in Fig.
8, probably does represent a real residual venous or tissue back pressure at the needle tip. Similarly with proximal venous occlusion, the IOP recorded may represent the best obtainable arterial supply pressure at that needle tip. For the first time, this simple approach gives a means of assessing the perfusion pressure range obtainable at the needle tip deep in the cancellous bone.
There are several potential limitations in this study. X-rays were not available for needle placement, but under direct vision hand placement of the needle tip was to within a few millimetres of the subchondral surface as illustrated in Fig.
1. The animals used were similar healthy adults but were not identical. Different sites were used, initially mainly the femoral head and later mainly at the femoral condyle and proximal tibia which were technically easier. We could not always obtain tracings from all sites in all subjects. The subject’s blood pressures inevitably varied during the experiments. Needle insertion could never be identical at all sites. The experiments varied in duration from about 15 min to over 2 h. Because IOP inevitably fluctuated during that time, an average of the early, middle and late values was used for each of the 44 sites. There was a learning curve but generally anaesthetic control and experimental duration increased with experience [
7,
20,
21]. Although the known experimental differences resulted in some variability in results, our experimental conditions were generally representative of the clinical situation. The analysis was based on data from different sites. It could be argued that these are not completely independent as in some subjects there were two or three sites. However, the IOP was not significantly different at different anatomical sites. If the analysis was based on subjects alone, the numbers would be smaller. The work was primarily qualitative and designed to develop a standardised model. Nevertheless, some useful insights on perfusion physiology have been derived from our study. This may in turn allow better use of IOP measurement in the study of circulation in diseased bone states such as osteoarthritis and avascular necrosis. Future work will measure IOP under load and with vascular occlusion to give a better understanding of subchondral perfusion physiology.
Acknowledgements
The authors would like to thank Dr. Joseph Pflug for his advice and support with early work at the Hammersmith Hospital and Dr. Jill Urban for her advice and support. We thank Mrs. Barbara Marks for her support and assistance with submission.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (
http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.