| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Orthopaedics, and 2Institute for Surgical Research, Klinikum Grosshadern, Ludwig-Maximilians University, 81377 Munich, Germany
Submitted 25 August 2003 ; accepted in final form 15 December 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
experimental study; osteoarthritis
To investigate the relevance of RBBF in skeletal disorders, long-term experiments are essential. The microsphere method offers several crucial advantages, which predispose this technique for application in long-term experiments. Due to the inherent principle of the method, cardiac output (CO) and blood flow in other organs can be determined simultaneously in a single measurement. Knowledge of CO as well as of regional blood flow values in other organs like the kidneys is essential to interpret the measured blood flow values correctly and to exclude methodological errors due to inadequate mixing or shunting of the spheres. The precision of the measurements can be easily assured by simultaneously injecting two different microsphere species. The microsphere method allows repetitive measurements without surgical manipulation of the bone. Finally, distribution of blood flow within long bones can be investigated. The reliability of the microsphere method for measurement of RBBF has been demonstrated for both radioactive (RM) and fluorescent microspheres (FM) in acute experiments (2, 3). In contrast to RM or colored microspheres, the applicability of FM in chronic experiments was validated for various organs in rabbits, except for osseous tissue (29). Besides other disadvantages inherent to handling of radioactivity, the main reason for the limited application of RM is that isotopes tend to leak from microspheres with time and, therefore, lead to an underestimation of blood flow (14, 29).
In summary, the FM method appears ideally suited to investigate the relevance of RBBF in the development of osteoarthritis. In our experiments, osteoarthritis was induced by performing a medial meniscectomy in the right knee of New Zealand white rabbits. Bone blood flow measurements were performed for up to 11 wk postoperatively. Two hypotheses were tested: 1) FM method, which was demonstrated to be a useful tool for measurement of RBBF in acute experiments is also valid and accurate for measurement of RBBF in a long-term study; and 2) development of osteoarthritis induced by meniscectomy leads to changes in RBBF in the periarticular regions but remains unchanged in other regions of the femur and tibia.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Twenty-four mature female New Zealand White rabbits (Charles River, Kisslegg, Germany) were used in this study. Mean body weight (BW) was 3.8 ± 0.2 kg in group 1, 3.8 ± 0.1 kg in group 2, and 3.9 ± 0.2 kg in group 3. Eight rabbits were assigned to one of three groups. Blood flow measurements were performed for up to 11 wk postoperatively. In group 1, blood flow was measured preoperatively as well as 4, 7, 14, and 21 days after meniscectomy. Measurements in group 2 were carried out weekly from postoperative weeks 3 to 7 and, in group 3, from postoperative weeks 7 to 11 (Fig. 1). The epiphyseal plate was closed in all rabbits, as ascertained during dissection of the bone samples. Pre- and postoperatively, animals were kept in groups on the ground (cage size 15 m2) and were allowed free movement. Standard laboratory rabbit chow and water were provided ad libitum. Postoperatively, wound healing and BW were assessed daily until full functional recovery.
|
Surgical procedure and microsphere technique. Animals were anesthetized for meniscectomy and for each injection of FM by an intramuscular injection of ketamine (15 mg/kg body wt; Pharmacia & Upjohn, Erlangen, Germany) and xylazine (2 mg/kg body wt; Bayer, Leverkusen, Germany). Rabbits were fixed in supine position and allowed to breathe room air and supplemental oxygen spontaneously.
The meniscectomy and the preparation of the right carotid artery for the injection of microspheres were performed under sterile conditions.
Medial meniscectomy was performed on the right knee according to the modified protocol of Messner et al. (20). The skin was incised anterior to the medial collateral ligament. After the joint capsule was opened vertically, the anterior insertional ligament of the medial meniscus was transected, and the anterior horn and medial part of the meniscus was dissected free from its capsular attachment and from the medial collateral ligament. After the posterior insertional ligament was dissected by means of incision of the posterior capsule, the medial meniscus was removed. After joint irrigation with normal saline, joint capsule and skin were closed with single sutures.
For the injection of microspheres, the right carotid artery was isolated and canulated with a catheter (flow rate: 10 ml/min; Cavafix MT; B. Braun Melsungen Aktiengesellschaft, Melsungen, Germany). The left ventricle was catheterized, and correct positioning of the catheter tip was ascertained by observing the typical waveform of the left ventricular pressure curve. Blood pressure and heart rate were continuously monitored throughout the experiment.
FM of seven different colors (blue, blue-green, yellow-green, orange, red, crimson, scarlet) 15.5 ± 0.3 µm in diameter (Fluo-Spheres, Molecular Probes, Leiden, The Netherlands) were used for the experiments. For each measurement, FMs of one color were injected in a dosage of 1 x 106 spheres/kg body wt. FM were withdrawn from the stock solution after being sonicated and vortexed and were diluted in a syringe with 0.9% NaCl to a total volume of 10 ml. To determine CO, 20-µl aliquots were taken from each suspension (I20µl) to calculate the total fluorescence intensity (Iinjected) of each FM species administered (Iinjected = 500 x I20µl). Before each injection, microspheres were again sonicated and vortex mixed.
To determine the accuracy of the FM method, the last injection in group 1 as well as the first and last injections in groups 2 and 3 were performed by simultaneously injecting FM of two different species (Fig. 1).
FMs were injected manually over a 60-s interval. Arterial reference blood samples were withdrawn from a cannula (flow rate: 31 ml/min; 22 gauge; Vasodrop, Bad Hersfeld, Germany), which was inserted in the left auricular artery. Arterial reference blood sampling started 15 s before the injection and was continued for a period of 120 s. The withdrawal rate of the Harvard pump (33 syringe pump, Föhr Medical Instruments, Egelsbach, Germany) was 3.54 ml/min.
Immediately after animals were killed, kidneys, lungs, brain hemispheres, humeri, femora, and tibiae were removed and cleaned of all connective tissue. Subsequently, kidneys were dissected into 8, lungs into up to 10, brain hemispheres into 4, femora into 8, tibiae into 7, and humeri into 5 tissue samples. The weight of the individual samples, which was determined immediately on dissection, was between 1 and 2 g. FMs were isolated from arterial reference blood, lung, brain, and kidney samples immediately after the experiment, and fluorescence intensity was measured. Fluorescence intensity of the bone samples was determined after the crystalline matrix was dissolved for 3 wk in hydrochloric acid (1 M). Measurement of fluorescence intensity was carried out by an automated system (28) using the sample-processing unit (Gaiser, Kappel-Grafenhausen, Germany) (22).
Fluorescence data were used to calculate the blood flow values for each tissue sample (
sample) as follows
 |
where reference is the withdrawal rate of the Harvard pump (3.54 ml/min), Isample is the fluorescence intensity of the individual tissue sample, and Ireference is the fluorescence intensity of the reference blood sample.
To allow comparison of different samples, sample was divided by the sample weight and normalized to 100 g.
CO was calculated by multiplying the reference blood withdrawal rate (3.54 ml/min) with the ratio of Iinjected and Ireference and divided by the BW (
).
Statistical analysis. Relative bone blood flow was calculated by dividing the measured fluorescence of each tissue sample by the mean fluorescence for the total organ. Relative blood flow values of different organs obtained by simultaneously injected FM were compared using least-squares linear regression analysis. The coefficient of correlation (r), the coefficient of determination (r2), the slopes, and the intercepts were computed and compared for a two-sided 95% confi-dence interval. The method of Bland and Altman (4) was used to evaluate absolute measurement error between two simultaneously injected types of FM.
ANOVA on ranks according to Friedman was used to determine significant differences between repeated measurements. Comparison of blood flow values of right- and left-sided samples was assessed by using Wilcoxon's signed rank test. A P value of <0.05 was considered to be significant.
The statistical analyses were carried out with SPSS software for Windows, version 11.5 (SPSS, Chicago, IL).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
No statistically significant differences were found between the weights of all right- and left-sided tissue samples (right kidney: 1.128 ± 0.023 g; left kidney: 1.127 ± 0.023 g; right lung: 1.443 ± 0.050 g; left lung: 1.398 ± 0.053 g; brain right hemispheres: 1.031 ± 0.034 g; brain left hemispheres: 1.040 ± 0.333 g). The weight of all bone samples was not statistically different when right- and left-sided samples were compared from corresponding regions (see Table 3).
|
Blood flow in right kidneys varied between 259 ± 24 (group 3, postoperative week 8) and 437 ± 26 ml·min-1·100 g-1 (group 1, postsurgery). No differences were observed when blood flow values of right and left kidneys were compared (Fig. 2). Blood flow in lungs varied between 65 ± 9 (group 3, postoperative week 11) and 111 ± 17 ml·min-1·100 g-1 (group 3, postoperative week 7). Absolute blood flow was slightly lower in the right compared with the left lung; however, differences did not reach statistical significance (Fig. 2). Blood flow in both brain hemispheres was identical and varied between 95 ± 18 (group 1, preoperative) and 177 ± 27 ml·min-1·100 g-1 (group 2, postoperative week 7) (Fig. 2).
|
Relative bone blood flow values obtained by simultaneously injected FM showed an excellent correlation both at the first and last measurement. Results from least squares linear regression analysis showed excellent correlation. The confidence intervals of the slope included 1 or were close to 1, and the confidence intervals of the intercept included zero or were close to zero (Table 2). Comparison by the method of Bland und Altman (4) resulted in a mean difference of zero and a uniform distribution of scatter above and below zero at each injection. The linear regression analysis and the Bland and Altman comparison of all bone blood flow values for the last injection in group 3 are presented in Fig. 3.
|
|
Mean blood flow in the right humerus ranged between 14.0 ± 1.3 (postoperative week 7) and 20.0 ± 2.3 ml·min-1·100 g-1 (postoperative week 5) and did not differ between the right- and left-sided bones. At each time point of measurement, mean blood flow in the humerus was higher than in the femur, which varied between 10.0 ± 0.7 (postoperative week 7) and 17.3 ± 1.2 ml·min-1·100 g-1 (postoperative week 2). Mean blood flow in the tibia was lower compared with the femur [between 6.2 ± 0.6 (postoperative week 7) and 11.5 ± 1.6 ml ·min-1·100 g-1 (postsurgery)] at all time points. In contrast to the humerus, we found slightly higher blood flow values in femur and tibia on the operated right side compared with the left side. The contralateral difference in the femur was statistically significant different at postoperative week 1 (right: 15.8 ± 1.1 ml·min-1·100 g-1; left: 13.6 ± 1.2 ml·min-1 ·100 g-1) and at postoperative week 3 (right: 16.3 ± 1.1 ml·min-1·100 g-1; left: 14.8 ± 1.1 ml·min-1·100 g-1). In the tibia, significantly lower RBBF was measured postsurgery (right: 9.1 ± 0.9 ml·min-1·100 g-1; left: 11.5 ± 1.6 ml·min-1·100 g-1). Significantly higher RBBF values were found in the operated tibia at postoperative weeks 3 (right: 9.5 ± 0.8 ml·min-1·100 g-1; left: 8.9 ± 0.7 ml·min-1·100 g-1), 7 (right: 8.0 ± 0.8 ml·min-1·100 g-1; left: 7.0 ± 0.7 ml·min-1·100 g-1), and 8 (right: 7.4 ± 0.7 ml·min-1·100 g-1; left: 6.2 ± 0.6 ml·min-1·100 g-1). Data are presented in Table 3.
In the humerus, blood flow values were found to be highest in the humeral head and in the greater tubercle (from 14 to 33 ml·min-1·100 g-1) and lower in the metaphysis, diaphysis, and distal epiphysis (from 7 to 14 ml·min-1·100 g-1). RBBF did not differ when the right- and left-sided identical regions of the humerus were compared. Data are presented in Table 3.
In the femur, the highest flow values were found in the femoral neck and the distal metaphysis (between 20 and 25 ml·min-1·100 g-1). About 15-20 ml·min-1·100 g-1 were measured in the greater trochanter, the proximal metaphysis, and the diaphysis, whereas RBBF in the femoral head and in the medial and lateral condyles was lowest at 
5-8 ml·min-1 ·100 g-1. The distribution of blood flow within the femur remained unchanged up to postoperative week 7. From the postoperative week 8 up to week 11, a significantly lower RBBF (5 ml·min-1·100 g-1) in the right distal femoral metaphysis was observed compared with the nonoperated left side. In contrast to that finding, an increase in RBBF in the right-sided femoral condyles was observed. This difference was significant at postoperative weeks 7 and 8 (Table 3).
In the tibia, blood flow values were highest in the proximal metaphysis and ranged from 12 to 21 ml·min-1·100 g-1. In the medial and lateral condyles, blood flow ranged from 7 to 16 ml·min-1·100 g-1. Lowest blood flow values were found in the distal diaphysis and in the epiphysis (between 2 and 8 ml·min-1·100 g-1). Postoperatively, in the right-sided medial and lateral tibial condyles, an increase of 3 ml·min-1·100 g-1 was observed compared with the nonoperated left side. At postoperative week 3, a significant increase in RBBF on the right medial and lateral tibial condyles compared with the nonoperated left side was found. This difference was statistically significant at up to postoperative week 8 but was consistently present during the whole time period of observation. Data of each region and each time point of measurement are presented in Table 3 and Fig. 4.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
In 1998, Van Oosterhout et al. (29) established the microsphere method using FM for blood flow measurements in long-term studies in numerous organs of rabbits except osseous tissue. They found that FMs are superior to RM in experiments lasting longer than 1 day. Our laboratory recently was able to demonstrate that the FM method is valid for RBBF measurements in acute experiments (2, 3). To establish the microsphere reference sample technique using FM for the measurement of RBBF in long-term studies, sources of errors have to be excluded. In the study presented here, stochastic and methodological errors were minimized by using the same experimental setup as described previously (2). To allow for the analysis of systematic errors, blood flow was determined in reference organs, and the accuracy of measurements was tested by simultaneous injection of differently labeled microspheres. The comparison of regional blood flow values of bilateral organs revealed no differences, indicating adequate mixing of FM in the blood. Shunting or dislodgment of spheres over time is unlikely. Dislodging of microspheres would have to be reflected in an increase of absolute blood flow in the lungs, since shunted microspheres are primarily trapped in the lung (15). We did not observe an increase of blood flow in the lungs over time. Blood flow in both brain hemispheres was equal at each measurement, indicating that repeated dissection of the right carotid artery did not influence cerebral blood flow.
The linear correlation between bone blood flow values calculated from simultaneous injections of differently colored FM was excellent and did not differ between the first and last measurements. The 95% confidence intervals of the slope included 1 or were close to 1, and the 95% confidence intervals of the intercept included 0 or were close to 0, indicating absence of systematic error. Intramethod comparison by the method according to Bland and Altman (4) yielded no difference between blood flow values calculated at the first and last measurements. The standard deviation of the difference as a parameter of the precision of the measurement was similar to what we estimated for RBBF measurements in acute experiments (2) and to what was known from other organs (25). The high consistency of relative blood flow values over the period of 4 wk suggests that errors due to leaking of dye or dislodgment of microspheres are unlikely and thus confirms the finding of van Oosterhout et al. (29). Furthermore, these results show that secondary effects on the microcirculation after repeated injection of high doses of microspheres are negligible.
It is well known that bone blood flow shows a variability between different individuals (5, 13, 19). To exclude changes of bone perfusion caused by altered hemodynamic conditions, hemodynamic parameters should be constant during the experiment. In our study, CO, arterial blood pressure, and heart rate were stable over time.
By comparing blood flow values of right- and left-sided bone samples from individual regions of the humerus, we were able to demonstrate that there are no side differences under constant hemodynamic and adequate experimental conditions in long-term experiments. This is in accordance with previous findings in acute experiments (2, 9, 11, 26).
In contrast, after meniscectomy, we observed an increase in RBBF values in the right tibial condyles as soon as postoperative week 1 compared with the nonoperated left side. This increase was consistent during the complete period of observation and was significant from the postoperative weeks 3 through 7. In the other regions of the tibia, no side difference in bone blood flow was observed at any time point. In the femoral condyles of the operated joint, we also observed an increase in RBBF values compared with the nonoperated left side. However, this increase was not statistically significant. Similar findings were reported by Shymkiw et al. (26) in a rabbit model. By use of colored microspheres, they found an increase in RBBF in the periarticular bone of the femoral condyles at 2 wk up to 48 wk after surgery. They also observed a significant decrease of periarticular bone mineral density after anterior cruciate ligament transsection and concluded that increased bone blood flow may be linked to mechanisms of bone adaptation secondary to an altered mechanical situation. A direct influence of the surgical procedure on RBBF is unlikely because Gross et al. (11) observed an increase in RBBF 7 days after osteotomy but not in the sham group in an animal model. They concluded that hyperemia precedes the onset of disuse-induced intracortical resorption.
Nevertheless, to interpret the increase in RBBF, some limitations of our study have to be taken into consideration. First, although all rabbits reached full weight bearing on the operated leg from postoperative day 7 onward and no protective limping was observed at any time during the entire period of observation, postoperative activity could have had an influence on RBBF (10). Second, we did not exclude possible effects of the surgery on RBBF by using a sham surgery. Finally, the microsphere method cannot distinguish between neoangiogenesis and vasodilatation. Hence, the observed increase in RBBF in the distal femoral condyles and the proximal tibial condyles could reflect an increased flow due to vasodilatation and/or neoangiogenesis.
In conclusion, our study demonstrates, that the microsphere method (reference sample technique) using FM is an accurate and reliable method to perform repetitive measurements of RBBF in long-term experiments. The increase in RBBF in periarticular regions such as the femoral and tibial condyles after meniscectomy indicates that hyperemia is locally mediated. RBBF was elevated temporarily for up to 3 mo after meniscectomy, supporting the assumption that RBBF is linked to the mechanism of bone adaptation in periarticular regions after mechanically induced osteoarthritis after meniscectomy.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
