According to the limited joint volume effect, a relative negative pressure was induced in the changes in joint volume using a joint distraction load. The manual loading device applied a fixed axial distracted displacement on the MCP joint specimen. Therefore, the joint volume changes were observed using the Micro CT images, and the joint volume was also calculated using the software CTAn from the Micro CT company.
Fig 3.4 (A) shows a plot for the displacement versus total volume. The results indicate that the total volume was linearly related to the amount of displacement.
The total volume in the middle finger was larger than the total volume in both the ring and the little finger due to anatomic morphology. Because the necking volume was the difference between the virtual vented volume and total volume, the necking volume was also linearly related to the amount of displacement (Fig 3.4 (B)).
(A)
39
(B)
Fig 3.4 (A) Total volume and (B) Necking volume are linear between the displacement and volume. (n=4)
3.3 Load, displacement and IAP
Four out of the nine measured IAPs from the MCP specimens were abandoned due to significant increases as compared with the intact condition of the displacement at the cross-point. This was presumably due to possible leakage resulting from pressure measurement. As a result, only five IAP measurements of MCP joints (one little finger, two middle, and two ring fingers) were obtained. The initial IAP value, without load, for each specimen was nearly 0mmHg relative to the atmospheric pressure. The maximum decreases in the IAP, when loaded under conditions of 16 kg of load, ranged from 11.59mmHg to 42.20mmHg. The IAP values against the load and displacement of these five specimens are plotted in Fig 3.5 and Fig 3.6, respectively. The IAP values at the cross-point of each specimen are also marked in the figures. The correlation coefficient between the IAP and load was -0.531 to -0.899 (r = -0.792 ± 0.151). The correlation coefficient between the
40
IAP and displacement was -0.942 to -0.993 (r = -0.963 ± 0.029). From Fig 3.5, it can easily be seen that the drop in IAP occurred mostly before the cross-point; there was a 45% to 80% decrease in the IAP before the cross-point. Highly non-linear curves are presented in the IAP load plots, as shown in Fig 3.5, while more linear segments can be observed in the IAP displacement plots shown in Fig 3.6.
Fig 3.5 Load versus IAP with the cross-point marked. Highly non-linear curves are presented in the IAP load plots (r = -0.792 ± 0.151). The IAP decreased rapidly before
the cross-point, during 5% to 10% total distraction of load. The IAP then slowly decreased with increasing load. (n=5)
41
Fig 3.6 Displacement versus IAP. The curve presented a more linear pattern than that of the IAP - load curve (r = -0.963 ± 0.029). (n=5)
The data from the pneumatic loading device was not collected on one middle finger due to a malfunction in the Micro CT scanner. Therefore, only data for four fingers (one little, two ring, and one middle fingers) were used for further changes in the joint volume analyses. The necking volume and IAP plot was shown in the Fig 3.7. Correlation among the all necking volume and IAP is highly linear in a negative relation (r= -0.882). The correlation coefficient between the necking volume and IAP in each specimen was range -0.920 to -0.982 (r = -0.951 ± 0.025).
The total volume and IAP plot is shown in the Fig 3.8. The correlation coefficient between the total volume and IAP (r = -0.762) was not high such as was observed in the relation between the necking volume and IAP (r = -0.882) or in the relation between the displacement and IAP (r = -0.838) in all specimens. However, the correlation between the necking volume and IAP in each specimens is highly linear in a negative relation (range: -0.960 to -0.994, r = 0.982 ± 0.015).
42
Fig 3.7 The necking volume versus IAP. The correlation between the necking volume and IAP in all specimens is highly linear in a negative relation (r= -0.882). (n=4)
Fig 3.8 The total volume versus IAP. The correlation coefficient between the total volume and IAP in all specimens was not highly correlation (r = -0.762). However,
the correlation between the necking volume and IAP in each specimens is highly linear in a negative relation (range: -0.960 to -0.994) (n=4)
43
3.4 Finite element analysis
In the preliminary finite element (FE) study, the material properties are assumed to be homogenous, linearly isotropic. The axial displacement is 5.01 mm under a 16 kg axial loading with corresponding values of negative IAP applied. (Fig 3.9 (A)) The axial displacement increased to 5.26 mm when a 16 kg axial loading was applied without the negative IAP (Fig 3.9 (B)). The displacement data are similar the experimental results.
The radial deformation of the capsule in the situation when both negative pressure and a 16kg distraction load were applied was 1.88 mm. This value was 1.72 mm when only 16 kg load was applied. The radial deformation ranged from 5 mm in Micro CT measurements. The pattern of capsule deformation was different between these two conditions. Major deformation was observed on the dorsal side and palmar side of the capsule when both the negative pressure and the 16 kg load were applied (Fig 3.10 (A)). The circular deformation on the center of the capsule was observed when only the 16 kg load was applied (Fig 3.10 (B)).
(A)
44
(B)
Fig 3.9 The axial displacement (A) Negative pressure and 16 kg axial loading is applied. (B) Only 16 kg is applied
(A)
45
(B)
Fig 3.10 The radial deformation (A) Negative pressure and 16 kg axial loading is applied. (B) Only 16 kg is applied
The original cross-section images of the joint capsule in initial condition from the results of FE model and the CT image are shown in Fig 3.11 (A). The cross-section images of the joint capsule with axial load distraction from the results of FE model and the CT image are shown in Fig 3.11 (B). A large different was found in the radial deformation of the joint capsule between the finite element model and the CT image with axial load distraction. The geometry and material of the joint capsule is different between the FEM model and specimen.
46
(A)
(B)
Fig 3.11 The cross-section image of MCP joint capsule in FE model and CT image (A) Initial condition (B) under axial distraction load
47
CHAPTER 4 Discussion
In the present study, the IAP was measured in the MCP joint under a distraction load. A Micro CT compatible loading device was developed to measure the changes in joint volume. The relation of the IAP and the changes in joint volume was investigated in this study. The effect of IAP on an MCP joint under a distraction load in intact and vented conditions was also investigated.
In the present study, the displacement value under 16 kg of load was 4.60 ± 0.80mm for the intact MCP joint. These values are similar to those reported in the literature as measured by radiography during a distraction load of 14-16 kg.
4.1 The measurement of displacement data from the loading device
A custom-made Micro CT compatible loading device was developed in this study. Although a commercially available micro CT compatible loading device exists, the specimen dimensions allowed (18-23mm length and 20mm object diameter) are too small, and therefore, it is not suitable for detecting the volume change in an MCP joint (Object stages for in situ examination: MATERIAL TESTING STAGE http://www.skyscan.be/products/stages.htm) (Evans et al., 2012).
The custom-made loading device provided a larger space for the specimen (30-40mm in length and 40mm in object diameter) for purpose of Micro CT scanning. Although the accuracy in displacement measured by a cable displacement
48
transducer and a micrometer is different from that of actual real displacement value, such a setup when coupled with a radiographic image, provides a more precise and concurrent measurement of distraction load, joint displacement, and joint volume.
The displacement profile was taken with a micrometer and a radiographic image before the 2 kg load was reached. However, as the loading was increased to 5-16 kg, the displacement difference increased to 0.15-1.5mm. There are two possible reasons for the difference found in the displacement profiles. First, a slight deformation from the load cell was included in the micrometer or cable displacement transducer measurements. The load cell converted the load into electrical signals, and metal foil strain gauges in the load cell deformed when the load was applied (Load cell http://www.transducertechniques.com/load-cell.aspx).
The cable displacement transducer or micrometer attached behind the load cell would include the deformation of the load cell in the overall displacement measurement. Second, the loading frame material was made of the carbon fiber. It is possible that the loading frame was deformed when the maximal a 16 kg load was applied. Two simple tests were executed to verify the source of error. A plastic bar was fixed on the stainless clamps on the loading device. The difference in distance between the pneumatic cylinder and the load cell was 0.33 mm and the difference in distance between the slider and the load cell was 1.17 mm at the 16kg load. (Fig 4.1) The total difference in the distance on the load cell was 1.47 mm, as derived from caliper measurements. Then, a 16 kg load was applied using the MTS. The deformation of the loading frame was 0.14 mm at the 16 kg load. (Fig 4.2) Therefore, this might account for the differences in displacement between the micrometer or cable displacement transducer on the radiographic image measurements.
49
Fig 4.1 To determine the error, the changes in the distance between the pneumatic cylinder and load cell as well as the distance between the slider and the load cell in the
loading device were measured.
Fig 4.2 A 16 kg load was applied by the MTS. The deformation of the loading frame was 0.14 mm at 16 kg of load.
A new design for a loading device intended to move the cable displacement transducer between the aluminum-made fixed plate and axial slide improved the displacement error. (Fig 4.3)
50
Fig 4.3 The position of the cable displacement transducer in the loading device was modified in order to improve the displacement error.
4.2 Load-displacement data from the loading device and MTS
No significant differences in displacement measurements from the loading device and MTS were found at a 16 kg load or in the stiffness at the terminal linear region. However, significant differences were found in the stiffness at the neutral zone as well as the toe region at the initial loading stage. The sample rate in the loading device (every 650 ms) was less than 1/10 that of the MTS (20Hz). The sampling data from the loading device was too small to affect the stiffness in the initial stage due to the fact that the number of regression points was too small. The distraction velocity of the MTS was fixed at 0.2mm/s while that of the pneumatic cylinder was 0.6mm/s at the initial stage and 0.01mm/s at the terminal stage.
Therefore, differences in distraction velocity between the MTS and the custom made loading device might have accounted for the difference in the stiffness parameters measured. The loading rate was different between the loading device and MTS units. Therefore, the stiffness was affected by the creeping and stress relaxation of the capsule on the MCP joint.
51
4.3 Load-displacement curve in intact and vented conditions
After vented the MCP joint and eliminating the negative IAP, the displacements at the cross-point as well as at a load of 16 kg were significantly increased by 16% and 13%, respectively. This result indicated that negative IAP affects the stability of the MCP joint during distraction. However, the displacement difference between the 16 kg of load and the cross-point were not statistically significant between the intact and vented MCP joints. This finding suggested that the influence of IAP is mostly on the initial phase of the distraction, before the cross point. This was further confirmed by the insignificant difference found in the terminal region between the intact and vented MCP joints. From a mechanical point of view, negative IAP and the capsule structure of the MCP joint are the major stabilizers during joint distraction. Both contribute to the passive stability of the MCP joint. Based on the results of the present study, it is suggested that the stability of the MCP joint under a distraction load is provided primarily by the joint capsule at the terminal linear region while IAP contributes to the initial stage.
4.4 The correlation of IAP and load, displacement and necking volume
The IAP of the MCP joint with respect to the atmospheric pressure has been investigated on the MCP joint in a previous study (Gaffney et al., 1995). Their results showed that the IAP was near or below atmospheric pressure in an in vivo MCP joint at rest, which is similar to the results of the present study. A high linear correlation between IAP and load was reported on the shoulder by Itoi et al. in 1993.
However, in this study, a non-linear curve relationship between the load and the IAP
52
were found in all specimens, while more linear patterns were presented in the IAP displacement curves. This might be due to the load size as well as to the limited joint volume effect. According to the limited joint volume effect, a relative negative pressure will be induced in the joint cavity by a joint distraction load. The displacement affects the volume of the joint cavity directly. Therefore, in this study, the IAP-displacement curve presented a near linear pattern.
It is clearly shown in Fig 3.5 that the IAP drop before the cross-point exploited only 5% to 10% of the total distraction load, consuming 45% to 80% of the total IAP drop. This matched well with the load-displacement curves, indicating that stiffness increased as load increased, as shown in Fig 2.11. Itoi et al. studied the shoulder joint, and the observed relationship between the load and IAP was found to be almost linear. This might be due to the fact that the applied loading was still before the cross-point. The stiffness of a joint depends on its size and geometry as well as the material properties and the laxity of the capsule. The IAP might not be the major stabilizer in terms of the load it can withstand. This is the reason why we excluded many tissues surrounding the MCP joint that are not directly attached on the bone or capsule of the MCP joint, in order to focus on the IAP response.
However, different tissues and mechanisms play different roles in the stabilization of the MCP joint. The maximum load to bear should not be the only index. Most soft tissues play major mechanical roles in the late stage, that is, after a certain amount of loading or displacement. However, the IAP contributes to joint stability in the early stage of the distraction by reducing the amount of stretching, which might be harmful to other tissues connecting the joint if it is overstretched. This research provided fundamental, if not complete, information about the mechanical role of IAP in the MCP joint.
53
Our results indicated that the correlation between the necking volume and IAP (Fig 3.7, r=-0.882) is higher than that between the displacement and IAP (Fig 3.8, r=-0.838). A virtual vented volume was defined in this study to replace the actual vented volume. As the joint is vented, air flows into the joint cavity, and the IAP will equilibrate with the atmospheric pressure. Therefore, the necking phenomenon does not exist in the vented condition. The changes in joint volume before and after vented are the result of losing negative IAP. The space allowed within the Micro CT scanner is too small to allow space for the instruments necessary for the measurement of the IAP. In addition, a metal needle is not suitable for insertion into the joint cavity while scanning to evaluate volume changes in the vented condition due to the metal scattering in CT images. We suppose that the virtual vented volume is similar to the real volume in the vented condition. The necking volume is the difference between the virtual vented volume and total volume, and the necking volume is the effect of the negative IAP on the joint capsule during MCP joint distraction.
The change in necking volume and IAP is highly correlated in this study.
Change in necking volume may be used to predict IAP for the MCP joint in the future. Although change in the necking volume and IAP are highly correlated, it is necessary to provide proof that change in the necking volume is a direct result of the negative IAP. Future efforts should be devoted to measuring the changes in vented volume and to conducting finite element analyses to investigate the effect of IAP on changes in joint volume.
The total volume was the volume of the total soft tissue of the capsule. The results showed the relation between the total volume and IAP in each specimen to be highly linear (r = -0.960 ~ -0.994) although the relation between the total volume
54
and IAP in all specimens was lower than the relationships between IAP and other parameters. This might be due to the fact that the limited joint volume effect is a factor that affect IAP in each specimen, as well as the fact that the materials, anatomic morphology, interaction between the synovial fluid and the capsule, and physiological mechanisms also affect the IAP of the MCP joint.
4.5 Finite element analysis
Although significant effort was devoted to formulating the detailed geometry of the joint capsule in the FEA model constructed for the present study, the capsular geometry of the MCP joint adopted was still very different from the actual capsule structure of the MCP joint. The results for the distracted distance and necking phenomena were similar to the FEA results with and without the negative IAP.
Changes in joint volume obtained from FEA were less than those measured from of the Micro CT images. Future studies will improve in regard to the detailed geometry and material properties of different portions of the capsule in FE models in order to present a more realistic model for the MCP joint.
4.6 Activities of daily living with MCP joint distraction
In previous studies, the biomechanical parameters were investigated in other movements (Howe et al., 1985; Minami et al., 1985; P ylios and Shepherd, 2007;
Randall et al., 1992; Yung et al., 1986). The major activities of the MCP joint were flexion, extension or lateral deviations. However, the MCP joint was distracted in climbing, pulling a horizontal bar and carrying a heavy bag. If the negative IAP did not contribute to the stability of MCP joint in the vented condition, loosening might
55
be observed on the MCP joint. The other soft tissue would provide more effort to stabilize the MCP joint. Overloading an MCP joint might cause further harm to the soft tissue surrounding the MCP joint. Therefore, a distracted load was applied on the MCP joint in this study. Habitually cracking the MCP joint is another problem in certain persons. Although some investigations have reported that this cracking is not a risk factor for hand osteoarthritis (Castellanos and Axelrod, 1990; Deweber et al., 2011), another study reported that habitually cracking joints was observed to cause ligamentous ossification and chondrocalcinosis in the MCP joint (Watson et al., 1989). This is another phenomenon is worthy of investigation. However, no cracking sound was detected during the distraction in the present study. Further studies will be executed to investigate this phenomenon.
4.7 Limitations
First, the number of specimens was too small even though significant differences were detected with regard to some parameters. Second, the specimens were dissected, preserving only the capsule of the MCP joint. Therefore, the magnitude of parameters assessed in the present study might differ from those of an
First, the number of specimens was too small even though significant differences were detected with regard to some parameters. Second, the specimens were dissected, preserving only the capsule of the MCP joint. Therefore, the magnitude of parameters assessed in the present study might differ from those of an