In regard to the ACL reconstruction using interference screws, there are still some concerns, such as: risk of early graft fixation failure, slippage, and laceration, which need to be addressed. Considering that an intact ACL experiences a gradual increase in stiffness as it gets closer to the point of insertion into the bone [29], it was hypothesized here that by increasing the slope of the interference screw, and thus mimicking a natural ACL structure, the stability of fixation will increase. In order to check the validity of the hypothesis, two custom-made metallic interference screws were designed and fabricated, i.e. lower slope tapered interference screw (LSTIS), and higher slope tapered interference screw (HSTIS), and performance of the fabricated screws were compared through experimental tests on graft-bone-interference screw constructs. The diameters of both screws in one third of their length, from the tip of the screw, was equal to the bone tunnel diameter, and they were gradually increased to 8.5 mm (in LSTIS), and 9.5 mm (in HSTIS), i.e. 1 and 2 mm greater than bone tunnel diameter (Fig. 1). Thus, in LSTIS group, compared to HSTIS group, the average of screw diameter in one third of its length from tip (Fig. 1, region A) was larger, and in the rest of its length (Fig. 1, regions B and C) was smaller.
To compare the capability of HSTIS and LSTIS in improving the initial stability of reconstructed anterior cruciate ligament, stiffness of each graft-bone-screw construct was measured through applying a sub-failure incremental cyclic loading. Results of this work proved superiority of HSTIS group, in terms of the graft-bone-screw stiffness, compared to LSTIS group, especially in cyclic loading with the peak values of: 100, 150 N (Fig. 4a). Stiffness of intact femur-ACL-tibia complex of human cadaver knee, under approximately similar incremental cyclic loading protocol, was measured in Scheffler et al.' study [22]. Their reported mean stiffness of the samples’ constructs at cycles with peak values of: 100, 200, and 300 N, were 43.8, 76.3, and 92.6 N/mm, respectively [22]. The mean stiffness values of the bone-graft-screw stiffness in HSTIS group of this work, for the same cycles as the ones used in Scheffler et al.' study, were 40.73, 68.99, and 102.24 N/mm. These values are closer to those of intact ACL in Scheffler et al.' study [22], compared with the LSTIS group, with mean stiffness of 27.82, 66.05 and 100.9 N/mm, respectively (Fig. 4a). Thus, it seems that fixation of the grafts with HSTIS better bio-mimicked intact ACL function, compared with LSTIS.
Graf laxity is another important concern associated with the ACL reconstruction when employing interference screws. The graft laxity can be caused by graft's fiber damages, and/or graft slippage from the bone tunnel, without including the elongation of the tendon graft itself. The graft laxities found in this work for both groups were initiated by loads that were well below the failure load (see Fig. 4b), and showed an increase when the peak load increased (see Fig. 4b). Moreover, mean graft laxity measured for graft fixed with HSTIS was less than that of LSTIS, especially at loading cycle with peak values of: 100 and 150 N, in which the difference between LSTIS and HSTIS laxities were significant (p < 0.05) (Fig. 4b). This observation regarding the graft laxity indicates that different body slopes of the screw will likely lead to different performance of the reconstructed ACL in early stage of rehabilitation. Nonetheless, results of this work showed that there is insignificance difference between graft laxities in two groups for the load greater than 150N.These insignificance differences may be due to the reduction of survived samples numbers in cycles with higher peak values, especially in LSTIS group, which can directly affect the statistical analysis’s results.
By comparing the graft laxity parameter in HSTIS group (Fig. 4.b) with those reported in Scheffler et al.’ study, using a similar protocol of loading, one may hold promise for superior behavior of the HSTIS to none-tapered metal interference screws [22]. In their study, the graft laxity was measured in the case of fixing Smith & Nephew RCI interference screw with a diameter of 7 mm and length of 25 mm in the bone tunnels, with diameters ranging from 8 to 9 mm [22]. The graft laxity, in load cycle with a peak value of 200 N, was reported to be 3.0 ± 3.8 mm [22], which is greater than the corresponding value for both HSTIS and LSTIS groups of current study, which were 1.16 ± 0.56 mm 2.49 ± 1.00, respectively. Furthermore, in Miccuci et al.’ study, graft laxities, in the case of fixation by screws with diameters equal to, 1 mm smaller, 1 mm and 2 mm greater than, the bone tunnel, were measured with a video analysis technique, and with photo-reflective markers, while the graft was experiencing a cyclic loading from 50 to 250 N at the frequency of 2 Hz, for a total of 1,500 cycles [3].The least graft slippage in their study was reported to be 2.65 ± 2.38 mm, for the screw with a diameter equals to the bone tunnel [3], which is greater than the graft laxity measured for HSTIS, in cycle with a peak value of 250 N, i.e. 2.54 ± 1.02 mm. However, Miccuci et al.’ results have been reported after applying 1500 cycles of loading, and the reported values for 100 cycle of loads in their study are less than the corresponding values for both HSTIS and LSTIS groups. As a result, preponderance of HSTIS to none-tapered bio-interface screws cannot be claimed in this study and it seems that manufacturing HSTIS with biodegradable materials can be deemed as a good option for improving its mechanical behavior.
Another cause of graft fixation failure in ACL reconstruction surgery through using interference screws, which can be observed clinically, is the graft laceration. In this study, in order to investigate the effect of body slope of the screws on grafts damage, graft mode of failure for each sample was recorded. It was found that the graft and screw engagement in region A was the weakest site in all constructs (Fig. 1). However, the mode of graft failure in HSTIS group was mostly necking of the grafts, and the samples did not fail due to screw threads cuts, which was mostly the mode of failure in LSTIS group (see Fig. 5a). In previous studies, type of graft failure was only determined in terms of graft slippage, deterioration of graft material or failure at the mid-substance of the graft [33, 34]. Thus, due to paucity of the data in current literature on the subject of grafts’ fibers damages, comparison with previous work is not possible here.
Based on the evidence provided in terms of lower stiffness, greater graft’s fibers damages, higher graft laxity and displacement in LSTIS group, compared with HSTIS group (Table 1, Figs. 4 and 5), it can be speculated that, in the ACL reconstruction surgery, through using a tapered interference screw, the risk of early graft fixation failure can be minimized through controlling the pressure and contact area of the screw and graft, by means of precisely determined body slope for the screw. It seems that in region B and C (Fig. 1), smaller average diameter of screw and lower body slope of LSTIS causes less friction, compared to HSTIS, which consequently lead to a greater displacement of the graft in the former group. Subsequently, the slippage of the graft in regions B and C will be transmitted to region A, near to the loading exertion point, due to the direction of the applied load, pulling the graft outside of the bone tunnel (Fig. 2). Finally, it can be deemed that this transmitted graft to region A, in LSTIS group, due to a larger mean diameter of screw (Fig. 1), compared to HSTIS, will be exposed to higher average contact pressure, which leads to a transverse cut of the graft's fibers (see Fig. 5). On the other hand, higher body slop of HSTIS group, in conjunction with a smaller mean diameter of the HSTIS in region A, compared to LSTIS, prevents transverse cut of the graft at the most vulnerable regions of the fixation (Fig. 5).Therefore, it can be suggested that in regions B and C, major slippage of the graft takes place, which could be transferred to hazardous region A that can cause further damages on graft fibers, thus an adequate contact pressure must be applied in region B and C, while high contact pressures should be avoided in region A.
The following points should be taken into consideration while one is trying to interpret results of this work. First, in-vitro tests' results can give us information about the initial stability, but they are unable to evaluate mechanical behavior of the bone-graft-interference screw construct after graft healing and remodeling processes, which can alter graft tissue's mechanical properties [35]. Secondly, stress distribution within the graft and on bone tunnel can have influence on bone tunnel widening during the healing and remodeling processes, and consequently can affect graft fixation stability, which was not taken into account in this study. Thirdly, it should be noted that extensor-digitrom of bovine [36, 37], instead of human hamstring tendon, was used in this investigation. Lastly, synthetic bone, similar to a dense cancellous bone, was used here, in order to avoid cadaver’s wide range variation in BMDs, as well as non-homogeneity of real bone, and thus make the comparison between the LSTIS and HSTIS more logical.