Changes in Local Bone Mineral Density in Patients With Rupture of the Anterior Cruciate Ligament and Its Guiding Signicance for Treatment

Background:It is recognized that there are many factors that affect the outcomes of anterior cruciate ligament(ACL) reconstruction. However, there were few studies on the effect of local bone conditions on the reconstruction of the ACL. This study aimed to summarize the changes in local bone mineral density (BMD) of the knee after rupture of the ACL and to guide the treatment. Methods: Eighty patients with ACL rupture treated in our department from January 2017 to April 2019 were routinely measured for local bone mineral density before surgery, and according to the bone mineral density, the appropriate method of ligament reconstruction and xation under arthroscopy was selected: if the local bone mineral density of the affected knee was not signicantly lower than that of the healthy side, squeeze xation was used, and suspension xation was used when the local bone mineral density of the affected knee was lower than that of the healthy knee. The conditions of tunnel cutting or screw splitting and tunnel enlargement or screw pull-out were observed during the operation, and the xation mode was adjusted in time according to the situation. The function of the knee joint was evaluated regularly by physical examination, imaging data, the IKDC scale and the Lysholm score table after the operation. Results:A total of 80 patients with unilateral ACL rupture were included in this group. There were 44 males and 36 females. Sixty-eight patients had decreased bone mineral density in the affected knee. The bone mineral density of patients with a history of more than 3 months was lower than that of patients with a history of less than 3 months. Tunnel enlargement and screw pull-out occurred in 2 patients, screw splitting occurred in 1 patient, and no adverse conditions occurred in the rest of the patients. In 3 patients, the xation mode was adjusted in time during the operation. The patients were followed for 12 months (mean 20.65±5.12 months). The IKDC score increased from 43.07±2.66 before the operation to 89.17±3.28 at the nal follow-up, and the Lysholm score increased from 43.49±2.38 to 89.67±2.97. Conclusions:The measurement of local bone mineral density before surgery is of guiding signicance for the selection of reconstruction and xation of the ACL. It is recommended that patients undergo surgical reconstruction within 3 months after injury. When the bone mineral density of the affected knee decreases signicantly to at least 70% lower than that of the healthy side, suspension xation is recommended and the brace xation time is prolonged.


Introduction
Anterior cruciate ligament (ACL) rupture is one of the most common injuries in sports medicine. With the development of minimally invasive concepts and the continuous maturity of arthroscopic technology, arthroscopic ACL reconstruction has become the main method for the treatment of this disease. However, there are still some patients who do not achieved very good results, and even need to undergo a second revision operation. How to reduce the failure rate of ACL reconstruction surgery is a hot topic in sports medicine at home and abroad. At present, it is recognized that the factors that affect the outcomes of anterior cruciate ligament reconstruction include the choice of graft [1] , the choice of xation mode, the location of the bone tunnel and so on. However, there are few studies on the effect of local bone conditions on the reconstruction of the ACL [2,3] . Some patients with abnormal bone around the knee joint will undergo screw splitting and pull-out during the operation [4] . Since January 2017, the local bone mineral density (BMD) of the affected knee has been measured routinely in patients with ACL rupture in our department, and the corresponding xation methods have been selected according to the bone conditions of the patients, achieving good results reported as follows.
1 Data And Methods

General information
From January 2017 to April 2019, 80 patients with ACL rupture were treated surgically in our department:44 males and 36 females. The age ranged from 19 to 58 years old, with an average of 37.68 ± 9.13 years old. The time of injury ranged from 3 days to 12 months. There were 56 patients with a history of 3 months, 31 males and 25 females, with an average age of 37.70 ± 9.47 years. There were 24 patients with a history of more than 3 months, 13 males and 11 females, with an average age of 37.63 ± 8.46 years. All patients were injured in one knee:27 cases on the left and 53 cases on the right. X-ray lm, magnetic resonance imaging and bone mineral density were measured in both knees routinely after admission. Case selection criteria [5] : 1. only ACL rupture (including meniscus tear) was found, excluding posterior cruciate ligament and collateral ligament rupture; 2. there was no history of knee joint trauma or surgery; 3. there was joint instability and a positive Lachman test or anterior drawer test; and 4. cases of primary osteoporosis (postmenopausal osteoporosis and senile osteoporosis) were excluded.
There is a standardized database reference for the bone mineral density T value of the lumbar vertebrae and hip joint, but there is no database comparison for the local bone mineral density of the knee joint. In this study, we compared the bone mineral density (g/cm 2 ) of the affected knee joint with that of the healthy knee joint [6] . The results were statistically analysed by paired t tests. The measuring instrument was American Norland XR-36 dual energy X-ray absorptiometry.

Selection of operative methods
In this study, patients were grouped according to the degree of decrease in local bone mineral density before the operation: the density of the affected knee was not signi cantly lower than that of the healthy side (normal bone mineral density group), the local bone mineral density of the affected knee was lower than 70% of the healthy side (low bone mineral density group), and the local bone mineral density of the affected knee decreased but not by more than 30% of the healthy side. If tunnel cutting or screw splitting and tunnel enlargement or screw pull-out occurred during the operation, the xation mode was adjusted in time according to the situation: an additional plate with a loop was used for suspension xation at the outer entrance of the tunnel, and if the tunnel was obviously enlarged, it was necessary to carry out arti cial bone grafting [7] .

Extrusion Fixation Method
After successful anaesthesia, the patient was placed in the supine position, and the affected limb was placed in the 90° exion position. Routine disinfection and towel laying were performed. An anterior medial L-shaped incision was made under the tibial tubercle of the knee, the semitendinosus tendon and gracilis tendon were cut, the muscle was removed, and the tendon line was woven at both ends. At the level of the knee joint space, an opening of 8 mm was made on both sides of the patellar ligament to the articular cavity, which was disposed into the arthroscope from the lateral entrance, additionally, the intercondylar fossa was cleared, and the scar tissue and residual ACL were removed. The tibial end of the ACL was located with a guide needle, and the tibia of the corresponding diameter was drilled to open the tibial tunnel. Then, the femoral footprint of the ACL was located with a guide needle, and the femur drill was used to open the femoral tunnel by approximately 35 mm. The braided tendon was introduced with the guide needle, and the knee was exed by 20°. Reverse traction was implemented at both ends of the femur and tibia to maintain a certain tension, and passive movement of the affected knee was carried out to provide appropriate tension of the tendon. The femoral tunnel was xed with a 25 mm long polylactic acid hydroxyapatite screw. The tibial tunnel was xed with a tibial xation wing and a 25 mm long polyetheretherketone screw (both screws and wings were produced by Smith & Nephew). The knee was moved passively, and the anterior drawer test was performed. If the knee joint was stable and the autologous tendon tension was moderate, then suturing and pressure bandaging were performed. If screw splitting or tendon cutting occurred during the operation, the mode of xation was adjusted to suspension xation. If the tunnel was enlarged or screws were pulled out due to osteoporosis during the operation, adjustment to suspension xation could be carried out. Alternatively, on the basis of extrusion xation, a loop steel plate was added to the outer entrance of the tunnel to carry out suspension xation.
If the tunnel was obviously enlarged, it was necessary to carry out arti cial bone grafting [8] .

Suspension Fixation Method
Harvesting of the transplanted tendon and the cleaning process of the joint cavity were carried out in the same manner as the extrusion xation method. After the joint cavity was cleaned, the guide needle was used to locate the tibial attachment of the ACL, and the tibia was drilled to expand the tibial tunnel. Then, the femoral footprint of the ACL was located, and the femur was drilled and enlarged by approximately 35 mm in a long tunnel. The guide needle was introduced into the autologous tendon, which was pulled at both ends of the tunnel to provide a certain tension; passive movement of the affected knee was carried out, providing the autologous tendon with the appropriate tension. A plate with a loop (Smith & Nephew) was used, the loop was cut, the tendon traction line was passed through both of the plate holes outside the thigh, and the plate was pushed forward to make sure the plate was adhered to the femoral cortex; multiple knots were made on the surface of the plate so that the plate was rmly stuck at the outer entrance of the femoral tunnel to achieve suspension xation. The outer entrance of the tibial tunnel was also suspended and xed with a plate with a loop. The affected knee was moved passively again, and the anterior drawer test was performed. If the knee was stable and the tension of the transplanted tendon was moderate, the knee was sutured and bandaged under pressure.

Postoperative Management
After the operation, a local cold compress was applied, the affected limb was raised, hinge braces were worn, and isometric contraction training of the quadriceps femoris and hamstring muscles was performed every day. One to two weeks after the operation, the affected limb was locked in a completely straight position, and the use of double crutches was supported until complete loading. Passive exion and extension could be performed daily starting 3 to 4 weeks after the operation, reaching knee exion ≥ 120°. Five to eight weeks after the operation, the affected knee joint could be actively moved to reach 90°o f knee exion, and semi-squatting exercises could be used to improve muscle strength. Starting at the 9th week after the operation, forward uniform speed running and backward running could be carried out.

Postoperative Follow-up And Evaluation Of The Curative Effect
The patients were followed 1 month, 2 months, 3 months, 6 months and 12 months after the operation.
All patients were examined by the anterior drawer test and Lachman test. Patients with suspension xation were re-examined by X-ray, and squeeze xation did not need to be photographed because the screw was not visible. The function of the knee joint was evaluated by the IKDC scale score and Lysholm knee joint scale score.

Statistical Processing
The data of this study were analysed by the statistical software SPSS 19.0, and the measurement data were expressed as the mean ± standard deviation. Paired t-tests were used to compare the bone mineral density of both knee joints and the functional score of the affected knee before and after the operation. Group t-tests were used to compare the decreasing degree of bone mineral density within 3 months and more than 3 months. P < 0.05 was considered statistically signi cant.

Results
The measurement of local bone mineral density of the knee joint in 80 patients showed that 68 patients had varying degrees of bone mass loss relative to the healthy side. The BMD of the contralateral knee joint was 1.256 ± 0.153 g/cm 2 , and that of the affected knee was 0.927 ± 0.283 g/cm 2 ; the difference was statistically signi cant (Table 1). There was no signi cant difference in age or sex between patients with a history of less than 3 months and those with a history of more than 3 months.According to the degree of decrease in bone mineral density, the local bone mineral density of the affected knee was less than 70% of the contralateral value in 28 patients and higher than 70% of the contralateral value in 40 patients. The longer the time after injury, the more obvious the decrease was in relative bone mineral density. In this group of patients, the BMD of the affected knee was 0.956 ± 0.217 g/cm 2 for those with a history of less than 3 months and was 0.914 ± 0.176 g/cm 2 for those with a history of more than 3 months. The difference was statistically signi cant ( Table 2). Table 1 Comparison of local BMD of bilateral knee joints(g/cm 2 ) Local BMD All patients were followed for 12 to 30 months (mean 20.65 ± 5.12 months). During the last follow-up, the symptoms of knee instability disappeared in all patients before the operation, 5 patients complained of mild knee pain, and physical examination showed that the results of the Lachman test and anterior drawer test were negative. At the last follow-up, the IKDC score of the affected knee was 89.17 ± 3.28, and the Lysholm knee score was 89.67 ± 2.97, which were signi cantly higher scores than those before the operation, and the difference was statistically signi cant ( Table 3). The range of motion and muscle function of the knee joints of all patients returned to normal. In the normal bone mineral density group, tunnel enlargement and screw pull-out occurred in 2 patients, and screw splitting occurred in 1 patient.
Among them, the patients with screw splitting and tunnel enlargement were adjusted to suspension xation in time during the operation, and the other patient with tunnel enlargement was xed by suspension with a plate with a loop on the basis of extrusion xation. All 3 patients recovered well after follow-up.

Discussion
The anterior cruciate ligament (ACL) is a statically stable structure of the knee joint that plays an important role in the stability of the knee joint. Once injured, it is di cult to heal and can even lead to progressive meniscus and articular cartilage wear, so it is usually necessary to reconstruct the ligament under arthroscopy. ACL attachment xation is the most important link in reconstruction surgery [9] , and most reconstruction failures are caused by unstable attachment xation [10] . Vopat et al. [11] believed that the failure of ACL reconstruction often occurs on the xed side, not on the transplanted tendon itself.
Through their experimental study, they believed that reliable postoperative immediate stability could be obtained by interfacial screw extrusion. Additionally, the operation of interface screw xation is relatively simple, as long as it is screwed into the tunnel along the guide needle. Therefore, in the past, when dealing with this kind of disease, our department usually chose the method of extrusion xation and achieved good results. However, with the increase in the number of surgical cases, we found that a small number of patients had symptoms of ligament relaxation and joint instability after the operation, especially in elderly women and patients with old injuries. Therefore, we inferred that the local bone mass of the affected knee decreased due to postmenopausal osteoporosis and disuse bone atrophy, thus reducing the holding force of the screw [12,13] . Therefore, starting from January 2017, we routinely measured the local bone mineral density of both knees and took an X-ray of the affected knee in the patients with ACL rupture in our department before surgery, evaluated the bone condition of the affected knee, selected the ACL attachment xation method according to the decreasing degree of bone mineral density, and achieved good results.
For ACL rupture, the common attachment xation methods in our department include extrusion xation and suspension xation [14] . For suspension xation with plates, it was previously believed that because it was far away from the normal ACL anatomical attachment, the longitudinal movement between the autogenous tendon and the bone canal could produce a "bungee jumping" effect, and the transverse movement could produce a "wiper" effect, which led to the enlargement of the bone tunnel, and synovial uid immersion would also affect the tendon-bone healing. However, through interface screw extrusion xation, we can operate under direct vision, according to hand feeling and mirror observation, and this method is more rm and reliable. The polylactide hydroxyapatite screw does not need to be removed for the second time and does not affect the magnetic resonance examination. Therefore, in general, we usually prefer the extrusion xation method. However, extrusion xation is not an omnipotent xation method, and its indication and curative effect are closely related to the local bone condition of the affected knee [15] . Through previous surgical observations, we found that some patients suffered from screw splitting and that the transplanted tendon was cut by the edge of the bone tunnel when squeezing the screw. In addition, some patients suffered from screw loosening or even pull-out and bone tunnel enlargement when the knee was passively moved during the operation [16] . Therefore, there are still some limitations in extrusion xation. During the operation, once a bad condition occurs during extrusion xation, immediate adjustment to the suspension xation mode should be carried out. When the plate with a loop is used for suspension xation, the attachment crosses the outer entrance of the bone tunnel, and there is no need to consider the effect of bone condition, so it can be used for patients with osteoporosis and osteosclerosis. The force provided by suspension xation and the contractile force of the transplanted tendon belong to a pair of acting forces and reaction forces, which are kept in a straight line in the opposite direction and have a strong anti-pull-out ability. The stress is concentrated between the plate and the outer entrance of the tunnel, and the suspension force is positively correlated with the contact area between them.
For some patients, especially those with old ACL fractures (course of disease > 3 months), rm xation with screws is usually di cult because of the local osteoporosis during the operation. Flexion and extension of the affected knee several times during the operation can lead to screw loosening, tendon relaxation and, nally, suspension xation. If the preoperative equipment is not fully prepared and the xed mode cannot be adjusted in time, bene ts can be lost and concerns will ensue.
Preoperative bone mineral density examination, combined with X-ray lm, can help doctors more accurately understand the bone condition around the knee joint and guide the preoperative preparation and selection of knee joint functional reconstruction. If the bone mineral density of patients is different, the appropriate individual xation method will accordingly be adopted. For patients with old injuries, due to a long-term lack of exercise, there is usually severe osteoporosis in the affected knee, so we can guide them to carry out functional exercises in the outpatient clinic, provide anti-osteoporosis treatment with drugs, after a period of preparation, and then perform the reconstructive operation in order to reduce the risk of extrusion xation failure. Even if the osteoporosis of the patient does not improve and the local bone mineral density of the affected knee is always less than 70% of the contralateral value, we can still complete ACL reconstruction by suspension xation. However, the time of external xation must be extended appropriately to ensure the curative effect.
We are still constantly evaluating how much the reduction of bone mineral density is suitable for each xation method due to the lack of clinical data. In addition, this was not a randomized controlled study, and the intraoperative xation method was changed according to the intraoperative conditions, which requires a certain degree of experience. According to the experiences and lessons of our department in recent years, we believe that for those with normal bone, extrusion xation should be selected.
Suspension xation should be used for those whose local bone mineral density is lower than 70% of the healthy side. For those whose local bone mineral density of the affected knee is decreased but not lower than 70% of the healthy side, extrusion xation is preferred. We suggest that anterior cruciate ligament reconstruction should be carried out within 3 months after injury to avoid a further decrease in the bone mass of the affected knee and to avoid affecting the outcomes of xation. Of course, even according to the above criteria, there will still be cases of xation failure in individual patients, so for each patient, we must consider the operation plan as a whole, including the patient's age, sex, past history, injury time, and so on. Only in this way can we maximize the effectiveness of the operation.