Biomechanical evaluation of native acromioclavicular joint ligaments and two reconstruction techniques in the presence of the sternoclavicular joint: A cadaver study

Introduction

From its physical appearance, the acromioclavicular (AC) joint seems to have a simple structure; however, a growing amount of laboratory-based research shows that its biomechanics are not yet fully understood. ¹ AC dislocation usually occurs as a result of direct force to the acromion with the arm in an adducted position. ² AC dislocation injuries are generally classified into six different types according to Rockwood. ^3,4 Most authors agree that Rockwood types I and II should be treated conservatively, and types IV–VI should be treated operatively. ^5,6 However, treatment of Rockwood type III injury remains controversial; recent literature advocates a conservative approach, with surgical reconstruction only conducted in particular cases. ^1,7 Despite the description of numerous techniques, treatment of AC joint dislocation remains controversial. ^{1,5,8 –10}

There have been a number of biomechanical studies conducted on cadaver shoulders. ^{6 –8,11 –22} In all previous studies, the scapula and clavicle were positioned anatomically, but the sternoclavicular (SC) joint was removed. The SC joint provides movement around the sagittal axis; during full elevation of the upper limb, the clavicle is raised to an angle of approximately 60°. ²³ It is not clear whether the presence of the SC joint influences the biomechanical properties of native AC ligaments and reconstruction techniques.

The purpose of the present study was to investigate the biomechanical properties of native AC ligaments and AC joint reconstruction techniques in cadavers with the SC joint still present. Our hypotheses were that (1) native AC ligaments would have better biomechanical performance than both AC joint reconstruction techniques and (2) the single coracoclavicular loop (SCL) technique would have better biomechanical properties than the two coracoclavicular loops (TCL) technique.

Materials and methods

Surgical techniques

Native ligaments group

Native AC ligaments.

SCL group

Two 2.5 mm drill holes were made; the first was 4.5 cm from the distal end of the clavicle in its anterior half, and the second was 2.5 cm distal from the first drill hole in an oblique direction in the posterior half of the clavicle. The position of the drill holes imitated the normal attachment points of the CC ligaments. ⁶ Double no. 2 polyester suture (Atramat, International Farmaceutica, Mexico City, Mexico) was passed under the coracoid process through the drill holes in the clavicle. A knot was made on the superior surface of the clavicle. Next, a 2-mm diameter hole was drilled horizontally from ventral to dorsal in the acromion and the sternal end of the clavicle. To secure the AC joint, no. 2 polyester suture was passed through these holes, and an additional figure of eight-shaped configuration suture was made (Figure 1).

OPEN IN VIEWER

Figure 1.

Schematic drawing of the single coracoclavicular loop technique (front view): (1) clavicle, (2) coracoid process, (3) acromion process, and (4) knots.

TCL group

The same drill holes were used for the SCL technique. This technique was performed similarly to previously described polydioxanone suture (PDS) reconstruction with additional AC cerclage ¹⁰ ; we modified this technique using polyester suture instead of PDS. The CC ligaments were repaired using two no. 2 polyester cords. Each of the cords was guided around the base of the coracoid and then passed through one of the drill holes in the clavicle. A figure of eight-shaped AC suture was then made as described previously (Figure 2). All sutures were tied by the same experienced surgeon.

OPEN IN VIEWER

Figure 2.

Schematic drawing of the two coracoclavicular loops technique (front view): (1) clavicle, (2) coracoid process, (3) acromion process, and (4) knots.

Biomechanical testing

A custom-built testing device was used (Figure 3). This device had three main parts: (1) the frame with the linear actuator attached to it (the frame was rigidly connected to the autopsy table, and the linear actuator was connected to the acromial end of the clavicle by a 2-mm diameter cable); (2) the plate for fixing the scapula to the autopsy table; and (3) the processor for linear actuator control and data acquirement from the force sensor (at a rate of 9.6 kHz). The force sensor was set in between the linear actuator and the clavicle. Digital calipers (Vagner SDH, Slovenia) with an accuracy of 0.01 mm were used to measure the distance between two points on the linear actuator: one point on the moving part and the other on the stable part. The change in the distance between these points was considered as elongation of the structures between the clavicle and the scapula.

OPEN IN VIEWER

Figure 3.

Biomechanical testing setup: (1) processor, (2) frame, (3) linear actuator, and (4) specimen.

Each specimen underwent three tests: one in its native state and one for each of the two different reconstruction techniques. After native state test, AC and CC ligaments were dissected. All specimens were preconditioned from 0 to 70 N for one cycle in a superior direction to eliminate creep phenomenon. The testing protocol included static and cyclic loading. In the static part, there were two measurements: the first was taken at a load of 10 N and the second at 70 N. To achieve equal tension of the structures, each load was applied for 5 s before measurement. The difference between the first and the second measurements was defined as translation in the superior direction. Immediately after the static testing, cyclic loading was started for 1000 cycles from 20 to 70 N at a rate of 1 Hz. The third measurement was made after the last cycle, at the peak of 70 N and 5 s hold. The difference between the second and the third measurements was defined as the translation increment.

Results

Native ligaments group

Native ligaments had a translation of 5.13 ± 1.17 mm and a translation increment of 0.74 ± 0.23 mm (Figure 4).

OPEN IN VIEWER

Figure 4.

Translation in the superior direction. *A statistically significant difference. SCL: single coracoclavicular loop; TCL: two coracoclavicular loops.

TCL group

The TCL failed during static loading in one specimen. The pattern of failure was a tear of one of the CC slings, followed by a tear of the other (it was not recorded which of the CC slings failed first) and complete cutoff of the figure of eight-shaped AC suture from the acromial end of the clavicle. The final sample size in the TCL group was therefore seven. TCL reconstruction had a translation of 8.33 ± 3.30 mm and a translation increment of 1.19 ± 1.07 mm.

Translation was significantly lower in the native ligaments group than in the SCL (p = 0.023) and the TCL (p = 0.046) groups. Translation increment was significantly higher than in the SCL group than the native ligaments group (p = 0.028), and there was no significant difference between the native ligaments group and the TCL group (p = 0.342; Figure 5). There was no significant difference between the SCL and the TCL group in terms of translation (p = 0.865) and translation increment (p = 0.113).

OPEN IN VIEWER

Figure 5.

Translation increment. *A statistically significant difference. SCL: single coracoclavicular loop, TCL: two coracoclavicular loops.

Discussion

To the best of our knowledge, this is the first report on biomechanical testing of the AC joint and its reconstruction techniques in the presence of the SC joint. The most important finding was that the presence of the SC joint did not have a measurable influence on testing data. Native AC ligaments had better static properties than both reconstruction techniques and worse dynamic biomechanical properties than the SCL technique. The SCL technique appeared to be more secure than the TCL technique.

In the literature, the average superior translation of native AC ligaments under 70 N loading varies from 2 to 8 mm, which is consistent with the 5.13 ± 1.17 mm found in the present study (Table 1). However, in both reconstruction groups, we observed a slight cutoff in the transverse tunnel of the figure of eight-shaped suture in the acromial end of the clavicle. This resulted in increased translation, and we did not have a method to measure its influence on the final data. We propose two possible reasons for this cutoff: (1) the acromial end of the clavicle mainly consists of trabecular bone, and no. 2 polyester suture might not have had enough surface contact with the bone. This size suture should be considered insufficient for use in the acromial end of the clavicle. (2) The SC joint provides motion in the sagittal axis; therefore, the acromial end has higher superior translation than the more proximal parts of the clavicle under loading. This might result in additional stresses on the AC site, which may help to explain the cutoff and the significantly higher translation that the reconstruction techniques had compared with native AC ligaments. It is also possible that the SC joint has an influence on the biomechanical behavior of native AC ligaments and AC reconstruction techniques, but our testing setup failed to detect this. It should be taken into account that specimens in this study had a mean age of 76 years and eight-shaped suture might be more beneficial in young persons with better quality of trabecular bone to maintain horizontal stability.

OPEN IN VIEWER

Table 1.

Review of the literature.

Study	Number of specimens	Displacement under 70 N (mm)
Mazzocca et al. 6	14	6.14 ± 3.70
	14	5.63 ± 2.14
	14	4.38 ± 2.45
Beitzel et al. 13	18	4.28 ± 1.81
Beitzel et al. 8	12	3.8 ± 1.48
Beitzel et al. 12	18	From 2 to 8
Present study	8	5.13 ± 1.17

The purpose of the present study was to investigate the biomechanical performance of native AC ligaments and AC joint reconstruction techniques. The results are partially in agreement with our first hypothesis; native AC ligaments had better static biomechanical properties, but they showed inferior cyclic results compared with the SCL group, and there was no difference compared with TCL group. The second hypothesis is rejected, as there was no significant difference between SCL and TCL reconstruction in terms of translation and translation increment. Although SCL tended to have higher estimates than TCL in cyclic loading, this difference was not statistically significant. However, TCL had higher variance around the mean in translation and translation increment than SCL, and TCL reconstruction failed during static loading in one specimen. This might be because both CC loops were knotted separately, and it was impossible to pre-tension them equally; one of the loops was more tensioned than the other, and thus sustained higher stresses under loading. Because of these abovementioned circumstances, SCL appeared to be more secure than TCL.

We chose 70 N as the maximum magnitude loading in this study because of the assumption that the scapula (coracoid process and acromion) and clavicle are rigid bodies with ligamentous structures between them, and this 70 N loading simulates arm weight. ^2,8,12,13,24 The prescribed cyclic testing protocol covers at least 10% physiological load situation in the early postoperative rehabilitation period. ^2,24 In the literature, under the same cyclic loading protocol, synthetic materials are reported to have superior cyclic biomechanical properties compared with biological material. ²² Our results confirm these previous findings; the native AC ligaments had a significantly higher translation increment than the SCL reconstruction technique.

One of the limitations of the present study is that each specimen was used for three tests. This iterative loading might have influenced the ligaments of the SC joint and caused additional translation increment in the group that was tested last. The second limitation is that we did not measure bone mineral density; however, two experienced surgeons visually inspected all specimens for observable diseases and cortical bone quality.