Tossy et al initially classified AC joint injuries into three categories in 1963.
This classification was primarily based on the degree of damage and/or rupture of the acromioclavicular (AC) and coracoclavicular (CC) ligaments.
Rockwood retained these three categories and modified the third category into a further three that were more specific to the degree of damage of the joint.(2) In the following images there is no displacement of the clavicle and no gross abnormalities.
Fig. 1: X-ray normal AP left shoulder
References: Department of Radiology, Royal Brisbane and Women's Hospital / Dr. Jeffrey Hocking
Fig. 2: X-ray normal lateral left shoulder
References: Department of Radiology, Royal Brisbane and Women's Hospital / Dr. Jeffrey Hocking
Fig. 3: X-ray normal axial view left shoulder
References: Department of Radiology, Royal Brisbane and Women's Hospital / Dr Jeffrey Hocking
Fig. 4: X-ray normal Zanca view left shoulder. This view employs a 15 degree cephalic tilt in order to remove overlying structures to improve visualization.
References: Department of Radiology, Royal Brisbane and Women's Hospital
Type I: The AC ligament is strained with no rupture of the CC ligament.
There will be no visible abnormalities on X-ray.
Refer Fig 1-4 to preview what type I injury would appear on plain X-ray.
Type II: The AC ligaments are completely torn with partial rupture of the CC ligaments.
There may be a slight raise in the clavicle on X-ray compared to the opposite side,
however the CC distance is not affected.
(Fig.
5)
Fig. 5: X-ray bilateral left shoulder type II injury.
AC ligament: complete tear.
CC ligament: partial rupture.
(White dotted line demonstrates normal clavicle location)
References: Department of Radiology, Royal Brisbane and Women's Hospital / Dr Jeffrey Hocking
Type III: Both the AC and CC ligaments are completely ruptured.
X-ray will show an elevated distal clavicle with a widening of the CC space by approximately 25-100%.
(Fig 6)
Fig. 6: X-ray bilateral left shoulder type III injury.
AC ligament: complete tear.
CC ligament: complete tear.
Elevated distal clavicle + widening of CC space
(white dotted line demonstrates normal clavicle location)
References: Department of Radiology, Royal Brisbane and Women's Hospital / Dr Jeffrey Hocking
Type IV: Both the AC and CC ligaments are completely ruptured.
X-ray will show the clavicle displaced posteriorly into the trapezius muscle.
Lateral radiograph demonstrates the clavicle overriding the acromion,
which is suggestive of a posterior dislocation (Fig 7-8).(15) However,
there are limitations in the use of plain radiographs for type IV injury.
This is due to the fact that the standard method of confirming a type IV injury is with the axial and/or scapular view on X-ray.
This can be difficult to obtain due to pain limitations in requiring the patient to manoeuvre their shoulder into the correct position for the radiographer.
There are also limitations in the radiograph’s failure to visualise soft tissue damage,
which is a significant feature of this class of injury.
(11) Finally,
a study has shown the propensity for the axillary view to exaggerate the appearance of posterior subluxation of the clavicle in non-pathological shoulders.
Axillary views were found to have high sensitivity but low accuracy in detecting type IV lesions.(16) This is problematic in the sense that this could lead to inappropriate surgical intervention in non-pathological shoulders.
Fig. 7: X-ray bilateral right shoulder type IV injury.
AC ligament: complete rupture.
CC ligament: complete rupture.
Posterior dislocation of clavicle cannot be appreciated on these views.
References: Department of Radiology, Royal Brisbane and Women's Hospital / Dr Jeffrey Hocking
Fig. 8: X-ray lateral/Scapular Y-view right shoulder demonstrating type IV injury.
Clavicle can be seen overriding the acromion. This is suggestive of a posterior clavicle dislocation.(15) Patient was unable to tolerate an axial view.
(White dotted line demonstrates normal clavicle location)
Note: Axial views or CT images are often needed to confirm this diagnosis. The Y-view will often demonstrate an overriding clavicle in other injury grades if the patient is not in a neutral position (e.g. Alexander stress View).
References: Department of Radiology, Royal Brisbane and Women's Hospital / Dr Jeffrey Hocking
Type V: Both the AC and CC ligaments are completely ruptured.
X-ray will show a large superior displacement of the distal clavicle accompanied by an increase of the CC distance by approximately 100-300% compared to the contralateral side.
(Fig 9)
Fig. 9: X-ray bilateral right shoulder type V injury.
AC ligament: complete rupture.
CC ligament: complete rupture.
Superior displacment of clavicle + increase of CC distance by 100-300%
(white dotted line demonstrates normal clavicle location)
References: Department of Radiology, Royal Brisbane and Women's Hospital / Dr Jeffrey Hocking
Type VI: Both the AC and CC ligaments are completely ruptured.
X-ray will show the clavicle displaced into the subacromial or subcoracoid position.
(Fig 10)
Fig. 10: X-ray right shoulder subcoracoid type VI injury
AC ligament: complete rupture
CC ligament: complete rupture
(white dotted line demonstrates normal clavicle location)
References: Modified with permission, Mohammad J. Emami, Ali Haghighat, Zahra Babanezhad, Arash Khozaei, Subcoracoid dislocation of the lateral end of the clavicle: Mechanism of injury, Injury Extra, Volume 41, Issue 5, May 2010, Pages 58-61, ISSN 1572-3461,
As mentioned previously,
the sole use of radiographs for some classes of AC joint injury has come into question.
Gausted et al found that analysis of the vertical plane using Zanca/AP views was accurate and reproducible,
while the horizontal plane did not demonstrate this.
They found that radiographs were more accurate for type II,
III,
V,
and VI injuries but not as useful for type IV as the axillary view,
which is the standard for this type of injury,
is in the horizontal plane.(11) In fact,
Rockwood’s seminal paper in which the classification system is modelled,
analysed 520 X-rays and only found four examples of type IV injuries.
This could suggest that type IV class is under detected using traditional axillary views.(2) This conveys
that adjuncts to plain film such as ultrasound,
CT,
MRI and 4D CT could be vital in properly diagnosing a type IV injury.
The idea of adding additional X-ray views from the standard ones mentioned earlier has also been investigated by some researchers.
Ibrahim et al analysed different views and their efficacy in accurately grading AC joint injury using the Rockwood classification.
They demonstrated that the simple addition of a bilateral view resulted in a type change for a significant amount of his patients.(6) They found that a bilateral view resulted in a type change for 74.5% of patients: 28/56 were upgraded from a I/II/III to a V and 2/3 were downgraded from a V to a II.
The addition of a weighted view upgraded a further 10 patients to a type V.(6)
Furthermore,
research has shown that the use of ultrasound can reveal vital information on the state of the soft tissue involved in the injury (Fig 11).
One study found examples of disruption of the insertions of the deltoid and trapezius muscle in Rockwood classes II,
III,
IV and V (there were no examples of VI).(7) This additional information on the state of the soft tissue of the site around the injury could help allow doctors to decide the management of type III injuries.
Currently,
type III injuries are either conservatively or surgically managed,
with no consensus on the benefits of one over the other.
Fig. 11: Ultrasound shoulder: C= Clavicle, White arrow= insertion of trapezius muscle, small white arrows= common fascia of the deltoid and the trapezius muscle
References: Heers G, Hedtmann A. Correlation of ultrasonographic findings to Tossy's and Rockwood's classification of acromioclavicular joint injuries. Ultrasound in medicine & biology. 2005;31(6):725.
There is limited evidence on the efficacy of analysing AC joint injuries using CT scan.
One study assessed whether the same observer would grade differently when shown the same injury in a different modality; in this case a CT scan compared to an X-ray.
The study found that while there was an improvement in intraobserver reliability with the use of CT,
the results were not statistically significant.(9) It was noted that CT imaging might prove beneficial in assessment of class IV injuries,
as they are more difficult to visualise on plain radiography.
Fig. 12: CT shoulder coronal view demonstrating superior displacement of the distal clavicle
References: Department of Radiology, Royal Brisbane and Women's Hospital
Fig. 13: CT shoulder in axial plane
References: Department of Radiology, Royal Brisbane and Women's Hospital
Fig. 14: CT shoulder in sagittal plane
References: Department of Radiology, Royal Brisbane and Women's Hospital
The use of MRI has been shown to be of value in helping doctors to accurately grade AC joint injury.
One study found that approximately 50% of patients had been classified to have the wrong type of AC joint injury with the use of T1 weighted MRI.
In these cases there were examples of both over-estimating and under-estimating.(8) Perhaps the use of MRI could prove beneficial in determining the management plan of particularly difficult cases.
Further study into this imaging modality could shed further light.
Fig. 15: Coronal proton density MRI demonstrates type III (borderline Type V) dislocation with complete rupture of AC and CC ligaments
References: Department of Radiology, Royal Brisbane and Women's Hospital
Dyer et al explored an entirely new option to AC joint analysis using an imaging modality known as wide field of view dynamic CT-scan (4D CT).
Dyer analysed the patient doing 3 dynamic shoulder movements: Forward flex 45°,
internal rotation 90° and adduction 45°.
He found that translations of the clavicle occurred that were previously unrecognised using traditionally imaging methods.
With this new information the patient who was previously diagnosed as having a type II/III AC injury which was conservatively treated,
was given surgical intervention.
Dyer has shown the potential benefit of using 4D CT over traditional imaging.
His study was limited to one person; a more comprehensive sample size may be able to shed further light on the benefits of this method.(14) A limitation of this method is the increased radiation dose of ~2-3 mSv the patient is exposed to as opposed to ~0.1 found in a traditional X-ray.
Fig. 16: 4D CT anterior view of the shoulder in neutral position showing a widening of the AC joint
References: Dyer DR, Troupis JM, Kamali Moaveni A. Wide field of view CT and acromioclavicular joint instability: A technical innovation. Journal of medical imaging and radiation oncology. 2015;59(3):326-30.
Fig. 17: 4D CT anterior view of the shoulder after internal rotation of 90 degrees showing the clavicle overriding the acromion
References: Dyer DR, Troupis JM, Kamali Moaveni A. Wide field of view CT and acromioclavicular joint instability: A technical innovation. Journal of medical imaging and radiation oncology. 2015;59(3):326-30.