Injury to the cervical spine involving all three columns is a serious problem in acute trauma scenarios that accounts for 2.4% of all blunt trauma [1] and 55% of all spinal cord injuries [2]. Commonly presenting injury patterns involving all three columns include flexion-distraction injuries (AO type B) and cervical fracture dislocations (AO type C). The most commonly reported sources of cervical spine injuries are vehicular accidents (41%), falls (27%), violence (15%), sports-related injuries (8%), and heavy objects falling onto the head [3].

Management of three-column injuries of the cervical spine (TCICS) demands achieving an adequate reduction with the reconstruction of the anterior and posterior tension bands, the decompression of neurological structures, and an emphasis on achieving early fusion for rehabilitation. Although combined anterior and posterior stabilization is considered the choice of surgery for these injuries, with improved instrumentation techniques and the right indications, anterior- or posterior-only stabilization to create instrumented fusion may prove to be an attractive alternative to combined surgery. Though biomechanical studies have shown that posterior stabilization allows for a more rigid fixation than anterior plates [4], the anterior approach allows direct decompression in cases with a protruded intervertebral disc or a fractured, retropulsed bone fragment, both of which may require the use of a large graft surface area, reducing the number of fused segments. The aim of this study is to assess the clinical and functional outcomes in patients undergoing anterior cervical discectomy/corpectomy and fusion (ACDF/ACCF) for TCICS.

In retrospective data retrieved after 2 years, 28 cervical trauma patients were operated on, out of which 22 patients met the inclusion criteria and were included in the study. One patient expired in the postoperative period due to pulmonary complications and was excluded from the study. The included patients had an approximate male-to-female ratio of 5:2 (15 males, six females) with a mean age of 45.3 years (range, 18 to 70 years). Out of 21 patients, 19 had AO type C injuries and were managed with ACDF, and 2 patients with AO type B injuries were managed with ACCF (Table 1). The levels involved were C3–C4 (1), C4–C5 (4), C5–C6 (10), and C6–C7 (6), with C5–C6 being the most common level. The mechanism of injury was observed to be a fall from height in 14 patients (66.6%) and road traffic accidents in seven patients (33.3%). The average duration of the follow-up period was 16.8 months (range, 11 to 29 months). Fifteen patients had a complete cord injury (ASIA A), while six patients had an incomplete cord injury (two ASIA B, three ASIA C, and one ASIA D). Out of 19 patients with AO type C injuries, 13 patients were found to have a bilateral facet dislocation, while six patients had a unilateral facet dislocation. Seventeen patients underwent a closed reduction under GA on the table before surgery, while two patients had to be put on Crutchfield tongs as their pulmonary condition needed optimization before surgery. Out of the 17 patients, a complete reduction was achieved in 12 cases, while another five needed intraoperative manipulation. In one patient, a reduction could not be achieved despite intraoperative manipulation. Staged posterior instrumentation with lateral mass screws and posterolateral fusion was done for the patient, and a complete fusion was achieved by the final follow-up (Fig. 2). Overall, 19 patients underwent ACDF with bone grafts from the iliac crest (15 patients) (Fig. 3) or a PEEK interbody cage (four patients), and two patients underwent ACCF with a mesh cage (Fig. 4).

The mean operative time was 177.3±33.8 minutes, and the average blood loss was observed to be 253.1±109.7 mL. There were no intraoperative complications. Seven patients reported an improvement in neurology by 1 or 2 grades on the ASIA scale, while 14 showed no improvement (Table 2). Radiologically, the mean segmental kyphosis angle at presentation was 12.2°±4.4°, while, in the immediate postoperative period, it was -7.2°±2.5° (p<0.05). At the last follow-up, the mean loss of correction was 2.1°±0.9°. All patients showed bony fusion of the operated levels at the final follow-up. There was no evidence of instability as seen in flexion-extension radiographs at the time of last follow-up in any of the patients.

The preoperative mean VAS was 6.8±2.1, which showed a gradual improvement that was recorded over the serial follow-ups with the final mean VAS being 1.8±1.1 (p<0.05). The mean ODI score showed a similar trend with a mean preoperative score of 59.7±11.6 and 34.9±13.4 at the final follow-up (p<0.05). The preoperative SCIM score was 24.8±17.2, which improved to 36.4±23.7 (p<0.05).

TCICS occurs as a result of hyperflexion with, or without, a rotational element to the cervical spine. The sub-axial spine is commonly involved, with nearly 70% of cases located in the C4–C7 region [6]. In our data, C5–C6 was the most common level. Falling from a height was found to be the most common mode of injury. Nineteen patients were found to have cervical fracture dislocations, while two patients had a flexion-distraction injury.

Achieving an early reduction of the dislocated facets is the first step in the management of patients with cervical fracture dislocations. Different authors have different recommendations with respect to the technique of the said reduction. A reduction of the dislocated facets can be achieved in the emergency room by the application of cervical traction [7], in the operating room under GA before surgery [8-10], or intraoperatively by distraction after a discectomy [11-13]. We achieved a reduction in 70.5% of patients (12 out of 17) by using a technique involving manipulation on the table under GA without any attempt at a closed reduction by traction; this was to avoid time loss and patient discomfort [12,13]. We achieved a reduction by intraoperative distraction after a discectomy in four patients, and in one patient with persistent dislocation, a staged posterior approach was deemed necessary. All five patients with failed attempts at closed reductions had injuries more than 72 hours old, which could explain the inability to achieve a reduction by manual traction.

Being the most mobile segment in the axial spine, the need for adequate stabilization cannot be overemphasized. Optimal stabilization for TCICS remains controversial due to the lack of randomized prospective trials. From the anterior-only, posterior-only, and combined anterior and posterior stabilization techniques, the combined approach is reported to be superior in terms of biomechanical stability [4,14,15]. Despite the theoretical biomechanical advantage, a number of studies have reported comparable results with anterior-only fixation [14,16]. Anterior-only surgery has less blood loss and a shorter operative time and avoids prone turning of the patient intraoperatively, something that poses a potential risk of losing reduction [7]. Facet reduction by the anterior approach can be achieved via intraoperative manipulation with a distraction technique [17] or, if required, by a facetectomy [18]. Moreover, the anterior-only approach offers direct decompression [19-22], by removing the offending agent and reducing the number of fused segments or the large surface area (i.e., the interbody area) to be fused, and has a lower infection rate [23].

The posterior approach is considered superior to the anterior approach in achieving reduction [24], but in cases with the presence of a retropulsed disc fragment, reduction by a posterior-only approach may lead to further neurological deterioration [25]. The posterior approach is reserved for cases with posterior cord compression due to fractured posterior elements or irreducible fracture dislocation with failed open, or closed, anterior reduction techniques. In our data, a staged posterior fixation was needed in one case with the dislocation persisting after an attempted closed reduction and an attempted anterior reduction with stabilization.

None of the patients in our data showed a worsening of their neurology. An improvement in neurology was seen in 33.3% of patients (7 out of 21). Among the patients with an incomplete spinal cord injury, four out of six patients showed an improvement in their ASIA grade, and five out of six patients achieved an ambulatory status after surgery. A significant improvement in VAS, ODI, and SCIM scores was seen at the final follow-up in all the patients. The neurological and functional outcomes were comparable to other studies describing the management of TCICS with posterior or combined approaches [7,12-14,16]. Radiologically, the segmental kyphosis angle showed significant improvement at the final follow-up (-5.1°±1.4°) when compared to the preoperative kyphosis angle (12.2°±4.4°, p<0.05). There was a mean loss of 2.1° of correction between the immediate postoperative period and the last follow-up. This can be explained due to interface bone resorption in cases that used an iliac crest bone graft or minimal cage subsidence in cases using a PEEK or mesh cage. Achieving and maintaining a reduction of the facet joints is the most important factor when determining the correction of kyphosis deformity in cases of cervical fracture dislocations. These results are comparable to other studies evaluating the radiological outcomes of surgically managed patients with cervical fracture dislocations [14,16]. A facet reduction was achieved in all our patients, and follow-up imaging showed bony fusion in all our cases. None of our patients had any instability or implant-related complications at the final follow-up.

This article has a few limitations: the study design being retrospective could lead to bias, the small sample size, and the short follow-up period.

Most patients with TCICS can be managed by anterioronly stabilization with good neurological, radiological, and functional outcomes with a minimal complication rate. Further randomized prospective studies are required for a better comparison with the combined approach.