NAPLES, FLA. – Computed tomography of the cervical spine had 100% specificity and sensitivity in ruling out clinically significant fracture and ligamentous injuries after blunt trauma in a prospective cohort of 5,676 patients.
Patients with negative CT scans do not need further workup for possible ligamentous injury, as those were only identified in patients with a positive CT scan, Dr. Poornima Vanguri said at the annual scientific assembly of the Eastern Association for the Surgery of Trauma.
"Patients with a normal CT do not need further imaging and collars should be cleared as soon as possible," the authors concluded.
The prospective study included all 5,676 blunt trauma alert patients seen at a Level 1 trauma center between January 2008 and December 2012. Their average age was 40.6 years, mean Glasgow Coma Scale (GCS) score 14.2, mean Injury Severity Score (ISS) 9.6, and mean length of stay 4.8 days.
Of these, 420 (7.2%) had any cervical spine injury and 2.6% died.
The incidence of fracture was 7.2% (409/5,676) and ligamentous injury 0.9% (52/5,676), said Dr. Vanguri, a general surgery resident at Virginia Commonwealth University in Richmond.
Patients with cervical spine injury versus those without were significantly older (46.3 years vs. 40.2 years), had significantly lower GCS scores (13.5 vs. 14.2), and had significantly higher ISS (18.9 vs. 8.9), average blood alcohol levels (666 mg/L vs. 538 mg/L), and average lactate levels (2.6 mmol/L vs. 2.4 mmol/L). Patients with cervical spine injuries also stayed significantly longer in the hospital (11.3 days vs. 4.3 days) and ICU (5.8 days vs. 1.4 days), and had more ventilator days (9.8 days vs. 4.1 days).
Of the 52 patients with ligamentous injury, 30 (57.7%) were suspected on CT, Dr. Vanguri noted. The remaining 22 without suspected ligamentous injury all had associated fractures identified by CT requiring further intervention. Thus, CT attained 100% sensitivity and specificity in ruling out cervical spine injuries, she said.
Notably, the incidence of ligamentous injury without fracture was 0.2% (10/5,676). Stepwise logistic regression identified only three independent predictors of ligamentous injury: cervical midline tenderness, abnormal alertness, and C-spine fracture on CT.
The poster presentation prompted a spirited debate at the meeting, with some attendees arguing that clearing collars in patients with a negative CT could leave practitioners and hospitals open to potential litigation if unstable fractures and/or ligamentous injuries are missed by not conducting further testing with magnetic resonance imaging (MRI).
The specter of litigation is always a possibility; however, the incidence of ligamentous injury without fracture was exceedingly low and prolonged C-collar use is not without consequences, argued senior author Dr. Therèse Duane, an ACS Fellow and trauma and critical care surgeon at the university.
"How many times do patients go home and 3 months later still have a collar on because they can’t get an MRI and now they have migraines, can’t work, and can’t drive? It’s huge," she said.
In the study, 77% of the 52 patients with ligamentous injury underwent MRI in addition to a CT scan, 48% had CT angiography of the neck, 40.4% had C-spine surgery, and 81% required prolonged collar use for treatment purposes of 6 weeks or longer.
Though current practice at the university is to clear collars in patients with a negative CT scan, research would be needed to determine whether this approach is feasible in the higher-risk patients with altered alertness, Dr. Duane observed.
Further studies are planned by the group looking at national databanks, as well as combining a decade worth of data at their institution, to define a C-spine clearance protocol.
Dr. Vanguri and her colleagues reported having no financial disclosures.