Note: Last week’s article, as selected by 52 in 52, was the NEXUS study methodology. I erroneously assumed that the other NEXUS article would be the results, but it was not. I don’t want to leave everyone with a cliffhanger, so here are the NEXUS study results: Sensitivity 98-99%, Specificity 12.9%, PPV 2.7%, NPV 99.8%.
Article Citation: Hendey GW, Wolfson AB, Mower WR, Hoffman JR; National Emergency X-Radiography Utilization Study Group. Spinal cord injury without radiographic abnormality: results of the National Emergency X-Radiography Utilization Study in blunt cervical trauma. Lancet J Trauma. 2002 Jul;53(1):1-4. PMID: 12131380
What we already know about the topic: Xray (and by extension CT) is much more sensitive to bony injury than soft tissue injury, particularly when no contrast is given. The NEXUS study authors were aware of pediatric case series of SCIWORA (spinal cord injury without radiographic abnormalities) and were interested in the rate identified in their NEXUS data. Studies reporting SCIWORA had largely been case series rather than prospective observational studies.
Why this study is important: It attempts to quantify the worry that a negative imaging study may be insufficient to identify spinal cord injury.
Brief overview of the study: This study was a secondary analysis of the NEXUS study database (see last week’s Pearl for full methods). 34,869 patients were recorded in the NEXUS study, of whom 818 (2%) were identified by some diagnostic method (imaging, clinically, both) as having a spinal cord injury. SCIWORA was defined by the authors as an MRI finding of spinal cord injury after a negative x-ray. If the patient went straight to CT or MRI, or if they were diagnosed on CT without MRI they were excluded. 27 cases of SCIWORA (0.008% of all participants, 3.3% of SCI cases) were identified. Interestingly, 40% of these did not have a focal neurological deficit documented on initial exam. Spinal cord edema was present in 93% of cases on MRI. None were in children. All of these patients screened positive by NEXUS.
Limitations: Perhaps not a limitation per se, but the definition of SCIWORA used here is different from the one that we have learned in pediatrics, wherein the CT findings are negative as well. Second, there was no protocol in place to ensure that patients with a persistent neurologic deficit after a negative adequate plain film proceeded to MRI, so some patients may have been discharged with an inconclusive diagnosis. Third, given that the standard for C-spine imaging has moved to CT from plain films, this study does not inform us about how many patients with negative CT c-spines are likely to still have underlying spinal cord injury (see the Bonus).
Take home points: The NEXUS criteria is adequately sensitive to pick up injuries that are non-bony. Unsurprisingly, some isolated spinal cord trauma without bony injury will not be apparent on plain films.
BONUS! : To bring this up to the modern era, I thought I would mention an observational study of 11644 adult and 3458 pediatric blunt traumas by Como et all out of Cleveland (1). Using CT as the screening mechanism, they identified 313 spinal cord injuries by clinical diagnosis, of whom 25 had negative CT c-spine studies, yielding an 8% rate of SCIWOCTET (SCI w/o CT e/o Trauma – they reserve the term SCIWORA for neg MRI). That is a concerning number.
(1) Como JJ1, Samia H, Nemunaitis GA, Jain V, Anderson JS, Malangoni MA, Claridge JA.The misapplication of the term spinal cord injury without radiographic abnormality (SCIWORA) in adults. J Trauma Acute Care Surg. 2012 Nov;73(5):1261-6.
