Implementing a clinical scoring system reduces CT use, but it comes with a cost – lower diagnostic accuracy.
Using a scoring system for adolescent and young adult patients who are suspected of appendicitis can be a trade-off – it can help identify individuals who are best suited to CT scans, but it can also significantly reduce diagnostic accuracy.
In recent decades, CT utilization has been on the rise, leading providers to search for more targeted ways to pinpoint which patients actually need the exam, as well as lower the needed radiation dose. Reaching this goal is important because CT pre-operative imaging is standard for an acute appendicitis diagnosis, said a team of investigators from Soongsil University in Seoul, South Korea, in a May 18 study published in Radiology.
“In this study, we measured the reduction of CT use and the diagnostic accuracy in simulated patient triaging using clinical scoring systems,” said the team led by Hyunjoo Song, from Soongsil University’s School of Computer Science and Engineering. “We hypothesized that patient triaging with clinical scoring systems could reduce CT use without a substantial loss of diagnostic accuracy in adolescents and young adults with suspected appendicitis.”
However, the results of the study yielded “disappointing results,” the team said.
For their retrospective study, conducted between December 2013 and August 2016, they tested five scoring systems (adult appendicitis, appendicitis inflammatory response, modified Alvarado, Broek, and Christian scores) and categorized 2,888 patients between ages 15 and 44 with suspected appendicitis into low-, intermediate-, and high-probability groups. Of the patients, 1,088 ultimately received a positive diagnosis; 1,800 did not.
According to their analysis, CT reduction rates for the five scoring systems ranged from 55.6 percent to 71.1 percent. However, those drops were also accompanied by lower sensitivity (48.7 percent to 81.2 percent) and specificity (79.0 percent to 97.8 percent).
“Targeting sensitivity of 97.6 percent and specificity of 94.9 percent, which were achieved when CT was used for all patients with suspected appendicitis, the CT reduction rates of all five clinical scoring systems were 0 percent,” the team said. “Even when both target sensitivity and specificity were lowered to 95.0 percent, the clinical scoring systems yielded small CT reduction rates ranging from 0.1 percent to 11.2 percent.”
These results, the team said, reiterate the findings from previous studies that support the routine – rather than selective – use of CT. However, in recently released guidelines, the World Society of Emergency Surgery supported implementing clinical scoring systems for selective use. But, the team explained, if they had applied the scoring systems based on those guidelines, a significant number of patients with appendicitis would have been misdiagnosed and would have experienced complications.
Related Content: Getting an Abdominal CT in the ED Could Depend on Age
The results of their study could potentially be explained, the team said, by CT’s significant diagnostic performance and its provision of alternative diagnoses that can frequently lead to hospitalization or surgery.
In an accompanying editorial, Vincent M. Mellnick, M.D., associate professor and abdominal imaging section chief at the Mallinckrodt Institute of Radiology at Washington University School of Medicine, agreed with the assessment of Song’s team – their findings further support widespread CT use.
Still, it is possible, he said that developing clinical scoring systems that work as intended for patients suspected of appendicitis could effectively reduce the number of patients who eventually undergo a CT scan.
Ultimately, he said, a hybrid approach could be beneficial.
“Perhaps the best approach is to use both clinical scoring and imaging in all patients, particularly in cases with equivocal imaging findings,” he said. “More study on this topic is warranted.”