REPORTING FROM AAD 18
SAN DIEGO (FRONTLINE MEDICAL NEWS) – A simple scoring system surpassed thermal imaging for diagnosing lower extremity cellulitis in a head-to-head, single-center comparison in 67 patients.
The ALT-70 score – which tallies points for asymmetry, leukocytosis, tachycardia, and age of at least 70 years – produced a positive predictive value for lower-extremity cellulitis (LEC) of 80.4% and a negative predictive value of 90.9%, compared with values of 75.5% and 57.1%, respectively, for thermal imaging when researchers applied both methods to 67 patients, said David G. Li , a clinical research fellow in the department of dermatology at Brigham and Women’s Hospital, Boston, where the study was conducted.
“We recommend ALT-70 for routine practice to reduce misdiagnosis of lower-extremity cellulitis,” said Mr. Li.
The senior author of Mr. Li’s report, Arash Mostaghimi, MD , director of the inpatient consultation service, department of dermatology at Brigham and Women’s, was also lead investigator for the team of dermatology researchers – from his center and from Massachusetts General Hospital in Boston – who recently devised the ALT-70 scoring system for diagnosing LEC ( J Amer Acad Dermatol. 2017 April;76[4]:618-25.e2 ).
The four-item survey can generate a score of 0-7, with a score of 0-2 suggesting need for additional monitoring, a score of 3-4 initiating a dermatology consult, and a score of 5-7 triggering immediate treatment for cellulitis, Mr. Li said. The 2017 review of ALT-70 showed that among 259 patients, those with a score of 0-2 had an 83% likelihood of having pseudocellulitis, while patients with a score of 5-7 had an 82% likelihood of having true cellulitis.
Thermal imaging of the lower extremity, which identifies cellulitis by a higher skin temperature compared with unaffected areas on the limb, has also recently gained currency as a way to objectively diagnose cellulitis ( J Invest Dermatol. 2018 March;138[3]:520-6 ).
The current study enrolled 67 patients who had a presumptive diagnosis of LEC while in the emergency department or inpatient wards during a 7-month period. In addition to undergoing blinded assessment by both thermal imaging and by ALT-70 scoring, all patients also underwent blinded assessment by a board-certified dermatologist, who provided the definitive diagnosis. The attending dermatologists determined that 46 of the patients had true LEC and 21 patients did not.
The calculated sensitivity of ALT-70 was 97.8%, compared with 87.0% for thermal imaging. Specificity was 47.6% for ALT-70 and 38.1% for thermal imaging, Mr. Li reported at the annual meeting of the American Academy of Dermatology.
He also presented an analysis of the results when he combined both methods, with a positive on both assessments required to produce a positive LEC diagnosis. This resulted in a positive predictive value of 86.7%, slightly higher than the 80.4% from ALT-70 alone, but the combination produced a negative predictive value of 68.2%, substantially less than the 90.9% rate with ALT-70 alone. This demonstrated the “marginal benefit” from combining the two methods, he said.
In a receiver operating characteristic curve analysis , in which the area under the curve (c-statistic) reflects a diagnostic test’s validity, ALT-70 produced a c-statistic of 0.85, thermal imaging had a c-statistic of 0.63, and when combined, the c-statistic was 0.88.
Mr. Li called for validation of the findings using larger and different patient populations.
He had no reported disclosures.
SOURCE: Li DG et al. AAD 18, Abstract 6744.
