Cellulitis is a bacterial infection of dermal and subcutaneous tissue characterized by redness, pain, and warmth. The legs are the most commonly affected sites. Unfortunately, cellulitis is often misdiagnosed in the clinical setting and proves costly to the healthcare system, resulting in increased medical error, medical burden, and unnecessary treatment. It is a frequent cause of hospital Emergency Department visits and inpatient admissions.
What makes diagnosis difficult is that the mimickers so closely resemble cellulitis. These mimickers are categorized together as pseudocellulitis. Cellulitis mimickers can include noninfectious inflammatory skin or subcutaneous conditions such as allergic and irritant contact dermatitis, panniculitis, stasis dermatitis, and deep venous thrombosis, as well as other infectious entities such as cutaneous fungal infections, etc. For a comprehensive differential, see VisualDx.
A New Predictive Model for Cellulitis: The ALT-70
In 2017, Dermatologists from Massachusetts General and Brigham and Women’s Hospital created a novel predictive model for the diagnosis of lower extremity cellulitis1. These parameters include
- Asymmetry: unilateral leg involvement (3 points)
- Leukocytosis: elevated white blood cell count >/= 10,000/uL (1 point)
- Tachycardia: heart rate >/= 90 beats per minute (1 point)
- Age: >/= 70 (2 points)
The outcome recommendations are:
- 0-2 points: cellulitis unlikely, reconsider diagnosis or consider causes of pseudocellulitis
- 3-4 points: indeterminate, obtain additional information and consider consultation
- 5-7 points: likely true cellulitis, consider empiric antibiotic therapy
Since then, the team concluded the ALT-70 score maintained its predictive value between initial ED presentation, 24 hours, and 48 hours2. The ALT-70 model also performed well in a single institution prospective study against thermal imaging3.
The studies are somewhat limited; however if further validated, this tool may serve to be a useful point of care predictive tool for objective diagnosis of cellulitis.
- Raff, Adam B et al. “A predictive model for diagnosis of lower extremity cellulitis: A cross-sectional study.” Journal of the American Academy of Dermatology vol. 76,4 (2017): 618-625.e2. doi:10.1016/j.jaad.2016.12.044
- Singer, Sean et al. “The ALT-70 cellulitis model maintains predictive value at 24 and 48 hours after presentation.” Journal of the American Academy of Dermatology vol. 81,6 (2019): 1252-1256. doi:10.1016/j.jaad.2019.03.050
- Li, David G et al. “The ALT-70 predictive model outperforms thermal imaging for the diagnosis of lower extremity cellulitis: A prospective evaluation.” Journal of the American Academy of Dermatology vol. 79,6 (2018): 1076-1080.e1. doi:10.1016/j.jaad.2018.06.062