Purpose: Bacterial orbital cellulitis (OC) and diffuse non-specific orbital inflammation (DNSOI) may be challenging to differentiate clinically. This study investigates the utility of systemic inflammatory markers, namely white cell count (WCC) and C-reactive protein (CRP), in differentiating between OC and DNSOI.
Methods: A single-centre retrospective study of patients diagnosed with OC or DNSOI, between 2003 to 2021, who had WCC and/or CRP obtained at presentation. The mean levels of these factors and the proportion of positivity were compared between OC and DNSOI. A receiver operating characteristic (ROC) analysis was conducted to calculate the specificity and sensitivity of WCC or CRP in each group.
Results: 49 patients were included in this study. The mean age was 56 ± 20 years, and 21 patients were females. 26 (53.1%) patients had OC, and 23 (46.9%) patients had DNSOI. Mean WCC for OC and DNSOI were 14.5 × 103/μL and 9.27 × 103/μL, respectively (P = 0.001). Mean CRP for OC and DNSOI were 104.4 mg/L and 10.0 mg/L, respectively (P < 0.001). The optimal CRP cut-off value of 20.2 mg/L demonstrated 90.9% sensitivity and 90.5% specificity (AUC = 0.946, P < 0.001) for differentiating between OC and DNSOI. CRP was more predictive of OC than WCC (P = 0.017). 7/26 (26.9%) OC patients with fever also had an elevated CRP, while 1/23 (4.3%) of DNSOI with fever had a normal CRP.
Conclusions: An elevated WCC is suggestive of OC. However, a normal WCC can neither exclude nor differentiate between OC and DNSOI. CRP may be a more accurate predictor of OC compared to WCC.
Keywords: C-Reactive protein; Non-specific orbital inflammation; inflammatory markers; orbital cellulitis; white-cell count.