Association of the Modified HALP Score with Disease Complexity in Deep Neck Infections: A Dual-Center Retrospective...

Background/Objectives: To evaluate the association between the modified hemoglobin-albumin-lymphocyte-platelet (mHALP) score and disease complexity in patients hospitalized with deep neck infections (DNIs) and to compare performance with HALP and individual laboratory markers. Methods: This dual-center retrospective study included 367 adults hospitalized with DNIs between December 2021 and September 2025. Patients were classified as having complicated or non-complicated disease. The mHALP score was calculated from admission laboratory values and incorporated C-reactive protein. ROC curve and multivariable logistic regression analyses were performed. The primary model evaluated natural-log-transformed, standardized mHALP after adjustment for age, maximum abscess diameter, and study center. Results: Fifty-nine patients (16.1%) had complicated DNIs. Median mHALP was lower in the complicated group than in the non-complicated group (0.0137 vs. 0.0486; p < 0.001). mHALP showed moderate discrimination (AUC = 0.756, 95% CI 0.677-0.834) and performed better than CRP alone (AUC difference = 0.077; p < 0.001), but not better than HALP, albumin, or lymphocyte count. An exploratory cut-off of ≤0.0302 yielded 74.6% sensitivity and 71.8% specificity. In the primary multivariable model, each one-standard-deviation increase in log-transformed mHALP was associated with lower odds of complicated disease (adjusted OR = 0.592, 95% CI 0.397-0.884; p = 0.010). Adding mHALP to the clinical model increased the AUC from 0.858 to 0.864, whereas adding HALP increased it to 0.894. Conclusions: Lower mHALP was independently associated with complicated DNIs, but its incremental value beyond clinical variables was modest, and it did not outperform HALP. mHALP may be considered an adjunctive, rather than stand-alone, marker, pending external validation.