Potential Prognostic Parameters from Patient Medical Files for Inhalation Injury Presence and/or Degree: A Single-Center Study.

(1) Background: Inhalation injury significantly worsens burn outcomes but lacks a standardized definition and diagnostic consensus, complicating prognosis. Existing diagnostic tools often show limited sensitivity and specificity, reducing clinical utility. This study aimed to identify potential clinical markers, recorded at or shortly after admission, for inhalation injury prognostication. (2) Methods: A retrospective cohort study of 59 burn patients admitted to Tygerberg Hospital's Burn Centre (South Africa) between 23 April 2016 and 15 August 2017 was conducted. Descriptive statistics were reported based on data type and distribution. Fisher's exact test, Spearman's rank correlation (rho), and partial least squares regression (VIP scores) assessed associations, correlations, and predictive value. p < 0.05 (two-tailed) denoted significance. (3) Results: Severe inhalation injury accounted for 61% of admissions (mean 11.2; CI = 9.5-12.9), with a 38.9% mortality rate. Significant associations (p ≤ 0.008) and positive correlations (p ≤ 0.06) were noted for total body surface area (rho = 0.357), complications (rho = 0.690), and burns intensive care unit length of stay (BICU LOS, rho = 0.908). Complications and BICU LOS showed the strongest predictive contributions (VIP = 1.229 and 1.372). Lactate (rho = 0.331, p < 0.011) and hoarseness (rho = -0.314, p < 0.015) correlated significantly but lacked association. (4) Conclusions: Findings suggest elevated lactate may serve as a prognostic marker, while BICU LOS and complications may reflect disease progression. A multi-marker approach is recommended.