Document Type : Original Article
Authors
1
Toxicological Research Center, Excellence Center of Clinical Toxicology, Department of Clinical Toxicology, Loghman Hakim Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran
2
Department of Emergency Medicine, School of Medicine, Iran University of Medical Sciences, Tehran, Iran
3
Rajaie Cardiovascular Medical and Research Institute,Tehran, Iran
4
Student Research Committee, School of Nursing and Midwifery, Shahid Beheshti University of Medical Sciences, Tehran, Iran
5
Laser Application in Medical Sciences, Shahid Beheshti University of Medical Sciences
Abstract
Introduction:
Early identification of acutely poisoned patients requiring intensive care remains a critical challenge in emergency toxicology. The Tanta University Risk Model (TURM) estimates ICU admission risk without requiring toxicological identification of the causative agent. This study examined the external validity of the TURM in a high-volume tertiary toxicology referral center in Iran.
Methods:
Medical records of 842 patients hospitalized for acute toxic exposure at Loghman Hakim Hospital, Tehran, between 2022 and 2023 were reviewed retrospectively. The TURM was computed from five admission parameters: GCS, diastolic blood pressure, peripheral oxygen saturation, respiratory rate, and serum bicarbonate. Discriminatory performance was quantified by area under the ROC curve, with threshold optimization via the Youden index. Model calibration was assessed using the Hosmer-Lemeshow test. The Poisoning General Index (PGI) was additionally evaluated for comparative purposes.
Results:
Of 842 patients (67.9% male; mean age 38.99 ± 16.46 years), 436 (51.8%) required ICU admission. The TURM score was significantly lower in ICU-admitted patients (median −73.76 [IQR: −78.73 to −66.77]) than non-ICU patients (−78.15 [IQR: −81.24 to −74.07]; p < 0.001), yielding an AUC of 0.62 (95% CI: 0.58–0.66), sensitivity of 65.1%, specificity of 51.4%, and overall accuracy of 58.2%. Calibration was adequate (Hosmer-Lemeshow χ²(8) = 6.9, p = 0.55). The PGI demonstrated inferior discrimination (AUC = 0.563).
Conclusion:
The TURM showed statistically significant but moderate discriminatory performance alongside good calibration, supporting its potential role as a supplementary triage instrument in resource-limited emergency settings. Prospective multicenter validation remains necessary to confirm generalizability.
Highlights
Nava Soleymani (Google Scholar) (PubMed)
Shahin Shadnia (Google Scholar) (PubMed)
Babak Mostafazadeh (Google Scholar) (PubMed)
Mitra Rahimi (Google Scholar) (PubMed)
Peyman Erfan Talab Evini (Google Scholar) (PubMed)
Seyed Mohammad Vahabi (Google Scholar) (PubMed)
Amirali Soheili (Google Scholar) (PubMed)
Aydin Feyzi (Google Scholar) (PubMed)
Mahsa Taremi (Google Scholar) (PubMed)
Zahra Razzaghi (Google Scholar) (PubMed)
Negar Sadat Mohajerani (Google Scholar) (PubMed)
Sayed Masoud Hosseini (Google Scholar) (PubMed)
Keywords
Subjects
