Background: Current severity scoring systems in intensive care units (ICUs) are complex and time-consuming, limiting their utility for rapid clinical decision-making. This study aimed to develop and validate simplified prediction models using readily available biomarkers for assessing in-hospital mortality risk. Methods: We analyzed 19,720 adult ICU patients in this retrospective study. Three prediction models were developed: a basic model using lactate-to-albumin ratio (LAR) and neutrophil percent-to-albumin ratio (NPAR) and two enhanced models incorporating mechanical ventilation and continuous renal replacement therapy. Model performance was evaluated against Sequential Organ Failure Assessment (SOFA) score and Acute Physiology and Chronic Health Evaluation (APACHE) II score using machine learning approaches and validated through comprehensive subgroup analyses. Results: Among individual biomarkers, SOFA score showed the highest discriminatory power (area under these curves [AUC] 0.931), followed by LAR (AUC 0.830), CAR (AUC 0.749), and NPAR (AUC 0.748). Our enhanced Model 3 demonstrated exceptional predictive performance (AUC 0.929), statistically comparable to SOFA (p = 0.052), and showed a trend toward superiority over APACHE II (AUC 0.900, p = 0.079). Model 2 performed comparably to APACHE II (AUC 0.913, p = 0.430), while Model 1, using only LAR and NPAR, achieved robust performance (AUC 0.898) despite its simplicity. Subgroup analyses across different ICU types demonstrated consistent performance of all three models, supporting their broad clinical applicability. Conclusions: This study introduces novel, simplified prediction models that rival traditional scoring systems in accuracy while offering significantly faster implementation. These findings represent a crucial step toward more efficient and practical risk assessment in critical care, potentially enabling earlier clinical interventions and improved patient outcomes.

Objective We aimed to investigate the causes and clinical and laboratory features of patients with ureteritis observed on intravenous contrast-enhanced abdominopelvic computed tomography (APCT) conducted in the emergency department (ED).Methods All APCTs conducted in the ED from November 2017 to November 2020 were investigated for the presence of ureteritis. The incidence of ureteritis, presumed cause of ureteritis, and clinical as well as laboratory features of patients with ureteritis were retrospectively analyzed.Results Ureteritis was observed in 422 out of 7,386 patients (5.7%) who underwent APCTs. The two main reasons for undergoing APCT in the ED were abdominal pain (49%) and infection focus workup (33%). The first major cause of ureteritis was urinary tract infection (UTI) (351 of 422, 83%). Most patients (85%) were febrile, but 208 (59%) exhibited no urinary symptoms such as dysuria, increased frequency, or residual urine sense. The second major cause of ureteritis was ureteral stones (42 of 422, 10%). Thirty-two of 42 patients (76%) had simple obstructive uropathy, while 24% of patients had a combined infection along with an obstruction. Other rare causes were malignancy and the spread of adjacent inflammation.Conclusion Ureteritis was a common finding observed in 5.7% of patients who underwent APCTs at the ED, and most of them were secondary to UTIs and ureteral stones. UTIs can cause ureteritis even without typical symptoms or signs suggestive of UTI, and diagnosis without an APCT can be difficult. More liberal use of APCTs should be considered when the cause of fever is difficult to diagnose.

Background/Objectives: Elevated intracranial pressure (ICP) drives secondary brain injury and poor outcome. The pressure reactivity index (PRx), the reference marker of cerebral autoregulation, requires high-frequency waveforms unavailable in many centers. We examined whether routinely charted hourly ICP and cerebral perfusion pressure (CPP) can stratify prognosis without the PRx. Methods: In a retrospective cohort of 990 adults with invasive ICP monitoring at a tertiary neurosurgical ICU (a predominantly non-traumatic case mix), we derived from hourly ICP the mean, variability (within-patient standard deviation), and peak burden (time above 22 mmHg), and from CPP the low-perfusion burden (time below 60 mmHg). Each metric (per 1 SD) was related to 28-day mortality and poor neurological outcome (Glasgow Outcome Scale 1–3), adjusted for age, sex, APACHE II, GCS, diagnosis, and mean ICP, with incremental value and sensitivity analyses. Results: In total, 86 of 990 patients died within 28 days. ICP variability was associated with both mortality and poor neurological outcome and, unlike peak burden, remained independent after adjustment for mean ICP. Peak burden lost significance once mean and variability were known. For CPP, hypoperfusion burden, but not CPP variability, was associated with mortality. CPP variability did not survive joint modeling with ICP variability. Excess risk was confined to low CPP, with no harm at high CPP. ICP variability added incremental value beyond a clinical model, robust to sensitivity analyses. Conclusions: Routinely available ICP variability and CPP hypoperfusion burden carry prognostic information beyond mean pressure. ICP variability predicted both outcomes, whereas hypoperfusion burden predicted mortality. These summaries may offer a pragmatic alternative where the PRx is unavailable, pending validation.