Background: In critically ill patients, tracheostomy decisions are driven by heterogeneous and dynamic clinical trajectories, and no universally accepted scoring system exists to reliably predict tracheostomy requirement. An accurate and interpretable prediction model could help earlier decision-making and potentially reduce prolonged mechanical ventilation (MV) and failed weaning. Methods: In this retrospective study, data from 6507 mechanically ventilated intensive care unit (ICU) patients were analyzed using an electronic clinical decision support system; 1049 patients required tracheostomy and 5458 did not. The primary outcome was the prediction of tracheostomy occurrence during ICU stay based on invasive mechanical ventilation (IMV) parameters obtained within the first five days. The secondary outcome was the identification of the most influential parameters guiding tracheostomy decision-making during early IMV. Ten machine learning algorithms were developed using an 80/20 train–test split. Model performance was assessed using discrimination, calibration, and clinical performance metrics. Explainability was evaluated using SHapley Additive exPlanations (SHAP) analysis. Results: Among all models, Gradient Boosting demonstrated strong discrimination and calibration performance (AUROC 0.92, AUPRC 0.56, specificity 97%, F1 score 0.46, Brier score 0.078). In the Gradient Boosting model, feature importance analysis demonstrated that secretion count was the strongest predictor of tracheostomy requirement, accounting for 14.72% of the model’s predictive contribution. This was followed by lactate level (6.12%), arterial pH (3.74%), and peak airway pressure (3.57%). SHAP-based analyses consistently identified secretion count as the strongest predictor of tracheostomy requirement, followed by lactate level, Glasgow Coma Scale (GCS), and arterial pH. In addition, SHAP provided clinically interpretable insights into the direction and magnitude of the effects of individual predictors. Conclusions: Machine learning models integrating early-phase ventilatory and physiological data may enable clinically meaningful prediction of tracheostomy requirement. The combination of strong performance and explainability suggests potential utility as a decision-support tool in critically ill patients requiring prolonged MV.

Background and Objectives: Neuroleptic malignant syndrome (NMS) is a rare but high-mortality clinical condition. The aim of this study was to describe the demographic characteristics, clinical and laboratory findings, management strategies, and clinical outcomes of adult patients with NMS treated in an adult intensive care unit. Materials and Methods: This retrospective descriptive study included adult patients admitted to a tertiary care ICU with a diagnosis of NMS. Data were obtained from medical records and included demographic characteristics, suspected NMS-related drug exposures, clinical and laboratory findings at ICU admission, disease severity indicators, treatments administered, and clinical outcomes. Continuous variables were expressed as median and interquartile range (IQR), and categorical variables as number and percentage. Results: A total of 42 patients were included. The median age was 48 years (IQR: 24.1–75.5), and 61.9% were male. Atypical antipsychotics were the most frequently implicated agents (69.0%). Altered mental status was observed in 92.9% of patients, tachycardia in 61.9%, and hyperthermia in 47.6%. Creatine kinase levels >1000 U/L were present in 57.1%, and leukocytosis in 71.4%. Mechanical ventilation was required in 59.5% of patients. Acute kidney injury developed in 31%, and continuous renal replacement therapy (CRRT) was initiated in 21.4%. Bromocriptine was administered in 54.8% of cases, dantrolene in 4.8%, and amantadine in 11.9%. The ICU mortality rate was 9.5%. Conclusions: Patients with neuroleptic malignant syndrome often present with a severe clinical course requiring advanced organ support in the intensive care unit. However, with early diagnosis and appropriate intensive care management, mortality can be maintained at acceptable levels.

Background and Objectives: The development of acute kidney injury (AKI) in the post-cardiopulmonary resuscitation (post-CPR) period is a common pathology that has not been adequately investigated but contributes significantly to morbidity and mortality. We aimed to investigate the causes of AKI in the early post-CPR period. Materials and Methods: This study was performed retrospectively in 82 adult patients who survived for at least 2 days out of 312 patients admitted to the intensive care unit after cardiac arrest in 2013–2022. AKI developed in 40 (48.7%) of these 82 patients (AKI 1–3 patient, respectively: 14, 13, 13). Binary logistic regression analysis was performed separately to determine the risk factors for AKI and mortality. Results: Each unit increase in BMI increased the risk of developing AKI by 1.272-fold, and the increase was statistically significant [OR (95%CI) = 1.272 (1.089–1486); p = 0.002]. The use of VSP and INO treatment alone increased the risk of AKI by approximately 14-fold, and this increase was statistically significant [OR (95%CI) = 14.225 (1.172–172.669); p = 0.037]. The combined use of VSP and INO treatment increased the risk of AKI by approximately 42-fold, and this increase was statistically significant [OR (95%CI) = 42.089 (2.683–660.201); p = 0.008]. The COVID-19 period alone increased the risk of developing AKI by 2.8-fold compared to the non-COVID-19 period, but the statistical significance of this increase was limited [OR (95%CI) = 2.801 (0.859–9.126); p = 0.088]. The development of AKI was not associated with mortality [OR (95%CI) = 2.194 (0.700–6.872); p = 0.178]. Conclusions: Having VSP and/or INO support and high BMI in the post-CPR period are the most important reasons for the development of AKI. COVID-19 may also increase the risk of developing AKI.

Aims: Early brain computed tomography following cardiac arrest can reveal alterations in gray–to-white matter differentiation, offering potentially important prognostic insights into neurological outcomes. This study examined the relationship between gray-to-white matter ratio (GWR) metrics and survival. Methods: Patients managed in the intensive care unit after cardiopulmonary resuscitation (CPR) between 2015 and 2025 were retrospectively reviewed. Patients aged >18 years, with witnessed in-hospital or out-of-hospital CPR and those with cranial CT scans within the first 48 hours after cardiac arrest, were included. To assess the association with survival, GWR measurements were calculated in six different regions across both hemispheres. Results: Overall, normal-range GWR measurements were observed across the cohort, with no statistically significant impact on survival outcomes. Independent risk factors for two-day survival were age (OR 1.039; p=0.005), absent pupillary light reflex (OR 5.43; p

Nekrotizan pankreatit mortalite ve morbiditesi yüksek olan, tedavisi konusunda belirsizlikler içeren, önemli bir pankreatit türüdür ve etiyolojisinde hipertrigliseridemi de olabilmektedir. Bölüm; hipertrigliseridemiye sekonder gelişen akut pankreatit olgumuzun yoğun bakım ünitesinde tedavi sürecini içermektedir. 35 yaş, erkek, ek hastalığı olmayan nekrotizan pankreatit hastası çoklu organ yetmezliği sebebiyle yoğun bakım ünitesine interne edilmiş ve 18 gün invaziv mekanik ventilasyon ile desteklenmiştir.16 seans plazma değişimi, 1 kez sürekli renal replasman tedavisi, hedeflenmiş terapötik ısı yönetimi, trigliserit düşürücü medikal tedavi, çoklu antibiyoterapi uygulanmış, ileri hemodinamik monitörizasyon ile yakın takip edilmiştir. Servis taburculuğu ardından sırasıyla girişimsel radyoloji tarafından perkütan drenaj ve genel cerrahi tarafından açık cerrahi ile nekroze koleksiyon drene edilmiştir. Uzayan tedavi süreci, servis yatışları sonunda (5 ay) kazanılmış güçsüzlük gelişen hasta fizik tedavi ve rehabilitasyon servisinde 2 ay yatırılarak tedavi edilip, bir yılın sonunda aktif yaşamına dönebilmiştir. Necrotizing pancreatitis is a severe form of pancreatitis with high mortality and morbidity rates and remains a clinical challenge due to uncertainties in its management. Hypertriglyceridemia may also be among its etiological factors. This chapter presents the treatment process of a case of acute pancreatitis secondary to hypertriglyceridemia managed in the intensive care unit. A 35-year-old, male patient with no known comorbidities was admitted to the ICU with necrotizing pancreatitis and multiple organ failure, requiring invasive mechanical ventilation support for 18 days. The patient underwent 16 sessions of plasma exchange, a single session of continuous renal replacement therapy (CRRT), targeted therapeutic temperature management, triglyceride-lowering medical treatment, and multiple antibiotic regimens, and was closely monitored using advanced hemodynamic monitoring techniques. Following discharge from the ICU, the patient underwent percutaneous drainage performed by interventional radiology and subsequently open surgical necrosectomy performed by general surgery. After a prolonged hospitalization period (5 months), the patient developed acquired weakness and was transferred to a physical therapy and rehabilitation unit, for 2 months. At the end of one year, the patient was able to return to an active lifestyle.

Aim: Postoperative intensive care unit (ICU) care after pancreaticoduodenectomy (PD) is essential for managing perioperative complications. This study evaluated ICU outcomes, mortality, reoperation, and major postoperative complications following PD.Material and Methods: This retrospective cohort study included patients undergoing PD between January 1, 2015, and August 1, 2025, who required postoperative ICU admission. Postoperative ICU admission was reserved for selected high-risk patients based on perioperative clinical assessment. Descriptive statistical analyses were performed. Demographic characteristics, comorbidities, need for continuous renal replacement therapy (CRRT), length of ICU stay, reoperation rates, major postoperative complications, and mortality were analyzed.Results: A total of 80 patients fulfilled the inclusion criteria among the 179 who underwent PD. The cohort had a median age of 65 years, with males constituting the majority of the population (53,66.2%). Common comorbidities included hypertension, diabetes mellitus, coronary artery disease, chronic obstructive pulmonary disease, and almost all patients were categorized as ASA physical status III. Preoperative epidural catheterization was performed in 22 patients (27%). Reoperation was required in 3 patients (3.8%), CRRT was required in 3 patients (3.8%), and 8 patients (10%) required ICU readmission, while ICU mortality occurred in 7 patients (8.8%).Conclusion: Due to its complexity, PD is frequently linked to notable morbidity and mortality rates and typically requires close postoperative monitoring in the ICU. This descriptive analysis provides insight into postoperative ICU outcomes following pancreaticoduodenectomy. Early recognition of clinical deterioration during postoperative ICU monitoring may facilitate timely intervention and potentially improve clinical outcomes.