Emre Karagöz

Background and Objectives: Tracheostomy is frequently performed in critically ill patients requiring prolonged invasive mechanical ventilation; however, factors associated with intermediate-term mortality after tracheostomy remain poorly characterized. This study aimed to identify clinical and procedural factors associated with 90-day all-cause mortality after tracheostomy in ICU patients. Materials and Methods: This retrospective multicenter cohort study included 292 adult patients who underwent tracheostomy in two tertiary ICUs between 1 October 2023 and 1 June 2025. Demographic characteristics, admission diagnoses, comorbidities, clinical severity scores, procedural variables, microbiological culture results, and survival data were collected. Univariate analyses, multivariable binary logistic regression, Cox proportional hazards regression, receiver operating characteristic (ROC) analysis, and Kaplan–Meier survival analysis were performed. Results: The overall 90-day all-cause mortality rate was 74.7%. Age, Charlson Comorbidity Index, and APACHE II score were significantly associated with 90-day mortality in univariate analyses, whereas tracheostomy timing and technique were not. In multivariable logistic regression analysis, the Charlson Comorbidity Index and APACHE II score were independently associated with mortality. Cox proportional hazards regression confirmed that both APACHE II score and Charlson Comorbidity Index were independently associated with mortality over time. ROC analysis demonstrated moderate discriminative performance for age, Charlson Comorbidity Index, and APACHE II score. Conclusions: In critically ill patients undergoing tracheostomy, 90-day mortality was high and was driven primarily by acute illness severity and comorbidity burden rather than procedural characteristics. These findings support incorporating biological vulnerability, expected recovery potential, and goals-of-care discussions into tracheostomy decision-making.

Background: Non-thyroidal illness syndrome is frequent in critically ill patients, but the prognostic value of dynamic changes in thyroid function tests remains unclear. This study evaluated whether serial measurements of thyroid-stimulating hormone (TSH) and free triiodothyronine (FT3) provide additional predictive value for 30-day mortality beyond conventional severity scores in ICU patients. Methods: This single-center retrospective observational study included 74 adult patients treated for ≥72 h in a general ICU who had TSH and FT3 measured within 24 h of admission and repeated at 48–72 h. Patients aged 18 years or above admitted to the intensive care unit were included in the study. Demographic characteristics, comorbidities, APACHE II, SOFA, modified NUTRIC (mNUTRIC) scores, and routine laboratory data (including albumin, CRP, and lactate) were recorded. The primary outcome was 30-day mortality. Between-group comparisons were performed using t-tests, Mann–Whitney U, and Chi-square tests. Variables significant in univariate analyses were entered into binary logistic regression models, and predictive performance was assessed using receiver operating characteristic (ROC) curves and the Youden index. Results: The mean age was 68.7 ± 14.7 years, and 41.9% (n = 31) of the patients died within 30 days. Non-survivors had higher APACHE II, SOFA, and mNUTRIC scores and lower albumin, lymphocyte count, and second FT3 levels compared with survivors (all p ≤ 0.003). Baseline FT3 and TSH were not associated with mortality, whereas both the subsequent FT3 measurements and the ΔT3 (variance in former to latter FT3) were remarkably predictive. The latter FT3 < 1.63 pg/mL produced an AUC of 0.835 (sensitivity: 77%, specificity: 74%), and a ΔT3 log ratio threshold of −0.09 (≈20% early FT3 decline) produced an AUC of 0.835 (sensitivity: 71%, specificity: 81%). The APACHE II + ΔT3 (numeric) model showed the best discrimination (AUC: 0.921; sensitivity: 87.1%, specificity: 81.4%), outperforming APACHE II alone (AUC: 0.861). Conclusions: In critically ill adult patients, dynamic T3 kinetics—particularly premature decline in FT3 within the first 72 h—provide incremental prognostic value for 30-day mortality beyond APACHE II. Serial FT3 monitoring may help identify high-risk patients whose endocrine adaptation to critical illness is failing.

Background and Objectives: Procalcitonin (PCT) kinetics are increasingly used as prognostic markers in sepsis, but their interpretation is confounded by dynamic changes in renal function during acute illness. This study evaluated the prognostic value of ΔPCT for 30-day mortality in critically ill patients with either sepsis or septic shock by incorporating serial kinetic eGFR measurements and renal function-adjusted ΔPCT cut-off values based on the mean kinetic eGFR during the first 48–72 h of ICU admission. Materials and Methods: This retrospective cohort study included 106 adult ICU patients with either sepsis or septic shock. Serial procalcitonin measurements were used to calculate ΔPCT as a ratio of follow-up to baseline values, while renal function was assessed using mean kinetic eGFR over the first 72 h of ICU admission. Results: Thirty-day mortality was 43.4%. ΔPCT was a strong independent predictor of mortality across all models. At 48 h, ΔPCT2 was independently associated with 30-day mortality in the overall cohort (AUC 0.793) and demonstrated independent prognostic significance only in patients with preserved renal function (GFR ≥ 30 mL/min/1.73 m²). The optimal ΔPCT2 cut-off corresponded to a 56% reduction in procalcitonin levels. At 72 h, ΔPCT3 emerged as an independent predictor of mortality regardless of renal function. ROC analysis identified an optimal ΔPCT3 cut-off corresponding to 62% procalcitonin reduction in the overall cohort, with renal function-specific thresholds of ~50% for patients with GFR < 30 mL/min/1.73 m² and ~73% for those with preserved renal function. The combination of APACHE II score and ΔPCT3 demonstrated the highest discriminative performance for mortality prediction (AUC 0.948). Conclusions: Procalcitonin kinetics provide clinically meaningful prognostic information in sepsis when interpreted alongside dynamic renal function. While 48 h procalcitonin kinetics offer prognostic value primarily in patients with preserved renal function, 72 h ΔPCT provides renal function-independent and superior mortality discrimination. Integrating serial kinetic eGFR measurements enables renal function-adapted ΔPCT threshold determination and may improve risk stratification in critically ill septic patients.

Background and Objectives: Procalcitonin (PCT) kinetics are increasingly used as prognostic markers in sepsis, but their interpretation is confounded by dynamic changes in renal function during acute illness. This study evaluated the prognostic value of ΔPCT for 30-day mortality in critically ill patients with sepsis or septic shock by incorporating serial kinetic eGFR measurements and renal function–adjusted ΔPCT cut-off values based on the mean kinetic eGFR during the first 48-72 hours of ICU admission. Materials and Methods: This retrospective cohort study included 106 adult ICU patients with sepsis or septic shock. Serial procalcitonin measurements were used to calculate ΔPCT as ratio of follow-up to baseline values, while renal function was assessed using mean kinetic eGFR over the first 72 hours of ICU admission. Results: Thirty-day mortality was 43.4%. ΔPCT was a strong independent predictor of mortality across all models. At 48 hours, ΔPCT2 was independently associated with 30-day mortality in the overall cohort (AUC 0.793) and retained independent prognostic significance only in patients with preserved renal function (GFR ≥30 mL/min/1.73 m²). The optimal ΔPCT2 cut-off corresponded to a 56% reduction in procalcitonin levels. At 72 hours, ΔPCT3 emerged as an independent predictor of mortality regardless of renal function. ROC analysis identified an optimal ΔPCT3 cut-off corresponding to 62% procalcitonin reduction in the overall cohort, with renal function–specific thresholds of ~50% for patients with GFR

Background : Non-thyroidal illness syndrome is frequent in critically ill patients, but the prognostic value of dynamic changes in thyroid function tests remains unclear. This study evaluated whether serial measurements of thyroid-stimulating hormone (TSH) and free triiodothyronine (FT3) provide additional predictive value for 30-day mortality beyond conventional severity scores in ICU patients. Methods : This single-center retrospective observational study included 74 adult patients treated for ≥72 hours in a general ICU who had TSH and FT3 measured within 24 hours of admission and again at 48–72 hours. Demographic characteristics, comorbidities, APACHE II, SOFA, modified NUTRIC (mNUTRIC) scores, and routine laboratory data (including albumin, CRP, and lactate) were recorded. The primary outcome was 30-day mortality. Between-group comparisons were performed using t-tests, Mann–Whitney U, and Chi-square tests. Variables significant in univariate analyses were entered into binary logistic regression models, and predictive performance was assessed using receiver operating characteristic (ROC) curves and the Youden index. Results: The mean age was 68.7 ± 14.7 years, and 41.9% (n = 31) of patients died within 30 days. Non-survivors had higher APACHE II, SOFA, and mNUTRIC scores and lower albumin, lymphocyte count, and second FT3 levels compared with survivors (all p ≤ 0.003). Baseline FT3 and TSH were not associated with mortality, whereas both the second FT3 measurement and the ΔT3 (second–first FT3) were strongly predictive. Second FT3 <1.63 pg/mL yielded an AUC of 0.835 (sensitivity 77%, specificity 74%), and a ΔT3 log ratio threshold of –0.09 (≈20% early FT3 decline) produced an AUC of 0.835 (sensitivity 71%, specificity 81%). The APACHE II + ΔT3 (numeric) model showed the best discrimination (AUC 0.921; sensitivity 87.1%, specificity 81.4%), outperforming APACHE II alone (AUC 0.861). Conclusions : In critically ill adults, dynamic T3 kinetics—particularly early decline in FT3 within the first 72 hours—provide substantial incremental prognostic value for 30-day mortality beyond APACHE II. Serial FT3 monitoring may help identify high-risk patients whose endocrine adaptation to critical illness is failing.

Background: Remifentanil, an ultra-short-acting µ-receptor agonist, is used with propofol for optimal laryngeal mask airway (LMA) insertion. However, no studies have assessed its effects when combined with thiopental on LMA conditions. The combined use of thiopental and remifentanil may offer advantages, such as enhanced cardiovascular and respiratory stability. This study aims to compare the administration of thiopental with different doses of remifentanil to assess their combined effects on LMA insertion conditions and success in a prospective, randomised double-blind study. Method: A total of 100 ASA I–II patients (18–65 years), including both male and female participants, were randomly assigned to four remifentanil dose groups (0.5–3 µg.kg⁻¹). Induction involved thiopental (5 mg.kg⁻¹) after remifentanil. LMA insertion conditions were evaluated using a six-variable scale. Systolic arterial pressure (SAP), diastolic arterial pressure (DAP), mean arterial pressure (MAP), heart rate (HR), and bispectral index monitor (BIS) values, as well as apnoea duration, eyelash reflex loss time, and insertion attempts, were recorded at baseline, before insertion, and at 5 min post-insertion. Results: Time to eyelash reflex loss and LMA insertion were shorter in Groups III and IV than in Groups I and II (p < 0.001). Apnoea duration was longest in Group IV, followed by Group III (p < 0.001). Groups III and IV had significantly better LMA placement, mouth opening, and ease of insertion (p < 0.05). Coughing and gagging were highest in Group I (p < 0.001). SAP, MAP, HR, and DAP were significantly lower in Group IV at various time points (p < 0.05). HR was significantly higher in Group I compared to Groups II and III at multiple time points (p < 0.05). Conclusions: The administration of 5 mg.kg⁻¹ thiopental with 2 μg.kg⁻¹ remifentanil has been found to provide a stable haemodynamic response and 96% excellent or satisfactory laryngeal mask insertion conditions without increasing the duration of apnoea.

Background and Objectives: Dexmedetomidine is a potent selective α₂ receptor agonist with analgesic and sedative effects. Many reports indicate that compared to fentanyl, the combination of dexmedetomidine with propofol provides comparably acceptable conditions for a laryngeal mask airway (LMA). However, no study has evaluated the effectiveness of combined dexmedetomidine and thiopental in LMA insertions compared to that of combined dexmedetomidine and propofol. This prospective, randomized, double-blind study aimed to compare the effects of dexmedetomidine with thiopental or propofol on LMA insertion conditions, hemodynamic responses, and pharyngolaryngeal morbidity, which in this study was defined as the presence of postoperative sore throat, dysphagia, or visible blood in the airway following a laryngeal mask airway (LMA) insertion. Materials and Methods: A total of 80 premedicated ASA I-II patients aged 18–65 years were randomized to the propofol group (Group P, n = 40) or thiopental group (Group T, n = 40). Anesthesia was induced by infusing 1 μg·kg⁻¹ dexmedetomidine over 10 min followed by 2.5 mg·kg⁻¹ propofol or 5 mg·kg⁻¹ thiopental. LMA insertion conditions were evaluated on a scale assessing six variables. Systolic arterial pressure (SAP), diastolic arterial pressure (DAP), mean arterial pressure (MAP), heart rate (HR), and bispectral index values were recorded at baseline; 1 min before; and at 1, 2, 3, 4, and 5 min after an LMA insertion. The baseline values for the systolic arterial pressure (SAP), diastolic arterial pressure (DAP), mean arterial pressure (MAP), heart rate (HR), and bispectral index (BIS) values were recorded before dexmedetomidine infusion. Measurements for all patients were then taken 1 min before and at 1, 2, 3, 4, and 5 min after the LMA insertion Results: Demographic data were similar between the groups. In Group P, the time to loss of eyelash reflex and LMA insertion time were significantly shorter, the apnea duration was significantly longer, and the rates of full jaw opening and optimal LMA insertion conditions were significantly higher when compared with those of Group T (p < 0.05). Group P showed a significantly greater percentage decrease in HR compared to that of Group T at 1 min before and 1, 2, and 3 min after the LMA insertion (p < 0.05). Group T had a greater decrease in SAP and MAP at 1 min before insertion, while the SAP decrease was lower in Group T at 3, 4, and 5 min after insertion. The MAP and DAP values after the LMA insertion showed a greater decrease in Group P compared to in Group T (p < 0.05). The incidence of bradycardia was significantly (p < 0.05) higher in Group P than in Group T. There was no significant difference between the groups in terms of the frequency of hypotension, sore throat, presence of blood, or dysphagia at discharge from the recovery unit (p > 0.05). Conclusions: This study showed that the use of dexmedetomidine with thiopental provided comparably acceptable LMA insertion conditions with more stable hemodynamics compared to propofol.