OBJECTIVE: To analyze the efficacy and access safety of extracorporeal membrane oxygenation (ECMO) in series with continuous renal replacement therapy (CRRT) access for critically ill patients using propensity score matching analysis, and to explore the potential influencing factors of infection. METHODS: A total of 200 critically ill patients who received both ECMO and CRRT treatment in the intensive care unit (ICU) of Huai'an Second People's Hospital from December 2020 to December 2024 were retrospectively selected as the research subjects. They were divided into the independent operation group (72 cases) and the series system group (128 cases) according to the access connection mode of ECMO and CRRT. Propensity score matching analysis was used to perform 1 : 1 matching for patients of the two groups. The general data [age, gender, body mass index (BMI), clinical diagnosis, underlying disease, intubation method, intubation position, disease severity, ECMO support duration, catheter indwelling duration, oxygenation index (PaO₂/FiO₂) at 48 hours after ECMO initiation, serum creatinine (SCr), procalcitonin (PCT), hemoglobin (Hb), white blood cell count (WBC), platelet count (PLT)], treatment status [ECMO initiation duration, ECMO operation duration, ECMO flow, left ventricular ejection fraction (LVEF), CRRT initiation duration, CRRT catheter indwelling duration, inflow and outflow volume of replacement fluid], clinical outcome indicators (28-day survival rate, length of ICU stay, renal function recovery, fluid balance compliance rate), and access safety indicators (incidence of ECMO access thrombosis, incidence of infection, and incidence of bleeding events) of all the patients were collected. Subgroup analysis was conducted based on the occurrence of infection, and multivariate Logistic regression analysis was used to screen the potential risk factors for infection in critically ill patients receiving both ECMO and CRRT treatment. RESULTS: Finally, a total of 120 patients were successfully matched, with 60 patients in both the independent operation group and the series system group. No statistically significant differences were observed in the general data between the two groups, indicating comparability. Compared with the independent operation group, the ECMO flow at 48 hours after ECMO initiation, SCr, and alanine transaminase (ALT) of the patients in the series system group were significantly decreased, while the LVEF at 48 hours after ECMO initiation was significantly increased, additionally, the CRRT initiation duration, CRRT catheter indwelling duration, and the length of ICU stay were significantly shortened, and the inflow and outflow volume of replacement fluid were significantly increased. The incidence of infection and bleeding events in the series system group was significantly lower than that in the independent operation group [infection incidence: 11.67% (7/60) vs. 36.67% (22/60), bleeding event incidence: 8.33% (5/60) vs. 48.33% (29/60), both P < 0.05]. No significant difference was found in the other general data, treatment status, clinical outcome indicators, or access safety indicators between the two groups. Among the 120 patients, 29 cases developed infection (accounting for 24.17%), and 91 cases had no infection (accounting for 75.83%). Compared with the non-infection group, the catheter indwelling duration was significantly prolonged and PCT was significantly increased in the infection group, while the PLT and the proportion of patients with ECMO and CRRT access connected via the series system were significantly decreased. Multivariate Logistic regression analysis showed that catheter indwelling duration [odds ratio (OR) = 1.277, 95% confidence interval (95%CI) was 1.001-1.629, P = 0.049], PCT (OR = 1.529, 95%CI was 1.222-1.914, P < 0.001], PLT (OR = 0.953, 95%CI was 0.926-0.981, P = 0.001), and access connection mode (OR = 0.289, 95%CI was 0.090-0.930, P = 0.037) were potential risk factors for infection in critically ill patients. CONCLUSIONS The ECMO-in-series CRRT access can accelerate the initiation of CRRT, avoid local bleeding, stabilize patients' cardiac, hepatic and renal functions, reduce potential infection risks, and improve the prognosis of patients.
Humans ; Extracorporeal Membrane Oxygenation/adverse effects* ; Critical Illness/therapy* ; Retrospective Studies ; Continuous Renal Replacement Therapy ; Male ; Female ; Intensive Care Units ; Propensity Score ; Middle Aged ; Renal Replacement Therapy ; Adult ; Aged ; Risk Factors

Objective To study the utility and feasibility of CRAMS score to assess prognosis of patients with acute trauma.Methods A retrospective review of 1 802 patients with acute trauma was conducted to calculate CRAMS and ISS score.Receiver operation characteristic curve (ROC) was used to measure the prognostic role of CRAMS in comparison with ISS.Results Area under the curve (RUC) was 0.885 for CRAMS (95 ％ CI 0.870-0.900) and was 0.792 for ISS (95％ CI 0.773-0.811),with statistical difference of the two scoring systems (Z=4.280,P ＜0.01).To identify patients with potential critical illness,optimal cut-off point was≤7 for CRAMS and≥24 for ISS.CRAMS presented better sensitivity (x2 =16.910,P ＜ 0.01),but lower specificity (x2 =5.260,P ＜ 0.05) and accuracy (x2 =0.693,P ＞ 0.05) for predicting mortality when compared with ISS.Conclusions CRAMS is better than ISS in predicting prognosis for patients with acute trauma and exhibits a high discrimination.RAMS has advantages of simple operation,easy grasping and accurate and timely reflection of illness severity,which facilitates the early detection and treatment of critical illness in inhospital trauma patients.

Objective To evaluate the application value of prehospital index (PHI) in hospitalized patients with acute trauma.Methods A study was done in 1802 hospitalized patients with acute trauma by random sampling.PHI and injury severity score (ISS) were made respectively.Receiver operating characteristic curve (ROC curve) was used for detecting optimal cut-off point by taking the date of discharge as the endpoint and the outcome as observed indicator.In order to compare the predicting prognosis value of PHI in hospitalized patients with acute trauma,the corresponding predicting indicators were calculated and ISS was used for reference.Results The area under the ROC curve was 0.871 (95％ CI:0.855-0.886) by the score of PHI and 0.792 (95％ CI:0.773-0.811) by the score of ISS,and there was statistically difference between the two scoring systems (Z =2.674,P =0.007),and the optimal cut-off point was used for judging the potential for critically ill patients when PHI was ≥ 4,ISS ≥ 22.The sensitivity predictors of critically ill patients death of PHI was superior to ISS (x2 =6.975,P =0.008),the specificity and the accuracy of PHI and ISS showed no significant difference (P ＞ 0.05).Conclusions PHI has high potential for assessing patient condition and predicting the death of hospitalized patients with acute trauma,and it is equivalent to ISS in prediction value.The advantages of PHI are simple in operation,easy to learn,reflecting the condition timely and reliably,suitable for dynamic evaluation and comparison,which is suitable for critical patients with trauma of preliminary screening.