Objective: To analyze the risk factors of acute pancreatitis complicated by acute respiratory distress syndrome(Acute respiratory distress syndrome,ARDS),and establish a risk assessment model for acute pancreatitis complicated by ARDS.Methods: Clinical data of patients admitted to the emergency room of General Hospital of Eastern Theater Command,PLA,from July 2019 to July 2020 who were diagnosed with acute pancreatitis and subsequently hospitalized were retrospectively collected.Followed up for 28 days,according to the diagnosis of acute pancreatitis after 2 weeks if complicated with ARDS is divided into ARDS group and Non-ARDS group,And record the patient’s vital signs and other clinical data,demographic data(including gender,age,smoking,drinking or not),laboratory indicators(including blood routine,blood biochemistry,blood gas analysis),etc.;within 2 hours of admission Perform APACHE Ⅱ score and SOFA score,and perform RANSON score within 48 hours;clinical data such as hospitalization days,secondary infection,and intubation.All statistical data were statistically processed by SPSS 25.0 and Med Calc 19.1 software.The clinical data of hospital Logistic analysis for single factors,and clinical significance of statistically differences of indicators and multi-factor Logistic analysis,risk assessment model is established,and draw the risk assessment model,APACHE Ⅱscores,RANSON score and SOFA score ROC curve.Through the comparison between different ROC curve and AUC,To observe the risk assessment model,APACHE Ⅱscores and SOFA score for the evaluation of patients with acute pancreatitis complicated with ARDS.Results: 238 patients with acute pancreatitis complicated with ARDS 59(24.8%),of which 27(45.8%)with mild ARDS,20(33.9%)and 18(30.5%)with moderate and severe ARDS,respectively;Compared the ARDS group with the non-ARDS group,the degree of acute pancreatitis in the ARDS group [3(3,3)VS.2(1,2),P <0.001],whether the tracheal intubation [36(55.4%)VS.0(0%),P < 0.001],28-day mortality [12(18.5%)VS.2(1.2%),P<0.001],etc.were significantly higher than the non-ARDS group;Compared with the non-ARDS group,the ARDS group has T,HR,RR,RDW,CRP,PT,ATⅢ,APTT,D-Dimer,Fi B,Lac,OI,PCT,IL-6,Alb,BUN,SCr,LDH,admission The differences between 6h blood glucose,secondary infection,ionized calcium,and serum calcium were statistically significant(P<0.05),where RDW>13.2%(OR =2.400,P=0.001),Lac>1.2mmol/L(OR=1.934,P=0.014),OI<217.24 mm Hg(OR=0.990,P=0.001),Alb<31.3g/L(OR=0.866,P=0.013)are related to acute pancreatitis complicated by ARDS,and the risk is established based on the above factors Evaluation model 1.After adding fluid resuscitation factors,RR>22 times/min(OR=1.109,P=0.034),RDW>13.2%(OR=2.154,P=0.004),D-Dimer>2.87mg/L(OR=1.111,P =0.05),Lac>1.2mmol/L(OR=2.269,P=0.006),OI<217.24 mm Hg(OR=0.992,P=0.018),8-hour colloid input(OR=1.013,P=0.01),total Cholesterol(OR=0.703,P=0.05)is related to acute pancreatitis complicated by ARDS,and a risk assessment model 2 is established.Risk assessment model 2 The area under the ROC curve(AUC=0.889,0.841~0.926)is larger than the risk assessment model 1(AUC=0.884,0.836~0.922),APACHEⅡ score(AUC=0.864,0.813~0.905),SOFA score(AUC=0.829),0.775~0.874)and RANSON score(AUC=0.816,0.761~0.863),The sensitivity of risk assessment model 1 to acute pancreatitis complicated by ARDS(85.94%)is greater than that of risk assessment model 2(82.81%),APACHEⅡ score(70.77%),SOFA score(67.69%)and RANSON score(84.62%).Conclusion:1.The assessment ability of risk assessment model 1 and risk assessment model 2for patients with acute pancreatitis complicated by ARDS is greater than APACHE Ⅱscore,SOFA score and RANSON score;2.Risk assessment model 1 has better sensitivity than risk assessment model 2,APACHEⅡ score,SOFA score and RANSON score,and is more suitable as a risk assessment model for acute pancreatitis complicated by ARDS. |