Construction Of A Risk Prediction Model For Artificial Airway-associated Pressure Injury In ICU Patients

Objectives:To analyze the risk factors for the occurrence of artificial airway-related pressure injury and construct a risk prediction model for artificial airway-related pressure injury in ICU patients.Risk assessment of ICU artificial airway patients by the constructed risk prediction line graph model,and early intervention for patients scoring as high risk to reduce the incidence of artificial airway-related pressure injury.By screening out high-risk risk factors,nurses’ awareness of prevention of artificial airway-related pressure injuries was enhanced.1.Risk assessment content and feasibility study of factors influencing pressure injury associated with artificial airway in ICU patients : Firstly,we searched domestic and abroad databases such as CNKI（China National Knowledge Infrastructure）,collect the document published literature on factors influencing pressure injury associated with artificial airway,to construct a draft of risk assessment.Secondly,an expert panel discussion was conducted on the draft risk assessment of pressure-related injury associated with an artificial airway.The risk factors of pressure injury associated with an artificial airway were determined through the expert group discussion.Finally,a pre-survey was conducted in the ICU ward of a tertiary care hospital in Zhejiang Province.2.Construction of a prediction model for the risk of pressure injury associated with an artificial airway in ICU patients:A literature study and expert panel discussion were used to determine the influencing factors of artificial airway-related pressure injuries,and 650 ICU artificial airway patients who met the inclusion and exclusion criteria and were admitted to a tertiary care hospital in Hangzhou,Zhejiang Province,from December 2020 to July 2021 were selected for the study.The indices of the group with the occurrence of pressure injury related to the artificial airway（131 cases）and the group without pressure injury（519 cases）were compared,and independent risk factors for the occurrence of pressure injury related to the artificial airway in ICU patients were explored by single-factor analysis and multi-factor logistic regression analysis,and risk prediction models were established and column plots were constructed using the HosmerLemeshow chi-square test was used to test the fit of the model;the consistency index was used to test the discriminatory ability of the model,and Bootstrap method was used to internally validate the prediction effect of the model,and the decision curve analysis was used to assess the clinical application value of the model.3.External validation of a predictive model for the risk of pressure-related injury in ICU patients with artificial airway: 300 ICU patients with artificial airway who met the inclusion and exclusion criteria admitted to a tertiary care hospital in Hangzhou,Zhejiang Province,from August to October 2021 were selected for the study.The external data were brought into the original line graph risk prediction model for validation,and the accuracy of the model prediction was judged by using the prediction accuracy rate;the discriminatory ability of the model was tested by using the consistency index;and the Bootstrap method was used to externally validate the model prediction effect,and the decision curve analysis was used to assess the clinical application value of the model.Results:Of the 650 patients with artificial airways included,131 artificial airway-related pressure injuries occurred,with an incidence of 20.2%,most of which were stage 2 and 3 pressure injuries.Multifactorial logistic regression analysis showed that the duration of retention of the artificial airway（OR=1.035）,the use of subglottic suction catheter（OR=2.878）,the position of the tracheal tube in the mouth（OR=4.937）,the moist score in the Braden score during intubation（OR=0.215）,the mobility score（OR=0.225）,the friction and shear scores（OR=0.170）were independent risk factors for the occurrence of pressure-related injuries in ICU patients with artificial airways.The column line graph model for predicting the risk of pressure-related injury associated with artificial airways in ICU patients was constructed based on the six risk factors mentioned above,with an H-L goodness of fit test（χ2=12.432,p=0.133）,a consistency index of 0.842（95%CI80.4%to 87.9%）,a sensitivity of 77.1%,a specificity of 78.4%,a maximum Yordon index of0.563,corresponding to a risk prediction value of 0.170 and a column line graph prediction threshold of 177 points.The internal validation results showed a consistency index of 0.835,a sensitivity of 79.2%,a specificity of 75.3%and a model Brier value of 0.112.The calibration curve showed that the prediction curve and the actual observed curve were in general agreement,and the decision curve analysis showed that the column line graph model had good clinical application;the external model validation results showed an unadjusted consistency index of0.832（95%CI was 0.781-0.882）,sensitivity was 77.8%,specificity was 73.8%,the calibration curve shown in the calibration graph was near the ideal curve,the model Brier value was 0.125,the accuracy of the model prediction was 76.3%,and the decision curve analysis showed that the column line graph model had a good clinical application value.Conclusion:In this study,we constructed a line graph model for predicting the risk of artificial airway-related pressure injury in ICU patients with good predictive efficacy,and the risk assessment of pressure injury during the patient’s indwelling artificial airway provides a reference for early intervention of artificial airway-related pressure injury.Patients with an artificial airway in the ICU were evaluated in a timely manner considering the duration of the artificial airway,the use of a subglottic catheter,the position of the tracheal tube in the mouth,the moist score in the Braden score,the mobility score,and the friction and shear scores,etc.enables early identification of patients at risk for artificial airway-related pressure injuries and early intervention for patients with a predicted score ≥177 to reduce the incidence of artificial airway-related pressure injuries.