Objective To investigate the clinical characteristics and influencing factors of pressure injury in patients with severe neurological diseases and to construct and evaluate a predictive model for it.Methods A retrospective research method was adopted to collect 250 patients with severe neuropathy admitted to the First Affiliated Hospital of Wenzhou Medical University from April 2020 to April 2024,and their clinical characteristics were collected.The patients were then divided into a pressure injury group(n=58)and a non-pressure injury group(n=192)based on whether they development pressure injury after treatment.Baseline data on patient coma or lethargy status,primary diagnosis requiring neurocritical care admission,and Acute Physiology and Chronic Health Evaluation(APACHE)Ⅱscores were collected.The area under the curve(AUC)of the receiver operating characteristic(ROC)curves for acute cerebrovascular disease,coma or lethargy status,and APACHE Ⅱ scores of the subjects was compared.Results Among the 250 patients with severe neurological diseases,58 had pressure injuries.Of these,35(60.34%)had mucosal pressure injuries,while 23(39.66%)had device-related pressure injuries.According to the National Pressure Injury Advisory Panel Pressure Injury Staging System,46 cases(79.31%)had stage 1 pressure injuries,8 cases(13.97%)had stage 2 pressure injuries,4 cases(6.90%)had stage 3 pressure injuries,and no patients had stage 4 pressure injuries.Logistic multivariate regression analysis showed that primary diagnosis requiring neurocritical care admission(odds ratio[OR]=3.102;95%CI,1.013-9.499),coma or lethargy status(OR=3.769;95%CI,1.237-11.478),and APACHE Ⅱ score(OR=0.201;95%CI,0.124-0.328)were influencing factors for pressure injury in patients with severe neurological diseases.The ROC results showed that the AUC of the prediction model combining the 3 influencing factors was 0.974(95%CI,0.957-0.992),and that the sensitivity and specificity were 91.40%and 93.70%,respectively.The prediction accuracy of the combination prediction model was 0.96,which was significantly higher than those of the prediction models based on the 3 separate influencing factors(P<0.05).The Hosmer-Lemeshow test showed that the model had a good fit(χ2=4.779,P=0.062),indicating that the model had a relatively high accuracy.Conclusion Acute cerebrovascular disease,coma or lethargy,and APACHE Ⅱ score have different predictive values for pressure injury in patients with severe neurological diseases.While acute cerebrovascular disease and coma or lethargy have the same predictive value separately,the combination prediction incorporating the 3 influencing factors demonstrated superior accuracy and holds considerable potential for clinical application.

Objective:To construct logistic regression, decision tree, and neural network models to predict pressure injury in neurocritically ill patients using machine learning algorithms, and compare the predictive performance of the three models.Methods:The clinical data of 341 neurocritically ill patients who received treatment in the Department of Neurosurgery at The First Affiliated Hospital of Wenzhou Medical University from May 2020 to February 2023 were collected retrospectively. The patients were randomly divided into a training set and a testing set in a 7:3 ratio. Univariate and multivariate analyses were conducted based on the clinical data from the training set. According to the results of the multivariate analysis, logistic regression, decision tree, and neural network models were constructed. The predictive performance of the three models was validated and compared using receiver operating characteristic curve analysis.Results:Among the 341 patients, 35 developed pressure injury (a total of 40 occurrences), with an incidence rate of 10.26%. Multivariate analysis indicated that incontinence ( OR = 47.32, 95% CI: 1.360-1 647.700), decreased albumin levels ( OR = 0.56, 95% CI: 0.360-0.870), increased sensory ability ( OR = 0.00, 95% CI: 0.000-0.190), and increased mobility ( OR = 0.03, 95% CI: 0.000-0.390) were independent risk factors for pressure injury in neurocritically ill patients (all P < 0.05). Based on these independent risk factors, logistic regression, decision tree, and neural network models were constructed. Receiver operating characteristic curve analysis revealed that the area under the curve for the three models was 0.987 (95% CI: 0.941-0.999), 0.945 (95% CI: 0.881-0.980), and 0.908 (95% CI: 0.834-0.956), respectively. These results suggest that all three models exhibited high predictive performance for pressure injury in neurocritically ill patients, with the logistic regression model showing a significantly greater area under the curve than the neural network model. Conclusions:The occurrence of pressure injury in neurocritically ill patients is closely related to incontinence, albumin levels, sensory ability, and mobility. Constructing predictive models using machine learning algorithms can provide valuable insights for the early prevention and management of pressure injury in neurocritically ill patients.

Objective:To construct logistic regression, decision tree, and neural network models to predict pressure injury in neurocritically ill patients using machine learning algorithms, and compare the predictive performance of the three models.Methods:The clinical data of 341 neurocritically ill patients who received treatment in the Department of Neurosurgery at The First Affiliated Hospital of Wenzhou Medical University from May 2020 to February 2023 were collected retrospectively. The patients were randomly divided into a training set and a testing set in a 7:3 ratio. Univariate and multivariate analyses were conducted based on the clinical data from the training set. According to the results of the multivariate analysis, logistic regression, decision tree, and neural network models were constructed. The predictive performance of the three models was validated and compared using receiver operating characteristic curve analysis.Results:Among the 341 patients, 35 developed pressure injury (a total of 40 occurrences), with an incidence rate of 10.26%. Multivariate analysis indicated that incontinence ( OR = 47.32, 95% CI: 1.360-1 647.700), decreased albumin levels ( OR = 0.56, 95% CI: 0.360-0.870), increased sensory ability ( OR = 0.00, 95% CI: 0.000-0.190), and increased mobility ( OR = 0.03, 95% CI: 0.000-0.390) were independent risk factors for pressure injury in neurocritically ill patients (all P < 0.05). Based on these independent risk factors, logistic regression, decision tree, and neural network models were constructed. Receiver operating characteristic curve analysis revealed that the area under the curve for the three models was 0.987 (95% CI: 0.941-0.999), 0.945 (95% CI: 0.881-0.980), and 0.908 (95% CI: 0.834-0.956), respectively. These results suggest that all three models exhibited high predictive performance for pressure injury in neurocritically ill patients, with the logistic regression model showing a significantly greater area under the curve than the neural network model. Conclusions:The occurrence of pressure injury in neurocritically ill patients is closely related to incontinence, albumin levels, sensory ability, and mobility. Constructing predictive models using machine learning algorithms can provide valuable insights for the early prevention and management of pressure injury in neurocritically ill patients.

Objective To explore the effect of using cluster nursing measures on expectoration in mechanical ventilation patients after craniocerebral operation. Methods Convenient sampling and controlled trials at not the same period were used. Sixty-seven mechanical ventilation patients after craniocerebral operation were selected as the research objects in Department of Neurosurgery Intensive Care Unit (ICU) of the First Hospital Affiliated to Wenzhou Medical College. Thirty-two patients treated from June 2015 to June 2016 were assigned in the control group, and they were given routine respiratory nursing care; 35 patients admitted and treated from July 2006 to July 2017 were included in the intervention group, and they were given evidence-based cluster nursing intervention measures on the basis of routine care. The differences in expectoration effect, arterial blood gas analysis index, incidence of pulmonary infection and prognosis of patients in two groups were compared. Results Compared with control group, the amount of expectoration in the intervention group was significantly increased (mL/d: 49.69±9.45 vs. 33.72±10.63, P < 0.05), while the daily number of sputum suction (times: 21.57±2.31 vs. 28.76±22.66), the time needed for each sputum suction(s: 6.81±1.74 vs. 9.28±2.52), respiratory frequency (bpm: 26.26±1.83 vs. 28.58±1.36), incidence of pulmonary infection [0 vs. 12.5% (4/32)], time of mechanical ventilation (days: 6.37±2.51 vs. 8.92±3.32), time of ICU stay (days: 7.49±3.87 vs. 10.33±2.12), time of hospital stay (days: 10.31±1.99 vs. 14.56±3.57), fatality rate [8.6% (3/35) vs. 21.9% (7/32)] in the intervention group were significantly decreased (all P < 0.05); after treatment the arterial partial pressure of oxygen (PaO2) and pulse oxygen saturation degree (SpO2) were significantly increased than those before treatment, and the arterial partial pressure of carbon dioxide (PaCO2) was significantly decreased than that before treatment, and the degrees of improvement in the intervention group on 5 days were significantly better than those in the control group [PaO2(mmHg, 1 mmHg = 0.133 kPa): 60.89±3.44 vs. 57.34±2.49, PaCO2(mmHg): 41.06±4.32 vs. 45.22±4.78, SpO2: 0.986±0.030 vs. 0.963±0.023, all P < 0.05]. Conclusion The cluster nursing measures can effectively improve the expectoration effect for mechanical ventilation patients after craniocerebral surgery, reduce the mortality and incidence of pulmonary infection, shorten the stay in ICU and improve the prognosis, suggesting that the measures be worthy to be applied widely in clinics.