Development and validation of CAVE score in predicting presence of pressure ulcer in intensive care patients

Risk predictions of hospital-acquired pressure injury in the intensive care unit based on a machine learning algorithm

Pressure injury (PI), or local damage to soft tissues and skin caused by prolonged pressure, remains controversial in the medical world. Patients in intensive care units (ICUs) were frequently reported to suffer PIs, with a heavy burden on their life and expenditures. Machine learning (ML) is a Section of artificial intelligence (AI) that has emerged in nursing practice and is increasingly used for diagnosis, complications, prognosis, and recurrence prediction. This study aims to investigate hospital-acquired PI (HAPI) risk predictions in ICU based on a ML algorithm by R programming language analysis. The former evidence was gathered through PRISMA guidelines. The logical analysis was applied via an R programming language. ML algorithms based on usage rate included logistic regression (LR), Random Forest (RF), Distributed tree (DT), Artificial neural networks (ANN), SVM (Support Vector Machine), Batch normalisation (BN), GB (Gradient Boosting), expectation-maximisation (EM), Adaptive Boosting (AdaBoost), and Extreme Gradient Boosting (XGBoost). Six cases were related to risk predictions of HAPI in the ICU based on an ML algorithm from seven obtained studies, and one study was associated with the Detection of PI risk. Also, the most estimated risksSerum Albumin, Lack of Activity, mechanical ventilation (MV), partial pressure of oxygen (PaO2), Surgery, Cardiovascular adequacy, ICU stay, Vasopressor, Consciousness, Skin integrity, Recovery Unit, insulin and oral antidiabetic (INS&OAD), Complete blood count (CBC), acute physiology and chronic health evaluation (APACHE) II score, Spontaneous bacterial peritonitis (SBP), Steroid, Demineralized Bone Matrix (DBM), Braden score, Faecal incontinence, Serum Creatinine (SCr) and age. In sum, HAPI prediction and PI risk detection are two significant areas for using ML in PI analysis. Also, the current data showed that the ML algorithm, including LR and RF, could be regarded as the practical platform for developing AI tools for diagnosing, prognosis, and treating PI in hospital units, especially ICU.

The Direct Effects of Norepinephrine Administration on Pressure Injuries in Intensive Care Patients: A Retrospective Cohort Study

OBJECTIVE

To estimate the direct effects of norepinephrine administration on pressure injury (PI) incidence in intensive care patients.

METHODS

This is a secondary and exploratory analysis of a retrospective cohort study of intensive care patients discharged in 2017 to 2018. Observational cases only included patients who received primary PI preventive care during intensive care (N = 479). As a first-choice vasopressor drug, norepinephrine administration was approximated with days of norepinephrine. Linear path models were examined from norepinephrine administration to PI development. The identification of confounding variables and instrumental variables was grounded on directed acyclic graph theory. Direct effects were estimated with instrumental variables to overcome bias from unobserved variables. As models were re-specified with data analysis, the robustness of path identification was improved by requiring graph invariance with sample split.

RESULTS

Norepinephrine caused PI development from one stage to another after 4.0 to 6.3 days of administration in this cohort as a total effect (90% CI). The direct effect was estimated to advance the stage of PI at a rate of 0.140 per day of norepinephrine administered (standard error, 0.029; P < .001). The direct effect accounted for about 70% of the total effect on PI development.

CONCLUSIONS

Estimations with instrumental variables and structural equation modeling showed that norepinephrine administration directly and substantially affected hospital-acquired PI incidence in intensive care patients in this cohort.

Retrospective study on the factors influencing the severity of pressure injuries among intensive care unit patients

AIMS AND OBJECTIVES

This study investigated the factors of pressure injuries in patients in intensive care units (ICU).

BACKGROUND

Pressure injuries among patients in ICUs can be prevented by the early assessment of risk factors and taking appropriate preventative measures.

DESIGN

A retrospective study.

METHODS

ICU patients who suffered from pressure injuries between January 2016 and August 2018 at a hospital in Taiwan were selected. Patient medical histories and data associated with pressure injuries and medical treatment were collected from electronic medical records. A total of 256 patients were included in our analysis. The study adhered to the STROBE checklist.

RESULTS

A multivariate model of multinomial logistic regression analysis indicated that when compared to Stage 1 pressure injuries, Stage 2-4 pressure injuries were associated with albumin levels. Patients with lower albumin levels were at greater odds of Stage 2-4 pressure injuries than Stage 1 pressure injuries. In addition, when compared to unstageable pressure injuries or deep tissue injuries, Stage 1 pressure injuries were associated with the use of fentanyl and haemoglobin levels. Patients using fentanyl were more likely to suffer from unstageable pressure injuries, and those with lower haemoglobin levels were more likely to suffer from unstageable pressure injuries or deep tissue injuries.

CONCLUSION

When patients were simultaneously on ventilators and taking midazolam or fentanyl, the incidence of pressure injuries with greater severity became higher. Furthermore, ICU patients with lower albumin and haemoglobin levels were more likely to suffer from pressure injuries of greater severity.

RELEVANCE TO CLINICAL PRACTICE

For critically ill patients on ventilators, it is recommended to devise a means of assessing each patient daily as well as systematically reduce their dosage of midazolam or fentanyl. Furthermore, regularly monitoring albumin and haemoglobin levels to understand their nutritional status is necessary.

Singular Predictors of Hospital-Acquired Pressure Injuries Under Intensive Care: A Retrospective Cohort Study in a Medium-Complexity Hospital

A retrospective cohort study of hospital-acquired pressure injuries (HAPI) reported an incidence rate of 34.3% based on 582 medical records of adult patients admitted to the intensive care unit (ICU) of a medium-complexity public hospital in 2017 and 2018. Sixty percent of the patients used respirators, 49.3% presented hypotension, and 48.1% used norepinephrine. The main individual predictors of HAPI in the ICU were "days of norepinephrine" with an odds ratio (OR) of 1.625 (95% CI: 1.473-1.792) and concordance statistic (AUC) of 0.818 (95% CI: 0.779-0.857), "days of mechanical ventilation" with an OR of 1.521 (1.416-1.634) and AUC of 0.879 (0.849-0.909), "ICU stay (days)" with an OR of 1.279 (1.218-1.342) and AUC of 0.846 (0.812-0.881), and "Braden's sensory perception" with an OR of 0.345 (95% CI: 0.278-0.429) and AUC of 0.760 (0.722-0.799). The duration of mechanical ventilation, norepinephrine administration, and ICU length of stay presented significant discriminative capacity for HAPI prediction.

Translation and piloting of the Chinese Mandarin version of an intensive care-specific pressure injury risk assessment tool (the COMHON Index)

Objective

Methods

Results

Conclusions

Limb edema in critically ill patients: Comparing intermittent compression and elevation

Intensive care unit (ICU) patients are at high risk for limb edema, which caused complications such as pain, joint contracture, limited range of motion and atrophy of the limbs. Thus, this study was conducted to compare ICU patients' upper limb edema between two groups with the intervention of limb elevation and intermittent pneumatic compression (IPC). In this quasi‐experimental before and after study, 40 patients were recruited. One upper limb was randomly assigned to the upper limb elevation (ULE) group and the other one was assigned to the IPC group. The circumference of the wrist and the middle of the arm were compared between and within groups. Results showed that in both groups of IPC and ULE and all five sessions (unless the second session of ULE), participants' arm and wrist edema were reduced significantly after the interventions (arm: P < .01; wrist: P < .0001). The differences between the two groups of ULE and IPC in regards to limb edema reduction were not significant. Although there was no significant difference between IPC and ULE intervention in the removal of edema, ULE seems to be more feasible and practical, which should be assessed in future studies.

Selected Biochemical Blood Parameters and a Risk of Pressure Ulcers in Patients Receiving Treatment in Intensive Care Units

Background and Objectives: This study aimed to assess the level of selected biochemical parameters in venous blood and their potential effects on the development of pressure ulcers in patients treated in intensive care settings. Materials and Methods: Fifty patients hospitalised in an intensive care unit (ICU) were enrolled for the study. The methods used included controlled observation, literature review and medical record analysis. The observation protocol applied in the study consisted of two parts comprising the basic information, sociodemographic data, results of laboratory tests (CRP, PCT, albumin, protein and haemoglobin concentrations) as well as the Braden Scale for Predicting Pressure Ulcer Risk. Results: The subjects presented moderate to high risk of pressure ulcers, reflected by the mean score of 8.18 ± 1.3 points, with minimum and maximum scores of 6 and 12 points, respectively. Normal albumin level was identified in only five subjects (10.0%) while 45 subjects (90.0%) were found with results below the norm. A statistical relationship was observed between such variables as albumin concentration (p < 0.01) and total protein level (p = 0.007). The findings show a strong correlation between the score in the Braden Scale and the level of albumins (R = 0.55). Conclusions: In our study, lower concentrations of albumins and total proteins correspond to a greater risk of pressure ulcers.