Heterogeneity hampers the identification of general pressure injury risk factors in intensive care populations: A predictive modelling analysis

Prediction model of pressure injury occurrence in diabetic patients during ICU hospitalization--XGBoost machine learning model can be interpreted based on SHAP

BACKGROUND

The occurrence of pressure injury in patients with diabetes during ICU hospitalization can result in severe complications, including infections and non-healing wounds.

AIMS

The aim of this study was to predict the occurrence of pressure injury in ICU patients with diabetes using machine learning models.

STUDY DESIGN

In this study, LASSO regression was used for feature screening, XGBoost was employed for machine learning model construction, ROC curve analysis, calibration curve analysis, clinical decision curve analysis, sensitivity, specificity, accuracy, and F1 score were used for evaluating the model's performance.

RESULTS

Out of the 503 ICU patients with diabetes included in the study, pressure injury developed in 170 cases, resulting in an incidence rate of 33.8 %. The XGBoost model had a higher AUC for predicting pressure injury in patients with diabetes during ICU hospitalization (train: 0.896, 95 %CI: 0.863 to 0.929; test: 0.835, 95 % CI: 0.761-0.908). The importance of SHAP variables in the model from high to low was: 'Days in ICU', 'Mechanical Ventilation', 'Neutrophil Count', 'Consciousness', 'Glucose', and 'Warming Blanket'.

CONCLUSION

The XGBoost machine learning model we constructed has shown high performance in predicting the occurrence of pressure injury in ICU patients with diabetes. Additionally, the SHAP method enables the interpretation of the results provided by the machine learning model.

RELEVANCE TO CLINICAL PRACTICE

Improve the ability to predict the early occurrence of pressure injury in diabetic patients in the ICU. This will enable clinicians to intervene early and reduce the occurrence of complications.

Medical Device-Related Pressure Injury Care and Prevention Training Program (DevICeU): Effects on intensive care nurses' knowledge, prevention performance and point prevalence

OBJECTIVE

To determine the effect of the training given to intensive care unit (ICU) nurses to prevent medical device-related pressure injuries (MDRPIs) on nurses' knowledge levels, their prevention performance, and the point prevalence (PP) of MDRPIs.

RESEARCH METHODOLOGY/DESIGN

A pre-post test intervention study without a control group.

SETTING

The study was conducted between May and July 2023 with ICU nurses in three phases: pre-training phase (E0) (104 nurses, 116 patients), training implementation phase (E) and post-training phase (E1) (89 nurses, 120 patients).

MAIN OUTCOME MEASURES

The data were collected by using the Patient (E0, E1) and Nurse (E0) Characteristic Forms, MDRPI Follow-up and Prevalence Form (E0, E1), D.E.V.I.C.E Performance Observation Checklist (E0, E1), MDRPI Knowledge Assessment Questionnaire (E0, E1), Braden Pressure Ulcer Risk Assessment Scale (E0, E1), Pressure Injury Grading Form (E0, E1), and Feedback Form about the Training Process (E).

RESULTS

The mean MDRPI knowledge score of the nurses increased significantly from E0 to E1 (13.23 ± 1.43 vs. 20.02 ± 1.30, p = 0.001), with the highest improvement in the staging and prevention themes. Nurses' MDRPI prevention performance increased significantly from E0 to E1 (2.15 ± 1.01 vs. 11.17 ± 1.65, p = 0.001). There was a significant difference between the PP rate at E0 (61.2 %) and E1 (27.5 %) (p = 0.001).

CONCLUSION

The study indicated that the training on MDRPIs given to ICU nurses increased their knowledge and prevention performance and decreased the prevalence of MDRPIs. However, further studies with a larger sample size are needed to confirm these findings.

IMPLICATIONS FOR CLINICAL PRACTICE

Since MDRPIs have more complex staging and prevention practices than conventional PIs, they require the adoption of a training approach that includes visual materials and practical methods in addition to theoretical knowledge. Accurate definitions of medical device dimensions and fixation, skin assessment, and prevention practices will lead to the desired outcome of reducing MDRPIs in ICUs.

Pressure injuries in mechanically ventilated COVID-19 patients utilising different prone positioning techniques - A prospective observational study

OBJECTIVES

To compare the incidence and distribution of pressure injuries (PIs) with two approaches to prone positioning for mechanically ventilated COVID-19 patients, and to determine the prevalence of these PIs on intensive care unit (ICU) and hospital discharge.

DESIGN

A prospective observational study.

SETTING

Adult patients admitted to a quaternary ICU with COVID-19-associated acute lung injury, between September 2021 and February 2022.

MAIN OUTCOME MEASURES

Incidence and anatomical distribution of PIs during ICU stay for "Face Down" and "Swimmers Position" as well as on ICU and hospital discharge.

RESULTS

We investigated 206 prone episodes in 63 patients. In the Face Down group, 26 of 34 patients (76 %) developed at least one PI, compared to 10 of 22 patients (45 %) in the Swimmers Position group (p = 0.02). Compared to the Swimmers Position group, the Face Down group developed more pressure injuries per patient (median 1 [1, 3] vs 0 [0, 2], p = 0.04) and had more facial PIs (p = 0.002). In a multivariate logistic regression model, patients were more likely to have at least one PI with Face Down position (OR 4.67, 95 % CI 1.28, 17.04, p = 0.02) and greater number of prone episodes (OR 1.75, 95 % CI 1.12, 2.74, p = 0.01). Over 80 % of all PIs were either stage 1 or stage 2. By ICU discharge, 29 % had healed and by hospital discharge, 73 % of all PIs had healed.

CONCLUSION

Swimmers Position had a significantly lower incidence of PIs compared to the Face Down approach. One-quarter of PIs had healed by time of ICU discharge and three-quarters by time of hospital discharge.

IMPLICATIONS FOR CLINICAL PRACTICE

There are differences in incidence of PIs related to prone positioning approaches. This study validates and helps better inform current prone position guidelines recommending the use of Swimmers Position. The low prevalence of PIs at hospital discharge is reassuring.

Incidence, characteristics, and associated factors of pressure injuries acquired in intensive care units over a 12-month period: A secondary analysis of a quality improvement project

OBJECTIVES

To determine the 12-month cumulative incidence, characteristics, and associated factors of pressure injuries acquired in Intensive Care Units.

SETTING

Four intensive care units in a Norwegian University Hospital.

RESEARCH METHODOLOGY

A prospective observational cohort study using data from daily skin inspections during a quality improvement project. We used descriptive statistics and logistic regression. Variables associated with the development of intensive care unit-acquired pressure injuries are presented with odds ratios (OR), and 95% confidence intervals.

RESULTS

The 12-month cumulative incidence of patients (N = 594) developing intensive care unit-acquired pressure injuries was 29 % (172/594) for all categories and 16 % (95/594) when excluding category I pressure injuries (no skin loss). Cumulative incidence for patients acquiring medical device-related pressure injuries was 15 % (91/594) and 11 % (64/594) for category II or worse. Compression stockings (n = 51) and nasogastric tubes (n = 22) were the most frequent documented medical devices related to pressure injuries. Development of pressure injuries category II or worse was significantly associated with vasoactive drug infusions (OR 11.84, 95 % CI [1.59; 88.13]) and longer intensive care unit length of stay (OR 1.06, 95 % CI [1.04; 1.08]).

CONCLUSION

The 12-month cumulative incidence of intensive care unit-acquired pressure injuries was relatively high when category I pressure injuries were included, but comparable to other studies when category I was excluded. Some medical device-related pressure injuries were surprisingly frequent, and these may be prevented. However, associated factors of developing pressure injuries were present and deemed non-modifiable.

IMPLICATIONS FOR CLINICAL PRACTICE

Awareness about pressure injury prevention is needed in the intensive care unit considering high incidences. Nurses can detect category I pressure injuries early, which may be reversed. Our findings show several factors that clinicians can control to reduce the risk of pressure injuries in the intensive care unit.

Exploring body morphology, sacral skin microclimate and pressure injury development and risk among patients admitted to an intensive care unit: A prospective, observational study

OBJECTIVE

To determine the association between body morphology, sacral skin microclimate and their impact on the development and risk of pressure injuries among patients in an intensive care unit.

METHODOLOGY

A prospective observational exploratory study was conducted over 30 weeks. Repeat study observations occurred multiple times a week for 28 days or until discharge. Participant inclusion criteria were ≥ 18 years of age, expected intensive care length of stay > 24 h and intact skin over the sacrum region.

SETTING

The study was conducted in a 36-bed intensive care unit of a major metropolitan public hospital in Queensland, Australia.

OUTCOME MEASURES

Pressure injuries were staged and independently verified according to the international pressure injury classification system. Pressure injury risk was determined by the Braden scale score and subepidermal oedema, using a subepidermal moisture scanner at the sacrum.

RESULTS

Of the 93 participants recruited, an inverted triangle body shape (p =.049), a BMI > 25 kg/m2 (p =.008), a standard foam mattress type (p =.017) and increased length of stay (p <.001) were associated with an increased pressure injury risk according to subepidermal oedema. Participants with increased sacral skin temperature (p <.001), mechanical ventilation (p <.001), vasoactive drugs administered (p =.003), increased sequential organ failure assessment score (p =.047), neurovascular diagnosis (p =.031) and increased length of stay (p =.027) were associated with increased pressure injury risk according to the Braden scale score.

CONCLUSION

Body morphology and skin microclimate are associated with pressure injury risk during critical illness.

IMPLICATIONS FOR CLINICAL PRACTICE

Subepidermal oedema was associated with a patient's shape, body mass index and mattress type, factors that directly influence the pressure loading and the skin, whereas the Braden scale was associated with sacral temperature and clinical measures of critical illness. Consideration of body morphology and skin microclimate in pressure injury risk assessment could lead to more specific prevention strategies targeting high risk patients.

Evaluating the Effectiveness of an Alternating Pressure Overlay in Patients Undergoing Cardiothoracic Surgery

LOCAL PROBLEM

In 2019 and the first half of 2020, our facility experienced an increase in the number and severity of hospital-acquired pressure injuries (HAPIs) among our cardiothoracic surgery population. Fifty percent of these HAPIs occurred within 72 hours of surgery. A review of the literature revealed that alternating pressure overlays (APOs) have been successfully used to prevent HAPIs in surgical patients.

PURPOSE

The primary purpose of our quality improvement (QI) project was to measure perioperative HAPI rates in cardiothoracic surgery patients after the addition of APOs to our HAPI prevention protocol. Our secondary purpose was to identify common factors among those patients who developed HAPIs.

METHODS

This QI project collected both pre- and postintervention data and compared the findings. A nurse-led team was responsible for measuring HAPI rates during the intervention-from July through October 2020-which involved placing an APO under cardiothoracic surgery patients during the 72-hour perioperative period. APOs were placed on all operating room (OR) tables and remained with the patients following surgery. Bed linens and skin care products were standardized for consistency. Lifts were used to reduce friction during repositioning.

RESULTS

During preintervention data collection, we identified 10 patients who developed HAPIs (seven out of 1,174 cardiothoracic surgery patients in 2019, for a HAPI rate of 0.6%, and three out of 333 patients in the first half of 2020, for a HAPI rate of 0.9%). During the four-month intervention period, in which APOs were used in 331 patients undergoing cardiothoracic surgery, no HAPIs developed.

CONCLUSION

Use of an APO in cardiothoracic ORs and critical care units may help reduce HAPI rates.

Exploration of pressure injury risk in adult inpatients: An integrated Braden scale and rough set approach

OBJECTIVE

This study aimed to develop an interpretive model with decision rules to assess the risk level of pressure injuries in adult inpatients and identify the critical risk factors associated with these injuries.

METHODS

The rough set approach was used to identify the critical risk factors associated with pressure injuries and demonstrate their behavioral patterns. The study focused on adult inpatients aged 18 or above who remained in bed for at least 24 hours after admission. The data was extracted from a nursing electronic medical record system of a hospital in Zhejiang Province, China, from 27 October 2019 to 1 November 2020.

RESULTS

The critical risk factors associated with pressure injuries in adult inpatients were identified as "Sensory perception," "Nutrition," and "Friction and shear." A prediction model with 89 decision rules was established and demonstrated reliable predictive capabilities. Nursing staff should focus more on high-risk and severe-risk rules (Rules 11 to 18) to reduce the likelihood of potential high-risk pressure injuries.

CONCLUSIONS

The prediction model established by the rough set approach can be used to identify the critical risk factors of pressure injuries and has good explanatory ability, which can complement and improve the predictive accuracy of the Braden Scale. The decision-making rules can help nurses improve work efficiency.

IMPLICATIONS FOR CLINICAL PRACTICE

Explanatory analysis can explain most inpatients' potential risk patterns and corresponding critical risk factors. Data-driven research models and results can help nurses understand patients' potential risks better. Additionally, these insights can be valuable in nursing education, aiding new nurses in comprehending and addressing the potential risks patients face.

Machine learning-derived blood culture classification with both predictive and prognostic values in the intensive care unit: A retrospective cohort study

OBJECTIVES

Diagnosis and management of intensive care unit (ICU)-acquired bloodstream infections are often based on positive blood culture results. This retrospective cohort study aimed to develop a classification model using data-driven characterisation to optimise the management of intensive care patients with blood cultures.

SETTING, METHODOLOGY/DESIGN

An unsupervised clustering model was developed based on the clinical characteristics of patients with blood cultures in the Medical Information Mart for Intensive Care (MIMIC)-IV database (n = 2451). It was tested using the data from the MIMIC-III database (n = 2047).

MAIN OUTCOME MEASURES

The prognosis, blood culture outcomes, antimicrobial interventions, and trajectories of infection indicators were compared between clusters.

RESULTS

Four clusters were identified using machine learning-based k-means clustering based on data obtained 48 h before the first blood culture sampling. Cluster γ was associated with the highest 28-day mortality rate, followed by clusters α, δ, and β. Cluster γ had a higher fungal isolation rate than cluster β (P < 0.05). Cluster δ was associated with a higher isolation rate of Gram-negative organisms and fungi (P < 0.05). Patients in clusters γ and δ underwent more femoral site vein catheter placements than those in cluster β (P < 0.001, all). Patients with a duration of antibiotics treatment of 4, 6, and 7 days in clusters α, δ, and γ, respectively, had the lowest 28-day mortality rate.

CONCLUSIONS

Machine learning identified four clusters of intensive care patients with blood cultures, which yielded different prognoses, blood culture outcomes, and optimal duration of antibiotic treatment. Such data-driven blood culture classifications suggest further investigation should be undertaken to optimise treatment and improve care.

IMPLICATIONS FOR CLINICAL PRACTICE

Intensive care unit-acquired bloodstream infections are heterogeneous. Meaningful classifications associated with outcomes should be described. Using machine learning and cluster analysis could help in understanding heterogeneity. Data-driven blood culture classification could identify distinct physiological states and prognoses before deciding on blood culture sampling, optimise treatment, and improve care.

Dependency-related skin lesions in the prone critical patient. Incidence study

OBJECTIVE

To determine the incidence of dependence-related skin lesions (DRSL) in patients in prone position (PP) and to identify the predisposing factors.

METHOD

Follow-up study in two polyvalent intensive care units. Patients undergoing invasive mechanical ventilation and PP with no skin lesions on admission were included. We recorded the 3 types of DRSL: (pressure ulcers [PU], moisture-associated skin damage [MASD] and friction injuries [FI]), demographic variables, diagnosis, length of stay, PP episodes, postural changes, APACHE II (Acute Physiology and Chronic Health Disease Classification System), prealbumin level on admission, body mass index (BMI), diabetes, hypertension, smoking, peripheral vascular disease (PVD), vasoactive drugs, Braden scale and mortality. Bivariate analysis: chi-square test, t-test or Mann-Whitney U test. Multivariate analysis: logistic regression.

RESULTS

Forty nine patients were included and 170PP were performed.Forty-one DRSL appeared in 22 patients with a cumulative incidence of 44.9% (95%CI: 31.6-58.7). PU accounted for 63.4% (73.1% facial; 76.9% stage II), 12.2% were MASD (60% inguinal; 60% stage II) and 24.4% were FI (50% thoracic; 70% stage III). The median age of the lesion group (LG) was 66.5 [61.8-71.3] vs. 64 [43-71] years old in the non-lesion group (NLG), p=0.04. Eighty percent of the LG had PVD vs. 20% of the NLG, p=0.03. The median total hours on PP of the LG was 96.9 [56.1-149.4] vs. 38.2 [18.8-57] of the NIG, p<0.001. Multivariate analysis selected total PP hours (OR=1.03; 95%CI:1.01-1.05) and PVD (OR=8.9; 95%CI:1.3-58.9) as predisposing factors for developing DRSL.

CONCLUSIONS

There is a high incidence of skin lesions related to prone decubitus dependence, mostly pressure lesions, although of low severity.The accumulated hours in probe position and peripheral vascular disease favor their development.

Systematic Review for Risks of Pressure Injury and Prediction Models Using Machine Learning Algorithms

Pressure injuries are increasing worldwide, and there has been no significant improvement in preventing them. This study is aimed at reviewing and evaluating the studies related to the prediction model to identify the risks of pressure injuries in adult hospitalized patients using machine learning algorithms. In addition, it provides evidence that the prediction models identified the risks of pressure injuries earlier. The systematic review has been utilized to review the articles that discussed constructing a prediction model of pressure injuries using machine learning in hospitalized adult patients. The search was conducted in the databases Cumulative Index to Nursing and Allied Health Literature (CINAHIL), PubMed, Science Direct, the Institute of Electrical and Electronics Engineers (IEEE), Cochrane, and Google Scholar. The inclusion criteria included studies constructing a prediction model for adult hospitalized patients. Twenty-seven articles were included in the study. The defects in the current method of identifying risks of pressure injury led health scientists and nursing leaders to look for a new methodology that helps identify all risk factors and predict pressure injury earlier, before the skin changes or harms the patients. The paper critically analyzes the current prediction models and guides future directions and motivations.

Prevalence of skin pressure injury in critical care patients in the UK: results of a single-day point prevalence evaluation in adult critically ill patients

OBJECTIVES

Hospital-acquired pressure injuries (PIs) are a source of morbidity and mortality, and many are potentially preventable.

DESIGN

This study prospectively evaluated the prevalence and the associated factors of PIs in adult critical care patients admitted to intensive care units (ICU) in the UK.

SETTING

This service evaluation was part of a larger, international, single-day point prevalence study of PIs in adult ICU patients. Training was provided to healthcare givers using an electronic platform to ensure standardised recognition and staging of PIs across all sites.

PARTICIPANTS

The characteristics of the ICUs were recorded before the survey; deidentified patient data were collected using a case report form and uploaded onto a secure online platform.

PRIMARY AND SECONDARY OUTCOME MEASURES

Factors associated with ICU-acquired PIs in the UK were analysed descriptively and using mixed multiple logistic regression analysis.

RESULTS

Data from 1312 adult patients admitted to 94 UK ICUs were collected. The proportion of individuals with at least one PI was 16% (211 out of 1312 patients), of whom 8.8% (n=115/1312) acquired one or more PIs in the ICU and 7.3% (n=96/1312) prior to ICU admission. The total number of PIs was 311, of which 148 (47.6%) were acquired in the ICU. The location of majority of these PIs was the sacral area, followed by the heels. Braden score and prior length of ICU stay were associated with PI development.

CONCLUSIONS

The prevalence and the stage of severity of PIs were generally low in adult critically ill patients admitted to participating UK ICUs during the study period. However, PIs are a problem in an important minority of patients. Lower Braden score and longer length of ICU stay were associated with the development of injuries; most ICUs assess risk using tools which do not account for this.

TRIAL REGISTRATION NUMBER

NCT03270345.

Neuromuscular blocking agents and opioids are major risk factors for developing pressure injuries in patients in the intensive care unit

BACKGROUND/AIMS

Patients in the intensive care unit (ICU) are at high risk for developing pressure injuries, which can cause severe complications and even increase the mortality risk. Therefore, prevention of pressure injuries is most important. In this study, we investigated the risk factors of pressure injury development in patients admitted to the ICU.

METHODS

We retrospectively analyzed patients ages &gt; 18 years admitted to the medical ICU in a tertiary hospital between January and December 2019. We collected patient baseline characteristics, medications received, mechanical ventilation or hemodialysis use, laboratory findings, and date of pressure injury onset and characteristics.

RESULTS

We analyzed 666 patients who did not have pressure injuries at ICU admission. Pressure injuries developed in 102 patients (15%). The risk of pressure injury development increased as the administration days for neuromuscular blocking agents (NMBAs; odds ratio [OR], 1.138; p = 0.019) and opioids (OR, 1.084; p = 0.028) increased, and if the patient had problem with friction and shear (OR, 2.203; p = 0.011).

CONCLUSION

The prolonged use of NMBAs and opioids can increase the risk of pressure injury development. Because these medications are associated with immobilization, using both should be minimized and patient early mobilization should be promoted. Among the Braden subscales, "friction and shear" was associated with the development of pressure injuries in ICU patients.

Clinical Coders' Perspectives on Pressure Injury Coding in Acute Care Services in Victoria, Australia

Pressure injuries (PIs) substantively impact quality of care during hospital stays, although only when they are severe or acquired as a result of the hospital stay are they reported as quality indicators. Globally, researchers have repeatedly highlighted the need to invest more in quality improvement, risk assessment, prevention, early detection, and care for PI to avoid the higher costs associated with treatment of PI. Coders' perspectives on quality assurance of the clinical coded PI data have never been investigated. This study aimed to explore challenges that hospital coders face in accurately coding and reporting PI data and subsequently, explore reasons why data sources may vary in their reporting of PI data. This article is based upon data collected as part of a multi-phase collaborative project to build capacity for optimizing PI prevention across Monash Partners health services. We have conducted 16 semi-structured phone interviews with clinical coders recruited from four participating health services located in Melbourne, Australia. One of the main findings was that hospital coders often lacked vital information in clinicians' records needed to code PI and report quality indicators accurately and highlighted the need for quality improvement processes for PI clinical documentation. Nursing documentation improvement is a vital component of the complex capacity building programs on PI prevention in acute care services and is relied on by coders. Coders reported the benefit of inter-professional collaborative workshops, where nurses and coders shared their perspectives. Collaborative workshops had the potential to improve coders' knowledge of PI classification and clinicians' understanding of what information should be included when documenting PI in the medical notes. Our findings identified three methods of quality assurance were important to coders to ensure accuracy of PI reporting: (1) training prior to initiation of coding activity and (2) continued education, and (3) audit and feedback communication about how to handle specific complex cases and complex documentation. From a behavioral perspective, most of the coders reported confidence in their own abilities and were open to changes in coding standards. Transitioning from paper-based to electronic records highlighted the need to improve training of both clinicians and coders.

Exploring medical device-related pressure injuries in a single intensive care setting: A longitudinal point prevalence study

OBJECTIVES

To report longitudinal prevalence rates of device-related pressure injuries in critically ill adult patients in the intensive care unit and to explore the patient characteristics associated with the development of device related pressure injuries.

RESEARCH DESIGN

A prospective observational design where observations of patients' skin integrity were conducted on one day each week for 52 weeks.

SETTING

The study was conducted in the 36-bed ICU of a major metropolitan tertiary referral hospital in Queensland, Australia. The sample included all patients aged 18 years or older admitted to the intensive care unit before midnight on the day preceding the observation, with a medical device in situ.

MAIN OUTCOME MEASURES

The primary outcome measure was device related pressure injuries identified at the weekly observations and defined as a pressure injury found on the skin or mucous membrane with a history of medical device in use at the location of the injury. Patient demographic and clinical characteristics were recorded.

RESULTS

Over the study period, 11.3% (71/631) of patients developed at least one hospital-acquired DRPI. The most common devices associated with injury were nasogastric/nasojejunal tubes (41%) and endotracheal tubes (27%). Significant predictors of device related pressure injuries were the total number of devices (OR 1.230, 95% CI 1.09-1.38, p < 0.001), the length of time in the ICU (OR 1.05, 95% CI 1.02-1.09, p = 0.003), male sex, (OR 2.099, 95% CI 1.18-3.7, p = 0.012), and increased severity of illness score on admission (OR 1.044, 95% CI 1.01-1.09, p = 0.013).

CONCLUSION

Device related pressure injuries are an all-too-common iatrogenic problem for this vulnerable patient cohort.