Yunseob Shin, Kyung-Jae Cho, Yeha Lee, Yu Hyeon Choi, Jae Hwa Jung, Soo Yeon Kim, Yeo Hyang Kim, Young A Kim, Joongbum Cho, Seong Jong Park, Won Kyoung Jhang
Acute Crit Care. 2022;37(4):654-666. Published online October 26, 2022
Background Early recognition of deterioration events is crucial to improve clinical outcomes. For this purpose, we developed a deep-learning-based pediatric early-warning system (pDEWS) and aimed to validate its clinical performance. Methods: This is a retrospective multicenter cohort study including five tertiary-care academic children’s hospitals. All pediatric patients younger than 19 years admitted to the general ward from January 2019 to December 2019 were included. Using patient electronic medical records, we evaluated the clinical performance of the pDEWS for identifying deterioration events defined as in-hospital cardiac arrest (IHCA) and unexpected general ward-to-pediatric intensive care unit transfer (UIT) within 24 hours before event occurrence. We also compared pDEWS performance to those of the modified pediatric early-warning score (PEWS) and prediction models using logistic regression (LR) and random forest (RF). Results: The study population consisted of 28,758 patients with 34 cases of IHCA and 291 cases of UIT. pDEWS showed better performance for predicting deterioration events with a larger area under the receiver operating characteristic curve, fewer false alarms, a lower mean alarm count per day, and a smaller number of cases needed to examine than the modified PEWS, LR, or RF models regardless of site, event occurrence time, age group, or sex. Conclusions: The pDEWS outperformed modified PEWS, LR, and RF models for early and accurate prediction of deterioration events regardless of clinical situation. This study demonstrated the potential of pDEWS as an efficient screening tool for efferent operation of rapid response teams.
Citations
Citations to this article as recorded by
Predicting cardiac arrest after neonatal cardiac surgery Alexis L. Benscoter, Mark A. Law, Santiago Borasino, A. K. M. Fazlur Rahman, Jeffrey A. Alten, Mihir R. Atreya Intensive Care Medicine – Paediatric and Neonatal.2024;[Epub] CrossRef
Volumetric regional MRI and neuropsychological predictors of motor task variability in cognitively unimpaired, Mild Cognitive Impairment, and probable Alzheimer's disease older adults Michael Malek-Ahmadi, Kevin Duff, Kewei Chen, Yi Su, Jace B. King, Vincent Koppelmans, Sydney Y. Schaefer Experimental Gerontology.2023; 173: 112087. CrossRef
Predicting sepsis using deep learning across international sites: a retrospective development and validation study Michael Moor, Nicolas Bennett, Drago Plečko, Max Horn, Bastian Rieck, Nicolai Meinshausen, Peter Bühlmann, Karsten Borgwardt eClinicalMedicine.2023; 62: 102124. CrossRef
A model study for the classification of high-risk groups for cardiac arrest in general ward patients using simulation techniques Seok Young Song, Won-Kee Choi, Sanggyu Kwak Medicine.2023; 102(37): e35057. CrossRef
An advanced pediatric early warning system: a reliable sentinel, not annoying extra work Young Joo Han Acute and Critical Care.2022; 37(4): 667. CrossRef
Background There are few studies on the effect of intensivist staffing in pediatric intensive care units (PICUs) in Korea. We aimed to evaluate the effect of pediatric intensivist staffing on treatment outcomes in a Korean hospital PICU.
Methods We analyzed two time periods according to pediatric intensivist staffing: period 1, between November 2015 to January 2017 (no intensivist staffing, n=97) and period 2, between February 2017 to February 2018 (intensivists staffing, n=135).
Results Median age at admission was 5.4 years (range, 0.7–10.3 years) in period 1 and 3.6 years (0.2–5.1 years) in period 2 (P=0.013). The bed occupancy rate decreased in period 2 (75%; 73%–88%) compared to period 1 (89%; 81%–94%; P=0.015). However, the monthly bed turnover rate increased in period 2 (2.2%; 1.9%–2.7%) compared to period 1 (1.5%, 1.1%– 1.7%; P=0.005). In both periods, patients with chronic neurologic illness were the most common. Patients with cardiovascular problems were more prevalent in period 2 than period 1 (P=0.008). Daytime admission occurred more frequently in period 2 than period 1 (63% vs. 39%, P<0.001). The length of PICU stay, parameters related with mechanical ventilation and tracheostomy, and pediatric Sequential Organ Failure Assessment score were not different between periods. Sudden cardiopulmonary resuscitations occurred in two cases during period 1, but no case occurred during period 2.
Conclusions Pediatric intensivist staffing in the PICU may affect efficient ICU operations.
Citations
Citations to this article as recorded by
The impact of pediatric intensivists on the management of pediatric diabetic ketoacidosis in pediatric intensive care units Ah Young Choi, Esther Park BMC Pediatrics.2023;[Epub] CrossRef
Impact of staffing model conversion from a mandatory critical care consultation model to a closed unit model in the medical intensive care unit Sung Jun Ko, Jaeyoung Cho, Sun Mi Choi, Young Sik Park, Chang-Hoon Lee, Chul-Gyu Yoo, Jinwoo Lee, Sang-Min Lee, Robert Jeenchen Chen PLOS ONE.2021; 16(10): e0259092. CrossRef
Intensivists in the pediatric intensive care unit: conductors for better operational efficiency June Dong Park Acute and Critical Care.2020; 35(2): 115. CrossRef
Kawasaki disease (KD) is an acute, systemic vasculitis of childhood. The early mortality of KD results from coronary complications, mainly aneurysmal thrombosis with myocardial infarction, and the subacute phase of KD has the highest risk of mortality. Although there have been reports of ischemic heart disease as late cardiologic sequelae of KD in young adults, acute myocardial infarction caused by coronary complications in the subacute phase of KD is rare. We experienced one pediatric patient who developed coronary artery aneurysm and acute myocardiac infarction (AMI) during the subacute phase of incomplete and intravenous immunoglobulin (IVIG)-nonresponsive KD. The patient was given a good prognosis due to close monitoring and early recognition of AMI. Physicians should carefully monitor KD patients who do not respond to initial IVIG therapy and who show progressive coronary artery dilatation. If such a patient complaints of chest pain and the ECG shows hyperacute T waves, the physician should suspect development of AMI.
A 9-month-old infant presented with cough, tachypnea, and grunting was admitted. The patient was revealed to have cardiomegaly, high NT-proBNP, and severe left ventricular dilation and dysfunction; she was subsequently diagnosed with acute myocarditis and congestive heart failure.
Intravenous immunoglobulin, inotropics, diuretics, angiotensin converting enzyme inhibitors and beta blocker were used. However, left hemiparesis suddenly developed at 30-day after treatment. Brain MRI showed high signal intensity in the right middle cerebral arterial territory on diffusion weighted brain MRI and in the left parietal lobe with gyral enhancement. Echocardiogram revealed no definite intraventricular thrombus. The patient was started on an antiplatelet agent only without anticoagulant therapy for the treatment of cerebral infarct in respect of the risk to the infant. Four years after the cerebral ischemic stroke (CIS), she showed complete recovery from hemiparesis, with no more CIS. In conclusion, severe ventricular dilatation and dysfunction can lead to thromboembolic events in infants. We should keep in mind that anticoagulant or antiplatelet agents can be used in specific situations.
Spontaneous pneumomediastinum (SPM) is a rare condition in children which is triggered by respiratory infection and inflammation, although it occurs most commonly in asthmatics. It is caused by alveolar rupture and dissection of air into the mediastinum and hilum, and the prognosis is usually benign. We report two cases of SPM and subcutaneous emphysema complicating pneumonia in children with severe H1N1 infection. The patients were admitted to the intensive care unit and treated with oxygen, inhalation of a bronchodilator, intravenous systemic corticosteroid (methyprednisolone, 2 mg/kg/day for 5 days) and antibiotics, together with antiviral therapy. On day 4 after admission, there was no further evidence of SPM. SPM associated with severe H1N1 infection in children resolves with aggressive supportive care, without progression to pneumothorax. We should remain aware of this air leak complication in children with severe respiratory infection.