Dataset Information

Prediction of Adverse Outcomes in Pediatric Acute Hematogenous Osteomyelitis.

ABSTRACT:

Background

Clinicians cannot reliably predict complications of acute hematogenous osteomyelitis (AHO).

Methods

Consecutive cases of AHO from 2 pediatric centers in the United States were analyzed retrospectively to develop clinical tools from data obtained within 96 hours of hospitalization to predict acute and chronic complications of AHO. Two novel composite prediction scores derived from multivariable logistic regression modeling were compared with a previously published severity of illness (SOI) score, C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR) using area under the receiver operating characteristic curve analyses.

Results

The causative organisms were identified in 73% of 261 cases. Bacteremia (45%), abscesses (38%), and associated suppurative arthritis (23%) were relatively common. Acute or chronic complications occurred in 24% and 11% of patients, respectively. Multivariable logistic regression identified bone abscess (odds ratio [OR], 2.3 [95% confidence interval {CI}, 1.0-5.2]), fever > 48 hours (OR, 2.7 [95% CI, 1.2-6.0]), suppurative arthritis (OR, 3.2 [95% CI, 1.3-7.5]), disseminated disease (OR, 4.6 [95% CI, 1.5-14.3]), and delayed source control (OR, 5.1 [95% CI, 1.4-19.0]) as strong predictors of acute complications. In a separate model, CRP ? 100 mg/L at 2-4 days after antibiotics (OR, 2.7 [95% CI, 1.0-7.3]), disseminated disease (OR, 3.3 [95% CI, 1.1-10.0]), and requirement for bone debridement (OR, 6.7 [95% CI, 2.1-21.0]) strongly predicted chronic morbidity. These variables were combined to create weighted composite prediction scores for acute (A-SCORE) and chronic (C-SCORE) osteomyelitis, which were superior to SOI, CRP, and ESR and had negative predictive values > 90%.

Conclusions

Two novel composite clinical scores were superior to existing tools to predict complications of pediatric AHO.

SUBMITTER: Alhinai Z

PROVIDER: S-EPMC7904074 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of Adverse Outcomes in Pediatric Acute Hematogenous Osteomyelitis.

Alhinai Zaid Z Elahi Morvarid M Park Sangshin S Foo Bill B Lee Brian B Chapin Kimberle K Koster Michael M Sánchez Pablo J PJ Michelow Ian C IC

Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 20201201 9

<h4>Background</h4>Clinicians cannot reliably predict complications of acute hematogenous osteomyelitis (AHO).<h4>Methods</h4>Consecutive cases of AHO from 2 pediatric centers in the United States were analyzed retrospectively to develop clinical tools from data obtained within 96 hours of hospitalization to predict acute and chronic complications of AHO. Two novel composite prediction scores derived from multivariable logistic regression modeling were compared with a previously published severi ...[more]

PMID: 32129457

Similar Datasets

Project description:Background:Children with acute hematogenous osteomyelitis (AHO) have a broad spectrum of illness ranging from mild to severe. The purpose of this study is to evaluate the impact of genomic variation of Staphylococcus aureus on clinical phenotype of affected children and determine which virulence genes correlate with severity of illness. Methods:De novo whole genome sequencing was conducted for a strain of Community Acquired Methicillin Resistant Staphylococcus aureus (CA-MRSA), using PacBio Hierarchical Genome Assembly Process (HGAP) from 6 Single Molecule Real Time (SMRT) Cells, as a reference for DNA library assembly of 71 Staphylococcus aureus isolates from children with AHO. Virulence gene annotation was based on exhaustive literature review and genomic data in NCBI for Staphylococcus aureus. Clinical phenotype was assessed using a validated severity score. Kruskal-Wallis rank sum test determined association between clinical severity and virulence gene presence using False Discovery Rate (FDR), significance <0.01. Results:PacBio produced an assembled genome of 2,898,306 bp and 2054 Open Reading Frames (ORFs). Annotation confirmed 201 virulence genes. Statistical analysis of gene presence by clinical severity found 40 genes significantly associated with severity of illness (FDR ?0.009). MRSA isolates encoded a significantly greater number of virulence genes than did MSSA (p < 0.0001). Phylogenetic analysis by maximum likelihood (PAML) demonstrated the relatedness of genomic distance to clinical phenotype. Conclusions:The Staphylococcus aureus genome contains virulence genes which are significantly associated with severity of illness in children with osteomyelitis. This study introduces a novel reference strain and detailed annotation of Staphylococcus aureus virulence genes. While this study does not address bacterial gene expression, a platform is created for future transcriptome investigations to elucidate the complex mechanisms involved in childhood osteomyelitis.

Project description:BackgroundThis study was conducted to evaluate how the preeclampsia integrated estimate of risk (fullPIERS) model performs in the prediction of adverse maternal outcomes when the predictor variables are all obtained within 24-h of admission for preeclampsia.MethodsA prospective cohort study on 323 women who fulfilled definite inclusion and exclusion criteria was conducted. Subjects were monitored for clinical symptoms of preeclampsia, biochemical parameters, and adverse maternal and neonatal outcomes. A risk prediction score was calculated using the fullPIERS calculator. Statistical analysis of rates and ratios was carried out by assessing ? (2) test and odds ratio.Results18.3 % (n = 60) had adverse maternal outcome and 42.8 % (n = 138) had adverse fetal outcome, and 43 (13.35 %) had combined adverse maternal and perinatal outcome. Dyspnea, visual disturbances, epigastric pain, and [Formula: see text] appeared to be highly significant risk factors. In the biochemical variables studied, serum creatinine and serum uric acid were found to have a significant association. The association between adverse perinatal outcome and vaginal delivery was highly significant (OR 0.35, 95 % CI 0.19, 0.63), and the P value was 0.0005. The likelihood ratio associated with the highest risk group (predicted probability of the outcome ?30 %) showed excellent performance (i.e., 17.5) of fullPIERS model as a rule in test.ConclusionThe fullPIERS model performed well in the prediction of adverse maternal outcomes in women with preeclampsia. It is easy to use. The model is based on the use of few important clinical and biochemical parameters and does not require extensive laboratory testing. Although it might be of limited use in a well-equipped tertiary care facility, this model can be utilized in the setting of district or sub-district level hospitals to identify patients who are at risk of complications due to preeclampsia. Timely referral to a higher center will help in reducing the morbidity and mortality associated with this condition.

Project description:The COVID-19 pandemic has devastated the world with health and economic wreckage. Precise estimates of adverse outcomes from COVID-19 could have led to better allocation of healthcare resources and more efficient targeted preventive measures, including insight into prioritizing how to best distribute a vaccination. We developed MLHO (pronounced as melo), an end-to-end Machine Learning framework that leverages iterative feature and algorithm selection to predict Health Outcomes. MLHO implements iterative sequential representation mining, and feature and model selection, for predicting patient-level risk of hospitalization, ICU admission, need for mechanical ventilation, and death. It bases this prediction on data from patients' past medical records (before their COVID-19 infection). MLHO's architecture enables a parallel and outcome-oriented model calibration, in which different statistical learning algorithms and vectors of features are simultaneously tested to improve prediction of health outcomes. Using clinical and demographic data from a large cohort of over 13,000 COVID-19-positive patients, we modeled the four adverse outcomes utilizing about 600 features representing patients' pre-COVID health records and demographics. The mean AUC ROC for mortality prediction was 0.91, while the prediction performance ranged between 0.80 and 0.81 for the ICU, hospitalization, and ventilation. We broadly describe the clusters of features that were utilized in modeling and their relative influence for predicting each outcome. Our results demonstrated that while demographic variables (namely age) are important predictors of adverse outcomes after a COVID-19 infection, the incorporation of the past clinical records are vital for a reliable prediction model. As the COVID-19 pandemic unfolds around the world, adaptable and interpretable machine learning frameworks (like MLHO) are crucial to improve our readiness for confronting the potential future waves of COVID-19, as well as other novel infectious diseases that may emerge.

Dataset Information

Prediction of Adverse Outcomes in Pediatric Acute Hematogenous Osteomyelitis.

Background

Methods

Results

Conclusions

Publications

Prediction of Adverse Outcomes in Pediatric Acute Hematogenous Osteomyelitis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure