[Skip to Navigation]
Views 1,291
Citations 0
Original Investigation
July 15, 2021

Clinical Prediction Models for Suspected Pediatric Foreign Body Aspiration: A Systematic Review and Meta-analysis

Author Affiliations
  • 1Department of Otolaryngology–Head and Neck Surgery, University of Toronto, Toronto, Ontario, Canada
  • 2Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada
  • 3Department of Otolaryngology–Head and Neck Surgery, The Hospital for Sick Kids, Toronto, Ontario, Canada
JAMA Otolaryngol Head Neck Surg. Published online July 15, 2021. doi:10.1001/jamaoto.2021.1548
Key Points

Question  What is the evidence for the use of clinical predictions models in diagnosing pediatric foreign body aspiration?

Findings  This systematic review and meta-analysis identified models using predictors based on clinical history, physical examination, and radiographic findings. A meta-analysis of model performance suggests that there is a potential benefit of such decision-making tools, but is overall limited because of concerns of model overfitting, inconsistent reporting of performance, and lack of model validation.

Meaning  Current prediction models for the diagnosing pediatric foreign body aspiration are at high risk of bias and are not recommended to guide clinical decision-making; thus, adherence to recognized guidelines for conducting and reporting modeling studies is recommended.

Abstract

Importance  Although various clinical prediction models (CPMs) have been described for diagnosing pediatric foreign body aspiration (FBA), to our knowledge, there is still no consensus regarding indications for bronchoscopy, the criterion standard for identifying airway foreign bodies.

Objective  To evaluate currently available CPMs for diagnosing FBA in children.

Data Sources  Performed in Ovid MEDLINE, Ovid Embase, PubMed, Web of Science, and CINAHL database with citation searching of retrieved studies.

Study Selection  Prediction model derivation and validation studies for diagnosing FBA in children were included. Exclusion criteria included adult studies; studies that included variables that were not available in routine clinical practice and outcomes for FBA were not separate or extractable.

Data Extraction and Synthesis  We followed the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies and the Prediction Model Risk of Bias Assessment Tool framework. Data were pooled using a random-effects model.

Main Outcomes and Measures  The primary outcome was the diagnosis of FBA as confirmed by bronchoscopy. Characteristics of CPMs and individual predictors were evaluated. The final model presentation with available measures of performance was provided by narrative synthesis. A meta-analysis of individual predictor variables and prediction models was performed.

Results  After screening 4233 articles, 7 studies (0.2%; 1577 patients) were included in the final analysis. There were 6 model derivation studies and 1 validation study. Air trapping (odds ratio [OR], 8.3; 95% CI, 4.4-15.5), unilateral reduced air entry (OR, 4.8; 95% CI, 3.5-6.5), witnessed choking (OR, 3.1; 95% CI, 1.0-9.6), wheezing (OR, 2.5; 95% CI, 1.2-5.2), and suspicious findings suggestive of FBA on radiography (OR, 18.5; 95% CI, 5.0-67.7) were the most commonly used predictor variables. Model performance varied, with discrimination scores (C statistic) ranging from 0.74 to 0.88. The pooled weighted C statistic score of all models was 0.86 (95% CI, 0.80-0.92). All studies were deemed to be at high risk of bias, with overfitting of models and lack of validation as the most pertinent concerns.

Conclusions and Relevance  This systematic review and meta-analysis suggests that existing CPMs for FBA in children are at a high risk of bias and have not been adequately validated. No current models can be recommended to guide clinical decision-making. Future CPM studies that adhere to recognized standards for development and validation are required.

Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    ×