PICADAR: a diagnostic predictive tool for primary ciliary dyskinesia.

Primary ciliary dyskinesia (PCD) remains a challenging condition to diagnose due to its nonspecific symptoms and limited referral guidelines. While diagnostic methods exist, they require specialized equipment and trained personnel, limiting accessibility. Prior research has shown that early symptoms may not reliably predict PCD, creating diagnostic uncertainty. This gap motivated the need for a practical clinical tool to guide referrals. Existing knowledge includes the association of PCD with chronic respiratory symptoms and situs inversus. However, no prior work had resolved how to translate these associations into a predictive score. The absence of a validated referral tool has hindered timely diagnosis. This paper's contribution is the development of a diagnostic prediction model based on clinical history data.

The aim of this study was to create a clinical prediction tool to identify patients likely to have PCD based on easily obtainable clinical features. The researchers sought to address the diagnostic uncertainty caused by PCD's nonspecific symptoms. They focused on developing a tool that could be used in respiratory centers before referring patients for specialized testing. The motivation came from the high cost and limited availability of diagnostic tests for PCD. The study aimed to use patient history to guide referrals more effectively. It also aimed to validate the tool in separate diagnostic centers to ensure generalizability. The researchers wanted to ensure the tool was practical and accessible to clinicians. The ultimate goal was to improve diagnostic accuracy and referral efficiency.

The researchers analyzed data from 641 consecutive referrals with confirmed diagnostic outcomes. They collected clinical history data from each patient and correlated it with diagnostic results. Logistic regression was used to identify the strongest predictors of PCD. The model was tested using receiver operating characteristic curve analysis to assess performance. The best-performing model was simplified into a clinical tool named PICADAR. Seven clinical features were selected as predictive parameters: gestation, neonatal chest symptoms, neonatal intensive care admission, chronic rhinitis, ear symptoms, situs inversus, and congenital cardiac defect. The tool was validated internally and then in an external center to assess reliability. The final model was designed to be user-friendly for clinical use.

Of the 641 referrals, 75 (12%) were confirmed as PCD cases. The PICADAR tool demonstrated a sensitivity of 0.90 and specificity of 0.75 at a 5-point cut-off score. The area under the curve for internal validation was 0.91, indicating strong predictive power. External validation showed an area under the curve of 0.87, confirming the tool's reliability. The seven predictive parameters were selected based on their statistical significance in the model. The tool's performance remained consistent across both validation sets. No additional parameters improved the model's accuracy. The results suggest that PICADAR can effectively guide referrals for PCD testing. The tool's simplicity and accuracy make it suitable for use in respiratory centers.

The authors propose that PICADAR is a valid and practical diagnostic prediction rule for PCD. They suggest that the tool's high sensitivity ensures fewer cases are missed in referrals. The specificity of 0.75 indicates that it reduces unnecessary referrals for testing. The authors state that the tool's performance was consistent in both internal and external validations. They propose that PICADAR is ready for implementation in respiratory centers referring patients to PCD diagnostic centers. The tool's reliance on clinical history data makes it accessible to clinicians without specialized training. The authors suggest that PICADAR improves the efficiency of PCD diagnosis by guiding appropriate referrals. They conclude that the tool's accuracy and simplicity make it suitable for widespread clinical use.