Objective:

To review the integration of multiomics profiling and artificial intelligence in the detection and subtyping of endocrine hypertension (EH), highlighting its potential impact on clinical outcomes.

Key Findings:

• EH accounts for 5% to 10% of hypertension cases, with primary aldosteronism (PA) potentially being 3- to 5-fold more prevalent than previously cited.
• AI models show diagnostic accuracy comparable to invasive standards for conditions like PA, indicating a shift in diagnostic approaches.
• Multiomics analyses can identify microRNA signatures that distinguish between unilateral and bilateral PA, enhancing diagnostic precision.
• Machine learning models integrating various data types have demonstrated high diagnostic performance for PA, suggesting a promising future for AI in clinical settings.

Interpretation:

The integration of AI and multiomics could lead to earlier detection and more personalized treatment strategies for EH, potentially reducing cardiovascular risk and improving patient outcomes.

Limitations:

• The narrative review format does not formally assess the quality of included studies, which may introduce selection bias.
• Challenges in real-world implementation include data privacy, poor data-sharing, limited clinician digital skills, and potential biases in AI models.

Conclusion:

The management of endocrine hypertension is likely to evolve towards dynamic, data-driven systems, enhancing diagnostic accuracy and treatment personalization, while addressing the identified limitations will be crucial for successful implementation.