Filtering clinically significant hypercalcaemia from non-significant hypercalcaemia at the laboratory level

Area of Science:

• Clinical Chemistry
• Medical Diagnostics
• Biochemical Analysis

Background:

• Hypercalcaemia diagnosis is often delayed in hospital settings, with only 30% identified.
• Clinical significance of hypercalcaemia varies, necessitating differentiation between critical and non-critical cases.

Purpose of the Study:

• To develop and validate a laboratory-based alarm filter to distinguish clinically significant hypercalcaemia.
• To improve the timely identification of critical hypercalcaemia cases.

Main Methods:

• A two-step discriminant function analysis was employed using routine laboratory variables (phosphate, albumin, protein, LD, haemoglobin, ALP) and age.
• The filter classified patients into categories: primary hyperparathyroidism, malignancy, and transient hypercalcaemia.
• Cross-validation with clinician diagnoses assessed the filter's classification efficiency.

Main Results:

• The alarm filter achieved high classification accuracy: 84% for women and 77% for men.
• Sensitivity and specificity were notable: 90% sensitivity and 73% specificity in women; 80% sensitivity and 67% specificity in men.
• The filter successfully differentiated clinically significant from non-significant hypercalcaemia.

Conclusions:

• The developed alarm filter is an effective tool for identifying clinically significant hypercalcaemia at the laboratory level.
• Implementation of this filter can aid in earlier diagnosis and management of critical hypercalcaemia cases.
• The study highlights the utility of routine biochemical markers in developing diagnostic algorithms for complex conditions.