Related Experiment Videos
Hormone pulsatility discrimination via coarse and short time sampling.
S M Pincus1, M L Hartman, F Roelfsema
1stevepincus@alum.mit.edu
The American Journal of Physiology
|November 24, 1999
Summary
Analysis of pulsatile hormone secretion using approximate entropy (ApEn) can be simplified. Shorter, less frequent sampling still accurately detects hormonal irregularities in disease states, making it a potential clinical tool.
Area of Science:
- Endocrinology
- Biostatistics
- Biomedical Engineering
Background:
- Pulsatile hormonal secretion is common in endocrinology but requires extensive sampling for analysis.
- Current pulsatility analysis methods are research-based due to high costs associated with frequent blood sampling and assays.
- Approximate entropy (ApEn) quantifies sequential irregularity and has shown promise in distinguishing pathological from healthy hormonal patterns.
Purpose of the Study:
- To determine if less frequent and shorter duration sampling can still effectively identify significant differences in hormonal pulsatility between healthy and diseased individuals using ApEn.
- To assess the potential for simplifying pulsatility analysis protocols for clinical application.
Main Methods:
- Reanalyzed data from two existing studies on 24-hour hormone profiles.
- Compared growth hormone secretion in healthy subjects versus acromegalic patients (originally sampled every 5 min).
- Compared ACTH and cortisol secretion in healthy subjects versus Cushing's disease patients (originally sampled every 10 min).
- Applied statistical analyses to ApEn values derived from data sampled at coarser intervals (e.g., 60 min) and shorter durations (2-4 h).
Main Results:
- Hormonal secretion patterns in patients with tumors (acromegaly, Cushing's disease) were consistently and significantly more irregular than in healthy controls (P < 0.0001).
- These significant differences were detectable even with very coarse sampling regimens (e.g., 60-min intervals) and short time frames (2-4 hours).
- The ApEn analysis demonstrated high sensitivity and specificity in distinguishing between patient groups and controls.
Conclusions:
- Less frequent and shorter duration sampling methods are sufficient for robustly detecting significant differences in hormonal pulsatility using ApEn.
- These findings suggest that ApEn analysis with reduced sampling requirements can be broadly utilized.
- Simplifying sampling protocols has the potential to transition pulsatility analysis from a research-only method to a clinically applicable diagnostic and therapeutic tool.