Decompression illness medically reported by hyperbaric treatment facilities: cluster analysis of 1929 cases

Area of Science:

• Diving Medicine
• Hyperbaric Medicine
• Biostatistics

Background:

• Decompression illness (DCI) lacks consensus on clinically useful subclasses.
• Objective classification of DCI manifestations is needed.

Purpose of the Study:

• To explore an objective DCI classification using multivariate statistics.
• To assess naturally associated clusters of DCI manifestations.
• To evaluate mapping onto other DCI classifications and therapeutic outcomes.

Main Methods:

• Utilized two-step cluster analysis and Bayesian information criterion.
• Confirmed with hierarchical clustering (squared Euclidean distances, Ward's method).
• Analyzed 1929 DCI cases from Divers Alert Network (DAN America) (1999-2003).

Main Results:

• Identified four robust and significant clusters of DCI manifestations.
• Cluster 1: Pain only.
• Cluster 2: Numbness, paresthesia, decreased skin sensitivity.
• Cluster 3: Malaise, paralysis, muscular weakness, bladder-bowel dysfunction.
• Cluster 4: Diverse symptoms including hearing loss, skin mottling, confusion, dyspnea, vision problems, altered consciousness, and vertigo.

Conclusions:

• Clusters mapped poorly to traditional DCI categories (AGE, Type I/II DCS) but better to the Perceived Severity Index (PSI).
• All classification methods predicted complete relief equally well.
• Cluster analysis provides an objective method for classifying DCI manifestations.