F Gregory Murphy’s scientific contributions

Introduction: Venous gas emboli (VGE) are a common surrogate experimental endpoint for decompression sickness (DCS). VGE numbers are graded, and the peak post-dive grade is associated with the probability of DCS (PDCS). VGE are typically graded with the subject at rest when bubble numbers are stable, and again after limb flexions which elicit a transient shower of bubbles. Detection of VGE using two-dimensional (2-D) echocardiography has become common, but the principal grading scales do not specify how to grade VGE after limb movement. Methods: This was a retrospective analysis of 1,196 man-dives following which VGE were detected using 2-D echocardiography and graded on a scale 0–4 and 41 cases of DCS occurred. PDCS was estimated for each peak post-dive VGE grade from the cumulative incidence of DCS. Two different definitions of movement VGE grades were assessed in 84 measurements; the grade was either the maximum VGE number sustained for one diastole (1-cycle) or for six cardiac cycles (6-cycle). Results: For each peak post-dive VGE grade (maximum of rest or movement) the cumulative incidences of DCS (%) were: grade 0 (0%); grade 1 (1.3%); grade 2 (2.5%); grade 3 (4.6%); grade 4 (5.7%). When grading movement VGE, 57% of 1-cycle grade 4 were reduced to grade 3 using the 6-cycle definition. Conclusions: There is a need for consensus in the research community on how to assign movement VGE grades when using 2-D echocardiography. Publications should carefully explain methodology for assigning VGE grades and consider differences in methodologies when comparing historical data sets.

Introduction: Venous gas emboli (VGE) are widely used as a surrogate endpoint instead of decompression sickness (DCS) in studies of decompression procedures. Peak post-dive VGE grades vary widely following repeated identical dives but little is known about how much of the variability in VGE grades is proportioned between-diver and within-diver. Methods: A retrospective analysis of 834 man-dives on six dive profiles with post-dive VGE measurements was conducted under controlled laboratory conditions. Among these data, 151 divers did repeated dives on the same profile on two to nine occasions separated by at least one week (total of 693 man-dives). Data were analysed for between- and within-diver variability in peak post-dive VGE grades using mixed-effect models with diver as the random variable and associated intraclass correlation coefficients. Results: Most divers produced a wide range of VGE grades after repeated dives on the same profile. The intraclass correlation coefficient (repeatability) was 0.33 indicating that 33% of the variability in VGE grades is between-diver variability; correspondingly, 67% of variability in VGE grades is within-diver variability. DCS cases were associated with an individual diver’s highest VGE grades and not with their lower VGE grades. Conclusions: These data demonstrate large within-diver variability in VGE grades following repeated dives on the same dive profile and suggest there is substantial within-diver variability in susceptibility to DCS. Post-dive VGE grades are not useful for evaluating decompression practice for individual divers.