Hydrostatic weighing without head submersion: description of a method

Hydrostatic weighing (HW) was performed at residual volume (RV) and total lung capacity without head submersion (TLCNS). Ninety-five males (25.6 +/- 4.9 yr) and 87 females (22.6 +/- 5.2 yr) were studied at two laboratory sites using identical protocols. Twenty males and 20 females were separated from the original group and randomly assigned to cross-validation groups. RVs were determined by the oxygen dilution method. Vital capacity was determined with the subject submerged in water to the shoulders. Underwater weight was determined using 10 trials at RV and 5 trials at TLCNS, with the order of methods randomly assigned. Regression analysis provided an equation to predict body density (pDb) at RV from body density (Db) at TLCNS. The equation for males was pDb (HW at RV) = 0.5829 (DbHW at TLCNS) + 0.4059, r = 0.88, SEE = 0.0067. The equation for females was pDb (HW at RV) = 0.4745 (DbHW at TLCNS) + 0.5173, r = 0.85, SEE = 0.0061. Cross-validation showed no significant differences using Db from HW at RV (males = 1.0626 g.ml-1, females = 1.0493 g.ml-1 and pDb from HW at TLCNS (males = 1.0625 g.ml-1, females = 1.0479 g.ml-1). The correlation coefficient SEE and total error for males were r = 0.95, 0.0043, and 0.0041, respectively and for females r = 0.82, 0.0084, and 0.0085, respectively. Mean percent fat for RV and TLCNS was identical for males and differed by 0.7% for females. Test-re-test data indicated the TLCNS procedure was reliable (r = 0.98).(ABSTRACT TRUNCATED AT 250 WORDS)