Predicted body weight during mechanical ventilation: using arm demispan to aid clinical assessment

Recent research suggests an association between the development of acute lung injury (ALI) and mechanical ventilation with tidal volumes > 6mL per kg of predicted body weight (BW). Specific subgroups (women and obese patients) may be at risk of unintentional delivery of excessive tidal volumes. We conducted a prospective audit of delivered tidal volumes (mL/kg) calculated using recorded BW and compared these to volumes calculated using predicted BW.

Patients requiring controlled mechanical ventilation admitted to a mixed intensive care unit in October 2006 were eligible. Exclusion criteria were ALI on admission or no recorded BW (defined as a weight measured by scales or a dietitian-estimated weight) for the current admission.

Arm demispan was used to calculate height, and predicted BW was derived using ARDSNet formulas. Hourly Day 1 tidal volumes were downloaded from the medical record, and the mean for each patient was calculated. Volumes (mL/kg) were calculated using predicted and recorded BW. Data are presented as mean (SD) or median (interquartile range) depending on normality. The Mann-Whitney rank sum test was used for comparisons.

34 patients were studied (20 men) with a mean age of 60.6 (SD, 13.3) years and mean APACHE II score of 19.5 (SD, 6.1). Predicted BW was lower than recorded BW (69.0 [61.0-74.8] v 75 [65-85] kg; P < 0.05). Median tidal volumes (mL) were higher for men than women (552 [530- 586] v 474 [424-500] ; P < 0.01). Tidal volumes expressed as mL/kg were higher when calculated from predicted BW than from recorded BW (7.8 [7.3-8.3] v 7.2 [6.3-7.9]; P<0.05). Volumes calculated using predicted BW were higher among women than men (8.2 [7.8-8.7] v 7.5 [6.8-8]mL/kg; P < 0.05). The difference in volume between the sexes using recorded weight was not significant (7.5 [6.6-8.6] v 6.9 [6.2-7.8]mL/kg; P=0.42).

Predicted BW was significantly less than recorded BW. Consequently, larger tidal volumes were delivered on a mL/kg basis when calculated using predicted BW than recorded BW. This was particularly so for women, who received higher volumes than men when using predicted BW. Calculating predicted BW using demispan as a surrogate marker of height is a cheap, easy and noninvasive tool for clinical assessment; its use in the ICU may result in the delivery of more appropriate tidal volumes.