Elucidating the fuzziness in physician decision making in ARDS

Abstract

The current standard of care for patients suffering from acute respiratory distress syndrome (ARDS) is ventilation with a tidal volume of 6 ml/kg predicted body weight (PBW), but variability remains in the tidal volumes that are actually used. This study aims to identify patient scenarios for which there is discordance between physicians in choice of tidal volume and positive end-expiratory pressure (PEEP) in ARDS patients. We developed an algorithm based on fuzzy logic for encapsulating the expertise of individual physicians regarding their use of tidal volume and PEEP in ARDS patients. The algorithm uses three input measurements: (1) peak airway pressure (PAP), (2) PEEP, and (3) arterial oxygen saturation (SaO₂). It then generates two output parameters: (1) the deviation of tidal volume from 6 ml/kg PBW, and (2) the change in PEEP from its current value. We captured 6 realizations of intensivist expertise in this algorithm and assessed their degree of concordance using a Monte Carlo simulation. Variability in the tidal volume recommended by the algorithm increased for PAP > 30 cmH₂O and PEEP > 5 cmH₂O. Tidal volume variability decreased for SaO₂ > 90 %. Variability in the recommended change in PEEP increased for PEEP > 5 cmH₂O and for SaO₂ near 90 %. Intensivists vary in their management of ARDS patients when peak airway pressures and PEEP are high, suggesting that the current goal of 6 ml/kg PBW may need to be revisited under these conditions.

Fig. 1

Fuzzy set structure. A) The ranges for the input parameters PAP, PEEP and SaO₂ are divided into 2, 3 and 2 overlapping sets, respectively. The sets shown in the figure are illustrative examples; the precise positions of the vertices defining the positions of the sets were provided by each of the physicians we studied and used as part of the setting up process for their specific algorithm. B) The ranges of the output parameters ΔV_t and ΔPEEP are each divided into 3 overlapping sets.

Fig. 2

SD-ΔV_T and SD-ΔPEEP plotted against each of the input parameters PAP, PEEP, and SaO₂. The darkness of the plot is proportional to the number of the 10,000 individual patient scenarios that fall on a given value.