Comparison of consciousness level assessment in the poisoned patient using the alert/verbal/painful/unresponsive scale and the Glasgow Coma Scale.

STUDY OBJECTIVE We determine how the alert/verbal/painful/unresponsive (AVPU) responsiveness scale (alert, responsive to verbal stimulation, responsive to painful stimulation, and unresponsive) corresponds to the Glasgow Coma Scale (GCS) when assessing consciousness level in the poisoned patient. METHODS Consciousness level was assessed using the AVPU responsiveness scale and the GCS in all patients admitted to the hospital during a 6-month period with deliberate or accidental poisoning. An AVPU responsiveness scale algorithm and details of the individual components of the GCS were provided. Data were recorded prospectively on admission to the toxicology ward by nursing staff in the majority of cases and from case records for the small number of patients admitted directly to the ICU. Nursing staff also recorded any difficulty assessing consciousness level using either scoring system. RESULTS Of the 1,384 patients studied, 1,138 patients were alert, 114 patients responded to a verbal stimulus, 87 patients responded to a painful stimulus, and 15 patients were unresponsive. The median GCS scores with interquartile ranges (IQR) for each AVPU responsiveness category were 15 (IQR 15), 13 (IQR 12 to 14), 8 (IQR 7 to 9), and 3 (IQR 3), respectively. There was a degree of overlap between the range of GCS scores for each category. Nursing staff recorded more difficulty using the GCS than the AVPU responsiveness scale. Alcohol-intoxicated patients proved to be the most difficult to assess. All patients who were unresponsive required intubation. No patient with a GCS score greater than 6 was intubated. CONCLUSION Each AVPU category can be shown to correspond to a range of GCS scores. The AVPU responsiveness scale appears to provide a rapid simple method of assessing consciousness level in most poisoned patients, but difficulty was still observed in assessing alcohol-intoxicated patients.