Pupillary abnormalities in non-selected critically ill patients: an observational study.

BACKGROUND: Repeated pupillary examination is a key element of neurologic surveillance in intensive care units (ICU). However, in non-selected critically ill patients, the clinical interest of monitoring pupillary diameter and light reflex is poorly documented. We aimed to determine the prevalence and the etiologies of pupillary abnormalities (PAs) in this ICU patient population.

METHODS: We performed a prospective, observational study in a medical university affiliated ICU over a 6-month period. All patients with at least one pupillary examination were included. PA was defined as areflexia and/or anisocoria present at the time of ICU admission or occurring during the ICU stay.

RESULTS: During the study period, we included 297 patients who had 6±9 pupillary examinations per day (totaling 11,360 pupillary assessments). The majority of patients (n=161, 54%) were admitted to the ICU for acute respiratory or cardiovascular failure. A total of 128 PAs were recorded in 109 patients: 78 areflexia alone (61%), 33 anisocoria alone (26%) and 17 (13%) with associated anisocoria and areflexia. The main causes of PAs were related to acute brain ischemia (n=41, 32%) and sedation/analgesia (n=50, 39%). Among the PAs, 59 (46%) were present upon ICU admission. The etiologies of the PAs at admission did not differ from those occurring during ICU stay (P=NS). Interestingly, 9 (7%) PAs were attributed to ipratropium nebulization in patients with chronic obstructive pulmonary disease exacerbation.

CONCLUSIONS: The high prevalence of PAs, frequently associated with both brain organic lesions and drug side effects, highlights the clinical interest of pupillary surveillance in non-selected critically ill patients.