PROJECT SUMMARY Based upon our previous work, one out of four intensive care unit (ICU) survivors will acquire cognitive impairment similar to Alzheimer’s disease and related dementias (ADRD) at 12-months. This leads to reduced employment, decreased quality of life, and loss of independence. This growing public health problem especially affects the 2 to 3 million ICU patients who are mechanically ventilated each year. Currently, mechanical ventilation of critically ill adults universally involves titration of the fraction of inspired oxygen (FiO2) to maintain arterial oxygen saturation (SpO2). The optimal SpO2 target, however, remains unknown; current recommendations range from tolerating SpO2 values as low as 88% (Acute Respiratory Distress Syndrome Network) to pursuing SpO2 values as high as 98% (British Thoracic Society). Because hypoxia is strongly implicated in ADRD pathogenesis, we hypothesize that targeting lower SpO2 during mechanical ventilation is significantly associated with worse long-term global cognition compared with higher SpO2 targets. To address this hypothesis, we propose this ancillary study to the NIH-funded Pragmatic Investigation of optimaL Oxygen Targets (PILOT) trial. The PILOT parent trial is a cluster-randomized, cluster-crossover trial comparing low (88% to 92%), intermediate (93% to 96%), and high (>96%) SpO2 targets during mechanical ventilation and will enroll 2,250 patients from the medical ICU over a three-year period. PILOT’s outcomes are ventilator-free days and mortality but is not assessing long-term global cognition. Therefore, we propose this ancillary study titled Cognitive Outcomes in PILOT (CO-PILOT) which has the following specific aims: (1) determine if the lower SpO2 target (88 to 92%) is significantly associated with worse 6-month global cognition compared with higher (>92%) SpO2 targets in mechanically ventilated ICU patients; and (2) determine how low SpO2 targets (88-92%) affects individual cognitive domains, particularly immediate and delayed memory, at 6-months compared with higher (>92%) SpO2 targets among mechanically ventilated ICU patients. CO-PILOT will enroll a subset of 189 survivors who are enrolled in PILOT over an 18-month period. This R21 will provide the funding and resources needed to obtain 6-month cognitive and secondary outcome measurements. Six-month global cognition and its individual cognitive domains (intermediate and delayed memory, attention, visuospatial construction, and language) will be determined by face-to-face neuropsychological testing using the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS). It is possible that the PILOT parent trial may report that a lower SpO2 target improves survival, but the CO-PILOT ancillary trial may reveal that this will come at the expense of long-term cognition. Because patients and their families likely have different preferences for what they consider the ideal outcome is, both studies are necessary to provide a comprehensive evaluation of SpO2 targets on a wide range of outcomes.

PROJECT SUMMARY One out of four intensive care unit (ICU) survivors will develop long-term cognitive impairment (LTCI) similar to Alzheimer’s disease and related dementias at 12-months. Hypoxia during mechanical ventilation is a risk factor for LTCI and may be a prominent driver for its pathogenesis. We propose this cluster-randomized, cluster- crossover study to determine if lower arterial oxygen saturation targets (88% to 92%) during mechanical ventilation is significantly associated with worse 6-month cognition compared with higher (>92%) oxygen saturation targets in patients admitted to the medical ICU.