A methodology to explore ventilatory chemosensitivity and opioid-induced respiratory depression risk

Reported incidence of postoperative opioid-induced respiratory depression (OIRD) ranges from 0.5-41% and is not reliably predicted by traditional risk factors. This study tests a new methodology to investigate ventilatory chemosensitivity as a new potential risk factor and explore OIRD distribution across sleep and wakefulness. Preoperative patient ventilatory chemosensitivity was quantified by hypercapnic ventilatory responses with (HCVRREMI, effect site concentration 0.7 or 2.0 ng/mL) and without (HCVRBL) remifentanil during hyperoxia and hypoxia. Postoperative opioid consumption was recorded during hospital stays. OIRD frequency was the primary outcome of the study, detected as incidences of respiratory rate < 60% of baseline, minute ventilation < 60% of predicted value, pulse oximetry SpO(2) < 90% (breathing room air) or 92% (supplemental O-2), transcutaneous PCO2 > 50 mmHg, and central and obstructive apnea/hypopnea. Sleep stages were recorded until the first postoperative morning to determine the OIRD sleep distribution as the secondary outcome. The methodology was feasible in implementation and posed no obstacles to standard care. In the nine patients studied (2 females, mean age 65 +/- 7.5 yr), remifentanil depressed HCVR to a highly variable degree. High OIRD frequency was generally observed with lower HCVRREMI. OIRD predominantly occurred during light sleep. This study supports ventilatory chemosensitivity as an important predictor of OIRD, lending a new perspective to classify risk for OIRD and detailing a methodology in which to pursue this investigation for future studies. NEW & NOTEWORTHY Our new and noteworthy methodology allows for exploration of preoperative ventilatory chemosensitivity, measured as the hypercapnic ventilatory response (HCVR), as a risk factor for postoperative opioid-induced respiratory depression (OIRD). This feasible and reliable methodology produced preliminary data that showed highly variable depression of HCVR by remifentanil, predominance of OIRD during light sleep, and potentially negative correlation between OIRD frequency generally and HCVR measurements when measured in the presence of remifentanil. Although the results are preliminary in nature, this novel methodology may guide future studies that can one day lead to effective clinical screening tools.