Previous Abstract | Next Abstract
Printable Version
A3015
October 15, 2018
10/15/2018 10:30:00 AM - 10/15/2018 12:00:00 PM
Room North, Room 22
The Clinical Application of Noninvasive Minute Ventilation Monitor in the Perioperative Setting: Preliminary Results from Four Sites within Kaiser Permanente Medical System
Chunyuan Qiu, M.D., M.S., Eugene Y. Cheng, M.D., Diana LaPlace, M.D., Jamie Lim, P.A., Mijiin Lee-Brown, M.D., Atef Morkos, M.D., Maria T. Enciso, B.S.N., Fang-Chen Hou, B.S.N., Shou Yen, B.S.N., Vimal N. Desai, M.D.
Kaiser Permanente, Baldwin Park, California, United States
Disclosures: C. Qiu: None. E.Y. Cheng: None. D. LaPlace: None. J. Lim: None. M. Lee-Brown: None. A. Morkos: None. M.T. Enciso: None. F. Hou: None. S. Yen: None. V.N. Desai: None.
Introduction: Combined effects of inhalation agents, narcotics, residual muscle relaxants, and neuroactive medications on respiratory function can be devastating. Avoidable respiratory depression is highlighted in the Anesthesia Closed Claim Project, detailing malpractice claims against anesthesiologists. Previously, patient monitoring lacked clinically useful, timely warning of impending respiratory compromise. Both pulse-oximetry and capnography are late indicators, fraught with nuisance alarms from patient motion or probe malposition. Recent introduction of noninvasive respiratory volume monitoring (RVM) has implications for managing respiration, non-invasively displaying of minute ventilation (MV) and respiratory rate. We evaluated the effectiveness of the RVM in reducing the incidence of nuisance alarms and alarm fatigue in one of the largest integrated healthcare system in US: Kaiser Permanente (KP).Methods: This observational study was conducted at four KP hospital sites. An RVM (ExSpiron1Xi, Waltham, MA) was used in the post-anesthesia-care-unit (PACU) and on the general hospital floor (GHF). RVM alarms and staff response to alarms were recorded and analyzed by non-study staff. Alarms were divided into four categories: 1) actionable and addressed, 2) actionable and not addressed, 3) self-corrected, and 4) technical. The action taken to resolve each alarm was recorded and further categorized. Self-corrected alarms were those that resolved without staff intervention, usually by the patient being stimulated by the audible RVM alarm. Technical alarms did not require intervention and were considered a nuisance. This study was initiated, funded and conducted in KP as part of a standard evaluation program. Results: 247 patients (age 60.9 ± 13.9 years (22-93), BMI = 32.7 ± 9.8 kg/m2 (14.8-75.8), 143 females) were enrolled. Of those, 104 patients were only monitored in the PACU and 109 patients were monitored in the PACU and GHF, for a total of 2321 hours. 605 total RVM alarms occurred, ~1 alarm every 4 patient-hours. 64% RVM alarms were actionable and addressed by hospital staff; 20% of actionable alarms were “not-addressed”; and 13% of alarms deemed “real” resolved without any intervention. Only 6% of RVM alarms were technical (nuisance) (Figure 1). The most common intervention was direct patient stimulation, accounting for ~2/3 of all interventions in the PACU and ~80% of all interventions on the GHF (Figure 1). Other interventions included patient repositioning and reduction in opioid dosing. With a focus on early warning, none of the patients with MV alarms had respiratory related negative events. Conclusions: Cardiac monitoring advances have proven invaluable for managing and safeguarding patients against cardiac-related complications. Unfortunately, respiratory monitoring has lagged behind and drug-related respiratory compromise has become the leading cause of preventable perioperative death. Using SpO2 and EtCO2 monitoring in an attempt to curtail these deaths has resulted in significant increase of non-actionable nuisance alarms and overburdening of clinical staff without clear improvement in mortality. Our results indicate that the RVM generates primarily actionable alarms, with a high intervention-to-false-alarm ratio. More data is required to evaluate whether this earlier identification of hypoventilation will lead to a meaningful improvement of outcomes.

References: 1. Best Pract Res Anaesthesiol 2011:263-76 2. J Neurosurg Anesthesiol 2017:298-303 3. Anesth Analg 2017:120-126

Figure 1

Copyright © 2018 American Society of Anesthesiologists