APSF Awards 2020 Grant Recipients

Steven K. Howard, MD

The APSF’s Investigator-Initiated Research Program supports the Mission Statement that includes the goal to continually improve the safety of patients during anesthesia care by encouraging and conducting safety research and education. The APSF has funded over 9 million dollars on patient safety research since 1987 to help achieve these goals. This year’s grants get at the heart of two long-standing anesthesia-related safety issues: malignant hyperthermia and evaluation of the airway.

The 2019–20 APSF Investigator-Initiated Research Grant Program received 27 letters of intent submitted in early February 2019. After a thorough evaluation, five teams were invited to submit full proposals. On October 19, the Scientific Evaluation Committee met in Orlando, FL, during the American Society of Anesthesiologists national meeting to make funding recommendations to the APSF Board of Directors. Two recommendations were reviewed and subsequently accepted.

The principal investigators of this year’s APSF grant provided the following description of their proposed work.

Sheila Riazi, MSc, MD, FRCPC

Sheila Riazi, MSc, MD, FRCPC

Sheila Riazi, MSc, MD, FRCPC

Associate Professor, Department of Anesthesia, University Health Network, University of Toronto, Toronto, Canada

Dr. Riazi’s proposal is entitled “A Minimally Invasive Diagnostic Test for Malignant Hyperthermia.”

Background: Malignant hyperthermia (MH) is a potentially fatal hereditary disorder that is induced by certain anesthetics. Although rare, a malignant hyperthermia crisis is one of the most feared adverse anesthetic outcomes because 1 in 10 patients die and 1 in 3 experience complications.1,2 Thus, preventing exposure to triggering anesthetics is crucial in patients who are susceptible to MH. However, screening patients for MH remains challenging as genetic testing identifies only half of all susceptible individuals.2 The current standard diagnostic test for MH susceptibility—the caffeine-halothane contracture test (CHCT)—is sensitive (97–100%) but invasive and costly. Furthermore, it requires travel to one of the few specialized centers worldwide because the test must be completed within 5 hours after the muscle biopsy. Therefore, only about 4% of those with suspected MH susceptibility undergo the standard diagnostic test.3

The calcium-induced calcium release (CICR) test is an alternative, less-invasive means to diagnose MH susceptibility. Unlike the CHCT, the CICR test requires a small muscle sample that could be harvested in a physician’s office and shipped for analysis at a specialized testing center up to 72 hours after the biopsy. In addition to these advantages, the CICR test is the standard for MH susceptibility diagnosis in Japan, despite lacking rigorous validation against the standard CHCT.4 Our MH testing center is currently the only site worldwide with the expertise to conduct both the CHCT and the CICR test, placing our research team in an advantageous position. We therefore propose a single-center, prospective cohort study to validate the alternative CICR test against the standard CHCT by performing both tests simultaneously in samples from every patient referred to our center.

Aims: Our overall goal is to demonstrate that the alternative CICR test is a suitable replacement for the standard CHCT for diagnosing MH susceptibility. Given the advantages of CICR over CHCT, our primary aim is to evaluate whether the sensitivity of the alternative CICR test is greater than 80%, using the CHCT as the reference standard. Our preliminary data have proved the feasibility of CICR and have shown promising results.

Implications: This research will be the first to validate CICR against CHCT, taking advantage of our unique position as the only laboratory worldwide with expertise in both tests. Not only could this work increase use of a less expensive and less invasive diagnostic test, it could also increase the uptake of MH-susceptibility testing beyond the current 4%, ultimately improving patient safety.

Funding: $137,449 (January 1, 2020–December 31, 2021). This grant was designated the APSF/ASA Presidents’ Research Award. Dr. Riazi is also the recipient of the Ellison C. “Jeep” Pierce, Jr., MD, Merit Award, which provides an additional, unrestricted amount of $5,000.

References

  1. Larach MG, Gronert GA, Allen GC, et al. Clinical presentation, treatment, and complications of malignant hyperthermia in North America from 1987 to 2006. Anesth Analg. 2010;110:498–507.
  2. Riazi S, Kraeva N, Hopkins PM. Malignant Hyperthermia in the postgenomic era: new perspective on an old concept. Anesthesiology. 2018;128:168–180.
  3. Jones PM, Allen BN, Cherry RA, et al. Association between known or strongly suspected malignant hyperthermia susceptibility and postoperative outcomes: an observational population-based study. Can J Anaesth. 2019;66:161–181.
  4. Oku S, Mukaida K, Nosaka S, et al. Comparison of the in vitro caffeine-halothane contracture test with the Ca-induced Ca release rate test in patients suspected of having malignant hyperthermia susceptibility. J Anesth. 2000;14:6–13.

 

Scott Segal, MD

Scott Segal, MD

Scott Segal, MD

Thomas H. Irving Professor and Chair, Department of Anesthesiology Wake Forest School of Medicine

Dr. Segal’s project is entitled “Development of machine learning algorithms to predict difficult airway management.”

Background: Successful airway management is fundamental to safe anesthetic performance, and airway management failure continues to be the one of the leading causes of anesthesia-related death and severe morbidity.1 While preoperative airway assessment is considered the worldwide standard of care, 75–93% of difficult intubations are unanticipated, and all easily performed airway examination systems in clinical practice perform only modestly to detect difficult intubations.2 We propose to create a machine learning system based on analysis of facial photographs which could outperform conventional bedside tests and human experts and improve airway management and patient safety. Previous work by our group has demonstrated that an algorithm based on supervised (i.e., human-assisted) computer analysis of facial images combined with thyromental distance (TMD) can outperform classical bedside tests and human experts.3 Here we propose to extend this work by the development of completely unsupervised computer algorithms based on feature extraction from facial photographs by convoluted neural networks (CNNs).

Aims: CNN technology already exists for highly accurate deterministic feature extraction of frontal views of the face and is widely employed in facial recognition applications. We will develop a similar CNN-based feature extractor from profile views of the face, which likely contains important information about potential intubation difficulty (Aim 1). We will then fuse this information with frontal facial information and patient demographics and bedside airway data (TMD and Mallampati class [MP]) and train an advanced algorithm to classify faces as easy- or difficult-to-intubate based on prospective observation of ground truth during induction of general anesthesia (Aim 2). We will compare performance of the derived algorithm to MP+TMD in both the derivation dataset as well as an independent validation dataset. We will test the hypothesis that the computer-derived algorithm will outperform classical bedside tests and improve prediction of difficult intubation. Finally, we will build a smartphone-based data entry tool to capture photographs, patient demographic information, and bedside airway examination data and transmit it to an HIPAA-compliant, encrypted online database (Aim 3). This will form the basis of a future completely automated airway prediction tool, based on our methods derived in this investigation.

Implications: Airway failure is still the #1 cause of anesthesia-related mortality, and most difficult intubations are unanticipated, but there are well-established guidelines and ever-expanding options for the management of the anticipated difficult airway.4 Therefore any improvement in airway prediction is likely to improve patient safety. Failed airway management is even more common outside the OR, and the safety improvement may be even more profound in the emergency department, ICU, or prehospital setting. Our proposal is closely aligned with the APSF’s current funding priorities, as it involves large numbers of patients, including the healthiest, uses advanced information technology to prevent harm, and focuses on a low-frequency but devastating complication, difficult or failed intubation. In the future, prediction of difficulty with additional aspects of airway management, including bag-mask ventilation, could also be modeled. The data entry tool could be coupled to a cloud-based feature extraction and prediction calculator, returning a prediction to end-users.

Funding: $150,000 (January 1, 2020–December 31, 2021). This grant was designated the APSF/Medtronic Research Award.

References

  1. Joffe AM, Aziz MF, Posner KL, et al. Management of difficult tracheal intubation: a closed claims analysis. Anesthesiology. 2019;131:818–29.
  2. Cook TM, MacDougall-Davis SR. Complications and failure of airway management. Br J Anaesth. 2012;109 Suppl 1:i68–i85.
  3. Connor CW, Segal S. Accurate classification of difficult intubation by computerized facial analysis. Anesth Analg. 2011;112:84–93.
  4. Apfelbaum JL, Hagberg CA, Caplan RA, et al. Practice guidelines for management of the difficult airway: an updated report by the American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology. 2013;118:251–270.

The APSF would like to thank the above researchers and all grant applicants for their dedication to improve patient safety.

 

Dr. Howard is staff anesthesiologist at the VA Palo Alto Health Care System, professor of Anesthesiology at Stanford University School of Medicine and chair of the APSF Scientific Evaluation Committee.


Dr. Howard serves on the Board of Directors of the APSF and has no other conflicts of interest to declare.