1. ORF PnP Program “Plug-and-Play” Medical Device Interoperability Standardization Framework for the Operating Rooms of the Future Julian M. Goldman, MD Depts. Of Anesthesia and Biomedical Engineering Massachusetts General Hospital Harvard Medical School A multidisciplinary program to standardize communication and control of medical devices to improve patient safety and healthcare efficiency C I M I T

2. Farnam Operating Amphitheatre, Yale New Haven http://www.ynhh.org/general/history/oldsur.html From This

3. To This

4. © 2005 Julian M. Goldman, MD www.orfpnp.org “minimally invasive surgery” yields “maximally invasive” technology!

5. © 2005 Julian M. Goldman, MD www.orfpnp.org What is an “OR of the Future”? The “Operating Room of the Future” (ORF) is not a specifically configured OR. “ORF” is shorthand for a constellation of emerging innovations in processes and technologies for perioperative care. The “ORF” may include: – Surgical robots – Minimally invasive surgery suites – Redesign of the perioperative environment

6. © 2005 Julian M. Goldman, MD www.orfpnp.org The ORF is a “state of mind”

7. CIMIT/MGH OR of the Future Project The ORF is a “living laboratory” to study the impact of process change, technology, and team work, on safety and productivity. Center for Integration of Medicine and Innovative Technology

8. Self contained suite of rooms Induction Emergence Control OR

9. W. Sandberg, PhD, MD Induction Room

10. Entering OR

11. Emergence Room

12. Click and go > 1 year development efforts!

14. Perioperative patient transport

15. Optimized monitor placement Could they self-configure?

16. Karl Storz OR1 surgical interface

19. © 2005 Julian M. Goldman, MD www.orfpnp.org Breaking news …

21. “ The President's Plan to Improve Care and Save Lives Through Health IT ” 1/27/05 The President is “ …committed to his goal of assuring that most Americans have electronic health records within the next 10 years” “Adopting Uniform Health Information Standards to allow medical information to be stored and easily shared electronically … standards for transmitting X- rays over the Internet; electronic lab results transmitted to physicians …” What’s missing?

22. © 2005 Julian M. Goldman, MD www.orfpnp.org

30. The “Last Mile” Proposed Health Information Technology innovations address many critical problems in medical record-related data communication Patient and clinician interaction with medical devices is not receiving the same attention Diagnosis and therapy is usually performed with medical devices!

31. © 2005 Julian M. Goldman, MD www.orfpnp.org Problem The OR is a complex and potentially hazardous environment We depend on teamwork and a patchwork of systems to mitigate hazards instead of using automated safety systems (e.g. “interlocks”) Clinicians are not empowered by information technology to achieve complete situational awareness, or to network and control medical devices in the environment – Absence of smart alarms and automated clinical decision support – Absence of technological infrastructure to implement the required solutions

32. © 2005 Julian M. Goldman, MD www.orfpnp.org Insufflation Vital Sign Changes Is the patient monitored?

33. © 2005 Julian M. Goldman, MD www.orfpnp.org Anesthesia Machine Safety Interlocks Vaporizers – only one may be on O 2 /N 2 O delivery “failsafe” must have high pressure O 2 source to deliver N 2 O

34. © 2005 Julian M. Goldman, MD www.orfpnp.org Safety Interlock Safety interlock at 100° F (38° C).

35. © 2005 Julian M. Goldman, MD www.orfpnp.org Interlocks Motels Individual devices Generally, NOT across medical device systems

36. Insufflation

37. Ventilation During Intraoperative Cholangiography Benefit of medical device interoperability: Synchronization

38. “With the advent of sophisticated anesthesia machines incorporating comprehensive monitoring, it is easy to forget that serious anesthesia mishaps still can and do occur.” APSF Newsletter Winter 2005 A 32-year-old woman had a laparoscopic cholecystectomy performed under general anesthesia. At the surgeon’s request, a plane film x-ray was shot during a cholangiogram. The anesthesiologist stopped the ventilator for the film. The x-ray technician was unable to remove the film because of its position beneath the table. The anesthesiologist attempted to help her, but found it difficult because the gears on the table had jammed. Finally, the x-ray was removed, and the surgical procedure recommenced. At some point, the anesthesiologist glanced at the EKG and noticed severe bradycardia. He realized he had never restarted the ventilator. This patient ultimately expired.

39. BP is low -> Stop Infusion Interoperability: Remote Control

40. © 2005 Julian M. Goldman, MD www.orfpnp.org Why is interoperability so difficult to implement? The OR is a clinical, non-engineered, environment ORF PnP problem space is at the nexus of hardware, software, and networking technologies Prior efforts failed to achieve interoperability Poorly-defined safety and business models

41. © 2005 Julian M. Goldman, MD www.orfpnp.org ORF PnP: The OR is a unique domain Real-time data/waveforms Real-time device control Zero error tolerance High production pressure Etc. etc. etc.

42. © 2005 Julian M. Goldman, MD www.orfpnp.org Clinical Hazard Mitigation: Examples of Proposed Safety “Interlocks” or guardrails HazardMitigation (example: Inadvertent car movement when engaging gear) (Cannot engage gear without depressing brake pedal) Abdominal CO 2 insufflation may cause adverse vital sign changes Ensure vital signs are monitored prior to insufflation (Smart alarm) Abdominal CO 2 insufflation cause adverse vital sign changes Halt insufflation if vital signs change (Device control) Starting surgical procedure with system unready Automated system readiness check Administration of contraindicated drugs (e.g. allergy) Automated drug ID cross- referenced to patient record w/alert

43. © 2005 Julian M. Goldman, MD www.orfpnp.org Smart Alarms & Decision Support Require data fusion Contextual information

44. SpO 2 (oxygen saturation measured by pulse oximetry) Time Blood Oxygen Desaturation Alarm limit Alarm Pulse Oximetry (SpO 2 ) monitor Question: Appropriate alarm, or nuisance alarm? Value of alarm depends on what will happen next … Example of the challenge of “Smart Alarms” J. Goldman, MD 2005

45. If the patient is an adult with Obstructive Sleep Apnea (OSA) SpO 2 Time Self-limiting desaturations Alarm limit J. Goldman, MD 2005 (Home monitoring or pre-operative setting)

46. If the patient is an adult with Obstructive Sleep Apnea (OSA) SpO 2 Time Self-limiting desaturations Alarm limit The two nuisance alarms reveal desaturations that occur every night, and probably do not require immediate intervention. J. Goldman, MD 2005

47. This patient has just received a narcotic for pain relief Time Ideal “ smart ” alarm threshold Alarm limit Narcotic Administration Dangerous oxygen desaturation SpO 2 Alarm heralds a potentially catastrophic desaturation! An intelligent system would have to know all the relevant contextual information. Context! J. Goldman, MD 2005

48. © 2005 Julian M. Goldman, MD www.orfpnp.org What contextual data was required to produce clinically meaningful alert? Oxygen saturation value (SpO 2 ) Patient demographics Co-morbidities (e.g. sleep apnea) Treatment environment (OR, hosp room, PACU, home, etc) Medications administered (dose, time)

49. © 2005 Julian M. Goldman, MD www.orfpnp.org Anesthesia Electronic Medical Record (EMR): Anesthesia Patient Safety Mandate (APSF) Vendors deal w/expensive headaches Institutions: same problem Operators: No standardization of data acquisition schemes, data averaging time, etc. No standardized structure to enable decision support systems

50. Cables required for MGH EMR

51. © 2005 Julian M. Goldman, MD www.orfpnp.org Future State with ORF PnP in place: Comprehensive Data Communication Comprehensive population of the EMR Enhanced clinical situational awareness Decision Support Quality assurance Process improvement Cost analysis Future State

52. © 2005 Julian M. Goldman, MD www.orfpnp.org ORF PnP at Work: Device Control Device-device control, safety interlocks Remote user actuation of devices – Bedside control of devices beyond reach, using innovative Human-Device Interfaces “Distributed” medical devices – distribute CPU, HCI, sensor networks, etc. – Physiologic Closed-Loop control Systems (ISO/IEC JWG5 and ASTM F29.21) Future State

53. EMR + Clinical “Rules” Interface Rules/Alarm settings, Process Exceptions “Ultimate State” ORF PnP at Work Patient data (demographics, labs) Clinical Context (e.g. stage of procedure) Image recognition, motes, etc. Vital Signs J. Goldman, MD 2005 Medical device state Remote alarms and data presentation

54. © 2005 Julian M. Goldman, MD www.orfpnp.org ORF PnP Program Goals Produce a framework for the delivery of safe and effective consensus medical device interoperability standards to industry, FDA, and clinical users. – Standards: Adopt, modify, create as needed – User Requirements – Engineering Requirements – ORF PnP Lab “sandbox” populated by medical devices in simulated OR/hospital environment Collaborate w/Industry, Academia (U Penn), Govnmt – HIMSS IHE “OR Profile” – Develop open-source standards “reference models” for assessment of safety – Create value statements/business case

55. ORF PnP System Attributes CapabilitySafetyEfficiency Ubiquitous data acquisition and presentation  Decrease technology deployment barriers  Enable safety interlocks  Extend connectivity of healthcare environment  Enable decision support  Enable adaptive alarms, closed loop control, sensor networks  Hot-swappable networked medical devices 

56. MeetingLocationAttendeesClinical/ Academic GovtIndustryCompanies 1 st PnP May 04 Cambridge844343721 At ASA Oct 04 Las Vegas2011-94 2 nd PnP Nov 04 FDA7528153222 At STA Jan05 Miami502 Phil Society Of Anes Phil30 U Wash March 05 Seattle25 SAGES April 05 Florida40 AAMI May 2005 Florida50 Meetings to Date

57. © 2005 Julian M. Goldman, MD www.orfpnp.org Directions for WGs on high-level clinical requirements The proposed standards could permit seamless connectivity of medical devices to allow data communication and control of medical devices. The unique challenge of this project is that we are striving to develop an infrastructure for innovation. Therefore, it is important to define the high-level general system requirements without getting bogged down in the details of technical specifications. Manufacturers/designers need to know what we need the system to do, so that they can determine how to best provide the desired functionality.

58. Think of examples of connectivity that could a) solve current clinical problems, b) improve safety or efficiency, or c) enable innovative clinical systems of the future. Assume that there are no technical, economic, legal, or regulatory obstacles to deploying a comprehensive ORF PnP system. What clinical challenges exist today that could be solved by the proposed system? Which obstacles to safety, efficiency, and teamwork could be reduced or eliminated by the proposed system? How would this approach affect the practice environment, both clinically and from a business perspective? What risks may be introduced by a PnP system, and how could they be mitigated?

59. www.ORFPnP.org For information about the OR of the Future project: www.CIMIT.org

61. AAMI BI&T May/June 2005

62. AAMI BI&T May/June 2005

63. © 2005 Julian M. Goldman, MD www.orfpnp.org Next ORF PnP Meeting June Monday June 6 – Tuesday June 7 Cambridge, Massachusetts (Next week!)

64. © 2005 Julian M. Goldman, MD www.orfpnp.org Thank you