Society for Technology in Anesthesia Section Editor: Maxime Cannesson
E Technical Communication
A Technical Evaluation of Wireless Connectivity from Patient Monitors to an Anesthesia Information Management System During Intensive Care Unit Surgery Allan F. Simpao, MD,*† Jorge A. Galvez, MD,*† W. Randall England, BA,† Elicia C. Wartman, BA,† James H. Scott, CNA,† Michael M. Hamid, Sr., CE,† Mohamed A. Rehman, MD,*† and Richard H. Epstein, MD‡ Surgical procedures performed at the bedside in the neonatal intensive care unit (NICU) at The Children’s Hospital of Philadelphia were documented using paper anesthesia records in contrast to the operating rooms, where an anesthesia information management system (AIMS) was used for all cases. This was largely because of logistical problems related to connecting cables between the bedside monitors and our portable AIMS workstations. We implemented an AIMS for documentation in the NICU using wireless adapters to transmit data from bedside monitoring equipment to a portable AIMS workstation. Testing of the wireless AIMS during simulation in the presence of an electrosurgical generator showed no evidence of interference with data transmission. Thirty NICU surgical procedures were documented via the wireless AIMS. Two wireless cases exhibited brief periods of data loss; one case had an extended data gap because of adapter power failure. In comparison, in a control group of 30 surgical cases in which wired connections were used, there were no data gaps. The wireless AIMS provided a simple, unobtrusive, portable alternative to paper records for documenting anesthesia records during NICU bedside procedures. (Anesth Analg 2016;122:425–9)
T
he use of an anesthesia information management system (AIMS) during bedside procedures in the intensive care unit (ICU) can be challenging when monitoring equipment needs to be cabled directly to the AIMS workstation. Because of logistical considerations, the workstation must be situated at some distance from the patient, creating potential tripping hazards because of the cables and the possible need for duplicate monitoring equipment and sensors attached to the patient.1 A wireless connection, if sufficiently reliable, would eliminate the need for such cabling and thus reduce the barrier to the use of the AIMS in the ICU and in other locations where space is at a premium. We established the need for such a solution through a paper survey distributed to 10 anesthesiologists from 10 pediatric hospitals who attended the Biomedical Informatics Special Interest Group meeting held at the Society for Pediatric Anesthesia’s International Assembly for Pediatric Anesthesia in Washington, DC, on October 11, 2012 (Table 1). Ten surveys from separate institutions were returned with 9 indicating that they would use their AIMS for all ICU patients
From the *Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania; †The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania; and ‡Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, Pennsylvania. Accepted for publication August 21, 2015. Funding: Departmental. The authors declare no conflicts of interest.
if a wireless solution were available and 4 indicating that they currently use their AIMS during bedside cases (Table 2). The Bluetooth wireless transmission protocol has been shown under laboratory conditions2 and in a small (n = 4 patient) operating room (OR) pilot study3 to be a potentially suitable replacement for transmission of data via serial data cables in the ICU. Here, we describe the implementation, technical assessment, and performance testing during 30 bedside ICU procedures using inexpensive, readily available Bluetooth wireless transponders.
METHODS
This study did not meet the definition of human subjects research for the IRB at The Children’s Hospital of Philadelphia, and the requirement for written informed consent was waived. We used Bluetooth wireless serial adapters (HdwBTRS232; WCSC Inc., Houston, TX) to transmit vital sign data (electrocardiogram, heart rate, respiratory rate, noninvasive blood pressure, Spo2, ventilator parameters, etc.) from the bedside monitor (Solar 8000i; GE Healthcare, UK)a to a portable computer thick client workstation (Lenovo M92 Tiny, Morrisville, NC; Microsoft Windows 7, Redmond, WA) running our AIMS (CompuRecord; Philips, Andover, MA). The wireless adapters were paired and configured using a PC HyperTerminal application (Hilgraeve, Monroe, MI): 19,200 bits per second, 8 data bits, no parity bit, 1 stop bit, and hardware flow control.b One wireless adapter was connected to the patient monitor’s RS-232
Reprints will not be available from the authors. Address correspondence to Allan F. Simpao, MD, Perelman School of Medicine, University of Pennsylvania, The Children’s Hospital of Philadelphia, 3401 Civic Center Blvd., Philadelphia, PA 19104. Address e-mail to simpaoa@ email.chop.edu. Copyright © 2016 International Anesthesia Research Society DOI: 10.1213/ANE.0000000000001064
February 2016 • Volume 122 • Number 2
GE Healthcare Solar 8000i Patient Monitor Manual. Available at: http://goo. gl/tvjT58. Accessed August 14, 2015. b Bluetooth Wireless Serial Port Adapter Support (HdwBTRS232). Available at: http://goo.gl/puOTHl. Accessed August 14, 2015. c Baytech M Series Data Acquisition and Control Owner’s Manual. Available at: http://goo.gl/eawYfH. Accessed August 14, 2015. a
www.anesthesia-analgesia.org
425
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
E Technical Communication Table 1. Wireless Intensive Care Unit Anesthesia Information Management System Needs Assessment Survey Questions 1. Does your institution have an anesthesia information management system (AIMS) currently installed (e.g., CompuRecord, Innovian, DocuSys)? Yes No 2. At your institution, approximately how many anesthesia-monitored cases are performed at the intensive care unit (ICU) bedside per year? ______ ICU bedside cases monitored by anesthesia providers per year 3. If your institution has an AIMS installed, please answer the following 3 questions: a. Are patient monitors connected directly to the AIMS workstation or to a central database? Direct to computer Central database via the network Don’t know b. What percentage of your institution’s bedside ICU cases are recorded in the AIMS? 0% 100% ____% (fill in the percentage) c. If there were a simple wireless connection that would allow connecting remotely the patient monitor to a mobile computer without the need to attach cables or provide duplicate monitors, would you use your AIMS in these locations for all patients? Yes No
Table 2. Pediatric Anesthesiologists’ Responses to Wireless Intensive Care Unit Anesthesia Information Management System Needs Assessment Survey Respondent 1 2 3 4 5 6 7 8 9 10
AIMS in use
No. of yearly bedside NICU cases
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
6 Unknown 50 5 0 40 10 500 10 250
Patient monitors connection Computer serial port Network Unknown X
No. of bedside NICU cases recorded in AIMS (%)
Would use wireless NICU AIMS if available
100 N/A 0 100 N/A 0 100 100 0 0
Yes Yes Yes Yes N/A Yes Yes Yes Yes Yes
X X
X X X X X X X X
AIMS = anesthesia information management system; NICU = neonatal intensive care unit; N/A = not applicable.
serial port, and a second paired wireless adapter was connected to the AIMS workstation through a data acquisition and control unit (M4 Series; BayTech, Long Beach, MS)c (Fig. 1). To test for interference, we activated an electrosurgical generator (Force FX; Valleylab, Boulder, CO)d located between the GE Solar and the AIMS workstation during data acquisition using vital sign simulator modules. The distance between the patient monitor and the AIMS workstation was 4 m. The electrosurgical generator was set at the maximal monopolar coagulation power setting (120 W) and then activated in 15-second bursts followed by 15-second rest periods over a 330-minute period. After we did pilot testing to confirm that electrical interference during electrocautery would not likely interfere with wireless transmission, the wireless AIMS was used during 30 bedside ICU procedures between July 2012 and June 2014. During the first 5 cases, “shadow charting” was performed with paper anesthesia records. The subsequent cases were performed without shadow charting after preliminary review of the AIMS records showed satisfactory results (i.e., no data gaps). During cases, the distance from the patient to the AIMS workstation was 4 to 8 m (Fig. 2). We queried the AIMS database for Spo2, heart rate, respiratory rate, end-tidal CO2, blood pressure, and temperature data and the “Induction” and “Surgery End” timestamps from the 30 wireless cases. We also extracted data from 30
426 www.anesthesia-analgesia.org
similar cases that were performed in the main ORs with conventional data cables. The patient monitors used in our neonatal intensive care unit (NICU) are of the same brand and model as those used in the ORs. Data were imported into a spreadsheet (Excel; Microsoft) for analysis. Our AIMS records vital signs at 15-second intervals with each set of transmitted parameters recorded with the same timestamp (i.e., a “data row”). If transmission were uninterrupted during a given interval, then the expected number of data rows should equal the number of seconds in the interval divided by 15. For example, if a case’s “Induction” timestamp was 14:20:00 and its “Surgery End” timestamp was 16:57:00, then the duration would be 9420 seconds and the expected number of data rows would be 628. Because all the vital signs were transmitted wirelessly, a data gap was defined as any data row that was either null or contained only manually entered values. When a case’s row count was less than expected, we located the vital sign gap(s) in the spreadsheet and then examined the AIMS record using the AIMS case browser application.
RESULTS
During the 330-minute electrosurgical generator testing period, there were 0 data gaps during the n = 330, 1-minute d Valleylab Force FX Electrosurgical Generator Owner’s Manual. Available at: http://goo.gl/Vzntbk. Accessed August 14, 2015.
anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Wireless Connectivity from Patient Monitors to an AIMS
batches of 15-second test periods (95% binomial upper confidence limit = 1.11%). During patient monitoring, 3 wireless AIMS cases had 1 consecutive data gap each (Table 3), whereas in the remaining 27 cases, all the data were received. In 2 of the cases, the interruptions were brief (15 and 75 seconds) and the source of the interruption could not be determined. In the remaining case, there was a consecutive 38-minute data gap because of a faulty power cable connection to one of the paired Bluetooth
Figure 1. Wireless anesthesia information management system configuration. The wireless anesthesia information management system (AIMS) setup consists of 2 major components: 1 is the bedside monitor and the ventilator connected to the patient and the other is the mobile AIMS workstation with the data acquisition and control (DAC) module. The Solar 8000i GE bedside monitor is connected via cables to the patient through the GE Tram-Rac module. The Drager ventilator is connected to a GE Unity Interface Device that allows third-party devices to send data through Cat-5 cables to the bedside monitor. The RS-232 transmitter module is connected to port 1 on the back of the Solar monitor. The AIMS mobile cart consists of the host mobile computer that is connected to the DAC switch. The RS-232 wireless receiver connects to the DAC switch in a preconfigured data port. The bedside monitor and ventilator are connected wirelessly to the AIMS mobile cart by the RS-232 transmitter and receiver pair. One RS-232 pair is used in this setup.
transponders; after the power cable was reconnected, there were no data gaps for the remainder of the procedure. During the 30 surgical procedures that were documented in the main OR using serial data cables, there were no cases with any data gaps (Table 4). The cases consisted entirely of diaphragmatic hernia repairs, matching 24 of the 30 NICU cases.
DISCUSSION
Our analysis determined that a wireless AIMS can provide a simple, feasible solution to logistical problems associated with documenting AIMS records in the ICU, where data are collected locally. Implementation of the system in an ICU replaced paper documentation with an unobtrusive, portable AIMS with its expected potential benefits such as facilitating the retrieval of procedural and physiologic data.4–6 Setup of the wireless AIMS at the bedside was simple and unobtrusive, requiring approximately 10 minutes to connect the wireless adapters and start the AIMS workstation. The wireless AIMS recorded data without dropping data in the presence of an electrocautery device operated at high power with long bursts over a prolonged period of time. Although the introduction of wireless transmitters in a perioperative critical care setting may raise concern regarding electromagnetic interference with existing bedside equipment, the literature7,8 and our experience support the notion that wireless data transmission in the critical care setting is generally safe and reliable. Electrosurgical units typically produce radiofrequency output between 300 kHz and 1.2 MHz (typically 500 kHz),9 >3 orders of magnitude below the Bluetooth frequency. The 30 OR cases that used serial data cables had zero data gaps, which suggests that the NICU cases’ data gaps were likely because of interference with wireless transmission unrelated to electrocautery. In contrast, the Bluetooth frequency is 2.4 to 2.495 GHz, close to the frequency at which microwave ovens, cordless phones, Wi-Fi adapters, wireless baby monitors, and other wireless devices transmit.10,11
Figure 2. Neonatal intensive care unit layout during wireless cases. The layout of a typical neonatal intensive care unit (NICU) patient room is a small 3-walled space that opens to a larger central area that is bordered by 3 other patient rooms. When a surgical case occurs in the NICU, the surgical and nursing teams surround the patient’s open care system and the anesthesia team is near the patient’s medication pumps and ventilator. The patient’s vital sign monitor is mounted on a wall behind the patient’s neonatal open care system, whereas the wireless anesthesia information management system (AIMS) workstation is usually located in the corner of the patient’s room approximately 4 to 8 m away. The electrosurgical unit is often situated near the patient monitor and AIMS workstation, because a sterile instrument table typically blocks the opposite side of the room.
February 2016 • Volume 122 • Number 2
www.anesthesia-analgesia.org 427
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
E Technical Communication Table 3. Wireless Intensive Care Unit Data Gap Analysis Results Case no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Procedure Diaphragmatic hernia repair Abdominal wound closure Thoracotomy and pneumonectomy Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Laparotomy, colostomy, bowel resection Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Re-exploration of abdomen Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Central line placement Thoracotomy, lung biopsy Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair
Case duration (min) 146 69 157 104 97 141 156 156 160 229 170 212 101 134 144 144 150 98 77 131 54 136 74 86 82 51 177 80 100 122
Table 4. Data Gap Analysis Results for Operating Room Cases Using Conventional Data Cables Case no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Procedure Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair
Case duration (min) 129 393 220 102 255 234 93 100 236 99 142 220 372 98 83 139 193 230 161 221 79 235 207 134 79 103 288 199 143 116
428 www.anesthesia-analgesia.org
Number of data gaps 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Number of consecutive data gaps 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Total data gap duration (min:s) 0 0 0 0 0 1:15 0 0 0 0 0 0:15 0 0 0 38:00 0 0 0 0 0 0 0 0 0 0 0 0 0 0
We attribute the 2 cases with brief data gaps to a transmission or data processing issue, not an artifact, because it is extremely unlikely that all the vital sign sensors (e.g., capnograph, pulse oximeter) failed simultaneously. However, the electrocautery unit did not appear to be the source of the interference given the results of the simulation. The 38-minute data gap was because of a faulty power cable connection resulting in a loss of power to the wireless transmitter. When the anesthesia team noticed after a few minutes that data were not being transmitted to the AIMS, they switched promptly to charting on a paper record while technical support personnel investigated the issue. The incident highlights one limitation of this system: each wireless adapter requires a power supply, and the power cable connections are a potential source of failure that must be considered when deciding to use such devices. Our system is designed for configurations where patient monitors are connected directly to the AIMS workstation rather than to central servers via the internal hospital network. The wireless system we describe may be useful in situations other than in ICUs where data are traditionally collected via a serial cable. Potential applications include capturing vital sign data electronically for laboratory and research applications where mobile workstations are used and where ad hoc connections to physiologic monitors need to be made. Even where data are collected via a network, the wireless approach that we describe can enable receiving and storing data locally at the monitors’ frequency of transmission (e.g., every 15 seconds) rather than the limit imposed by the software application (at most every 1 minute in some systems).
anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Wireless Connectivity from Patient Monitors to an AIMS
To enhance network security, the Bluetooth devices were turned on and paired before the procedure and turned off at the completion of the procedure in accordance with the U.S. National Institute of Standards and Technology’s best practices for Bluetooth security.e Bluetooth links use authentication and encryption algorithms that are widely considered acceptably strong when both implemented and used correctly.f,g The strength of Bluetooth security relies primarily on the length and randomness of the passkey used for Bluetooth pairing.g In our system, we used wireless adapters with user-defined passkey functionality and used long, randomly generated passkeys. Furthermore, our system transmits only vital sign data from the monitor without any protected health information. Our study had several limitations. First, only 1 brand and type of wireless adapter was studied; the reliability of other adapters may vary. Second, our ESU testing consisted of only one ESU unit that has 3 output sinusoidal waveform settings: 240, 390, and 470 kHz.h ESU units that operate at other frequencies might affect wireless transmission to a greater degree. Third, only 1 AIMS was used in our study, and protocol drivers from other AIMS vendors may produce different results. Fourth, because our system consisted of vital sign transmission to an AIMS running on a full-featured, thick client, workstation computer, our findings may not be generalizable to those using thin client or virtual machine AIMS configurations.12 Last, only 1 model of AIMS workstation was studied (e.g., units with poorer insulation might have performed worse). Because of these limitations, we recommend that facilities considering the use of wireless transmission for clinical purposes verify that their combination of devices provides reliable data capture. Although the wireless AIMS provides convenience, it is not as robust as the wired connections in the main OR, and data loss may occur. For this reason, when practical, we recommend the use of wired connections. Anesthesiologists and other users of this system should be aware of this potential issue during the use of wireless transmission and be prepared to manually enter missing values in the event of a data gap, because failure to recognize the loss of incoming data can increase medical liability exposure.13 E DISCLOSURES
Name: Allan F. Simpao, MD. Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. Attestation: Allan F. Simpao approved the final manuscript and reviewed the original study data and data analysis. Allan F. Simpao attests to the integrity of the original data and the analysis reported in this manuscript. Allan F. Simpao is the archival author. Name: Jorge A. Galvez, MD. Contribution: This author helped design the study, conduct the study, and write the manuscript. e Scarfone K, Padgette J, Chen L. Guide to Bluetooth security. NIST Special Publication, 121. Available at: http://www.mcs.csueastbay.edu/~lertaul/ BluetoothSECV1.pdf. Accessed August 14, 2015. f Bluetooth Enabled Medical Devices: A Growing Market. Available at: http://www.bluetooth.com/Pages/Medical-Devices-2.aspx. Accessed August 14, 2015. g Bluetooth Security. Available at: https://www.nsa.gov/ia/_files/factsheets/i732-016r-07.pdf. Accessed August 14, 2015. h Valleylab Force FX Electrosurgical Generator Technical Specifications. Available at: http://goo.gl/jdWI5p. Accessed August 14, 2015.
February 2016 • Volume 122 • Number 2
Attestation: Jorge A. Galvez approved the final manuscript. Name: W. Randall England, BA. Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. Attestation: W. Randall England approved the final manuscript. Name: Elicia C. Wartman, BA. Contribution: This author helped conduct the study, analyze the data, and write the manuscript. Attestation: Elicia C. Wartman approved the final manuscript. Name: James H. Scott, CNA. Contribution: This author helped design the study, conduct the study, and write the manuscript. Attestation: James H. Scott approved the final manuscript. Name: Michael M. Hamid, Sr., CE. Contribution: This author helped design the study, conduct the study, and write the manuscript. Attestation: Michael M. Hamid, Sr., approved the final manuscript. Name: Mohamed A. Rehman, MD. Contribution: This author helped design the study, conduct the study, and write the manuscript. Attestation: Mohamed A. Rehman approved the final manuscript. Mohamed A. Rehman attests to the integrity of the original data and the analysis reported in this manuscript. Name: Richard H. Epstein, MD. Contribution: This author helped design the study, analyze the data, and write the manuscript. Attestation: Richard H. Epstein approved the final manuscript. This manuscript was handled by: Maxime Cannesson, MD, PhD. REFERENCES 1. Paksuniemi M, Sorvoja H, Alasaarela E, Myllyla R. Wireless sensor and data transmission needs and technologies for patient monitoring in the operating room and intensive care unit. Conf Proc IEEE Eng Med Biol Soc 2005;5:5182–5 2. Wallin MK, Wajntraub S. Evaluation of Bluetooth as a replacement for cables in intensive care and surgery. Anesth Analg 2004;98:763–7 3. Øyri K, Balasingham I, Samset E, Høgetveit JO, Fosse E. Wireless continuous arterial blood pressure monitoring during surgery: a pilot study. Anesth Analg 2006;102:478–83 4. Balust J, Macario A. Can anesthesia information management systems improve quality in the surgical suite? Curr Opin Anaesthesiol 2009;22:215–22 5. Shah NJ, Tremper KK, Kheterpal S. Anatomy of an anesthesia information management system. Anesthesiol Clin 2011;29:355–65 6. Stabile M, Cooper L. Review article: the evolving role of information technology in perioperative patient safety. Can J Anaesth 2013;60:119–26 7. Lapinsky SE, Easty AC. Electromagnetic interference in critical care. J Crit Care 2006;21:267–70 8. Boyle J. Wireless technologies and patient safety in hospitals. Telemed J E Health 2006;12:373–82 9. Munro MG. Fundamentals of electrosurgery Part I: principles of radiofrequency energy for surgery. In: Feldman L, Fuchshuber P, Jones DB, eds. The SAGES Manual on the Fundamental Use of Surgical Energy (FUSE). New York, NY: Springer Publishing, 2012:19, 26–27 10. The United States of America National Institute for Occupational Safety and Health. EMF basics. In: Questions and Answers: EMF in the Workplace. Darby, PA: Diane Publishing, 1996:8 11. Dodd AZ. Spectrum allocation. In: The Essential Guide to Telecommunications. Upper Saddle River, NJ: Prentice Hall Publishing, 2002:377–8 12. Nikolik D. Network architecture. In: A Manager’s Primer on e-Networking: An Introduction to Enterprise Networking. New York, NY: Springer Publishing, 2003:41–49 13. Vigoda MM, Lubarsky DA. Failure to recognize loss of incoming data in an anesthesia record-keeping system may have increased medical liability. Anesth Analg 2006;102:1798–802
www.anesthesia-analgesia.org 429
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
E Technical Communication
A Technical Evaluation of Wireless Connectivity from Patient Monitors to an Anesthesia Information Management System During Intensive Care Unit Surgery Allan F. Simpao, MD,*† Jorge A. Galvez, MD,*† W. Randall England, BA,† Elicia C. Wartman, BA,† James H. Scott, CNA,† Michael M. Hamid, Sr., CE,† Mohamed A. Rehman, MD,*† and Richard H. Epstein, MD‡ Surgical procedures performed at the bedside in the neonatal intensive care unit (NICU) at The Children’s Hospital of Philadelphia were documented using paper anesthesia records in contrast to the operating rooms, where an anesthesia information management system (AIMS) was used for all cases. This was largely because of logistical problems related to connecting cables between the bedside monitors and our portable AIMS workstations. We implemented an AIMS for documentation in the NICU using wireless adapters to transmit data from bedside monitoring equipment to a portable AIMS workstation. Testing of the wireless AIMS during simulation in the presence of an electrosurgical generator showed no evidence of interference with data transmission. Thirty NICU surgical procedures were documented via the wireless AIMS. Two wireless cases exhibited brief periods of data loss; one case had an extended data gap because of adapter power failure. In comparison, in a control group of 30 surgical cases in which wired connections were used, there were no data gaps. The wireless AIMS provided a simple, unobtrusive, portable alternative to paper records for documenting anesthesia records during NICU bedside procedures. (Anesth Analg 2016;122:425–9)
T
he use of an anesthesia information management system (AIMS) during bedside procedures in the intensive care unit (ICU) can be challenging when monitoring equipment needs to be cabled directly to the AIMS workstation. Because of logistical considerations, the workstation must be situated at some distance from the patient, creating potential tripping hazards because of the cables and the possible need for duplicate monitoring equipment and sensors attached to the patient.1 A wireless connection, if sufficiently reliable, would eliminate the need for such cabling and thus reduce the barrier to the use of the AIMS in the ICU and in other locations where space is at a premium. We established the need for such a solution through a paper survey distributed to 10 anesthesiologists from 10 pediatric hospitals who attended the Biomedical Informatics Special Interest Group meeting held at the Society for Pediatric Anesthesia’s International Assembly for Pediatric Anesthesia in Washington, DC, on October 11, 2012 (Table 1). Ten surveys from separate institutions were returned with 9 indicating that they would use their AIMS for all ICU patients
From the *Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania; †The Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania; and ‡Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, Pennsylvania. Accepted for publication August 21, 2015. Funding: Departmental. The authors declare no conflicts of interest.
if a wireless solution were available and 4 indicating that they currently use their AIMS during bedside cases (Table 2). The Bluetooth wireless transmission protocol has been shown under laboratory conditions2 and in a small (n = 4 patient) operating room (OR) pilot study3 to be a potentially suitable replacement for transmission of data via serial data cables in the ICU. Here, we describe the implementation, technical assessment, and performance testing during 30 bedside ICU procedures using inexpensive, readily available Bluetooth wireless transponders.
METHODS
This study did not meet the definition of human subjects research for the IRB at The Children’s Hospital of Philadelphia, and the requirement for written informed consent was waived. We used Bluetooth wireless serial adapters (HdwBTRS232; WCSC Inc., Houston, TX) to transmit vital sign data (electrocardiogram, heart rate, respiratory rate, noninvasive blood pressure, Spo2, ventilator parameters, etc.) from the bedside monitor (Solar 8000i; GE Healthcare, UK)a to a portable computer thick client workstation (Lenovo M92 Tiny, Morrisville, NC; Microsoft Windows 7, Redmond, WA) running our AIMS (CompuRecord; Philips, Andover, MA). The wireless adapters were paired and configured using a PC HyperTerminal application (Hilgraeve, Monroe, MI): 19,200 bits per second, 8 data bits, no parity bit, 1 stop bit, and hardware flow control.b One wireless adapter was connected to the patient monitor’s RS-232
Reprints will not be available from the authors. Address correspondence to Allan F. Simpao, MD, Perelman School of Medicine, University of Pennsylvania, The Children’s Hospital of Philadelphia, 3401 Civic Center Blvd., Philadelphia, PA 19104. Address e-mail to simpaoa@ email.chop.edu. Copyright © 2016 International Anesthesia Research Society DOI: 10.1213/ANE.0000000000001064
February 2016 • Volume 122 • Number 2
GE Healthcare Solar 8000i Patient Monitor Manual. Available at: http://goo. gl/tvjT58. Accessed August 14, 2015. b Bluetooth Wireless Serial Port Adapter Support (HdwBTRS232). Available at: http://goo.gl/puOTHl. Accessed August 14, 2015. c Baytech M Series Data Acquisition and Control Owner’s Manual. Available at: http://goo.gl/eawYfH. Accessed August 14, 2015. a
www.anesthesia-analgesia.org
425
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
E Technical Communication Table 1. Wireless Intensive Care Unit Anesthesia Information Management System Needs Assessment Survey Questions 1. Does your institution have an anesthesia information management system (AIMS) currently installed (e.g., CompuRecord, Innovian, DocuSys)? Yes No 2. At your institution, approximately how many anesthesia-monitored cases are performed at the intensive care unit (ICU) bedside per year? ______ ICU bedside cases monitored by anesthesia providers per year 3. If your institution has an AIMS installed, please answer the following 3 questions: a. Are patient monitors connected directly to the AIMS workstation or to a central database? Direct to computer Central database via the network Don’t know b. What percentage of your institution’s bedside ICU cases are recorded in the AIMS? 0% 100% ____% (fill in the percentage) c. If there were a simple wireless connection that would allow connecting remotely the patient monitor to a mobile computer without the need to attach cables or provide duplicate monitors, would you use your AIMS in these locations for all patients? Yes No
Table 2. Pediatric Anesthesiologists’ Responses to Wireless Intensive Care Unit Anesthesia Information Management System Needs Assessment Survey Respondent 1 2 3 4 5 6 7 8 9 10
AIMS in use
No. of yearly bedside NICU cases
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
6 Unknown 50 5 0 40 10 500 10 250
Patient monitors connection Computer serial port Network Unknown X
No. of bedside NICU cases recorded in AIMS (%)
Would use wireless NICU AIMS if available
100 N/A 0 100 N/A 0 100 100 0 0
Yes Yes Yes Yes N/A Yes Yes Yes Yes Yes
X X
X X X X X X X X
AIMS = anesthesia information management system; NICU = neonatal intensive care unit; N/A = not applicable.
serial port, and a second paired wireless adapter was connected to the AIMS workstation through a data acquisition and control unit (M4 Series; BayTech, Long Beach, MS)c (Fig. 1). To test for interference, we activated an electrosurgical generator (Force FX; Valleylab, Boulder, CO)d located between the GE Solar and the AIMS workstation during data acquisition using vital sign simulator modules. The distance between the patient monitor and the AIMS workstation was 4 m. The electrosurgical generator was set at the maximal monopolar coagulation power setting (120 W) and then activated in 15-second bursts followed by 15-second rest periods over a 330-minute period. After we did pilot testing to confirm that electrical interference during electrocautery would not likely interfere with wireless transmission, the wireless AIMS was used during 30 bedside ICU procedures between July 2012 and June 2014. During the first 5 cases, “shadow charting” was performed with paper anesthesia records. The subsequent cases were performed without shadow charting after preliminary review of the AIMS records showed satisfactory results (i.e., no data gaps). During cases, the distance from the patient to the AIMS workstation was 4 to 8 m (Fig. 2). We queried the AIMS database for Spo2, heart rate, respiratory rate, end-tidal CO2, blood pressure, and temperature data and the “Induction” and “Surgery End” timestamps from the 30 wireless cases. We also extracted data from 30
426 www.anesthesia-analgesia.org
similar cases that were performed in the main ORs with conventional data cables. The patient monitors used in our neonatal intensive care unit (NICU) are of the same brand and model as those used in the ORs. Data were imported into a spreadsheet (Excel; Microsoft) for analysis. Our AIMS records vital signs at 15-second intervals with each set of transmitted parameters recorded with the same timestamp (i.e., a “data row”). If transmission were uninterrupted during a given interval, then the expected number of data rows should equal the number of seconds in the interval divided by 15. For example, if a case’s “Induction” timestamp was 14:20:00 and its “Surgery End” timestamp was 16:57:00, then the duration would be 9420 seconds and the expected number of data rows would be 628. Because all the vital signs were transmitted wirelessly, a data gap was defined as any data row that was either null or contained only manually entered values. When a case’s row count was less than expected, we located the vital sign gap(s) in the spreadsheet and then examined the AIMS record using the AIMS case browser application.
RESULTS
During the 330-minute electrosurgical generator testing period, there were 0 data gaps during the n = 330, 1-minute d Valleylab Force FX Electrosurgical Generator Owner’s Manual. Available at: http://goo.gl/Vzntbk. Accessed August 14, 2015.
anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Wireless Connectivity from Patient Monitors to an AIMS
batches of 15-second test periods (95% binomial upper confidence limit = 1.11%). During patient monitoring, 3 wireless AIMS cases had 1 consecutive data gap each (Table 3), whereas in the remaining 27 cases, all the data were received. In 2 of the cases, the interruptions were brief (15 and 75 seconds) and the source of the interruption could not be determined. In the remaining case, there was a consecutive 38-minute data gap because of a faulty power cable connection to one of the paired Bluetooth
Figure 1. Wireless anesthesia information management system configuration. The wireless anesthesia information management system (AIMS) setup consists of 2 major components: 1 is the bedside monitor and the ventilator connected to the patient and the other is the mobile AIMS workstation with the data acquisition and control (DAC) module. The Solar 8000i GE bedside monitor is connected via cables to the patient through the GE Tram-Rac module. The Drager ventilator is connected to a GE Unity Interface Device that allows third-party devices to send data through Cat-5 cables to the bedside monitor. The RS-232 transmitter module is connected to port 1 on the back of the Solar monitor. The AIMS mobile cart consists of the host mobile computer that is connected to the DAC switch. The RS-232 wireless receiver connects to the DAC switch in a preconfigured data port. The bedside monitor and ventilator are connected wirelessly to the AIMS mobile cart by the RS-232 transmitter and receiver pair. One RS-232 pair is used in this setup.
transponders; after the power cable was reconnected, there were no data gaps for the remainder of the procedure. During the 30 surgical procedures that were documented in the main OR using serial data cables, there were no cases with any data gaps (Table 4). The cases consisted entirely of diaphragmatic hernia repairs, matching 24 of the 30 NICU cases.
DISCUSSION
Our analysis determined that a wireless AIMS can provide a simple, feasible solution to logistical problems associated with documenting AIMS records in the ICU, where data are collected locally. Implementation of the system in an ICU replaced paper documentation with an unobtrusive, portable AIMS with its expected potential benefits such as facilitating the retrieval of procedural and physiologic data.4–6 Setup of the wireless AIMS at the bedside was simple and unobtrusive, requiring approximately 10 minutes to connect the wireless adapters and start the AIMS workstation. The wireless AIMS recorded data without dropping data in the presence of an electrocautery device operated at high power with long bursts over a prolonged period of time. Although the introduction of wireless transmitters in a perioperative critical care setting may raise concern regarding electromagnetic interference with existing bedside equipment, the literature7,8 and our experience support the notion that wireless data transmission in the critical care setting is generally safe and reliable. Electrosurgical units typically produce radiofrequency output between 300 kHz and 1.2 MHz (typically 500 kHz),9 >3 orders of magnitude below the Bluetooth frequency. The 30 OR cases that used serial data cables had zero data gaps, which suggests that the NICU cases’ data gaps were likely because of interference with wireless transmission unrelated to electrocautery. In contrast, the Bluetooth frequency is 2.4 to 2.495 GHz, close to the frequency at which microwave ovens, cordless phones, Wi-Fi adapters, wireless baby monitors, and other wireless devices transmit.10,11
Figure 2. Neonatal intensive care unit layout during wireless cases. The layout of a typical neonatal intensive care unit (NICU) patient room is a small 3-walled space that opens to a larger central area that is bordered by 3 other patient rooms. When a surgical case occurs in the NICU, the surgical and nursing teams surround the patient’s open care system and the anesthesia team is near the patient’s medication pumps and ventilator. The patient’s vital sign monitor is mounted on a wall behind the patient’s neonatal open care system, whereas the wireless anesthesia information management system (AIMS) workstation is usually located in the corner of the patient’s room approximately 4 to 8 m away. The electrosurgical unit is often situated near the patient monitor and AIMS workstation, because a sterile instrument table typically blocks the opposite side of the room.
February 2016 • Volume 122 • Number 2
www.anesthesia-analgesia.org 427
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
E Technical Communication Table 3. Wireless Intensive Care Unit Data Gap Analysis Results Case no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Procedure Diaphragmatic hernia repair Abdominal wound closure Thoracotomy and pneumonectomy Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Laparotomy, colostomy, bowel resection Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Re-exploration of abdomen Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Central line placement Thoracotomy, lung biopsy Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair
Case duration (min) 146 69 157 104 97 141 156 156 160 229 170 212 101 134 144 144 150 98 77 131 54 136 74 86 82 51 177 80 100 122
Table 4. Data Gap Analysis Results for Operating Room Cases Using Conventional Data Cables Case no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Procedure Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair Diaphragmatic hernia repair
Case duration (min) 129 393 220 102 255 234 93 100 236 99 142 220 372 98 83 139 193 230 161 221 79 235 207 134 79 103 288 199 143 116
428 www.anesthesia-analgesia.org
Number of data gaps 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Number of consecutive data gaps 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Total data gap duration (min:s) 0 0 0 0 0 1:15 0 0 0 0 0 0:15 0 0 0 38:00 0 0 0 0 0 0 0 0 0 0 0 0 0 0
We attribute the 2 cases with brief data gaps to a transmission or data processing issue, not an artifact, because it is extremely unlikely that all the vital sign sensors (e.g., capnograph, pulse oximeter) failed simultaneously. However, the electrocautery unit did not appear to be the source of the interference given the results of the simulation. The 38-minute data gap was because of a faulty power cable connection resulting in a loss of power to the wireless transmitter. When the anesthesia team noticed after a few minutes that data were not being transmitted to the AIMS, they switched promptly to charting on a paper record while technical support personnel investigated the issue. The incident highlights one limitation of this system: each wireless adapter requires a power supply, and the power cable connections are a potential source of failure that must be considered when deciding to use such devices. Our system is designed for configurations where patient monitors are connected directly to the AIMS workstation rather than to central servers via the internal hospital network. The wireless system we describe may be useful in situations other than in ICUs where data are traditionally collected via a serial cable. Potential applications include capturing vital sign data electronically for laboratory and research applications where mobile workstations are used and where ad hoc connections to physiologic monitors need to be made. Even where data are collected via a network, the wireless approach that we describe can enable receiving and storing data locally at the monitors’ frequency of transmission (e.g., every 15 seconds) rather than the limit imposed by the software application (at most every 1 minute in some systems).
anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Wireless Connectivity from Patient Monitors to an AIMS
To enhance network security, the Bluetooth devices were turned on and paired before the procedure and turned off at the completion of the procedure in accordance with the U.S. National Institute of Standards and Technology’s best practices for Bluetooth security.e Bluetooth links use authentication and encryption algorithms that are widely considered acceptably strong when both implemented and used correctly.f,g The strength of Bluetooth security relies primarily on the length and randomness of the passkey used for Bluetooth pairing.g In our system, we used wireless adapters with user-defined passkey functionality and used long, randomly generated passkeys. Furthermore, our system transmits only vital sign data from the monitor without any protected health information. Our study had several limitations. First, only 1 brand and type of wireless adapter was studied; the reliability of other adapters may vary. Second, our ESU testing consisted of only one ESU unit that has 3 output sinusoidal waveform settings: 240, 390, and 470 kHz.h ESU units that operate at other frequencies might affect wireless transmission to a greater degree. Third, only 1 AIMS was used in our study, and protocol drivers from other AIMS vendors may produce different results. Fourth, because our system consisted of vital sign transmission to an AIMS running on a full-featured, thick client, workstation computer, our findings may not be generalizable to those using thin client or virtual machine AIMS configurations.12 Last, only 1 model of AIMS workstation was studied (e.g., units with poorer insulation might have performed worse). Because of these limitations, we recommend that facilities considering the use of wireless transmission for clinical purposes verify that their combination of devices provides reliable data capture. Although the wireless AIMS provides convenience, it is not as robust as the wired connections in the main OR, and data loss may occur. For this reason, when practical, we recommend the use of wired connections. Anesthesiologists and other users of this system should be aware of this potential issue during the use of wireless transmission and be prepared to manually enter missing values in the event of a data gap, because failure to recognize the loss of incoming data can increase medical liability exposure.13 E DISCLOSURES
Name: Allan F. Simpao, MD. Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. Attestation: Allan F. Simpao approved the final manuscript and reviewed the original study data and data analysis. Allan F. Simpao attests to the integrity of the original data and the analysis reported in this manuscript. Allan F. Simpao is the archival author. Name: Jorge A. Galvez, MD. Contribution: This author helped design the study, conduct the study, and write the manuscript. e Scarfone K, Padgette J, Chen L. Guide to Bluetooth security. NIST Special Publication, 121. Available at: http://www.mcs.csueastbay.edu/~lertaul/ BluetoothSECV1.pdf. Accessed August 14, 2015. f Bluetooth Enabled Medical Devices: A Growing Market. Available at: http://www.bluetooth.com/Pages/Medical-Devices-2.aspx. Accessed August 14, 2015. g Bluetooth Security. Available at: https://www.nsa.gov/ia/_files/factsheets/i732-016r-07.pdf. Accessed August 14, 2015. h Valleylab Force FX Electrosurgical Generator Technical Specifications. Available at: http://goo.gl/jdWI5p. Accessed August 14, 2015.
February 2016 • Volume 122 • Number 2
Attestation: Jorge A. Galvez approved the final manuscript. Name: W. Randall England, BA. Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript. Attestation: W. Randall England approved the final manuscript. Name: Elicia C. Wartman, BA. Contribution: This author helped conduct the study, analyze the data, and write the manuscript. Attestation: Elicia C. Wartman approved the final manuscript. Name: James H. Scott, CNA. Contribution: This author helped design the study, conduct the study, and write the manuscript. Attestation: James H. Scott approved the final manuscript. Name: Michael M. Hamid, Sr., CE. Contribution: This author helped design the study, conduct the study, and write the manuscript. Attestation: Michael M. Hamid, Sr., approved the final manuscript. Name: Mohamed A. Rehman, MD. Contribution: This author helped design the study, conduct the study, and write the manuscript. Attestation: Mohamed A. Rehman approved the final manuscript. Mohamed A. Rehman attests to the integrity of the original data and the analysis reported in this manuscript. Name: Richard H. Epstein, MD. Contribution: This author helped design the study, analyze the data, and write the manuscript. Attestation: Richard H. Epstein approved the final manuscript. This manuscript was handled by: Maxime Cannesson, MD, PhD. REFERENCES 1. Paksuniemi M, Sorvoja H, Alasaarela E, Myllyla R. Wireless sensor and data transmission needs and technologies for patient monitoring in the operating room and intensive care unit. Conf Proc IEEE Eng Med Biol Soc 2005;5:5182–5 2. Wallin MK, Wajntraub S. Evaluation of Bluetooth as a replacement for cables in intensive care and surgery. Anesth Analg 2004;98:763–7 3. Øyri K, Balasingham I, Samset E, Høgetveit JO, Fosse E. Wireless continuous arterial blood pressure monitoring during surgery: a pilot study. Anesth Analg 2006;102:478–83 4. Balust J, Macario A. Can anesthesia information management systems improve quality in the surgical suite? Curr Opin Anaesthesiol 2009;22:215–22 5. Shah NJ, Tremper KK, Kheterpal S. Anatomy of an anesthesia information management system. Anesthesiol Clin 2011;29:355–65 6. Stabile M, Cooper L. Review article: the evolving role of information technology in perioperative patient safety. Can J Anaesth 2013;60:119–26 7. Lapinsky SE, Easty AC. Electromagnetic interference in critical care. J Crit Care 2006;21:267–70 8. Boyle J. Wireless technologies and patient safety in hospitals. Telemed J E Health 2006;12:373–82 9. Munro MG. Fundamentals of electrosurgery Part I: principles of radiofrequency energy for surgery. In: Feldman L, Fuchshuber P, Jones DB, eds. The SAGES Manual on the Fundamental Use of Surgical Energy (FUSE). New York, NY: Springer Publishing, 2012:19, 26–27 10. The United States of America National Institute for Occupational Safety and Health. EMF basics. In: Questions and Answers: EMF in the Workplace. Darby, PA: Diane Publishing, 1996:8 11. Dodd AZ. Spectrum allocation. In: The Essential Guide to Telecommunications. Upper Saddle River, NJ: Prentice Hall Publishing, 2002:377–8 12. Nikolik D. Network architecture. In: A Manager’s Primer on e-Networking: An Introduction to Enterprise Networking. New York, NY: Springer Publishing, 2003:41–49 13. Vigoda MM, Lubarsky DA. Failure to recognize loss of incoming data in an anesthesia record-keeping system may have increased medical liability. Anesth Analg 2006;102:1798–802
www.anesthesia-analgesia.org 429
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
