Surgical Energy-Based Device Injuries and Fatalities Reported to the Food and Drug Administration

Surgical Energy-Based Device Injuries and Fatalities Reported to the Food and Drug Administration Douglas M Overbey, MD, Nicole T Townsend, MD, Brandon C Chapman, MD, Daine T Bennett, MD, Lisa S Foley, MD, Aline S Rau, MD, Jeniann A Yi, MD, Edward L Jones, MD, Greg V Stiegmann, MD, FACS, Thomas N Robinson, MD, MS, FACS Energy-based devices are used in virtually every operation. Our purposes were to describe causes of energy-based device complications leading to injury or death, and to determine if common mechanisms leading to injury or death can be identified. STUDY DESIGN: The FDA’s Manufacturer and User Facility Device Experience (MAUDE) database was searched for surgical energy-based device injuries and deaths reported over 20 years (January 1994 to December 2013). Device-related complications were recorded and analyzed. RESULTS: We analyzed 178 deaths and 3,553 injuries. Common patterns of complications were: thermal burns, 63% (n ¼ 2,353); hemorrhage, 17% (n ¼ 642); mechanical failure of device, 12% (n ¼ 442); and fire, 8% (n ¼ 294). Events were identified intraoperatively in 82% (3,056), inpatient postoperatively in 9% (n ¼ 351), and after discharge in 9% (n ¼ 324). Of the deaths, 12% (n ¼ 22) occurred after discharge home. Common mechanisms for thermal burn injuries were: direct application, 30% (n ¼ 694); dispersive electrode burn, 29% (n ¼ 657); and insulation failure, 14% (n ¼ 324). Thermal injury was the most common reason for death (39%, n ¼ 70). The mechanism for these thermal injuries was most frequently direct application (84%, n ¼ 59, p < 0.001 vs all other mechanisms). Fires were most common with monopolar “Bovie” instruments (88%, n ¼ 258, p < 0.001 vs all other devices) when they were used in head and neck operations (66%, n ¼ 193, p < 0.001 vs all other locations). CONCLUSIONS: Complications due to energy-based devices occur from 4 main causes: thermal burn, hemorrhage, mechanical failure, and fire. Thermal direct application injuries are the most common reason for both injury and death. (J Am Coll Surg 2015;221:197e206.
2015 by the American College of Surgeons)

BACKGROUND:

Energy-based devices are used in virtually every operation in modern medicine. Commonly used types of energy-based

devices include the monoplar “Bovie” instrument, traditional biplor instruments, plasma beam monopolar devices, advanced bipolar devices, ultrasonic devices, and radiofrequency/microwave ablation devices. Despite nearly ubiqutious use, surgeons have a relative lack of understanding of the mechanisms by which surgical energy-based devices cause injuries and deaths.1 Energy-based device complications are vital to understand because they represent an enormous source of preventable injuries and deaths from a public health perspective. The rate of energy-based device complications in an individual surgeon’s practice is low (best estimates have found that surgical energy injuries occur in 1 to 2 per 1,000 operations).2 However, from a population standpoint, with more than 50 million inpatient procedures performed annually in the United States, energybased device complications are an important target for

Disclosure Information: Nothing to disclose. Disclosures outside the scope of this work: Dr Robinson’s institution received grant money from Medtronic and Covidien. Dr Robinson is the principal investigator on research grants from Medtronic and Covidien. Presented at the Western Surgical Association 122nd Scientific Session, Indian Wells, CA, November 2014. Received January 9, 2015; Revised February 19, 2015; Accepted March 3, 2015. From the Departments of Surgery, University of Colorado School of Medicine, Aurora, CO (Overbey, Townsend, Chapman, Bennett, Foley, Rau, Yi, Stiegmann, Robinson) and Ohio State University, Columbus, OH (Jones). Correspondence address: Thomas N Robinson, MD, MS, FACS, 12631 East 17th Ave, MS C313, Aurora, CO 80045. email: thomas.robinson@ ucdenver.edu

ª 2015 by the American College of Surgeons Published by Elsevier Inc.

197

http://dx.doi.org/10.1016/j.jamcollsurg.2015.03.031 ISSN 1072-7515/15

198

Overbey et al

Energy-Based Device Injuries and Deaths

education and research efforts to reduce surgical complications. Understanding the patterns by which surgical energy complications occur can alter surgeon practice to avoid these injuries. To our knowledge, no large series exists describing energy-based device complications and their patterns of patient injury. The specific aims of this study were to examine the patterns of all injuries and deaths related to electrosurgical devices reported to the FDA, and to describe the common mechanisms leading to patient injury or death. Our purposes were to describe causes of energy-based device complications leading to injury or death; to determine if common mechanisms leading to injury or death can be identified; and to delineate which mechanisms are most commonly reported per device.

METHODS Database The US Food and Drug Administration’s (FDA) Manufacturer and User Facility Device Experience (MAUDE) database archives medical device reports of suspected medical device-associated malfunctions, injuries, and deaths.3 This publicly available database includes records from mandatory reporters (manufacturers and device using facilities) and voluntary reporters (health care professionals, patients, and consumers). Institutional review board approval was waived because this database is publicly available and contains no patient identifiers. Search strategy Consecutive records were reviewed over 20 years, between January 1, 1994 and December 31, 2013. Reports from the MAUDE database were downloaded for injury and death by searching product codes GEI, LFL, JOS, NUJ, MUK, PDG, JOT, ONQ, DWG, and ODR to encompass all relevant energy-based device codes (Product regulation: Electrosurgical cutting and coagulation device and accessories, Appendix 1, online only). Each record was individually examined by a trained researcher. Of these reports, 7,717 were coded as injury and 238 as deaths.

J Am Coll Surg

4. Pattern of injury recorded information was different based on the type of complication (eg, thermal burn, hemorrhage, etc). For thermal burns, the pattern of injury included direct application (defined as thermal burn from direct extension of heat during activation of instrument), inadvertent activation (defined as unintended activation of the energy device), grounding failure (thermal injury at dispersive electrode site), insulation failure (a defect of the insulation sheath on the active electrode), capacitive coupling (thermal injury of tissue adjacent to the shaft of the active electrode when no insulation defect was noted), or residual heat (a burn caused by tip of energy-based device from retained heat after device activation completed). For hemorrhage, data recorded included arterial bleed, venous bleed, or bleeding source unknown. For fires, the fuel source recorded was oxygen, drapes, or unknown. 5. Patient injury: for thermal burns, the injuries were classified as visceral injury (damage to an organ within a body cavity), visceral perforation (mention of a hole in a hollow viscous), skin burn minor (burn limited within surgical field) or major (burn extended outside surgical field). For hemorrhage, bleeding was defined as minor (no transfusion), major (requiring transfusion), or bleeding causing death. For fires, minor fires were limited to those within the surgical field and major fires extended outside the surgical field. 6. Timing of complication was recorded as intraoperative, postoperative during inpatient stay, or postoperative after discharge. 7. Severity of injury was defined by the previously described Accordion Severity Classification of Postoperative complications (1, mild; 2, moderate; 3, severe; 4, death).4 8. The anatomic location of the operation was recorded.

Data extraction We developed a taxonomy of description to codify events. Information extracted from each record included:

Exclusion criteria Reports were excluded if there was an inadequate description to categorize the energy-based, device-related adverse event, including 2 or more of the categories listed previously that could not be found from the record, or the injury was explicitly described as not device related. Duplicated reports (identical date and event description) were also excluded.

1. Date report received. 2. Category of device (monopolar, traditional bipolar, advanced bipolar, ultrasonic, plasma energy, ablation, or other). 3. Type of complication: thermal burn, hemorrhage, fire, mechanical failure or other.

Interobserver reliability To ensure that all researchers collected data recorded similar information, all individuals coding the reports received structured training in terminology and coding practices from the principal investigator. A pilot study of

Vol. 221, No. 1, July 2015

Overbey et al

Energy-Based Device Injuries and Deaths

199

Figure 1. Event report inclusion flow diagram. See Appendix 1, online only, for code descriptions.

Figure 2. Number of energy-based device reports involving injury or death per year from January 1994 to December 2013.

recording the results from the same 50 records was completed by each of the 8 researchers who recorded data and compared with the coding performed by the principal investigator. All data extractionists were surgery residents PGY 2 or higher. High inter-rater reliability was assured with a kappa coefficient > 0.80 for all 8 researchers, a level of agreement that is considered exceptional.5

Overall, monopolar instrument complications were most commonly reported (n ¼ 1,670, 45%), followed by ablation devices (n ¼ 728, 20%), and tissue sealing devices, divided into advanced bipolar (n ¼ 538, 14%) and ultrasonic energy (n ¼ 450, 9%).

Statistical analysis Descriptive statistics were generated for all variables. Nominal variables were expressed as proportions and percentages. Between-group differences for proportions were tested using 2-tailed chi-square analysis. All statistical tests were conducted at the 5% significance level. All data management and statistical analysis were performed using SPSS (IBM Corp. Released 2012. IBM SPSS Statistics for Windows, Version 21.0.).

RESULTS Report demographics A total of 178 deaths and 3,553 injury reports were included (Fig. 1). The high rate of exclusion is a function of the diversity of reporters and open-ended descriptions provided. Over the 20 years included, annual reports for included product codes increased from a low of 24 in 1995 to a high of 991 in 2012. The annual breakdown of these energy-based device records is reported in Figure 2. The increase in reports per year is likely multifactorial, resulting from both increasing use of electrosurgical devices as well as growing popularity of the database.

Combined injury and death report summary As shown in Table 1, cumulative patterns of complications were: thermal burns, 63%; hemorrhage, 17%; mechanical failure, 12%; and fire, 8%. Severe injury or death (Grade 3 or 4 Accordion classified complication) accounted for 47% of all reports. Of all events reported, 18% were recognized after conclusion of the operation. Thermal burns Table 2 shows common mechanisms for thermal burn injuries. Direct application (32%), grounding failure (29%), and insulation failure (14%) were the 3 most common mechanisms. Although dispersive electrode failures and insulation failures account for 43% of burn injuries, these rarely translate into deaths. Deaths due to thermal injury are most commonly via direct application (84%, n ¼ 59, p < 0.001 vs all other mechanisms). Although monopolar devices most commonly cause thermal burns, death due to thermal injury is most frequently reported from an ablation device. Hemorrhage As shown in Table 3, bleeding was most frequently reported with the advanced bipolar devices (47%).

200

Table 1.

Overbey et al

J Am Coll Surg

Energy-Based Device Injuries and Deaths

Cumulative Injuries and Deaths

Variable

Injury pattern Fire Thermal burn Hemorrhage Mechanical failure Severity of injury Minor (Accordion 1) Moderate (Accordion 2) Severe (Accordion 3) Death (Accordion 4) Type of energy device Monopolar “Bovie” Traditional bipolar Advanced bipolar Ultrasonic Ablation Argon beam coagulator Other Timing of event Intraoperative Postoperative inpatient Postoperative at home

Injuries (n ¼ 3,553) n %

Deaths (n ¼ 178) n %

Combined (n ¼ 3,731) n %

279 2,283 579 412

8 64 16 12

15 70 63 30

8 39 35 17

294 2,353 642 442

8 63 17 12

1,589 398 1,566 0

45 11 44 0

0 0 0 178

0 0 0 100

1,589 398 1566 178

43 11 42 5

1,632 267 508 328 654 152 12

46 8 14 9 18 4 0

38 3 30 22 74 11 0

21 2 17 12 42 6 0

1,670 270 538 350 728 163 12

45 7 14 9 20 4 0

2976 275 302

84 8 8

80 76 22

45 43 12

3056 351 324

82 9 9

However, ablation devices were again more commonly reported with fatality (40%) (p < 0.001 compared with advanced bipolar). Bleeding events leading to death were most common in abdominal operations (60%) followed by noncardiac thoracic operations (19%). Only 27% of the time was the source of bleeding specifically described as arterial or venous. Bleeding events required transfusion in 50% of events. Forty percent of hemorrhage complications were postoperative, with 10% occuring after discharge.

Fires Table 4 shows that fires were most common with monopolar “Bovie” devices (88%; p < 0.001 vs composite of all other devices). Traditional bipolar (5%) and argon beam (4%) were then next 2 most common instrument types reported to cause fires. Fires were most common in head and neck operations 67% (p < 0.001 vs composite all other locations). Oxygen was 4 times more likely to be reported as the oxidizer in comparison to drapes as the fuel (66% vs 16%, p < 0.001). Fires were most commonly limited to inside the surgical field and led to severe injury or death in 42% of cases.

Mechanical failure Mechanical failures of the device resulted in 442 (12%) of included reports, including 30 deaths. Mechanical failures were included only if they resulted in a patient injury or death, with most reports indicating malfunction without clinical significance (Fig. 2). Mechanical failures were most commonly seen with ablation devices (n ¼ 120, 27%) followed by monopolar (n ¼ 91, 21%) and ultrasonic energy (n ¼ 86, 19%). When broken down into mechanism, a “broken device” was most frequently reported with advanced bipolar (n ¼ 48, 48% of broken devices); “retained object” was most frequently reported with ultrasonic energy (n ¼ 79, 37% of retained object reports). Patterns of injury by device Interaction analyses with each device were performed to identify specific device-related trends. Monopolar devices were most frequently associated with dispersive electrode thermal injuries. Traditional bipolar device injuries were diversified, but were most commonly due to direct application injury (n ¼ 81, 30%). Advanced bipolar device events were mostly bleeding (n ¼ 305, 57%), but if thermal burn occurred, it was most commonly by direct

Overbey et al

Vol. 221, No. 1, July 2015

Table 2.

Energy-Based Device Injuries and Deaths

201

Thermal Burns Due to Surgical Energy-Based Devices Injuries (n ¼ 2,283)

Variable

Pattern of thermal burn Direct application Inadvertent application Grounding failure Insulation failure Capacitive coupling Residual heat Other Location of thermal injury Skin/mucosal burn in surgical field Skin/mucosal burn outside field Visceral injury Visceral perforation Other Severity of injury Minor (Accordion 1) Moderate (Accordion 2) Severe (Accordion 3) Death (Accordion 4) Type of energy device Monopolar “Bovie” Traditional bipolar Advanced bipolar Ultrasonic Ablation Argon beam coagulator Other Anatomic location of operation Abdominal, open or not specified Abdominal, laparoscopic Cardiovascular Endoscopy Head and neck Orthopaedic Soft tissue Thoracic-noncardiac Urology or gynecologic Unknown Timing of event Intraoperative Postoperative inpatient Postoperative at home

Deaths (n ¼ 70)

Combined (n ¼ 2,353)

n

%

n

%

n

%

694 182 657 324 203 190 33

30 8 29 14 9 8 1

59 3 1 2 1 2 2

84 4 1 3 1 3 3

753 185 658 326 204 192 35

32* 8 28 14 9 8 1

815 863 221 368 16

36 38 10 16 1

0 0 31 37 2

0 0 44 53 3

815 863 252 405 18

35 37 11 17 1

1,095 265 923 0

48 12 40 0

0 0 0 70

0 0 0 100

1,095 265 923 70

47 11 39 3

1,221 155 144 131 513 112 7

53 7 6 6 22 5 0

17 2 4 12 29 6 0

24 3 6 17 41 9 0

1,238 157 148 143 542 118 7

53 7 6 6 23 5 0

588 424 278 335 550 315 135 62 255 611

26 19 12 15 24 14 6 3 11 27

26 15 11 11 0 1 0 4 2 0

37 21 16 16 0 1 0 6 3 0

614 439 289 346 550 316 135 66 257 611

26 19 12 15 23 13 6 3 11 26

2,008 107 168

88 5 7

34 22 14

49 31 20

2,042 129 182

87 5 8

*Denotes statistical significance (p < 0.001) compared with composite of others.

application (n ¼ 64, 12%). Ultrasonic energy device injuries reported were also most commonly associated with bleeding (n ¼ 120, 19%), but were also due to

retained object (n ¼ 79, 23%) and thermal burn from residual heat (n ¼ 72, 21%). Ablation device grounding failures (n ¼ 241, 33%) were followed by direct

202

Table 3.

Overbey et al

J Am Coll Surg

Energy-Based Device Injuries and Deaths

Bleeding Due to Surgical Energy-Based Devices Injuries (n ¼ 579)

Variable

Source of bleeding Arterial Venous Uncertain Severity of bleeding Did not require transfusion Required transfusion Unknown Severity of injury Minor (Accordion 1) Moderate (Accordion 2) Severe (Accordion 3) Death (Accordion 4) Type of energy device Monopolar “Bovie” Traditional bipolar Advanced bipolar Ultrasonic Ablation Argon beam coagulator Anatomic location of operation Abdominal, open or not specified Abdominal, laparoscopic Cardiovascular Endoscopy Head and neck Orthopaedic Soft tissue Thoracic-noncardiac Urology or gynecologic Unknown Timing of event Intraoperative Postoperative inpatient Postoperative at home

Deaths (n ¼ 63)

Combined (n ¼ 642)

n

%

n

%

n

%

107 48 414

18 8 72

16 2 45

25 3 71

123 50 459

19 8 71

301 261 17

52 45 3

1 62 0

2 98 0

302 323 17

47 50 3

152 78 349 0

26 13 60 0

0 0 0 63

0 0 0 100

152 78 349 63

24 12 54 10

78 44 283 111 38 25

13 8 49 19 7 4

5 1 21 9 25 2

8 2 33 14 40 3

83 45 304 120 63 27

13 7 47 19 10 4

132 136 44 40 66 11 7 13 62 68

23 23 8 7 11 2 1 2 11 12

38 3 2 0 6 0 0 12 0 2

60 5 3 0 10 0 0 19 0 3

170 139 46 40 72 11 7 25 62 70

26 22 7 6 11 2 1 4 10 11

368 152 59

64 26 10

17 42 4

27 67 6

385 194 63

60 30 10

application burn (n ¼ 129, 18%). Argon beam had the most consistent singular pattern of direct application (n ¼ 101, 62%). Timing Overall events occurred most frequently intraoperatively (82%); postoperative events were inpatient in 9% of instances and postdischarge in the remaining 9%. Twelve percent of all deaths occurred after discharge home, with the remainder split evenly between intraoperative (45%) and postoperative while inpatient (43%).

DISCUSSION Energy-based devices have been an essential tool in the operating room over the past 20 years. Despite the growing and widespread use, complication profiles and mechanisms of different types of energy devices have not been well studied. This study identified thermal burn, hemorrhage, mechanical failure, and fire as 4 unifying patterns of energy-based device injuries and deaths. One-half of all reported events resulted in a serious injury or death. One in 10 deaths attributed to energy-based device complications occur after the patient is discharged

Overbey et al

Vol. 221, No. 1, July 2015

Table 4.

Energy-Based Device Injuries and Deaths

203

Fires Due to Surgical Energy-Based Devices

Variable

Fuel/oxidizer for fire Oxygen Drapes Unknown Location of fire Limited to surgical field Extended outside field Other Severity of injury Minor (Accordion 1) Moderate (Accordion 2) Severe (Accordion 3) Death (Accordion 4) Type of energy device Monopolar “Bovie” Traditional bipolar Advanced bipolar Ultrasonic Ablation Argon beam coagulator Anatomic location of operation Head and neck Abdomen Endoscopy Other/unknown

Injuries (n ¼ 279) n %

n

Deaths (n ¼ 15) %

Combined (n ¼ 294) n %

185 44 49

66 16 18

8 3 3

53 20 20

193 47 52

66 16 18

172 91 16

62 33 6

7 5 3

47 33 20

179 96 19

61 33 6

126 44 109 N/A

45 16 39

0 0 0 15

0 0 0 100

126 44 109 15

43 15 37 5

246 16 3 1 2 10

88 6 1 0 1 4

12 0 0 0 1 2

80 0 0 0 7 13

258 16 3 1 3 12

88* 5 1 0 1 4

187 11 15 66

67 4 5 24

6 3 0 6

40 20

193 14 15 72

66* 48 5 24

40

*Denotes statistical significance, p < 0.001 compared with composite of others.

home. An overview of the key findings of this study are reported in Table 5. The surgical community’s knowledge of mechanisms of energy-based device complications relies on experts’ interpretation of individual energy-based injury reports, based on their knowledge of fundamental principles of energy. The currently understood major mechanisms of injury patterns related to energy-based devices include direct application, inadvertent application, dispersive electrode failure, insulation failure, capacitive coupling, and residual heat. Direct application injury occurs when the active electrode directly causes thermal injury to tissue adjacent to the active electrode tip during activation of the device. This category was the broadest and therefore most commonly categorized in this study, accounting for the majority of events, and most happened with likely wellfunctioning electrosurgical devices but poor handling. Second, inadvertent application is the unintended activation of the active electrode. This was the least commonly described complication in this study and typically denotes operator error. Third, dispersive electrode failure is

described as a thermal burn underneath the dispersive electrode. This is most common with monopolar and ablation devices, and results in an alternate site injury. Fourth, insulation failure is a break or defect in the insulation that coats the active electrode.6 This was the third most common injury pattern identified, perhaps due to the fact that defining this complication requires careful inspection of the device post-injury. Fifth, capacitive coupling is current transferring from the active electrode, through intact insulation, into adjacent conductive material without direct contact.7 This was also an uncommonly described event (9% of total), but dangerous in that it is difficult to predict and cannot be eliminated.8 Sixth, residual heat describes injury from a heated electrode after the completion of activation. Eight percent of injuries were attributed to residual heat, which varies based on instrument type, but was most commonly noted with ultrasonic energy, concordant with previous results.9 These 6 mechanisms were able to describe 99% of the reports reviewed. Other mechanisms, such as direct coupling and antennae coupling, have been described, but were difficult to

204

Overbey et al

Energy-Based Device Injuries and Deaths

Table 5. Main Findings Relating to Type of Energy-Based Device and Adverse Events Monopolar “Bovie” Responsible for 88% of reported operating room fires. Most commonly injures through dispersive electrode burn (24%) or direct application injury (20%). Traditional bipolar Infrequently reported complication (7%). Events are most frequently injuries from direct application. Advanced bipolar Responsible for many reported bleeding events (47%). Infrequently causes thermal burns or fire. Ultrasonic Second most common cause of bleeding events (19%). Frequently a source of retained object (23% of ultrasonic reports). Ablation Less commonly reported to cause bleeding or burns overall, but when events occur they are more frequently fatal. Dispersive electrode failure is most frequent mechanism to burn. Argon beam coagulator Direct application is most common mechanism of injury. Percentages are % of occurrences for type of energy-based device.

elucidate from these reports and therefore were not categorized in this study.6,7,9-11 Bleeding events were most commonly reported with newer advanced bipolar and ultrasonic energy devices and were likely due to these instruments being used for dividing blood vessels, which has been described in detail elsewhere.12 This report highlights new mechanisms of energy-based device injuries, such as fires and mechanical failure, which are not typically listed as common causes of energy-based device complications. Fires have an estimated frequency of 20 to 650 per year, with diathermy and lasers as the major ignition sources.13 These results confirm head and neck operations are the most common location of an operating room fire, but they further subcategorize the associated instrument as monopolar in 88% of reported fires. Mechanical failures are possible with any complex device, but our results indicate that 12% of devicerelated injury reports were from mechanical failure events (retained object, broken device), with ultrasonic energy devices demonstrating the highest rate (24% of ultrasonic energy reports). This study is important because it assembles and analyzes the largest group of energy-based device complications published in the surgical literature. The surgical literature is limited to case reports8,14 and small case series1,2,10,12,15-17 regarding clinical complications resulting from energy-based devices. Patterns of complications cannot be recognized from these studies because too few complications are reported. This study, which includes 3,731 reported cases, allows patterns of complications, timing of complications, and severity of

J Am Coll Surg

complications to be analyzed across multiple device types and procedures. A second reason this study is important is that it includes 294 operating room fires. Operating room fires have been recognized as having a similar incidence to wrong-site operations, although with much less publicity.13 This series subdivides these fires into mechanism and severity to more accurately describe these devastating events. This study includes 2 main limitations. First, the MAUDE database is a voluntary medical device report database. The accuracy and completeness of reports cannot be established or confirmed.3 The MAUDE database recordings do not allow for an incidence of energybased devices to be calculated because the total number of uses of energy-based devices is not recorded and the event records are incomplete of all energy-based device complications occurring in the United States. Second, because surgical complications occur for many reasons, it cannot be definitively concluded that an energy-based device caused the injuries and death events reported on the MAUDE database. However, the unique strength of this study is the large sample size of these rarely occurring energy-based device complications.

CONCLUSIONS Energy-based devices are tools used by all surgical and procedural specialties. The risk of injury from electrosurgical instruments to the general population of surgical patients is significant and warrants continued research and education. This study provides insight into the reproducible patterns by which energy-based device injuries and deaths occurda fact that could allow surgeons to alter their use of surgical energy-based devices to avoid or at least minimize the occurrence of these adverse events. Additionally, the taxonomy of description presented in this study could be used to better categorize and report patient injury from electrosurgical instruments. Surgeon awareness of the common injury patterns is essential to minimize complications due to surgical energy-based devices. Author Contributions Study conception and design: Overbey, Townsend, Robinson Acquisition of data: Overbey, Townsend, Chapman, Bennett, Foley, Rau, Yi Analysis and interpretation of data: Overbey, Jones, Stiegmann, Robinson Drafting of manuscript: Overbey, Jones, Townsend, Stiegmann, Robinson Critical revision: Robinson

Vol. 221, No. 1, July 2015

REFERENCES 1. Feldman LS, Fuchshuber P, Jones DB, et al. Surgeons don’t know what they don’t know about the safe use of energy in surgery. Surg Endosc 2012;26:2735e2739. 2. Nduka CC, Super PA, Monson JR, Darzi AW. Cause and prevention of electrosurgical injuries in laparoscopy. J Am Coll Surg 1994;179:161e170. 3. Publicly available Manufacturer and User Facility Device Experience Database. Available at: http://www.accessdata.fda. gov/scripts/cdrh/cfdocs/cfmaude/search.cfm. Accessed October 8, 2014. 4. Strasberg SM, Linehan DC, Hawkins WG. The Accordion severity grading system of surgical complications. Ann Surg 2009;250:177e186. 5. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159e174. 6. Montero PN, Robinson TN, Weaver JS, Stiegmann GV. Insulation failure in laparoscopic instruments. Surg Endosc 2010; 24:462e465. 7. Feldman LS, Fuchshuber PR, Jones DB, eds. The SAGES Manual on the Fundamental Use of Surgical Energy (FUSE). New York: Springer-Verlag; 2012. 8. Vancaillie TG. Active electrode monitoring: How to prevent unintentional thermal injury associated with monopolar electrosurgery at laparoscopy. Surg Endosc 1998;12:1009e1012. 9. Govekar HR, Robinson TN, Steigmann GV, McGreevy FT. Residual heat of laparoscopic energy devices: how long must the surgeon wait to touch additional tissue? Surg Endosc 2011;25:3499e3502. 10. Wu MP, Ou CS, Chen SL, et al. Complications and recommended practices for electrosurgery in laparoscopy. J Am Coll Surg 2000;179:67e73. 11. Robinson TN, Barnes KS, Govekar HR, et al. Antenna coupling-a novel mechanism if radiofrequency electrosurgery complication. Ann Surg 2012;256:213e218. 12. Newcomb WL, Hope WW, Schmelzer TM, et al. Comparison of blood vessel sealing among new electrosurgical and ultrasonic devices. Surg Endosc 2009;23:90e96. 13. Yardley IE, Donaldson LJ. Surgical fires, a clear and present danger. Surgeon 2010;8:87e92. 14. Gilbert TB, Shaffre M, Matthews M. Electrical shock by dislodged spark gap in bipolar electrosurgical device. Anesth Analg 1991;73:355e357. 15. Brill AI, Feste JR, Hamilton TL, et al. Patient safety during laparoscopic monopolar electrosurgery- principles and guidelines. JSLS 1998;2:221e225. 16. Bishoff JT, Allaf ME, Kirkels W, et al. Laparoscopic bowel injury: incidence and clinical presentation. J Urol 1999;161:887e890. 17. Sankaranarayanan G, Resapu RR, Jones DB, et al. Common uses and cited complications of energy in surgery. Surg Endosc 2013;27:3056e3072.

Discussion DR JOHN GARRY (Fresno, CA): The stated purposes of the paper are to describe the causes of energy-based device applications leading to injury or death and to determine if common mechanisms leading to injury or death can be identified. Four main causes were identified: thermal burn, hemorrhage, mechanical failure,

Overbey et al

Discussion

205

and fire. Thermal direct application was identified as the most common reason for both injury and death. Data were extracted from the FDA Manufacturer and User Facility Device Experience (MAUDE) database. Of 6,719 reports coded as injury, 3,553 injury reports were used in this report. Also, 200 deaths were extracted from the FDA’s database, with 178 clearly deemed appropriate for this report. The high rate of exclusion is disconcerting as it relates to MAUDE. The causes of injury were clearly delineated. Interestingly, bleeding was described most commonly with advanced bipolar devices. One would wonder whether this was related to the speed of use and is most likely user related and not device failure. Actual mechanical failures of devices accounted for 12% of the injuries, 442 of included reports, including 30 deaths. Eighty-two percent of the overall events occurred intraoperatively, with 9% postoperatively while still inpatient status, with 9% postdischarge. Forty-five percent of the deaths occurred intraoperatively, with 43% postoperatively as inpatient; 12% of deaths occurred postdischarge. Direct application, inadvertent application, dispersive electrode failure, insulation failure, capacitive coupling, and residual heat are the understood major mechanisms of injury patterns related to energy-based devices. Direct application injury causing thermal injury to tissue adjacent to the tip accounted for the majority of events that occurred with well-functioning devices but with poor handling by the operating surgeon. From an epidemiologic standpoint, the occurrence rate of 1 to 2 per 1,000 operative procedures in a population of 50 million inpatient cases per year leads one to conclude there is an urgent need for improvement. Operating room protocols need to be addressed and physician education and training in the use of these devices needs to be improved. Education and training should not just be on the job type training in the operating room with a device, but should be in a skills lab. I would say that this is a prime target area for Centers for Medicare and Medicaid Service’s financial recovery program. This is an excellent delineation of device-related injuries occurring during a procedure. An overarching issue raised by this study in conjunction with MAUDE is the implication of this with Patient Safety Indicator 15 and ICD 9 code 998.2 (accidental laceration coding, which includes bleeding), which then affects the physician’s report card as well as the facility’s Hospital Acquired Conditions Reduction Program. What would be your recommendations to help reduce this occurrence and its effect on pay for performance? Do you think that reusable and reprocessed energy source devices are more of a risk to cause issues, as opposed to using new disposable devices (including reprocessed endoscopic forceps)? What changes in the MAUDE reporting structure should be made to more accurately report incidents? DR THOMAS ROBINSON (Aurora, CO): The first question asked what our recommendation would be to reduce the occurrence of adverse events that are caused by surgical energy-based devices. Improved education of the surgical community regarding surgical energy-based devices, the mechanisms by which they work, and the common profiles or patterns of complications that occur from energy-based devices is probably the most logical starting

Overbey et al

Vol. 221, No. 1, July 2015

Appendix 1. Category Code Used by the MAUDE Database for Regulation Description: Electrosurgical Cutting and Coagulation Device and Accessories Code

GEI LFL JOS ODR NUJ

DWG MUK

PDG JOT ONQ

Description4

Electrosurgical, cutting, coagulation, accessories Instrument, ultrasonic surgical Electrode, electrosurgical Electrosurgical patient return electrode Electrosurgical, cutting and coagulation accessories, laparoscopic and endoscopic, reprocessed Electrosurgical device Electrosurgical radiofrequency system, stress urinary incontinence, female, transvaginal or laparoscopic, pelvic tissue Electrosurgical vessel and/or tissue sealer with built-in generator Electrode, gel, electrosurgical Electrosurgical coagulation for aesthetic

MAUDE, Manufacturer and User Facility Device Experience.

Energy-Based Device Injuries and Deaths

205.e1