Source-linked AI summary
Adverse Events in Robotic Surgery: A Retrospective Study of 14 Years of FDA Data
Homa Alemzadeh, Ravishankar K. Iyer, Zbigniew Kalbarczyk, Nancy Leveson, Jaishankar Raman
TL;DR
Robotic-surgery adverse events can affect patients and surgical teams, motivating analysis of their patterns and causes. A 14-year FDA MAUDE review found injury and death events per procedure remained relatively constant, with higher event probabilities in complex specialties.
Problem
The study examines how robotic-surgery adverse events affect patients and teams to inform safer system design.
Method
The authors retrospectively analyzed robotic-system malfunctions during surgical procedures and their patient impacts.
Results
Injury and death events per procedure stayed relatively constant, while complex specialties had higher event probabilities than other specialties.
Takeaways & Limitations
Uniform standards and safety-based design techniques should be developed to reduce incident rates as robotic systems evolve.
Takeaways & Limitations
The MAUDE database suffers from underreporting and inconsistencies.
Abstract
from arXiv · showhide
Understanding the causes and patient impacts of surgical adverse events will help improve systems and operational practices to avoid incidents in the future. We analyzed the adverse events data related to robotic systems and instruments used in minimally invasive surgery, reported to the U.S. FDA MAUDE database from January 2000 to December 2013. We determined the number of events reported per procedure and per surgical specialty, the most common types of device malfunctions and their impact on patients, and the causes for catastrophic events such as major complications, patient injuries, and deaths. During the study period, 144 deaths (1.4% of the 10,624 reports), 1,391 patient injuries (13.1%), and 8,061 device malfunctions (75.9%) were reported. The numbers of injury and death events per procedure have stayed relatively constant since 2007 (mean = 83.4, 95% CI, 74.2-92.7). Surgical specialties, for which robots are extensively used, such as gynecology and urology, had lower number of injuries, deaths, and conversions per procedure than more complex surgeries, such as cardiothoracic and head and neck (106.3 vs. 232.9, Risk Ratio = 2.2, 95% CI, 1.9-2.6). Device and instrument malfunctions, such as falling of burnt/broken pieces of instruments into the patient (14.7%), electrical arcing of instruments (10.5%), unintended operation of instruments (8.6%), system errors (5%), and video/imaging problems (2.6%), constituted a major part of the reports. Device malfunctions impacted patients in terms of injuries or procedure interruptions. In 1,104 (10.4%) of the events, the procedure was interrupted to restart the system (3.1%), to convert the procedure to non-robotic techniques (7.3%), or to reschedule it to a later time (2.5%). Adoption of advanced techniques in design and operation of robotic surgical systems may reduce these preventable incidents in the future.
Introduction
Robotic minimally invasive surgery has expanded rapidly, yet no comprehensive safety and reliability study had been performed. This study analyzes FDA MAUDE adverse events from 2000–2013 across six surgical specialties to assess patient impact and inform safer future system design.
- Background: 1.74 million robotic procedures were performed in the U.S. between 2007 and 2013, including 86% in gynecology and urology.Procedures in all other specialties combined accounted for 14%.
- Study scope: The study analyzes all robotic surgical adverse events collected in the FDA MAUDE database during 2000–2013 across six major surgical specialties.The analysis includes gynecology, urology, general, colorectal, cardiothoracic, and head and neck surgery, covering deaths, injuries, and device malfunctions.
- Study objectives: The analysis examines whether evolving robotic technologies and features improved system safety and effectiveness across surgical specialties.It measures the causes and impact of safety-related incidents on patients and surgical progress.
- Research gap: Previous studies focused mainly on gynecology and urology or on small subsets and specific device failure modes.Reported failure modes included electrocautery, electrosurgical injuries, and instrument failures.
- Study objectives: The study aims to inform future surgical-system design using advanced safety mechanisms, improved human-machine interfaces, and regulated operational practices.The stated goal is to minimize adverse impacts on patients and surgical teams.
Methods
The study analyzed FDA MAUDE reports on robotic surgery systems and instruments from 2000–2013, combining structured fields with parsed and manually validated narrative data. Event rates were compared across specialties and years, while manual review characterized causes of injuries and deaths and malfunction trends.
- Data extraction: Automated narrative analysis used domain-specific dictionaries, pattern-matching rules, POS and negation taggers, and manual review for accuracy and validity.The extraction covered patient injuries and deaths, specialties and procedure types, malfunction types, and surgery interruptions.
- Rate estimation: Cardiothoracic and head and neck procedure volumes were estimated by assigning most procedures outside gynecology, urology, and general surgery to those specialties.This assumption followed manufacturer reports identifying them as the only other specialties using the robot.
- Assumptions and limitations: Injury and death underreporting was assumed to be low and surgery-independent, but changing reporting rates limited comparisons across years; specialty comparisons used whole-period totals.Two-sided P values below 0.05 and 95% confidence intervals determined statistical significance.
- Cause and trend analysis: Manual review of all pre-2013 event descriptions characterized causes attributed to injury and death events, while malfunctions per procedure assessed trends from 2004–2013.MAUDE was treated as a sample set for estimating lower bounds on adverse-event prevalence and identifying major causes and impacts.
Results
Across 10,624 FDA reports from 2000–2013, device malfunctions predominated, while adverse-event rates and impacts varied substantially by specialty and malfunction type. Injuries and deaths were linked mainly to malfunctions, surgical risks, and operator errors.
- Overall adverse events: 10,624 events were extracted, including 1,535 (14.4%) with significant negative patient impacts, 8,061 (75.9%) device malfunctions, and 1,028 events with unavailable or “Other” event types.The negative-impact events included 1,391 injuries and 144 deaths; 160 injury events were reported as “Malfunction” or “Other.”
- Adverse events across surgical specialties: 52.2 per 100,000 procedures vs. 5.7: estimated death rates were higher in cardiothoracic and head and neck specialties than in gynecology, urology, and general surgery.Injury rates were 91.0 versus 71.5, and conversion rates were 89.7 versus 29.2 per 100,000 procedures.
- Procedure impact: 1,104 cases (10.4% of all the adverse events) interrupted procedures for troubleshooting, system resets, conversion to traditional techniques, or rescheduling.System errors and video/imaging problems contributed to 787 (7.4%) adverse events and were major contributors to resets, conversions, and aborted or rescheduled procedures.
- Patient impacts and causes: 62% of injury events involved device malfunctions; deaths were related to inherent surgical risks or complications in 33.7% of cases and operator mistakes in 7%.Other injury causes included operator errors (7.1%), improper positioning or port incisions (6.3%), inherent surgical risks (3.9%), and grounding problems (1.5%).
Discussion
MAUDE report counts cannot establish event-rate trends because reporting is affected by use, reporting practices, and awareness, so the study assessed events per procedure. Injury/death rates per procedure remained constant since 2007, while complex specialties had higher estimated event burdens and the authors propose improved safety controls.
- Interpretation and limitations: MAUDE report increases cannot establish changing event rates because reporting varies with system use, manufacturer practices, and public awareness.The database also suffers from underreporting and inconsistencies.
- Interpretation and limitations: Injury/death events per procedure stayed constant since 2007, despite a relatively high number of reports and successful procedures comprising the vast majority.The estimated event counts per procedure are likely lower than the actual numbers because of MAUDE underreporting.
- Differences across specialties: Complex specialties, including cardiothoracic and head and neck surgery, had higher estimated injury, death, and conversion burdens than gynecology, urology, and general surgery.Possible explanations include greater procedural complexity, less frequent robotic use, and less robotic expertise, although these factors could not be definitively assessed.
- Safety improvements: Improved safety practices and controls could prevent some events through procedure-monitoring safety engines, safe-path feedback, and better interfaces and simulators.Proposed systems would monitor surgeon, patient, and device status, provide troubleshooting feedback, model anatomy and robotic motions, and train teams to handle technical problems.
- Interpretation and limitations: The reports lacked sufficient detail to determine exact event causes or assess sensitivity to reporting mechanisms, team expertise, and inherent surgical risks.Procedure complexity itself carries inherent risks, particularly in complex procedures.
Conclusions
The study found that injury and death events per procedure remained relatively constant, while complex specialties had higher event probabilities and device malfunctions affected patients and surgical teams. It recommends standardized training, improved interfaces and reporting, and safety-based design to reduce future incidents.
- Injury and death events per procedure stayed relatively constant over the years.
- Complex cardiothoracic and head and neck specialties had higher event probabilities than other specialties.
- Device and instrument malfunctions affected thousands of patients and surgical teams by causing complications and prolonged procedure times.
- Reducing future incident rates requires uniform team training, advanced human-machine interfaces, improved accident investigation and reporting, and safety-based design.
Appendix · Underreporting
The appendix frames underreporting as incomplete event counting or recording errors and explains that prevalence estimates rely on representative samples, sufficient sample sizes, and assumptions about missing cases. For this study, procedure volumes were reconstructed from manufacturer sources to estimate adverse-event prevalence while accounting for possible underreporting.
- Underreporting: Underreporting occurs when event counts are incomplete or outcomes are recorded incorrectly.The passage identifies this as a common issue across social sciences, public health, criminology, microeconomics, and other fields.
- Underreporting: Prevalence estimates assume that sampled events share the actual population’s characteristics and distributions, including underreported and missing cases.This assumption supports estimating event prevalence from a random sample.
- Underreporting: Sample-size calculations use confidence intervals to make estimated proportions precise and likely close to the actual proportions.The method uses normal approximations to the binomial distribution for large and finite populations.
- Underreporting: The study estimated adverse-event prevalence from a sufficiently large sample under the assumption that observed and actual event distributions do not differ significantly.The authors state that they are investigating extensions of the proposed inference approach.
- Underreporting: 2010–2013 procedure counts came from manufacturer annual reports, whereas 2004–2009 counts were estimated from investor-presentation graphs.When sources disagreed or reported only worldwide totals, the maximum annual procedure count was selected to obtain a lower bound on event likelihood.
- Underreporting: A 4-degree polynomial fit estimated weekly procedure counts from annual totals with R2 = 0.999.Weekly numbers were calculated using the area under the fitted curve for each week.
- Underreporting: 254 (62.0%) of 410 injury reports were attributed to device malfunctions.Other listed causes included surgeon or staff mistakes, improper patient positioning, inherent surgical risks, and tissue burning.
- Underreporting: 15 (17.4%) of the 410 injury reports involved after-procedure infection/sepsis or heavy bleeding, while 77 (18.8%) had no available cause.The examples illustrate that reported patient impacts and causes were not uniformly specified.