The effect of alcohol on surgical skills

Abstract

INTRODUCTION

The General Medical Council states ‘a surgeon must not work when their health state is adversely influenced by fatigue, disease, drugs or alcohol’. However, there are no defined criteria for acceptable blood alcohol levels when operating. The aim of this study was to measure the effect of varying amounts of alcohol on surgical dexterity, cognitive abilities and the social interactions required to ensure patient safety during a routine ear, nose and throat (ENT) operation.

METHODS

ENT surgeons were asked to perform a microlaryngoscopy with excision of a predetermined glottic lesion on a validated laryngeal model. The procedure was repeated four times over a period of four hours with varying doses of alcohol (no alcohol control, one glass, three glasses and six glasses of wine). The parameters recorded included theatre etiquette, surgical time, operative skills and patient safety. Scores were adjudicated by two independent observers.

RESULTS

The more glasses of wine consumed, the more detrimental the effect was on the surgical performance of all participants. There was a global reduction in ability of 7.25% after three glasses and 19.25% after six glasses of alcohol. No domain showed an improvement following sequential increase in blood alcohol concentration.

CONCLUSIONS

This study suggests that there are no deleterious effects on surgical performance following the consumption of one glass of wine 45 minutes prior to microlaryngoscopy among ENT surgeons of varying experience. However, there is clear evidence that with three or more glasses of wine there is an adverse effect on performance, with decreased surgical dexterity, cognition functions and professionalism.

Abstinence from alcohol is not a legal requirement for surgeons on duty despite the proven adverse effects of alcohol on psychomotor performance and cognitive abilities.^1–3 The Royal College of Surgeons of England states in Good Surgical Practice: ‘Surgeons must not compromise patient safety because of ill health, fatigue or the effects of drugs or alcohol.’⁴ While noble in their cause, these guidelines remain open to interpretation and safe alcohol limits for on-duty doctors therefore remain a contentious issue.

Our personal communications with senior ear, nose and throat (ENT) surgeons have revealed great variability in attitudes towards alcohol consumption during on-call shifts, ranging from the prohibition of alcohol consumption to an acceptance of levels below the legal limit to operate a motor vehicle. One long retired surgeon admitted having needed to take a taxi to the hospital on several occasions to complete his on-call duties because he was over legal driving limit. Whether the claimed habits of a retired surgeon were bravado or truth, this situation raises serious safety concerns.

Borne from a strong evidence base, the aviation industry has strict guidelines that advise pilots to abstain from alcohol for at least 24 hours before flying.^5,6 The professional activities of pilots share many parallels with those of surgeons: both perform complex procedures requiring manual dexterity in a pressurised environment in which errors could result in the loss of human life. Currently, however, there is no equivalent evidence base documenting the effect of alcohol on surgical performance.

A moderate dose of alcohol has been shown to severely impair simulated surgical laparoscopic performance among medical students.⁷ Furthermore, the ‘hangover’ effect has been shown to adversely affect surgical dexterity among post-fellowship surgeons.⁸ However, to date, there appears to be no published evidence documenting the relationship between acute alcohol consumption and surgical performance among surgeons.

The specific aims of this study were to measure the effect of varying amounts of alcohol on surgical dexterity, cognitive abilities and the social interactions required to ensure patient safety during a routine ENT operation using a model larynx. The independent variable was quantified in glasses of wine consumed, as opposed to blood alcohol concentration (BAC), to allow the results to be more relevant and accessible to surgeons. It is hoped that the results will serve as a guide to safer use of alcohol among surgeons during their time on call.

Methods

Participants

Four healthy ENT surgeons (all male, aged 28–51 years) were screened to exclude teetotallers, those with a body mass index (BMI) of >30kg/m² and those with first-degree relatives with a history of alcoholism. Participants were social drinkers (average of 14 units per week) with varied surgical experience: Participant 1 was a senior house officer who had completed no microlaryngoscopy procedures prior to the study; Participants 2 and 3 were registrars and had completed >100 and 30 microlaryngoscopy procedures respectively; Participant 4 was a consultant who had completed >500 microlaryngoscopy procedures. All participants gave their informed consent prior to their inclusion in the study.

Procedure

Participants were instructed to abstain from the use of alcohol for at least 24 hours prior to the experiment and had not eaten in the previous 4 hours before the beginning of the experiment. Each surgeon performed the surgical simulation four times during a four-hour period. The first procedure was performed before any alcohol was ingested to provide a baseline result. Subsequent simulations were carried out 45 minutes after ingestion of one (1.6 units), three (4.8 units) and six standard glasses (9.75 units) of a Sancerre 2003 (14%) wine. Participants began consuming the next beverage immediately after completing a simulation, over a period of ten minutes. No food was consumed during the period of the experiment.

Simulated microlaryngoscopy task

The simulated microlaryngoscopy task was performed in an operating theatre approved for laser surgery. It involved using a laryngeal training mannequin with predetermined pathological glottic findings. The surgical mannequin had been used previously as a validated training tool and was approved for the use of laser surgery simulation. Independent assessors and safety officers were present throughout and no contact with patients or the public was permitted. The study was performed out of normal working hours and participants were kept isolated from the main theatre complex to ensure patient safety.

The participants were marked by the independent assessor at the time of the procedure using predetermined surgical scoring criteria (Table 1). In addition, the operative procedure was recorded on a video stack system and copied on to DVDs that were later randomised and reviewed by two separate ENT surgeons.

Table 1.

Surgical scoring criteria

Variable	Scoring	Description
Position	0–2	Positions the patient correctly
Scope	0–1	Chooses the appropriate endoscope for task
Microscope	0–1	Sets up microscope safely
Magnification	0–1	Shows awareness of correct set up of microscope
Focal length	0–1	Shows awareness of correct set up of microscope
Lip care	0–1	Ensures care is taken not to traumatise lips
Teeth care	0–1	Ensures care is taken not to traumatise teeth
Lubrication	0–3	Ensures endoscope is correctly lubricated
Laser safety	0–3	Ensures patient and team conform to laser precautions
Laser technique	0–3	Uses laser safely
Instrument selection	0–3	Selects appropriate endoscopic instruments
Manipulation	0–3	Manipulates specimen carefully
Scrub nurse	0–3	Demonstrates professionalism with theatre team
Checklist	0–1	Discusses World Health Organization safety checklist

Results

Simulator performance

All surgeons completed the experiment successfully. The variance in scores during the control setting reflected the experience level of the surgeon. The scores did not correlate to age, BMI or average weekly alcohol intake.

The relationship between overall surgical performance and alcohol dose is summarised in Figure 1. All participants achieved, or matched, their best operating score 45 minutes after the administration of one glass of wine. Participant 1 achieved his lowest overall score in the sober simulation and his highest score following one glass. Subsequent scores, however, showed a downwards trend with increasing alcohol consumption. The performance of Participant 4 following three and six glasses of wine was markedly impaired compared with the control simulation, whereas the performance of Participants 2 and 3 did not show significant impairment compared to the control simulation until the administration of six glasses.

Figure 1.

Effects of alcohol ingestion on overall simulator performance

Across all participants, several areas of surgical expertise remained constant despite increasing amounts of alcohol: positioning, instrument selection and laser setting. There were no skills that showed overall improvement with increasing units across participants. Increasing amounts of alcohol appeared to have the greatest adverse effect on laser technique, then on insertion and lip care respectively.

Non-technical skills

In the sober simulation and following one glass of wine all participants demonstrated appropriate and professional interactions with the scrub nurse. However, following three glasses, participants began to use inappropriate language and demonstrated delayed responses to communication. After six glasses, participants began to forget the names of regularly used surgical instruments, with inappropriate language and frustration being evident during communication.

Feedback from independent adjudicators

The two independent observers found no deleterious effects in surgical performance following administration of one glass of wine and therefore deemed the surgery at the correct standard for clinical practice. However, following administration of three and six glasses, all participants' performance was judged as being unsafe and both independent observers stated that they would have stopped the surgeon from operating in a real life clinical setting.

Discussion

The results suggest that one glass of wine consumed 45 minutes prior to operating has no adverse effects on surgical ability among surgeons performing microlaryngoscopy. In fact, despite consuming one glass of wine, the two most junior surgeons improved their performance compared with the control simulation. Improvements might be explained by a learning curve effect, in addition to an increase in confidence without any perceived loss of judgement. The downwards trend in the house officer's performance following three and six glasses may demonstrate a reversal of the learning curve due to impairment by intoxication. Following the administration of three standard glasses of wine, all surgeons were deemed to be unsafe to perform the operation by independent observers.

The ability of surgeons to communicate with the scrub nurse was affected negatively by three or more glasses of wine. This impairment of non-clinical (social and cognitive) skills is in line with previous research documenting that alcohol selectively impairs fundamental mechanisms of behavioural control and that it can lead to aggression and other socially deviant behaviours.^9,10 Non-technical skills such as communication and teamworking are known to play a significant role in ensuring patient safety during surgical procedures.¹¹ The impairment of non-technical skills following three or more glasses is therefore a significant safety concern.

The question of what to do if a senior colleague has the aroma of alcohol on his or her breath while attending an emergency is a well recognised topic in senior medical interviews. The expected answer is that one would prevent the said surgeon from operating or attending a patient. However, there is no clearly defined protocol for those surgeons who have consumed small amounts of alcohol during extended periods of on-call duties. Other high risk industries such as aviation and the nuclear power industry have implemented clearly defined safety strategies in relation to alcohol consumption prior to and during working hours. Clear, transparent guidelines directed at alcohol use for on-duty doctors may help improve patient safety.

Study limitations

While the present study provides important insights, its limitations must be recognised. A key weakness is the lack of a control group, which is necessary to isolate the effects of alcohol and account for the learning curve effect. Extending the parameters of the study to employ a between-subjects design with an alcohol and placebo condition is necessary in order to discern the effect of other confounding variables (eg fatigue).

Under the current legislation, self-regulation is relied on to ensure surgeons drink responsibly during their on-call duties. Without clear guidelines, individuals may use internal cues to assess their level of intoxication. Future research should therefore use tools to measure self-reported intoxication (eg on a Likert-type visual analogue scale). This should be compared with actual intoxication levels such as breathalyser readings.

This study provides information regarding estimates of the safe number of glasses of wine that ENT surgeons can consume while on call. Nevertheless, caution must be taken when drawing conclusions based on the results of a single experiment. Considerable research is needed to fully characterise the nature of alcohol-induced impairment on surgical performance. Studies aiming to negotiate definitive guidelines on alcohol use while on call will need to document the safe BAC limit, thereby allowing for individual differences in alcohol metabolism. In addition, testing a dose response effect of alcohol in smaller increments is needed in order to decipher the safe BAC limit for operating.

Further key strategies to improve on the present study include increasing the sample size and the inclusion of women to ensure that samples better reflect the UK population of ENT surgeons. Of course, the on-call duties of ENT surgeons extend beyond those in the operating theatre and, consequently, research on the adverse effects of alcohol consumption prior to ward-based duties may be of relevance to potential guidelines.

Conclusions

This study suggests that there are no deleterious effects on surgical performance following the consumption of one glass of wine 45 minutes prior to microlaryngoscopy among ENT surgeons of varying experience. However, there is clear evidence that consuming three or more glasses of wine does adversely affect performance, with decreased surgical dexterity, cognition functions and professionalism. The results of this study augment the innate knowledge among members of the surgical community and mirror the current General Medical Council guidance. Further research into safe alcohol limits for surgeons during on-call shifts can only serve to facilitate education, allowing surgeons to recognise their own limitations and increase patient safety.