Ear, Nose & Throat Journal2023, Vol. 102(2) 85–89© The Author(s) 2020Article reuse guidelines:sagepub.com/journals-permissionsDOI: 10.1177/0145561320980242journals.sagepub.com/home/ear
Background: Fatigue is thought of as a leading cause of iatrogenic accidents. A significant deterioration in qualitative balance function has been shown in sleep deprived individuals. Aim/Objectives: To quantify the degree to which balance is impaired by sleep deprivation (SD) in post-call medical residents. Methods: Medical residents voluntarily underwent computed dynamic posturography (CDP) before and after an on-call night, at an identical time of the day. Order of test performance was random to avoid behavioral learning. Each participant served as his or her own control. Results: Seventeen residents were enrolled (median age 32years). Average sleeping duration the night before and during the night shift was 6.5 and 1 hour, respectively. The average response times difference between alert and fatigued was 10.15 milliseconds (95% CI: 6.81-13.49 milliseconds), yielding a significantly prolonged response times from 120 milliseconds before to 130 milliseconds after the night shift (P < .001). Comparison of additional measurements of CDP performance did not differ between test conditions. Conclusion: Medical residents are fatigued due to the effect of on-call nights. Sleep deprivation prolongs response times to vestibular stimuli. This finding probably has an effect on execution of manual skills and may reflect a more generalized slowing of responses and overall performance impairment. Significance: The vestibular system is susceptible to SD.
Even though social rights regarding working and resting times have been established in the modern times,1 extended working ours among medical personnel (mainly residents), selected industrial fields workers, and air and ground transportation workers are common.2,3 Naturally, fatigue has been hypothized to be the leading cause of occupational accidents in general4-6 and in medicine in particular.7-10 The great importance of fatigue related accidents is there preventable nature, once fatigue is avoided.
Several attempts have been made to develop a simple objective fatigue screening test (eg, electrocardiogram, electroencephalogram, etc) among employees in high fatigue-risk professions, without any practical significance.11 The ideal screening test should be noninvasive, available, non-time consuming, and cheap to perform. Even though computed dynamic posturography (CDP) doesn’t meet all these criteria, it was found to be fatigue-sensitive.11,12 Only one study investigated the CDP as a tool for fatigue screening in the ‘‘real-world’’ among post-call medical residents; this study had small participants number and didn’t reach statistical significance, maybe due to heterogeneous fatigue levels among the post call residents because of various sleeping hours during the call.13 The aim of this study is to quantify the degree to which balance is impaired by sleep deprivation in post-call medical residents in a referral tertiary medical center.
The study has been approved by the research ethics committee of a tertiary referral medical center. Medical residents were recruited from the internal and surgical division at ‘‘Sourasky’’ Medical Center. Informed consent has been obtained before participation in the study. Each participant fulfilled a demographics questionnaire, as well as 2 sets of CDP tests, which was performed in the outpatient clinic, at the same medical center where the residents work. Each participant served as his or her own control. One CDP test (CDPa-CDP when alert) was performed after an overnight sleep at home, with a minimum of 72 hours from the first morning after the weekend, or last night shift, in order to ensure optimal alertness. The other CDP test (CDPt-CDP when tired) was performed immediately after the night shift, at the same hour as the CDP (08:00-08:30 AM) to avoid performance change due to circadian rhythm. As a referral tertiary center, the night shifts are very busy and stressed and usually characterized by continuous work without the possibility to nap. In fact, the night shift is in continuity to a regular 8 hours’ working day and begins at 15:00 until 08:00 the following day, so actually the resident performed the CDPt in conditions of severe sleep deprivation, following 25 hours’ work.
Some residents performed CDPa before CDPt, and vice versa, according to the logistic constraints of the Hearing and Speech Unit. For each participant, at least 2 weeks have passed between the CDPs examinations to minimize behavioral learning. Excluded residents who declared the use vestibularaltering medications, consume alcohol regularly or suffer a condition that might be associated with a vestibulopathy (eg, vestibular migraine, vestibular schwannoma, peripheral neuropathy, pregnancy, visual impairment, other central nervous system conditions, or musculoskeletal disability). All residents were instructed to avoid caffeinated drinks 18 hours before each CDP. Residents, who declared to sleep less than 6 hours before CDPa, as well as more than 2 hours before CDPt, performed the respected test in a later occasion. All patients were tested on the same CDP instrument (EquiTest; Neurocom Int Inc), with a single examiner who was blind to their sleeping deprivation status.
Categorical variables were described as frequency and percentage and continuous variables as median, interquartile range (IQR), and range. Generalized estimating equations (GEE) model was used to study the association between each of the predictor and the response time before a night shift. In order to evaluate the change in each predictor before and after a night shift, we used a GEE model with interactions variables between the night shift and each predictor as well as the night shift and the predictor. We also performed a multivariate model that included all the night shifts, all the predictors, and interactions between each of the predictors with the night shift. All the statistical analyses were 2-tailed. A P value less than .05 was considered statistically significant. SPSS (IBM SPSS Statistics for Windows, Version 22.0; IBM Corp) was used for all the statistical analyses.
Of the 17 residents enrolled to the study, 13 (76.5%) were males and 4 (23.5%) were females. The median (range) age was 32 (26-37) years. Ten (58.8%) residents recruited from the surgical division and 7 (41.2%) from the internal division. Minimal and maximal (median, IQR) response times throughout the study were 90 milliseconds and 200 milliseconds (130, 120-140 milliseconds), respectively. The mean sleeping duration before CDPa and CDPt was 6.53 and 0.95 hours, ranging from 5 to 8 and 0 to 2.50 hours, SD 0.92 and 0.95 hours, all respectively. Response times in CDPa ranged (median, IQR) from 90 to 160 milliseconds (120, 120-130 milliseconds), and in CDPt from 100 to 200 milliseconds (130, 120-140 milliseconds; P < .001). The average difference in response times (ΔCDPt-CDPa) was 10.15 milliseconds (95% CI: 6.81-13.49 milliseconds; Figure 1). The CDPa response times between genders and age didn’t reach statistical significance (P = .399 and .753, respectively). ΔCDPt-CDPa response times were not affected differently between genders, with a mean difference of 0.08 milliseconds (95% CI: −7.643 to 7.803; P = .984). ΔCDPt-CDPa response times changed differently with age of participants (P = .021). For each year of life, CDPt response time decreased in 1.217 milliseconds (95% CI: 0.0182-2.252 milliseconds; Figure 2).
CDPa failed to detect a significant change in response times between directions (ie, anterior vs posterior) or sides (right vs left) of the force plate displacement (P = .099 and .166, respectively). The change of power was associated with a significant change in CDPa response times: even though mild power, with any combination, was not significantly affected response times (P = .19 with moderate, and P = .809 with severe), the change from moderate to severe power displacement affected response time significantly (P = .017). Putting all 3 power combinations together, it was found to be insignificant in relation to response time (P = .025; Table 1).
Analyzing the effect of sleep deprivation on CDP performance (CDPt), using CDPa as control for each participant, indicates that the side of the force plate displacement is not associated with significant response time difference (ΔCDPt-CDPa), mean = 2.451 milliseconds (95% CI: −2.791 to 7.693 milliseconds; P = .359). On the other hand, direction affected response time difference significantly, mean = 6.373 milliseconds (95% CI: 1.464-11.281 milliseconds; P = .011). The change of power was associated with a significant response time difference. ΔCDPt-CDPa response time between mild and moderate force plate displacement had a mean of 14.265 milliseconds (95% CI: 9.675-18.854 milliseconds; P<.001). A similar phenomenon was observed with mild to severe force plate displacement, with a mean of 14.412 milliseconds (95% CI: 8.785-20.038 milliseconds; P<.001). Moderate to severe force plate displacement was not associated with significant ΔCDPt-CDPa response time difference, with a mean of 0.147 milliseconds (95% CI: −3.510 to 3.804 milliseconds; P = .937). Putting all 3 power combinations together, it was found to be significant in relation to response time difference (P < .001).
A multivariate analysis for interaction of the participant’s characteristics as well as different parameters of force plate displacement with ΔCDPt-CDPa found significant dependency only in age (1.228 milliseconds reduction for each life year, P = .002), power (P <.001), and direction (P = .011; Table 2).
Fatigue is considered to be a potential cause of accidents, and is one of the most common reason for iatrogenic accidents and human errors.14 In the modern industrial cultures, some disciplines are active 24/7 as in medicine, ground and air transportation, and industry. These working schedule patterns expose not only the employee but also the customer to the results of fatigue-related accident. Even though alcohol was considered as the most important factor for ground transportation traffic accident in the past,15 it is now evident that fatigue is the leading cause of fatal traffic accidents and contribute greatly to ones which were classified as ‘‘alcohol-related.’’16 In fact, 24 hours’ sleep deprivation is associated with reduced psycho-motor performance equivalent to moderate alcohol intoxication.17
Although recent restrictions partially introduced working hours limits among medical personnel,18 several parameters other than the total hours worked are considered as factors for fatigue-related accident. For example, night-time errors are more common than would be expected by chance alone, even in non-fatigued personnel.19
In contrast to air and ground transporters and certain industrial field workers, who must adhere to strict resting schedule, there is no such regulations for a medical resident. As a result, fatigue among medical residents, is not uncommon and might put the patient in danger. One of the most known fatiguerelated fatal medical accident is the monumental event of ‘‘Libby Zion’’ which took place at March 1984.20 Based on this tragic incident, many working regulations were emerged, among them, one can find restriction of 80 working hours a week, and 24 hours consecutively.18 However, these regulations are restricted to several states and not well accepted in the Western medical centers.
In this study, CDP performance was studied among medical residents, pre and post night shift. The tertiary referral center where the study was conducted is very busy at nights, and the night shifts are considered highly active. This point is additive to the fact that each night shift begins at 4 PM, after 8 hours of regular working day, and ends at 8 AM, 24 hours after working actually was begun.
Even though stability limits didn’t differ significantly as a function of sleep deprivation, reaction times to vestibular stimuli differ significantly. The association between delayed reaction time and fatigue, studied extensively in the past, and carry significant attention among pilots and drivers. Reaction time is defined as the time lag between external stimuli, to a response, usually motor one. Several types of stimuli exist: auditory, visual tactile, or their combination. Our study demonstrate that sleep deprivation prolongs response times to vestibular stimuli.
When the methodology of this study was designed, we decided to avoid controlling other variables but pre- or postnight shift (eg, caffeine consumption, sleeping duration before the night shift etc). Even though it would be logic to keep all variables controlled, the aim of this study was to reflect the ‘‘real-world’’ conditions.
This finding probably has an effect on execution of manual skills and may reflect a more generalized slowing of responses and overall performance impairment.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Omer J. Ungar https://orcid.org/0000-0002-5904-1291
1 Department of Otolaryngology Head and Neck Surgery and Maxillofacial Surgery, Tel-Aviv Sourasky Medical Center, Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
Received: October 22, 2020; revised: November 15, 2020; accepted: November 18, 2020
Corresponding Author:
Ophir Handzel, MD, Department of Otolaryngology-Head and Neck Surgery and Maxillofacial Surgery, Tel-Aviv Sourasky Medical Center, 6 Weizman Street, Tel-Aviv 6423906, Israel.
Email: ophir.handzel@ozen.co.il
