TECHNICAL REVIEW

Ergonomics and GI endoscopy

Amandeep K. Shergill, MD, Kenneth R. McQuaid, MD, David Rempel, MD, MPH
San Francisco, Berkeley, California, USA

Gastroenterologists are at risk for overuse injuries, such METHODS

as carpal tunnel syndrome (CTS), DeQuervain’s tenosyno-
vitis, and lateral epicondylitis due to the repetition and A systematic review was performed of published stud-
prolonged awkward postures associated with endoscopy. ies that evaluated the ergonomics of endoscopy and fo-
Much attention has been paid to the safety of patients cused on current video endoscopy technology. Thus,
during endoscopic procedures, but the safety of the phy- 1990 was chosen as the first year of literature review, be-
sician is often forgotten. Indeed, the rigorous training and cause this is when videoendoscopy became widely avail-
increasing demand for colonoscopy have made the gastro- able. MEDLINE (1990 to October 2008) was searched for
enterologist a commodity in today’s workplace. Survey- English language publications by using the following
based studies estimate a 37% to 89%1-7 prevalence of mus- MeSH terms and search strategy: (‘‘human engineering’’
culoskeletal pain among gastroenterologists, although the or ‘‘cumulative trauma disorders’’ or ‘‘musculoskeletal
proportion of occupation-related injuries is unknown. Ac- disease’’ or ‘‘occupational disease’’ or ‘‘equipment de-
cording to a recent American Society for Gastrointestinal sign’’) and (‘‘gastroenterology’’ or ‘‘endoscopy, digestive
Endoscopy (ASGE) survey, gastroenterologists spend system’’ or ‘‘endoscopy, gastrointestinal’’ or ‘‘colono-
43% of their time performing endoscopy,8 and increased scopy’’), and the following key words: ‘‘ergonomics of en-
endoscopy volume is associated with an increased risk doscopy,’’ ‘‘endoscopist injury,’’ ‘‘medical ergonomics,’’
of musculoskeletal complaints.1,5,6 ‘‘endoscopy and musculoskeletal strain,’’ ‘‘musculoskele-
Ergonomics is the study of the physical and cognitive tal injury and endoscopists,’’ ‘‘occupational diseases and
demands of a task in relation to an individual’s capacity. endoscopy,’’ ‘‘cumulative trauma disorder and endos-
In effect, ergonomics evaluates how a job can best be fit copy,’’ and ‘‘repetitive strain injury and endoscopy.’’ A to-
to an individual, instead of forcing an individual to fit tal of 9279 articles were retrieved. To broaden the results
into a job. Because work-related injuries can be devastat- and to allow for inclusion of abstracts, further searches
ing to a physician’s livelihood, it is important for physi- were performed on Google Scholar by using the key
cians to be educated on ergonomic principles to words listed above. References from all relevant articles
minimize the risk for endoscopy-related injury. We herein were also reviewed to identify additional articles. The
review the available literature on the prevalence, risk fac- principle investigator (A.S.) searched all titles and re-
tors, and potential mechanisms for upper extremity and viewed abstracts to determine eligibility. Full articles
neck injuries in endoscopists and propose general ergo- were retrieved for all relevant titles for which an abstract
nomic guidelines to reduce these risks. was not available and for all abstracts that appeared to
meet inclusion criteria. Articles were included if they
were determined to be related to the ergonomic evalua-
tion of GI endoscopy. More specifically, articles were in-
Abbreviations: ACGIH, American Conference of Industrial Hygienists; cluded if they addressed the prevalence, risk factors, or
ASGE, American Society for Gastrointestinal Endoscopy; CTS, carpal mechanism of injury in endoscopists; if they measured
tunnel syndrome; EMG, electromyography; PAL, progressive addition posture or forces during endoscopy; or if they provided
lenses. recommendations on tactics to reduce injury in endo-
DISCLOSURE: The following author disclosed financial relationships scopists. Studies that addressed ergonomic issues associ-
relevant to this publication: A. K. Shergill: recipient of an ASGE Cook ated with fiberoptic endoscopy were excluded from
Endoscopy Research Career Development Award, which provides
review.
salary support and protected time, and the ASGE Olympus
Endoscopic Research Award, which supports her ongoing research in After review of all titles and abstracts by the primary in-
the ergonomics of colonoscopy. All other authors disclosed no vestigator, only 12 articles initially were identified through
financial relationships relevant to this publication. MEDLINE that fulfilled these criteria.1,6,7,9-17 After review
Copyright ª 2009 by the American Society for Gastrointestinal Endoscopy of Google Scholar results and the references from relevant
0016-5107/$36.00 articles, an additional 9 articles were identified.2-5,18-22
doi:10.1016/j.gie.2008.12.235 One article, previously published in abstract form, was

www.giejournal.org Volume 70, No. 1 : 2009 GASTROINTESTINAL ENDOSCOPY 145

Ergonomics and GI endoscopy Shergill et al

TABLE 1. Summary of endoscopist surveys

Type of Response Number of Age Sex

Study study rate (%) endoscopists (y) (% men)

Buschbacher,1 1994 Survey of ASGE members 72 265 Mean (SD) 95.1

47.8  8.6

O’Sullivan Survey of 74 114 – –

et al,7 2002 Canadian
ERCP endoscopists

Liberman Survey of worldwide members of 28 608 Mean (SD) 89.3

et al,6 2005 American Society for 48  9.5
Colon and
Rectal Surgery
Keate,4 2006 Online survey of ASGE N/A 237 – –
members
Hansel et al,3 2007 E-mail survey, case-control study of GI-group, 63 72 – –
Mayo Clinic gastroenterologists
(GI group) and nonprocedure-oriented non-GI 10
internists and subspecialists (non-GI group, 45
group)
Lee and Valiozis,5 Survey of members of 12.4 94 88% between ages 31 and 60 84
2007 Gastroenterological Society of Australia
Byun et al,2 2008 Endoscopists practicing in general N/A 55 Median age 39 67
hospitals or health promotion centers
in Korea willing to participate
ASGE, American Society for Gastrointestinal Endoscopy; N/A, not applicable; CTS, carpal tunnel syndrome.
*The total of 12% reported missing work because of injury and/or requiring surgery for injury; surgery rate not specifically defined.

included in this review, and has subsequently been these injuries are more common among gastroenterolo-
published.21 One abstract was provided by one of the gists than among other internal medicine specialists.3
investigators in this study (K.M.).23 The risk of injury appears to be related to endoscopy
The full texts of all selected articles were reviewed by volume.1,5,6
the primary investigator (A.S.) for extraction of data. A survey of 400 randomly selected gastroenterologists
Seven articles1-7 addressed the prevalence of endoscopist from the ASGE in 1994 reported 57% of respondents
injury. Five articles9,11,17,18,22 evaluated or discussed possi- had a main musculoskeletal complaint, with many having
ble risk factors and mechanisms of injury. One online more than one complaint.1 Of those with pain, 85% re-
study19 evaluated postures during endoscopy, whereas 4 ported that their condition bothered them at work, and
articles10,12,20,21 evaluated the forces applied during en- 44% had suffered from the pain for more than 6 months.
doscopy, and 6 articles13-16,22,23 provided recommenda- Physicians who performed more endoscopies (measured
tions to reduce injury during endoscopy. in terms of hours per week, number of procedures per
week, or percentage of work time devoted to endoscopy)
were more likely to complain of thumb, hand, elbow, or
RESULTS low back pain. Thumb, wrist, and hand pain more com-
monly occurred on the left and elbow pain on the right.
Prevalence of endoscopist injury Physicians who complained of thumb, hand, and neck
A total of 7 articles1-7 were included (Table 1). These pain tended to attribute their symptoms to performing
studies differed in the population being studied, the sur- endoscopy.1
vey instruments, and the types of procedures performed, A study of surgeons who performed colonoscopy re-
which precluded pooling of data. Therefore, a descriptive ported that 39% of surgical endoscopists had at least
analysis is provided. In total, these studies demonstrate one injury or pain related to colonoscopy, whereas the
a prevalence of musculoskeletal complaints among endo- risk increased to 47% among surgeons who performed
scopists ranging from 37% to 89%.1-7 Common regions of more than 30 colonoscopies per week.6 Most of the right
pain are the left thumb, right wrist, neck, and back, and upper-extremity injuries were attributed to torquing of the

146 GASTROINTESTINAL ENDOSCOPY Volume 70, No. 1 : 2009 www.giejournal.org

Shergill et al Ergonomics and GI endoscopy

TABLE 1 (Continued)

Years in Mean (SD) time % Reporting Type of % Requiring % Time

practice spent in endoscopy injury injury (%) surgery off work

All O0.5 y 12.4 h/wk 57 Low back (13), thumb (10), neck (10), 3.2 –
elbow (8), CTS (4), shoulder (3), hand
numbness (3), other (6)

Mean (SD): injured, – 67 Back (57), neck (46), hand pain (36), 8 –
14.7 (7) y; noninjured, shoulder (16), elbow (8), other (5)
11.6 (5.9) y
Mean (SD): 14.8 2.4 39 Right foot (12.5), neck (10.7), right hand 2.7 1.3
(8.6) y (1.9) (8.7), back (8.6), right fingers (7.2), left
d/wk fingers (6.4), left-hand pain (6.3), left
thumb (5.3), right thumb (3.3), CTS (2)

– – 78 Hand or CTS (43), (!12%)* 12

back (29), neck (28), other (16)
– – 74 GI group: thumb (19), hand (17), – 13.2
back (12),
35 neck (10)

– – 37 Thumb (70) – 9

Median duration 19.5 (7.7) 89 Right shoulder (31), left shoulder (29), 0 2
3.25 y h/wk left finger (29), right wrist (26), neck
(18), left wrist (18), right finger (16)

colonoscope, and the left thumb, finger, and hand injuries  7.7 hours per week and a mean (SD) number of 270.2
were attributed to turning the control dials and gripping  153 endoscopies per month. This survey also included
the colonoscope handle.6 the largest percentage of women endoscopists (33%). Se-
In an online survey of 237 ASGE members, 78% of re- vere pain (pain greater than 55 mm on a 100-mm visual
spondents reported one or more injury.4 Most respon- analog scale) was reported by 47.3% of endoscopists
dents (85%) reported pain when performing endoscopy. (26/55), with more women than men reporting severe
The endoscopic activities associated with pain included pain (61% vs 40%, respectively; P Z .15).2
applying torque to the colonoscope (29%) and manipulat- The type of injury is an important factor in determining
ing the colonoscope head (17%) (left thumb pain).4 the duration of disability. A retrospective cohort study of
A case-control study compared gastroenterologists at disability claims in Washington State determined that a di-
the Mayo Clinic with nonprocedure-oriented internal agnosis of CTS and back-neck sprain predicted a longer
medicine specialists and subspecialists.3 The incidence of duration of disability. For instance, fewer than 50% of
any musculoskeletal injury was higher in the GI group workers diagnosed with CTS returned to work within 1
(74%) compared with the non-GI group (35%). In this month of their injury.24 This is in contrast to disability
study, there was no significant association between endos- from more acute injuries, such as fractures and non-
copy volume or years performing endoscopy and injury.3 back–related sprains and strains, in which up to 70% of
This is in contrast to the 1994 ASGE survey,1 the survey workers had returned within 1 month and 80% to 95%
of surgical endoscopists,6 and a recent study of members by 2 months.24 Disability data were not consistently pro-
of the Gastroenterological Society of Australia, in which vided in the various studies. Surgery was required in 0%
volume of endoscopy was a significant predictor of injury.5 to 8%1,2,4,6,7 of respondents who reported injury in the
Recently, 55 Korean endoscopists participated in a sur- various studies, and 2% to 13%2-6 required time off work.
vey in which 89% of respondents reported at least one Despite the differences in survey methods, these stud-
musculoskeletal complaint, and 73% reported more than ies document an association between endoscopy and
one injury.2 These endoscopists devoted substantial time musculoskeletal complaints, and they show a consistently
to endoscopy, with a mean (SD) procedure time of 19.5 high prevalence of musculoskeletal complaints among

www.giejournal.org Volume 70, No. 1 : 2009 GASTROINTESTINAL ENDOSCOPY 147

Ergonomics and GI endoscopy Shergill et al

endoscopists. Surgery and time off work was required in tip during examination. A recent case report describes
a minority of respondents, although it is still a concerning ‘‘colonoscopist’s thumb,’’ a tenosynovitis of the left thumb
finding. Men predominate in these surveys; therefore, the attributed to the repeated application of force by the
prevalence and site of injury in women is not as clear. left thumb during colonoscopy,11 previously described as
‘‘endoscopists’ thumb.’’9,17 Although the majority of the
Applying ergonomics to gastroenterology: risk force is applied by the left thumb, the digits of the left
factors and potential mechanisms of injury hand may provide additional force if the endoscopist
Ergonomic evaluations involve understanding how has a large enough hand. Coupled with up-down tip de-
a job affects people, both physically and emotionally, flection by the left hand, movement of the endoscope
with the goal of increasing worker safety, productivity, tip also can be achieved by application of torque to the
and job satisfaction.25 It is thought that overuse injuries insertion tube by the right hand and arm. In addition to
occur from repeated microtrauma to a tendon, ligament, the torque and push-pull forces applied by the right
or joint, or repeated ischemia to peripheral nerves. The hand, the right hand is also important for threading tools
repeated loading, injury, and repair may lead to degenera- through the biopsy channel. One prior report describes
tive changes over time.26-28 These degenerative changes ‘‘biliary endoscopist’s knuckle,’’ a traumatic arthritis of
are in contrast to acute injuries from tripping, falling, or the metacarpophalangeal joint attributed to repeated
other sudden high loads. high-force gripping to push large-caliber prostheses
The risk factors associated with overuse injuries of the through biliary and pancreatic strictures.17
upper extremities include high pinch force, repetitive A survey of U.S. gastroenterology fellows asked respon-
hand activities, awkward postures, vibration, and contact dents their impression of how their hand size (median sur-
stress.27,28 GI endoscopy involves several of these fac- gical glove size 7.5) affected their capacity to learn
tors.22 Gastroenterologists perform an average of 12 endoscopy.18 The response rate was 17.5% (227 fellows).
EGDs and 22 colonoscopies per week,29 resulting in Forty-one percent considered their hand too small for
frequent hand-intensive activities throughout the day. a standard endoscope’s control section and 62% thought
The few studies that directly evaluated hand forces during their hand size impaired their ability to perform endos-
endoscopy demonstrated high forces because of gripping copy. Sixty-three percent of respondents with a surgical
or pinching the endoscope, and pushing-pulling and glove size !6.5 (mostly women) would opt for an endo-
torquing of the insertion tube.10,12,21 There may also be scope with a smaller handle if available compared with
sustained awkward postures during endoscopy, including 28.3% of fellows with a larger glove size (P % .001).18
extension or rotation of the spine if the video monitor is No consideration to date has been given in endoscope de-
placed too high or to the side of the endoscopist.14,19 In sign to accommodate variability in hand size or the differ-
a survey of endoscopists who perform ERCP, 67% of rooms ence in hand sizes between men and women. A pilot study
reviewed by a study kinesiologist and occupational thera- evaluated the use of an angulation dial adapter (Olympus
pist were thought to be poorly designed because of awk- America, Center Valley, Pa) for hand spans (defined as the
ward placement of the monitor or use of nonadjustable thumb to the fifth digit)!19 cm.23 Although no significant
patient beds, and ergonomic room design was signifi- difference was found in procedure time or ease of proce-
cantly associated with no complaint of injury.7 Awkward dure, retroflexion was rated significantly easier with the
postures may also be related to endoscope manipulation. adapter by all endoscopists. The investigators concluded
During endoscopy, the left hand grips and stabilizes the that further evaluation of endoscope design may reduce
control section, the left thumb manipulates the control di- hand fatigue and injury, given the angulation dial showed
als, the right hand pinches or grips the insertion tube, and a trend toward decreased procedure time in physicians
the right arm pushes, pulls, and applies torque to the en- with small hands.23
doscope. These activities may require extreme or pro- The application of basic ergonomic principals, such as
longed wrist flexion or extension and/or radial or ulnar maintaining neutral wrist, neck, and shoulder postures
deviation during endoscopy that can decrease pinch and during endoscopy, keeping hand forces low, and optimiz-
grip strength.30,31 High finger forces in association with ing endoscope design, may reduce the risk of injury and
awkward wrist postures can further increase the risk of make endoscopy more comfortable for the gastroenterol-
overuse injury. ogist, meriting further investigation.
Human engineering is another important aspect of er-
gonomics.25 The interaction between the operator and
their tools, workstation, and tasks should be optimized Measurement of forces during endoscopy
to reduce overuse injury. Although the optics of endo- Limited data are available on the forces exerted during
scopes continue to improve, the basic shape and design endoscopy. Four articles, describing 3 studies, were in-
of the instrument have not changed for over 20 years. Cur- cluded for review.10,12,20,21 All articles specifically evaluated
rent endoscope design requires mechanical application colonoscopy. However, these studies differed in their mea-
of force to the control dials to deflect the insertion-tube surement tools, types of forces being evaluated, and use of

148 GASTROINTESTINAL ENDOSCOPY Volume 70, No. 1 : 2009 www.giejournal.org

Shergill et al Ergonomics and GI endoscopy

in vivo versus in vitro colonoscopy, precluding pooling of tivity level action limit, ie, a level at which general controls
data. or surveillance are warranted according to the ACGIH.
One study assessed forces applied by the right hand Peak muscle activity in the left-wrist extensors, which sta-
during tube insertion as a proxy for forces applied to bilize the instrument-control section, approached the AC-
the colon wall and the risk of bowel perforation.10,20 Per- GIH hand-activity–level threshold limit, a level at which
foration occurs at direct colonic-wall forces greater than redesign of the task is recommended by the ACGIH to re-
53.9 N. The investigators designed a tubular hand grip duce hand force and the risk of overuse injury.21 Further
that wrapped around the insertion tube, which contained study is needed to confirm these findings in a larger
strain gauges to measure force and torque. The mean cohort of endoscopists.
peak pushing force was 17.9 N when the tip was in the sig- In summary, the studies evaluating the forces exerted
moid colon and at the hepatic and splenic flexures; how- during colonoscopy vary in the measurement tool used
ever, forces greater than 10 N occurred only 5% of the and the types of forces recorded. However, these studies
total time.10 The peak torque forces occurred when the document potentially high peak forces during colonoscopy
tip was in the sigmoid colon and during shortening of that may reach levels associated with an increased risk of
the instrument.10 musculoskeletal injury of the thumb and wrist. Mean forces
A second study evaluated the forces applied during co- throughout colonoscopy appear low. The greatest loads
lonoscopy in an in vitro model by using a transducer occur during insertion, especially into the sigmoid colon,
sheath that was integrated into 50 cm of a colonoscope in- presumably because of looping. Only one study evaluated
sertion tube, proximal to the distal 11-cm flexible tip.12 A EMG activity of the distal upper extremity during colono-
3-dimensional electromagnetic image-sensing catheter scopy, and loads of the left-wrist extensors, the right-wrist
was placed into the biopsy channel to evaluate looping extensors, and the left-thumb extensors approached levels
during the experiment. The outfitted colonoscope was at which general controls or surveillance was warranted.
used by an experienced endoscopist during a series of co-
lonoscopies limited to the left side of the colon in a model Optimizing the ergonomics of endoscopy
that this team of researchers previously used and validated No guidelines or published endoscopic studies were
for measuring torque forces of the right hand.32 Right- identified that evaluated optimal room setup, monitor
hand grip forces were also recorded. Grip forces steadily height, or bed height during endoscopy. Therefore, the
increased as the colonoscope was inserted into the following recommendations for endoscopy are generally
model’s sigmoid colon (maximum 9.3 N). The maximum extrapolated from the laparoscopic surgery and general
force measured along sensors in the insertion tube was ergonomics literature.
12.7 N, with an average force of 0.28 N. Forces on the co- Monitor location and height. The main determi-
lonoscope appeared greatest during looping in the sig- nants of upper-body postures are the location of the pa-
moid colon. The forces measured on the colonoscope tient, placement of the endoscopy equipment, and
insertion tube were less than the grip forces applied exter- location of the monitor. The monitor placement is an es-
nally by the right hand.12 pecially important determinant of torso and head posture.
A recently completed study addressed more directly Optimal monitor position was studied in the surgical lap-
whether the musculoskeletal load during colonoscopy ap- aroscopy literature. Monitors should be placed directly in
proached levels associated with risk of overuse injuries.21 front of the endoscopist while in the position of ‘‘work,’’
Right-thumb pinch forces and right and left forearm mus- to avoid rotation and flexion of the cervical spine and
cle electromyography (EMG) activity was evaluated in 3 should be adjustable to eye level, as has been anecdotally
male endoscopists during 9 colonoscopies. Peak pinch recommended.16,22 One study of simulated laparoscopic
forces of the right hand during insertion into the left suturing evaluated subjective and EMG measurements in
and right colon exceeded 10 N, a level that is associated 3 different monitor positions: in front at eye level, in front
with increased risk of musculoskeletal injury of the thumb at the height of the operating field, and 45 degrees to the
and wrist.33,34 Peak muscle activity in the forearm muscles right at eye level. The neck and shoulder muscle activity
also exceeded safety thresholds established by the Ameri- was significantly lower when the monitor was positioned
can Conference of Industrial Hygienists (ACGIH).35 The in front at eye level,36 a position in which the cervical
ACGIH hand activity level is an ergonomic risk-assessment spine is in a neutral posture, with little lateral spine
tool that compares hand activity and applied force, as rotation. The findings were confirmed in a second laparos-
a percentage of the maximal strength of the individual copy study.37 An unpublished study of gastroenterologists
subject, to established thresholds above which a risk of in- in Lisbon, Portugal19 evaluated endoscopists’ postures and
jury exists. Activity of the left-wrist extensors (active in perceived level of discomfort while performing EGD and
holding the control section of the colonoscope), the colonoscopy within different room configurations. During
right-wrist extensors (active in torquing the insertion EGD, when the monitor was at the head of the patient’s
tube), and the left-thumb extensors (active in manipulat- bed, endoscopists were in cervical right rotation 77% of
ing the colonoscope dials), exceeded the ACGIH hand ac- the time and cervical extension 29% of time. When the

www.giejournal.org Volume 70, No. 1 : 2009 GASTROINTESTINAL ENDOSCOPY 149

Ergonomics and GI endoscopy Shergill et al

monitor was placed directly in front of the endoscopist at clarity, and endoscopist’s preference.42 Larger monitors
a fixed height during colonoscopy, neutral cervical posture can be placed at even longer viewing distances.
was achieved 93% and cervical extension 25% of the time. Overall, these findings suggest that monitor height
Overall, perceived discomfort during both procedures was should be adjustable so that it can accommodate the
low.19 Extrapolating from these studies, it appears that height of the endoscopist and the preferred viewing dis-
placement of the monitor directly in front of the endo- tance. Anthropometric data25 can be applied to determine
scopist achieves a neutral neck posture and minimizes what degree of adjustability is required of a monitor in
right and left cervical rotation. This position is likely asso- a typical endoscopy suite within the constraints of these
ciated with less discomfort and with optimal performance, viewing angles and monitor distances. To accommodate
although this has not been systematically studied among the 5th percentile female to the 95th percentile male
endoscopists. eye height, the monitor should be adjustable such that
Several studies evaluated optimal monitor height. A the center of the monitor can be adjusted between 93
study of 8 laparoscopic surgeons found that the favored to 162 cm above the floor.
monitor position, when fixed at a distance of 120 cm, Optimal bed height. In a study of surgeons simulat-
was achieved by placing the middle of the screen at ap- ing laparoscopic tasks, the optimal operating table height
proximately 20 cm lower than the height of the surgeon.38 was a compromise between spine and arm position.43 The
A randomized study of endoscopists, which compared optimal height was between elbow height and 10 cm be-
a video headset versus a conventional video monitor, pro- low elbow height. If physicians had to flex their trunks
vides additional insight into optimal monitor height.14 Im- during a task to accommodate a lower table height, then
age quality and comfort level were assessed with the 2 there was an increase in discomfort with higher difficulty
image display systems in 5 endoscopists during 96 colo- ratings. At higher bed heights, increased deltoid and tra-
noscopies. The video monitor was mounted from the ceil- pezius EMG muscle activity was seen because of shoulder
ing at a fixed height, with its center 200 cm above the abduction.43 Another study, of laparoscopic surgeons,
floor. The mean height of the endoscopists was 173  5 found that optimal table height was 70% to 80% of elbow
cm. The headset was associated with reduced neck strain, height, which allowed the joints to stay in a neutral pos-
although the video monitor had superior video image ture over 90% of the procedure time.44 By extrapolating
quality and was associated with improved comfort for all from laparoscopy to endoscopy, we believe that the en-
other parameters measured, including blurry vision and doscopy gurney should be adjusted to allow holding of
eye strain.14 The laparoscopy study cited above38 suggests the endoscope between elbow height and 10 cm below el-
that the optimal viewing height would be at eye level or bow height to minimize forward flexion of the back and
lower (ie, !173 cm), indicating that the monitor was shoulder abduction. To accommodate the elbow height
mounted too high for the endoscopists in this study. of the 5th percentile female to the 95th percentile
High monitor placement likely led to excessive cervical ex- male,25 the examination bed should be adjustable be-
tension and neck strain.38,39 tween 85 and 120 cm.
Optimal monitor height also depends on the distance of
the monitor from the endoscopist. The optimal monitor Procedural specific recommendations
distance can be extrapolated from the computer and lapa- EGD. EGDs generally are short procedures, with
roscopic literature. The optimal viewing angle for computer a mean (SD) procedure time of 14  6 minutes.45 In
monitor use is 15 to 25 degrees below the horizon of the this short time frame, it is unlikely that there will be sig-
eyes.25 In a study of the effects of visual display distance nificant prolonged loads on the endoscopist. Thus, the
on eye accommodation, head posture, and vision and most important factor is maintenance of neutral body
neck symptoms in computer users, the optimal distance postures.
was between near (52 cm) and middle (73 cm) distance Colonoscopy. The mean (SD) times for a diagnostic
from the eyes (taking into account screen-font size and colonoscopy without or with biopsy are 23.4  11.1 min-
the subject visual acuity).40 A study of optimal viewing dis- utes and 26.4  12.5 minutes, respectively.45 Colonoscopy
tance among 14 surgeons during simulated laparoscopic can lead to overuse injury because of repetition, high forces
surgery that used a 34-cm cathode-ray–tube monitor (14- while maneuvering the colonoscope tip, and prolonged
inch diagonal) concluded that the optimal distance was be- awkward postures. The left-wrist extensors (which stabilize
tween 90 and 182 cm for close-up viewing (defined as the the instrument control section) appear to be at greatest risk
minimum view distance below which image degradation for injury.21 In a case study, ‘‘colonoscopist’s thumb’’ or
was experienced) and 139 to 303 cm for maximal distance DeQuervain’s tenosynovitis (wrist tendonitis at the first ex-
viewing (defined as the maximal distance at which the finest tensor compartment) was attributed to left-thumb strain
details of an image could still be seen).41 Taken together, we because of repeated turning of the dials of the control sec-
estimate that the optimal viewing distance of video moni- tion of the colonoscope.11 At this time, there are limited op-
tors during endoscopy is between 52 and 182 cm. This is tions to reduce hand or wrist loads during endoscopy. If
a broad range, which will depend on monitor size, image a procedure requires additional use of the right-left dial

150 GASTROINTESTINAL ENDOSCOPY Volume 70, No. 1 : 2009 www.giejournal.org

Shergill et al Ergonomics and GI endoscopy

to negotiate a turn, then a technique called the ‘‘left hand

shaft grip’’ has been proposed.15 This technique consists
of holding the control section with the left thumb, index
finger, and middle finger while simultaneously supporting
the insertion tube with the left-hand fourth and fifth fin-
gers, so that the right hand is free to assist in turning the
right-left dials during sharp turns or for added precision
during endoscopic therapy. A variation of this technique,
the ‘‘pinkie maneuver,’’13 also has been described. A 2-
handed technique has been postulated to reduce the risk
of left-thumb injury but has not been formally studied
and, in general, is not thought to be as efficient as the
one-handed technique to control the dials.22 In an attempt
to minimize the risk associated with the repetition of colo-
noscopy, it is important to be aware of hand and wrist pos-
tures during endoscopy and to recognize the potential for
fatigue in the hand and forearm muscles. Recovery time is
important. A break from endoscopy allows the heavily
used muscle groups to recover.25 This ‘‘break’’ can be the
time used to complete an endoscopy report. Neither 2-per-
son colonoscopy nor seated colonoscopy has been studied.
Changes in colonoscope design, such as the use of a joy-
stick, trackballs, or pushbutton devices, offer theoretical er-
gonomic advantages. New endoscopic devices are under Figure 1. Recommended endoscopy suite layout with adjustable height
development,46 although their ergonomic impact has not patient bed and adjustable monitor. The monitor should be positioned
directly in front of the endoscopist while endoscopy is performed. Mon-
been formally studied. itor height should be just below eye level, with an optimal viewing angle
ERCP. In addition to the factors discussed above per- of 15-25 degrees below the horizon from the eyes to the center of the
taining to upper endoscopy and colonoscopy, ERCP uses monitor and an optimal viewing distance between 52 and 182 cm. To ac-
an elevator, which increases the left-thumb load. In addi- commodate the 5th percentile female to the 95th percentile male height,
tion, ERCP involves the static load of the lead apron. the monitor height should be adjustable from 93 to 162 cm. The exami-
nation table should be at or below elbow height (0-10 cm below the el-
Lead aprons can weigh up to 6.8 to 9.1 kg (15-20 pounds) bow). To accommodate the 5th percentile female to the 95th percentile
but can apply a load of approximately 2068 kPa (300 male elbow height, the examination table height should be adjustable
pounds per square inch) on the intervertebral disk from 85 to 120 cm.
spaces.7 Eleven percent of 237 endoscopists who partici-
pated in an online ASGE survey reported problems with 0.9 kg), and produce increased static loads on the left
lead aprons.4 In a survey of interventional cardiologists, hand and wrist compared with electronic echoendoscopes
approximately 42% of 424 respondents reported spinal (weight of control section approximately 0.5 kg). Endo-
problems.47 In a case-control study of interventional cardi- scopists should preferentially use the electronic echoen-
ologists (n Z 385), orthopedic surgeons (n Z 131), and doscopes whenever possible.
rheumatologists (n Z 198), significantly more interven-
tional cardiologists (7%) reported cervical disk disease
than orthopedists (0.3%) or rheumatologists (0%).48 Prior Personal factors
studies documented an increased risk of acute prolapsed Use of corrective lenses. Our visual ability to accom-
cervical intervertebral disk associated with frequent lifting modate decreases as we age, resulting in presbyopia. This
of 11.3 kg (25 pounds),48 and it can be argued that the leads to the use of bifocal or progressive addition lenses
weight of a lead apron coupled with cervical disk exten- (PAL) that allow the user to correct both near and far vi-
sion caused by improper monitor placement can increase sual acuity with one pair of eyeglasses. Endoscopists
cervical intradisc pressure39 and risk of injury.48 Although who wear bifocal or PAL lenses during endoscopy may
this issue has not been systematically studied in ERCP, need to extend their neck to see through the bottom of
2-piece lead shields may more evenly distribute the load the lenses, depending upon their distance from the mon-
across the spine and pelvis. As noted, it is important to itor. If the monitor is positioned too high, this can lead to
place the endoscopic and fluoroscopic monitors in front prolonged and excessive neck extension. To prevent this,
of the endoscopist and at the proper height to avoid endoscopists should consider the use of dedicated mono-
neck extension or lateral rotation.7 focal lenses or so-called ‘‘endoscopy lenses,’’ which func-
EUS. Mechanical echoendoscopes are much heavier tion much like ‘‘computer glasses’’ to provide the needed
(the weight of the control section is approximately correction for monitor viewing at a comfortable distance

www.giejournal.org Volume 70, No. 1 : 2009 GASTROINTESTINAL ENDOSCOPY 151

Ergonomics and GI endoscopy Shergill et al

in order to maintain a neutral neck and resting-eye posi- 8. American Society for Gastrointestinal Endoscopy. Media Back-
tion. For endoscopists who wish to use bifocals or PAL grounder, 2008. Available at: http://www.asge.org/PressroomIndex.
aspx?id=544. Accessed October 5, 2008.
lenses, the height of the monitor should be lowered to 9. Calebaugh DL. Afflicted endoscopists [letter]. Gastrointest Endosc
prevent neck extension. 1990;36:633.
10. Appleyard MN, Mosse CA, Mills TN, et al. The measurement of forces
Summary exerted during colonoscopy. Gastrointest Endosc 2000;52:237-40.
11. Cappell MS. Colonoscopist’s thumb: DeQuervains’s syndrome (teno-
In summary, the prevalence of musculoskeletal com-
synovitis of the left thumb) associated with overuse during endos-
plaints has been shown to be higher for endoscopists copy. Gastrointest Endosc 2006;64:841-3.
than for other medical specialties, with a range of 37% 12. Dogramadzi S, Virk GS, Bell GD, et al. Recording forces exerted on the
to 89%. GI endoscopy is associated with several risk fac- bowel wall during colonoscopy: in vitro evaluation. Int J Med Robot
tors for overuse injury: repetitive hand motion, high 2005;1:89-97.
13. Guelrud M. Improving control of the colonoscope: the ‘‘pinkie maneu-
hand forces, and awkward wrist, shoulder, and neck pos-
ver’’ [letter]. Gastrointest Endosc 2008;67:388-9:author reply 389.
tures. Further study is needed to determine how worksta- 14. Lee LS, Carr-Locke DL, Ookubo R, et al. Randomized trial of a video
tion design and tool design changes can reduce these headset vs. a conventional video monitor during colonoscopy. Gastro-
injuries. To optimize the ergonomics of endoscopy, we intest Endosc 2005;61:301-6.
recommend the following: 15. Rex DK. Maximizing control of tip deflection with sound ergonomics:
the ‘‘left hand shaft grip’’ [letter]. Gastrointest Endosc 2007;65:950-1:
d The endoscopy suite should be set up with the monitor
author reply 951.
positioned directly in front of the endoscopist while en- 16. Riemann JF. Ergonomics in gastrointestinal endoscopy: what do we
doscopy is performed (Fig. 1). need for the future? Endoscopy 1993;25:369-70.
d Monitor height should be just below eye level, with an 17. Siegel JH, Kasmin EE, Cohen SA. Health hazards and endoscopy: the
optimal viewing angle of 15 to 25 degrees below the ho- known and newly experienced: a personal report. Endoscopy 1994;26:
545-8.
rizon from the eyes, with a viewing distance of 52 to 182
18. Cohen DL, Naik JR, Tamariz LJ, et al. The perception of gastroenterol-
cm, depending on monitor size and endoscopist prefer- ogy fellows towards the relationship between hand size and endo-
ence. To accommodate the 5th percentile female to the scopic training. Dig Dis Sci 2008;53:1902-9.
95th percentile male eye height, the monitor height 19. Cotrim T, Francisco C, Freitas JL, et al. Ergonomic analysis of postural
should be adjustable from 93 to 162 cm. workload during endoscopies, 2002. Available at: http://web.wits.ac.
za/NR/rdonlyres/8E097A19-845E-40D9-A0BE-3E110EE2F321/0/med2.pdf.
d The examination table should be at or below elbow
Accessed October 5, 2008.
height (0-10 cm below the elbow). To accommodate 20. Mosse CA, Mills TN, Bell GD, et al. Device for measuring the forces ex-
the 5th percentile female to the 95th percentile male el- erted on the shaft of an endoscope during colonoscopy. Med Biol Eng
bow height, the examination table height should be ad- Comput 1998;36:186-90.
justable from 85 to 120 cm. 21. Shergill AK, Asundi KR, Barr A, et al. Pinch force and forearm muscle
load during colonoscopy: a pilot study. Gastrointest Endosc 2009;69:
d During ERCP, a 2-piece apron should be used to reduce
142-6.
loads on the upper back and cervical spine. 22. Siegel J. Risk of repetitive-use syndromes and musculoskeletal injuries.
d Electronic EUS scopes should be used in favor of the Tech Gastrointest Endosc 2007;9:200-4.
mechanical EUS scopes whenever possible. 23. Akerkar GM, McQuaid KR, Terdiman JP, et al. An angulation dial
d Recovery time between endoscopies is important. A adapter to facilitate endoscopy [abstract]. Gastrointest Endosc 1999;
49:AB120.
break from endoscopy allows the muscle groups to
24. Cheadle A, Franklin G, Wolfhagen C, et al. Factors influencing the dura-
rest and ‘‘pay back’’ some of the lactic-acid debt that tion of work-related disability: a population-based study of Washington
was incurred during a work cycle of endoscopy. State workers’ compensation. Am J Public Health 1994;84:190-6.
25. Chengular SR, Rodgers SH, Bernard TE, editors. Kodak’s ergonomic de-
sign for people at work, 2nd ed Hoboken: Wiley and Sons; 2004.
REFERENCES 26. Diao E, Shao F, Liebenberg E, et al. Carpal tunnel pressure alters me-
dian nerve function in a dose-dependent manner: a rabbit model for
1. Buschbacher R. Overuse syndromes among endoscopists. Endoscopy carpal tunnel syndrome. J Orthop Res 2005;23:218-23.
1994;26:539-44. 27. Rempel DM, Harrison RJ, Barnhart S. Work-related cumulative trauma
2. Byun YH, Lee JH, Park MK, et al. Procedure-related musculoskeletal disorders of the upper extremity. JAMA 1992;267:838-42.
symptoms in gastrointestinal endoscopists in Korea. World J Gastroen- 28. Uhl TL, Madaleno JA. Rehabilitation concepts and supportive devices
terol 2008;14:4359-64. for overuse injuries of the upper extremities. Clin Sports Med 2001;20:
3. Hansel SL, Di Baise JK, Crowell MD, et al. Prevalence and impact of 621-39.
musculoskeletal injury among endoscopists: a controlled study 29. Cohen LB, Wecsler JS, Gaetano JN, et al. Endoscopic sedation in the
[abstract]. Gastrointest Endosc 2007;65:AB104. United States: results from a nationwide survey. Am J Gastroenterol
4. Keate R. Occupational injuries to endoscopists: report from the ASGE 2006;101:967-74.
Web survey [abstract]. Gastrointest Endosc 2006;63:AB111. 30. Halpern CA, Fernandez JE. The effect of wrist and arm postures on
5. Lee T, Valiozis I. Occupational related injury in endoscopists [abstract]. peak pinch strength. J Hum Ergol (Tokyo) 1996;25:115-30.
J Gastroenterol Hepatol 2007;22:A338. 31. Mogk JP, Keir PJ. The effects of posture on forearm muscle loading
6. Liberman AS, Shrier I, Gordon PH. Injuries sustained by colorectal sur- during gripping. Ergonomics 2003;46:956-75.
geons performing colonoscopy. Surg Endosc 2005;19:1606-9. 32. Hancock J, Nylander D, Abro H, et al. Basic colonoscopy: Hoken colon
7. O’Sullivan S, Bridge G, Ponich T. Musculoskeletal injuries among ERCP model or computer simulator for teaching torque steering skills
endoscopists in Canada. Can J Gastroenterol 2002;16:369-74. [abstract]? Gut 2004;53:A48.

152 GASTROINTESTINAL ENDOSCOPY Volume 70, No. 1 : 2009 www.giejournal.org

Shergill et al Ergonomics and GI endoscopy

33. Roquelaure Y, Mechali S, Dano C, et al. Occupational and personal risk 44. van Veelen MA, Kazemier G, Koopman J, et al. Assessment of the
factors for carpal tunnel syndrome in industrial workers. Scand J Work ergonomically optimal operating surface height for laparoscopic
Environ Health 1997;23:364-9. surgery. J Laparoendosc Adv Surg Tech A 2002;12:47-52.
34. Silverstein BA, Fine LJ, Armstrong TJ. Hand wrist cumulative trauma 45. McCashland T, Brand R, Lyden E, et al. The time and financial impact of
disorders in industry. Br J Ind Med 1986;43:779-84. training fellows in endoscopy. CORI Research Project. Clinical Out-
35. American Conference of Governmental Industrial Hygienists (ACGIH). comes Research Initiative. Am J Gastroenterol 2000;95:3129-32.
ACGIH 2008 TLVs and BEIs, based on the documentation of the thresh- 46. Rösch T, Eickhoff A, Fritscher-Ravens A, et al. The new scopes: broad-
old limit values for chemical substances and physical agents. Cincin- ening the colonoscopy marketplace. Digestion 2007;76:42-50.
nati: ACGIH; 2008. p. 189. 47. Goldstein JA, Balter S, Cowley M, et al. Occupational hazards of inter-
36. Matern U, Faist M, Kehl K, et al. Monitor position in laparoscopic sur- ventional cardiologists: prevalence of orthopedic health problems in
gery. Surg Endosc 2005;19:436-40. contemporary practice. Catheter Cardiovasc Interv 2004;63:407-11.
37. Haveran LA, Novitsky YW, Czerniach DR, et al. Optimizing laparoscopic 48. Ross AM, Segal J, Borenstein D, et al. Prevalence of spinal disc disease
task efficiency: the role of camera and monitor positions. Surg Endosc among interventional cardiologists. Am J Cardiol 1997;79:68-70.
2007;21:980-4.
38. Zehetner J, Kaltenbacher A, Wayand W, et al. Screen height as an er-
gonomic factor in laparoscopic surgery. Surg Endosc 2006;20:139-41.
39. Pospiech J, Stolke D, Wilke HJ, et al. Intradiscal pressure recordings in Received October 31, 2008. Accepted December 22, 2008.
the cervical spine. Neurosurgery 1999;44:379-84.
Current affiliations: Department of Medicine, Division of Gastroenterology
40. Rempel D, Willms K, Anshel J, et al. The effects of visual display dis-
(A.K.S., K.R.M.), University of California, San Francisco, California, Medical
tance on eye accommodation, head posture, and vision and neck
Service, Gastroenterology Section (A.K.S., K.R.M.), San Francisco Veterans
symptoms. Hum Factors 2007;49:830-8.
Affairs Medical Center, San Francisco, California, Department of Medicine,
41. El Shallaly G, Cuschieri A. Optimum view distance for laparoscopic sur-
Division of Occupational Medicine (D.R.), University of California, San
gery. Surg Endosc 2006;20:1879-82.
Francisco, San Francisco, California, Department of Bioengineering (D.R.),
42. Cuschieri A. Epistemology of visual imaging in endoscopic surgery.
University of California, Berkeley, Berkeley, California, USA.
Surg Endosc 2006;20(Suppl 2):S419-24.
43. Berquer R, Smith WD, Davis S. An ergonomic study of the optimum Reprint requests: Amandeep Shergill, MD, GI Section, San Francisco VA
operating table height for laparoscopic surgery. Surg Endosc 2002; Medical Center, 4150 Clement St, Bldg 203 (111B1), San Francisco, CA
16:416-21. 94121.

www.giejournal.org Volume 70, No. 1 : 2009 GASTROINTESTINAL ENDOSCOPY 153