Quantifying and communicating peri‐operative risk


Title Quantifying and communicating peri-operative risk

Author(s) Irwin, MG; Kong, VKF

Citation Anaesthesia (Oxford), 2014, v. 69 n. 12, p. 1299-1303

Issued Date 2014

URL http://hdl.handle.net/10722/215071

Rights Creative Commons: Attribution 3.0 Hong Kong License



Editorial

Quantifying and communicating peri-operative risk

Shallow men believe in luck.

Strong men believe in cause

and effect

—Ralph Waldo Emerson

Life is risky
Risk is the potential that a chosen

action or activity (including the

choice of inaction) will lead to a spe-

cific outcome, and implies that the

choice has an influence on the out-

come. Most definitions are synony-

mous with the possibility of an

adverse event but, of course, a risk

can also be taken in the hope of a

favourable outcome, particularly with

investment. There is also a personal

perspective on risk. A fatalist person-

ality may be very accepting and

unconcerned about risk whereas

more pragmatic individuals know

that there can be modifiable factors

involved. This can apply to health-

care; for example, even though sur-

gery may be necessary in a patient,

there may be pharmaceutical inter-

ventions that could reduce morbidity.

Almost any human endeavour carries

some risk. Staying in hospital is far

riskier than travelling by aeroplane. A

recent study showed that a one-night

stay in hospital carries a 11.1% risk of

nosocomial infection, a 3.4% risk of

an adverse drug reaction related to

human error or allergy, and a 0.4%

risk of pressure ulcer due to immobil-

isation [1]. In 2007 in the USA, there

were 1.31 fatal crashes per 100 000

flight hours for non-commercial

flights and 0.016 per 100 000 for

major airlines [2]. Despite efforts to

the contrary, healthcare is an intrinsi-

cally hazardous business.

Anaesthesia is a medical spe-

cialty very much focused on risk

management and patient safety and,

consequently, the mortality risk

attributable to anaesthesia itself has

dropped dramatically over the years,

from about one death in 1000

anaesthetic procedures in the 1940s

to one in 100 000 in the early 2000s

[3]. However, although anaesthesia

is relatively safe, surgery can be very

dangerous. In 2000, the 30-day mor-

tality risk in the UK was one death

in 34 emergency operations (2.9%)

and 1:177 after elective surgery

(0.6%) [4]. The European Surgical

Outcomes Study was an observa-

tional study in which data were col-

lected on 46 539 patients aged ≥
16 years undergoing non-cardiac

surgery, over a seven-day period, in

498 hospitals across 28 European

nations [5]. There was considerable

variability from country to country

but median death rates were 3% for

elective and 10% for emergency sur-

gery. Anaesthesia has an excellent

track record for patient safety and

has been described as the leading

medical specialty in addressing such

issues [6], yet it is apparent that the

peri-operative process still has great

potential for hazard from a host of

factors, of which anaesthesia is but

one.

Ronald A. Howard, a pioneer

of decision analysis, wanted to

develop a scale that would more

clearly confer risk rather than per-

centages. He coined the term

‘microprobability’ to refer to an

event with a chance of one in a

million. From this concept, a ‘micro-

mort’ (from ‘micro’ and ‘mortality’)

is then a one in a million chance of

death [7]. We face risk simply by

being alive and this may be exacer-

bated by indulging in various activi-

ties. A mobile app is now available

for illustrating how many micro-

morts are involved in our daily activ-

ities (see https://play.google.com/

store/apps/details?id=com.zanzibar-

tech.micromorts). The use of micro-

morts then allows us actually to

quantify risk and translate it into

whole numbers. A micromort

denotes a one in a million chance of

death from one-time dangerous

events, a concept that can be easily

understood and compared. A one in

a million chance is, of course, rare

but also an everyday occurrence. For

example, the chance of a particular

individual’s winning the weekly

lottery is less than 1 in 32 million

© 2014 The Association of Anaesthetists of Great Britain and Ireland 1299

Anaesthesia 2014, 69, 1299–1313



people, but conversely this jackpot

gets won almost every week by

somebody. Micromorts are an

expression of acute risks, such that

once that event has been completed

the risk has gone. The risk of surgi-

cal anaesthesia is in the range of var-

ious day-to-day activities. Audits of

the risk of death from a general

anaesthetic alone vary considerably

geographically, but may be around

one death in 100 000 operations in a

developed country [8], which

equates to 10 micromorts per opera-

tion. This is the same risk of death,

on average, as riding a motorcycle

for 60 miles or skydiving. Micro-

morts rely on aggregated risk data

for calculations, so their applicability

to specific circumstances or individ-

uals is limited. In contrast with mi-

cromorts, there is also a unit called

microlife which is a risk (or gain)

representing a 30-minute change of

life expectancy [9]. It is a way of mea-

suring the impact of long-term hab-

its on the human body. For example,

smoking two cigarettes will ‘cost’ one

microlife; whereas bonus life can be

gained by taking a statin daily (one mi-

crolife per day) or doing 20 minutes of

moderate exercise daily (two microlives

per day). A user-friendly microlife cal-

culator (see http://journals.bmj.com/

site/microlives) is available. People in

general are notoriously bad at calculat-

ing risk. The concept of micromort and

microlife can be useful for explaining

various risks in our daily peri-operative

practice to the general public.

The blind leading the
blind?
Shared decision-making in the

healthcare context very often

depends on the understanding of

numerical information, in either text

or graphical format. The perception

of harm and benefit associated with

particular options is important for

many health decisions. Surprisingly,

not only patients but many doctors

have severe problems mastering a

host of numerical concepts that are

prerequisites for understanding

information about the harm and

benefit of medical treatments [10].

Highly educated people can still have

difficulty with relatively simple

numeracy questions [11]. Numeracy

influences the processing of both

numerical and non-numerical infor-

mation. Less numerate individuals

are more susceptible to framing

effects, more easily affected by non-

numerical information such as mood

states, and less sensitive to different

levels of numerical risk [12]. The

Berlin Numeracy Test is a new psy-

chometric instrument for assessing

statistical numeracy and risk literacy

in an educated population [13]. It

typically takes three minutes to com-

plete and an online version is now

also available (see http://www.risklit-

eracy.org).

Statisticians, clinicians and psy-

chologists have recommended the

use of numerical as opposed to ver-

bal descriptions for risk communica-

tion [14–16]. In addition to

probability information, the way

people perceive a risk message may

be influenced by the framing of risk

information, risk comparisons, the

message’s qualitative content and

trust [17]. Conveying relative risks

alone without absolute risk or base-

line risk is an example of non-trans-

parent framing. Comparing benefits

and harm using different scales, such

as reporting benefits in big numbers

by relative risk reduction and harm

in small numbers by absolute risk

increases, is another way of altering

risk perception. An example is the

1995 contraceptive pill scare in the

UK, where an alarming figure of a

100% increase (relative) in thrombo-

sis caused by third-generation oral

contraceptive pills was much more

terrifying than a humble increment

in absolute risk, from one in 7000 to

two in 7000 [18]. Humans can be

prone to unwittingly tricking them-

selves with representative bias in risk

assessment in gambling pursuits

such as purchasing lottery tickets. If

the odds of winning a lottery are one

in a million, then buying two tickets

will ‘double’ the chance to two in a

million (an apparent 100% increase).

However, the odds of not winning

the jackpot by buying two lottery

tickets hardly changes at all (from

99.9999% to 99.9998%; a change of

0.0001%).

Non-transparent and mis-

matched framings are common

phenomena, even for scientific

research published in leading medi-

cal journals. Studies have revealed

that up to half of articles report

only relative risks or odds ratios,

and about one third adopt mis-

matched framing for risk-benefit

discussion [19, 20]. Risk communi-

cation with incomplete and mis-

leading numerical descriptions

hinders shared decision-making.

Patients are likely to be familiar

with the concept of risk, but human

nature is such that many do not

understand the relativity or perhaps

even choose to ignore it. To use

gambling again as an example, a

recent $640 million lottery in the

USA created much excitement and

1300 © 2014 The Association of Anaesthetists of Great Britain and Ireland

Anaesthesia 2014, 69, 1299–1313 Editorial



a scramble to buy tickets when the

odds of winning were approxi-

mately one in 175 million. That

number may not mean much in

itself, but in relative terms the

chance of winning is 175 times less

than that of being struck by light-

ning in a given year, a fact that

helps conceptualise probability.

As in life, there is no zero risk

and no certainty in any branch of

medicine, but only risks that are

more or less acceptable. Communi-

cating risk information is important

but, unfortunately, more difficult

than might be expected. Patients’

values and preference are essential

elements of shared decision-making.

In 2011, a report on the peri-opera-

tive care of surgical patients pub-

lished by the National Confidential

Enquiry into Patient Outcome and

Death principally recommended

that “an assessment of mortality risk

should be made explicit to the

patient and recorded clearly on the

consent form and in the medical

record” [21]. A spreadsheet has

been developed to quantify mortal-

ity risk before and after surgery by

calculating mortality rates, and life

expectancy with adjustment for var-

ious parameters such as age, sex,

co-morbidities, renal function,

physical fitness, and body mass

index [22]. An on-line calculator is

available for estimating peri-opera-

tive mortality in order to assist

shared decision-making between

patients and their doctors when

non-surgical intervention is an

option (see https://sites.google.com/

site/informrisk/).

The complex nature of the

peri-operative period gives rise to

the potential for significant risk to

all patients. Directing efforts

towards patient safety can be

uncomplicated and inexpensive, yet

significantly improve the quality of

peri-operative care. The World

Health Organization’s surgical

safety checklist is an example of a

simple, cheap and effective method

of reducing avoidable complications

resulting from surgery [23]. More-

over, the use of a checklist is likely

to provide a net financial benefit to

the healthcare system because the

cost of the intervention is low

under all scenarios and there should

be a reduction in morbidity and

medicolegal claims. Anaesthesia is a

medical discipline of applied science

related to the art of peri-operative

risk reduction by identification,

intervention, and prevention.

To test or not to test
High-risk patients account for no

more than 15% of all surgical pro-

cedures but over 80% of deaths

[24]. However, it is still a major

challenge to identify accurately and

reliably patients who are at high

risk of postoperative mortality and

morbidity. Cardiopulmonary exer-

cise testing (CPET) is a measure of

aerobic capacity that is becoming

more widely employed, with the

estimated number of tests per-

formed in the UK alone estimated

to be in excess of 14 000 per year

[25]. Based on the possible associa-

tion of pre-operative aerobic fitness

with subsequent survival after

surgery, CPET has been used for

triaging patients with occult cardio-

respiratory disease for further

investigation and optimisation strat-

egies before major operations [26].

Nonetheless, none of the derived

variables, such as ventilatory anaer-

obic threshold or maximal oxygen

uptake, can be regarded as a cor-

nerstone for the prediction of sur-

vival after major surgery [27]. The

European Society of Cardiology’s

guidelines on pre-operative cardiac

risk assessment and peri-operative

cardiac management, published in

2009, questioned the role of CPET

in risk assessment before surgery

and emphasised that it is not a sub-

stitute for stress testing in routine

practice [28]. A scientific statement

from the American Heart Associa-

tion has also highlighted the lack of

randomised trials to support recom-

mendations for diagnostic and

prognostic applications of CPET

[29]. Reliance upon any single fac-

tor in predicting future risk is more

like gambling than rational assess-

ment, since no test is infallible. No

single variable can estimate the

extent of survival and quality of life,

although physical fitness appears to

be an important component [30,

31]. Pre-operative testing should

not be a screening exercise for sta-

ble patients but a strategic part of

the peri-operative risk reduction

programme for susceptible patients.

Risk assessment should always be

tailored to individual patients, and

pre-operative tests only reserved for

those in whom test results would

positively influence and change

peri-operative management. Our

drive to ‘optimise risk’ can lead to

unnecessary investigation and inter-

vention when we should really be

optimising ‘risk assessment’ through

more comprehensive history taking

and physical examination. Concerns

over malpractice liability result in

excessive and unnecessary consulta-

© 2014 The Association of Anaesthetists of Great Britain and Ireland 1301

Editorial Anaesthesia 2014, 69, 1299–1313



tion, hospitalisation, testing, and

treatment can, paradoxically, be a

safety hazard for both patients and

doctors, with false positive findings

leading to costly and possibly harm-

ful treatments or further investiga-

tions and delays in surgery. The

quantity of tests should not be con-

fused with the quality of care.

Are drugs the answer?
Is there a pharmacological panacea

for peri-operative risk optimisation?

One of the main objectives of risk

identification is to determine which

individuals could benefit from a

protective, therapeutic intervention.

If that intervention is potentially

dangerous in itself, e.g. coronary

artery stenting, then careful selec-

tion is absolutely essential. There

are, however, fairly safe and simple

pharmacological treatments that are

very promising in this regard. It

may not be necessary to investigate

patients aggressively with expensive

and even hazardous techniques if

the indication for surgery is very

strong. Why not assume the worst

and instigate protective measures

anyway? Over the last decade, sta-

tins have been investigated exten-

sively for their potential multimodal

effects in modifying a number of

aspects of peri-operative morbidity

and mortality [32–34]. Possible

pleiotropic effects of statin therapy

are reduction of myocardial infarc-

tion and stroke, prevention of atrial

fibrillation, improvement of vascu-

lar draft survival, protection from

renal insufficiency, and inhibition

of malignant cell growth [34]. All

peri-operative applications of statins

are ‘off-label’, as their primary indi-

cation is lipid-lowering, but the

drugs, which are now available in

generic form, are relatively inexpen-

sive with a very good safety profile

[35]. However, in the latest Cochra-

ne review of statins for vascular

surgery, there was insufficient evi-

dence to conclude that use of sta-

tins resulted in either a reduction

or an increase in any of the out-

comes examined. It was also

observed that the widespread use of

statins in the population now will

make it difficult for researchers to

undertake the large randomised tri-

als needed to demonstrate any

effect [36].

Peri-operative pharmacological

interventions, while having certain

benefits, may themselves be associ-

ated with risk that could outweigh

such advantages. Data from the

POISE trial suggested that routine

administration of peri-operative

beta-blockers in an un-titrated, rela-

tively high dose starting on the day

of operation, increased the risk of

stroke and overall mortality for

non-cardiac surgery in the presence

of favourable outcomes on other

cardiovascular parameters [37, 38].

POISE II has generated similar con-

troversy recently over the use of

peri-operative aspirin [39].

Conclusions
Anaesthesia, as a service-based spe-

ciality dealing with specific inci-

dent-related risks, has not yet

reached its peak despite being

acknowledged as the leading medi-

cal speciality in designing fail-secure

systems and probably the only spe-

cialty in healthcare to have reached

the critical target of six sigma defect

rate [6, 40]. The peri-operative

period could be made safer by

revolutionising risk management

tactics with corresponding interven-

tions in preventing complications

and improving patient outcomes. It

has been suggested that the techni-

cal aspects of anaesthesia could be

delegated to robots [41], so are we

ready for our new role as peri-

operative risk strategists?

Competing interests
No external funding and no com-

peting interests declared.

M. G. Irwin
Professor
V. K. F. Kong
Honorary Assistant Professor
Department of Anaesthesiology
University of Hong Kong,
Hong Kong
Email: mgirwin@hku.hk

References
1. Hauck K, Zhao X. How dangerous is a

day in hospital? a model of adverse
events and length of stay for medical
inpatients. Medical Care 2011; 49:
1068–75.

2. Li G, Baker SP. CRash risk in general
aviation. Journal of the American Medi-
cal Association 2007; 297: 1596–8.

3. Li G, Warner M, Lang BH, Huang L, Sun
LS. Epidemiology of anesthesia-related
mortality in the United States, 1999–
2005. Anesthesiology 2009; 110: 759–
65.

4. Jenkins K, Baker AB. Consent and
anaesthetic risk. Anaesthesia 2003; 58:
962–84.

5. Pearse RM, Moreno RP, Bauer P, et al.
Mortality after surgery in Europe: a
7 day cohort study. Lancet 2012; 380:
1059–65.

6. Leape LL. Error in medicine. Journal of
the American Medical Association
1994; 272: 1851–7.

7. Howard RA. Microrisks for medical
decision analysis. International Journal
of Technology Assessment in Health
Care 1989; 5: 357–70.

8. Aitkenhead AR. Injuries associated with
anaesthesia. A global perspective. Brit-
ish Journal of Anaesthesia 2005; 95:
95–109.

1302 © 2014 The Association of Anaesthetists of Great Britain and Ireland

Anaesthesia 2014, 69, 1299–1313 Editorial



9. Spiegelhalter D. Using speed of ageing
and microlives to communicate the
effects of lifetime habits and environ-
ment. British Medical Journal 2012;
345: e8223.

10. Galesic M, Garcia-Retamero R. Using
analogies to communicate information
about health risks. Applied Cognitive
Psychology 2013; 27: 33–42.

11. Lipkus IM, Samsa G, Rimer BK. General
performance on a numeracy scale
among highly educated samples. Medi-
cal Decision Making 2001; 21: 37–44.

12. Peters E. Beyond comprehension: the
role of numeracy in judgments and
decisions. Current Directions in Psycho-
logical Science 2012; 21: 31–5.

13. Cokely ET, Galesic M, Schulz E, Ghazal
S, Garcia-Retamero R. Measuring risk
literacy: the Berlin numeracy test.
Judgment and Decision Making 2012;
7: 25–47.

14. Smith A. Foreword to the papers on
’The communication of risk’. Journal of
the Royal Statistical Society: Series A
(Statistics in Society) 2003; 166: 205–6.

15. Edwards A. Communicating risks. Brit-
ish Journal of Anaesthesia 2003; 327:
691–2.

16. Gigerenzer G, Gaissmaier W, Kurz-Mil-
cke E, Schwartz LM, Woloshin S. Help-
ing doctors and patients make sense
of health statistics. Psychological Sci-
ence in the Public Interest 2007; 8:
53–96.

17. Visschers VHM, Meertens RM, Passchier
WWF, De Vries NNK. Probability infor-
mation in risk communication: a
review of the research literature. Risk
Analysis 2009; 29: 267–87.

18. Furedi A. The public health implications
of the 1995 ‘pill scare’. Human Repro-
duction Update 1999; 5: 621–6.

19. Schwartz LM, Woloshin S, Dvorin EL,
Welch HG. Ratio measures in leading
medical journals: structured review of
accessibility of underlying absolute
risks. British Medical Journal 2006;
333: 1248.

20. Sedrakyan A, Shih C. Improving depic-
tion of benefits and harms: analyses of
studies of well-known therapeutics and
review of high-impact medical journals.
Medical Care 2007; 45: S23–S8.

21. Findley G, Goodwin A, Protopapa K,
Smith N, Mason M. Knowing the risk: a
review of the peri-operative care of sur-
gical patients. London: NCEPOD, 2011.

22. Carlisle JB. Pre-operative co-morbidity
and postoperative survival in the
elderly: beyond one lunar orbit. Anaes-
thesia 2014; 69: 17–25.

23. Haynes AB, Weiser TG, Berry WR, et al.
A surgical safety checklist to reduce
morbidity and mortality in a global
population. New England Journal of
Medicine 2009; 360: 491–9.

24. Pearse R, Harrison D, James P, et al.
Identification and characterisation of
the high-risk surgical population in the
United Kingdom. Critical Care 2006; 10:
R81.

25. Huddart S, Young EL, Smith R-L, Holt
PJ, Prabhu PK. Preoperative cardiopul-
monary exercise testing in England–a
national survey. Perioperative Medicine
2013; 2: 4.

26. Kothmann E, Danjoux G, Owen SJ,
Parry A, Turley AJ, Batterham AM. Reli-
ability of the anaerobic threshold in
cardiopulmonary exercise testing of
patients with abdominal aortic aneu-
rysms. Anaesthesia 2009; 64: 9–13.

27. Colson M, Baglin J, Bolsin S, Grocott
MPW. Cardiopulmonary exercise testing
predicts 5 yr survival after major sur-
gery. British Journal of Anaesthesia
2012; 109: 735–41.

28. Poldermans D, Bax JJ, Boersma E, et al.
Guidelines for pre-operative cardiac
risk assessment and perioperative car-
diac management in non-cardiac sur-
gery: The Task Force for Preoperative
Cardiac Risk Assessment and Perioper-
ative Cardiac Management in Non-car-
diac Surgery of the European Society of
Cardiology (ESC) and endorsed by the
European Society of Anaesthesiology
(ESA). European Heart Journal 2009;
30: 2769–812.

29. Balady GJ, Arena R, Sietsema K, et al. Clini-
cian’s guide to cardiopulmonary exercise
testing in adults: a scientific statement
from the American Heart Association. Cir-
culation 2010; 122: 191–225.

30. Carlisle JB. The quick and the dead.
Anaesthesia 2013; 68: 799–803.

31. Lee L, Schwartzman K, Carli F, et al.
The association of the distance walked
in 6 min with pre-operative peak oxy-
gen consumption and complications
1 month after colorectal resection.
Anaesthesia 2013; 68: 811–6.

32. Biccard BM. A peri-operative statin
update for non-cardiac surgery. Part II:
Statin therapy for vascular surgery and

peri-operative statin trial design.
Anaesthesia 2008; 63: 162–71.

33. Biccard BM. A peri-operative statin
update for non-cardiac surgery. Part I:
The effects of statin therapy on athero-
sclerotic disease and lessons learnt
from statin therapy in medical (non-
surgical) patients. Anaesthesia 2008;
63: 52–64.

34. Chan WW, Wong GT, Irwin MG. Periop-
erative statin therapy. Expert Opinion
on Pharmacotherapy 2013; 14: 831–
42.

35. Minder CM, Blumenthal RS, Blaha MJ.
Statins for primary prevention of car-
diovascular disease: the benefits out-
weigh the risks. Current Opinion in
Cardiology 2013; 28: 554–60.

36. Sanders RD, Nicholson A, Lewis SR,
Smith AF, Alderson P. Perioperative sta-
tin therapy for improving outcomes
during and after noncardiac vascular
surgery. Cochrane Database of System-
atic Reviews 2013; 7: CD009971.

37. Devereaux P, Yang H, Yusuf S, et al.
Effects of extended-release metoprolol
succinate in patients undergoing non-
cardiac surgery (POISE trial): a rando-
mised controlled trial. Lancet 2008;
371: 1839–47.

38. Wijeysundera DN, Duncan D, Nkonde-
Price C, et al. Perioperative beta block-
ade in noncardiac surgery: a systematic
review for the 2014 ACC/AHA Guide-
line on Perioperative Cardiovascular
Evaluation and Management of
Patients Undergoing Noncardiac Sur-
gery: a report of the American College
of Cardiology/American Heart Associa-
tion Task Force on Practice Guidelines.
Journal of the American College of Car-
diology 2014 Aug 1; doi 10.1016/j.
jacc.2014.07.939.

39. Devereaux PJ, Mrkobrada M, Sessler DI,
et al. Aspirin in patients undergoing
noncardiac surgery. New England Jour-
nal of Medicine 2014; 370: 1494–503.

40. Amalberti R, Auroy Y, Berwick D, Bar-
ach P. Five system barriers to achieving
ultrasafe health care. Annals of Inter-
nal Medicine 2005; 142: 756–64.

41. Hemmerling TM, Terrasini N. Robotic
anesthesia: not the realm of science
fiction any more. Current Opinion in
Anesthesiology 2012; 25: 736–42.

doi:10.1111/anae.12881

© 2014 The Association of Anaesthetists of Great Britain and Ireland 1303

Editorial Anaesthesia 2014, 69, 1299–1313