BJA Education, 19(1): 1e6 (2019)

doi: 10.1016/j.bjae.2018.09.005
Advance Access Publication Date: 6 November 2018

Matrix codes: 1A01,

1A02, 2A03, 3A01

Smoking and anaesthesia

M.A. Carrick J.M. Robson and C. Thomas*
St James’s University Hospital, Leeds Teaching Hospitals NHS Trust, Leeds, UK
*Corresponding author: [email protected]

Learning objectives Key points

By reading this article, you should be able to:  Smoking increases the risk of perioperative
 Discuss the physiological effects of smoking. morbidity and mortality in a dose-dependent
 Discuss the effects of smoking on perioperative manner.
morbidity and mortality.  The Royal College of Anaesthetists advises that
 Explain the rationale behind the timing of people who smoke should quit several weeks
smoking cessation and how cessation may be before surgery and should especially be
achieved. encouraged not to smoke on the day of an
operation.
 There are more than 4500 chemicals in cigarette
smoke, and the majority of these have detri-
‘Smoking’ in this article describes the cigarette smoking of
mental effects on human body systems.
tobacco. The WHO has described tobacco as, ‘the only legal
drug that kills many of its users when used exactly as inten-
 The preoperative assessment clinic provides an
ded by manufacturers’.1 Smoking is the primary cause of opportunity to discuss and encourage smoking
preventable illness and premature death in the UK, account- cessation.
ing for almost 100,000 deaths per year and 6 million deaths per  Electronic cigarettes are used by 2.3 million adults
year worldwide. Although the prevalence of smoking in many in the UK, but the long-term harm caused is
countries is decreasing, in some areas of Africa and the unknown.
Eastern Mediterranean it is rapidly increasing. By 2025, it is
projected that across the world, there will be 1.1 billion people
Smoking carries a considerable burden of comorbidity
who smoke.2 In 2013, the World Health Assembly, under a
and is estimated to cost the NHS around £2 billion each
United Nations mandate, set a global voluntary tobacco target
year. The NHS Five-Year Forward View states that ‘the sus-
to help reduce and prevent premature and avoidable mortality
tainability of the NHS relies on a radical upgrade in disease
from smoking. The agreed target is a 30% relative reduction in
prevention and public health’.3 This includes a reduction in
the prevalence of tobacco use.1
smoking-related ill health. The joint briefing produced by
Action on Smoking and Health (ASH), the Royal College of
Anaesthetists (RCoA), the Royal College of Surgeons of
Edinburgh, and the Faculty of Public Health provides advice
Michael Carrick FRCA is a specialty registrar in anaesthesia at the and examples of good practice in relation to smoking and
Leeds Teaching Hospitals NHS Trust. surgery.4
Jonathan Robson MRCP (Resp) is a consultant in respiratory and Quitting smoking before surgery leads to a reduced
general medicine at St James’s University Hospital, Leeds, whose incidence of postoperative complications. The longer the
special interests include lung cancer, pleural disease, and interven- period of cessation before surgery, the greater the benefit.
tional pulmonology. The RCoA advises that people who smoke should quit
smoking for at least several weeks before surgery and may
Caroline Thomas BSc (Hons), FRCA is a consultant in anaesthesia at benefit from abstaining on the day of surgery.4
St James’s University Hospital, Leeds, whose special interests
include perioperative medicine and anaesthesia for thoracic and
colorectal surgery.

Accepted: 26 September 2018

© 2018 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.
For Permissions, please email: [email protected]

1
Smoking and anaesthesia

History of smoking
Documented instances of tobacco smoking date from around
5000 BC when it was used by the Mayans during religious
rituals. Smoking spread to Western civilisations in the 16th
century after the colonisation of South America. It became
more widespread during the First World War and reached
peak prevalence in the UK after the Second World War.5 By
1949, 81% of men and 39% of women smoked regularly. Doll
and Hill’s6 reports in the 1950s of the adverse effects of
smoking marked the beginning of a decline in its popularity.
Earlier sceptics existed, however, including James I of England
who in 1604 wrote in his ‘Counterblaste to Tobacco’ that
‘Smoking is a custom loathsome to the eye, hateful to the
nose, harmful to the brain…and…dangerous to the lungs’.5

Constituents of cigarettes and cigarette

smoke
Fig 1 The P50 of the COHb dissociation curve is reduced in comparison with that
Tobacco is prepared from the leaves of the tobacco plant of the oxyhaemoglobin dissociation curve, that is the affinity of carbon mon-
through a curing process. The tobacco plant is part of the oxide for haemoglobin is far greater than the affinity of oxygen for haemoglobin.
genus Nicotiana. More than 70 species of tobacco are known
but the main commercial crop is N. tabacum. A more potent
variant, N. rustica, may also be used.7 Components of a ciga- In addition, COHb adversely affects oxygen delivery to the
rette typically include tobacco rolled inside cigarette paper tissues through its effect on the oxyhaemoglobin dissociation
with a filter covered in a tipping paper through which to curve. The presence of COHb causes a left shift (i.e. a reduction
inhale. Cigarette smoke is a concentrated aerosol of liquid in P50) of the oxyhaemoglobin dissociation curve partly
particles suspended in an atmosphere consisting mainly of because of a reduction in 2,3-di-phosphoglycerate (2,3-DPG)
nitrogen, oxygen, carbon monoxide, and carbon dioxide; it levels. This reduces the ability to unload oxygen in the tissues.
comprises a gaseous phase and a particulate phase. The par- For these reasons, hypoxaemia occurs when one breathes
ticulate phase is defined as that which is eliminated on air in the presence of supranormal levels of COHb, and a pa-
passing through a filter of pore size 0.1 mm. This is the Cam- tient who smokes has reduced physiological reserve to
bridge filter. This is a different entity to the filter tip of a maintain their PO2 at times of physiological stress. Although
cigarette described above which allows the passage of parti- many arterial blood gas analysers provide a percentage for
cles of much greater diameter.5 COHb, most oxygen saturation measurement devices are un-
More than 4500 chemicals are present in cigarette smoke, able to distinguish oxyhaemoglobin from COHb and give an
and many of these have adverse effects on human body sys- erroneously high value for oxygen saturation in the presence
tems. The main component of the gaseous phase is carbon of significant levels of COHb.
monoxide and of the particulate phase, nicotine.

Nicotine
Carbon monoxide
Nicotine is the main component of the particulate phase of
Carbon monoxide is the main component of the gaseous cigarette smoke. Tobacco leaves contain many different al-
phase of cigarette smoke and its presence adversely affects kaloids, of which nicotine is the most prevalent. Nicotine
oxygen delivery to the tissues. The inhalation of carbon content varies depending on where the nicotine leaf is
monoxide leads to increased formation of carbox- attached to the tobacco plant and the blend used by different
yhaemoglobin, COHb. In people who smoke, the percentage tobacco companies. Nicotine is addictive in humans; the
of COHb in arterial blood is 2e12%, compared with <1.5% in chemical structure of nicotine is similar to that of acetylcho-
non-smokers. A greater percentage of COHb significantly line and so plays a role in cerebral neurotransmission.
reduces the capacity of haemoglobin to bind and carry ox- A typical cigarette contains around 2 mg nicotine. This is
ygen. Carbon monoxide has around a 300-fold greater af- readily absorbed across the alveolar membrane. Nicotine
finity for haemoglobin than the affinity of oxygen for crosses the bloodebrain barrier and enters the cerebral cir-
haemoglobin, and therefore the formation of COHb is fav- culation within 20 s. It stimulates nicotinic acetylcholine re-
oured over the formation of oxyhaemoglobin.8 Because of ceptors and through a variety of second messengers,
its much greater affinity for haemoglobin, the COHb disso- stimulates the secretion of neurotransmitters such as
ciation curve is shifted to the left of the oxyhaemoglobin noradrenaline, adrenaline, vasopressin, serotonin, dopamine,
dissociation curve (Fig. 1). P50 is the partial pressure of ox- and b-endorphin. Nicotine increases cardiac output and the
ygen (PO2) at which the haemoglobin saturation is 50% un- risk of tachydysrhythmias. At increasing doses the stimulant
der standard conditions; the P50 of the COHb dissociation effects of nicotine diminish; high doses have a sedative and
curve is greatly reduced in comparison with the P50 of the depressant effect. Nicotine has a half-life of 30 min and is
oxyhaemoglobin dissociation curve, which adversely affects metabolised by the cytochrome P450 enzyme system (mainly
oxygen delivery. The dissolved oxygen content is un- via CYP2A6 and CYP2B6) to a number of different metabolites,
changed, but this represents a small contribution only to including cotinine, an active metabolite, which remains in the
blood oxygen content. bloodstream for up to 20 h.6,9

2 BJA Education - Volume 19, Number 1, 2019

 Smoking and anaesthesia

Additional chemicals tissue.12 Smoking causes an acute reduction in airway diam-

eter as a bronchoconstrictive reflex to inhaled particles. Pre-
The classes of other chemical constituents of tobacco smoke
mature small airway closure occurring during expiration
and their effects are shown in Table 1.
results in an increase in closing volume and altered ventila-
tion/perfusion relationships. Smoking is associated with hy-
Tar persensitive airway reflexes. An increased incidence of
The nicotine-free remainder of the particulate phase of ciga- adverse events perioperativelydfor example cough, breath
rette smoke is known as ‘tar’. The chemical components in tar holding, and laryngospasmdis seen. Mucus production and
and their toxicity vary widely across tobacco from different viscosity are increased, whereas mucus clearance is impaired
sources. Measurement of tar is therefore a crude measure of through damage to ciliary structure and function.13 This pro-
the relative toxic potential of tobacco combustion products. motes sputum retention and increases the risk of developing
Tar yields of cigarette brands are measured using a stand- pneumonia and respiratory failure.
ardised method involving a smoking machine. On the basis of
these results, cigarette brands have been classified as high, Gastro-intestinal
medium, or low yield cigarettes. However, a criticism is that
Smoking causes relaxation of the lower gastro-oesophageal
the smoking machine does not accurately simulate human
sphincter and an increased incidence of gastro-oesophageal
smoking and also that smokers have ways of increasing their
reflux disease and peptic ulcer disease. There appears to be
intake, for example by blocking ventilation holes and taking
a reduced rate of postoperative nausea and vomiting amongst
deeper or more frequent puffs.10
smokers possibly because of increased metabolism of volatile
agents by CYP2E1.9 Smoking is associated with an increased
Relevant clinical effects of cigarette smoke incidence of Crohn’s disease and a reduced incidence of ul-
cerative colitis.11
Cardiovascular
Smoking is the largest preventable cause of cardiovascular
Other systems
morbidity and mortality. The effects are well documented and
widely appreciated by anaesthetists. Smoking is associated Cigarette smoking inhibits immune function, and this results
with a three-to four-fold increase in coronary heart disease. in poorer wound healing and increased wound infection rates.
The sympathomimetic effects of nicotine and the reduction in Patients who smoke have abnormal bone metabolism, and
oxyhaemoglobin caused by carbon monoxide adversely affect fracture healing may be delayed.
oxygen supply and demand to the myocardium. An increased
incidence of cardiac dysrhythmias with arterial blood COHb Involuntary or passive smoking
values as low as 4e5% has been described. Smoking causes
adverse effects on an individual’s lipid profile, endothelial The effects of smoking are not limited to smokers themselves.
injury, and the development of atherosclerotic plaques. The WHO defines passive smoking as ‘exposure to second-
Smoking is an independent risk factor in the development of hand tobacco smoke, which is a mixture of exhaled main-
peripheral vascular disease, thromboembolic disease, and stream smoke and side stream smoke released from a
stroke. A strong positive dose-dependent correlation exists smouldering cigarette and diluted with ambient air’. Invol-
between smoking and subarachnoid haemorrhage, and the untary smoking involves inhaling carcinogens and other toxic
association appears to be greater amongst women than components that are present in second-hand tobacco
amongst men. The risk reduces in a dose-dependent manner smoke.14
with smoking cessation.1
Pharmacological effects of smoking
Respiratory
Nicotine and polycyclic aromatic hydrocarbons induce the
Smoking is the cause of 90% of lung cancers. Around 20% of cytochrome p450 system, particularly CYP1A1, CYP1A2, and
smokers develop chronic obstructive pulmonary disease. This CYP2E1. The metabolism of many drugs is altered including
is characterised by a small airway obstruction and a reduction that of theophylline, caffeine, haloperidol, propranolol, and
in forced expiratory volume in 1 s (FEV1), and may be associ- volatile agents. Greater postoperative opioid requirements are
ated with emphysematous and bullous changes to lung described consistently amongst patients who smoke although

Table 1 Classification of chemical constituents of cigarette smoke

Chemical group Examples Common biological effects

Polycyclic hydrocarbons Naphthalene, fluorene, phenanthrene Respiratory tract inflammation and liver dysfunction
Nitrosamines Nicotine-derived nitrosamine A procarcinogen and immunosuppressant via tumour
ketone (NNK) necrosis factor-a and interleukin modulation
Aza-arenes Quinolene Hepatic carcinogen demonstrated in animal studies
Aromatic amines Toluidine, anisidine Bladder carcinogen
Ammonia Corrosive to mucous membranes at high levels;
respiratory tract inflammation
Pyridine Headache; dizziness; amnesia; irritant to eyes, nose,
throat, and skin
Other gases Butadiene, acrolein, isoprene, benzene Carcinogens

BJA Education - Volume 19, Number 1, 2019 3

Smoking and anaesthesia

the mechanism for this is poorly understood. It is not simply surgery, smoking is an independent risk factor for anasto-
because of the increased metabolism of substrates, and motic breakdown. Preoperative smoking has also been shown
additional factors (e.g. altered pain thresholds and receptor- to be associated with an increased risk of admission to the
mediated effects) are likely to have a role.9 Chronic nicotine ICU, emergency readmission to hospital, and longer inpatient
use may have an effect on the number and sensitivity of postoperative stays.13 By quantifying the increased likelihood
nicotinic acetylcholine receptors at the postsynaptic mem- of 30-day mortality and highlighting the broad range of
brane. A number of small studies investigating the potentially serious possible smoking-related complications, the clini-
altered pharmacodynamic effects of neuromuscular blocking cian’s ability to motivate smoking cessation in patients
agents have shown inconsistent results, and there is currently scheduled for elective surgery may be improved.
no clear evidence that patients who smoke require altered
dosing of these drugs.9
Smoking cessation
Perioperative timing of smoking cessation
Perioperative complications associated with Stopping smoking before surgery reduces the risk of post-
smoking operative complications. Evidence varies as to the optimum
Intraoperative complications time to quit. Studies of patients undergoing cardiac surgery in
the 1980s suggested that quitting within 8 weeks of surgery led
A study of a little more than 26,000 patients of whom 26% were
to increased pulmonary complications. However, a more
patients who smoke found an increased incidence of all spe-
recent meta-analysis found no increase in complications
cific respiratory adverse events in the group who smoked.15
amongst smokers who quit within 2 months of surgery.17
The respiratory events investigated included reintubation af-
Trials of at least 4 weeks smoking cessation had a signifi-
ter planned extubation, laryngospasm, bronchospasm, aspi-
cantly larger treatment effect in terms of the perioperative
ration, hypoventilation and hypoxaemia, and pulmonary
morbidity and mortality than shorter trialsdthat is, the longer
oedema. Those at the greatest risk of the adverse events were
the period of cessation before surgery, the better.
younger patients who smoke (aged 16e39 yr) and those who
It is likely that even a brief period of smoking cessation
were obese. The relative risk of developing one of the above
may confer some benefit, given the acute effects of nicotine
complications was 1.8 across all smokers, 2.3 in the younger
and carbon monoxide on the cardiovascular system. As the
population, and 6.3 in young, obese patients who smoke. The
half-life of carbon monoxide is 4 h when breathing air and the
relative risk of perioperative bronchospasm was found to be
half-life of nicotine is 30 min, even a relatively short period of
25.7 in younger patients who smoked and who also had
abstinence from smoking helps to avoid some of the adverse
chronic bronchitis.
effects.
It is not clear whether smoking cessation within a few
hours or days reduces perioperative complications, but there
Postoperative complications
is no clear evidence of harm either, and smoking cessation
Recent meta-analyses have demonstrated that people who should be encouraged at any time.
smoke have increased postoperative mortality and an
increased rate of all cardiac, pulmonary, and septic compli-
Effects of smoking cessation
cations (see Table 2 for details). In addition, there is a clear
doseeresponse relationship between amount smoked and After quitting, the symptoms of cough and wheeze decrease
morbiditydthat is morbidity is increased in smokers in a within weeks. Mucociliary clearance starts to improve after a
dose-dependent manner.16 Current smokers (defined as those week but lung inflammation takes much longer to subside.
having smoked in the preceding year compared with never- Goblet cell hyperplasia regresses and alveolar macrophages
smokers) are 1.38 times more likely to die within 30 days decrease, but alveolar destruction, smooth muscle hyperpla-
(95% confidence interval, 1.11e1.72). The findings regarding sia, and fibrosis may be permanent.18 Smoking cessation de-
morbidity are consistent with previous studies on patients creases all-cause mortality in patients with coronary artery
undergoing cardiac, vascular, thoracic, general, urology, or- disease by approximately one third. It is estimated that it
thopaedic, and plastic reconstructive surgery. In general takes several months for this risk to decrease after the patient
has quit smoking. The risk of coronary heart disease and ce-
rebrovascular disease approaches the risk of never-smokers
Table 2 Adverse effects of smoking on postoperative 30-day within 10e15 yr.19 The rate of decline of FEV1 amongst
morbidity smokers increases as FEV1 becomes worse. However, the
younger the patient at the time of quitting, the slower the rate
Morbidity Odds ratio (95% of decline, eventually approaching the rate in never-
confidence interval) smokers.12 Smoking cessation reduces mortality rates, with
the largest benefit being in those who quit under the age of 30
Pneumonia 2.09 (1.80e2.43)
Unplanned intubation 1.87 (1.58e2.21) yr, but even those who quit at 60 yr of age are likely to have
Mechanical ventilation 1.53 (1.31e1.79) survival benefit of up to 3 yr.20
Cardiac arrest 1.57 (1.10e2.25)
Myocardial infarction 1.80 (1.11e2.92)
Stroke 1.73 (1.18e2.53) Smoking cessation advice and programmes
Superficial wound infection 1.30 (1.20e1.42) Guidance from the National Institute of Health and Care
Deep wound infection 1.42 (1.21e1.68)
Excellence recommends that when patients who smoke ac-
Organ space infection 1.38 (1.20e1.60)
Septic shock 1.55 (1.29e1.87) cess secondary care services, they should be identified and
offered intensive support to quit.21 Patients who smoke are

4 BJA Education - Volume 19, Number 1, 2019

 Smoking and anaesthesia

more likely to quit if they are offered a combination of in- Declaration of interest
terventions, with combined behavioural support and phar-
The authors declare that they have no conflicts of interest.
macotherapy. Healthcare professionals should be trained to
give brief advice on stopping smoking and should have con-
tact with the local NHS Stop Smoking Service to which they MCQs
can refer. If patients do not wish to attend the service, they
should be offered brief advice and support to help them quit, The associated MCQs (to support CME/CPD activity) will be
and pharmacotherapy as appropriate. accessible at www.bjaed.org/cme/home by subscribers to BJA
Use of the Very Brief Advice (VBA) tool is encouraged. This Education.
comprises a three-step approach: ask, advise, and act.4

 ‘Ask’ and record smoking history. Smoking history is References

typically reported in ‘pack-years’ where the number of
1. World Health Organisation. WHO global report of trends in
packs of cigarettes smoked per day is multiplied by the
prevalence of tobacco smoking. Geneva: WHO Press; 2015
number of years the person has smoked (e.g. 1 pack-year
2. Bilano V, Gilmour S, Moffiet T. Global trends and pro-
represents smoking 20 cigarettes per day for 1 yr).
jections for tobacco use, 1990e2025: an analysis of
 ‘Advise’ that the most effective way to quit is with a
smoking indicators from the WHO comprehensive infor-
combination of medication and specialist support.
mation systems for tobacco control. Lancet 2015; 385:
 ‘Act’ on the patient response. Give information, refer, and
966e76
prescribe.
3. NHS England. NHS Five-year forward view. London. 2014
Brief advice offered by a physician has been shown to in- 4. Action on smoking and health. London: the Royal College of
crease quit rates.22 Trials using preoperative intensive smok- Anaesthetists, the Royal College of Surgeons of Edinburgh
ing cessation interventions found a 10-fold relative risk of and the Faculty of Public Health. Joint briefing: Smoking
smoking cessation compared with usual care. Studies looking and surgery; 2016. Available from, https://www.rcoa.ac.
at brief interventions found a smaller effect of smoking uk/sites/default/files/Joint-briefing-Smoking-Surgery.pdf.
cessation rates although quit rates were high in the control [Accessed 11 October 2018]
groups, likely reflecting the fact that people who smoke have a 5. Lumb AB. Smoking and air pollution. In: Lumb AB, editor.
high motivation to quit ahead of major surgery.23 This Nunn’s applied respiratory physiology. 8th ed. London:
‘teachable moment’ should be used to good effect.4 Elsevier; 2017. p. 281e90
6. Doll R, Hill AB. Smoking and carcinoma of lung. Br Med J
1950; 2: 739e48
Pharmacological options
7. Fewster H. The royal horticultural society encyclopaedia of
Three drugs are licensed in the UK for the support of smoking
gardening. 4th ed. London: Dorling Kindersley; 2008
cessation and have proven efficacy.24
8. Thomas C, Lumb AB. Physiology of haemoglobin. Cont Ed
 Nicotine replacement therapy; available as patches or in Anaesth Crit Care Pain 2012; 12: 251e6
shorter acting forms, for example lozenges, chewing gum, 9. Sweeney BP, Grayling M. Smoking and anaesthesia: the
or nasal sprays. pharmacological implications. Anaesthesia 2009; 64:
 Oral bupropion; a nicotinic receptor antagonist with 179e86
dopaminergic and adrenergic actions; it may work by 10. Wright C. Standardized methods for the regulation of
blocking the effects of nicotine, relieving withdrawal or cigarette-smoke constituents. Trends Anal Chem 2015; 66:
reducing depressed mood. 118e27
 Oral varenicline; a nicotinic receptor partial agonist that 11. Lakier J. Smoking and cardiovascular disease. Am J Med
binds less effectively than nicotine. 1992; 93: S8e12
12. Fletcher C, Peto R. The natural history of chronic airflow
obstruction. Br Med J 1977; 1: 1645e8
Electronic cigarettes 13. Grønkjær M, Eliasen M, Skov-Ettrup LS et al. Preoperative
There has been an increase in recent years in the popularity of smoking status and postoperative complications: a sys-
electronic cigarettes (e-cigarettes), or ‘vaping’. They are tematic review and meta-analysis. Ann Surg 2014; 259:
currently used by 2.3 million adults in the UK. E-cigarettes use 52e71
a battery to heat a solvent and disperse an aerosol that con- 14. World Health Organisation. Protection from exposure to
tains nicotine, water, and sometimes flavouring. Nicotine can second-hand tobacco smoke. Geneva: WHO Press; 2007
therefore be delivered to the respiratory tract without 15. Schwilk B, Bothner U, Schraag S, Georgieff M. Periopera-
combustion. tive respiratory events in smokers and nonsmokers un-
The potentially harmful effects of vaping are an area of dergoing general anaesthesia. Acta Anaesthesiol Scand
interest. Current opinion is that e-cigarettes are likely to be 1997; 41: 348e55
less harmful than cigarettes, but studies have found the 16. Turan A, Mascha EJ, Roberman D et al. Smoking and
aerosol to contain heavy metals, polycyclic hydrocarbons, perioperative outcomes. Anaesthesiol 2011; 114: 837e46
nitrosamines, volatile organic compounds, and inorganic 17. Mills E, Eyawo O, Lockhart L et al. Smoking cessation re-
compounds.25 duces postoperative complications: a systematic review
It is clear that e-cigarettes help some people to quit and meta-analysis. Am J Med 2011; 124: 144e54
smoking, although no randomised controlled trials have been 18. Willemse BW, Postma DS, Timens W et al. The impact of
carried out to compare the effectiveness of e-cigarette use smoking cessation on respiratory symptoms, lung func-
with more established pharmacotherapy to promote and tion, airway hyperresponsiveness and inflammation. Eur
maintain smoking cessation. Respir J 2004; 23: 464e76

BJA Education - Volume 19, Number 1, 2019 5

Smoking and anaesthesia

19. Godtfredsen NS, Prescott E. Benefits of smoking cessation 23. Thomsen T, Villebro N, Møller AM. Interventions for pre-
with focus on cardiovascular and respiratory comorbid- operative smoking cessation. Cochrane Database Syst Rev
ities. Clin Respir J 2011; 5: 187e94 2014; 3, CD002294
20. World Health Organisation International Agency for 24. Cahill K, Stevens S, Perera R et al. Pharmacological in-
Research on Cancer. Reversal of risk after quitting terventions for smoking cessation: an overview and
smoking. In: Handbooks of cancer prevention, vol. 11. Lyon: network meta-analysis. Cochrane Database Syst Rev 2013;
WHO Press; 2007 5, CD009329
21. National Institute for Health and Care Excellence. Public 25. Shahab L, Goniewicz M, Blount B et al. Nicotine, carcin-
health guideline PH48: smoking in acute, maternity and mental ogen and toxicant exposure in long-term e-cigarette and
health services. London. 2013 nicotine replacement therapy users: a cross-sectional
22. Stead LF, Buitrago D, Preciado N et al. Physician advice for study. Ann Intern Med 2017; 166: 390e400
smoking cessation. Cochrane Database Syst Rev 2013; 5,
CD000165

6 BJA Education - Volume 19, Number 1, 2019