Abstract
Multiple systematic reviews have investigated the effectiveness of preoperative interventions for smoking-cessation, although relatively few have focused on the prevention of surgical complications. This overview of systematic reviews aimed to describe the types of smoking interventions studied to prevent postoperative complications, summarize the results, and evaluate the quality of the reviews and strength of evidence to inform clinicians, health practitioners, policy developers, and government bodies. Comprehensive searches of Cochrane Library, MEDLINE, EMBASE, CINAHL, and Johanna Briggs Institute databases were conducted to identify systematic reviews of preoperative smoking-cessation interventions to prevent surgical complications (inception—May 14, 2024). Search results were independently screened by 2 reviewers for articles meeting the eligibility criteria. Data on key review characteristics and included studies were extracted: aim, search strategy, included studies, risk of bias, population, sample size, intervention, comparator, main findings, and conclusions. Quality appraisal of the reviews was undertaken using the AMSTAR 2 tool and evidence certainty was evaluated using Grading of Recommendations Assessment, Development and Evaluation (GRADE). Sixty-seven full-text articles from 838 citations were screened, resulting in 6 included systematic reviews with 12 primary studies reporting postoperative complications. Four reviews which included all primary studies, provided moderate to high strength of evidence. There was high-certainty evidence suggesting interventions started ≥4 weeks before surgery incorporating multiple behavioral support sessions and pharmacotherapy are needed to prevent postoperative complications, particularly the incidence of surgical site infections. High-certainty evidence also indicates the duration of smoking-cessation is important, with benefits amplified for longer periods. However, medium- to high-certainty evidence suggests interventions initiated <4 weeks before surgery even with multiple behavioral support sessions (with or without pharmacotherapy), and interventions commenced >4 weeks before surgery but with only 1 interventional component, increase quit rates but do not reduce complications. This overview provides the most up-to-date summary and quality assessment of systematic review evidence on the effectiveness of preoperative smoking-cessation interventions to prevent surgical complications. The evidence supports providing smoking-cessation interventions which include multiple behavioral support sessions and pharmacotherapy implemented at least 4 weeks before surgery to reduce postoperative complications. Consequently, anesthesiologists need to work with primary care physicians, consultants, and surgeons to optimize smoking-cessation interventions way in advance of surgery.
People who smoke are at increased risk of anesthetic, postanaesthetic, and postsurgical complications,1 including delayed wound healing, pulmonary complications, surgical site infections (SSIs), and mortality.1–3 Postoperative complications account for 80% of surgery-related deaths and reduce long-term median survival by up to 70%.4 Recent studies on the prevalence of smoking among surgical patients estimated 1 in 4 patients smoked cigarettes at the time of surgery in Michigan, United States,5 and 1 in 5 in Karachi, Pakistan.6 In Australia, approximately 24% of people undergoing surgical procedures each year smoke and in 2016 to 2017, approximately 40,593 Australians who had a surgical procedure experienced an SSI leading to an additional 101,888 days in hospital.7,8 Postoperative complications are expensive to health care systems, and place significantly psychological, financial, and physical burdens on surgical patients.9,10
Smoking-cessation can reduce some, but not all, of the smoking-induced pathophysiological changes that lead to heightened risk of perioperative complications.11 Preoperative smoking-cessation is associated with reduced postoperative complications3 and a systematic review by Myers et al12 clarified stopping smoking shortly before surgery does not worsen clinical outcomes. Perioperative care provides a unique opportunity to incentivize patients to quit smoking, presenting a teachable moment where patients are more receptive to health advice11,13,14 and surgery has been found to precipitate spontaneous uptake of risk-reducing health behaviors such as smoking-cessation.13 Given a large proportion of people who smoke will require surgery at some point during their lives, health care professionals in the preoperative clinic are presented with an opportunity to further influence patient health outcomes via smoking-cessation interventions, and reduce long-term health costs.11
During a search for effectiveness data for an upcoming economic evaluation of preoperative smoking-cessation to prevent postoperative complications,14 it became apparent that there were at least 3 relevant systematic reviews.1,4,9 Review of reviews is particularly beneficial when faced with multiple reviews of uncertain quality addressing a similar research question.15–21 Consequently, an overview of systematic reviews was undertaken to examine and juxtapose the methodologies, quality standards, outcomes, and conclusions across several reviews.
METHODS
The systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines22 (Supplemental Digital Content 1, Supplementary Table S1, https://links.lww.com/AA/E990) and is registered in the International Prospective Register of Systematic Reviews (PROSPERO CRD42021267813).23
Search Strategy
Comprehensive electronic searches of The Cochrane Library, MEDLINE, EMBASE, CINAHL, and The Johanna Briggs Institute were conducted from database inception until May 14, 2024 to identify systematic reviews on smoking-cessation interventions implemented in the preoperative setting to prevent surgical complications. The search for literature was conducted with the assistance of a Deakin University research librarian. Search terms and keywords were informed by previous reviews1,9,10 and grouped into 4 categories, smoking-cessation, preoperative care, postoperative complications, and systematic review. The search strategies are presented in Supplemental Digital Content 2, Supplementary Material S2, https://links.lww.com/AA/E991.
After deduplication, the titles, and abstracts of search results were screened separately by the 2 reviewers (R.F. and N.M.) using Covidence software24 to assess eligibility for inclusion. Full-text articles were retrieved where there was insufficient information in the abstract to reach a decision. After abstract screening, full-text articles were analyzed independently by the reviewers for inclusion. Any disagreements from abstract to full-text screening were resolved by discussion between the reviewers. The reference lists of the included reviews and forward citations were checked to identify additional literature unidentified by the search. Results of the study selection process are presented in the Figure and a list of excluded articles is provided in Supplemental Digital Content 3, Supplementary Material S3, https://links.lww.com/AA/E992.
Eligibility Criteria
Systematic reviews, defined as a review where systematic searches were conducted in at least 2 sources (with at least 1 being an electronic database),26 assessing the effectiveness of any type of smoking-cessation intervention started preoperatively to reduce postoperative complications in smokers undergoing any type of surgery were included in the review. Reviews with inadequate reporting of complications, intervention details, and duration of treatment were excluded. Narrative reviews, consensus statements, guidelines, and multicomponent prehabilitation reviews were also excluded. Postoperative complications included wound complications (including delayed wound healing, dehiscence, SSI, hernia, and necrosis), pulmonary or cardiovascular complications, prolonged hospital stay, all-inclusive postoperative complications, and mortality. Reviews not published in the English language were excluded.
Data Extraction and Synthesis
A narrative summary of the results from the reviews was conducted. The following information was extracted from each individual systematic review: first author’s name, country, review aim, search strategy, number and types of included studies, risk of bias, population characteristics (surgery type), types and intensity of interventions, timings of intervention implementation, and outcomes (postoperative complications, smoking-cessation rates). Data extraction was checked by the second reviewer (N.M.) and any disagreements were resolved through discussion.
Further, the results of the primary studies reporting postoperative complications were divided into 3 categories based on intervention intensity as per Wong et al’s Society for Perioperative Assessment and Quality Improvement Consensus Statement27 to facilitate presentation and analysis of the findings: high-intensity (intervention started more than 4 weeks before surgery and includes multiple contact points providing behavioral support with or without pharmacotherapy); moderate-intensity (intervention implemented 4 or less weeks before surgery and includes multiple contact points providing behavioral support, with or without pharmacotherapy or commenced >4 weeks before surgery but only have 1 interventional component); and low intensity (interventions that include only 1 smoking-cessation component such as pharmacotherapy, and interventions where only 1 instance of counseling is delivered). A meta-analysis was not conducted due to the potential for data duplication across individual studies within systematic reviews, which could result in inaccurate estimations of the true effect size.28
Quality Appraisal and Assessment of Evidence
Assessment of review quality was conducted independently by the reviewers using the AMSTAR 2 tool29 which provides a comprehensive checklist of 16 questions tailored to assess the methodological quality of systematic reviews. The AMSTAR 2 tool rates the quality of reviews as high, moderate, low, or critically low based on the number of critical domains met. Any discrepancies were resolved by discussion between the reviewers.
| Downgrade | Criteria |
|---|---|
| 1 level | |
| Imprecision | Sample size 100–199 participants in relevant reviews |
| Risk of bias (study level)a | High risk of bias in randomization and blinding for >75% included studies |
| Inconsistencyb | High heterogeneity (I 2 >75%) |
| AMSTAR 2 “no” response to 1 item | A priori research design or comprehensive literature review or duplicate study setting or duplicate study abstraction |
| 2 levels | |
| Imprecision | Sample size <100 participants in relevant reviews |
| AMSTAR 2 “no” response to ≥2 items | A priori research design or comprehensive literature review or duplicate study setting or duplicate study abstraction |
The AMSTAR 2 responses were used to inform Grading of Recommendations Assessment, Development and Evaluation (GRADE) certainty ratings (high, moderate, low, or very low certainty of evidence).30,31 GRADE is a systematic approach to rating the quality of evidence and the strength of clinical practice recommendations.31,32 The quality of the systematic reviews and strength of evidence signifies the extent to which confidence in an effect estimate is justified and therefore informs clinicians and health practitioners about the level of confidence in the findings from evidence synthesis.31 High-quality evidence suggests that further research is unlikely to change the estimate of effect and clinical practice recommendations, whereas low-quality evidence indicates that future research could significantly alter guidance. Initially, reviews were ranked with “high certainty” and then downgraded for serious or very serious methodological concerns and “no” responses for AMSTAR 2 items (see Table 1).28,29,33,34 The GRADE levels of evidence were computed as high (no downgrades), moderate (1 or 2 downgrades), low (3 or 4 downgrades), or very low (5 or 6 downgrades).33
RESULTS
Search Results
Sixty-seven full-text articles were screened from 838 citations, resulting in 6 included systematic reviews, Gavilan et al,35 Mills et al,1 Ricker et al,36 Scott et al,4 Sorensen,10 Thomsen et al,9 and 61 excluded articles (Figure).
No additional systematic reviews were identified from forwards or backward citation tracing. The key characteristics of the included systematic reviews are presented in Table 2.
| First author (y) | Country | Review aim | Search strategy | Included studies | Risk of bias for primary studies | Author conclusion |
|---|---|---|---|---|---|---|
| Gavilan et al (2023)35 | Spain | To assess the effectiveness of presurgical interventions to support smoking-cessation and reduce surgical complications. | MEDLINE (via PubMed), WOS and CINAHL. Search dates, March 2009 to April 2021. Search terms defined. Search criteria: RCTs that assessed any type of presurgical intervention, with no limitations on the definition of smoker, age, follow-up time or surgical population. | 9 RCTs | Low risk of bias | Intensive interventions (health care professional visits, psychoeducational support, and pharmacotherapy) were most effective in reducing postoperative complications, even when instigated very shortly before surgery. |
| Mills et al (2011)1 | Canada | To review randomized control trials and observational evidence to establish the effect of preoperative smoking-cessation on postoperative complications and to determine if there is an optimal cessation period before surgery. | Medline, EMBASE, CENTRAL, AMED, CINAHL, TOXNET, Development and Reproductive Toxicology, Hazardous Substances, Databank, PsychInfo, WOS, ScienceDirect and Ingenta and hand searching of relevant bibliographies and health technology assessments. No search start or end data. Search terms not defined. Search criteria: observational and RCTs that evaluated the incidence of postoperative complication among population who achieved smoking-cessation at a defined point before surgery. Studies could include any type of cessation and any surgical population. They had to report on the proportion of clinical complications of past smokers and current smokers. There were no search limitations. | 6 RCTs, 10 prospective comparisons, 5 retrospective analyses | Low risk of bias | Each week of cessation resulted in a larger effect size (B coefficient – 0.191, 95% CI, −0.368 to −0.014, P = .03). |
| Ricker et al (2024)36 | United States | To provide a summary and comparative evaluation of prehospital smoking-cessation interventions on smoking-cessation rates. | CINAHL, EMBASE, MEDLINE, PsychINFO, PubMed, Web of Science databases from 1998 to 2019. Search terms defined. Search criteria: studies including patients receiving preoperative smoking-cessation intervention. Studies evaluating self-reported outcomes and those with historical controls were excluded. |
8 RCTs | Low risk of bias reported according to ROB2 tool. | Preoperative smoking-cessation improves perioperative outcomes and complication rates. |
| Scott et al (2007)4 | Australia | To provide an evidence-based update in several key areas in the management of the perioperative patient. | Medline, Cochrane Library, CENTRAL, PubMed. No evidence of hand searching. No search start of end date Search terms defined. Search criteria not specified. Search limited to being published between 1985 to September 2006. Exclusion criteria: any study relating to pediatric, transplant or cardiac surgery | 1 RCT relating to smoking-cessation. Total number of studies not reported. | No comment on included articles risk of bias | Wound complications significantly reduced with the intervention but rates of perioperative pulmonary complications unchanged. |
| Sorenson (2012)10 | Denmark | To clarify the evidence on smoking and postoperative healing complications across surgical specialities and to determine the impactive of perioperative smoking-cessation intervention | CENTRAL, Medline, EMBASE and hand searching. No search start date. Search end date May 2010 for cohort studies and January 2011 for RCTs. Search terms defined, no limitations described. Search criteria: cohort studies with 100 participants or more assessing healing complications in smokers and former smokers. Additionally, RCTs assessing the effect of perioperative smoking-cessation on postoperative healing complications were included, with a minimum of 1 wk of preoperative intervention and assessment of healing outcomes after elective surgery. RCTs with a dropout rate >40% were excluded. | 140 cohort studies including 479,150 patients 4 RCTs including 416 participants |
All RCTs had a low risk of bias, and a high Jadad score of 6 | No significant reduction in healing complications combined by perioperative smoking-cessation. However, significant reduction in surgical site infections. |
| Thomsen et al (2014)9 | Denmark | The objectives of the review are to assess the effect of preoperative smoking intervention on smoking-cessation at the time of surgery and 12 mo postoperatively, and on the incidence of postoperative complications. | Cochrane Tobacco Addiction Group Specialised Register, CENTRAL, PsycINFO, EMBASE, MEDLINE and CINAHL. No search start or end date reported. Search terms defined. Search criteria: RCTs that recruited participants who smoked before surgery, offered a smoking-cessation intervention, measured preoperative and long-term quit rates from smoking or the incidence of postoperative complications or both. | 13 RCTs enrolling 2010 participants all together met inclusion criteria | Low risk of bias for majority | Intensive interventions beginning ≥4 wk before surgery, including multiple contacts with behavioral support, the offer of pharmacotherapy reduce the incidence of complication, reduce smoking in the perioperative environment and in the long term. Brief interventions offered closed to surgery offer small benefits on smoking behavior but have not demonstrated a reduction in complications. |
The 6 systematic reviews were published between 2007 and 2024; 2 reviews were conducted in Denmark,9,10 and 1 review in Australia,4 Canada,1 Spain,35 and the United States.36 Surgery types in the reviews included colorectal, joint replacement, general, cancer, hernia, laparoscopic, cholecystectomy, orthopedic, ophthalmologic, gynecologic, neurosurgical, urologic, ear, nose, throat, plastic, faciomaxillary, vascular and cardiovascular. Two reviews reported no funding support or had no financial disclosures to note4,10; two reviews did not explicitly mention funding9,36; one review reported the manuscript was sponsored by a pharmaceutical company1 and the final review was funded by the Spanish government.35 The primary studies included in the reviews are listed in Supplemental Digital Content 4 and 5, Supplementary Material S4, https://links.lww.com/AA/E993, and Supplementary Table S5, https://links.lww.com/AA/E994, (n = 23), and the characteristics for studies reporting postoperative complications (n = 12) are presented in Supplemental Digital Content 6, Supplementary Table S6, https://links.lww.com/AA/E995.
Quality Appraisal
Table 3 summarizes the findings of the AMSTAR 2 quality assessments.
| First author (y) | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | Overall confidence in the results |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gavilan et al (2023)35 | Y | Y | N | Y | Y | Y | N | Y | Y | N | N/A | N/A | Y | N | N/A | Y | Low |
| Mills et al (2011)1 | Y | N | N | Y | Y | Y | N | Y | Y | Y | Y | Y | Y | Y | N | Y | Critically low |
| Ricker et al (2024)36 | Y | Y | N | P | Y | Y | N | N | P | N | N/A | N/A | N | Y | N/A | Y | Critically low |
| Scott et al (2007)4 | Y | N | N | P | N | N | N | Y | N | N | N/A | N/A | N | Y | N/A | Y | Critically low |
| Sorenson (2012)10 | Y | Y | N | P | N | N | N | Y | Y | N | Y | Y | Y | Y | Y | Y | Low |
| Thomsen et al (2014)9 | Y | Y | N | P | Y | Y | Y | Y | Y | N | Y | Y | Y | Y | N | Y | Low |
All reviews had at least 1 critical flaw. Three systematic reviews, Gavilan et al,35 Sorensen,10 and Thomsen et al,9 had 1 critical flaw, that is, a “low” confidence rating, meaning the results may not offer a precise and complete summary of the available evidence.29 The remainder had more than 1 critical domain unmet, that is, a “critically low” confidence rating, meaning results do not provide rigorous and thorough summaries of the available studies.29
The reviews ranked from highest to lowest AMSTAR 2 confidence ratings are as follows: Gavilan et al35 and Sorenson,10 1 critical domain unmet (item 7), and Thomsen et al,9 1 critical domain unmet (item 15); Mills et al,1 3 critical domains unmet (items 2, 7, and 15); Ricker et al,36 2 critical domains unmet (items 7 and 13) and Scott et al,4 4 critical domains unmet (items 2, 7, 9, and 13).
Strength of Evidence
The strength of evidence according to the GRADE algorithm is presented in Table 4.
| Downgrades to levels of evidence | ||||||
|---|---|---|---|---|---|---|
| First author (y) | Imprecision | Risk of bias (study quality) | Inconsistency | Risk of bias (review quality) | Total | Levela |
| Gavilan et al (2023)35 | 0 | 0 | 0 | 1 | 1 | Moderate |
| Mills et al (2011)1 | 0 | 0 | 0 | 1 | 1 | Moderate |
| Ricker et al (2024)36 | 0 | 0 | 0 | 0 | 0 | High |
| Scott et al (2007)4 | 1 | 1 | 0 | 2 | 4 | Low |
| Sorenson (2012)10 | 0 | 1 | 0 | 2 | 3 | Low |
| Thomsen et al (2014)9 | 0 | 0 | 0 | 0 | 0 | High |
Thomsen et al9 and Ricker et al36 were rated with high strength of evidence based on GRADE; Gavilan et al35 and Mills et al1 were rated as moderate, and the remaining systematic reviews low strengths of evidence. Collectively, Thomsen et al,9 Ricker et al,36 Gavilan et al,35 and Mills et al1 include all of the primary studies reporting postoperative complications.
High-Intensity Interventions
High-strength evidence from 1 review, moderate strength evidence from 2 reviews and low-strength evidence from 2 reviews including mixed surgical populations indicates that high-intensity interventions have the greatest effect on postoperative complications, particularly infectious healing complications.
A randomized controlled trial (RCT) conducted by Moller et al37 with a low risk of bias (9, 10), included in all reviews except Gavilan et al,35 demonstrated that SSIs (odds ratio [OR], 0.12, 95% confidence interval [CI], 0.03–0.58)10(8) and wound-related complications (5% in the intervention group vs 31% in the control, P < .01)9 were reduced when smokers received weekly smoking-cessation counseling with a nurse and were offered nicotine replacement therapy (NRT) for 6 to 8 weeks before surgery to 10 days postoperatively. Rates of pulmonary complications were unchanged in this study,4 contrary to the findings from the primary study by Wong et al38 where smokers who quit 4 or more weeks before surgery had fewer reported pulmonary complications. Thomsen et al9 reported high-intensity interventions had a significant effect on any postoperative complication (relative risk [RR], 0.42, 95% CI, 0.27–0.65) and wound complications (RR, 0.31, 95% CI, 0.16–0.62) and Gavilan et al35 observed significant differences in postoperative complications at 30 days (20% vs 38%, P = .04).
Sorenson10 and Thomsen et al9 concluded that at least 4 weeks of cessation should be encouraged before surgery and include multiple contact points for behavioral support and optional pharmacotherapy, and longer periods of smoking-cessation may be needed to reduce all types of complication. Further, Scott et al4 and Thomsen et al9 agreed that high-intensity preoperative smoking programs appear to reduce overall wound complications.
Moderate-Intensity Interventions
High-strength evidence from 2 reviews, moderate-strength evidence from 2 reviews, and low-strength evidence from 1 review including mixed surgical populations suggests moderate-intensity interventions do not reduce postoperative complications.
An RCT conducted by Lindstrom et al with a low risk of bias suggested the relative risk of postoperative complications was 0.51 (95% CI, 0.27–0.97) for telephone counseling or weekly meeting delivered by a nurse, a telephone support service, and free NRT compared with standard care.1,39 However, Lee et al40 reported no statistically significant difference in postoperative complications between a multicomponent intervention including behavioral support sessions, NRT and telephone services, and standard care (RR, 0.79, 95% CI, 0.38–1.63). One RCT investigating the administration of varenicline 1 week before surgery and 12 weeks postsurgery with 1 15-minute counseling session delivered 8 to 30 days before surgery, and one 15-minute counseling session postoperatively reported no effect on postoperative complications (varenicline 12.6% vs placebo 13.3%, P = .88).9,38 Similarly, 2 further studies investigating the effectiveness of 5 behavioral counseling sessions over a 6-week period and an offer of NRT,41 and 1 brief counseling session plus varenicline and a referral to a telephone Quitline or brief advice and self-referral to a telephone Quitline (7–60 days before surgery)42 reported no significant differences in postoperative complications.35
Low-Intensity Interventions
High-strength evidence from 2 reviews, moderate-strength evidence from 2 reviews and low-strength evidence from 1 review including mixed surgical populations indicates low-intensity interventions increase abstinence but have no effect on postoperative complications.
Four of the 8 brief RCTs analyzed by Thomsen et al9 showed a modestly significant effect on abstinence at the time of surgery, but no impact on postoperative complications. Similarly, pooled analysis by Thomsen et al9 revealed no reduction in complications (RR, 0.92; 95% CI, 0.72–1.19) or wound complications (RR, 0.99; 95% CI, 0.70–1.40). Brochures used in conjunction with brief counseling, a free 6-week supply of NRT, and a referral to a national telephone Quitline demonstrated a 10% increase in quit rates compared with the control group on the day of surgery, and an 18% increase at 30 days postoperatively.37
Mills et al1 reported the results of a placebo-controlled trial43 where bupropion was administered for 7 weeks preoperatively. Bupropion improved quit rates and reduced cigarette consumption at the time of surgery and 3 weeks postsurgery but there was no difference detected in postoperative complications. However, the sample size was small with variable compliance rates (intervention group, 92% [n = 18]; placebo group 61% [n = 12]).
Finally, no significant differences were reported for brief intervention in the studies included in the review by Gavilan et al.35
Duration
High-strength evidence from 1 review and medium-strength evidence from 1 review suggest interventions started at least 4 weeks before surgery have a greater impact on reducing postoperative complications than those commencing closer to surgery.
In the meta-analysis by Mills et al,1 interventions started at least 4 weeks before surgery reduced postoperative complications but not cessation for shorter durations compared with no intervention (more than 4 weeks RR, 0.45, 95% CI, 0.30–0.68; <4 weeks RR, 0.92, 95% CI, 0.53–1.60; P value for the difference.04). The meta-regression analysis suggested each additional week of cessation resulted in a larger reduction in total postoperative complications (B coefficient 0.191, 95% CI, 0.37–0.01, P = .03) and aligns with evidence from the observational studies. Sorensen10 and Thomsen et al9 also concluded providing high-intensity interventions for at least 4 weeks before surgery may be required to reduce postoperative complications. One study conducted in the emergency trauma setting by Nasell et al35 suggested high-intensity interventions initiated shortly before surgery may reduce the number of patients experiencing at least 1 complication (20% intervention vs 38% control, P = .048). However, a secondary regression analysis did not confirm a statistically significant difference in the odds of having a postoperative complication between the groups.
DISCUSSION
The findings provide a narrative of the current body of review evidence on preoperative smoking-cessation to prevent postoperative complications, describing the perioperative smoking interventions that have been studied, summarizing the findings of published reviews, evaluating the quality of existing evidence and assessing the certainty of the effect of preoperative smoking-cessation on complications. This information provides an up-to-date resource for clinicians, health practitioners, policy makers, and government bodies.
The findings suggest high-intensity interventions are needed to prevent postoperative complications. Further, the evidence indicates that duration of smoking abstinence preoperatively is important, with longer periods of cessation associated with better outcomes possibly due to the ability to incorporate multiple face-to-face behavioral components and pharmacotherapy. Of note, Gavilan et al35 concluded intensive interventions (health care professional visits, psychoeducational support, and pharmacotherapy) were most effective in reducing postoperative complications, even when instigated very shortly before surgery. However, findings from this overview suggest smoking-cessation interventions need to be commenced >4 weeks before surgery to reduce postoperative complications. Naturally, urgent and emergency surgery should not be delayed to allow sufficient time to implement smoking-cessation interventions. Rather, the pre, peri, postoperative, and follow-up periods present excellent opportunities to support smokers to quit and prevent longer term adverse health outcomes such as stroke, chronic obstructive pulmonary disease and lung, pancreas, cervix, bladder, and kidney cancers.44,45
Findings are predicated on a limited amount of eligible and reliable systematic reviews; all reviews had low or critically low confidence rating as per the AMSTAR 2 critical appraisal instrument. Judgments about the quality of the reviews rely on adequate reporting and low confidence ratings may reflect insufficient details, inadequate study conduct or both. Most reviews did not justify the reasons for excluding individual studies (item 7) or report sources of funding (item 10), and none explained the selection of included study designs (item 3). Consequently, the appropriateness and impact of excluded studies from the reviews and the potential influence of industry-funded studies cannot be evaluated.29 Future reviews should endeavor to explicitly report all items listed in the AMSTAR 2, particularly commonly missing items 3, 7, and 10 to increase confidence in the findings. Based on the GRADE framework, Thomsen et al,9 Ricker et al,36 Gavilan et al,35 and Mills et al1 provided moderate-to-high strength of evidence. Of note, these 4 reviews include all the primary studies reporting postoperative complications. Consequently, findings from these reviews have been prioritized.
Unsurprisingly, many primary studies were included across multiple reviews; there were 12 unique studies reporting postoperative complications in the 6 reviews. Of note, a large proportion of studies had small sample sizes, potentially underpowered to detect differences in healing and total postoperative complications. The interventions were tested in heterogenous cohorts such as general, orthopedic, cancer, and gynecological surgery, among others. Distinct surgical procedures and underlying pathologies may differentially affect postoperative complication rates46,47 and motivations to quit.48 Many primary studies offered NRT but did not report on the uptake. More research is needed to determine exactly what type of pharmacotherapy is most effective and cost-effective in the preoperative setting.8,49
The primary studies typically recruited patients in the preoperative clinic. In some hospitals, preadmission appointments occur just 1 week before surgery, providing insufficient time to optimize smoking-cessation interventions. The best time to counsel surgical patients and start smoking-cessation interventions is at the time of referral to the surgeon as this is usually far in advance of the surgical procedure. Smoking-cessation interventions could be initiated before preadmission appointments by allied health, general practitioners (GPs), or practice nurses to increase the length of cessation preoperatively and reduce the likelihood of postoperative complications. For example, simple advice from GPs can increase quitting rates by up to 3%.50 Further, telephone counseling such as the Victorian Quitline service which provides affordable and accessible cessation support improve quit rates by up to 25% more than pharmacotherapy alone and could be integrated into coordinated care with health professionals.51 Opportunities are also presented during the postoperative period, possibly by utilizing a wide range of health care professionals such as nurses, surgeons, respiratory therapists, anesthetists, and pharmacists. Embedding best practice hospital-based models of care for smoking cessation would promote timely and systemic cessation support across the health system, particularly during potentially teachable moments such as after a cardiac event or diagnosis of cancer.51
None of the reviews included e-cigarettes as a means of smoking cessation to reduce postoperative complications amongst the presurgical population likely due to the relatively recent introduction of these devices and an absence of evidence of effectiveness.27
Further research into the costs and benefits of delaying elective surgery to provide people who smoke with sufficient time to quit before their operation may be warranted given the substantial health, social, and economic burden imposed by smoking in the shorter and longer term.7,49,52–54
Limitations
The number of reviews included in this analysis was limited by the provision of insufficient details on the type, intensity, and duration of smoking-cessation interventions in primary systematic reviews and information on postoperative complications. Quality appraisal and judgments about the strength of evidence are inherently subjective and AMSTAR 2 and GRADE ratings may not reflect broader opinion. Hence AMSTAR 2 and GRADE results are provided to explicitly present the way the reviewers assessed the reviews and to enable other researchers to consider how the evaluations might align with their own interpretation. Two reviewers independently appraised the reviews and discussed findings, supporting the validity of the results. Recent evidence suggests moderate agreement between raters of AMSTAR 2, similar to other instruments used to critically appraise systematic reviews.29,55 Some relevant reviews may have been missed by the searches. However, all key publications should have been detected through the combination of the searches and forward and backward citation tracing. Further, search strings and databases were constructed and chosen with the advice of the health liaison librarian, promoting the rigor of the review process. Multicomponent prehabilitation reviews were excluded and may have contained relevant information on preoperative smoking interventions.
In conclusion, the findings from this overview of systematic reviews suggest, based on limited, high-certainty evidence that high-intensity interventions incorporating multiple contacts for behavioral support for at least 4 weeks are needed to prevent postoperative complications and benefits appear amplified for each additional week of abstinence from smoking. Interventions initiated <4 weeks before surgery or which do not include behavioral support improve quit rates but do not reduce postoperative complications. Anesthesiologists need to work with primary care physicians, consultants, and surgeons to optimize smoking-cessation interventions way in advance of surgery. Future systematic reviews would benefit from adhering to PRISMA and AMSTAR 2 guidelines and applying the GRADE framework.
REFERENCES