Smokers have increased risk of soft-tissue complications following primary elective TKA

Tobacco smoking has been identified as a risk factor for adverse postoperative outcomes, including wound-related complications, surgical-site infection, and cardiopulmonary complications [1, 2]. Hawn et al. revealed an increased risk of postoperative complications for smokers regardless of surgical specialty and case complexity [1]. Total knee arthroplasty (TKA) is a frequently performed procedure with approximately 700.000 implanted in the USA yearly, with estimations suggesting a continuing increase [3‐5]. Previous studies have confirmed the hypothesis that tobacco smoking increases the risk of overall postoperative complications after elective orthopedic surgery and TKA [6‐8]. Regarding prosthesis-related complications, Singh et al. discovered smoking as a risk factor for deep infection and implant revision after primary TKA and THA [9]. Lim et al. have concluded a higher risk of earlier revision in smokers [10]. However, Matharu et al. did not find an increase in long-term revision rates [11]. Despite Matharu et al. revealing no clinically significant differences in postoperative patient-reported outcome measures (PROMs) between smokers, former smokers, and never smokers, literature on PROMs after TKA is scarce [11].

The literature describes a decrease in smoking and alcohol consumption in the initial 12-month post-THA and TKA [12]. The cited literature has a follow-up of 30 days up to 10 years, whereas long-term follow-up over 15 years is missing [6‐11].

The aim of this study was to identify whether smoking status (active, former, never smoker) at time of surgery influences the outcome of TKA in terms of prosthesis-related complications and postoperative PROMs. The hypothesis was that current smokers have a higher risk of complications, revision surgery, and decreased outcome.

The most important findings of this study were a significantly higher soft-tissue complication rate for current smokers in comparison with never smokers. Current smokers were more likely to undergo revision surgery due to restricted movement and hematoma as compared with never smokers.

Earlier findings were an increased overall revision risk for active smokers against never smokers after primary elective TKA [1, 7‐10]. In contrast to the existing evidence [7, 8, 20], we observed the risk of infection not significantly increased in active smokers compared to never smokers. Furthermore, former smokers did not have an overall increased revision likelihood compared to never smokers, as supported by other investigations [10]. We found current smokers (57 ± 9.7 years) and former smokers (63.3 ± 7.7 years) to undergo primary TKA surgery 12.6 years (± 8.2; p < 0.001) and 6.3 years (± 7.7; p < 0.001) earlier than never smokers (69.6 ± 8.2 years), which might be due to faster biological aging of smokers in contrast to non-smokers [21] or due to smokers reporting higher musculoskeletal pain ratings, therefore needing an earlier definitive surgical treatment [22].

Some studies have demonstrated a beneficial effect of smoking cessation prior to surgical interventions in general [23, 24]. The results of this study are supported by earlier findings, that smoking is major contributor to delayed wound healing and consecutively also infection [25, 26]. Smoking seems to have an all-or-nothing effect, with several studies suggesting to quit smoking rather than reduce the number of cigarettes per day [27, 28]. Studies have found cessation to be tremendously beneficial before the age of 40, almost eliminating the risk of losing a life-decade, and to be beneficial after the age of 70 [29, 30]. This evidence supports our finding that former smokers have similar complication risks than never smokers and emphasizes the beneficial effects of smoking cessation.

Apart from smoking status, the ASA status, the surgeon’s expertise, case complexity, the hospital environment, and other patient factors (comorbidities) may be important confounding variables, for which we were not able to do adjusted analyses due to a small sample size. Regarding comorbidities, the authors were unable to gather enough precise information to adequately include an analysis. A differentiation in preexisting disease and disease having developed postoperatively was not possible, wherefore proper interpretation would not have been meaningful. The importance of ruling out confounding bias should be emphasized, but the similar results of previous papers support our findings although not having been able to adequately rule out confounding bias.

This study has revealed higher revision rates (18.2%) than mentioned in the previous literature with 6.45% at 5 years and 3–12.9% at 10 years [31, 32]. This could be attributed to a longer follow-up of up to 17 years, a combination of short- and long-term complication-based revisions or an inclusion of complications as defined by Goslings and Gouma and Sokol and Wilson [14, 33]. The higher rate of soft-tissue complications than reported in previous investigations [31, 32] could be explained by smoking being a major cause of soft-tissue complications and a smoking prevalence of 25% in Austria [25, 26].

Previous investigations have reported an age difference between smokers and never smokers regarding primary TKA [7, 9, 20, 34], indicating a particular need to investigate whether smoking should be defined as an independent risk factor of knee osteoarthritis. We did not collect information about the Kellgren-Lawrence classification of individual patients. Therefore, we cannot say whether active smokers had minor or more severe grades of osteoarthritis possibly leading to TKA.

Interestingly, our secondary outcomes resulted in current smokers scoring significantly better at KSFS and SF-12PCS and reporting higher pain ratings prior to surgery. Active smokers score 15.3 points higher than never smokers for KSFS and 4.4 points higher for SF-12PCS. The minimally clinically important difference (MCID) for primary TKA was found to be 9 points for KSKS and 10 points for KSFS [35]. For the SF-12PCS, it was 4.5 and 4.8 points for the pain relief and function sections, respectively [36], rendering these findings as not clinically important. However, active smokers seem to have a clinically better functional outcome than never smokers. Previous research by Matharu et al. found no clinically important differences in patient-reported outcome measures between active smokers, former smokers, and never smokers [11]. Smokers tend to have an unhealthier lifestyle than never smokers and are more satisfied with a lower functional level, possibly distorting the test results [37]. Furthermore, smokers tend to receive primary TKA at a younger age than never smokers. As smokers generally tend to report higher pain scores than never smokers, the younger, actively smoking patients may achieve—owing to their better health status than older never smokers requiring TKA—higher functional ratings before TKA as well as during follow-up [22, 38].

The findings of this study indicate an overall beneficial effect of TKA on functional outcome regardless of smoking status, represented in a nearly even distribution of PROMs across all smoking status groups. However, as PROMs prior to surgery were not available, the authors could not evaluate whether smoking status groups respond differently to TKA.

The following limitations must be underlined: The retrospective design of the study produces a low level of evidence, as does the limited number of smoking patients, thus eventually reducing the generalization and reproducibility of results obtained. A risk analysis regarding the number of cigarettes smoked per day was not possible due to the algorithm during initial assessment only revealing smoking status. The percentage of smokers could have been underestimated, as patients could have defined themselves as a non-smoker either due to having stopped smoking or smoking cigarettes occasionally. It is possible that the percentage of patients who had to be revised and were smokers is overestimated as patients tend to start smoking in adolescence, but more likely stop smoking with growing age rather than starting it [39, 40]. The limited patient number impairs the comparability of the SSI rate obtained in comparison with the literature. Yet, an advantage of information obtained from the present cohort can be seen in its uniformity. As a benefit we want to mention that all complications had to be recorded due to the authors’ healthcare system, which allows reimbursement after adequate classification of the diagnosis-related groups only.

Although the overall revision risk was not significantly higher in active smokers, soft-tissue complications were significantly more common in active smokers than never smokers. The slightly better results observed for smokers regarding functional outcome after primary elective TKA warrant further research to define the significance. We strongly recommend surgeons to advise their patients toward quitting smoking to maximize the success of primary TKA and minimize complication risks. The present findings underline the detrimental effects of smoking on postoperative complications, and ongoing research on the effects of nicotine abuse in orthopedic patients further strengthens the stance against smoking.