Abstract
Background:
Cigarette smoking is known to increase perioperative complication rates, but no study to date has examined its effect specifically in forefoot surgery. The purpose of this study was to determine whether cigarette smoking increased complications after forefoot surgery.
Methods:
The records of 602 patients who had forefoot surgery between 2008 and 2010, and for whom smoking status was known, were reviewed. Patients were categorized into 3 groups based on smoking status: active smoker, smoker in the past, or nonsmoker. Medical records were reviewed for occurrence of complications, including nonunion, delayed union, delayed wound healing, infection, and persistent pain.
Results:
Active smokers were found to have a notably higher complication rate (36.4%) after forefoot surgery than patients who previously (16.5%) or never (8.5%) smoked. Patients who continued to smoke in the perioperative period had the highest percentage of delayed union (3.0%), infection (9.1%), delayed wound healing (10.6%), and persistent pain (15.2%). Active cigarette smokers were 4.3 times more likely to have a complication than nonsmokers. Patients who smoked at any point in the past but quit prior to surgery were 1.9 times more likely than nonsmokers to incur a complication. The average time of smoking cessation for patients who had smoked at any point in the past but had quit prior to surgery was 17 years. For active smokers, those with a complication smoked an average of 18 cigarettes daily, while those without a complication smoked 14 cigarettes daily.
Conclusions:
Before forefoot surgery, surgeons should educate patients who smoke about their increased risk of complications and encourage smoking cessation.
Level of Evidence:
Level III, retrospective comparative study.
The negative effects of cigarette smoking on overall health have been known since the 1960s, and recent literature has reported the specific effects on the musculoskeletal system and on outcomes of orthopedic procedures.3,4,8,11,13,14,16,18,23,28 The 2011 National Health Interview Survey found that 19% of adults smoked cigarettes, with no notable change in prevalence from 2010. 3 Over $96 billion in direct medical expenses and $97 billion in lost annual productivity are attributed to cigarette smoking in the United States. 4 Nicotine, 1 of more than 4000 chemicals released during cigarette smoking, has addictive dopaminergic effects in the central nervous system. 13 Combined with carbon monoxide, this causes peripheral vasoconstriction, which reduces the amount of oxygen available to tissues and leads to microclotting. 16 This negative effect has been shown to be reversible, though. 11 Hydrogen cyanide, found in the volatile phase of cigarette smoke, interferes with cellular metabolism, and smoking negatively affects the immune response through its effect on immunoglobulins and natural killer-cell activity.8,14,23
Animal studies have demonstrated the negative effect of nicotine on bone and skin healing,6,7 and clinical studies have demonstrated the deleterious effects of cigarette smoking on outcomes in arthroplasty, spine surgery,1,2,21 and a variety of other orthopedic conditions and procedures, including fracture healing, hand and foot surgery, and ligament and cartilage repair. Slower healing has been reported in smokers than in nonsmokers after transmetatarsal amputation, bunionectomy, and hindfoot fusion,15,17,33 but no study to date has examined the effect of smoking on the complication rate after forefoot surgery.
Two meta-analyses in the nonorthopedic literature clearly indicate benefits of smoking cessation in reducing postoperative complications. Both Mills et al 20 and Wong et al 31 reported notable reductions in wound healing complications with at least 4 weeks of smoking cessation. Møller et al 22 and Lindström et al 19 found noteworthy reductions in wound-related complications in patients who had total hip or knee replacement with a 6- to 8-week smoking cessation program before and/or after surgery. In 160 hindfoot fusions, Ishikawa et al 15 found that the nonunion rate decreased from 19% to 11% in those who quit smoking before surgery. A 2014 Cochrane Database Review 30 concluded that smoking cessation interventions beginning 4 to 8 weeks before surgery could notably reduce postoperative complications.
The purpose of our study was to determine the effect of smoking status on surgical outcomes in a large elective foot and ankle practice in order to provide practitioners with general numbers from which they would be able to formulate strategies to better educate their smoking patients before surgery. We hypothesized that smokers would have a higher total complication rate than nonsmokers after surgery of the forefoot and that quitting smoking before surgery would lower the complication rate.
Methods
This study was approved by our institution’s institutional review board before data collection began. All patients who had forefoot surgery and for whom cigarette use could be determined from the medical record were included. A Current Procedural Terminology (CPT) code search for all forefoot operative procedures performed between 2008 and 2010 was used to compile a list of patients for a retrospective medical record review (Table 1).
CPT Codes and Procedures Included.
Abbreviations: CPT, Current Procedural Terminology; IP, interphalangeal; MTP, metatarsophalangeal.
All surgeries were done by 3 fellowship-trained orthopedic foot and ankle surgeons. Medical records were reviewed to determine patient demographics, including age, sex, and presence of comorbidities. Comorbidities recorded were those that were known to have a negative impact on operative outcomes, including diabetes mellitus, rheumatoid arthritis, peripheral vascular disease, peripheral neuropathy, and chronic steroid use as detailed on the intake history and physical. All patients had palpable pulses in the operative foot before surgery, or vascular consultation was obtained before any surgical intervention. Smoking status was determined from the intake history, and subjects were classified as active smoker, smoker in the past, or nonsmoker as reported by the patient. Patients were classified into 3 groups based on chronology of cigarette smoking related to their forefoot procedure. Group 1 patients had no history of cigarette smoking. Group 2 included patients who previously smoked cigarettes but had stopped at some point before the date of surgery. Group 3 patients continued to smoke in the perioperative period. Medical records were then reviewed from the immediate postoperative visit through the latest follow-up visit. The length of follow-up was recorded for each patient.
The retrospective search of CPT codes identified 633 patients who had forefoot procedures between 2008 and 2010. Status of cigarette smoking could be determined from the medical records for 602 patients (95%); 21 patients did not answer the smoking questions on the intake history and were excluded from the study. The average age of the 475 females and 127 males was 53.1 years. Group 1 (nonsmokers) contained 457 patients; group 2 (previous smokers), 79 patients; and group 3 (current smokers), 66 patients (Table 2). Group 2 had a statistically higher percentage of patients with diabetes and peripheral neuropathy (P = .046 and .0003, respectively); however, the percentages of patients with rheumatoid arthritis or reporting chronic steroid use were similar (Table 3).
Demographics of the Patient Groups.
Group 1, nonsmokers; group 2, previous smokers; group 3, current smokers.
Comorbidities (%) in Each Patient Group.
Values are percentages. Group 1, nonsmokers; group 2, previous smokers; group 3, current smokers.
Outcome measures were complications that occurred including nonunion, infection, delayed wound healing, delayed union, and persistent pain as detailed by the surgeon in the medical record. The presence of each complication was recorded to allow calculation of total complications and complication rate. Nonunion, delayed wound healing, and delayed union were considered to be present when the primary surgeon had documented each in the medical record during a follow-up visit. Infection was considered to be present when documented by the primary surgeon at follow-up and treated with antibiotics. Persistent pain was defined as pain great enough to be reported by the patient as leaving him or her dissatisfied with the outcome at latest follow-up in the absence of any other complication. To account for patients who may have had more than one complication and to avoid artificial inflation of the complication rate, the complication rate was calculated as the number of patients with any complication divided by the total number of patients. A power analysis using a beta of 20% and a P value of .05 was performed, as were a chi-square analysis and Fisher’s exact test, with a P value of less than .05 indicating significance. A multiple logistic regression analysis was performed on the individual data at subject level with complications as the outcome and the smoking groups and medical comorbidities as explanatory variables. Relative risk also was calculated.
Results
The multiple logistic regression analysis demonstrated that medical comorbidities had no significant effect on complications when smoking was considered (Table 4). Power analysis confirmed adequate sample size to detect significance. The numbers of patients with any complication in each group were as follows: group 1, 39 (8.5%); group 2, 13 (16.5%); and group 3, 24 (36.4%). Delayed union, nonunion, infection, delayed wound healing, and persistent pain in the absence of any other complication were all significantly more frequent in group 3 (Table 5 and Figure 1). Chi-square analysis showed a significant difference in total complication rate between groups (P = .0001) in addition to a significant difference in each type of complication between groups. Compared with nonsmokers, patients who actively smoked in the perioperative period were 4.3 times more likely to have a complication after forefoot surgery. Patients who smoked at any point in the past but quit before surgery were 1.9 times more likely than nonsmokers to incur a complication. The relative risk of active smokers (compared with nonsmokers) to have each complication was 6.9 for delayed union, 6.9 for nonunion, 4.6 for infection, 9.7 for delayed wound healing, and 3.8 for persistent pain. The median duration of cessation prior to surgery for patients who had smoked at any point in the past but quit prior to surgery was 17 years. For active smokers, those with a complication smoked an average of 18 cigarettes daily while those without a complication smoked 14 cigarettes daily.
Multiple Logistic Regression Analysis.
Complications in Each Patient Group.
Values are percentages. Group 1, nonsmokers; group 2, previous smokers; group 3, current smokers.
Pain, delayed wound healing, and infection were dramatically increased in patients who smoked.
Discussion
Cigarette smoking has been shown to negatively affect outcomes of orthopedic procedures, but its effect on forefoot surgery has not been previously reported. In our findings, active cigarette smokers were 4.3 times more likely to have a complication after forefoot surgery than were nonsmokers, with a 36.4% total complication rate. Patients who continued to smoke in the perioperative period had the highest percentages of delayed union (3.0%), infection (9.1%), delayed wound healing (10.6%), and persistent pain (15.2%). It is unclear why group 2 had a slightly higher percentage of nonunion (2.5%) than group 3 (1.5%), but this may be related to the notably higher percentage of patients with diabetes in group 2.
In an effort to stratify patients by risk, a multiple logistic regression analysis was performed using complications as the outcome and medical comorbidities and smoking groups as the explanatory variables. As demonstrated in Table 4, medical comorbidities had no statistically significant impact on complications when smoking was considered. Although our overall sample size was relatively large, a larger, more focused study would be needed to examine the relationship between specific comorbidities and smoking on outcomes. For example, of the 602 patients included in the study, 66 identified as active smokers and only 4 patients in this group were diabetic. Although 2 of the 4 smokers with diabetes developed a complication, concluding that diabetic smokers have a 50% complication rate in forefoot surgery would be inaccurate as the study design was not powered to draw such conclusions.
Persistent pain was the most frequent complication in each group. Persistent pain was recorded as a complication only when it occurred in the absence of another complication. This eliminated any potential overlap with a patient who may have had pain due to nonunion, delayed union, or a wound problem. A notable increase in complication rate between groups remained even when persistent pain was removed from the calculation. Other studies have shown that smokers are more likely to have chronic back pain as well as persistent pain after lumbar spine surgery, but the occurrence of chronic pain has not been demonstrated with forefoot surgery.9,10,12,29 While the cause of this pain is unknown, it has been postulated that this amplified response to pain may be due to elevated levels of proinflammatory mediators seen in smokers.26,32
Medical records for group 2 were further reviewed to determine whether a link could be made between the period of time between smoking cessation and surgery and the risk of complications associated with surgery. The median period of time between smoking cessation and the date of surgery was 17 years. Six patients who stopped smoking at least 17 years before surgery had a complication, while 7 patients who stopped smoking less than 17 years before surgery had a complication. No direct correlation between duration of time between smoking cessation and the date of surgery could be drawn; a larger study would be needed to investigate this link. Similarly, group 3 was further analyzed to determine whether a higher daily quantity of cigarette smoking increased the complication rate. Patients were asked to estimate how many cigarettes they smoked each day. Patients in group 3 who had a complication reported smoking an average of 18 cigarettes per day, while those without a complication reported using 14 per day. Although patients with a complication smoked more cigarettes than those without a complication, the difference was not large and it is unclear how reliable a conclusion can be drawn because of reliance on patient self-reported tobacco use.
Other authors have studied the effect of cigarette smoking on complications after procedures in the hindfoot and ankle. Ishikawa et al, 15 in a study of 160 patients undergoing hindfoot fusions, showed that smokers had a notably higher rate of nonunion than nonsmokers. The relative risk of nonunion was 2.7 times higher for smokers than nonsmokers in their study. These authors noted a trend toward a higher nonunion rate in patients who quit smoking before surgery, but this did not reach statistical significance. No notable difference was noted between smokers and nonsmokers with regard to infection or wound healing. Cobb et al 5 found that the relative risk of nonunion was 3.75 times higher for smokers than nonsmokers in their study of 44 patients with ankle fusion. Nåsell et al 25 reported the effect of smoking on complications in 906 operatively treated ankle fractures with 6-week follow-up. Multivariate analysis showed that smokers had 6 times higher odds of developing a deep infection than did nonsmokers.
Although limited by its retrospective nature, our study does clearly show an increased risk of complications associated with cigarette smoking in patients with forefoot surgery. Smoking status could be determined from the medical records for 95% of the 633 patients; the exclusion of 21 patients who did not answer this question in the medical record could have introduced bias. The 11% of patients who identified themselves as smokers is lower than the 2011 Census data showing an average of 19% tobacco use in the United States. 3 It is possible that some patients may not have accurately reported their cigarette smoking, which would bias the collected data. If patients who smoked small quantities identified themselves as nonsmokers, then group 3 would be overrepresented by heavy cigarette smokers, which could overestimate the impact of cigarette smoking on the complication rate. If patients who actively smoked perioperatively identified themselves as nonsmokers but consumed the same average daily quantity of cigarettes as those who correctly identified themselves as active smokers, then we may have underestimated the impact of cigarette smoking on the complication rate. Although nicotine serum testing would have increased the accuracy of the data, this was not possible because of the number of patients involved in the study, as well as its retrospective design.
Previous studies have demonstrated that cessation of cigarette smoking before elective orthopedic surgery has a beneficial effect on reducing the postoperative complication rate. In a randomized trial of patients with hip and knee arthroplasty, 56 patients who underwent a smoking cessation intervention 6 weeks preoperatively had a notably lower complication rate (18%) than 52 patients in the control group (52%). 22 A randomized trial studying the effect of a smoking cessation program initiated during the acute hospitalization period for fracture surgery showed a complication rate of 20% in the cessation group compared with 38% in the control group. 24
Our study demonstrated that smokers have a significantly higher complication rate after forefoot surgery than do nonsmokers. Given the available evidence, every effort should be made by physicians to encourage smoking cessation. Although some surgeons believe that cessation counseling is best carried out by the patient’s primary care practitioner, a study of 10,000 spine patients showed that when surgeons placed a “high priority” on smoking cessation compared with a “low priority,” the quit rate increased from 19.5% to 35.6%. 27 The importance of surgeons’ taking the time to encourage smoking cessation cannot be overemphasized. Methods to consider include referral to online resources (www.smokefree.gov) and telephone hotlines (1-800-QUITNOW) and coordination with the patient’s primary care practitioner for medical treatment. The surgeon may code tobacco use disorder as International Classification of Diseases, Ninth Revision (ICD-9) code 305.1, and some insurance plans will reimburse for proper CPT coding. CPT code 99406 covers smoking cessation counseling of 3 to 10 minutes duration, while code 99407 covers more than 10 minutes of counseling. It is important for the practitioner to document the duration of counseling in addition to the resources provided.
In conclusion, although numerous factors can affect outcomes of forefoot surgery, smoking is perhaps the most easily modified by the patient. This is the first study to examine the effects of cigarette smoking on complications after forefoot surgery. In our findings, active cigarette smokers were 4.3 times more likely to have a complication after forefoot surgery than were nonsmokers, with a 36.4% total complication rate; persistent pain was the most frequently encountered complication. Before forefoot surgery, surgeons should educate patients who smoke regarding the increased risk of complications and should support patients through smoking cessation. Larger studies are needed to determine what duration of smoking cessation is necessary to return the complication rate to that of a nonsmoker.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.