The Effect of Smoking on Sternal Scar Healing: A Prospective Cohort Study

Introduction. Cardiothoracic surgery with a median sternotomy is an electing factor for the development of a hypertrophic scar. Hypertrophic scars, characterized by an increased vascularity, often result in aesthetic and functional problems. Smoking, due to its negative effects on vascularization, could therefore have an effect on scar healing. Objective. A prospective cohort study was conducted to evaluate the effect of smoking on scar healing after cardiothoracic surgery with a median sternotomy incision. Materials and Methods. One hundred patients who underwent cardiac surgery with a median sternotomy were divided into 3 groups: smokers, ex-smokers, and nonsmokers. Erythema values of the scar were measured with a colorimeter on 3 standardized parts of the scar. Scar evaluation was performed at 6 weeks, 3 months, 6 months, and 12 months after surgery. Results. During 1 year, a total of 90 patients were followed after a median sternotomy; 10 patients were lost to follow-up. There were 23 smokers, 52 ex-smokers, and 15 nonsmokers with an overall mean age of 61.5 ± 8.83 years. No significant difference in redness as a parameter for hypertrophic scarring was observed between the 3 groups. Nevertheless, a trend in favor of the smokers was seen, as they developed less hyperemic scars. The caudal part of the scar showed a significantly higher incidence of hypertrophy compared with the middle and cranial part of the scar (P < .001) at all time points. Conclusions. It is presumed that a large sample size with younger patients is needed to confirm the results herein. Furthermore, more caudally located skin, especially the subxiphoidal part, is prone to hypertrophic scarring and should, for that reason, be avoided in the incision.

RELATED CONTENT

Temporal Activation of Senescence-associated Secretory Proteins in a Post Burn Hypertrophic Scarring Model

Review of Silicone Gel Sheeting and Silicone Gel for the Prevention of Hypertrophic Scars and Keloids

The Thioredoxin and Glutaredoxin Systems in Smoking Cessation and the Possible Relation to Postoperative Wound Complications

Presternal scarring after cardiothoracic surgery is known for its aberrant scar formation that can lead to hypertrophic or keloid scar formation. These scars are notorious for their functional and aesthetic problems and associated reduced quality of life.^1-3

Hypertrophic scars develop frequently after full-thickness burn injuries and surgery performed at locations where skin tension is largest.^4,5 The border of a hypertrophic scar remains typically within the confines of the original lesion in contrast to keloids, which are characterized by expansive growth and a genetic predisposition. The thorax is one of the preferential locations for hypertrophic scar formation. These scars are seen in 10% to 57% of Caucasian patients after cardiothoracic surgery.^2,6,7

The exact etiology for aberrant scar formation remains unclear. One of the predicting factors is an increased vascular density compared with normotrophic scars.³ Smoking may have a negative effect on the vascularization of a wound due to endothelial dysfunction and is thought to be responsible for a significant increase in postoperative wound complications.^8,9 However, as a side effect, due to its impact on vessels, smoking also may lower the risk of developing hypertrophic scarring.

To examine this hypothesis, a prospective study was conducted in which both redness, as parameter of hypertrophy, and the incidence of scar hypertrophy among smokers, ex-smokers, and nonsmokers after a median sternotomy were evaluated.

From September 2010 until September 2011, a total of 100 patients were followed after cardiothoracic surgery at Maastricht University Medical Center (Maastricht, The Netherlands) and included in the study. A power analysis was performed using a previous pilot study,¹⁰ in which the redness of the scar after reduction mammoplasty was measured. The Medical Ethical Committee of Maastricht University Medical Center approved the study (project no. METC 10-4-009), and patients were included only after written, informed consent was obtained.

Cardiothoracic patients were contacted on postoperative day 5 before discharge. Patient inclusion criteria were age > 18 years and cardiac surgery through a median sternotomy incision. Patients with a history of hypertrophic scarring or keloid formation, diabetes mellitus, degenerative or metabolic diseases, or preexisting skin disorders and patients using corticosteroid or immunosuppressive medication were excluded from the study. Ex-smokers were defined as having quit smoking at least 1 month prior to the operation, whereas smokers had to be smoking on the day of surgery or quit smoking within 1 month prior to surgery. 

Objective measurement of the scar was performed by measuring the redness of the scar at 6 weeks and at 3, 6, and 12 months postop. The exact day after surgery was documented at every time point. A Minolta CR-300 (Minolta Camera Co, Ltd, Osaka, Japan) was used for measurements. These measurements were performed under standardized conditions at 3 locations on the scar: 2 cm from the cranial end of the scar; at nipple height (or in women with ptotic breasts, a height corresponding with half the length of the humerus was used as normal nipple height); and 5 cm from the caudal end of the scar.

Coloric measurement was not performed in case of incomplete wound healing of the wound in a particular area. Normal skin color was measured near the scar at each time point and subtracted from the measured values of the different measurement points of the scar. The Patient and Observer Scar Assessment Scale (POSAS) was used to evaluate the scar at each time point. Six weeks post surgery, patients started to apply a hydrating cream (Alhydran; BAP Medical, Apeldoorn, Overijssel, The Netherlands) on the scar twice daily. In case of hypertrophic scar formation, treatment was not deferred and an alternative treatment, such as silicone dressings (Scarban Elastic; SCARPRO N.V., De Pinte, Belgium), was proposed. Hypertrophic scarring was defined as a situation in which the scar, or a part of the scar, is raised above the level of the surrounding skin. The size of the scar had to remain within the original border of the incision. If this last condition was not fulfilled, a scar was deemed as having more keloid instead of hypertrophic characteristics.

Statistical analysis
The analysis was based on patients with known smoking status (smoker, ex-smoker, nonsmoker). Four response variables were analyzed: “base redness,” which was calculated as measured redness of the scar minus the redness of the normal skin measured at the corresponding location (cranial, nipple, caudal), and 3 POSAS components (color, thickness, and irregularity). Time (6 weeks, 3 months, 6 months, 12 months) was scaled to months for computational reasons and the evolution of scar redness (with normal skin measurement subtracted) and the 3 components of the POSAS by smoking status over time was analyzed using a linear model correcting for patient age, location (only for base redness), and length and width of the scar. Time was modeled continuously in the mean evolution. The assumption that time was linear was tested by adding a squared time effect, including the interactions with smoking status and location (only for base redness), to the models and jointly testing the null hypothesis that time was linear. An unstructured variance-covariance matrix was assumed using a categorical version of time point (6 weeks, 3 months, 6 months, 12 months).

The POSAS was summarized descriptively by mean, standard deviation, median, quartiles, range of numbers, and percentage, whichever was appropriate.

A Fisher’s exact test was used to analyze the difference between location (cranial vs. middle together with the caudal part) within every patient with a hypertrophic scar.

One hundred patients with a median sternotomy were recruited and provided informed consent for study participation. During the study, a total of 10 patients were lost to follow-up. Therefore, the study sample consisted of 90 patients. Of the enrolled patients, the distribution of men versus women was 81.1% (73) versus 18.9% (17), respectively. There were 23 smokers, 52 ex-smokers, and 15 nonsmokers with an overall mean age of 61.5 ± 8.83 years. No single wound infection occurred in the patient population. Patients’ baseline characteristics are shown in the Table. No significant difference was found in regression of redness over time as an indicator for hypertrophic scarring among the groups (Figure 1). There was no significant difference in incidence of hypertrophic scars between the 3 groups (Figure 2).

The caudal part of the scar showed a significantly greater incidence of hypertrophy than the middle and cranial parts of the scar (P < .001) at all time points (Figure 3). There was no significant difference between the groups for the subjective and objective scar assessment scale (POSAS).

In this prospective study, the difference in scar healing between smokers, ex-smokers, and nonsmokers after median sternotomy was studied. No significant difference in redness, as an indicator for hypertrophic scarring,¹¹ was seen among the groups. There are several factors that may have had an effect on the outcome of the present study.

At 6 months postop, a trend towards hyperemic scars in nonsmoking patients was observed. The possible positive influence of smoking on reducing hypertrophic scarring can be explained; smoking reduces the systemic inflammatory response as nicotine, a vasoconstrictor, reduces proliferation and migration of macrophages and fibroblasts, and subsequently, a reduction in collagen deposition occurs.¹²

van der Veer et al³ investigated the time course of the angiogenic response in wound healing in both hypertrophic and normotrophic scars of human median sternotomy incisions. They established that hypertrophic scars were associated with increased neovascularization compared with normotrophic scars. Microvessel density in the deeper part of the reticular dermis was significantly increased in the hypertrophic group at 12 weeks and 52 weeks. Unfortunately, in that study,³ no measurements were taken at 6 months. Nevertheless, and in accordance with the present study, it shows there is an increased vascular density after 3 months in hypertrophic scarring compared with normotrophic scarring in median sternotomy incisions. In addition, the present authors have shown nicotine, a major constituent of tobacco smoke, has inhibitory effects on wound growth factor expression that can result in reduced scar neovascularization and thus reduced scar hypertrophy.¹³

Although the present authors did not find statistical evidence that smokers are less susceptible to hypertrophic scars after a median sternotomy, a consistent trend indicates that it should be possible to confirm this hypothesis with a larger sample of patients. One of the reasons for lack of power might be that the power analysis was based on a population of women between 18 and 55 years old.¹⁰ In this study, the mean age of women was 37.4 ± 13.2 years for smokers and 32.6 ± 12.7 years for nonsmokers. This is in contrast with the current study consisting of a mixed population with a mean age of 61.5 years, which is more sensitive to develop hypertrophic scars because this form of scarring has the highest incidence in the second decade of life.¹⁴

Although it was not a separate item in this study, the authors found the caudal part of the sternotomy scars showed twice the incidence of hypertrophy (P < .001) than the cranial and central parts. This is in accordance with the literature, in which a tendency towards hypertrophic scar formation is seen in the lower half of the incision.^2,15 Nevertheless, skin tension is supposed to be less over the caudal part of the sternum, and therefore, a better cosmetic result should be achieved in this area. However, increased mobility seems to be the predominating factor for the development of a hypertrophic scar.¹⁶ Based on the literature and present findings, the authors suggest minimizing the length of the incision where the caudal end has to be proximal of the xiphoid process.

Although this study is the first to evaluate scars by means of an objective measure, no strict specification in regard to the results was utilized. An alternative objective measurement could have been performed, with biopsies being most specific for determination of scar activity. Besides, the results show the study population does primarily consist of male patients, while in females more traction on a sternal wound could be assumed; further, the patients were relatively elderly. However, these 2 characteristics of the study population are inherent to the procedure studied. Altogether, with an insufficient power to draw strong conclusions from, it is recommended that further extensive research with sufficient power on a more representative (ie, a younger and more sex-balanced) study population should be performed.

No significant difference in redness as a parameter for hypertrophic scarring was seen between the 3 groups. However, a trend in favor of smokers developing less hyperemic scars was seen in this study and, therefore, a larger sample of patients is needed to confirm these results.

In addition, the caudal part of the scar, especially the subxiphoidal part, is prone to hypertrophic scarring and should be avoided in planning the incision.

Authors: A.E.K. Deliaert, MD, PhD^1,2; J.F. Mermans, MD³; S.J. Schop, MD, MSc²; T.S. Dormaar, MD⁴; E.M. Heerdt, BSc⁵; S.A. Xanthoulea, PhD²; J.G. Maessen, MD, PhD⁵; E. van den Kerckhove, PhD^2,6,7; and R.R.W.J. van der Hulst² 

Affiliations: ¹Department of Plastic, Reconstructive, and Hand Surgery, VieCuri Medical Center, Venlo, The Netherlands; ²Department of Plastic, Reconstructive, and Hand Surgery, Maastricht University Medical Center, Maastricht, The Netherlands; ³Department of Plastic, Reconstructive, and Hand Surgery, VU Medical Center, Amsterdam, The Netherlands; ⁴Department of Plastic, Hand, and Microsurgery, Friederikenstift, Hannover, Germany; ⁵Department of Cardiothoracic Surgery, Maastricht University Medical Center; ⁶Department of Rehabilitation Sciences and Kinesiology, KU Leuven, Leuven, Belgium; and ⁷Department of Physical Medicine and Rehabilitation and Burns Centre, KU Leuven

Correspondence: A.E.K. Deliaert, MD, PhD, Department of Plastic, Reconstructive, and Hand Surgery, Maastricht University Medical Center, Maastricht, The Netherlands; andeliaert8@hotmail.com

Disclosure: The authors disclose no financial or other conflicts of interests.