CORRELATION BETWEEN COMPLICATION RATE AND PERIOPERATIVE RISK-FACTORS IN SUPERIOR PEDICLE REDUCTION MAMMAPLASTY: OUR EXPERIENCE IN 127 PATIENTS

INTRODUCTION

Reduction mammaplasty, one of the most common plastic surgery procedures, has been shown to confer significant sustained health benefits for patients with symptomatic breast enlargement (26, 29) providing a remedy for back, neck and shoulder pain. Unfortunately, complications and poor aesthetic outcome including infection, hematoma, seroma, dehiscence, fat necrosis, and skin loss may occur in as many as 50% of patients (8). Unacceptable scarring has also been reported (5).

In addition, today the goals of breast reduction have evolved to include optimal shape, minimal scarring and preservation of the sensibility and function of the breast.

The delineation of variables associated with complications following breast reduction surgery remains questioned. Several studies of breast reduction operations documented that increasing age, obesity, smoking status, body mass index, weight of breast tissue resected, pedicle type, keyhole incision, free nipple grafting, operative time, use of epinephrine, drains and liposuction all appear to affect either the rate of development of postoperative complications and subsequent healing. Complication data reveal several significant features, although there is not a great deal of objective evidence to support this (6, 10, 18, 20, 30, 33). The aim of this study was to find some variables implicated in morbidity following breast reduction operations.

MATERIALS AND METHODS

Over the period 2004–2008, 127 consecutive patients were admitted for breast reduction surgery. Patients treated after radical mastectomy for breast cancer were excluded from this study.

We employed a superior pedicle inverted T. All patients underwent bilateral reduction mammaplasty without local vasoconstrictor infiltration and without drains. An elastic bandage was applied. Intravenous antibiotics were administered pre- and post-operatively. Operative details such as the length of the procedure, the intra-operative blood loss and the weight of tissue excised were recorded, as well as hospitalization, and the development of early (within 30 days) postoperative complications were noted. Fifty-one patients also underwent an ultrasound examination of the breasts at 48 h postoperatively to identify the presence of clinically undetectable collections. After discharge from the hospital, every fourth day the wound was reviewed at the dressing clinic for 2 weeks after the operation. In case of any wound-related complication(s), the patient was followed up and the progress of healing recorded.

Patients were divided into groups for each of the following variables: smoking, obesity, age, co-morbidities (especially diabetic status) and breast-volume reduction. Body mass index (kg/m²) was calculated for every woman at the time of surgery. Women were classified as being of normal weight (BMI 20–25), overweight (BMI 26–30), or obese (BMI greater than 30) (28). Data were analyzed retrospectively to determine whether any correlation could be found between complication rate and perioperative risk-factors using a multifactorial ANOVA F-test (1, 11, 13, 31). Complication rates were analyzed according to each risk-factor, and risk-factors of patients with complications were compared with those of patients without complications. The nominal data were cross-classified in a contingency table and analyzed. Power analysis was based on the comparison of proportions. Single one-way contrast between proportions was used to test the trend. P < 0.05 was considered statistically significant.

RESULTS

Bilateral breast reduction was performed in all patients: in total surgery was performed on 254 breasts. The mean age of our patients was 40 years (range 18–65 years), their mean body weight was 73 kg (range 50–123 kg). Total time in theatre ranged from 85 min to 250 min with a mean of 119 min and a mean blood loss of 175 ml per patient. The mean mass of tissue excised was 780 g. The mean hospital stay, including the day of operation, was 3 days (range 2–4 days). Forty-seven patients had at least one complication. Forty-three patients (33.8 percent) had one complication, and four (3 percent) had two complications. The overall complication rate was 20% in terms of breasts undergoing surgery and 37% in terms of patients. The incidences of specific complications were as follows: 30 delayed wound healing (23.6 percent), 8 T-junction necrosis (6.2 percent), two hematoma (1.5 percent), one partial nipple necrosis (0.8 percent), five partial necrosis of one areola (3.9 percent), two fat necrosis (1.5 percent), one seroma (0.8 percent), and two infections (1.5 percent).

One expanding haematoma required urgent surgical intervention and was evacuated under intravenous sedation anaesthesia. All the other complications recovered without interventions. There were six cases of partial nipple or areola necrosis which healed uneventfully with conservative treatment. A seroma, extending for 2–3 cm along the deep part of the incision, developed 10 days postoperatively. Successful aspiration of a small amount of fluid was performed by a consultant radiologist as an outpatient procedure. One patient developed an abscess 6 days postoperatively, requiring incision and drainage under general anaesthesia. In the two cases identified, fat necrosis was a clinical or mammographic diagnosis applied to small discrete masses superior to the nipple-areolar complex (presumably at the glandular pedicle). Delayed healing at the T-junction occurred in 8 patients. Two of these required secondary excision under local anaesthesia. Therefore a total of 11 patients underwent minor revision for hypertrophic scar or contour deformities (8.7%). Hypertrophic scars, occurring in the first postoperative year, were the most common complication after delayed wound healing and T-junction necrosis seen in this series, occurring in 24% of these patients (9/38). One patient developed major distortion of the areola that also required revision of the circumareolar scar. In the two cases identified, fat necrosis was a clinical or mammographic diagnosis applied to small discrete masses superior to the nipple-areolar complex (presumably at the glandular pedicle). The follow-up period ranged from 6 months to 4 years, and no further problems were identified during this time. Nipple and areolar sensibility was unchanged after surgery in all patients. Complications were summed up for each risk factor, and risk factor-related events were analyzed in separate repeated analyses of variance (ANOVAs) because they represented specific areas relevant to the hypotheses. Unvariate analysis of factors such as age, smoking status, weight of breast tissue resected and operative time failed to demonstrate a correlation with an increased incidence of complications. Comparative unvariate analysis of associated variables did not reveal significant differences in age (p=0.09), smoking status (p=0.38), resection weight (p=0.07) or operative time (p=0.46). Just over a quarter of patients in the study were smokers (27.5%) with the rest being non-smokers (72.5%). Both groups were statistically similar with respect to age and BMI. The mean number of cigarettes consumed by smokers was 15.7. Two patients in the study were diabetic, both in the non-smoker group. 27% of smokers and 39% of non-smokers had wound complication(s). Overall smokers were not more likely to develop any complication (p=0.38). Analysis of variables associated with delayed wound healing, the most common complication, revealed the following: the average age of patients with delayed wound healing was 24.4 in comparison with an average age of 42 for those who did not have delayed wound healing (p=0.02). Average resection weight of breast tissue was 950 g for those with delayed wound healing versus 805 g for those without delayed healing (p=0.07). Although not statistically significant, the strength of the association of weight tissue resected with complications was considerable.

Analysis of variables associated with complications showed that, after adjusting for age and smoking status, only BMI was associated with any complication (p<0.05).

DISCUSSION

Obesity is widely recognized as epidemic in the Western world. There has been a dramatic rise in the worldwide prevalence of obesity, leading to a World Health Organization (WHO) declaration that obesity is a major killer disease of the millennium on par with HIV and malnutrition. With an increasing obese population, plastic surgeons are consulted more frequently by women requesting more substantial breast reductions. The BMI of patients requesting reduction mammaplasty is greater than in the control population, while an increase in mean weight and a decrease in mean height are noticed in mammaplasty patients (3). Obesity is frequently characterized by the regain of body weight following weight reduction, so-called “weight cycling”. Obese and formerly obese patients requesting mammaplasty represent an objective deviation from “normality” rather than a subjective alteration of their body image. There are various indications for mammaplasty, including potential psychological benefits. The reported complication rates for reduction mammaplasty range from 5% (26) to 53% (8).

The overall complication rate of 37% in this series is consistent with that reported in the literature.

The correlation between obesity and complications is corroborated by our study. Increasing BMI was associated with an increased risk and absolute number of complications.

The average BMI in the current series was 28.33, which is similar to that reported by other authors: 27.5 by Schumacher (28), 26.7 by Chan (4) and 29.7 by Cunningham (6).

Obesity and specimen weight have both been associated with a higher incidence of complications for patients undergoing reduction mammaplasty. Previous retrospective analyses of postoperative sequelae following breast reduction operations revealed two distinct and statistically significant associations between increasing resection weight of breast tissue, obesity and postoperative complication rate (7, 10, 16, 18, 19, 30, 33). In 2000 Zubowski et al. (32) analyzed 395 reduction mammaplasty procedures over a 10-year period and found that, although lean patients had fewer complications than obese patients, there was no increasing incidence of complication with increasing degrees of obesity. Other studies failed to consistently correlate obesity with an increased risk of complications (22, 32). In 2008 Roehl et al. also noted that there was no significant increase in the rate of complications for each body mass index group based on the reduction mass (24).

In 1964 Strombek (30) analyzed 1042 breast reductions and found that resection weights greater than 500 g per breast were at an increased risk for complications. Dabbah (7) also noted a significant increase in complication rate when more than 1000 g of tissue was removed from each breast; this study did not differentiate between procedure subtypes. In 1997 Schnur et al. (26) noted a trend toward an increased complication rate among 328 patients, where the mean resection weight for patients with complications was 1500 g versus 1329 g for those without complications (p=0.082). In 2000 Zubowski et al. (33) described the relationship between resection weight and both complication rate and major complications (p=0.0003 and 0.0002). Finally, Menke et al. (20), in a prospective study of 799 patients, noted an increased complication rate with larger reductions. Our retrospective study failed to show any statistically significant difference in the complication rate between smokers and non-smokers. The relationship between smoking and complications previously documented was not apparent in our data (2). There was no association between smoking status and incidence of complications, total number of complications, or specific complications such as delayed healing. Nevertheless the detrimental effect of smoking on various plastic surgical procedures (15, 23) has been reported in the past with relation to wound healing (12, 21). A statistically significant correlation has been established between patient age, measured preoperative BMI and incidence of complications. The association between youth and delayed healing contradicts the other studies literature, in which both age and procedure time have been found to impair wound healing (9, 14, 17).

CONCLUSION

It is certain that the definition of risk factors of reductive mammaplasty will continue to be a source of continuing research, and we hope this paper will contribute usefully to the debate for the ultimate benefit of patients. No previous study has conclusively identified risk factors and clearly demonstrated the potential effect of complications on patient satisfaction with the operation or with potential effect on health. This is due to the fact that it is very difficult to accurately define and comment on delayed healing and cosmetic outcome of breast reduction surgery in a retrospective studies, and thus the need for controlled, prospective data arises.

A great many differences in operative technique, including free nipple grafts, pedicle type, operative time and drains, may be significantly associated with complications. A search of issues relating to breast reduction techniques reveals approximately 72 different ones. We prefer the superior pedicle procedure; however, other procedures are also reliable, although accompanied by different complications. The literature features a large number of retrospective studies showing that too many variables impair wound healing. Therefore many studies have several limitations. A non-random sample is subject to selection bias, and small sample size limits the possibility of generalizing the results. In our study the overall complication rate was 20% in term of number of breasts undergoing surgery and 37% in terms of the number of patients. Analysis of variables associated with complications showed that, after adjusting for age and smoking status, only BMI was associated with any complication (p<0.05).

Address for correspondence:

Fabio Caviggioli, M.D.
Cattedra di Chirurgia Plastica, Università degli Studi di Milano – Unità Operativa di Chirurgia Plastica 2,
IRCCS Istituto Clinico Humanitas
Via Manzoni 56
20089 Rozzano (MILANO), Italy
E-mail: fabio.caviggioli@humanitas.it