Indian Journal of Cancer Medknow Publications on behalf of Indian Cancer Society ISSN: 0019-509X EISSN: 1998-4774 Vol. 44, Num. 3, 2007, pp. 99-103

Indian Journal of Cancer, Vol. 44, No. 3, July-September, 2007, pp. 99-103

Original Article

Factors predicting seroma formation after mastectomy for Chinese breast cancer patients

Loo WingsTY, Chow LouisWC

Comprehensive Centre For Breast Diseases, UNIMED Medical Institute, 10/F, Luk Kwok Centre, 72 Gloucester Road, Wanchai
Correspondence Address:Comprehensive Centre For Breast Diseases, UNIMED Medical Institute, 10/F, Luk Kwok Centre, 72 Gloucester Road, Wanchai, lwcchow@unimed.hk

Code Number: cn07016

Abstract

Keywords: Breast cancer, mastectomy, operation, postoperative drainage, seroma

Introduction

Materials and Methods

Patients intended for mastectomy for the treatment of primary breast cancer were recruited. The group developing seroma was compared with regard to epidemiological factors, operative factors and drainage output with a group without seroma [Table - 1].

Operation

Total mastectomy was performed as described by Forrest. ^[5] Axillary sampling was taken as a level I dissection and clearance was taken at all three levels. Reconstruction was performed using the pedicle Transverse Rectus Abdominis Myocutaneous (TRAM) flap technique. Each patient was managed with Hemovac closed suction drainage with two catheters from one suction device (Hemovac 400 ml Evacuator: Hemovac, Synder laboratories, USA). The catheters were inserted into the chest wound at the end of operation. Drains were retained until the 24h total output was less than or equal to 30 ml.

Seroma formation and its management

All patients were reviewed within two weeks after leaving hospital, unless seroma formation was detected before discharge. The diagnosis of seroma was made clinically when a collection was detected beneath the skin flaps. Aspiration under sterile condition was performed for both diagnostic confirmation and management. The aspirate was sent for bacterial culture and aspiration repeated weekly until no collection could be detected.

Statistical analysis

Student′s t test was used with continuous variables and the X ² test for categorical situations. Fisher′s exact test was applied when small numbers were encountered. A two-tailed test of P < 0.05 was considered significant. Univariate analyses were performed. The number of patients with seroma formation did not enable meaningful multivariate analysis to be carried out.

Results

A total of 119 patients with a mean age of 56.2 years (range: 26-92 years) were included in this study. Ten patients (8%) developed seroma after mastectomy, eight of whom required repeated aspirations [Table - 2]. ′Recovery′, defined as the day of drain removal in the control group and the day when the last aspiration was performed in those with seroma formation, was 4.5 (s.d.= 1.7) and 32.7 (s.d.= 20.7) days, from the day of operation, for the respective groups. The patients without seroma recovered significantly earlier than those with seroma (Wilcoxon rank sum W test, P =0.000).

Epidemiological factors

Patients older than 45 years old had a statistically significant higher chance of developing seroma than younger patients (Fisher′s exact test, P =0.02). However, no significant correlation was found between menstrual status and seroma formation.

Patients with hypertension were also more likely to develop seroma after mastectomy (X ² = 13.86, P = 0.001), but no such difference was found in patients with diabetes mellitus. Furthermore, statistical correlation was not found between the preoperative serum albumin concentration, hemoglobin concentration, disease stage, body weight and the development of seroma.

Operative factors

Of the four types of operation performed [Table - 3], the at-risk group for seroma were patients who had undergone axillary dissection and immediate reconstruction with the TRAM flap obviated this tendency reconstruction (X ² = 12.29, P = 0.007). Neither axillary status nor the type of axillary disturbance (no axillary dissection, axillary sampling or axillary clearance) were demonstrable as significant variables. Other operative factors such as intraoperative blood loss, transfusion requirements and operation time did not correlate with seroma formation. The relative experience of surgeons in terms of their grades was also not a significant variable.

Drainage output

Drains were kept for significantly longer in the seroma forming group, 6.6 days (range: 3-12 days), than in the control group, 4.5 days (range: 1-8 days) (t test, P= 0.008). When a drain was required for more than eight days, a significantly higher chance of seroma formation was found (Fisher′s exact test, P = 0.005). Drain output of more than 500 ml total for the first three days was associated with seroma formation (Fisher′s exact test, P = 0.049). Although the total drain output showed a strong linear relationship with the duration of drainage (Linear regression: B ₀ = -150.73, sigT= 0.000; B ₁ = 98.53, sigT= 0.000. R ² =0.69), neither the total drain output nor the late drain output, i.e., the drain output for the last three days revealed any correlation with seroma formation. Two kinds of drain output trends were identified: continuously decreasing and undulating. No statistically significant difference was observed between these two types of drainage output and seroma formation.

[Table - 4] lists the identified significant factors affecting seroma formation and their relative risk estimates.

Discussion

Although seroma formation may be considered as a trivial complication after mastectomy, it is the commonest complication and may be a nuisance. The incidence of seroma formation varies quite remarkably from 8-85% [Table - 5], a median rate of 20% often being quoted. Seroma may lead to delayed wound healing and infection. ^[6] Besides the economic loss due to prolonged hospital stay and delay in rehabilitation, seroma formation also adds to psychological trauma. This is, in addition, often to the embarrassment of the operating surgeon, whose experience in surgery does not influence the incidence of seroma after mastectomy. ^[7] Reasons that may account for the occurrence of seroma include a large operative field, division of lymphatic channels, the loose axillary skin hollow that follows surgical resection and the highly mobile, dependent nature of the area. ^[8] Seroma is often encountered for no obvious reason and without prodromal warning.

Several prophylactic measures have been proven to be effective in decreasing the incidence of seroma formation. Tranexamic acid was employed by Oertli et al , ^[3] peri- and postoperatively in a dose of 1g three times daily in a randomized double-blind trial. A significant reduction in the mean postoperative drainage volume as well as the incidence of seroma formation was observed in the treatment group.

The efficacy of shoulder immobilization has been investigated by Knight et al . ^[2] Although seroma resulted in delay in return to normal shoulder mobility, no patient sustained long-term musculoskeletal dysfunction.

Retention of the drain in-situ for a longer period seems to be a logical measure as formed seroma usually subsides with aspiration. ^[9] Estes and Glover ^[4] described the use of a vacutainer suction device as a convenient method of resolving seroma. This is no different in principle to re-establishment of the operative drain, a suction drain left in-situ for a prolonged period maintains physical contact between contiguous surfaces so as to facilitate adhesion. Conveney et al. ^[10] and O′Dwyer et al. , ^[11] demonstrated that both drainage and seroma formation were significantly less when dead space was obliterated by suturing the skin flaps to muscle. A similar flap tacking procedure was advocated by Chilson et al. , ^[1] as of proven value in seroma prevention. Lindsey et al. , ^[12] applied topical fibrin glue in the operative site in a Sprague-Dawley rat model and similarly decreased the incidence of seroma following mastectomy.

We have identified both age and hypertension to be risk factors for seroma formation. Patients older than 45 with hypertension are at a greater risk of developing seroma than their younger counterparts without hypertension. From the age correlation noted it has been speculated that the menstrual status might be relevant in seroma formation. No significant conclusion could, however, be drawn from our data. Alternatively, a decrease in elastic properties of the overlying skin flaps and less rapidity in re-draping to fill up the resultant space from surgical extirpation may be another possible factor. Similar findings have also been reported by Kumar et al., ^{[13] who postulated that a higher tendency to continuous exudation at the operative site is responsible for the association between hypertension and seroma formation.}

Operative blood loss, transfusion requirements and operation time are parameters which may reflect technical difficulties. Their relationship to seroma formation is of interest but no statistically significant relationship was found. Similarly, the experience of surgeons was not a significant factor. The same observation has also been reported by Funnell et al. , ^[7] who found that seroma formation was not totally avoidable with good technique.

It is logical to suspect that the larger the magnitude of the surgery, particularly when axillary lymphatics are disturbed, the greater the likelihood of seroma formation. This hypothesis is, however, not supported by the present set of data. The tendency towards seroma formation, though, was obviated by the addition of foreign well-vascularised tissue. The TRAM flap presumably provided an additional site for fluid resorption and represents an additional benefit for immediate breast reconstruction with living tissues. Another inference is that seroma formation is a manifestation of an imbalance between fluid resorption and excessive production. In the presence of a good fluid resorption mechanism, excessive fluid production is overshadowed and no clinical seroma formation occurs. A similar reduction in postoperative seroma was observed by O′Brien et al. , ^[14] employing immediate breast reconstruction with prostheses. A possible explanation here might be a reduction in the tendency to excessive exudation as a result of tension in the breast pocket.

A drain retained in-situ for longer than eight days due to a persistently high output is probably a direct reflection of an underlying tendency to excessive fluid production. Early drainage output in excess of 500 ml should alert the clinician as early as the fourth post-operative day.

In conclusion, there are identifiable peri-operative risk factors which may help to avoid prolonged hospital stay and morbidity. The presence of the factors identified should alert the clinician to possible seroma formation following mastectomy. Preventive measures may be warranted so as to reduce the risk of developing this not uncommon complication.^[16]

References

Copyright 2007 - Indian Journal of Cancer

The following images related to this document are available:

Photo images