Rationale: Seroma formation and its sequelae form the mainstay of complications in breast cancer surgery. Seroma has an incidence of 3% to 85%. Complications vary from delayed wound healing, infection, skin flap necrosis, patient discomfort and repeated visits to the out patient clinic. The key to reducing seroma formations seems to partly lie in the obliteration of dead space. The use of electrocautery has been demonstrated to increase seroma formation following mastectomy, however no other surgical devices (laser scalpel, argon diathermy and ultrasonic scalpel) or substances have proven to be superior in seroma reduction.

No prospective randomized controlled trials have been able to demonstrate which techniques are superior in reducing seroma and as a consequence patient discomfort in patients undergoing mastectomy. In a previous retrospective observational study these investigators demonstrated that mastectomy flap fixation significantly reduced seroma formation and the number of seroma aspirations. In a pilot study that was recently performed in one of the investigators hospitals, ARTISS tissue glue was used for skin flap fixation and showed promising results. The investigators hypothesize that obliteration of the dead space following mastectomy will significantly reduce seroma formation, its complications and the discomfort it causes patients undergoing mastectomy.

1. INTRODUCTION AND RATIONALE

   Seroma formation is a common side effect after surgery for breast cancer, with an incidence of 3% to 85%. Seroma is a collection of serous fluid that contains blood plasma and/ or lymph fluid. Some surgeons regard seroma formation as a necessary evil after breast cancer surgery. Seroma formation and its sequelae form the mainstay of complications in breast cancer surgery. Complications vary from delayed wound healing, infection, skin flap necrosis, patient discomfort and repeated visits to the out patient clinic to deal with seroma and its sequelae.

   The key to reducing seroma formations seems to partly lie in the obliteration of dead space. However, the techniques used to achieve this goal are subjects of much controversy and debate. In a randomized controlled study it was difficult to elucidate whether reducing the dead space or ligation of lymphatics or a combination of both were responsible of reduction of seroma formation.

   The use of electrocautery has been demonstrated to increase seroma formation following mastectomy, however no other surgical devices (laser scalpel, argon diathermy and ultrasonic scalpel) or substances have proven to be superior in seroma reduction. Seroma formation after axillary dissection for breast cancer cannot be avoided but hopefully it can be minimized by mechanical dead space closure. This systematic review highlights the fact that seroma formation is multifactorial and prevention of seroma is difficult at present. Certain prospective trials have been performed that demonstrate that flap anchoring and therefore dead space reduction could be very beneficial. The trial performed by Laurence et al analysed flap fixation and closed suction drainage. In this study there was no difference in seroma rates, but patients without drains were discharged earlier. In the trial performed by Sakkary et al the amount of fluid drained was significantly less in the flap fixation group however patient discomfort (seroma aspirations and out patient visits) were not assessed.

   A Cochrane review was published in 2013 regarding Fibrin Glue (Tissuecol), Fibrin Glue did not influence the incidence of postoperative seroma, the mean volume of seroma, wound infections, complications and the length of hospital stays in people undergoing breast cancer surgery. Due to significant methodological and clinical diversity among the included studies this conclusion may be considered weak and biased. Therefore, a major multicentre and high-quality RCT is required to validate these findings. ARTISS tissue glue is a two-component fibrin sealant matrix of human fibrinogen and human thrombin wich has the advantage of setting in 60 seconds. This means that the surgeon has 60 seconds to correctly position the skin flaps before the tissue sealant sets,

   The investigators hypothesize that obliteration of the dead space following mastectomy will significantly reduce seroma formation. This study aims to reduce seroma formation after mastectomy, proving that obliteration of dead space following mastectomy is vital in preventing seroma formation and its sequelae.

   The investigators current standard of treatment for patients undergoing mastectomy in Atrium Medical Centre Heerlen is flap fixation using interrupted Vicryl sutures and closed suction drainage. In Orbis Medical Centre Sittard current standard of wound closure in patients undergoing mastectomy is ARTISS Tissue Glue.

   A retrospective study (13-N-77) has been conducted in our centre in which flap fixation seemed to be an effective surgical technique in reducing dead space and therefore seroma formation in patients undergoing mastectomy for invasive breast cancer. Patients undergoing flap fixation required fewer seroma aspirations and had fewer seroma related complications (infection).

2. OBJECTIVES

Primary Objective:

To assess:

1. The number of seroma aspirations

Secondary Objective(s):

To assess:

1. The number of out patient clinic visits
2. Infection rate
3. Shoulder function
4. Cosmesis
5. Quality of life, costs and cost-effectiveness (cost/QALY)

3. STUDY DESIGN

Study design is a double-blind randomized controlled trial. All patients are > 18 years and will be enrolled after obtaining informed consent in the surgical out patient clinics of Zuyderland Medical Center (Heerlen and Sittard), Maastricht University Medical Center, St Jans Gasthuis, Weert and Albert Schweitser Hospital, Dordrecht. All patients have an indication to undergo mastectomy due to invasive breast cancer or DCIS.

Follow up will be performed until 1 year after surgery. Patients will be evaluated in the out patient clinic 2 weeks, 6 weeks, 3 months, 6 months and 12 months postoperatively.

Study population:

Female patients > 18 years diagnosed with invasive breast cancer or DCIS with an indication to perform mastectomy

Intervention:

1. Conventional mastectomy with closed suction drainage
2. Mastectomy with flap fixation using absorbable sutures and closed suction drainage
3. Mastectomy with flap fixation using ARTISS tissue glue and closed suction drainage

Main study parameters/endpoints:

Primary endpoint: Number of seroma aspirations

Nature and extent of the burden and risks associated with participation, benefit and group relatedness: Patients will be informed about the study before inclusion in the out patient clinic. Informed consent will be obtained in the out patient clinic a week after patients were initially informed. Postoperative check ups will de done more frequently.

Standard postoperative check ups:

2 weeks, 3 months, 12 months

Additional study postoperative check ups:

6 weeks, 6 months Therefore patients will be required to undergo two additional check ups. During out patients' visits, patients will hand in questionnaires regarding shoulder function and a questionnaire scale regarding cosmesis.

Patients will be clinically examined as they usually would be.

Out-patient follow up identical for all three groups

2 weeks post operatively, DASH + cosmesis questionnaire, clinical assessment seroma 6 weeks post operatively, DASH + cosmesis questionnaire + EQ-5D, clinical assessment seroma 3 months post operatively, DASH + cosmesis questionnaire + EQ-5D + cost questionnaire, clinical assessment seroma 6 months post operatively, DASH + cosmesis questionnaire + EQ-5D + cost questionnaire, clinical assessment seroma 12 months post operatively, DASH + cosmesis questionnaire +EQ-5D + cost questionnaire, clinical assessment seroma

EXIT FROM TRIAL 1 YEAR POST OPERATIVELY 4. STUDY POPULATION 4.1 Population (base ) All patients will be recruited from the surgical breast cancer clinic after evaluation for invasive breast cancer or DCIS. Patients will be recruited from two breast cancer clinics, treating 500 patients annually for breast cancer. Patients undergoing mastectomy or modified radical mastectomy are eligible for inclusion.

4.2 Inclusion criteria Older than 18 years Female sex Indication for mastectomy or modified radical mastectomy

4.3 Exclusion criteria Patients undergoing breast conserving therapy Patients undergoing direct breast reconstruction Unable to comprehend implications and extent of study and sign for informed consent

4.4 Sample size calculation The number of seroma aspirations is not normally distributed. Therefore, sample size estimations based on normally distributed continuous variables cannot be used. It is possible to use formulas based on non-parametric analysis methods, estimating the chance of a random patient in the treatment group having fewer aspirations than a random patient in the control group. However, since the majority of patients will have the same number of aspirations (i.e. 0), this method of sample size estimation does not seem to be applicable to our study.

Sample size estimation based on ordinal regression is in line with the data distribution and limited possibilities (maximum number of aspirations in the observational retrospective study was 4) of the outcome variable (Walters SJ. Health Qual Life Outcomes 2004: 2(1); 26). Using alpha=0.025 (correction for 2 comparisons with 3 study groups), beta=0.10, and a proportional odds ratio=2.67 (corresponding to an absolute difference of 20% in the need for seroma punctions), the sample size is estimated at 112 patients per study group.

5. TREATMENT OF SUBJECTS Group 1: conventional treatment and closed suction drainage Group 2: flap fixation with vicryl sutures and closed suction drainage Group 3: Artiss tissue glue and closed suction drainage

Inclusion Criteria:

• Older than 18 years
• Female sex
• Indication for mastectomy or modified radical mastectomy

Exclusion Criteria:

• Patients undergoing breast conserving therapy
• Patients undergoing direct breast reconstruction
• Unable to comprehend implications and extent of study and sign for informed consent

Female

Randomisation is achieved using a web based randomization programme (RANDI2). Randomization will take place on the day of surgery, after the start of the operation and 30 minutes before wound closure. Blocked randomisation will take place with randomly selected block sizes (6/9/12) with an allocation ratio of 1:1:1. Randomisation will be stratified per site. Both patients and surgeons will be blinded. Patients will be informed preoperatively that their post-operative check ups will NOT be performed by the surgeon that performed the surgery. Randomisation will take place 30 minutes before wound closure in the operating theatre. Post-operatively, patients will not be evaluated by their own surgeon. The OR report will contain the following information (in Dutch) so as to ensure that blinding of the physicians will not be compromised.