Journal of Plastic, Reconstructive & Aesthetic Surgery (2014) 67, 226e230
The drain game: Back drains for latissimus
dorsi breast reconstruction
B.H. Miranda*, K. Amin, J.S. Chana
Plastic & Reconstructive Surgery Department, Royal Free London NHS Foundation Trust, Royal Free
Hospital, London NW3 2QG, UK
Received 10 July 2013; accepted 11 October 2013
KEYWORDS
Breast;
Latissimus dorsi;
Flap;
Drain;
Complications;
Seroma
Summary Introduction: The pedicled latissimus dorsi myocutaneous (LD) flap is a popular
breast reconstruction choice, representing approximately 50% of procedures undertaken in
the UK. Donor site drain use may reduce complication rates, however no evidence exists
regarding the duration of back drain use for LD flap breast reconstruction and calls have been
made in the literature to investigate this further.
Aim: To compare inpatient hospital stay, drainage parameters and donor-site complications
associated with closed suction back drain removal by post-operative day (POD) 3 regardless
of output (early group), with removal after POD 3 where instructions were documented by
drainage volume/24 h output consistency (late group), in post-mastectomy LD reconstruction donor sites.
Method: A retrospective review of LD breast reconstruction procedures, performed between
January 2010 and July 2011, was undertaken to ensure 1 year minimum follow-up per patient.
Results: There were 81 patients who underwent unilateral LD breast reconstructions; 78 hospital records contained complete documentation. There were 48 patients in the late removal
group and 30 patients in the early removal group. The mean drain removal day (5.42 0.17
days v s. 2.8 7 0 . 06 days , p < 0 .001 ), total d rainage (907 .71 76.0 7 ml vs .
492.67 35.15 ml, p < 0.0001) and hospital inpatient stay (4.60 0.19 days vs.
3.63 0.17 days, p < 0.001) were greater for patients in the late group, versus the early
group. There were no differences in total complications (16.67%(8/48) vs. 10%(3/30),
p Z 0.41), seroma (6.25%(3/48) vs. 6.67%(2/30), p Z 0.94), dehiscence (4.17%(2/48) vs.
3.33%(1/30), p Z 0.85) or haematoma rates (10.42%(5/48) vs. 0%(0/30), p Z 0.07) between
patients in the late and early groups; seroma sub-analysis also indicated no differences in
number of seroma aspirations, duration of drainage (months) and mean total drainage (ml)
prior to resolution.
* Corresponding author.
E-mail address: [email protected] (B.H. Miranda).
1748-6815/$ - see front matter Crown Copyright ª 2013 Published by Elsevier Ltd on behalf of British Association of Plastic, Reconstructive and
Aesthetic Surgeons. All rights reserved.
http://dx.doi.org/10.1016/j.bjps.2013.10.010
The drain game: Back drains for LD breast reconstruction
227
Discussion: These data suggest significant advantages for patients who have back drains
removed by POD 3, without increased post-operative complications including seroma rates,
and we recommend drain removal and patient discharge by POD 3.
Crown Copyright ª 2013 Published by Elsevier Ltd on behalf of British Association of Plastic,
Reconstructive and Aesthetic Surgeons. All rights reserved.
Introduction
In 2010 93,083 women underwent breast reconstruction in the
USA, with tissue expander-based reconstructions performed
for 65,391 and autologous flap reconstruction procedures
performed for over 18,500 patients.1 In the UK 2010 national
mastectomy and breast reconstruction audit of 18,216 patients, 3389 opted for immediate reconstruction, while 1731
opted for delayed reconstruction.2 Of those who underwent
immediate reconstruction, 21.7% (735/3389) had pedicled
flap þ implant/expander and 27.5% (932/3389) had pedicled
flap reconstructions. Of those who underwent delayed
reconstruction,
25.3%
(438/1731)
had
pedicled
flap þ implant/expander and 25.8% (446/1731) had pedicled
flap reconstructions. The pedicled latissimus dorsi myocutaneous (LD) flap is a popular choice for breast reconstruction,
representing approximately 50% of procedures undertaken in
the UK.3 Tansini first reported using the LD, type V musculocutaneous flap for post-mastectomy reconstruction.4,5
The most frequently reported LD donor-site complication is seroma, with a reported incidence of up to 72%,
depending on flap size and type.6e9 Management is labourintensive, often involving multiple visits for repeat
drainage, often in the outpatient setting compressive
dressings to prevent re-accumulation.2,9e11 As such,
numerous surgical techniques, to reduce seroma rates at
the donor site, are described in the literature e.g. drain
insertion, quilting sutures and fibrin glue.12e14 It is generally recognised that seromas still occur post drain removal
in most cases.15 Opinions however vary considerably, with
no clear evidence-based practice on the timing of postoperative LD donor site drain removal and seroma development, such that further investigation is required.10,16
‘LD’ to generate a spreadsheet of procedures performed
between January 2010 and July 2011. These dates were
chosen to ensure adequate volume of performed procedures and a minimum 1 year follow-up for all patients
included in the study. The inclusion criteria were all LD
breast reconstructions, with donor site drains in situ, performed between the previously mentioned dates and the
exclusion criteria were non-breast reconstructions or those
procedures where full data were unavailable.
Hospital records and plastic surgery dressing clinic notes
were retrospectively analysed for the following variables;
age, date of birth, operation date, number of drains,
drainage volume by day, day of hospital discharge, day of
drain removal and donor-site complications including
seroma, flap-related or systemic complications. Definitions
were as follows; haematoma (a post-operative collection
due to the extravasation of blood requiring surgical evacuation), seroma (a post-operative collection of a fluctuant
mass yielding straw-coloured serous fluid requiring aspiration), dehiscence (post-operative wound separation due to
any underlying cause). Seroma sub-analysis of the following
variables was also undertaken prior to resolution; number
of seroma aspirations, duration of drainage and mean total
drainage. When more than 1 back drain was removed on the
day of drain removal, output was added for each drain and
recorded accordingly. If patients were discharged with
drains in situ, after appropriate drain care education, they
telephoned the ward daily for nurse-led drain output
recording. Patients were then recalled for drain removal
once output had reached target drainage volume over 24 h.
Statistics were analysed using SPSS with t-tests for parametric continuous data and Chi2 tests for discrete data.
Results
Aim
The primary aim was to compare the donor-site complications associated with closed suction back drain removal by
post-operative day (POD) 3 regardless of output (early
group), with removal after POD 3 (late group), in postmastectomy LD reconstruction donor sites. This cut-off was
chosen due to the practice amongst several departmental
Consultants of early drain removal. Secondary aims
included analysis of drain output and inpatient hospital
stay. The null hypothesis was that no differences would be
found between these 2 groups (early vs. late).
Methods
After clinical governance registration, hospital database
information was accessed using the clinical operative code
There were 81 patients who underwent unilateral LD breast
reconstructions all of whom were followed up for a minimum of 1 year. Operations were performed, in standard
manner without use of donor site adjunct techniques e.g.
quilting, by 8 consultants, 4 of whom had patients with
drain removal instructions by day 3. There were 78 sets of
hospital records and plastic surgery dressing clinic notes
with complete documentation. The average age of patients
enrolled was 52.67 1.50 years (Mean SEM), their hospital stay was 4.23 0.14 days and they had drains
removed at 4.44 0.18 days.
There were 48 patients in the late drain removal group
(52.5 2.19 years) and 30 patients in the early group
(52.83 2.07 years) and both groups were matched for age
(p Z 0.91) and number of drains; (77.08% (34/48) vs.
56.67% (17/30)), and (22.92% (11.48) vs. 43.33% (13/30)) of
patients in the late vs. early groups had 1 or 2 drains
228
Table 1
B.H. Miranda et al.
Drain removal group comparison.
Number of patients
Age (years)
1 or 2 back drains (%)
Drain removal (day)
Drain removal output (ml)
Total drain output (ml)
Hospital stay (days)
Late group drain
removal after day 3
Early group drain
removal by day 3
p-Value
48
52.5 2.19
1 Z 77.08
2 Z 23.92
5.42 0.17
75.42 9.85
907.71 76.07
4.60 0.19
30
52.83 2.07
1 Z 56.67
2 Z 43.33
2.87 0.06
121 15.98
492.67 35.15
3.63 0.17
e
0.91
0.06
<0.001
0.02
<0.0001
<0.001
There were 48 patients in the late drain removal group (52.5 2.19 years) and 30 patients in the early group (52.83 2.07 years), both
groups were matched for age (p Z 0.91) and number of drains; (77.08% (34/48) vs. 56.67% (17/30)), and (22.92% (11.48) vs. 43.33%
(13/30)) of patients in the late vs. early groups had 1 or 2 drains respectively (p Z 0.06). The mean drain removal day was greater for the
late group (5.42 0.17 days) than the early group (2.87 0.06 days) (p < 0.001). Drain output on the day of removal was less for the late
group (75.42 9.85 ml) than the early group (121 15.98 ml) (p Z 0.02), however the total drainage for the late group was greater
(907.71 76.07 ml vs. 492.67 35.15 ml p < 0.0001). Patients in the late group were hospital inpatients for longer than those in the
early group (4.60 0.19 days vs. 3.63 0.17 days, p < 0.001).
respectively (p Z 0.06) (Table 1). In all cases where drains
were removed after day 3 (late group), drain removal instructions were by volume output consistency (i.e. a
change from fresh blood to serosanguinous fluid), with the
most common removal instruction being 30 ml/24 h
(72.92%, 35/48). In 4 cases, drains fell out or were removed
on day 2 (drainage range Z 60e320 ml/24 h) and these
patients were included in the early group. The mean drain
removal day was greater for the late group (5.42 0.17
days) than the early group (2.87 0.06 days) (p < 0.001)
(Table 1). As expected, the drain output on the day of
removal was less for the late group (75.42 9.85 ml) than
the early group (121 15.98 ml) (p Z 0.02), however the
total drainage for the late group was greater
(907.71 76.07 ml vs. 492.67 35.15 ml p < 0.0001) (Table
1). Additionally, patients in the late group were hospital
inpatients for longer than those in the early group
(4.60 0.19 days vs. 3.63 0.17 days, p < 0.001) (Table 1).
There were no differences found in total complications
(16.67% (8/48) vs. 10% (3/30), p Z 0.41), seroma (6.25% (3/
48) vs. 6.67% (2/30), p Z 0.94), dehiscence (4.17% (2/48)
vs. 3.33% (1/30), p Z 0.85) or haematoma rates (10.42% (5/
48) vs. 0% (0/30), p Z 0.07) between patients in the late
and early removal groups (Table 2). Seroma sub-analysis
indicated no differences in the number of seroma aspirations (5 2 vs. 2.67 0.88, p Z 0.48), duration of drainage
(4.5 2.5 months vs. 1.67 0.67 months, p Z 0.47) and
mean total drainage (650 440 ml vs. 455 244.56 ml, p Z
0.74) between patients in the late and early drain removal
groups prior to resolution (Table 3).
Discussion
Much debate exists in the literature regarding the use of
drains for breast reconstruction and there are subsequently
no definitive guidelines, including for post-operative
removal day and seroma development, such that calls
have been made throughout the literature for further
investigation.10,16 In a survey of 4669 American Society and
Canadian Society of Plastic Surgeons, >81% reported closed
suction drain use in breast reconstruction, with >93% using
volume criteria for drain removal, most commonly (>86%)
Table 3
Table 2
Complications between drain removal groups.
Total
complications
Seroma
Dehiscence
Haematoma
Late group
drain removal
after day 3
Early group
drain removal
by day 3
p-Value
16.67% (8/48)
10% (3/30)
0.41
6.25% (3/48)
4.17% (2/48)
10.42% (5/48)
6.67% (2/30)
3.33% (1/30)
0% (0/30)
0.94
0.85
0.07
There were no differences found in total complications (16.67%
(8/48) vs. 10% (3/30), p Z 0.41), seroma (6.25% (3/48) vs. 6.67%
(2/30), p Z 0.94), dehiscence (4.17% (2/48) vs. 3.33% (1/30),
p Z 0.85) or haematoma rates (10.42% (5/48) vs. 0% (0/30),
p Z 0.07) between patients in the late and early drain removal
groups.
Seroma sub-analysis.
Late group
Early group
p-Value
drain removal drain removal
after day 3
by day 3
Number of
52
aspirations
Drainage duration 4.5 2.5
(months)
Mean total
650 440
drainage (ml)
2.67 0.88
0.48
1.67 0.67
0.47
455 244.56 0.74
There were no differences in the number of seroma aspirations
(5 2 vs. 2.67 0.88, p Z 0.48), duration of drainage
(4.5 2.5 months vs. 1.67 0.67 months, p Z 0.47) and mean
total drainage (650 440 ml vs. 455 244.56 ml, p Z 0.74)
between patients in the late and early drain removal groups
prior to resolution.
The drain game: Back drains for LD breast reconstruction
recommending removal when drain output was 30 ml/
24 h. Interestingly, antibiotic cover was used by 98%,
however discontinuation was split with 46% preferring
discontinuation with drain removal and 52% preferring
removal on a particular post-operative day.16 Of particular
relevance, 90% of those surveyed indicated use of drains at
the donor site in cases of autologous flap reconstructions,
with 86.4% recommending removal with drainage 30 ml/
24 h. However, it should be noted that in cases where more
than one back drain is placed, and drainage removal instructions are 30 ml/24 h, both drains could be removed
on the same day with 30 ml/24 h of drainage per drain. In
this example, as both back drains would be sited in the
same cavity, the overall drainage would be 60 ml/24 h. The
implication of this alone challenges the rationale for drain
removal with 30 ml/24 h of drainage, the consistency of
findings between studies and indeed between those patients included within individual studies. For this reason,
we cumulatively recorded drain output on the day of back
drain removal (75.42 9.85 ml) in the late group, where
instructions were given by volume.
With respect to antibiotic use, common protocols
include cover until drain removal or a completion of a 5 or 7
day course even after drain removal, however there is no
general consensus or evidence supporting either
practice.11,16e19 It has also been suggested by several articles that there is no supportive evidence for prolonged
antibiotic use with drains across surgical specialities.20e22
In a systematic review of antibiotic use in breast reconstruction surgery (81 studies included), infection rates
decreased between groups as follows; no antibiotic cover
(29.83%), prophylactic antibiotic cover (11.36%), antibiotic
cover until drain removal (8.91%).23 Of note, the average
infection rate was 5.83% and no difference was demonstrated between groups receiving <24 h (5.76%) and >24 h
(5.78%) of antibiotic cover. However, all reconstructive
options were pooled from the spectrum of studies included
in this review whose authors indicated multiple confounding factors between studies and a greater prevalence of
infection with prosthesis or drain use. Furthermore, a 3e4
times increase in surgical site infections and reoperations
has been demonstrated, in a study of prosthetic breast
reconstruction, in patients who received a single dose of
pre-operative antibiotics versus those who received preand post-operative antibiotics until all drains were
removed.24 Although no clear evidence exists for antibiotic
use to cover LD donor site drains, and further investigation
is required, available general evidence suggests appropriate cover while the drain is in situ.
A Cochrane review on wound drain usage for plastic and
reconstructive surgery of the breast has further highlighted
the limited available evidence for drain use and outcomes in
reduction mammaplasty, and a likely increase in associated
hospital stay when used (average inpatient stay Z 3 days).10
It was also concluded that the lack of data for drain use and
outcomes in breast augmentation or reconstruction surgery
warrants further investigation. The data presented in this
study show no differences in post-operative complications,
most interestingly seroma rates, between patients who had
back drain removal by day 3 regardless of output (early),
compared to those whose drain removal instructions were by
volume such that their drains were removed after day 3
229
(late). This is further reinforced by the fact that both groups
were matched for age and number of donor site back drains,
and when high output drains (range Z 60e320 ml/24 h) fell
out or were removed on day 2, these data were included in
the day 3 group (4/30 patients). Although seroma subanalysis revealed no differences in number of aspirations,
duration and mean total drainage between patients in the
late and early drain removal groups prior to resolution, the
small sample size precludes any reliable conclusions. What is
also apparent is that early removal by day 3 resulted in a
shorter hospital inpatient stay and less total drainage output
per patient. The significance of these findings imply less
necessity for prolonged antibiotic use to cover in situ donor
site drains for LD breast reconstruction patients who have
drain removal by day 3 regardless of output, less hospital
inpatient expenditure, less inpatient exposure to nosocomial
infection, less physiological fluid shift insult and are therefore highly supportive of early drain removal.
Conclusion
These results suggest no differences in complication rates,
including seroma, between patients who have back drain
removal early (by day 3), compared to those who have drain
removal late (after day 3); although we highlight the
benefit of a randomised controlled trial design in the
future. As a result of early drain removal, significant patient benefits were also demonstrated such as a shorter
hospital inpatient stay; this benefits hospitals by reducing
LD breast reconstruction inpatient costs. The data, as
presented in this paper, suggest the recommended guideline of early drain removal and patient discharge by day 3;
this is common practice for several consultants within our
unit. These evidence-based guidelines fall within the
important concept of fast-track surgery and enhanced recovery as patients would be encouraged to eat, drink and
sit up on POD 1, with physiotherapy and occupational
therapy-managed mobilisation on POD 2, and drain removal
and discharge by POD 3.25e27
Conflicts of interest
None.
Sources of funding
None.
References
1. American Society of Plastic Surgeons. American Society of Plastic
Surgeons report of the 2010 plastic surgery statistics. American
Society of Plastic Surgeons. http://www.plasticsurgery.org/
documents/news-resources/statistics/2010-statisticss/toplevel/2010-us-cosmetic-reconstructive-plastic-surgery-minimally-invasive-statistics2.pdf; 2010.
2. The NHS Information Centre. National mastectomy and breast
reconstruction audit 2010. http://www.rcseng.ac.uk/surgeons/
research/surgical-research/docs/national-mastectomy-andbreast-reconstruction-audit-third-report-2010; 2010.
230
3. Llewellyn-Bennett R, Greenwood R, Benson JR, et al. Randomized clinical trial on the effect of fibrin sealant on latissimus dorsi donor-site seroma formation after breast
reconstruction. Br J Surg 2012;99(10):1381e8.
4. Tansini I. Sopra il mio nuovo processo sli amputazione della
mamella. Goz Med Ital 1906;57:141.
5. Mathes SJ, Nahai F. Classification of the vascular anatomy of
muscles: experimental and clinical correlation. Plast Reconstr
Surg 1981;67(2):177e87.
6. Tomita K, Yano K, Masuoka T, Matsuda K, Takada A,
Hosokawa K. Postoperative seroma formation in breast reconstruction with latissimus dorsi flaps: a retrospective study of
174 consecutive cases. Ann Plast Surg 2007;59(2):149e51.
7. Jeon BJ, Lee TS, Lim SY, et al. Risk factors for donor-site
seroma formation after immediate breast reconstruction with
the extended latissimus dorsi flap: a statistical analysis of 120
consecutive cases. Ann Plast Surg 2012;69(2):145e7.
8. Davis JP, Nield DV, Garth RJ, Breach NM. The latissimus dorsi
flap in head and neck reconstructive surgery: a review of 121
procedures. Clin Otolaryngol 1992;17:487e90.
9. Clough KB, Louis-Sylvestre C, Fitoussi A, Couturaud B, Nos C.
Donor site sequelae after autologous breast reconstruction
with an extended latissimus dorsi flap. Plast Reconstr Surg
2002;109(6):1904.
10. Stojkovic CA, Smeulders MJC, Van der Horst CM, Khan SM.
Wound drainage after plastic and reconstructive surgery of the
breast. Cochrane Database Syst Rev 2013;28(3):CD007258.
11. Barwell J, Campbell L, Watkins RM, Teasdale C. How long
should suction drains stay in after breast surgery with axillary
dissection? Ann R Coll Surg Engl 1997;79(6):435e7.
12. Titley OG, Spyrou GE, Fatah MFT. Preventing seroma in the
latissimus dorsi flap donor site. Br J Plast Surg 1997;50(2):
106e8.
13. Weinrach JC, Cronin ED, Smith BK, Collins DRJ, Cohen BE.
Preventing seroma in the latissimus dorsi flap donor site with
fibrin sealant. Ann Plast Surg 2004;53:12e6.
14. Thekkinkattil DK, Hussain T, Mahapatra TK, McManus PL,
Kneeshaw PJ. Feasibility of use of a barbed suture (v-loc 180)
for quilting the donor site in latissimus dorsi myocutaneous flap
breast reconstruction. Arch Plast Surg 2013;40(2):117e22.
15. Ferbeyre-Binelfa L. Lattissimus dorsi myocutaneous flap in
head and neck surgery. Cir Cir 2010;78:485e91.
B.H. Miranda et al.
16. Phillips BT, Wang ED, Mirrer J, et al. Current practice among
plastic surgeons of antibiotic prophylaxis and closed-suction
drains in breast reconstruction. Ann Plast Surg 2011;66(5):
460e5.
17. Scevola S, Youssef A, Kroll SS, Langstein H. Drains and seromas
in TRAM flap breast reconstruction. Ann Plast Surg 2002;48(5):
511e4.
18. Reid RR, Dumanian GA. A minimalist approach to the care
of the indwelling closed suction drain-a prospective analysis
of local wound complications. Ann Plast Surg 2003;51(6):
575e8.
19. Murray JD, Elwood ET, Jones GE, Barrick R, Feng J. Decreasing
expander breast infection: a new drain care protocol. Can J
Plast Surg 2009;17(1):17e21.
20. Vilar-Compte D, Rosales S, Hernandez-Mello N, Maafs E,
Volkow P. Surveillance, control, and prevention of surgical site
infections in breast cancer surgery: a 5-year experience. Am J
Infect Control 2009;37(8):674e9.
21. Hedrick TL, Smith PW, Gazoni LM, Sawyer RG. The appropriate
use of antibiotics in surgery: a review of surgical infections.
Curr Probl Surg 2007;44(10):635e75.
22. Bratzler DW, Houck PM. Antimicrobial prophylaxis for surgery: an advisory statement from the National Surgical
Infection Prevention Project. Clin Infect Dis 2004;38(12):
1706e15.
23. Phillips BT, Bishawi M, Dagum AB, Khan SU, Bui DT. A systematic review of antibiotic use and infection in breast reconstruction: what is the evidence? Plast Reconstr Surg 2013;
131(1):1e13.
24. Clayton JL, Bazakas A, Lee CN, Hultman CS, Halvorson EG.
Once is not enough: withholding postoperative prophylactic
antibiotics in prosthetic breast reconstruction is associated
with an increased risk of infection. Plast Reconstr Surg 2012;
130(3):495e502.
25. Kehlet H, Wilmore DW. Evidence-based surgical care and the
evolution of fast-track surgery. Br J Surg 2008;248(2):89e98.
26. Kehlet H, Wilmore DW. Fast-track surgery. Br J Surg 2005;92:
3e4.
27. Association of Surgeons of Great Britain and Ireland. Guidelines
for the implementation of enhanced recovery protocols.
http://www.improvement.nhs.uk/enhancedrecovery2/Portals/
2/RAD1526_IIPP_Dec_09.pdf; 2009.