39

CLINICS IN
PLASTIC
SURGERY
Clin Plastic Surg 34 (2007) 39–50

Immediate Versus Delayed

Reconstruction
Steven J. Kronowitz, MD, FACS

- Immediate breast reconstruction The potential need for postmastectomy

- Delayed breast reconstruction radiation therapy
- Delayed-immediate breast reconstruction The effect of the clinical stage of breast
- Clinical considerations in determining the cancer
appropriate timing for breast The sequencing of chemotherapy:
reconstruction adjuvant or neoadjuvant chemotherapy
The impact of axillary sentinel lymph-node - Recommendations for clinical practice
biopsy - References

The increasing use of postmastectomy radiother- Center, the multidisciplinary institutional philoso-
apy (PMRT) in patients with early-stage breast can- phy is to avoid immediate breast reconstruction in
cer has increased the complexity of planning for patients who will require PMRT. To optimize re-
immediate reconstruction. Unfortunately, the in- construction in patients at risk for requiring
ability to know either preoperatively or intraopera- PMRT (stage II) when the need for PMRT is not
tively which patients will or will not require PMRT known at the time of mastectomy, the center im-
has led to uncertainty in regards to the appropriate plemented a two-stage approach, which is called
sequencing of breast reconstruction. If PMRT is re- delayed-immediate breast reconstruction. With
quired, delayed reconstruction is usually the best this approach, patients who do not require PMRT
course. If PMRT is not required, immediate recon- can achieve aesthetic outcomes that are essentially
struction is appropriate and permits better aes- the same as those with immediate reconstruction,
thetic outcomes. However, because of the and patients who require PMRT can avoid the aes-
inability first to detect nodal metastases in clini- thetic and radiation-delivery problems that can oc-
cally node-negative breast cancer patients before cur after an immediate breast reconstruction.
mastectomy and secondly to precisely evaluate
the presence of micrometastases intraoperatively,
it is often not known until several days after mas- Immediate breast reconstruction
tectomy whether PMRT will be required. If breast Immediate breast reconstruction is usually reserved
reconstruction is performed at the time of mastec- for patients with clinical stage-I breast cancer and
tomy and the patient is found postoperatively to some patients with clinical stage-II breast cancer
have lymph-node involvement, PMRT may ad- who do not have an increased risk of requiring
versely affect the aesthetic outcome, and the recon- PMRT. Unfortunately, although the risk of requir-
structed breast may cause technical difficulties with ing PMRT can be predicted before surgery, the
radiation delivery. At M.D. Anderson Cancer need for PMRT cannot be definitively determined

Department of Plastic Surgery, The University of Texas, M.D. Anderson Cancer Center, 1515 Holcombe Blvd.,
Unit 443, Houston, TX 77030, USA
E-mail address: skronowi@mdanderson.org

0094-1298/07/$ – see front matter ª 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.cps.2006.11.006
plasticsurgery.theclinics.com
40 Kronowitz

until the final pathologic evaluation is complete, breast cancer team, which includes a radiation on-
approximately 1 week after mastectomy. cologist. Patients deemed to be at increased risk
Immediate breast reconstruction offers many ad- for conditions necessitating PMRT and who desire
vantages over delayed reconstruction, including breast reconstruction, are considered eligible for de-
a better aesthetic outcome due to preservation of layed-immediate breast reconstruction (Fig. 1).
the three-dimensional breast skin envelope [1] Stage one of this two-stage approach consists of
and the psychological benefit of awakening from skin-sparing mastectomy with insertion of a filled
mastectomy with a reconstructed breast [2]. Imme- textured saline tissue expander to preserve the
diate breast reconstruction also enables athletic shape and dimensions of the breast skin envelope
patients to more easily resume their activities, until the results of the permanent pathology are
avoiding the potentially embarrassing situation of known. After review of permanent sections, patients
an external prosthesis dislodging during exercise. who do not require PMRT undergo stage 2 of de-
layed-immediate reconstruction, while patients
Delayed breast reconstruction who require PMRT complete this therapy and
then undergo a skin-preserving delayed reconstruc-
Delayed breast reconstruction is usually reserved for
tion. Clinicians at M.D. Anderson prefer to perform
patients who will require PMRT. At M.D. Anderson,
stage 2 of delayed-immediate reconstruction (re-
clinicians prefer not to use breast implants in pa-
moval of the tissue expander along with definitive
tients who have received PMRT. This is because
reconstruction) within approximately 2 weeks after
acute problems with wound healing and long-
mastectomy to avoid delays in the initiation of ad-
term problems with capsular contracture can result
juvant chemotherapy and to avoid capsule forma-
in implant displacement and a painful constriction
tion around the tissue expander, preserving the
across the chest wall. Thus, clinicians at M.D. An-
elasticity of the breast skin.
derson usually perform delayed breast reconstruc-
In patients who require PMRT, the tissue ex-
tion using autologous tissue.
pander remains inflated until just before CT simula-
Many of the aesthetic outcomes of delayed recon-
tion to design the radiation fields. At that time, the
struction, even when it is performed by experienced
tissue expander is deflated in an office-based setting
surgeons, are satisfactory at best [1,2]. However, pa-
to result in a flat chest wall surface for radiation de-
tients who undergo delayed reconstruction after
livery. The flat chest wall surface allows for the treat-
PMRT are the most appreciative because they have
ment of internal mammary nodes without excessive
had to experience the difficulties of not having
injury to the heart or lungs. Approximately 2 weeks
a breast. The retained, irradiated, and scarred breast
after the completion of PMRT, the tissue expander is
skin located between the mastectomy scar and the
gradually reinflated to the predeflation saline vol-
inframammary fold is usually resected at the time
ume. Clinicians usually perform delayed recon-
of delayed reconstruction because the skin is inflex-
struction using the preserved breast skin envelope
ible and does not allow for the reconstruction of
approximately 4 to 6 months after the completion
a curved and ptotic-appearing breast. This resection
of PMRT. With skin-preserving delayed reconstruc-
not only requires a much larger volume of flap tis-
tion, clinicians prefer the use of autologous tissue
sue because of the need for skin replacement but
flaps to avoid the problems associated with the
also requires the entire three-dimensional contour
use of breast implants within an irradiated opera-
of the breast to be recreated. The need to replace
tive field.
the inferior breast skin means that more flap skin
Delayed-immediate reconstruction allows pa-
is visible. Because of the increased skin require-
tients to review their final pathology report with a
ments, often three quarters of the transverse rectus
radiation oncologist before committing to delayed-
abdominis myocutaneous (TRAM) or deep inferior
immediate or delayed reconstruction. During that
epigastric perforator (DIEP) flap must be used to re-
time, as the fully inflated expander placed in stage
construct the breast, leaving inadequate tissue for
one prevents retraction of the mastectomy skin and
bilateral reconstruction. With delayed breast recon-
loss of breast shape, clinicians have the opportunity
struction, clinicians at M.D. Anderson also rely
to revise any irregularities of the inframammary fold
more significantly on the ability to perform a con-
and debride any nonviable mastectomy skin before
tralateral mastopexy (breast lift) to obtain symme-
insetting of an autologous tissue flap. Delayed-im-
try because it is more difficult to match the ptotic
mediate reconstruction can be adapted to any clini-
shape of a contralateral native breast.
cal practice and modified to comply with various
institutional guidelines for PMRT.
Delayed-immediate breast reconstruction
With the delayed-immediate approach, patients
At M.D. Anderson, patients with clinical stage-II who do not require PMRT can achieve aesthetic out-
breast cancer are evaluated by a multidisciplinary comes similar to those of immediate reconstruction
 Immediate Versus Delayed Reconstruction 41

STAGE 1 Fig. 1. Delayed-immediate

Skin-Sparing Mastectomy breast reconstruction proce-
+ dure for patients with unilat-
Subpectoral Tissue Expander
Intraoperative Saline Filling to 100 of Manufacturer’s Suggested Fill Volume
eral breast cancer not treated
with prophylactic contralateral
mastectomy and for patients
Assessment of Permanent Sections with bilateral breast cancer. In
patients with unilateral breast
cancer who elect prophylactic
Postoperative Radiation Consultation
contralateral mastectomy, con-
tralateral mastectomy and
immediate reconstruction are
No PMRT PMRT
performed at the time of de-
finitive reconstruction of the
STAGE 2 Deflate Tissue Expander* breast with cancer. *After the
Definitive Breast Reconstruction‡ completion of chemotherapy
† but before the initiation of ra-
Re-inflate Expander after PMRT
diation therapy. If patient had
TRAM or SGAP LD & Implant Permanent Implant neoadjuvant chemotherapy,
leave expander inflated during
Skin-Preserving Delayed Reconstruction the 4- to 6-week period before
the initiation of radiation ther-
apy. yAllow several weeks for
TRAM or SGAP LD & Implant skin desquamation to resolve.
Results regarding expander
re-inflation are pending. zUsually performed 2 weeks after mastectomy and stage one of delayed-immediate re-
construction to prevent a delay of chemotherapy. If patient had neoadjuvant chemotherapy, definitive recon-
struction may be delayed longer than 2 weeks. LD, latissimus dorsi myocutaneous flap; SGAP, superior gluteal
artery perforator flap. (From Kronowitz SJ, Hunt KK, Kuerer HM, et al. Delayed-immediate breast reconstruction.
Plast Reconstr Surg 2004;113:1620; with permission.)

(Figs. 2 and 3), and patients who require PMRT can initial operation. Unfortunately, the intraoperative
avoid problems associated with PMRT after an im- examination of sentinel lymph nodes with frozen
mediate breast reconstruction. Delayed-immediate section analysis, imprint cytology techniques, or
reconstruction offers the opportunity for a better both does not reveal all micrometastases [5,6].
aesthetic outcome than is achieved with standard Conducting an axillary lymph-node dissection after
delayed reconstruction. Specifically, re-expansion sentinel node biopsy and immediate autologous
of the mastectomy skin after PMRT (Fig. 4) pro- breast reconstruction have taken place can compro-
vides additional usable breast skin to perform de- mise the blood supply to the reconstructed breast,
layed breast reconstruction. Delayed-immediate particularly the thoracodorsal vascular system, parts
reconstruction provides an additional option that of which are often used as recipient vessels for a free
broadens patients’ treatment choices and allows pa- TRAM flap or as a vascular pedicle for a latissimus
tients to participate fully in treatment and recon- dorsi myocutaneous flap [7].
struction decisions. Several recent studies [7,8] have evaluated clini-
copathologic factors that may help identify preop-
Clinical considerations in determining eratively which clinically node-negative patients
the appropriate timing are at risk for undetectable micrometastatic axillary
for breast reconstruction disease. A recent report from M.D. Anderson [7]
demonstrated that patients who were 50 years of
The impact of axillary sentinel lymph-node age or younger, patients who had tumors larger
biopsy than 2 cm, and patients who had lymphovascular
The current recommendation when a sentinel node invasion detected in the initial biopsy specimen
is found to be positive is to perform a completion were at higher risk for harboring axillary metastases.
level-I and -II axillary node dissection because addi- However, although these factors can help identify
tional nodes will be involved in up to 40% of such high-risk patients, the ability to consistently predict
patients [3,4]. Current practice dictates that if the and quantify axillary involvement before surgery is
intraoperative assessment of the sentinel lymph limited.
node is positive, a completion level-I and -II axillary As use of the axillary sentinel-node biopsy tech-
nodal dissection is performed at the time of the nique in conjunction with breast reconstruction
42 Kronowitz

Fig. 2. Delayed-immediate breast reconstruction in a 55-year-old woman with stage-II (T2multifocalN0M0) left
breast cancer. (A) Preoperative view after neoadjuvant chemotherapy. (B) Four weeks after a left skin-sparing
total mastectomy with axillary sentinel lymph-node biopsy and subpectoral placement of a textured saline tissue
expander (700 mL) expanded intraoperatively to the saline-fill volume of 700 mL. (C) Intraoperative view during
complete axillary lymph-node dissection performed 10 days after mastectomy. (D) Ten days after TRAM flap re-
construction. (E) Thirteen months after TRAM flap reconstruction and 6 months after a left vertical breast reduc-
tion for symmetry.

continues to increase (Fig. 5), the current approach shoulder dystocia) on patients whose axillary nodes
to immediate breast reconstruction should be may turn out to be negative for disease on perma-
reevaluated for patients at high risk of axillary in- nent pathological analysis [9]. Delayed breast re-
volvement to avoid possibly compromising the vas- construction may also be an option; however,
cularity to the reconstructed breast. In a study immediate breast reconstruction has well-recog-
published in 2002 [7], the author and colleagues nized benefits in terms of aesthetics and lessening
proposed an algorithm for decision making for the psychological effects of mastectomy [10]. Al-
breast reconstruction in clinically node-negative though postoperative axillary radiation may be
breast cancer patients (Fig. 6) [7]. The risk to an im- a consideration for locoregional control when the
mediate breast reconstruction could be avoided sim- sentinel node is found to be positive on permanent
ply by performing an initial complete level-I and -II histopathological analysis, it does not provide the
axillary dissection because all nodes will be removed important prognostic information obtained from
without the worry about coming back for more axil- the additional nodal tissue [11]. When immediate
lary surgery. However, this practice could impose sig- breast reconstruction is performed with either a mi-
nificant surgical morbidity (eg, lymphedema and crovascular TRAM flap (with the most commonly
 Immediate Versus Delayed Reconstruction 43

Fig. 3. Delayed-immediate breast reconstruction in a 39-year-old woman with previous bilateral breast augmen-
tation who presented with clinical stage-II (T2multifocalN0M0) left breast cancer. (A) Preoperative view after
neoadjuvant chemotherapy. (B) Postoperative view 2 weeks after a left skin-sparing total mastectomy with ax-
illary sentinel lymph-node biopsy and subpectoral placement of a textured saline tissue expander (800 mL) ex-
panded intraoperatively to the saline-fill volume of 800 mL. (C) Two years after latissimus dorsi myocutaneous
flap and silicon breast implant (800 mL) reconstruction.

used recipient vessels being the thoracodorsal artery proceed with an immediate breast reconstruction or
and vein) or a pedicled latissimus dorsi myocutane- to delay reconstruction until after the results of the
ous flap, the vascular pedicle may be at risk if sub- permanent pathology are known will need to have
sequent axillary surgery is required. There are been made beforehand by the multidisciplinary
alternatives to the use of these vessels for immediate breast cancer team and the patient. At M.D. Ander-
autologous breast reconstruction in patients under- son, existing treatment guidelines greatly influence
going mastectomy and axillary sentinel-node bi- the decision of whether to proceed with breast re-
opsy that may minimize the risk of vascular construction if the sentinel node is positive or to de-
damage on reoperation. One of these alternatives lay the reconstruction [7].
is the use of the internal mammary artery and
vein as recipient vessels for a microvascular or a ped- The potential need for postmastectomy
icled TRAM flap [7]. With the use of axillary senti- radiation therapy
nel-node biopsy now routine at M.D. Anderson, The increasing use of PMRT in patients with early-
the internal mammary vessels are often the first stage breast cancer, along with the inability to deter-
choice in immediate microvascular TRAM flap mine preoperatively which patients will require
breast reconstruction to avoid the potential for vas- PMRT, has increased the complexity of planning
cular injury to the TRAM flap if subsequent axillary for immediate breast reconstruction. There are
surgery is required. two potential problems with performing an imme-
Another confounding factor is the possible need diate breast reconstruction in a patient who will re-
for postoperative axillary radiation when the senti- quire PMRT. First, an immediate breast
nel node is found to be positive [11]. At the time of reconstruction can interfere with the delivery of
surgery, it is not known whether postoperative radi- PMRT. Second, PMRT can adversely affect the aes-
ation will be required because findings from the thetic outcome of an immediate breast reconstruc-
pathologic examination of the additional tissue tion. Because the potential need for PMRT is one
are not available until several days after the surgery. of the most important considerations affecting the
Therefore, if the intraoperative examination reveals timing and technique of breast reconstruction, the
a positive sentinel node, the decision of whether to multidisciplinary breast team must work together
44 Kronowitz

in planning surgery for patients with breast cancer and on the technical problems with delivery of radi-
who desire reconstruction after mastectomy. The ation to a reconstructed breast [13].
preoperative consultation with the patient should Recently, both the American Society for Thera-
include emphasis on the potentially adverse effects peutic Radiology and Oncology [11] and the Amer-
that radiation treatment can have on aesthetic out- ican Society of Clinical Oncology [14] published
come of an immediate breast reconstruction [12] consensus statements regarding PMRT. Both groups
 Immediate Versus Delayed Reconstruction 45

700 Fig. 5. The increasing use of

sentinel-node biopsy in
600 combination with breast
No. of Procedures

500 reconstruction at M.D.

Anderson Cancer Center.
400 SLNB, sentinel lymph-node
biopsy.
300

200

100

0
1998 1999 2000 2001 2002 2003
Year
All SLNB Procedures SLNB with Any Breast Reconstruction

currently recommend PMRT in patients with four of the radiation fields for PMRT [17,18]. The previ-
or more positive lymph nodes or advanced tumors. ously mentioned randomized trials that reported
However, on the basis of recent prospective, ran- a survival advantage with PMRT [15,16] included
domized controlled trials (the so-called ‘‘Danish’’ the internal mammary nodes within the radiation
and ‘‘Canadian’’ trials [15,16]) that demonstrated fields. To treat these areas and minimize the dose
superior locoregional control, disease-free survival, to the heart and lungs, a separate electron beam
and overall survival in breast cancer patients with on the medial chest wall is often required to match
T1 or T2 disease and one to three positive lymph the laterally placed opposed tangent fields [17].
nodes with the addition of PMRT to mastectomy Some anatomic configurations make it difficult to
and chemotherapy, both societies emphasize the successfully deliver PMRT using such a separate me-
need for additional prospective data concerning dial electron-beam field [18]. The sloping contour
the use of PMRT in these patients. In the future, de- of a reconstructed breast leads to an imprecise geo-
pending on the outcome of ongoing trials, PMRT metric matching of the medial and lateral radiation
may be widely recommended in patients with fields. Alternative radiation fields will result in ei-
early-stage breast cancer. Some institutions have ther exclusion of the internal mammary nodes or
already instituted routine PMRT in patients with increased irradiation of normal tissues.
early-stage disease.
PMRT may adversely affect the aesthetic
Immediate breast reconstruction outcome of an immediate breast
may interfere with the delivery of PMRT reconstruction
An important issue in immediate breast reconstruc- Experience at the M.D. Anderson Cancer Center
tion is whether the reconstructed breast will impair [12,19,20] and many other experiences reported
the delivery of PMRT. Immediate breast reconstruc- in the literature [19,21–24] indicate that autologous
tion can cause technical problems with the design tissue is preferable for breast reconstruction in

=
Fig. 4. Skin-preserving delayed free microvascular TRAM flap reconstruction after stage one of delayed-immedi-
ate reconstruction and PMRT in a 52-year-old woman with a clinical stage-II (T2N1M0) right breast cancer. (A)
Preoperative view after neoadjuvant chemotherapy. (B) Postoperative view 3 weeks after right skin-sparing
modified radical mastectomy and stage one of delayed-immediate reconstruction with placement of a subpec-
toral textured saline tissue expander with an intraoperative saline-fill volume of 600 mL. The permanent pathol-
ogy after mastectomy upstaged the patient to stage III (T2N2M0). (C, D, and E) Before the start of
postmastectomy radiation therapy, the patient had complete deflation of the expander with removal of 600 cc
in the M.D. Anderson clinic. Complete deflation of the expander during postmastectomy radiation therapy
allows for treatment of the internal mammary lymph nodes without excessive injury to the heart and lungs
and avoids uneven radiation dose distribution. Shown is the three-dimensional planning for the design of
the radiation treatment fields using CT (axial and three-dimensional images). (F) Several weeks after the
completion of postmastectomy radiation therapy, the tissue expander was progressively reinflated to the
predeflation volume of 600 mL. (G) Seven months after completion of the postmastectomy radiation ther-
apy, the patient underwent a skin-preserving delayed breast reconstruction with a microvascular TRAM
flap. Postoperative view 6 months after breast reconstruction. (H) Six months after the skin-preserving de-
layed breast reconstruction, a scar revision of the right reconstructed breast and a left vertical mastopexy
was performed to achieve symmetry. Postoperative view 2 weeks after the symmetry procedure. Further
descent of the left breast is expected, which should improve the result.
46 Kronowitz

Fig. 6. An algorithm for decision

making in clinically node-nega-
tive breast cancer patients who
desire autologous breast recon-
struction and sentinel-node
biopsy. SLN, sentinal lymph
node. (From Kronowitz SJ,
Chang DW, Robb GL, et al.
Implications of axillary sentinel
lymph node biopsy in im-
mediate autologous breast re-
construction. Plast Reconstr
Surg 2002;109:1888–96; with
permission.)

patients who have received PMRT and that breast re- In 2001, investigators at M.D. Anderson [12]
construction should probably be delayed in pa- published a retrospective study comparing immedi-
tients who are known preoperatively to require ate and delayed free TRAM flap breast reconstruc-
PMRT. Unfortunately, evaluation of complication tion in patients receiving PMRT. In this study, 32
rates and aesthetic outcomes is extremely difficult patients had immediate TRAM flap reconstruction
because of significant variation in the sequencing before radioptherapy, and 70 patients had PMRT
of PMRT and reconstruction, the administration before TRAM flap reconstruction. The mean fol-
of systemic therapy, the duration of follow-up, low-up times after the end of treatment for the im-
and the techniques of radiation delivery and breast mediate and delayed reconstruction groups were 3
reconstruction. and 5 years, respectively. The incidence of early
In 1997, Williams and colleagues [24] from flap complications (vessel thrombosis and partial
Emory University compared outcomes of pedicled or total flap loss) did not differ significantly be-
TRAM flap breast reconstruction in 19 patients tween the two groups. However, the incidence of
who received PMRT after reconstruction and 108 late complications (eg, fat necrosis, flap volume
patients who received PMRT before reconstruction loss, and flap contracture) was significantly higher
with outcomes in 572 patients who underwent in the immediate reconstruction group than in the
TRAM flap breast reconstruction without PMRT. delayed reconstruction group (87.5% versus 8.6%;
At a mean follow-up time after reconstruction of P<.001). Furthermore, 28% of the patients with im-
47.6 months, 52.6% of the patients who received mediate reconstruction required an additional flap
PMRT after TRAM flap reconstruction demonstrated to correct the distorted contour that resulted from
postirradiation changes, and 31.6% required surgi- flap shrinkage and severe flap contracture after
cal intervention. PMRT.
Spear and Onyewu [21] published a review in In 2002, Rogers and Allen [25] published the
2000 evaluating the effects of irradiation on results of a study on the effects of PMRT on breasts
outcomes after two-stage breast reconstruction reconstructed with a DIEP flap. In this study,
with saline-filled implants. These investigators ret- a matched-pairs analysis was performed of 30 pa-
rospectively compared 40 patients who underwent tients who had breast reconstruction with a DIEP
two-stage saline-filled–implant breast reconstruc- flap and PMRT and 30 patients who underwent
tion followed by irradiation with 40 other patients DIEP flap reconstruction without PMRT. Patients
who underwent the same reconstruction procedure who received PMRT had higher incidences of fat ne-
without irradiation. The incidence of complications crosis in the DIEP flap (23.3% versus 0%; P5.006),
was significantly higher in the irradiated group than fibrosis and shrinkage (56.7% versus 0%; P<.001),
in the control group (52.5% versus 10%; P<.001). and flap contracture (16.7% versus 0%; P5.023).
Thirty-two percent of the irradiated patients had In 2005, Spear and colleagues [26] found that pa-
symptomatic capsular contractures, whereas no tients who had TRAM flap reconstruction before ir-
contractures occurred in the control group. Forty- radiation had worse aesthetic outcomes, symmetry,
seven percent of the 40 irradiated breasts needed and contractures than did patients who underwent
flap procedures, whereas only 10% of the nonirradi- irradiation before TRAM flap breast reconstruction.
ated breasts needed flaps. These investigators recommended that TRAM flap
 Immediate Versus Delayed Reconstruction 47

reconstruction be postponed in patients known or problems with wound healing [32–35], no delays
expected to require PMRT. in the initiation or resumption of chemotherapy
as a result of wound-healing problems or infections
The effect of the clinical stage of breast [32–35], and no need for premature cessation of
cancer chemotherapy as a result of wound-healing prob-
PMRT is given to some patients after mastectomy to lems [33–35].
reduce the risk of locoregional recurrence in high- A study by Yule and colleagues [32] evaluated 46
risk patients. Currently, indications for PMRT in- patients who underwent immediate breast recon-
clude large tumor size or direct skin involvement struction with tissue expanders and subsequent per-
(T3 or T4 tumors) or documented lymph-node in- manent implants. Twenty-three patients received
volvement in four or more lymph nodes. Therefore, adjuvant chemotherapy, and 23 did not. The inves-
the stage of the breast cancer is critical in recon- tigators reported no statistically significant differ-
structive planning. Patients with clinical stage-I ences in wound healing, wound infection, or
breast cancer are considered to be at low risk for re- capsular contracture between the patients treated
quiring PMRT and are therefore considered candi- with chemotherapy and those who did not receive
dates for immediate breast reconstruction using chemotherapy. Several studies [33,34] have shown
any of the available approaches. that patients who undergo immediate breast recon-
Some patients with clinical stage-II breast cancer struction are not predisposed to delays in the initi-
have a borderline elevated risk of requiring PMRT, ation of adjuvant chemotherapy compared with
and thus these are the patients for whom it is patients who have mastectomy without reconstruc-
most difficult to formulate recommendations re- tion. Schusterman and colleagues [36] compared
garding timing of breast reconstruction [27,28]. It the free TRAM flap with the pedicled TRAM flap
is essential that these patients have a careful preop- in patients requiring postoperative chemotherapy
erative evaluation for risk factors for occult axillary and found that 29% of patients who underwent re-
nodal involvement (age younger than 50 years, construction with a pedicled TRAM flap had a delay
lymphovascular invasion in the initial biopsy spec- in the start of chemotherapy, compared with only
imen, and T2 tumor) [7]. In patients with any of 14% of the patients who underwent reconstruction
these risk factors, it may be preferable to avoid the with a free TRAM flap. In another study, by Caffo
use of breast implants, to perform delayed recon- and colleagues [35], patients who underwent im-
struction, or to use a delayed-immediate approach mediate breast reconstruction did not require
(described below). more frequent adjustments in the dose intensity
In patients with clinical stage-III breast cancer of their chemotherapy as compared with patients
(locally advanced), it may be preferable to delay re- who underwent mastectomy without reconstruc-
construction until after mastectomy and PMRT to tion. Caffo and colleagues also observed that che-
avoid potential problems with radiation delivery motherapy did not influence the interval between
and to avoid the possibility of adverse affects of surgery and the start of expander inflation.
PMRT on an immediately reconstructed breast Neoadjuvant (preoperative) chemotherapy is be-
[12,19,20,22–24,26,27]. Breast reconstruction has ing used with increasing frequency in breast cancer
not been found to delay diagnosis or decrease sur- patients with stage-II and -III disease. The effect of
vival in patients who present with stage-III disease neoadjuvant chemotherapy on surgical outcome is
and later develop a local recurrence [29]. of particular concern in these patients, as is the po-
tential for wound-healing problems that may delay
The sequencing of chemotherapy: adjuvant any subsequent adjuvant therapy. As the interval
or neoadjuvant chemotherapy between chemotherapy and surgery increases, the
An increasing number of breast cancer patients with impact on wound healing diminishes [37]. That
stage-I disease are being treated with neoadjuvant is, the white blood cell count nadir occurs at 10
(postoperative) chemotherapy. Concerns have to 14 days after the last chemotherapy treatment,
been raised that the cytotoxic and myelosuppressive and recovery occurs by 21 days [38]. It is clinically
effects of chemotherapy may result in poor wound recognized that an absolute neutrophil count <500
healing or an increased incidence of postoperative cells per cubic millimeter is detrimental to wound
wound infections [30,31] after breast reconstruc- healing and wound strength [39]. Wound healing
tion. Concerns have also been raised that complica- can usually occur normally when the white blood
tions of immediate breast reconstruction may cell count is >3,000 cells per cubic millimeter [39].
interfere with the subsequent administration of Therefore, the timing of mastectomy with immedi-
adjuvant chemotherapy [32–35]. Most studies of ate breast reconstruction after neoadjuvant chemo-
chemotherapy in patients treated with breast therapy is important to avoid wound-healing
reconstruction have found no significant prolonged problems.
48 Kronowitz

Deutsch and colleagues [40] evaluated 31 pa- aware that the aesthetic results of delayed breast re-
tients who underwent immediate reconstruction construction are often less optimal than those of
with a TRAM flap after neoadjuvant chemotherapy. immediate reconstruction [1,10]. Throughout the
Seventeen patients had postoperative complica- patient-education process, it is prudent to obtain
tions, but only 2 had a delay in the start of adjuvant appropriate patient consent and document it in
chemotherapy. Seven patients were smokers, 5 of the medical record.
whom had complications. Both delays in chemo- Careful planning before surgery is required to
therapy occurred in smokers. Neither delay was lon- minimize adverse effects of PMRT on breast recon-
ger than 6 weeks after the normal 4-week interval. struction, which can result in significant patient dis-
The investigators found no correlation between satisfaction. During planning for immediate breast
the number of preoperative chemotherapy cycles, reconstruction, it is imperative to carefully review
interval from chemotherapy to surgery, or stage of the stage of disease and the likelihood that the pa-
disease and surgical outcome. There was no statisti- tient will require PMRT. If PMRT is planned, the
cally significant difference in the incidence of com- use of an implant for breast reconstruction should
plications between pedicled and free TRAM flap be strongly discouraged because of the risk of
reconstructions. The investigators concluded that capsular contracture [1,20,22,23,42]. Capsular
immediate breast reconstruction with the TRAM contracture can distort the appearance of the
flap can be performed safely in patients who receive reconstructed breast and cause chronic chest-wall
neoadjuvant chemotherapy but that the combina- pain and tightness [43]. Furthermore, the addition
tion of neoadjuvant chemotherapy and smoking of a latissimus dorsi flap does not protect against
may significantly increase the risk of complications. the negative effects of radiation on breast implants
The increasing use of neoadjuvant chemotherapy [20]. Even though autologous tissue alone is pre-
has also made it more difficult to predict which pa- ferred in an irradiated patient, autologous recon-
tients will require PMRT. The complexity relates to structions can also be adversely affected by PMRT
patients who present with an equivocal indication [1,12,19,21,24,25]. Contracture of the breast skin
for PMRT before neoadjuvant chemotherapy and and atrophy of the flap [43] can result in anatomic
who after neoadjuvant chemotherapy are found to distortion of the reconstructed breast that can prog-
have (through ultrasonography) either completely ress over time, resulting in displacement of the flap
or partially responded to this therapy with signifi- superiorly [43]. Restoring symmetry can be
cant reduction or complete resolution of the breast extremely difficult [43]and, although a local flap
tumor or any locoregional lymph-node metastases. may occasionally correct a small contour deformity,
Unfortunately, the recommendation for PMRT is an additional flap is often required to restore breast
usually not made until after review of the perma- shape and allow adequate healing.
nent pathology of the mastectomy specimen, ap- In all cases of decision making about possible
proximately 1 week after surgery. Although PMRT and whether or not to perform immediate
neoadjuvant chemotherapy can further complicate breast reconstruction, the situation should be dis-
the decision to perform immediate, delayed-imme- cussed at a multidisciplinary conference or ad-
diate, or delayed breast reconstruction, it often re- dressed among the various medical, surgical, and
duces the bulk of the tumor and therefore the radiation teams, with active participation by the
extent of the extirpative procedure required, there- patient. M.D. Anderson’s multidisciplinary philos-
fore allowing for improved outcomes of breast ophy is to avoid immediate breast reconstruction
reconstruction. in patients who will definitely require or have
a high likelihood of requiring PMRT. At M.D. An-
derson, delayed-immediate breast reconstruction is
Recommendations for clinical practice
often used for patients at high-risk for requiring
The multidisciplinary breast team should educate PMRT. In this approach, clinicians delay immedi-
patients about breast reconstruction and increase ate reconstruction until after review of the final pa-
their awareness of the interplay among the thology report on the mastectomy specimen and
currently evolving diagnostic and treatment the axillary lymph nodes. With the now routine
modalities. All patients who are candidates for im- use of axillary sentinel-node biopsy in breast can-
mediate breast reconstruction should be made cer patients, clinicians at M.D. Anderson now com-
aware that, if it is determined after reconstruction monly use the internal mammary vessels as their
that PMRT is required, the presence of the recon- first choice in immediate free TRAM or DIEP breast
structed breast could decrease the quality of the aes- reconstruction [7], which in addition to other ben-
thetic outcome [1,12,19–25,41–44] and cause efits has eliminated the potential for vascular in-
technical difficulties with radiation delivery jury to the reconstructed breast when the sentinel
[17,18]. However, patients should also be made lymph node is found to be positive on review of
 Immediate Versus Delayed Reconstruction 49

permanent sections and additional axillary nodal mastectomy in breast cancer. Oncologist 2002;
surgery is required [7]. As the indications for 7:539–46.
PMRT and other treatment modalities continue [14] Recht A, Edge SB, Solin LJ, et al. Postmastectomy
to change, plastic surgeons will have to adapt their radiotherapy: clinical practice guidelines of the
American Society of Clinical Oncology. J Clin
approach to breast reconstruction to maintain an
Oncol 2001;19:1539–69.
appropriate balance between minimizing the risk
[15] Overgaard M, Hansen PS, Overgaard J, et al.
of recurrence and providing the best possible Postoperative radiotherapy in high-risk premen-
aesthetic outcome. opausal women with breast cancer who receive
adjuvant chemotherapy. N Engl J Med 1997;
337:949–55.
References [16] Ragaz K, Jackson SM, Le N, et al. Adjuvant radio-
therapy and chemotherapy in node-positive pre-
[1] Kroll SS, Coffey JA, Winn RJ, et al. A comparison menopausal women with breast cancer. N Engl J
of factors affecting aesthetic outcomes of TRAM Med 1997;337:956–62.
flap breast reconstructions. Plast Reconstr Surg [17] Buchholz TA, Kronowitz SJ, Kuerer HM. Immedi-
1995;96:860–4. ate breast reconstruction after skin-sparing mas-
[2] Miller MJ, Rock CS, Robb GL. Aesthetic breast re- tectomy for treatment of advanced breast cancer:
construction using a combination of free trans- radiation oncology considerations. Ann Surg
verse rectus abdominis musculocutaneous flaps Oncol 2002;9:820–1.
and breast implants. Ann Plast Surg 1996;37: [18] Strom E. Radiation therapy for early and ad-
258–64. vanced breast disease. In: Hunt KK, Robb GL,
[3] Giuliano AE, Jones RC, Brennan M, et al. Senti- Strom EA, et al, eds. Breast cancer. New York:
nel lymphadenctomy in breast cancer. J Clin Springer-Verlag; 2001. p. 255–85. Buzdar AU,
Oncol 1997;15:2345–50. Freedman RS, eds. The M.D. Anderson Cancer
[4] Van Diest PJ, Torrenga H, Borgstein PJ, et al. Re- Care Series, vol. 1.
liability of intraoperative frozen section and cy- [19] Kroll SS, Schusterman MA, Reece GP, et al. Breast
tological investigation of sentinel lymph nodes reconstruction with myocutaneous flaps in pre-
in breast cancer. Histopathology 1999;35:14–8. viously irradiated patients. Plast Reconstr Surg
[5] Wiser MR, Montgomery LL, Susnik B, et al. Is 1994;93:460–9.
routine intraoperative frozen-section examina- [20] Evans GRD, Schusterman MA, Kroll SS, et al. Re-
tion of sentinel lymph nodes in breast cancer construction and the radiated breast: is there a role
worthwhile? Ann Surg Oncol 2000;7:651–5. for implants? Plast Reconstr Surg 1995;96:1111–5.
[6] Turner RR, Hansen NM, Stern SL, et al. Intrao- [21] Spear SL, Onyewu C. Staged breast reconstruc-
perative examination of the sentinel lymph tion with saline-filled implants in the irradiated
node for breast carcinoma staging. Am J Clin breast: recent trends and therapeutic implica-
Pathol 1999;112:627–34. tions. Plast Reconstr Surg 2000;105:930–42.
[7] Kronowitz SJ, Chang DW, Robb GL, et al. Impli- [22] Williams JK, Bostwick J III, Bried JT, et al. TRAM
cations of axillary sentinel lymph node biopsy in flap breast reconstruction after radiation treat-
immediate autologous breast reconstruction. ment. Ann Surg 1995;221:756–66.
Plast Reconstr Surg 2002;109:1888–96. [23] Kraemer O, Andersen M, Siim E. Breast recon-
[8] Katz A, Strom EA, Buchholz TA, et al. Locore- struction and tissue expansion in irradiated ver-
gional recurrence patterns after mastectomy and sus not irradiated women after mastectomy.
doxorubicin-based chemotherapy: implications Scand J Plast Reconstr Hand Surg 1996;30:201–6.
for postoperative irradiation. J Clin Oncol [24] Williams JK, Carlson GW, Bostwick J III, et al.
2000;18:2817–27. The effects of radiation treatment after TRAM
[9] Ivens D, Hoe AL, Podd TJ, et al. Assessment of flap breast reconstruction. Plast Reconstr Surg
morbidity from complete axillary dissection. Br 1997;100:1153–60.
J Cancer 1992;66:136–8. [25] Rogers NE, Allen RJ. Radiation effects on breast
[10] Miller MJ. Immediate breast reconstruction. Clin reconstruction with the deep inferior epigastric
Plast Surg 1998;25:145–56. perforator flap. Plast Reconstr Surg 2002;109:
[11] Harris J, Halpin-Murphy P, McNeese M, et al. 1919–24.
Consensus statement on postmastectomy radia- [26] Spear SL, Ducic I, Low M, et al. The effect of
tion therapy. Int J Radiat Oncol Biol Phys 1999; radiation therapy on pedicled TRAM flap breast
44:989–90. reconstruction: outcomes and implications. Plast
[12] Tran NV, Chang DW, Gupta A, et al. Comparison Reconstr Surg 2005;115:84–95.
of immediate and delayed TRAM flap breast re- [27] Kronowitz SJ, Robb GR. Breast reconstruction
construction in patients receiving postmastec- with postmastectomy radiation therapy: current
tomy radiation therapy. Plast Reconstr Surg issues. Plast Reconstr Surg 2004;114:950–60.
2001;108:78–82. [28] Kronowitz SJ, Hunt KK, Kuerer HM, et al. De-
[13] Buchholz TA, Strom EA, Perkins GH, et al. Con- layed-immediate breast reconstruction. Plast
troversies regarding the use of radiation after Reconstr Surg 2004;113:1617–28.
50 Kronowitz

[29] Taylor W, Horgan K, Dodwell D. Oncological as- the conventional TRAM flap? Plast Reconstr
pects of breast reconstruction. Breast 2005;14: Surg 1992;90:255–61.
118–26. [37] Drake D, Oishi S. Wound healing considerations
[30] Ferguson MK. The effect of antineoplastic agents in chemotherapy and radiation therapy. Clin
on wound healing. Surg Gynecol Obstet 1982; Plast Surg 1995;22:31–7.
154:421–9. [38] Ariyan S, Kraft RL, Goldberg NH. An experimental
[31] Lawrence WT, Talbot TL, Norton JA. Preoperative model to determine the effects of adjuvant
or postoperative doxorubicin hydrochloride: therapy on the incidence of postoperative wound
which is better for wound healing? Surgery 1986; infections II: evaluating preoperative chemother-
100:9–13. apy. Plast Reconstr Surg 1990;65:338–45.
[32] Yule GJ, Concannon MJ, Croll GH, et al. Is there [39] Springfield D. Surgical wound healing. In:
liability with chemotherapy following immedi- Verweij J, Pinedo HM, Suit HD, editors. Multidis-
ate breast reconstruction? Plast Reconstr Surg ciplinary treatment of soft tissue sarcomas. Bos-
1996;97:969–73. ton: Luwer Academic Publishers; 1993. p. 81–98.
[33] Furey PC, Macgillivray DC, Castiglione CL, et al. [40] Deutsch MF, Smith M, Wang B, et al. Immediate
Wound complications in patients receiving adju- breast reconstruction with TRAM flap after neo-
vant chemotherapy after mastectomy and imme- adjuvant therapy. Ann Plast Surg 1999;42:240–4.
diate breast reconstruction for breast cancer. [41] Vandeweyer E, Deraemaecker R. Radiation ther-
J Surg Oncol 1994;55:194–7. apy after immediate breast reconstruction with
[34] Allweis TM, Boisvert ME, Otero SE, et al. Imme- implants. Plast Reconstr Surg 2000;106:56–60.
diate reconstruction after mastectomy for breast [42] Robb GL. Reconstructive surgery. In: Hunt KK,
cancer does not prolong the time to starting Robb GL, Strom EA, et al, eds. Breast cancer.
adjuvant chemotherapy. Am J Surg 2002;183: New York: Springer-Verlag; 2001. p. 255–85.
218–21. Buzdar AU, Freedman RS, eds. The M.D. Ander-
[35] Caffo O, Cazzolli D, Scalet A, et al. Concurrent son Cancer Care Series, vol. 1.
adjuvant chemotherapy and immediate breast [43] Hartrampf CR, Bennett GK. Autogenous tissue
reconstruction with skin expanders after mastec- reconstruction in the mastectomy patient. Ann
tomy for breast cancer. Breast Cancer Res Treat Surg 1987;205:508–19.
2000;60:267–75. [44] Rosato RM, Dowden RV. Radiation therapy as
[36] Schusterman M, Kroll S, Weldon M. Immediate a cause of capsular contracture. Ann Plast Surg
breast reconstruction: why the free TRAM over 1994;323:342–5.