1. www.PRSJournal.com 761
BREAST
A
utologous fat grafting is an increasingly
popular technique with which to improve
aesthetic outcomes in breast reconstruc-
tion. First described by Czerny in 1895, fat transfer
to the breast was infrequent until a rapid renewal
of interest occurred in the early 1980s.1–4
Today,
fat grafting is frequently used for breast augmen-
tation and to correct contour deformities in autol-
ogous and prosthetic breast reconstruction.5–10
However, significant controversy regarding
the safety of fat grafting has mirrored significant
advancements in its application. In 1987, the
American Society of Plastic Surgeons released
a position paper stating the following: “The
committee is unanimous in deploring the
use of autologous fat injection ... the known
physiological response to necrosis of this tissue is
scarring and calcification. As a result, detection
Differentiating Fat Necrosis from Recurrent
Malignancy in Fat-Grafted Breasts: An Imaging
Classification System to Guide Management
Rajiv P. Parikh, B.A.
Erin L. Doren, M.D.
Blaise Mooney, M.D.
Weihong V. Sun, M.D., M.S.
Christine Laronga, M.D.
Paul D. Smith, M.D.
Tampa, Fla.
Background: In breast reconstruction with autologous fat grafting, concerns per-
sist about the ability to differentiate palpable masses representing fat necrosis
from recurrent cancer. The authors’ objective was to develop standardized imag-
ing classifications to distinguish benign from malignant lesions after fat grafting.
Methods: A database of 286 breast reconstruction patients undergoing fat graft-
ing from 2006 to 2011 was retrospectively reviewed to identify patients with
imaging of clinically palpable masses. All images were reviewed independent-
ly by a radiologist blinded to prior results. Lesions were classified, using the
American College of Radiology Breast Imaging Reporting and Data System ul-
trasound lexicon, as follows: A, solid mass, hypoechoic; B, solid mass, isoechoic;
C, solid mass, hyperechoic; D, solid mass, complex echogenicity; E, anechoic
mass with posterior acoustic enhancement; F, cystic mass with internal echoes;
and G, negative. Evolutions in lesions on follow-up ultrasound were recorded.
Images were correlated with histopathologic results.
Results: On ultrasound, 66 lesions were visualized in 37 patients with palpable
masses. Twenty-two lesions (33 percent) were Breast Imaging Reporting and
Data System category 4 lesions; biopsies were performed on all of them. Histo-
pathologic results revealed that 85.7 percent (six of seven) with classification
D and 100 percent with classifications A, B, C, E, F, and G were fat necrosis.
The one malignant lesion (classification D) exhibited vascularity and angular
margins on ultrasound and was not in the location of fat injection. Negative
predictive value of avascularity and circumscribed margins for malignancy was
100 percent. Follow-up ultrasound of 29 lesions at a median of 6.5 months
revealed that no masses increased in size or developed vascularity.
Conclusion: Ultrasound analysis, with a standardized classification system, is
reliable at differentiating benign from malignant lesions after fat grafting in
breast reconstruction. (Plast. Reconstr. Surg. 130: 761, 2012.)
CLINICAL QUESTION/LEVEL OF EVIDENCE: Diagnostic, III.
Disclosure: The authors have no financial interests
or commercial associations related to this article. No
funds were used to support the research for this article.
From the Division of Plastic Surgery, Department of Surgery,
University of South Florida, and the Department of Radiol-
ogy and the Comprehensive Breast Program, H. Lee Moffitt
Cancer Center.
Received for publication April 8, 2012; accepted April 19,
2012.
Presented in part at the Seventh Annual Academic Surgical Con-
gress, in Las Vegas, Nevada, February 14 through 16, 2012.
Copyright © 2012 by the American Society of Plastic Surgeons
DOI: 10.1097/PRS.0b013e318262f03b

2. 762
Plastic and Reconstructive Surgery • October 2012
of early breast carcinoma through xerography
and mammography will become difficult and
the presence of disease may go undiscovered.”11
Twenty years later, scientific reports of favorable
outcomes led the American Society of Plastic
Surgeons to convene a task force to reassess
the safety of fat grafting.5–7,12
Chief among their
conclusions was that further research is needed to
“assess the effect of fat grafting on breast cancer
detection and treatment.”12
To date, concerns persist about our ability
to differentiate palpable lesions representing fat
necrosis from palpable lesions representing recur-
rent cancer, with some going so far as to propose
prohibiting fat grafting entirely.13
The rationale
for this is the dearth of published reports on how
to objectively, effectively, and safely distinguish
benign from malignant lesions. This uncertainty
can contribute to an increase in unnecessary
biopsies and exacerbate patient anxiety about
recurrence.
To address these concerns, this study aims
to develop imaging classifications that can be
applied objectively to accurately differentiate fat
necrosis from recurrent cancer. Ultimately, we
hypothesize that a postoperative approach that
includes systematic imaging analysis, integration
of our medical colleagues, and increased patient
awareness will enable the accurate identification
of a palpable mass after fat grafting with minimal
detrimental complications.
PATIENTS AND METHODS
An institutional review board–approved data-
base of 286 prospectively gathered breast recon-
struction patients undergoing fat grafting at a large
tertiary care cancer center was reviewed (approval
for this study was obtained from the University of
South Florida College of Medicine and H. Lee
Moffitt Cancer Center Institutional Review Board,
no. MS7_Pro00001277). A retrospective study of a
subset of this population having imaging analysis
of clinically palpable subcutaneous masses after fat
grafting was conducted between January of 2006
and August of 2011. To be included in the study,
patients had to have their surgical procedure for
breast cancer or risk reduction followed by breast
reconstruction at our institution, postoperative
imaging, and at least 1-year follow-up. Fat graft-
ing was indicated to correct contour deformities
in reconstructed breasts. All procedures were
performed by the same surgeon using the same
principal surgical technique: sites of anticipated
fat procurement and fat injection were marked
preoperatively with the patient standing. All pro-
cedures were performed under general anesthe-
sia, with maintenance of strict sterile technique.
Fat was most commonly harvested from the lateral
thigh, abdominal, and flank donor sites. Before
lipoaspiration, the donor site was infiltrated with
tumescent solution (1000 cc saline, 10 cc of 1%
lidocaine, and 1:100,000 epinephrine) using a
small-bore cannula. After 15 minutes, fat was aspi-
rated by conventional liposuction technique using
a 4-mm blunt cannula attached to a liposuction
machine on low negative-pressure setting [−250
mmHg (0.33 atm)]. Harvested fat was transferred
immediately into 10-cc syringes that were subse-
quently capped. The senior author (P.D.S.) has
previously demonstrated that minimal prepara-
tory manipulation results in superior graft survival;
therefore, we prefer allowing fat to purify by gravity
sedimentation over centrifugation.14
The length of
extracorporeal adipocyte time was minimized to
that necessary for supernatant separation. Fat was
not concentrated on cotton gauze or Telfa pads in
an effort to minimize handling and the amount of
open air exposure. After discarding a layer of oil
anddebris,fractionaldispersionofsmallaliquotsof
fat occurred in multiple passes and overcorrection
was avoided to ensure favorable graft-recipient sur-
face-to-volume ratio. Fat was injected subdermally,
with a blunt 2.5-mm cannula attached to a 10-cc
syringe, through small incisions in premarked
areas of the breast with contour irregularities. The
recipient sites were managed postoperatively with
light, nonpressurized dressings. The location of
injection within the breast and the amount of fat
injected were recorded in the operative report.
Exclusion criteria were patients not experiencing
palpable masses, lack of imaging, and/or having
less than 1-year follow-up.
Descriptive Radiographic Analysis
All patients in this study had postoperative
imaging of palpable lesions with focused breast
ultrasonography. Ultrasonographic findings were
initially classified into four categories according to
the American College of Radiology Breast Imaging
Reporting and Data System.15
To develop a more
comprehensive and objective approach to the
imaging of palpable lesions after fat grafting, all
ultrasound records were reviewed independently
by a single breast radiologist (B.M.) with extensive
experience. Blinded to prior radiologic and
pathologic results, the radiologist characterized
lesions according to echo texture, location, size,
vascularity, and margins in accord with published
guidelines in the literature.15
Ultrasound lexicon

3. Volume 130, Number 4 • Analysis of Fat-Grafting Complications
763
was used to develop classifications for lesions as
follows: A, solid mass, hypoechoic; B, solid mass,
isoechoic; C, solid mass, hyperechoic; D, solid
mass, complex echogenicity; E, anechoic mass
with posterior acoustic enhancement/typical
cystic appearance; F, cystic mass with internal
echoes; and G, negative findings.
Radiographic and Histopathologic Follow-Up
The time interval to and the number of follow-
up ultrasound examinations were recorded. Evo-
lution in lesion characteristics, size, and vascularity
was documented as follows: (1) disappearance of
mass, (2) stable findings, (3) decrease in size, (4)
transition to more cystic appearance, (5) decrease
in size and transition to more cystic appearance,
(6) increase in size, and/or (7) development of
vascularity. Percutaneous biopsies of all Breast
Imaging Reporting and Data System class 4 lesions
were performed. Occasionally, biopsies of Breast
Imaging Reporting and Data System class 2 and 3
lesions were warranted because of patient and/or
clinician preference. An experienced pathologist
analyzed tissue specimens and diagnosed lesions.
RESULTS
Patient Information
Thirty-seven patients constituted the study
cohort after inclusion and exclusion criteria were
applied. Table 1 lists patient demographic infor-
mation and surgical procedures. The average age
was 53 years (range, 40 to 71 years). Of the original
286 patients undergoing fat-grafting procedures,
90 (31.5 percent) had undergone irradiation
before fat grafting. There was no statistically sig-
nificant relationship between the incidence of fat
necrosis and previous breast radiation therapy in
our population (p = 0.8605). There were 69 total
reconstructive procedures in these 37 patients: 14
autologous free flaps (20.3 percent) (10 transverse
rectus abdominis myocutaneous flaps and four
deep inferior epigastric artery perforator flaps);
nine latissimus dorsi flaps with implants (13.0
percent), and 46 tissue expander/implant recon-
structions (66.7 percent). Fat injections occurred
in aesthetic breast subunits. The superior breast
pole (38 percent) and upper inner quadrant
(34 percent) were the most common recipient
units, followed by the upper outer quadrant (10
percent), the lower outer quadrant (1 percent),
and the lower inner quadrant (1 percent); 16 per-
cent of the patients had injections in more than
one location. The average volume injected was
42.77 ± 25.99 cc.
Qualitative Ultrasound Analysis of Masses
Ultrasound of clinically palpable lesions
occurred at a median 9 months after fat grafting. A
total of 66 lesions were visualized, most commonly
in the superior breast (12-o’clock position) and
upper inner quadrant subunit (left breast, 10- to
11-o’clock position; right breast, 1- to 2-o’clock
position). The superior breast pole was the
location for 33 percent of the lesions; 35 percent
were found in the upper inner quadrant, 15
percent were found in the upper outer quadrant,
11 percent were found in the lower inner
quadrant, and 6 percent were found in the lower
outer quadrant. There were 32 (48.5 percent)
Breast Imaging Reporting and Data System class
2, 12 (18.2 percent) class 3, and 22 (33.0 percent)
class 4 lesions initially classified on ultrasound. On
blinded retrospective analysis, lesion echo texture
and echogenicity were as classified in Table 2.
The most common feature was a typical cystic
appearance of an anechoic mass with posterior
acoustic enhancement (40.9 percent) (Fig. 1,
above, left). We observed 25 (37.9 percent) solid
appearing masses. Of these, 15 (22.7 percent)
were hypoechoic (Fig. 1, above, right), seven
(10.6 percent) were of complex echogenicity
(Fig. 1, center, left), two (3.0 percent) were
hyperechoic (Fig. 1, center, right), and one (1.5
percent) was isoechoic (Fig. 1, below, left). There
were 10 masses (15.2 percent) that had a cystic
appearance with internal echoes (Fig. 1, below,
Table 1. Patient Demographics and Surgical Data
Value (%)
Patient population 37
Age, yr
Mean 53
Range 40–71
BMI, kg/m2
Mean 27.4
Range 19.6–42.1
Race
Caucasian 31 (83.8)
African American 3 (8.1)
Hispanic 3 (8.1)
Family history of breast cancer 21 (56.8)
Personal history of breast cancer 9 (24.3)
Active smokers 4 (10.8)
Preprocedure radiation therapy 12 (32.4)
Bilateral mastectomy 28 (75.7)
Unilateral mastectomy 9 (24.3)
Total reconstructive procedures 69
Autologous only (TRAM, FTRAM, DIEP) 14 (20.3)
TE/LD 9 (13.0)
TE/implant 46 (66.7)
Total fat-grafting procedures 82*
BMI, body mass index; TRAM, transverse rectus abdominis muscu-
locutaneous; FTRAM, free TRAM; DIEP, deep inferior epigastric per-
forator; TE, tissue expander; LD, latissimus dorsi.
*2.22 per patient.

4. 764
Plastic and Reconstructive Surgery • October 2012
right). We were unable to visualize a radiographic
abnormality for four lesions documented on initial
ultrasound report. Of note, there was no statistical
correlation between specific lesion classifications
(A to F), margins, or internal blood flow signal
demonstrated on ultrasonography and any
patient demographic factors (body mass index,
family or personal history of breast cancer, and/
or tobacco use) or any procedural factors (prior
breast irradiation and/or type of reconstructive
procedure). Margins were circumscribed in 59 of
66 masses (89.4 percent) and not circumscribed
(angular) in three of 66 masses (4.5 percent).
Similarly, internal blood flow signal was avascular/
absent in 59 of 66 masses (89.4 percent) and
vascular in three of 66 masses (4.5 percent). The
average size of Breast Imaging Reporting and
Data System class 2, 3, and 4 lesions was 5.3, 6.6,
and 15.9mm, respectively.
Relationship between Ultrasound Characteristics
and Histopathologic Analysis
Nineteen of 37 patients (51.4 percent) under-
went biopsies of 31 masses for histopathologic
diagnosis of 22 class 4, two class 3, and seven
class 2 lesions according to the Breast Imaging
Reporting and Data System. Table 3 demon-
strates the relationship between biopsy results and
Table 2. Ultrasound Characteristics of Masses
Classification Description
No. of
Cases (%)
A Solid mass, hypoechoic 15 (22.7)
B Solid mass, isoechoic 1 (1.5)
C Solid mass, hyperechoic 2 (3.0)
D Solid mass, complex 7 (10.6)
E Typical cystic appearance/
anechoic mass with posterior
acoustic enhancement
27 (40.9)
F Cystic mass with internal echoes 10 (15.2)
G Negative ultrasound findings 4 (6.1)
Fig. 1. Ultrasound images that represent lesion classifications developed: (above, left) solid mass, hypoechoic; (above, right) solid
mass, isoechoic; (center, left) solid mass, hyperechoic; (center, right) solid mass, complex echogenicity; (below, left) anechoic mass
with posterior acoustic enhancement/typical cystic appearance; (below, right) cystic mass with internal echoes.

5. Volume 130, Number 4 • Analysis of Fat-Grafting Complications
765
ultrasound features. Of the 22 class 4 lesions, 85.7
percent (six of seven) with classification D and
100 percent of the palpable masses with classifica-
tions A, B, C, E, F, and G were benign fat necro-
sis/oil cysts on histopathology. Nineteen of 22
(86.4 percent) were avascular and 19 of 22 (86.4
percent) were circumscribed. Of the Breast Imag-
ing Reporting and Data System class 2 and 3
lesions, two were class A, six were class E, and one
was class F. All nine (100 percent) were benign fat
necrosis/oil cysts on histopathology.
One patient presented with two palpable
lesions on clinical examination. This patient
underwent fat grafting to the left breast upper
inner quadrant. On ultrasound, two masses were
detected. One was located at the 10-o’clock posi-
tion (upper inner quadrant) in the location of fat
injection. This was categorized as Breast Imaging
Table 3. Correlation between Ultrasound Features and Pathologic Diagnosis after Biopsy
Ultrasound
Classification*
Internal
Blood Flow Margins Diagnosis Histopathologic Description
A Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
B Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
D Avascular Not circumscribed FN/OC Extensive fibrosis with multinucleated giant cells,
fibroblasts, and lipid-filled macrophages
A Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
A Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
A Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
F Avascular Circumscribed FN/OC Lipid cyst with minimal fibrosis at periphery
F Avascular Circumscribed FN/OC Lipid cyst with minimal fibrosis at periphery
D Vascular Circumscribed FN/OC Extensive fibrosis with multinucleated giant cells,
fibroblasts, and lipid-filled macrophages
E Avascular Circumscribed FN/OC Lipid cyst with minimal fibrosis at periphery
E Avascular Circumscribed FN/OC Lipid cyst with minimal fibrosis at periphery
F Avascular Circumscribed FN/OC Lipid cyst with minimal fibrosis at periphery
D Avascular Circumscribed FN/OC Extensive fibrosis with multinucleated giant cells,
fibroblasts, and lipid-filled macrophages
A Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
A Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
D Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
E Avascular Circumscribed FN/OC Fibrosis and lipid-filled macrophages
E Avascular Circumscribed FN/OC Lipid cyst, minimal fibrosis at periphery,
multinucleated giant cells
E Avascular Circumscribed FN/OC Foreign body giant cells and adipose tissue
E Avascular Circumscribed FN/OC Foreign body giant cells and adipose tissue
D Vascular Not circumscribed Recurrent
cancer
Well-differentiated invasive lobular carcinoma
F Avascular Circumscribed FN/OC Extensive fibrosis with multinucleated giant cells,
fibroblasts, and lipid-filled macrophages
C Vascular Circumscribed FN/OC Lipid cyst, no fibrosis
C Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
E Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
A Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
D Avascular Circumscribed FN/OC Multinucleated giant cells, fibroblasts, and lipid-filled
macrophages
D Avascular Circumscribed FN/OC Extensive fibrosis with multinucleated giant cells,
fibroblasts, and lipid-filled macrophages
A Avascular Not circumscribed FN/OC Extensive fibrosis with multinucleated giant cells,
fibroblasts, and lipid-filled macrophages
E Avascular Circumscribed FN/OC Lipid cyst with minimal fibrosis at periphery
F Avascular Circumscribed FN/OC Lipid cyst with minimal fibrosis at periphery
FN/OC, fat necrosis/oil cyst.
*Ultrasound classifications: A, solid mass, hypoechoic; B, solid mass, isoechoic; C, solid mass, hyperechoic; D, solid mass, complex echogenic-
ity; E, typical cystic appearance/anechoic mass with posterior acoustic enhancement; F, cystic mass with internal echoes.

6. 766
Plastic and Reconstructive Surgery • October 2012
Reporting and Data System class 2 and further
classified as a circumscribed, avascular, cystic mass
with internal echoes (class F) (Fig. 1, below, right).
In contrast, the other mass was at the 3-o’clock
position (not in the location of fat injection).
This was categorized as Breast Imaging Reporting
and Data System class 4 and classified as a solid
mass with complex echogenicity (class D) (Fig. 2,
above). This mass exhibited heightened vascularity
and angular margins (Fig. 2, below). Both lesions
were submitted to biopsy without delay and, on
histopathologic evaluation, the lesion at the
3-o’clock position was recurrent invasive lobular
carcinoma and the lesion at the 10-o’clock posi-
tion was an oil cyst.
Avascularity and circumscribed margins were
highly predictive of benign pathology, with nega-
tive predictive values for malignancy of 100 per-
cent (28 of 28). Vascularity and angular margins
each independently had a positive predictive
value for malignancy of 33 percent (one of three).
However, the combination of both vascularity and
angular margins had a positive predictive value
for malignancy of 100 percent (one of one).
Evolution in Lesion Characteristics on Follow-Up
Ultrasound
Follow-up ultrasound occurred at a median
6.5 months in 17 patients, and evolution in 29
masses was documented (Table 4). Twenty masses
remained stable, five decreased in size, two transi-
tioned to a more cystic appearance, and one both
decreased in size and transitioned to a more cys-
tic appearance (Figs. 3 and 4). One mass disap-
peared and 12 masses were downgraded in Breast
Imaging Reporting and Data System category on
follow-up analysis. No masses increased in size,
developed vascularity, or prompted biopsy.
DISCUSSION
With any innovative technique in plastic sur-
gery, it is essential to demonstrate that patient
safety is not compromised for improved aesthetic
results. For a procedure or technique to gain wide-
spread acceptance in breast reconstruction, onco-
logic safety must first be established. Recently,
reports in the literature have alleviated concerns
that fat grafting contributes to an increased risk of
recurrence in the breast cancer population.16,17
To
date, there appear to be no in vivo data to suggest
a relationship between fat grafting and locore-
gional recurrence.
Despite this, several concerns persist. Chief
among these is that recurrent breast cancer pres-
ents similarly to fat necrosis/oil cysts, a potential
sequela of fat grafting, and it may be difficult to
distinguish between the two on imaging. In addi-
tion, logistical issues have been raised regarding
the surveillance of patients injected with fat, such
as (1) what the optimal postoperative follow-up is;
and (2) when patients are followed by their local
health care providers, whether they are equipped
with the knowledge and experience to differen-
tiate benign from malignant lesions and, if not,
determining the best way of educating and inte-
grating our medical colleagues in breast surgical
oncology, radiology, and general practice to safely
manage complications.18,19
This report strives
to resolve several of the issues complicating the
follow-up and management of patients after fat
grafting by detailing a sequential approach that
includes the following: (1) strict patient follow-
up at periodic intervals; (2) ultrasound analysis
with objective classification of palpable lesions
detected on follow-up; and (3) integrated man-
agement of palpable lesions by a team that incor-
porates the patient, an experienced radiologist, a
breast surgeon, and a plastic surgeon.
Currently, there are no National Compre-
hensive Cancer Network guidelines that define a
screening protocol following mastectomy with or
without breast reconstruction.20
At most centers,
surveillance is by physical examination. Suspi-
cious examination findings then undergo imaging
Fig. 2. Ultrasound images of biopsy-proven recurrent malig-
nancy. (Above) Ultrasound image shows a solid mass with com-
plex echogenicity, classification D. (Below) Solid mass exhibits
heightened vascularity and angular margins.

7. Volume 130, Number 4 • Analysis of Fat-Grafting Complications
767
analysis. At our institution, tailored diagnostic
imaging with a focused breast ultrasonogram is
standard protocol. Some authors have suggested
that mammography is an effective imaging modal-
ity following breast reconstruction.6,21
However,
in the last year alone, contradictory reports have
Table 4. Evolution in Ultrasound Characteristics on Follow-Up Imaging
Ultrasound
ID No.
Initial Ultrasound
Classification
Initial
BI-RADS
Category
Lesion Size
(mm)
Time Interval to
Follow-Up
Ultrasound (mo)
Follow-Up
Ultrasound
Characteristics
BI-RADS
Category after
Follow-Up
2
Anechoic mass with posterior
acoustic enhancement (cystic
appearance) 3 8 5 Stable findings 3
2 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
3 4 5 Stable findings 3
4 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
2 5 6 Decrease in size 2
5 Cystic mass with internal echoes 3 8 7 Transition to
more cystic
2
5 Cystic mass with internal echoes 3 7 7 Stable findings 3
6 Solid mass, hypoechoic 3 13 7 Stable findings 3
15 Solid mass, hypoechoic 4 3 5 Stable findings 3
15 Solid mass, hypoechoic 4 7 5 Stable findings 3
19 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
3 9 2 Decrease in size 2
22 Solid mass, hypoechoic 4 14 6 Stable findings 3
22 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
3 4 6 Stable findings 3
25 Cystic mass with internal echoes 4 8 1 Stable findings 3
28 Negative 1 0 6 Stable findings 1
29 Solid mass with complex
echogenicity
4 11 6 Stable findings 3
30 Solid mass, hypoechoic 2 7 6 Transition to
more cystic
2
31 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
2 6 12 Stable findings 2
31 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
2 12 12 Stable findings 2
31 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
2 5 12 Stable findings 2
32 Solid mass with complex
echogenicity
4 15 12 Disappearance
of abnormality
1
33 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
4 12 7 Decrease in size 2
33 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
2 4 7 Stable findings 2
33 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
2 7 7 Stable findings 2
36 Solid mass with complex
echogenicity
4 17 6 Decrease in size
and transition
to more cystic
2
40 Solid mass, hypoechoic 3 5 8 Stable findings 3
42 Solid mass with complex
echogenicity
4 18 4 Stable findings 4
42 Cystic mass with internal echoes 2 8 4 Stable findings 2
43 Solid mass, hypoechoic 3 4 6 Decrease in size 2
44 Solid mass, hypoechoic 3 6 12 Decrease in size 2
44 Anechoic mass with posterior
acoustic enhancement (cystic
appearance)
3 5 12 Stable findings 3
BI-RADS, Breast Imaging Reporting and Data System.

8. 768
Plastic and Reconstructive Surgery • October 2012
been published regarding the ability of mammog-
raphy to accurately distinguish recurrent malig-
nancy from benign fat necrosis.13,22
This makes it
a necessity to evaluate whether other frequently
used imaging modalities are more reliable. The
goals of imaging should be to (1) classify lesions
with an objective lexicon that can be standardized
across institutions, (2) accurately rule out recur-
rence, and (3) establish a diagnosis that guides
management. For the postoperative workup of fat
necrosis, there are a limited number of studies that
have achieved these goals. What minimal reports
exist are primarily composed of data obtained
after fat grafting for cosmetic augmentation in
women with healthy breast tissue.21–24
Applying
these conclusions to the breast reconstruction
population has questionable validity given the
inherent risk of tumor recurrence and unique
recipient environment for injected fat in patients
with a history of breast cancer. These studies are
also limited by not correlating imaging features
with pathologic diagnoses. This has led experts
to call for the development of new imaging stan-
dards that can distinguish benign from malignant
lesions in fat-grafted breasts.25
Our study is the first in the English literature
to apply an objective imaging classification sys-
tem to evaluate clinically palpable masses after
fat grafting in breast reconstruction patients. We
used the American College of Radiology ultra-
sound lexicon to develop a standardized classi-
fication system based on lesion characteristics.
Breast Imaging Reporting and Data System cat-
egorization alone is insufficient because it is often
subjective, vague, and highly dependent on expe-
rience.26
By developing a systematic approach,
the translation of ultrasound findings into Breast
Imaging Reporting and Data System categories
can be standardized across institutions. This can
improve patient safety by minimizing unnecessary
biopsies and patient anxiety.
The previous standard has been to perform
a biopsy on palpable lesions that appear “suspi-
cious” on imaging after fat grafting.18
Unfortu-
nately, little guidance has been given as to what
constitutes a suspicious finding. Our results
resolve this issue by correlating imaging classifi-
cations with histopathologic results. Therefore,
recommendations can be made for when sus-
picion should be heightened and when doubt
should be minimized. Foremost, masses with
both uncircumscribed margins and vascular
internal blood flow signals should be submitted
to biopsy promptly. These findings are highly pre-
dictive for recurrent carcinoma, with a positive
predictive value of 100 percent in our study.27–29
Malignant lesions infiltrate tissue, developing
characteristic flow on ultrasound to reflect neo-
angiogenesis necessary for tumor growth.29
Pal-
pable lesions in areas of the breast not injected
with fat should also undergo biopsy; clearly, these
new lesions cannot be attributed to grafted fat.
In our population, one patient experienced can-
cer recurrence. This patient presented with two
palpable lesions on physical examination that
were clearly differentiated on ultrasound. One
mass was in a location where fat was not injected
and exhibited typical malignant features (Fig. 2).
The other mass exhibited typical benign features
Fig. 3. Ultrasound images for a lesion that decreased in size on
follow-up analysis. (Above) Initial ultrasound image obtained
9 months after the fat-grafting procedure. (Below) Ultrasound
image obtained at 6-month follow-up visit.
Fig. 4. Ultrasound images for a lesion that transitioned to a
more cystic appearance on follow-up analysis. (Above) Initial
ultrasound image obtained 10 months after the fat-grafting
procedure. (Below) Ultrasound image obtained at the 6-month
follow-up visit.

9. Volume 130, Number 4 • Analysis of Fat-Grafting Complications
769
(Fig. 1, below, right). These features include cir-
cumscribed margins and avascularity, which had
a negative predictive value for malignancy of
100 percent. Lesions with benign features, clas-
sified as an anechoic mass with posterior acous-
tic enhancement (class E) or a cystic mass with
internal echoes (class F), should be categorized
Breast Imaging Reporting and Data System class
2; biopsy and further imaging are not necessary.
These lesions have a classic presentation on ultra-
sound that correlates with the histopathologic
appearance of a lipid cyst contained in a thin rim
of fibrosis.
An additional contribution of this study is how
to deal with previously indeterminate findings. To
date, many have considered biopsy the next step
in management for all circumscribed solid masses
detected on ultrasound and therefore character-
ized these as Breast Imaging Reporting and Data
System class 4 lesions.30,31
However, biopsy is not
warranted in all solid masses. This practice reflects
the lack of published data proving the safety of
following lesions in fat-grafted breasts with imag-
ing. Faced with the choice of establishing a certain
diagnosis by means of biopsy or the uncertainty of
noninvasive monitoring, most have chosen a “safe
rather than sorry” approach. However, additional
invasive procedures in breast cancer patients
carry a substantial risk of physical and psychologi-
cal harm.32,33
Our data indicate that solid masses
with classifications A to D not exhibiting both
vascularity and irregular margins were confirmed
to be benign by biopsy. The solid appearance on
ultrasound represents the presence of inflam-
mation and extensive fibrosis on histopathology.
These lesions can be managed by periodic imag-
ing surveillance instead of biopsy, and the Breast
Imaging Reporting and Data System classification
should reflect that. We have shown that imaging
follow-up of palpable masses 6 months after fat
grafting is both safe and effective. In our study, no
lesions increased in size or developed vascularity.
Benign lesions remained stable, transitioned to a
more cystic appearance, or decreased in size on
follow-up ultrasound.
The ability to differentiate benign lesions
from malignant lesions with imaging marks signif-
icant progress in ensuring the safety of fat graft-
ing. For some, without this assurance, the mere
potential of fat necrosis/oil cysts is a contraindi-
cation.13
This should not be the case. Fat necrosis
is a potential sequela of any operative interven-
tion in the breast, including biopsy, reduction,
augmentation, and reconstruction.34–37
The most
accepted theory is that insufficient vascularization
and ischemia are primarily responsible. The
senior author (P.D.S.) has extensive fat-grafting
experience and strives to minimize complica-
tions by injecting small aliquots of fat into several
planes of the breast through several passes and
restricting the volume of fat injected in a single
procedure.5–8
The advantage of these refinements
is that injected fat maintains an optimal surface
area interface with the recipient environment’s
native vascularity. Unfortunately, regardless of
technique or surgeon experience, fat necrosis
is an inherent complication reported in almost
every clinical study on fat grafting to the bre
ast.5–9,12,13,16–18,22,38,39
In a prior animal model inves-
tigation, the senior author (P.D.S.) demonstrated
that no specific technique of harvesting or pro-
cessing fat was more advantageous for fat graft
viability.14
These findings confirmed a separate
experiment by Rohrich et al.40
Understanding the
potential for a palpable mass to occur has indica-
tions for postoperative management.
Breast surgeons instruct patients to perform
self-breast examinations to detect potential
recurrence. Thus, the occurrence of a new mass
can have a detrimental psychological effect.41–43
If educated that fat necrosis can present similarly
on physical examination, patients are prepared
for rather than blind-sided by this complication.
Postoperatively, in an ideal situation, patients
return at regular intervals for monitoring. The
ideal is not always real. At tertiary care centers,
patients come from a wide geographic region.
If complications arise, local providers, with little
fat-grafting experience and minimal awareness of
the patient’s history, are often tasked with manag-
ing them. If fat necrosis occurs, patients may have
imaging at an easily accessible radiology practice,
albeit one with limited experience interpreting
findings in fat-grafted breasts. We address this
issue by emphasizing an integrated approach
that relies on patient and physician awareness.
All patients are provided extensive details on the
potential complications of fat grafting and given
supplemental materials to reference.44,45
Patients
are also given a card, after authorizing the release
of medical information, they can present to their
local provider if complications develop (Fig. 5).
This card contains details of their fat-grafting
operation. It encourages providers to refer
patients back to our institution to facilitate the
management of complications. When returning
to our care is not possible, providers are advised
to consult the operating surgeons with any ques-
tions. Finally, our senior breast radiologist (B.M.)
is made available to reassess any questionable

10. 770
Plastic and Reconstructive Surgery • October 2012
images before biopsy to ensure the accuracy of
ultrasound interpretation.
CONCLUSIONS
Our study indicates that ultrasound analysis
is reliable at differentiating palpable masses rep-
resenting fat necrosis from palpable masses rep-
resenting recurrent cancer after autologous fat
grafting in breast reconstruction. The standard-
ized classification system described in this study
should mitigate the bias of physician experience
and enable radiologists across institutions to
accurately identify palpable lesions. Continued
research efforts consisting of prospective studies
that use this imaging classification system and
compare imaging modalities are necessary to
confirm the reproducibility and validity of our
findings.
‍REFERENCES
1. Czerny V. Plastischer Ersatz der Brustdruse durch ein Lipom.
Zentralbl Chir. 1895;27:72–75.
Fig. 5. Patient card provided to each patient after autologous fat grafting and
following authorization to release medical information. Card contains details of
operation, including location and volume of fat injected. Patients are encouraged
to present this card to their local provider if complications develop. Contact num-
bers for the operating breast and plastic surgeon are included, as are resources
for providers to access if assistance is needed.
Paul D. Smith, M.D.
Division of Plastic Surgery
University of South Florida College of Medicine
Harbourside Medical Tower
4 Columbia Drive, Suite 650
Tampa, Fla. 33606
pausmith@mac.com

11. Volume 130, Number 4 • Analysis of Fat-Grafting Complications
771
2. BircollM,NovackBH.Autologousfattransplantationemploy-
ing liposuction techniques. Ann Plast Surg. 1987;18:327–
329.
3. Bircoll M. Cosmetic breast augmentation utilizing autolo-
gous fat and liposuction techniques. Plast Reconstr Surg.
1987;79:267–271.
4. Illouz YG. The fat cell “graft”: A new technique to fill depres-
sions. Plast Reconstr Surg. 1986;78:122–123.
5. Spear SL, Wilson HB, Lockwood MD. Fat injection to correct
contour deformities in the reconstructed breast. Plast Recon-
str Surg. 2005;116:1300–1305.
6. Coleman SR, Saboeiro AP. Fat grafting to the breast revisited:
Safety and efficacy. Plast Reconstr Surg. 2007;119:775–85; dis-
cussion 786.
7. Delay E, Garson S, Tousson G, Sinna R. Fat injection to the
breast: Technique, results, and indications based on 880 pro-
cedures over 10 years. Aesthet Surg J. 2009;29:360–376.
8. Kanchwala SK, Glatt BS, Conant EF, Bucky LP. Autolo-
gous fat grafting to the reconstructed breast: The manage-
ment of acquired contour deformities. Plast Reconstr Surg.
2009;124:409–418.
9. Sinna R, Delay E, Garson S, Delaporte T, Toussoun G. Breast
fat grafting (lipomodelling) after extended latissimus dorsi
flap breast reconstruction: A preliminary report of 200 con-
secutive cases. J Plast Reconstr Aesthet Surg. 2010;63:1769–1777.
10. Del Vecchio DA, Bucky LP. Breast augmentation using
preexpansion and autologous fat transplantation: A
­clinical radiographic study. Plast Reconstr Surg. 2011;127:
2441–2450.
11. Report on autologous fat transplantation. ASPRS Ad-Hoc
Committee on New Procedures, September 30, 1987. Plast
Surg Nurs. 1987;7:140–141.
12. Gutowski KA; ASPS Fat Graft Task Force. Current applica-
tions and safety of autologous fat grafts: A report of the ASPS
fat graft task force. Plast Reconstr Surg. 2009;124:272–280.
13. Wang CF, Zhou Z, Yan YJ, Zhao DM, Chen F, Qiao Q. Clini-
cal analyses of clustered microcalcifications after autologous
fat injection for breast augmentation. Plast Reconstr Surg.
2011;127:1669–1673.
14. Smith P, Adams WP Jr, Lipschitz AH, et al. Autologous
human fat grafting: Effect of harvesting and preparation
techniques on adipocyte graft survival. Plast Reconstr Surg.
2006;117:1836–1844.
15. American College of Radiology. BI-RADS: Ultrasound.
In: Breast Imaging Reporting and Data System: BI-RADS
Atlas. 4th ed. Reston, Va: American College of Radiology;
2003.
16. Petit JY, Lohsiriwat V, Clough KB, et al. The oncologic out-
come and immediate surgical complications of lipofilling in
breast cancer patients: A multicenter study–Milan-Paris-Lyon
experience of 646 lipofilling procedures. Plast Reconstr Surg.
2011;128:341–346.
17. Petit JY, Botteri E, Lohsiriwat V, et al. Locoregional recur-
rence risk after lipofilling in breast cancer patients. Ann
Oncol. 2012;23:582–588.
18. Hyakusoku H, Ogawa R, Ono S, Ishii N, Hirakawa K. Com-
plications after autologous fat injection to the breast. Plast
Reconstr Surg. 2009;123:360–70; discussion 371.
19. Walden JL. Complications after autologous fat injection to
the breast. Plast Reconstr Surg. 2009;124:326–327.
20. Zakhireh J, Fowble B, Esserman LJ. Application of screen-
ing principles to the reconstructed breast. J Clin Oncol.
2010;28:173–180.
21. Carvajal J, Patiño JH. Mammographic findings after breast
augmentation with autologous fat injection. Aesthet Surg J.
2008;28:153–162.
22. Veber M, Tourasse C, Toussoun G, Moutran M, Mojallal A,
Delay E. Radiographic findings after breast augmentation by
autologous fat transfer. Plast Reconstr Surg. 2011;127:1289–
1299.
23. Kwak JY, Lee SH, Park HL, Kim JY, Kim SE, Kim EK. Sono-
graphic findings in complications of cosmetic breast aug-
mentation with autologous fat obtained by liposuction. J Clin
Ultrasound. 2004;32:299–301.
24. Wang H, Jiang Y, Meng H, Zhu Q, Dai Q, Qi K. Sonographic
identification of complications of cosmetic augmentation
with autologous fat obtained by liposuction. Ann Plast Surg.
2010;64:385–389.
25. Del Vecchio DA. Discussion: Clinical analyses of clustered
microcalcifications after autologous fat injection for breast
augmentation. Plast Reconstr Surg. 2011;127:1674–1676.
26. Baker JA, Kornguth PJ, Soo MS, Walsh R, Mengoni P.
Sonography of solid breast lesions: Observer variability
of lesion description and assessment. AJR Am J Roentgenol.
1999;172:1621–1625.
27. Moss HA, Britton PD, Flower CD, Freeman AH, Lomas DJ,
Warren RM. How reliable is modern breast imaging in differ-
entiating benign from malignant breast lesions in the symp-
tomatic population? Clin Radiol. 1999;54:676–682.
28. Stavros AT, Thickman D, Rapp CL, Dennis MA, Parker SH,
Sisney GA. Solid breast nodules: Use of sonography to dis-
tinguish between benign and malignant lesions. Radiology.
1995;196:123–134.
29. Taylor KJ, Ramos I, Carter D, Morse SS, Snower D, Fortune
K. Correlation of Doppler US tumor signals with neovascular
morphologic features. Radiology. 1988;166(1 Pt 1):57–62.
30. Donegan WL. Evaluation of a palpable breast mass. N Engl J
Med. 1992;327:937–942.
31. Sickles EA, Parker SH. Appropriate role of core breast biopsy
in the management of probably benign lesions. Radiology.
1993;188:315.
32. Lindfors KK, O’Connor J, Acredolo CR, Liston SE. Short-
interval follow-up mammography versus immediate core
biopsy of benign breast lesions: Assessment of patient stress.
AJR Am J Roentgenol. 1998;171:55–58.
33. Sheehan J, Sherman KA, Lam T, Boyages J. Association of
information satisfaction, psychological distress and monitor-
ing coping style with post-decision regret following breast
reconstruction. Psychooncology. 2007;16:342–351.
34. Chala LF, de Barros N, de Camargo Moraes P, et al. Fat
necrosis of the breast: Mammographic, sonographic, com-
puted tomography, and magnetic resonance imaging find-
ings. Curr Probl Diagn Radiol. 2004;33:106–126.
35. Mandrekas AD, Assimakopoulos GI, Mastorakos DP, Pantza-
lis K. Fat necrosis following breast reduction. Br J Plast Surg.
1994;47:560–562.
36. Handel N, Jensen JA, Black Q, Waisman JR, Silverstein MJ.
The fate of breast implants: A critical analysis of complica-
tions and outcomes. Plast Reconstr Surg. 1995;96:1521–1533.
37. Kroll SS. Fat necrosis in free transverse rectus abdominis
myocutaneous and deep inferior epigastric perforator flaps.
Plast Reconstr Surg. 2000;106:576–583.
38. Missana MC, Laurent I, Barreau L, Balleyguier C. Autologous
fat transfer in reconstructive breast surgery: Indications,
technique and results. Eur J Surg Oncol. 2007;33:685–690.
39. Losken A, Pinell XA, Sikoro K, Yezhelyev MV, Anderson E,
Carlson GW. Autologous fat grafting in secondary breast
reconstruction. Ann Plast Surg. 2011;66:518–522.
40. Rohrich RJ, Sorokin ES, Brown SA. In search of improved fat
transfer viability: A quantitative analysis of the role of centrif-
ugation and harvest site. Plast Reconstr Surg. 2004;113:391–
395; discussion 396–397.

12. 772
Plastic and Reconstructive Surgery • October 2012
41. Mayor S. National audit of care after mastectomy points to need
for better information on reconstruction. BMJ. 2010;340:c3506.
42. Reaby LL. Reasons why women who have mastectomy decide
to have or not to have breast reconstruction. Plast Reconstr
Surg. 1998;101:1810–1818.
43. Lee CN, Hultman CS, Sepucha K. What are patients goals
and concerns about breast reconstruction after mastectomy?
Ann Plast Surg. 2010;64:567–569.
44. American Society of Plastic Surgeons. Fat transfer/fat graft
and fat injection ASPS guiding principles. Avail­­able at:
http://www.plasticsurgery.org/Documents/medical-profes-
sionals/health-policy/guiding-principles/ASPS-Fat-Trans-
fer-Graft-Guiding-Principles.pdf. Accessed August 2, 2011.
45. O’Connor AM, Rostom A, Fiset V, et al. Decision aids for
patients facing health treatment or screening decisions:
­Systematic review. BMJ. 1999;319:731–734.
Article Collections – Cosmetic Breast
The Cosmetic Breast article collection on PRS-
Journal.com represents a pre-made article
search on relevant topics in Cosmetic Breast,
as evaluated and chosen by the PRS Editorial
Board and the PRS Section Editors. The col-
lection contains some of the most educational
and very best articles published in Plastic and
Reconstructive Surgery over the last 10 years. This
is just one of 15 articles in the collection.
See more at www.PRSJournal.com