Keywords
fluorescence imaging - intraoperative - ambient light compatible - breast reconstruction
- nipple necrosis
Introduction
Nipple necrosis unfortunately complicates 3 to 37% of nipple-sparing mastectomies
(NSMs) making preoperative discussion critical as postoperative development of this
complication can result in emotional stress, prolonged surgical management, and increased
cost.[1] Known surgical risk factors for nipple–areolar complex (NAC) necrosis include higher
mastectomy weight, thinner mastectomy skin flaps, and a periareolar incision.[2]
[3] However, compared to women who undergo skin-sparing mastectomies, NSMs provide significant
psychological benefits such as improved postoperative body image and breast satisfaction.[4]
Along with clinical judgment, intraoperative assessment of nipple perfusion during
NSMs using fluorescence-guided surgery (FGS) indocyanine green (ICG) angiography can
be performed, with prior studies demonstrating a strong correlation between ICG findings
to areas of the mastectomy skin that ultimately develop skin necrosis.[1]
[5] Implementation of ICG angiography in breast cancer surgery has also resulted in
a decrease in mastectomy skin necrosis rates, with Duggal et al demonstrating a reduction
in necrosis requiring reoperations from 14.1 to 5.9% (p = 0.009).[6] However, some currently available ICG systems, when operated in the presence of
ambient light, experience light contamination, thereby compromising image quality
and detection sensitivity of the near-infrared signal from the dye.[7] Notably, the U.S. Food and Drug Administration recently cleared the OnLume Avata™
System (OnLume Surgical, Madison, WI). The FGS technology of the Avata System utilizes
transient lighting to allow for simultaneous capture and display of color and monochrome
fluorescence images in ambient light without impact on image quality. This allows
for high-definition resolution, real-time overlay visualization and can be utilized
at a variety of working distances thanks to its electronic zoom and focus features.
The Avata System's large range of field of views, variable working distance, and adjustable
camera angle allows it to easily adapt to different regions of interest and integrate
into a variety of surgical scenarios. In this case report, we describe the use of
the OnLume Avata System to predict nipple necrosis and help guide surgical and postoperative
decision-making in two patients undergoing bilateral breast reconstruction with NSM.
Cases
Patient 1
The first patient was a 48-year-old, nonsmoking woman with ductal carcinoma in situ
in bilateral breasts. On preoperative exam, she was noted to have a body mass index
(BMI) of 29.6, excess subcutaneous abdominal tissue, and size B cup breasts with minimal
ptosis. Given her cancer characteristics and physical exam findings, we determined
that she was a candidate for NSMs. The patient underwent bilateral NSM with immediate
breast reconstruction using autologous deep-inferior epigastric artery perforator
(DIEP) flaps performed by the senior author (S.O.P.).
Both NSMs were performed with a periareolar incision. Her mastectomy weights were
552 and 534 g from the right and left breast, respectively. Before the DIEP flap inset,
there was a clinical concern for hypoperfusion of the nipples bilaterally. Therefore,
vascular perfusion imaged in real time using the OnLume Avata System, which demonstrated
global hypoperfusion of her mastectomy skin flaps and dark, poorly perfused areas
of the NACs ([Fig. 1]). Given these findings, the senior author elected to bank a skin paddle on the DIEP
flap below the NACs. For the first two postoperative days, to help improve vascular
perfusion, nitropaste was applied to the bilateral mastectomy skin flaps and the NACs.
Despite this intervention, she continued to have progressive worsening of NAC ischemia
in both nipples, ultimately developing full-thickness necrosis of the entire NAC in
both breasts by postoperative day 3. Prior to discharge, the senior author (S.O.P.)
reviewed the intraoperative ICG findings, confirmed the irreversible necrosis of the
NAC, and discussed the need for surgery within the coming weeks. Approximately 3 weeks
after her initial surgery, the patient underwent surgical excision of necrotic nipples
and inset of banked DIEP skin flap. Five months later, the patient had nipple reconstruction
for both breasts to complete her reconstructive course.
Fig. 1 Patient 1 bilateral mastectomy skin flap fluorescence (top left: blue false-color
fluorescence on color image, top right: monochrome) and clinical color (bottom left)
images captured intraoperatively demonstrating focal areas of ischemia to the nipple–areolar
complex with global hypoperfusion of her remaining mastectomy skin flaps. Post hoc
surgeon annotations: green, “likely well perfused;” blue, “questionably perfused;”
red, “underperfused/ischemic.”
Patient 2
The second patient was a 40-year-old, nonsmoking woman who presented for a discussion
regarding breast reconstruction options with right breast lobular carcinoma in situ.
The cancer was identified within her breast tissue specimens after undergoing a breast
reduction 3 months earlier. On preoperative exam, she was noted to have a BMI of 34.0,
excess subcutaneous abdominal tissue, and size C cup breasts. The patient strongly
preferred to attempt NSMs despite her larger breast size. After a thorough discussion
regarding the risks and benefits, both she and her surgeons agreed to attempt NSMs
with immediate DIEP flaps, given that her prior breast reduction could result in a
delay phenomenon for her skin. The same reconstructive team performed the DIEP flaps.
The NSMs were performed with an inframammary incision bilaterally. Her mastectomy
weights were 836 and 808 g. There was clinical concern for bilateral NAC ischemia;
therefore, an assessment of vascular perfusion was performed utilizing the OnLume
Avata System. Her imaging findings showed bilateral hypoperfusion of the NAC and mastectomy
skin immediately below the NAC. However, the remainder of the mastectomy skin flaps
were well-perfused ([Fig. 2]). To preserve the NAC on both breasts, the senior author elected to remove the lower,
hypoperfused mastectomy skin and inset the DIEP flap with a skin paddle that could
be used for nipple reconstruction. Postoperatively, nitropaste was ordered for the
bilateral NACs and continued until postoperative day 3. The patient had progressive
resolution of hypoperfusion with normal capillary refill of both areolas noted on
postoperative day 2. On the discharge date, necrosis was localized to the tips of
both nipples, and no additional surgical interventions for the nipples were required.
Fig. 2 Patient 2 bilateral mastectomy skin flap fluorescence (top left: blue false-color
fluorescence on color image, top right: monochrome) images captured intraoperatively
demonstrating ischemia to the nipple–areolar complex bilaterally but robust perfusion
of the remaining mastectomy skin flaps. Post hoc surgeon annotations: green, “likely
well perfused;” blue, “questionably perfused;” red: “underperfused/ischemic.”
Comparison of OnLume Findings
Institutional Review Board (IRB) approval and subject consent was obtained prior to
patient data collection (ID: 2020-0906). Informed consent was obtained from both patients
to utilize clinical photos and information for this manuscript. Post hoc analysis
of OnLume Avata images for both patients, obtained at the point of maximal perfusion,
was performed by a plastic surgeon not involved in the patient's initial operation
(E.C.S.) to allow for unbiased analysis of the images. This surgeon had significant
prior clinical experience with reviewing fluorescence images for vascular perfusion.
Areas of the mastectomy skin and NAC were subjectively assessed and outlined as either
“likely well perfused,” “questionably perfused,” or “underperfused/ischemic.” Relative
fluorescence intensity values were normalized between images to allow for comparison
and correlate to the degree of vascular perfusion within the tissues. Patient 1, who ultimately developed full-thickness NAC necrosis in both breasts, demonstrated
an overall lower mean relative intensity of their mastectomy skin flaps (left breast:
0.41, right breast: 0.40), compared to Patient 2, whose nipples ultimately survived (left breast: 0.57, right breast: 0.55; [Fig. 3]). Furthermore, the majority of Patient 1's mastectomy skin flap was identified to be either “questionably perfused” or “underperfused/ischemic”
despite portions of her NAC being labeled as “likely well perfused” ([Table 1]).
Table 1
Relative areas of vascular perfusion
|
|
NAC necrosis?
|
Relative area (%)
|
|
|
Likely well perfused
|
Questionably perfused
|
Underperfused/Ischemic
|
|
Patient 1
|
Left breast
|
Yes
|
5.4
|
92.8
|
1.8
|
|
Right breast
|
Yes
|
3.1
|
96.3
|
0.6
|
|
Patient 2
|
Left breast
|
No
|
83.6
|
5.6
|
10.8
|
|
Right breast
|
No
|
92.5
|
0.0
|
7.5
|
Abbreviation: NAC, nipple–areolar complex.
Fig. 3 Comparison of relative fluorescence intensity values between patients. Patient 1
ultimately developed full-thickness nipple–areolar complex necrosis requiring additional
surgical intervention, while Patient 2 had recovery of perfusion.
Discussion
Nipple loss in patients undergoing NSM is an unfortunate complication, given that
it is devastating to the patient and decreases their quality of life after reconstruction.
Methods to predict NAC necrosis are critical to help guide intraoperative decision-making
and postoperative care. In this report, we describe the use of a high-resolution,
ambient light-compatible FGS technology to evaluate two patients with intraoperative
nipple hypoperfusion and demonstrate differences in their mastectomy flap vascularity.
Interpretation of these images and close postoperative monitoring permitted informed
patient counseling when deciding if additional surgical management was required.
When reviewing the OnLume Avata fluorescence images of both patients' mastectomy skin
flaps, there is a clear difference in the relative intensities and areas of adequate
perfusion. Despite Patient 2's NAC appearing to have a greater area of hypoperfusion, the majority of her mastectomy
skin flaps were well-perfused, which may have contributed to the nipples' ultimate
survival. Furthermore, the senior author made the intraoperative decision to remove
the small area of hypoperfused mastectomy skin, just inferior to the nipple, with
inset of the DIEP flap with a skin paddle. This is in contrast to Patient 1, who not only had areas of ischemia identified on the NAC but was also noted to have
hypoperfused mastectomy skin flaps. The senior author ultimately made the intraoperative
decision to bank a large area of DIEP skin in order to perform nipple reconstruction
at a later date. Sadly, this was eventually needed, prolonging the patient's reconstructive
course.
ICG imaging, combined with clinical assessment, results in high sensitivity in predicting
areas of nipple necrosis.[1] Its routine use has been shown to decrease the incidence of complications related
to mastectomy skin necrosis, as described by Duggal et al.[6] However, unlike traditional FGS technology, the OnLume Avata System allows for the
simultaneous assessment of the clinical anatomy and vascularity, given that both fluorescence
and color images can be displayed at video rate in real time with the operating room
lights on, therefore potentially increasing its sensitivity to identify areas of underperfusion.
Furthermore, the Avata System provides a repeatable and reproducible fluorescence
quantitation, as described by Seets et al, whereas the current SPY-Elite and SPY-PHI
systems cannot.[8]
[9] Additionally, there are potential cost savings with implementation of the OnLume
Avata System. Despite their limitations, use of previous generation fluorescence systems
to assess vascular perfusion in autologous breast reconstruction operations has resulted
in a decreased incidence of revision operations, ultimately saving patients an average
of $1,400 to $9,550.[5]
[10] Given that the OnLume System provides even greater reliability, the potential cost
savings could be theoretically larger.
Lastly, individual patient characteristics and surgical techniques can influence the
incidence of nipple necrosis in patients undergoing NSMs. Patient characteristics
such as smoking history, prior radiation, increasing BMI, and large ptotic breasts
have demonstrated an increased risk.[11]
[12] Given that radiotherapy is recommended in more advanced cancers or those with high-grade
features, that may also result in a greater risk of nipple necrosis. However, this
was not the case for either of our patients who were diagnosed with low-grade in situ
cancers. Regarding surgical risks, an extensive systematic review by Lee and Mun demonstrated
that incisions involving the NAC, specifically periareolar incisions and mastectomies
performed with diathermy, resulted in higher rates of nipple loss.[13] Additional studies have also noted greater mastectomy specimen weight and thinner
mastectomy skin flaps also increase the risk of nipple necrosis.[14]
[15] Regarding our patients, despite Patient 2 having larger breasts, a higher BMI, and greater mastectomy tissue resection weight,
the use of an inframammary incision may have been a protective factor in her NAC ultimately
recovering its perfusion. Furthermore, her history of prior breast reduction also
may have resulted in the delay phenomenon with enhanced vascularity to the skin flaps.
We recognize that tissue perfusion is dynamic, and therefore, a static measurement
may not be entirely predictive of necrosis. Therefore, as with previous fluorescence
technologies, FGS imaging should be used as a supplement to clinical assessment. Unlike
conventional “lights-off” intraoperative angiography systems, the OnLume Avata imaging
system was easily integrated into the surgical workflow to permit the detection of
nipple ischemia and assisted with real-time intraoperative decisions in the setting
of ambient light.
