ME69CH09_Mehrara ARI 28 August 2017 11:49

Annual Review of Medicine

Lymphedema: Pathogenesis
and Novel Therapies
Joseph H. Dayan,1,2 Catherine L. Ly,1
Raghu P. Kataru,1 and Babak J. Mehrara1,2
Access provided by Griffith University on 09/08/17. For personal use only.

1
Plastic and Reconstructive Surgery Service, Department of Surgery, Memorial Sloan Kettering
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

Cancer Center, New York, NY 10065; email: dayanj@mskcc.org, lyc@mskcc.org,

katarur@mskcc.org, mehrarab@mskcc.org
2
Plastic and Reconstructive Surgery Service, Department of Surgery, Weill Cornell Medical
College, New York, NY 10065

Annu. Rev. Med. 2018. 69:9.1–9.14 Keywords

The Annual Review of Medicine is online at CD4+ cells, complete decongestive therapy, lymphovenous bypass, lymph
med.annualreviews.org
node transfer
https://doi.org/10.1146/annurev-med-060116-
022900 Abstract
Copyright  c 2018 by Annual Reviews. Lymphedema affects up to 1 in 6 patients who undergo treatment for a
All rights reserved
solid tumor in the United States. Its prevalence has increased as more ef-
fective oncologic therapies have improved patient survival, but there re-
mains no definitive cure. Recent research has elucidated new details in the
pathogenesis of the disease and has demonstrated that it is fundamentally
an immunologic progress that ultimately results in inflammation, fibroadi-
pose deposition, impaired lymphangiogenesis, and dysfunctional lymphatic
pumping. These findings have allowed for the development of novel medical
and surgical therapies that may potentially alter the standard of care for a
disease that has largely been treated by compression. This review seeks to
provide an overview of the emerging therapies and how they can be utilized
for effective management of lymphedema.

9.1

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

INTRODUCTION
Lymphedema is a chronic disease of the lymphatic system characterized by swelling and fibroad-
ipose tissue deposition. Developmental abnormalities of the lymphatic system resulting in poorly
developed or dysfunctional lymphatics are the cause of primary lymphedema. These disorders
may present clinically at birth or, more commonly, in adolescence or later in life. Secondary
lymphedema is more common and results from injury, obstruction, or infection of the lymphatic
system. It is estimated that 300 million patients in developing countries suffer from filariasis, a
disease caused by infection with bloodborne nematodes. Patients with filariasis develop massive
swelling of the limbs and genitals due to direct lymphatic obstruction by the parasites. In con-
trast, lymphadenectomy and radiation for cancer treatment are usually the precipitating factors
for lymphedema in developed countries.
In the United States, breast cancer is the most common cause of lymphedema owing to its
prevalence. It is estimated that 20–50% of patients who undergo complete axillary lymph node
dissection (ALND) for breast cancer go on to develop lymphedema. The variability in rate is due, at
Access provided by Griffith University on 09/08/17. For personal use only.

least in part, to differences in the diagnostic criteria and the length of follow-up (1). The advent of
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

sentinel lymph node biopsy has decreased the need for ALND and reduced the rate of lymphedema
to 5–7%. However, it is likely that breast cancer–related lymphedema (BCRL) will remain a
significant problem in the foreseeable future because of improved survival due to more effective
oncologic therapies and to the increasing incidence of major risk factors such as obesity and
radiation. Lymphedema is also commonly associated with melanoma, sarcoma, and gynecological,
head and neck, and urologic tumors, and it is estimated that as many as 1 in 6 patients who undergo
treatment for a solid tumor go on to develop lymphedema (2). As a result, lymphedema is major
biomedical burden afflicting 5–6 million Americans, a number that exceeds the total number of
those who suffer from Parkinson’s disease, ulcerative colitis, rheumatoid arthritis, systemic lupus
erythematosus, and multiple sclerosis combined. The negative impact on quality of life in patients
with only mild swelling is commonly underappreciated. The permanence of the disease, time
burden of treatment, and likelihood of progression can profoundly alter patients’ lifestyles. Thus,
development of effective therapies for lymphedema is an important clinical goal.

PATHOPHYSIOLOGY OF LYMPHEDEMA

Symptoms and Diagnosis

Patients with lymphedema complain of pain, heaviness, swelling, decreased limb function, and
decreased quality of life. They may develop hyperkeratosis, acanthosis, skin ulcerations, papillo-
matous plaques, and recurrent infections (3). Infections presenting as cellulitis or lymphangitis
commonly require intravenous antibiotics and, in some cases, long-term prophylaxis. Importantly,
infections are often associated with exacerbation of the disease, resulting in more pronounced
symptoms.
Swelling from lymphedema should be differentiated from other disorders that may present in
a similar manner, including deep vein thrombosis, venous insufficiency, congestive heart failure,
recurrent or primary malignancy causing lymphatic obstruction, and acute infections. Venous
insufficiency is particularly important because a major portion of postcapillary fluid transport is
accomplished by the venous system.
Limb circumference or volume measurements are commonly used for diagnosis and for follow-
ing disease progression. Circumference measurements, performed at defined anatomic locations
(e.g., 5 cm above and below the olecranon), are simple to perform, and a difference of 2 cm or
more compared to the contralateral limb is considered diagnostic. However, these measurements

9.2 Dayan et al.

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

can have high inter-rater variability, and their utility is limited in patients with a high body mass
index (a 2-cm difference is much more significant in a thin patient than in someone who is mor-
bidly obese). Volume measurements are considered superior to circumference measurements and
can be performed with water displacement or limb scanners such as the perometer. A practical
and inexpensive alternative involves measuring circumference every 4 cm and calculating volume
based on the truncated cone formula previously described by Brorson & Hoijer (4). Most authors
consider volume differences of 200 cc or 10% the threshold for lymphedema diagnosis (5).
Various technologies have been utilized in the diagnosis of lymphedema. Bioimpedance can be
used for early diagnosis, as the rate of electrical current transmission through the lymphedematous
limb is more rapid than in the normal limb (6). Lymphoscintigraphy is the most commonly
used radiologic method of diagnosis; it measures the uptake of 99m Tc by draining lymph node
basins after peripheral injection (7). Dermal backflow resulting from lymphatic obstruction and
accumulation of the tracer in the skin is consistent with lymphedema. However, this test has a
relatively low sensitivity (0.61) in early-stage disease, thus limiting its utility (8). Newer techniques
such as magnetic resonance lymphangiography have uncovered lymphatic architecture and disease
Access provided by Griffith University on 09/08/17. For personal use only.
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

with greater fidelity but are not yet widely available (9).
Attempts at analyzing the physiologic function of the lymphatics have led to the development
of indocyanine green (ICG) near-infrared lymphangiography. In contrast to previous methods
used for lymphangiography, this approach is nontoxic and enables visualization of the capillary
and superficial collecting lymphatic vessels following intradermal injection of the ICG, which is
preferentially taken up by the lymphatics. Although this is an off-label use of ICG, which is cur-
rently approved by the US Food and Drug Administration for intravenous injection, recent reports
suggest that ICG lymphangiography has a high degree of sensitivity and specificity for diagnosing
lymphedema. Pathologic changes in the lymphatic system consisting of dermal backflow, punc-
tate lymphatic vessel leaks, and hyperplastic and abnormal lymphatic vessels are often observed
in patients who have symptoms of lymphedema but do not meet diagnostic criteria based on limb
volume measurements or even biompedance or lymphoscintigraphy (8). Isolated clinical reports
and experimental studies have also shown that this approach can be used to estimate lymphatic
pumping rate and velocity of lymphatic flow, but this application is not widely adopted owing to
technical issues (10). Future studies should focus on more sensitive and quantitative methods that
can be used to diagnose lymphedema and, more importantly, analyze responsiveness to treatments.

Staging
Although various staging systems have been proposed, the most widely used is the International
Society of Lymphology (ISL) staging system. This scheme is based on clinical features of the
disease and classifies patients according to limb swelling and presence or absence of pitting edema
(Table 1) (11).

Table 1 International Society of Lymphology staging of lymphedema

Stage Description
0 (Latent lymphedema) Patients who have had lymphatic injury and have symptoms of lymphedema but no measurable
changes in limb volume/circumference
1 (Spontaneously reversible) Measurable limb swelling and pitting edema that resolves with compression
2 (Not spontaneously reversible) Limb swelling due to fibroadipose deposition that does not resolve with compression
3 (Lymphostatic elephantiasis) End-stage disease with severe swelling and skin changes

www.annualreviews.org • Pathogenesis and Treatment of Lymphedema 9.3

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

Natural History
Contrary to popular belief, few patients are diagnosed with BCRL immediately following ALND.
In most cases, BCRL develops in a delayed manner—on average, eight months after surgery,
although some patients develop the disease years later (12). Nearly 80% of patients who develop
BCRL do so within the first three years following surgery. In contrast, development of lower-
extremity lymphedema after pelvic or inguinal lymph node dissection is more rapid, occurring in
many cases within 3–6 months of surgery (13, 14).
Once lymphedema develops, the severity and rate of progression are variable. Some patients
have low-grade swelling limited to a localized portion of the limb (e.g., forearm, hand, or even a
single digit), whereas others have nearly uniform swelling of the entire limb. In some patients, the
disease progresses rapidly, with severe swelling developing months after the initial diagnosis. In
other cases, lymphedema progresses slowly, smoldering along over years, even despite noncom-
pliance with therapy.
The variability in the timing and rate of progression of disease, together with the fact that
Access provided by Griffith University on 09/08/17. For personal use only.

lymphedema develops only in a subset of patients who undergo lymphadenectomy, suggests that
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

lymphatic injury is simply an initiating event in the pathology of lymphedema. This pattern
of disease development suggests that additional pathologic steps are necessary for lymphedema
to become manifest once the capacity of the lymphatic system to transport interstitial fluid is
overcome. Understanding the cellular mechanisms that regulate development of these “second
hit” phenomena is therefore important to identifying novel treatment options for this morbid and
common disease.

CELLULAR MECHANISMS OF LYMPHEDEMA

The cellular mechanisms that translate lymphatic injury to the development of lymphedema com-
prise an important area of research. Understanding how lymphatic injury and accumulation of
interstitial fluid impair lymphatic function and promote fibroadipose tissue deposition may enable
development of targeted therapies that can be used to prevent lymphedema in high-risk patients
or reverse its pathology once it is diagnosed.

Accumulation of Interstitial Fluid

The initial presentation of lymphedema is accumulation of protein-rich interstitial fluid in the
subcutaneous and subfascial tissues resulting in symptomatic changes (e.g., heaviness or tightness)
and pitting edema. Chronic interstitial fluid stasis promotes activation of chronic inflammatory
pathways and adipose deposition. These changes further decrease lymphatic function, thereby
activating a feed-forward mechanism and disease progression. As a result, control of interstitial
fluid accumulation by early intervention is an important step in lymphedema management.

Chronic Inflammation, Fibrosis, and Regulation of Lymphangiogenesis

Chronic inflammation in the dermis and subcutaneous tissues is a histological hallmark of lym-
phedema. Comparison of biopsy specimens obtained from the normal and lymphedematous limbs
of patients with unilateral lymphedema as well as animal models have shown that CD4+ cells
comprise nearly 70% of all inflammatory cells in lymphedematous tissues, and that the number
of CD4+ cells is positively correlated with disease severity (15). Animals lacking T cells in gen-
eral or CD4+ cells in particular fail to develop lymphedema after injury. Similarly, depletion of

9.4 Dayan et al.

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

CD4+ cells (but not CD8+ cells, macrophages, or B cells) with neutralizing antibodies or with
topical tacrolimus to prevent T cell proliferation and differentiation potently prevents develop-
ment of the disease (15, 16). Infiltrating CD4+ cells display a T helper 2 (Th2)-biased mixed
Th1/Th2 response, and inhibition of Th2 differentiation with neutralizing antibodies that block
interleukin-4 (IL-4) or IL-13 is also highly effective in preventing and treating lymphedema in
preclinical models (17). This approach has been translated into an ongoing early-stage clinical
trial.
Th2 inflammatory responses cause progressive lymphatic dysfunction and eventual develop-
ment of lymphedema by exerting a variety of effects, including progressive fibrosis, inhibition of
collateral lymphatic vessel formation, and inhibition of lymphatic pumping function. Anatomic
clinical research and animal studies suggest that lymphedema is end-organ failure of the lymphatic
system in which functional parenchyma is progressively replaced by scar tissue. These studies have
shown that lymphatic vessels become progressively encased in and replaced by fibrous tissues, re-
sulting in loss of functional initial lymphatics and luminal obliteration of collecting vessels (8).
Thus, similar to end-organ failure in cirrhosis, pulmonary fibrosis, scleroderma, and other fibro-
Access provided by Griffith University on 09/08/17. For personal use only.
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

proliferative syndromes, symptoms occur in lymphedema when a critical threshold of function is

breached. This hypothesis is supported by the fact that fibrosis in lymphedema, like other fibro-
proliferative disorders, is regulated by chronic Th2-biased inflammatory responses and expres-
sion of profibrotic growth factors including IL-4, IL-13, and transforming growth factor beta-1
(TGF-β1) (18).
This pathology may also explain the variable nature of the disease. Fibrosis in other fibropro-
liferative diseases is modulated by genetic and environmental factors, thus providing a rationale
for the known risk factors of the disease, such as radiation, obesity, and infections, all of which
directly or indirectly increase fibrosis.
T cell-derived cytokines including IL-4, IL-13, interferon gamma (IFN-γ), and TGF-β1 are
also key negative regulators of lymphangiogenesis. They decrease collateral lymphatic vessel for-
mation by hindering lymphatic endothelial cell (LEC) proliferation, as well as tubule formation,
migration, and function. Inhibition of anti-lymphangiogenic cytokines to promote collateral lym-
phatic vessel formation is an active area of research and has several advantages over approaches that
rely on delivery of vascular endothelial growth factor-C (VEGF-C) or other pro-lymphangiogenic
cytokines. A key advantage of this approach is that inhibition of Th2 cytokines does not increase
the risk of cancer metastasis or tumor growth. In fact, some studies have suggested that anti-Th2
therapies may play an adjunctive role in the treatment of solid malignancies because these cy-
tokines have immunosuppressive effects that decrease tumor-specific immune responses (19). In
contrast, lymphangiogenic growth factors increase tumor metastasis and malignant behavior in a
variety of solid tumors by regulating infiltration and migration of tumor-associated macrophages
in addition to other effects on the tumor microenvironment (20). Furthermore, T cell-derived
cytokines decrease responsiveness of LECs to lymphangiogenic growth factors, thus making the
delivery of VEGF-C ineffective even with supraphysiologic doses (21). This concept is supported
by the fact that the expression of VEGF-C is markedly increased in lymphedematous tissues,
suggesting that formation of collateral lymphatics in this scenario is actively inhibited (22).
A final mechanism by which chronic inflammation impairs lymphatic function is by regulating
collecting lymphatic vessel pumping. Interstitial fluid enters capillary lymphatics and is trans-
ported to precollectors and collecting lymphatics through a successive series of larger vessels that
use unidirectional valves to regulate fluid flow. Collecting lymphatics propagate interstitial fluid
to regional lymph nodes and eventually back to the circulation as a result of intrinsic pump-
ing mechanisms relying on perilymphatic smooth muscle cells, and, to a lesser extent, passive
forces generated by extrinsic skeletal muscle action. Intrinsic lymphatic pumping is regulated by

www.annualreviews.org • Pathogenesis and Treatment of Lymphedema 9.5

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

gradients of nitric oxide (NO) produced by LEC-derived nitric oxide synthase, which enables
rhythmic vessel dilatation and contraction. After lymphatic injury, inflammatory cells accumulate
around lymphatic channels and upregulate the expression of induced nitric oxide synthase (iNOS).
High NO levels produced by extrinsic iNOS thus disrupt the intrinsic NO gradient, thereby de-
creasing collecting lymphatic contractility and transport potential. These molecular mechanisms
may account for the finding that patients with lymphedema have decreased lymphatic pumping
pressures.
In summary, an increasing body of evidence suggests that anti-inflammatory and antifibrotic
approaches aimed at preventing Th2 responses may be a novel means of preventing or treating
lymphedema.

MEDICAL TREATMENTS OF LYMPHEDEMA

Weight Loss and Exercise

Access provided by Griffith University on 09/08/17. For personal use only.
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

Obesity and postoperative weight gain are major risk factors for the development of secondary
lymphedema, increasing the risk of disease development by three- to fivefold (23, 24). This finding
is likely related to a bidirectional interaction between the lymphatic system and adipose deposition
such that, on the one hand, lymphatic defects promote adipose deposition, and on the other,
inflamed adipose tissues impair lymphatic function. Support for this hypothesis is derived from the
natural history of lymphedema, which is characterized by progressive adipose deposition locally in
the lymphedematous limb, suggesting that lymphedema is a form of localized or regional obesity
promoting adipose deposition preferentially in the affected tissues. This concept is supported
by experimental studies using mice with developmental abnormalities of the lymphatic system,
which have demonstrated that lymphatic leakiness results in adult-onset obesity due to increased
adipocyte proliferation and differentiation (25). Other studies comparing obese mice with lean
controls, as well as lean mice versus mice with high-fat-diet-induced obesity, have shown that
obesity independently decreases macromolecule clearance by increasing lymphatic vessel leakiness
and decreasing collecting lymphatic pumping (26).
Not surprisingly, weight loss is a highly effective method of treating lymphedema. Several
randomized clinical trials have shown significant reductions in upper-extremity volumes in patients
who dieted for 12 weeks and lost weight as compared to matched controls who were only given
nutritional advice (27). Similarly, recent monitored exercise studies have shown that both aerobic
exercise and resistance training can decrease the severity of lymphedema and that these effects may
be at least partly due to changes in body composition (28). Experimental evidence also suggests that
aerobic exercise may increase lymphatic function by decreasing subcutaneous tissue inflammation,
thereby minimizing lymphatic leakiness and pumping abnormalities. Thus, nutritional and exercise
counseling is important for patients with lymphedema as well as those who are at risk for developing
the disease.

Decongestive Therapy, Compression, and Intermittent Pneumatic Pumps

Complete decongestive therapy (CDT) is the mainstay of lymphedema management in most
patients with secondary lymphedema. CDT uses lymphatic massage and compression to decrease
fluid accumulation in the lymphedematous tissues. Importantly, CDT is palliative in nature,
aiming to reduce symptoms and prevent disease progression rather than cure lymphedema. CDT
comprises multiple interventions, including manual lymphatic drainage (MLD), skin care, com-
pression, and light exercises. MLD, which is a key component, is a form of soft-tissue massage that

9.6 Dayan et al.

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

gently stretches the initial lymphatics and is thought to increase lymphatic contractility. The
therapist or patient performs light strokes in a directional manner to move lymphatic fluid flow
away from the damaged lymphatics to alternative drainage basins. Although MLD in isolation
is not effective in decreasing limb volumes, some studies have shown that the combination of
MLD with compression is more effective in improving the symptoms of lymphedema than is
compression alone (29).
Short-stretch bandages, which are commonly used in the intensive phase of CDT, differ from
the more widely available long-stretch bandages because they are less likely to impinge on skin
folds and provide a more rigid structure for skeletal muscles to press against, thereby increasing in-
terstitial fluid transport through the exogenous pump mechanism. These bandages are commonly
used in conjunction with padding (so-called multilayer compression) and are more effective than
compression garments in decreasing limb volumes. However, these bandages are cumbersome
and may impair quality of life. Nevertheless, compliance with compression is an important factor
because therapeutic success is directly correlated with prescribed treatment.
Intermittent pneumatic compression devices have one or more pneumatic cuffs that sequen-
Access provided by Griffith University on 09/08/17. For personal use only.
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

tially inflate and stimulate the natural pump effect of muscular contraction to promote lymphatic
drainage. Many devices are currently on the market with variable pressures and cycle times. Al-
though there is some debate as to the efficacy of this intervention, recent meta-analyses and
systematic reviews have suggested that intermittent pneumatic compression may be effective for
volume reduction in patients with secondary lymphedema when used alone or combined with
other treatment modalities (30, 31). Unfortunately, these treatments can be expensive and may
not be covered by insurance.

Low-Level Laser Therapy

Low-level laser therapy (LLLT) relies on low-power (or “cold”) lasers to stimulate tissue changes
via nonthermal mechanisms. Recent evidence suggests that these treatments may be moderately
effective in treating patients with secondary lymphedema. A systematic review and meta-analysis
of nine studies in women with BCRL found that LLLT, either alone or in combination with other
modalities, significantly decreased both pain and swelling (32). Patients treated with LLLT and
other supportive modalities had an average of 90 cc volume reduction. Importantly, although the
number of patients in these reports was small (N = 10–64), only one patient developed cellulitis,
indicating that this therapy is safe. These results, combined with experimental studies in mouse
models demonstrating that LLLT has anti-inflammatory and antifibrotic effects, suggest that this
modality may be a useful adjunct for lymphedema treatment (33).

Stem Cell Therapy

The use of bone marrow or adipose tissue-derived mesenchymal stem cells has recently been
reported and requires additional investigation. Hou et al. (34) compared outcomes in patients with
BCRL treated with autologous bone marrow-derived mesenchymal cells injected intramuscularly
in their lymphedematous upper extremity (N = 10) to a control group of patients (N = 35) treated
with CDT. The stem cell-treated patients were not treated with compression or other modalities.
Follow-up at 1, 3, and 12 months after treatment revealed significant volume reductions in both
groups (81% reduction in stem cell-treated patients versus 55% in CDT-treated patients with
one-year follow-up). In addition, patients reported significantly decreased pain and symptoms of
lymphedema. One mitigating factor in the study was the use of granulocyte colony-stimulating
factor in stem cell-treated patients to mobilize and increase the number of CD34+ stem cells in

www.annualreviews.org • Pathogenesis and Treatment of Lymphedema 9.7

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

the bone marrow aspirate. Although the treatment seemed highly effective, the authors have not
provided additional follow-up or more recent studies.
Maldonado et al. (35) used similar methodology in 20 patients with BCRL, who were divided
equally between bone marrow-derived stem cell therapy (without compression) and CDT alone
and followed for 12 weeks. These authors noted significant (though more modest than in Reference
34) reductions in upper-extremity volume, with no significant difference noted between stem cell-
treated patients and those treated with CDT. However, two patients in the stem cell-treated
group developed recurrent cellulitis that required antibiotic therapy. Similarly, Toyserkani and
colleagues (36) documented a case report in which freshly isolated adipose-derived stem cells were
injected in the axilla of a single patient with BCRL; the authors noted improvements in volume
and symptomatology four months after treatment.
Although the cellular mechanisms by which stem cell therapy may be beneficial in patients with
lymphedema remain unknown, in vitro and laboratory studies suggest that mesenchymal stem cells
can differentiate into lymphatic endothelial-like cells in specialized in vitro culture conditions and
can increase drainage of interstitial fluid when injected in vivo (37). Nevertheless, although these
Access provided by Griffith University on 09/08/17. For personal use only.
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

results are encouraging, additional clinical studies are necessary to validate these findings and help
translate them into the clinic.

VEGF-C Therapy
VEGF-C is a critical regulator of lymphangiogenesis and, through vascular endothelial growth
factor receptor 3 (VEGFR3) signaling, regulates a wide range of effects in LECs, including dif-
ferentiation, survival, and migration. Disturbances in the VEGF-C/VEGFR3 signaling cascade
significantly impair lymphatic development and repair. In fact, heterozygous inactivating muta-
tions of VEGFR3 have also been identified clinically and are the cause of Milroy’s disease, an
autosomal dominant form of primary lymphedema.
Many reports have evaluated the potential for exogenously administered VEGF-C to improve
lymphatic function in animal models of lymphatic injury. These studies have, by and large, shown
that local delivery of supraphysiologic doses of VEGF-C or gene therapy with adenoviral vec-
tors in a variety of animal models of primary or secondary lymphedema significantly increases
lymphangiogenesis and decreases swelling (38, 39). However, clinical translation of VEGF-C (or
any other lymphangiogenic therapy) in cancer survivors has been hampered by the fact that these
same growth factors are key regulators of the tumor microenvironment and may increase the risk
of tumor recurrence or metastasis (39).

SURGICAL TREATMENTS OF LYMPHEDEMA

Surgical treatment of lymphedema includes techniques that simply remove excess fibroadipose
tissues (reductive techniques) and physiologic methods in which lymphatic continuity and function
is restored. Both approaches are useful, and selection of the procedure depends primarily on the
stage of disease and the degree of fibrofatty tissue deposition.

Reductive Methods
Reductive techniques, including direct excision and liposuction, are appropriate in patients whose
lymphedema has a predominantly fibrofatty component. Direct excision techniques such as the
Charles or Sistrunk procedures have been described for the treatment of end-stage upper- and
lower-extremity or genital lymphedema (Figure 1a). These operations are still used in extreme

9.8 Dayan et al.

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

a b c d

Figure 1
Surgical therapies for lymphedema. (a) The Charles procedure for lower-extremity lymphedema, in which excess tissue is excised
circumferentially and defects are covered using split-thickness skin grafts. (b) Liposuction for upper-extremity lymphedema.
(c) Lymphovenous bypass procedure for upper-extremity lymphedema, in which the vein is shown in blue and the lymphatic vessel in
Access provided by Griffith University on 09/08/17. For personal use only.

green. (d) Transplantation of inguinal lymph nodes to the axilla (orthotopic transplantation) or forearm (heterotopic transplantation)
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

for upper-extremity lymphedema.

cases and are effective in situations where few other options exist. However, these invasive proce-
dures can result in significant morbidity, including pain, infections, and delayed healing.
The advent of liposuction for cosmetic procedures led some surgeons to apply this technique
to patients with lymphedema. Using small incisions and a thin metal cannula, the surgeon can
aspirate fat while sparing the skin (Figure 1b). This concept is consistent with the pathology of lym-
phedema as “regional obesity.” Liposuction is indicated in patients with more advanced, nonpitting
lymphedema with significant adipose deposition (>600 cc volume differential) (40). Prospective
studies of patients with BCRL have shown that this technique can, on average, decrease excess
volume by 50–100%, and improvements are maintained in the long term if compression garments
are used continuously. A recent systematic review identified four high-quality studies reporting
an average excess volume reduction of 97% (95% CI 86.2–107, I2 = 0.0%) (41). Symptomatic
relief and decreased rates of cellulitis have also been reported. These procedures are generally
safe and can be performed in an outpatient setting. Complications that do occur tend to be minor
and self-limited; these include paresthesia and wound-healing problems. However, liposuction is
not a cure for lymphedema, and fibrofatty tissue deposition recurs rapidly (within three months)
if patients are not compliant with constant use of tight-fitting compression garments.

Physiologic Methods
Physiologic procedures are aimed at treating the fluid component of lymphedema. These can be
broadly divided into two categories: lymphovenous bypass (LVB) procedures, which reroute lym-
phatic drainage to regional veins, and vascularized lymph node transplantation (VLNT), in which
healthy lymph nodes are transferred from an unaffected region of the body to the lymphedema-
tous limb. Both techniques are commonly used with different advantages and disadvantages. Some
surgeons combine the procedures to improve outcomes, and this technique has gained increas-
ing adoption in light of technological improvements and deeper understanding of lymphedema
pathology. The effect of combining LVB and VLNT is an active area of research.

Lymphovenous bypass (LVB). The advent of microsurgery, a technique enabling surgeons to

repair small vessels with fine suture, led to the development of procedures that reroute congested

www.annualreviews.org • Pathogenesis and Treatment of Lymphedema 9.9

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

lymphatic vessels to nearby veins, providing an exit for lymph. These LVB procedures involve
anastomosis of superficial lymphatic collectors to adjacent venules using high magnification and
superfine sutures in a technique called supermicrosurgery (Figure 1c). This approach avoids the
disappointing results reported in the late 1970s, when collecting lymphatics were connected to
large superficial veins, because smaller venules have lower pressure gradients and favor improved
lymphatic drainage. LVB procedures are most effective in patients with early-stage lymphedema
and decrease excess fluid volume in the majority of cases.
This conclusion is supported by a recent prospective study of 100 consecutive patients in which
the authors noted that 96% of patients with upper-extremity lymphedema had symptomatic im-
provement (lighter, softer, less painful upper extremities) and 74% had quantitative reduction
in excess volume (42). At one-year follow-up, patients with stage I/II disease experienced a 61%
decrease in excess volume. In contrast, patients with more severe disease had more modest im-
provements (17% volume decrease), suggesting that patients with early-stage disease are more
likely to respond. Additionally, although there were relatively few patients with lower-extremity
lymphedema in this study, the authors concluded that LVB is less effective for these patients,
Access provided by Griffith University on 09/08/17. For personal use only.
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

as only 57% had symptomatic improvement. In contrast, Campisi et al. (43) reported equally
beneficial outcomes in upper- and lower-extremity lymphedema. In their experience with >2,600
patients treated for upper- and lower-extremity lymphedema over a 40-year period, 84% had
long-term improvement in excess limb volume, >86% with early-stage disease no longer needed
compression garments, and >90% experienced decreased incidence of cellulitis. However, the au-
thors used a variety of techniques, and follow-up and outcome measures were somewhat variable.
Additionally, studies reporting on symptomatic relief in patients undergoing LVB have relied on
nonvalidated questionnaires, so it is difficult to critically analyze these reports. Finally, limb volume
measurements can be influenced by postoperative compression and therapy. Future prospective
outcome studies with improved standardization and validated lymphedema questionnaires will
facilitate comparison between groups.
Although the benefits of LVB continue to be clarified, these procedures appear to be safe and
well-tolerated (44, 45). Most surgeons perform these operations as an outpatient procedure and
reported complications are generally minor and self-limited.
LVB procedures are also being used prophylactically in patients undergoing lymph node dis-
section, and outcomes are promising. In the largest series to date, Boccardo et al. (46) reported
four-year outcomes on 74 patients treated with prophylactic LVB, a procedure they termed
LYMPHA (“LYmphatic Microsurgical Preventive Healing Approach”). The authors used vol-
umetric analysis and lymphoscintigraphy to prospectively follow patients after ALND and
LYMPHA. Lymphedema developed in only 4% of patients (three individuals), in contrast to
historical rates of 13–65%. Additionally, lymphoscintigraphy suggested that the lymphovenous
anastomoses were patent as long as four years postoperatively. In a similar though smaller study,
Feldman and colleagues (47) compared 27 patients who underwent ALND and LYMPHA to 10
patients who underwent ALND without LYMPHA for technical reasons. Although the follow-
up time was relatively short (mean six months), the authors noted lymphedema development in
12.5% of patients who underwent successful LYMPHA versus 50% of patients in whom LYMPHA
could not be performed. More recently, LYMPHA procedures have also been used in an effort
to prevent development of lower-extremity lymphedema in patients treated with inguinofemoral
lymphadenectomy (48). These studies have relatively short follow-up and small patient numbers
(27 patients in the largest series), but the results are encouraging, with lymphedema development
in only one patient (3%).

9.10 Dayan et al.

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

Vascularized lymph node transplantation (VLNT). VLNT is a procedure in which lymph

nodes are harvested from an unaffected region and transferred to the lymphedematous area using
microsurgical techniques to reconnect the blood supply (Figure 1d). The afferent and efferent
lymphatic vessels are not reconnected but rely on spontaneous regeneration through lymphangio-
genesis. Lymph nodes may be transplanted to the region where lymph node dissection was per-
formed (orthotopic transplantation) or to a distal region of the lymphedematous limb (heterotopic
transplantation). Although the precise mechanism of action of VLNT remains unknown, experi-
mental and clinical studies have shown that transplanted lymph nodes spontaneously re-establish
afferent and efferent connections with shunting of lymph into the venous system through filtering
of interstitial fluid in lymph node high endothelial venules (49, 50).
Various donor areas have been described for harvesting lymph nodes for VLNT, including the
groin, axilla, lateral thoracic area, cervical area, submental area, and omentum. A systematic review
of 24 studies reporting on 271 VLNTs noted that inguinal lymph nodes were most commonly
harvested (68%) (51). However, more recent studies have shown a trend away from the groin
due to donor site complications, including seroma (7.6%), chronic pain (1.6%), and, worryingly,
Access provided by Griffith University on 09/08/17. For personal use only.
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

donor site lymphedema (1.6%). Reverse lymphatic mapping has reduced the risk of iatrogenic
lymphedema by allowing the surgeon to identify and avoid lymph nodes draining the extremity
during VLNT (52). Importantly, the omentum eliminates the risk of donor site lymphedema
completely, and lymph nodes can be harvested using minimally invasive techniques. However,
intra-abdominal surgery has potential risks, such as hernia, small bowel obstruction, and pancre-
atitis in a small number of patients (53). These donor site issues have motivated recent attempts
at engineering lymph node tissue in the laboratory, but clinical application of these techniques is
not likely to be possible in the near future (54).
Analysis of outcomes following VLNT is complicated by the heterogeneity of surgical tech-
niques and outcome measures. Some studies use circumference measurements whereas others rely
on volumetric changes. Additionally, some studies use adjunctive measures such as lymphoscintig-
raphy. Few studies have used validated patient-reported outcome questionnaires for analyzing
outcomes; in general, most have been based on nonvalidated questionnaires. A lack of uniformity
also applies to follow-up times and use of compression bandages after surgery. Nevertheless, a
systematic review of 18 studies reporting on outcomes in 305 patients treated with VLNT re-
ported encouraging results (55). For example, in studies that used circumference measurements
(N = 182 patients), >90% of patients had improvements after surgery. Similarly, in studies that
used volumetric analysis as the outcome measure, >86% of patients improved following surgery.
These findings are supported by a study of vascularized omental transplants in 42 patients not
included in the systematic review (53). In this study, 83% of patients had subjective improvement
in upper-extremity lymphedema symptoms, with a 22% reduction in mean volume. In a systematic
review of seven studies with patient-reported outcomes (N = 105 patients), 93% of respondents
were satisfied or highly satisfied with surgery and the remainder of patients were no better or
worse following surgery (55). More than 50% were able to discontinue the use of compression
garments and in studies (N = 6 with 106 patients) that followed development of cellulitis, there
was a 90% decrease in the incidence of infections following VLNT (55).
Taken together, studies suggest that VLNT is a promising technique that can be used to restore
lymphatic circulation in patients with lymphedema. Like other physiologic procedures, it is most
effective in patients with early-stage disease. Combination of VLNT with other procedures such
as LVB or liposuction may be more effective depending on the patient’s anatomy and degree of
adipose deposition.

www.annualreviews.org • Pathogenesis and Treatment of Lymphedema 9.11

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

CONCLUSIONS
Lymphedema is fundamentally an immunologic phenomenon resulting in mechanical obstruction.
Because it is a morbid disease with an expected increase in prevalence, the development of effective
preventative and therapeutic approaches is of paramount importance. Based on research into the
cellular mechanisms underlying the disease, combining surgical procedures with medical therapies
designed to augment lymphangiogenesis, decrease fibrosis, and enhance lymphatic function may
lead to better outcomes than surgery alone.

DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.
Access provided by Griffith University on 09/08/17. For personal use only.

LITERATURE CITED
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

1. Petrek JA, Heelan MC. 1998. Incidence of breast carcinoma-related lymphedema. Cancer 83:2776–81
2. Cormier JN, Askew RL, Mungovan KS, et al. 2010. Lymphedema beyond breast cancer: a systematic
review and meta-analysis of cancer-related secondary lymphedema. Cancer 116:5138–49
3. Olszewski WL, Jamal S, Manokaran G, et al. 1993. Skin changes in filarial and non-filarial lymphoedema
of the lower extremities. Trop. Med. Parasitol. 44:40–44
4. Brorson H, Hoijer P. 2012. Standardised measurements used to order compression garments can be used
to calculate arm volumes to evaluate lymphoedema treatment. J. Plast. Surg. Hand Surg. 46:410–15
5. Armer JM, Stewart BR. 2005. A comparison of four diagnostic criteria for lymphedema in a post-breast
cancer population. Lymphat. Res. Biol. 3:208–17
6. Ward LC. 2006. Bioelectrical impedance analysis: proven utility in lymphedema risk assessment and
therapeutic monitoring. Lymphat. Res. Biol. 4:51–56
7. Karacavus S, Yilmaz YK, Ekim H. 2015. Clinical significance of lymphoscintigraphy findings in the
evaluation of lower extremity lymphedema. Mol. Imag Radionucl. Ther. 24:80–84
8. Mihara M, Hara H, Hayashi Y, et al. 2012. Pathological steps of cancer-related lymphedema: histological
changes in the collecting lymphatic vessels after lymphadenectomy. PLOS ONE 7:e41126
9. Liu N, Wang C, Sun M. 2005. Noncontrast three-dimensional magnetic resonance imaging vs lym-
phoscintigraphy in the evaluation of lymph circulation disorders: a comparative study. J. Vasc. Surgery
41:69–75
10. Sevick-Muraca EM, Kwon S, Rasmussen JC. 2014. Emerging lymphatic imaging technologies for mouse
and man. J. Clin. Investig. 124:905–14
11. Bernas MJ, Witte CL, Witte MH. 2001. The diagnosis and treatment of peripheral lymphedema: draft
revision of the 1995 Consensus Document of the International Society of Lymphology Executive Com-
mittee for discussion at the September 3–7, 2001, XVIII International Congress of Lymphology in Genoa,
Italy. Lymphology 34:84–91
12. Kwan ML, Darbinian J, Schmitz KH, et al. 2010. Risk factors for lymphedema in a prospective breast
cancer survivorship study: the Pathways Study. Arch. Surg. 145:1055–63
13. Abu-Rustum NR, Alektiar K, Iasonos A, et al. 2006. The incidence of symptomatic lower-extremity
lymphedema following treatment of uterine corpus malignancies: a 12-year experience at Memorial Sloan-
Kettering Cancer Center. Gynecol. Oncol. 103:714–18
14. Ryan M, Stainton MC, Slaytor EK, et al. 2003. Aetiology and prevalence of lower limb lymphoedema
following treatment for gynaecological cancer. Aust. NZ J. Obstet. Gynaecol. 43:148–51
15. Zampell JC, Yan A, Elhadad S, et al. 2012. CD4+ cells regulate fibrosis and lymphangiogenesis in response
to lymphatic fluid stasis. PLOS ONE 7:e49940
16. Gardenier JC, Kataru RP, Hespe GE, et al. 2017. Topical tacrolimus for the treatment of secondary
lymphedema. Nat. Commun. 8:14345

9.12 Dayan et al.

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

17. Avraham T, Zampell JC, Yan A, et al. 2013. Th2 differentiation is necessary for soft tissue fibrosis and
lymphatic dysfunction resulting from lymphedema. FASEB J. 27:1114–26
18. Wynn TA. 2008. Cellular and molecular mechanisms of fibrosis. J. Pathol. 214:199–210
19. Shiao SL, Ganesan AP, Rugo HS, Coussens LM. 2011. Immune microenvironments in solid tumors: new
targets for therapy. Genes Dev. 25:2559–72
20. Skobe M, Hawighorst T, Jackson DG, Prevo R, Janes L, et al. 2001. Induction of tumor lymphangiogenesis
by VEGF-C promotes breast cancer metastasis. Nat. Med. 7:192–98
21. Savetsky IL, Ghanta S, Gardenier JC, et al. 2015. Th2 cytokines inhibit lymphangiogenesis. PLOS ONE
10:e0126908
22. Rutkowski JM, Moya M, Johannes J, et al. 2006. Secondary lymphedema in the mouse tail: lymphatic
hyperplasia, VEGF-C upregulation, and the protective role of MMP-9. Microvasc. Res. 72:161–71
23. McLaughlin SA, Wright MJ, Morris KT, et al. 2008. Prevalence of lymphedema in women with breast
cancer 5 years after sentinel lymph node biopsy or axillary dissection: objective measurements. J. Clin.
Oncol. 26:5213–19
24. Werner RS, McCormick B, Petrek J, et al. 1991. Arm edema in conservatively managed breast cancer:
obesity is a major predictive factor. Radiology 180:177–84
Access provided by Griffith University on 09/08/17. For personal use only.

25. Escobedo N, Proulx ST, Karaman S, et al. 2016. Restoration of lymphatic function rescues obesity in
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

Prox1-haploinsufficient mice. JCI Insight 1:e85096

26. Garcia Nores GD, Cuzzone DA, Albano NJ, et al. 2016. Obesity but not high-fat diet impairs lymphatic
function. Int. J. Obes. 40:1582–90
27. Shaw C, Mortimer P, Judd PA. 2007. A randomized controlled trial of weight reduction as a treatment
for breast cancer-related lymphedema. Cancer 110:1868–74
28. Kwan ML, Cohn JC, Armer JM, et al. 2011. Exercise in patients with lymphedema: a systematic review
of the contemporary literature. J. Cancer Surviv. 5:320–36
29. Lasinski BB. 2013. Complete decongestive therapy for treatment of lymphedema. Semin. Oncol. Nurs.
29:20–27
30. Feldman JL, Stout NL, Wanchai A, et al. 2012. Intermittent pneumatic compression therapy: a systematic
review. Lymphology 45:13–25
31. Shao Y, Qi K, Zhou QH, Zhong DS. 2014. Intermittent pneumatic compression pump for breast cancer-
related lymphedema: a systematic review and meta-analysis of randomized controlled trials. Oncol. Res.
Treat. 37:170–74
32. Smoot B, Chiavola-Larson L, Lee J, et al. 2015. Effect of low-level laser therapy on pain and swelling in
women with breast cancer-related lymphedema: a systematic review and meta-analysis. J. Cancer Surviv.
9:287–304
33. Assis L, Moretti AI, Abrahao TB, et al. 2013. Low-level laser therapy (808 nm) contributes to muscle
regeneration and prevents fibrosis in rat tibialis anterior muscle after cryolesion. Lasers Med. Sci. 28:947–55
34. Hou C, Wu X, Jin X. 2008. Autologous bone marrow stromal cells transplantation for the treatment
of secondary arm lymphedema: a prospective controlled study in patients with breast cancer related
lymphedema. Jpn. J. Clin. Oncol. 38:670–74
35. Maldonado GE, Perez CA, Covarrubias EE, et al. 2011. Autologous stem cells for the treatment of
post-mastectomy lymphedema: a pilot study. Cytotherapy 13:1249–55
36. Toyserkani NM, Jensen CH, Sheikh SP, Sorensen JA. 2016. Cell-assisted lipotransfer using autologous
adipose-derived stromal cells for alleviation of breast cancer-related lymphedema. Stem Cells Transl. Med.
5:857–59
37. Conrad C, Niess H, Huss R, et al. 2009. Multipotent mesenchymal stem cells acquire a lymphendothelial
phenotype and enhance lymphatic regeneration in vivo. Circulation 119:281–89
38. Baker A, Kim H, Semple JL, et al. 2010. Experimental assessment of pro-lymphangiogenic growth factors
in the treatment of post-surgical lymphedema following lymphadenectomy. Breast Cancer Res. 12:R70
39. Hartiala P, Saarikko AM. 2016. Lymphangiogenesis and lymphangiogenic growth factors. J. Reconstr.
Microsurg. 32:10–15
40. Brorson H, Svensson H. 1998. Liposuction combined with controlled compression therapy reduces arm
lymphedema more effectively than controlled compression therapy alone. Plast. Reconstr. Surg. 102:1058–
67

www.annualreviews.org • Pathogenesis and Treatment of Lymphedema 9.13

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)
 ME69CH09_Mehrara ARI 28 August 2017 11:49

41. Carl HM, Walia G, Bello R, et al. 2017. Systematic review of the surgical treatment of extremity lym-
phedema. J. Reconstr. Microsurg. 33:412–25
42. Chang DW, Suami H, Skoracki R. 2013. A prospective analysis of 100 consecutive lymphovenous bypass
cases for treatment of extremity lymphedema. Plast. Reconstr. Surg. 132:1305–14
43. Campisi CC, Ryan M, Boccardo F, Campisi C. 2016. A single-site technique of multiple lymphatic-
venous anastomoses for the treatment of peripheral lymphedema: long-term clinical outcome. J. Reconstr.
Microsurg. 32:42–49
44. Campisi C, Davini D, Bellini C, et al. 2006. Lymphatic microsurgery for the treatment of lymphedema.
Microsurgery 26:65–69
45. Chang DW. 2010. Lymphaticovenular bypass for lymphedema management in breast cancer patients: a
prospective study. Plast. Reconstr. Surg. 126:752–58
46. Boccardo F, Casabona F, De Cian F, et al. 2014. Lymphatic microsurgical preventing healing approach
(LYMPHA) for primary surgical prevention of breast cancer-related lymphedema: over 4 years follow-up.
Microsurgery 34:421–24
47. Feldman S, Bansil H, Ascherman J, et al. 2015. Single institution experience with lymphatic microsurgical
preventive healing approach (LYMPHA) for the primary prevention of lymphedema. Ann. Surg. Oncol.
Access provided by Griffith University on 09/08/17. For personal use only.
Annu. Rev. Med. 2018.69. Downloaded from www.annualreviews.org

22:3296–301
48. Boccardo F, Valenzano M, Costantini S, et al. 2016. LYMPHA technique to prevent secondary lower
limb lymphedema. Ann. Surg. Oncol. 23:3558–63
49. Aschen SZ, Farias-Eisner G, Cuzzone DA, et al. 2014. Lymph node transplantation results in spontaneous
lymphatic reconnection and restoration of lymphatic flow. Plast. Reconstr. Surg. 133:301–10
50. Raju A, Chang DW. 2015. Vascularized lymph node transfer for treatment of lymphedema: a compre-
hensive literature review. Ann. Surg. 261:1013–23
51. Scaglioni MF, Arvanitakis M, Chen YC, et al. 2016. Comprehensive review of vascularized lymph node
transfers for lymphedema: outcomes and complications. Microsurgery doi: 10.1002/micr.30079
52. Dayan JH, Dayan E, Smith ML. 2015. Reverse lymphatic mapping: a new technique for maximizing safety
in vascularized lymph node transfer. Plast. Reconstr. Surg. 135:277–85
53. Nguyen AT, Suami H, Hanasono MM, et al. 2017. Long-term outcomes of the minimally invasive free
vascularized omental lymphatic flap for the treatment of lymphedema. J. Surg. Oncol. 115:84–89
54. Cuzzone DA, Albano NJ, Aschen SZ, et al. 2015. Decellularized lymph nodes as scaffolds for tissue
engineered lymph nodes. Lymphat. Res. Biol. 13:186–94
55. Ozturk CN, Ozturk C, Glasgow M, et al. 2016. Free vascularized lymph node transfer for treatment of
lymphedema: a systematic evidence based review. J. Plast. Reconstr. Aesthet. Surg. 69:1234–47

9.14 Dayan et al.

Review in Advance first posted on

September 6, 2017. (Changes may
still occur before final publication.)