Articles in PresS. Am J Physiol Heart Circ Physiol (August 31, 2012). doi:10.1152/ajpheart.00522.2012
1
A Chronic and Latent Lymphatic Insufficiency
2
Follows Recovery from Acute Lymphedema in
3
the Rat Foreleg
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Uziel Mendez, Emily M. Stroup, Laura L. Lynch, Anna B.
Waller, and Jeremy Goldman
Biomedical Engineering Department, Michigan Technological University,
Houghton, MI
Address for Correspondence:
Jeremy Goldman, Ph.D., Associate Professor
Biomedical Engineering Department
Michigan Technological University
Houghton, MI, 49931 USA
ph: (906) 487-2851
Fax: (906) 487-1717
Email: [email protected]
Number of text pages: 24
Number of tables: 0
Number of figures: 4
28
29
Running Title: Latent Lymphatic Insufficiency Follows Acute Lymphedema
Copyright © 2012 by the American Physiological Society.
30
Abstract
31
Secondary lymphedema in humans is a common consequence of axillary lymph node dissection
32
(ALND) to treat breast cancer. Remarkably, the appearance of lymphedema in the human
33
generally follows a delay of over a year and can be triggered suddenly by an inflammatory insult.
34
However, it remains unclear why the apparently functional lymphatic system is unable to
35
accommodate an inflammatory trigger. In order to provide insight into the mechanism of
36
delayed and rapid human secondary lymphedema initiation, we compared the ability of the
37
ALND-recovered rat foreleg lymphatic system to prevent edema during an inflammatory
38
challenge to that of the uninjured lymphatic system. At 73 days post-surgery, the forelegs of
39
ALND(-) and ALND(+) sensitized rats were exposed to the pro-inflammatory agent oxazolone,
40
which was found to reduce fluid drainage and increase skin thickness in both ALND(-) and
41
ALND(+) forelegs (p<0.05). However, drainage in the ALND-recovered forelegs was more
42
severely impaired than ALND(-) forelegs, as visualized by indocyanine green lymphography in
43
the live animal.
44
inflammation-induced edema with the oxazolone exposure (p<0.05), the peak tissue swelling in
45
the ALND(+) group was significantly greater than that of the ALND(-) forelegs (arm area
46
peaked at ~13.4% vs. ~5.7% swelling, respectively, p<0.005; wrist diameter peaked at 9.7% vs.
47
2.2% swelling, respectively, p<0.005). The findings demonstrate that outward recovery from
48
ALND in the rat foreleg masks an ensuing chronic and latent lymphatic insufficiency, which
49
reduces the ability of the foreleg lymphatic system to prevent edema during an acute
50
inflammatory process.
Although both ALND(+) and ALND(-) forelegs experienced significant
51
52
53
Key words: lymphangiogenesis; lymphatic drainage; axillary lymph node dissection; secondary
54
lymphedema;
inflammation;
oxazolone
55
Introduction
56
Secondary lymphedema in humans is a common consequence of axillary lymph node dissection
57
(ALND) to treat breast cancer, which disconnects lymphatic vessels and excises extracellular
58
matrix from the axilla (6). The intervening surgical space forms into scar tissue, a matrix
59
material that has been demonstrated to hinder lymphatic regeneration and interstitial flows (2,
60
26, 35, 41, 42). An important characteristic of secondary lymphedema is that 25% of patients
61
experience chronic life-long swelling, which generally appears within several years after the
62
initial lymphatic related injury has healed, although there may be a period of acute, transient
63
swelling that appears immediately following the surgery (4, 6, 16, 20, 21).
64
65
Because most studies of lymphedema are conducted on humans with the chronic swelling,
66
relatively little is known about lymphatic function before edema appears. A lack of lymphatic
67
drainage impairment from the operated limb was reported in women assessed pre- and 3 month
68
post-ALND (27). A 7 months post-ALND study showed no deterioration in the muscle and
69
subcutis drainage in the arms of women who ultimately incurred lymphedema relative to women
70
who did not incur lymphedema (33). We have found in the mouse that lymph drainage recovers
71
within several weeks following ALND (19). How lymphatic drainage is maintained in the post-
72
ALND limb despite an inability to achieve sufficient physiological lymphangiogenesis across the
73
wound site has been an open question. In the absence of adequate lymphatic regeneration across
74
the wound site, there is evidence that accumulating lymph may experience compensatory
75
drainage at the expense of higher outflow resistance through a reduced number of surviving local
76
lymphatics, rerouted lymphatics that bypass the obstructive tissue, or lymphovenous
77
communications (20, 34). Each of these may help resolve any acute lymphedema that develops
78
and restore local lymph drainage, yet fail to prevent eventual lymphatic pump failure and chronic
79
lymphedema (34, 37).
80
81
Although there is evidence that an inability of the lymphatic system to adequately regenerate
82
during normal wound repair may predispose the tissue to swell during secondary lymphedema
83
(26, 41), the series of events that causes the initiation of secondary lymphedema in the human
84
has been relatively unexplored. Inflammation has been shown to play an important role in
85
determining the amount of swelling in lymphedema (22, 45, 46) and may be an important
86
stimulus promoting the appearance and perpetuation of the swelling (8). Indeed, the initial
87
appearance of the chronic form of lymphedema in the human can occur rapidly (within a few
88
hours to days) following an acute inflammatory insult (4), although subsequently it may slowly
89
worsen. It is presently unclear why the apparently functional lymphatic system is unable to
90
accommodate an acute inflammatory stimulus. We recently found that the functional recovery of
91
fluid drainage and the resolution of acute lymphedema produced by ALND in the mouse foreleg
92
are not hindered by VEGFR-3 neutralization (which completely prevents any lymphatic
93
regrowth) (19), demonstrating that recovery of fluid drainage and resolution of tissue swelling in
94
acute lymphedema are lymphangiogenesis-independent. This finding raises the possibility that
95
the lymphatic system may remain chronically impaired even though the lymphatic system
96
continues to effectively drain capillary filtrate and prevent edema.
97
98
We hypothesized that a chronic lymphatic impairment may be a natural consequence of the poor
99
physiological lymphatic regeneration that follows ALND. This chronic impairment may create a
100
latent condition of reduced lymphatic reserve capacity (the difference between the normal and
101
maximum lymph flow rate), which may reduce the ability of the tissue to successfully resolve an
102
inflammatory response to foreign antigen. This lymphatic deficiency may be undetectable under
103
normal conditions post-ALND, yet predispose the tissue to chronic swelling following an
104
inflammatory stimulus. In order to test our hypothesis, we compared the ability of the lymphatic
105
system that regenerates in the rat foreleg following ALND to prevent edema during an acute
106
inflammatory challenge to that of the normal foreleg lymphatic system. Such a comparison is
107
necessary to reveal the presence of any latent lymphatic functional deficiency that may follow
108
the physiological lymphatic regeneration process and to clarify the mechanism of secondary
109
lymphedema initiation in the human, which may follow ALND and a coincidental acute
110
inflammatory stimulus.
111
112
113
114
115
Methods
116
Foreleg Rat lymphedema Model
117
Female Sprague Dawley rats at 250-300 grams were used for all experiments. Rats were
118
anesthetized with 2.5% isoflurane mixed with oxygen gas for surgical procedures and were killed
119
at experimental endpoints by CO2 asphyxiation. All protocols were approved by the Animal
120
Care and Use Committee of Michigan Technological University. Axillary lymph nodes were
121
removed from female rats similar to what has been described previously for mice (39). Briefly, a
122
10 mm long surgical incision through the dermis was placed across the axilla on the right side
123
and the axillary lymph nodes were identified and excised along with visible portions of pre and
124
post-nodal collecting vessels under an Olympus SZX7 stereo microscope. The dermis was
125
sutured with 6-0 suture thread (Harvard Apparatus, Holliston, MA).
126
Oxazolone Exposure
127
Oxazolone (Sigma, catalog number E0753) is a potent pro-inflammatory agent. Oxazolone
128
induces contact hypersensitivity, an immunological response against the foreign chemical.
129
During the sensitization phase, cells of the innate immune system take up the oxazolone and
130
promote the formation of antigen-specific memory T cells (15). Subsequent contact with the
131
oxazolone activates the memory T cells and causes the recruitment of inflammatory cells to the
132
exposure site, which becomes outwardly manifest as tissue inflammation (14, 28). We allowed
133
rats to fully recover from axillary lymph node dissection (ALND). At 66 days post-ALND
134
surgery, at which time there was no edema (data not shown), the rats in the ALND (-) Oxazolone
135
(+) and ALND (+) Oxazolone (+) groups were sensitized by applying 300 μl of 2% oxazolone in
136
ethanol to the abdomen. A 1" x 1" area was shaved on the rat abdomen and 300 ul of the
137
oxazolone solution was evenly applied throughout the shaved area using a micropipette. The rat
138
was kept anesthetized until the applied oxazolone had dried. To initiate dermatitis, 300 μl of a
139
1.6% oxazolone solution of acetone and olive oil (4:1) was applied to the forelegs in the same
140
groups as before starting at day 7 post-sensitization (day 73 post-ALND surgery) and proceeding
141
every three days thereafter until day 82 post-surgery, for a total of 4 applications. For these
142
applications, rats were anesthetized and had their foreleg shaved. Then 300 uL of the 1.6%
143
oxazolone solution was evenly applied to the entire foreleg up to the shoulder using a
144
micropipette and allowed to dry before the rat was allowed to regain consciousness. Optical
145
images of the rat forelegs were collected before each application of oxazolone. Finally, rats were
146
imaged via ICG lymphography, injected with lymphatic fluid tracer, and euthanized for
147
cryosectioning of the foreleg on day 85 post-ALND.
148
Quantifying Interstitial Transport and Lymphatic Function
149
Tetramethylrhodamine-conjugated lysine fixable dextran 70,000 MW (Invitrogen, Carlsbad, CA,
150
catalog number D1818) at 1mg/mL in PBS was used as a fluorescent lymph tracer to quantify
151
fluid drainage in the rat foreleg. At the specified day post surgery, 15 μL of fluorescent tracer
152
solution was injected intradermally into the posterior of both foreleg paws. We allowed all rats
153
to recover for 6 hours following dextran injections to quantify lymph drainage post-surgery.
154
Because the presence and distribution of the tracer across the foreleg depends upon lymphatic
155
drainage, the coverage of fluorescent tracer that is measured later in foreleg cross sections can
156
serve to quantify interstitial transport and lymphatic function across the foreleg. The extent of
157
fluid marker coverage in cross sections made at the wrist is a direct measurement of the ability of
158
fluid marker to spread interstitially from the paw injection site towards the lymphatics proximal
159
to the wrist.
160
foreleg is indicative of fluid accumulation. At all locations, the absence of fluid marker can be
The extent of fluid marker coverage at the middle and upper locations of the
161
an indication of either highly efficient or highly dysfunctional lymphatics. Collected forelegs
162
were cryo-sectioned to produce 100 µm cross-sections at the elbow joint (designated as the upper
163
location), midway between the elbow and wrist (middle location), and near the wrist (lower
164
location).
165
(DAPI; Vector Laboratories, Burlingame, CA, catalog number H-1200) and imaged under an
166
Olympus BX51 Fluorescent Microscope. Fluorescent tracer area of coverage was quantified
167
using Metamorph Offline 6.3r7 software and expressed as a percentage of total cross-sectional
168
area of the foreleg tissue section.
169
Fluorescence Imaging
170
Fluorescent imaging was conducted on foreleg specimens cut into 100-µm cross sections. The
171
path taken by lymph through the foreleg following the injection of 70,000 MW
172
tetramethylrhodamine-conjugated dextran was identified in the cross-sections by immobilizing
173
the lysine-fixable fluid tracer. Cell nuclei were counterstained with DAPI. Fluorescent images
174
were captured with a Zeiss MRm camera on a Zeiss Axiovert 200M fluorescence microscope
175
with the Apotome system. This system collects a stack of two- dimensional images that are then
176
compressed into a single image.
177
Physiological measurements
178
Foreleg wrist thickness was measured in Metamorph from digital images of the rat foreleg.
179
Foreleg area was measured in Metamorph from digital images of the right rat foreleg by
180
outlining the paw, wrist, and leg 2.5 cm proximally from the wrist. Foreleg wrist thickness and
181
foreleg area measurements for the Oxazolone (+) groups were normalized to the Oxazolone (-)
182
condition. Skin thickness of the swollen and non-swollen ALND (-) Oxazolone (-) arm of each
183
rat was measured with MetaMorph Imaging software (Molecular Devices, PA) from sections
Sections were counterstained for cell nuclei with 4’, 6-diamino-2-phenylindole
184
obtained 4mm distal to the elbow of each arm. Skin thickness of the edematous Oxazolone (+)
185
groups was normalized to the Oxazolone (-) condition.
186
Imaging of functional lymphatic vessels via ICG fluorescence lymphography
187
We employed indocyanine green (ICG) fluorescence lymphography to identify functional
188
lymphatic vessels following ALND and oxazolone exposure (n=5 per group). An imaging
189
system recently developed by Drs. N. Unno, F. Ogata, and Eva M. Sevick-Muraca (23, 24, 31,
190
32, 40) was used to detect functional lymphatic vessels in the rat foreleg. 7.5 μl of a 5 mg/ml
191
solution of the fluorescent near-infrared dye Indocyanine Green (ICG; Akorn, catalog number
192
17478-701-02) was injected into the rat paw. Detection of ICG was performed with an electron
193
multiplying charge-coupled device (CCD) (Hamamatsu, C9100-13) using Hamamatsu
194
recommendations as described previously (40). ICG was illuminated with an array of 760 nm
195
LEDs (Epitex Inc) placed before a 760 nm band pass filter (model 760FS10-50, Andover
196
Corporation) to provide the excitation light for activating ICG. A 785 nm and 763 nm custom
197
made holographic notch band rejection filter (model HNPF-785.0-2.0 and HNPF-763.0-2.0,
198
Kaiser Optical Systems) and a 830 nm image quality bandpass filter (model 830FS20-25,
199
Andover Corporation) were placed before the camera lens to selectively reject the excitation
200
light and pass the emitted 830 nm wavelength. Following microdose paw injections of ICG dye,
201
images were collected at 10 minutes to visualize the path of ICG drainage through the rat
202
foreleg. Careful attention was made to inject the ICG dye at the proper depth in the paw dermis.
203
For this, the needle insertion was made parallel to the plane of the paw, so that the needle tip was
204
fully covered yet within the dermis.
205
Statistical Methods
206
Five animals were used for each experimental group. Data are presented as means with standard
207
deviations (SD). P values were calculated using ANOVA or Mann-Whitney U test.
208
209
Results
210
Increased Inflammation-Induced Edema Post-ALND
211
In order to reveal the presence of a chronic latent lymphatic deficiency, we compared the ability
212
of the lymphatic system that had regenerated in the rat foreleg following ALND to prevent
213
edema during an acute inflammatory challenge with that of the uninjured foreleg lymphatic
214
system. Following the ALND surgery and at all other times up to day 73 (measurements made
215
every five days) there was no detectable swelling in any of the forelegs (data not shown).
216
Beginning at 73 days post-ALND, long after quiescence of the acute wound healing phase, the
217
forelegs of sensitized rats were exposed to the pro-inflammatory agent oxazolone to challenge
218
the foreleg lymphatic system. Such an approach was intended to increase lymphatic load to
219
failure in order reveal the presence of latent lymphatic deficiencies in the ALND-regenerated
220
foreleg. We detected significant swelling in the ALND (+) Oxazolone (+) foreleg relative to
221
ALND (-) Oxazolone (+) forelegs by day 82, which was further increased at day 85 (Figure 1), at
222
which time the rats were euthanized for histological analyses. At day 82, the foreleg wrist
223
swelling was significantly increased in ALND (+) Oxazolone (+) forelegs relative to ALND (-)
224
Oxazolone (+) (7.46 +/- 2.39 % vs. 0.06 +/- 4.43 %, respectively, p<0.05 by ANOVA). At day
225
85, the increase in wrist and foreleg area swelling were both highly significant in the ALND (+)
226
Oxazolone (+) forelegs relative to ALND (-) Oxazolone (+) (9.71 +/- 2.55 % for ALND (+) wrist
227
swelling vs. -0.08 +/- 4.46m % for ALND (-), p<0.005 by ANOVA; 13.4 +/- 4.5 % for ALND
228
(+) foreleg area swelling vs. 4.91 +/- 2.04 % for ALND (-), p<0.005 by ANOVA). ALND (-)
229
Oxazolone (+) mean foreleg area was found to have peaked at ~5.7% swelling on day 76,
230
whereas ALND (+) Oxazolone (+) mean foreleg area was highest at ~13.4% swelling on day 85.
231
ALND (-) Oxazolone (+) mean wrist diameter peaked at ~2.2% swelling on day 79, whereas
232
ALND (+) Oxazolone (+) mean wrist diameter peaked at ~9.7% swelling at day 85. The peak
233
trends suggest that edema in the ALND (-) Oxazolone (+) foreleg had stabilized by as early as
234
days 76 - 79, whereas edema in the ALND-recovered foreleg had not yet stabilized even at the
235
time of euthanization (day 85), indicating that swelling in the ALND-recovered forelegs may
236
have continued to increase over time. The data demonstrate a significantly reduced capacity of
237
the ALND-regenerated lymphatic system to prevent edema during the oxazolone-induced acute
238
inflammation.
239
Appearance of Edematous Skin During Acute Inflammation
240
In order to identify histological changes associated with the foreleg swelling differences that
241
were measured in Figure 1, we measured skin thickness in histological foreleg cross sections.
242
Skin thickness was found to have significantly increased in both oxazolone-treated foreleg
243
groups relative to the ALND (-) Oxazolone (-) foreleg by approximately 35% at the lower
244
location, 45% at middle location, and 75% at the upper location (Figure 2D; p<0.05, by
245
ANOVA), suggesting that the oxazolone treatment induced long-term cellular and structural
246
changes in the skin irrespective of the presence of a healed lymphatic injury. The increase in
247
skin thickness was predominantly seen in the dermal and epidermal layers, with less expansion
248
seen in the hypodermis. However, there were no significant skin thickness differences between
249
the ALND-recovered and ALND (-) Oxazolone (+) forelegs (p>0.05). There was a significant
250
increase for both oxazolone treated foreleg groups in the upper foreleg location relative to the
251
lower location (Figure 2D; p<0.05, by ANOVA). This may have occurred because there are
252
more soft tissues in upper forelegs locations, so the ability to accumulate fluid in this area is
253
greater.
254
Reduced Lymphatic Drainage During Post-ALND Inflammation
255
Because it was evident that oxazolone application had resulted in greater edema in the ALND (+)
256
Oxazolone (+) forelegs relative to ALND (-) Oxazolone (+) forelegs (Figure 1), we asked
257
whether the increased swelling was a consequence of deficient lymphatic drainage. We therefore
258
visualized lymphatic drainage with indocyanine green (ICG) fluorescent lymphography at day
259
85, prior to euthanization. Indocyanine Green (ICG) dye in ALND (-) Oxazolone (-) forelegs
260
was seen to drain from the injection site by large lymphatic vessels leading to the axillary lymph
261
nodes (Figure 3A), indicative of normally functioning lymphatic vessels in the foreleg. ALND (-
262
) Oxazolone (+) forelegs were found to have incurred deficient lymph drainage, as the small ICG
263
dye was seen mainly to spread interstitially throughout the paw and leg, reaching up to the elbow
264
of the foreleg (Figure 3B). However, an even more dramatically impaired lymph drainage was
265
seen in the ALND-recovered forelegs treated with oxazolone. ICG dye in these forelegs was
266
mainly restricted to the paw, near the injection site (Figure 3C), indicative of a more serious
267
lymphatic transport impairment. Thus, whereas reduced lymphatic drainage was apparent in the
268
ALND (-) Oxazolone (+) forelegs due to the oxazolone-induced inflammation, a greater
269
lymphatic drainage impairment was seen in the ALND-recovered forelegs.
270
lymphography images correspond with the swelling data (Figure 1) and confirm lymphatic
271
drainage deficiency as a potential contributing source of the increased inflammation-induced
272
edema in the ALND-recovered forelegs.
273
Reduced Interstitial Transport and Lymphatic Function During Acute Inflammation
274
A second method was employed to assess the degree of lymph drainage impairment in the rat
275
forelegs. Fluorescent rhodamine-dextran was injected into rat paws to trace the path of lymph
276
drainage through the foreleg and quantitatively assess the ability of the lymphatic system to drain
277
macromolecules from the paw by measuring fluorescent fluid tracer area coverage in histological
The ICG
278
foreleg cross sections at the elbow (upper location), midway between the elbow and wrist
279
(middle location), and at the wrist (lower location). Fluid marker was found to be strongly
280
present interstitially and subcutaneously in the lower location, near the injection site of ALND
281
(-) Oxazolone (-) forelegs (Figure 4A). Fluid marker was more weakly present but was also seen
282
spreading interstitially and subcutaneously in the lower location of ALND (+/-) Oxazolone (+)
283
forelegs (Figure 4B & 4C). Fluid marker coverage in the lower location of ALND (+/-)
284
Oxazolone (+) forelegs was significantly lower than in the control foreleg (Figure 4D; p<0.05,
285
by Mann-Whitney U test). The data may indicate a reduced interstitial transport of fluid marker
286
from the injection site in the paw to more proximal lymphatic vessels due to the oxazolone
287
treatment. However, the data may also indicate an enormously increased interstitial transport due
288
to a dramatically increased blood capillary permeability, which has been shown to occur from
289
the oxazolone (9). Importantly, we also found that the ALND (+) Oxazolone (+) fluid marker
290
area coverage was significantly lower relative to the ALND (-) Oxazolone (+) condition,
291
significantly reducing by ~80% (p<0.05 by Mann-Whitney U test). Because the significantly
292
decreased fluid marker coverage was associated with the lymphatic injury, this suggests that a
293
reduction in fluid marker coverage indicates a reduced lymphatic function. Indeed, a reduced
294
lymphatic flow rate has been recently shown by direct lymphatic flow measurement conducted in
295
a model of chronic skin inflammation induced by oxazolone (11). Fluid marker was scarcely
296
detectable in the cross sections of all three groups at the middle and upper locations (data not
297
shown). We have found in a recent study in the mouse foreleg that fluid marker signal increases
298
at the middle and upper locations of mildly edematous tissues, indicative of fluid accumulation
299
(19). We also found that signal is near zero at the middle and upper locations of healthy tissues
300
as the lymphatic system is highly effective in clearing fluid marker from the limb once the
301
marker has entered the lymphatic system (19). We were surprised in the present study to find a
302
near zero coverage of fluid marker in the edematous limb at the middle and upper locations.
303
When taken together with the presence of edema in the oxazolone treated limb, a worsened
304
edema in the post-ALND foreleg relative to the ALND (-) Oxazolone (+) foreleg, the dramatic
305
lymph drainage impairment that was apparent with the ICG imaging, and the drastically reduced
306
interstitial drainage that we measured from the reduced marker coverage in the wrist cross-
307
sections (including the significant 80% reduction in ALND (+) Oxazolone (+) signal coverage
308
relative to the ALND (-) Oxazolone (+) condition), we conclude that lymph drainage is severely
309
impaired in the oxazolone treated forelegs and is more impaired in the ALND (+) Oxazolone (+)
310
foreleg relative to the ALND (-) Oxazolone (+) foreleg.
311
312
Discussion
313
Although secondary lymphedema is a common consequence of surgery to treat breast cancer, it
314
remains unclear why the apparently functional lymphatic system in the human arm is unable to
315
accommodate an acute inflammatory stimulus, which can induce chronic lymphedema. Here, we
316
are the first to demonstrate that the lymphatic-independent recovery of acute lymphedema in the
317
foreleg masks an ensuing chronic and latent lymphatic insufficiency, which reduces the ability of
318
the foreleg lymphatic system to prevent edema during acute inflammation. The present study
319
represents, to our knowledge, the first experimental evidence in support of the largely
320
unexplored clinical observation that secondary lymphedema is often preceded by an acute
321
inflammatory stimulus (18, 21, 29). Indeed, we showed that an acute inflammatory challenge
322
significantly reduced lymphatic drainage, even in the lymphatic-intact foreleg, leading to an
323
increased post-ALND foreleg edema. Our findings therefore demonstrate a clear link between
324
inflammatory stimulation and increased post-ALND fluid accumulation, and demonstrate that
325
insufficient lymphatic drainage may be a potential contributing mechanism.
326
327
The similar lymphatic injury, period of latency, and disease onset between the post-ALND rat
328
foreleg and the human arm suggests that the post-ALND rat foreleg may be a useful model for
329
clarifying the lymphatic deficiency that pre-exists human secondary lymphedema and its
330
response to acute inflammation. Although lymphatic function may be chronically impaired by
331
the ALND surgery, there was no measured edema or deficient lymphatic drainage prior to the
332
inflammatory stimulus.
333
foreleg edema supports the hypothesis that the ALND surgery produces a chronic latent
334
lymphatic functional deficiency that is not adequately restored by the natural wound healing
335
process. We show here that the chronic latent deficiency becomes outwardly manifest during
336
periods of acute inflammation. This chronic latent lymphatic functional deficiency may consist,
337
at least in part, of a reduced maximum lymph flow rate caused by an insufficient wound healing
338
lymphangiogenesis. Indeed, we and others have reported deficient physiological wound healing
339
lymphangiogenesis in the mouse (2, 3, 41). Because the onset of secondary lymphedema in the
340
human arm also follows a time delay (25) during which time lymph continues to drain normally
341
(27, 33, 34), a similar latent lymphatic deficiency may be present in the post-ALND human arm,
342
which predisposes the arm to chronically swell following an acute inflammatory stimulus. A
343
straightforward explanation for the increased post-ALND edema is that lymphatic obstruction in
344
the rat foreleg reduces the maximum lymphatic flow rate (i.e. lymphatic reserve) without altering
345
lymph outflow under normal conditions. In conditions of acute inflammation, the reduced
The appearance of an increased inflammation-induced post-ALND
346
lymphatic reserve results in a greater lymphatic flow reduction and an increased fluid
347
accumulation relative to what occurs in a healthy lymphatic system.
348
349
It has recently been demonstrated that oxazolone mediated inflammatory responses increase
350
capillary permeability (9), spur inducible nitric oxide (NO) release (17), and reduce lymphatic
351
flow (11). NO is a key molecular regulator of lymphatic contractility (1) that is known to slow
352
lymphatic drainage by reducing lymphatic contraction (30, 43). Reduced lymphatic pumping
353
activity due to increased NO production may explain the appearance of edema in the ALND (-)
354
Oxazolone (+) forelegs. A reduced maximum lymphatic flow rate in the post-ALND limb due to
355
an increased lymphatic outflow resistance may explain the further increase in edema that we saw
356
in the ALND (+) Oxazolone (+) forelegs. Thus, we hypothesize that the increased
357
intralymphatic pressure and weakening of lymphatic contractility and pump function caused by
358
the increased regional outflow lymphatic network resistance after ALND may be worsened by
359
acute inflammation due to an accelerated pump failure.
360
361
It has long been hypothesized that an increased lymphangiogenesis may be beneficial for
362
increasing fluid drainage in humans with secondary lymphedema. A number of groups have
363
reported that lymphatic growth promoting molecular therapies hasten the natural resolution of
364
experimental acute lymphedema (5, 13, 38, 44), thereby raising the prospect that similar
365
approaches may be useful for resolving chronic lymphedema in the human. However, we have
366
found that acute lymphedema in experimental models resolves via a lymphangiogenesis-
367
independent mechanism (19, 26). Therefore, conclusions regarding the therapeutic effectiveness
368
of stimulated lymphangiogenesis drawn from acute lymphedema models may not necessarily
369
apply to chronic lymphedema patients.
370
371
In contrast to acute lymphedema, the inflammation-induced swelling and lymphatic deficiency
372
that we have found in the post-ALND rat foreleg is independent of an ongoing wound repair
373
process and is entirely dependent on the functional status of the foreleg lymphatic system as it
374
had regenerated following the surgery to remove the axillary lymph nodes.
375
experimental model more closely parallels the human disease etiology relative to previously
376
employed acute lymphedema animal models in that there is a delay between the surgery and the
377
appearance of swelling as opposed to an immediate edema onset, an inflammatory stimulus that
378
initiates the swelling as opposed to a wound obstruction, the lack of an ongoing wound repair
379
process with active remodeling of multiple inter-related systems, and an edema caused by a
380
chronically (as opposed to acutely) impaired lymphatic system. We have used this more realistic
381
animal model to provide novel insights into the latent lymphatic deficiency that precedes
382
secondary lymphedema.
The present
383
384
Presently, therapies for treating secondary lymphedema have largely been directed at increasing
385
lymphatic drainage to clear excess fluid after disease onset. Interestingly, it has recently been
386
reported in the mouse ear that an increased lymph flow may help resolve acute inflammation and
387
reduce inflammation-induced edema (10), whereas decreased lymphangiogenesis or lymphatic
388
function may have the opposite effect (7, 36). It may therefore be beneficial to apply pro-
389
lymphangiogenic therapy prophylactically in order to increase lymphatic vasculogensis to reduce
390
susceptibility for developing lymphedema during periods of acute inflammation, as capillary
391
lymphatics by themselves will not be able to conduct all upper limb tissue fluid/lymph.
392
However, it should be noted that lymphatic vasculogensis has not been shown to occur in
393
humans, although it has been shown to occur in the mouse (12, 19). The present experimental
394
model may be more ideal than existing models for evaluating experimental lymphatic growth
395
promoting therapies, which could be applied to the post-ALND foreleg either during the wound
396
healing phase or during the subsequent period of latency and then tested at a later time by
397
challenging the lymphatic system with an acute inflammatory stimulus, as we have done here.
398
399
GRANTS
400
This work was funded by National Institutes of Health Grants R21-AR-053094, R21-HL-093568
401
and R15-HL-093705.
402
403
DISCLOSURES
404
No conflicts of interest, financial or otherwise, are declared by the author(s).
405
406
Figure Legends
407
Figure 1. Inflammation-Induced Edema is Worsened in the Post-ALND Rat Foreleg.
408
Shown are representative images of ALND (-) Oxazolone (-) (A – top row), and ALND (-)
409
Oxazolone (+) (A – middle row), and ALND (+) Oxazolone (+) (A – bottom row) rat forelegs at
410
days 73, 79, and 85 post-surgery. Black scale bar in A, lower right panel = 10 mm. Foreleg arm
411
area (B) and wrist diameter (C) were measured for all forelegs and graphed with means and
412
standard deviations. * indicates statistical significance and ** indicates high statistical
413
significance relative to the control condition. n = 5 forelegs per group.
414
Figure 2. Foreleg Skin Thickness Increases During Acute Inflammation.
415
Shown are fluorescent images of rat foreleg skin cryosections stained with DAPI to identify cell
416
nuclei (blue) at 85 days post-surgery and at the middle foreleg location for ALND (-) Oxazolone
417
(-) (A), ALND (-) Oxazolone (+) (B), and ALND (+) Oxazolone (+) (C) skin. Black scale bar in
418
C = 1 mm. Skin thickness was measured and graphed for all groups at all foreleg locations (D).
419
* indicates statistical significance relative to the normal condition.
420
significance relative to the lower location within a particular condition. n = 5 forelegs per group.
421
Figure 3. Acute Inflammation Impairs Post-ALND Lymph Drainage.
422
Shown are ICG fluorescence lymphography images captured from ALND (-) Oxazolone (-) (A),
423
ALND (-) Oxazolone (+) (B), and ALND (+) Oxazolone (+) forelegs (C) at 85 days post-
424
surgery. Images were captured at 10 minutes following the ICG paw injection. Red arrows
425
identify the paw injection site. Green arrows identify the elbow. Yellow arrows identify the
426
axillary region. White scale bar in C = 10 mm. n = 5 forelegs per group.
427
Figure 4. Acute Inflammation Impairs Interstitial Drainage.
# indicates statistical
428
Dextran fluid marker was imaged in cross-sections from the ALND (-) Oxazolone (-) (A),
429
ALND (-) Oxazolone (+) (B), and ALND (+) Oxazolone (+) forelegs at the lower location. Red
430
color in each image is fluorescent dextran lymph fluid tracer. Blue color in each image is DAPI
431
labeled cell nuclei. Yellow arrows identify fluid marker. Black scale bar in C = 1 mm. Fluid
432
marker coverage and peak intensity are graphically depicted (D).
433
difference.
n
=
5
forelegs
* indicates statistical
per
group.
434
References
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
1.
Akl TJ, Nagai T, Cote GL, and Gashev AA. Mesenteric lymph flow in adult and aged rats. Am J
Physiol Heart Circ Physiol 301: H1828-1840, 2011.
2.
Avraham T, Clavin NW, Daluvoy SV, Fernandes J, Soares MS, Corderio AP,
and Mehrara BJ. Fibrosis Is a Key Inhibitor of LYmphatic Regneration. Plastic and
Reconstructive Surgery 124: 438-450, 2009.
3.
Avraham T, Daluvoy S, Zampell J, Yan A, Haviv YS, Rockson SG, and Mehrara BJ. Blockade of
transforming growth factor-beta1 accelerates lymphatic regeneration during wound repair. Am J Pathol
177: 3202-3214, 2010.
4.
Bates DO. An interstitial hypothesis for breast cancer related lymphoedema. Pathophysiology
2009.
5.
Cheung L, Han J, Beilhack A, Joshi S, Wilburn P, Dua A, An A, and Rockson SG. An experimental
model for the study of lymphedema and its response to therapeutic lymphangiogenesis. BioDrugs 20:
363-370, 2006.
6.
Clarke D, Martinez A, and Cox RS. Analysis of Cosmetic Results and Complications in Patients
with Stage-I and Stage-Ii Breast-Cancer Treated by Biopsy and Irradiation. Int J Radiat Oncol 9: 18071813, 1983.
7.
Guo RL, Zhou Q, Proulx ST, Wood R, Ji RC, Ritchlin CT, Pytowski B, Zhu ZP, Wang YJ, Schwarz
EM, and Xing LP. Inhibition of Lymphangiogenesis and Lymphatic Drainage via Vascular Endothelial
Growth Factor Receptor 3 Blockade Increases the Severity of Inflammation in a Mouse Model of Chronic
Inflammatory Arthritis. Arthritis Rheum 60: 2666-2676, 2009.
8.
Halsted WS. The swelling of the arm after operation for cancer of the breast, elephantiasis
chirurgica, its causes and prevention. Bull Johns Hopkins Hosp 32: 309-313, 1921.
9.
He CF, Young AJ, West CA, Su M, Konerding MA, and Mentzer SJ. Stimulation of regional
lymphatic and blood flow by epicutaneous oxazolone. Journal of Applied Physiology 93: 966-973, 2002.
10.
Huggenberger R, Siddiqui SS, Brander D, Ullmann S, Zimmermann K, Antsiferova M, Werner S,
Alitalo K, and Detmar M. An important role of lymphatic vessel activation in limiting acute
inflammation. Blood 117: 4667-4678, 2011.
11.
Huggenberger R, Ullmann S, Proulx ST, Pytowski B, Alitalo K, and Detmar M. Stimulation of
lymphangiogenesis via VEGFR-3 inhibits chronic skin inflammation. J Exp Med 207: 2255-2269, 2010.
12.
Ikomi F, Kawai Y, Nakayama J, Ogiwara N, Sasaki K, Mizuno R, and Ohhashi T. Critical roles of
VEGF-C-VEGF receptor 3 in reconnection of the collecting lymph vessels in mice. Microcirculation 1-13,
2008.
13.
Jin da P, An A, Liu J, Nakamura K, and Rockson SG. Therapeutic responses to exogenous VEGF-C
administration in experimental lymphedema: immunohistochemical and molecular characterization.
Lymphat Res Biol 7: 47-57, 2009.
14.
Kaplan DH, Igyarto BZ, and Gaspari AA. Early immune events in the induction of allergic contact
dermatitis. Nat Rev Immunol 12: 114-124, 2012.
15.
Kaplan DH, Kissenpfennig A, and Clausen BE. Insights into Langerhans cell function from
Langerhans cell ablation models. Eur J Immunol 38: 2369-2376, 2008.
16.
Kissin MW, Querci della Rovere G, Easton D, and Westbury G. Risk of lymphoedema following
the treatment of breast cancer. Br J Surg 73: 580-584, 1986.
17.
Liao S, Cheng G, Conner DA, Huang YH, Kucherlapati RS, Munn LL, Ruddle NH, Jain RK,
Fukumura D, and Padera TP. Impaired lymphatic contraction associated with immunosuppression. P
Natl Acad Sci USA 108: 18784-18789, 2011.
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
18.
Mak SS, Yeo W, Lee YM, Tse SM, Ho FP, Zee B, and Chan E. Risk factors for the initiation and
aggravation of lymphoedema after axillary lymph node dissection for breast cancer. Hong Kong Med J
15: 8-12, 2009.
19.
Mendez U, Brown EM, Ongstad EL, Slis JR, and Goldman J. Functional Recovery of Fluid
Drainage Precedes Lymphangiogenesis in Acute Murine Foreleg Lymphedema. Am J Physiol Heart Circ
Physiol 302: H2250-2256, 2012.
20.
Mortimer PS. The pathophysiology of lymphedema. Cancer 83: 2798-2802, 1998.
21.
Mortimer PS, Bates DO, Brassington HD, Stanton AWB, Strachan DP, and Levick JR. The
prevalence of arm oedema following treatment for breast cancer. Qjm-Mon J Assoc Phys 89: 377-380,
1996.
22.
Nakamura K, Radhakrishnan K, Wong YM, and Rockson SG. Anti-Inflammatory
Pharmacotherapy with Ketoprofen Ameliorates Experimental Lymphatic Vascular Insufficiency in Mice.
Plos One 4: 2009.
23.
Ogata F, Azuma R, Kikuchi M, Koshima I, and Morimoto Y. Novel lymphography using
indocyanine green dye for near-infrared fluorescence labeling. Ann Plast Surg 58: 652-655, 2007.
24.
Ogata F, Narushima M, Mihara M, Azuma R, Morimoto Y, and Koshima I. Intraoperative
lymphography using indocyanine green dye for near-infrared fluorescence labeling in lymphedema. Ann
Plast Surg 59: 180-184, 2007.
25.
Olszewski W. On the pathomechanism of development of postsurgical lymphedema.
Lymphology 6: 35-51, 1973.
26.
Ongstad EL, Bouta EM, Roberts JE, Uzarski JS, Gibbs SE, Sabel MS, Cimmino VM, Roberts MA,
and Goldman J. Lymphangiogenesis-independent resolution of experimental edema. Am J Physiol Heart
Circ Physiol 299: H46-54, 2010.
27.
Pain SJ, Barber RW, Solanki CK, Ballinger JR, Britton TB, Mortimer PS, Purushotham AD, and
Peters AM. Short-term effects of axillary lymph node clearance surgery on lymphatic physiology of the
arm in breast cancer. J Appl Physiol 99: 2345-2351, 2005.
28.
Saint-Mezard P, Berard F, Dubois B, Kaiserlian D, and Nicolas JF. The role of CD4+ and CD8+ T
cells in contact hypersensitivity and allergic contact dermatitis. Eur J Dermatol 14: 131-138, 2004.
29.
Sakorafas GH, Peros G, Cataliotti L, and Vlastos G. Lymphedema following axillary lymph node
dissection for breast cancer. Surg Oncol 15: 153-165, 2006.
30.
Saul ME, Thomas PA, Dosen PJ, Isbister GK, O'Leary MA, Whyte IM, McFadden SA, and van
Helden DF. A pharmacological approach to first aid treatment for snakebite. Nat Med 17: 809-811, 2011.
31.
Sevick-Muraca EM, Sharma R, Rasmussen JC, Marshall MV, Wendt JA, Pham HQ, Bonefas E,
Houston JP, Sampath L, Adams KE, Blanchard DK, Fisher RE, Chiang SB, Elledge R, and Mawad ME.
Imaging of lymph flow in breast cancer patients after microdose administration of a near-infrared
fluorophore: feasibility study. Radiology 246: 734-741, 2008.
32.
Sharma R, Wang W, Rasmussen JC, Joshi A, Houston JP, Adams KE, Cameron A, Ke S, Kwon S,
Mawad ME, and Sevick-Muraca EM. Quantitative imaging of lymph function. Am J Physiol Heart Circ
Physiol 292: H3109-3118, 2007.
33.
Stanton AW, Modi S, Bennett Britton TM, Purushotham AD, Peters AM, Levick JR, and
Mortimer PS. Lymphatic drainage in the muscle and subcutis of the arm after breast cancer treatment.
Breast Cancer Res Treat 117: 549-557, 2009.
34.
Stanton AW, Modi S, Mellor RH, Levick JR, and Mortimer PS. Recent advances in breast cancerrelated lymphedema of the arm: lymphatic pump failure and predisposing factors. Lymphat Res Biol 7:
29-45, 2009.
35.
Suami H, Pan WR, and Taylor GI. The lymphatics of the skin filled by a dermal backflow: an
observation in a scarred cadaver leg. Lymphology 40: 122-126, 2007.
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
36.
Sugaya M, Kuwano Y, Suga H, Miyagaki T, Ohmatsu H, Kadono T, Okochi H, Blauvelt A, Tamaki
K, and Sato S. Lymphatic Dysfunction Impairs Antigen-Specific Immunization, but Augments Tissue
Swelling Following Contact with Allergens. Journal of Investigative Dermatology 132: 667-676, 2012.
37.
Szuba A, Pyszel A, Jedrzejuk D, Janczak D, and Andrzejak R. Presence of functional axillary
lymph nodes and lymph drainage within arms in women with and without breast cancer-related
lymphedema. Lymphology 40: 81-86, 2007.
38.
Szuba A, Skobe M, Karkkainen MJ, Shin WS, Beynet DP, Rockson NB, Dakhil N, Spilman S, Goris
ML, Strauss HW, Quertermous T, Alitalo K, and Rockson SG. Therapeutic lymphangiogenesis with
human recombinant VEGF-C. FASEB J 16: 1985-1987, 2002.
39.
Tammela T, Saaristo A, Holopainen T, Lyytikka J, Kotronen A, Pitkonen M, Abo-Ramadan U,
Yla-Herttuala S, Petrova TV, and Alitalo K. Therapeutic differentiation and maturation of lymphatic
vessels after lymph node dissection and transplantation. Nat Med 13: 1458-1466, 2007.
40.
Unno N, Inuzuka K, Suzuki M, Yamamoto N, Sagara D, Nishiyama M, and Konno H. Preliminary
experience with a novel fluorescence lymphography using indocyanine green in patients with secondary
lymphedema. J Vasc Surg 45: 1016-1021, 2007.
41.
Uzarski J, Drelles MB, Gibbs SE, Ongstad EL, Goral JC, McKeown KK, Raehl AM, Roberts MA,
Pytowski B, Smith MR, and Goldman J. The resolution of lymphedema by interstitial flow in the mouse
tail skin. Am J Physiol Heart Circ Physiol 294: H1326-1334, 2008.
42.
Warren AG, and Slavin SA. Scar lymphedema - Fact or fiction? Ann Plas Surg 59: 41-45, 2007.
43.
Weiler M, Kassis T, and Dixon JB. Sensitivity analysis of near-infrared functional lymphatic
imaging. J Biomed Opt 17: 066019, 2012.
44.
Yoon YS, Murayama T, Gravereaux E, Tkebuchava T, Silver M, Curry C, Wecker A, Kirchmair R,
Hu CS, Kearney M, Ashare A, Jackson DG, Kubo H, Isner JM, and Losordo DW. VEGF-C gene therapy
augments postnatal lymphangiogenesis and ameliorates secondary lymphedema. J Clin Invest 111: 717725, 2003.
45.
Zampell JC, Elhadad S, Avraham T, Weitman E, Aschen S, Yan A, and Mehrara BJ. Toll-like
receptor deficiency worsens inflammation and lymphedema after lymphatic injury. Am J Physiol Cell
Physiol 302: C709-719, 2012.
46.
Zampell JC, Yan A, Avraham T, Daluvoy S, Weitman ES, and Mehrara BJ. HIF-1alpha coordinates
lymphangiogenesis during wound healing and in response to inflammation. Faseb J 26: 1027-1039, 2012.