Abstract
Objective:
Herein we used near-infrared fluorescence lymphatic imaging in a pilot study to assess lymphatics in pre-ulcerative (C2-C4) venous insufficiency and determine whether involvement and/or degradation of lymphatic anatomy or function could play a role in the progression of chronic venous insufficiency. We also explored the role of lymphatics in early peripheral arterial disease.
Methods:
After informed consent and following intradermal injections of indocyanine green for rapid lymphatic uptake, near-infrared fluorescence lymphatic imaging was employed to assess lymphatic anatomical structure and quantify lymphatic propulsion rates in subjects with early venous insufficiency. Anatomical observations included interstitial backflow, characterized by the abnormal spreading of indocyanine green from the injection site primarily into the surrounding interstitial tissues; dermal backflow, characterized by the retrograde movement of dye-laden lymph from collecting lymphatics into the lymphatic capillaries; and lymphatic vessel segmentation and dilation.
Results:
Ten venous insufficiency subjects were enrolled, resulting in two legs with C2 disease, nine legs with C3 disease, eight legs with C4 disease, and one leg with C5 disease. Interstitial and/or dermal backflow were observed in 25%, 33%, and 41% of injection sites in each limb with C2, C3, and C4 disease, respectively. Distinct vessel segmentation and dilation was observed in limbs with C3 and higher classification, and dermal backflow proximal to the injection sites was observed in two legs with C4 disease and in the inguinal region of the C5 study subject. Overall average lymph propulsion rates were 1.3±0.4 contractile events/min, 1.2±0.7 events/min, and 0.8±0.5 events/min for limbs with C2, C3, and C4 disease respectively. One subject with peripheral arterial disease, who had previously undergone bypass surgery, presented with extensive dermal backflow and lymphatic reflux.
Conclusions:
Near-infrared fluorescence lymphatic imaging demonstrate that, compared to normal health subjects, lymphatic anatomy and contractile function generally degrade with severity of venous insufficiency. Lymphatic abnormalities mimic those in early cancer-acquired lymphedema subjects, as previously observed by us and others. Additional studies are needed to decipher the relationship, including any causality, between lymphatic dysfunction and peripheral vascular disease and venous insufficiency.
Keywords: venous insufficiency, venous disease, lymphatics, lymphatic imaging, indocyanine green
Table of Contents Summary
Lymphatic dysfunction occurs in early (C2-C4) venous insufficiency and, when taken together with prior studies showing lymphatic dysfunction in advanced (C5-C6) venous insufficiency, suggests a lymphatic component to progressive venous insufficiency. Treating lymphatic dysfunction in early peripheral arterial disease and chronic venous insufficiency could potentially slow progression and improve outcomes.
Introduction
Chronic venous insufficiency (CVI), is a common condition encompassing a wide range of symptoms ranging from unsightly superficial telangiectases to venous ulceration that, left untreated, can result in infection and, in rare cases, amputation1. The disease results from a degradation of the ability of veins to efficiently return blood to the heart and is most prevalent in the legs, where the veins must contend with increased gravitational effects associated with the vertical distance from the feet to the heart. Venous insufficiency is descriptively classified using the clinical section of the CEAP classification system and includes seven main classifications: C0 – no visual evidence of venous insufficiency; C1 – telangiectases (spider veins) and/or reticular veins; C2 – varicose veins; C3 – edema of the ankle; C4 – skin changes such as pigmentation or eczema (C4a), lipodermatosclerosis and/or atrophie blanche (C4b), corona phlebectatica (C4c); C5 – healed venous ulcer; and C6 – active ulceration.2,3 Varicose veins affect up to 56% of males and 73% of females, while CVI with edema and/or more advanced symptoms affects up to 17% and 40% of males and females, respectively.4 Risk factors for the disease include age, sex, body mass index (BMI), family history, pregnancy, vein inflammation or leg injury, a sedentary lifestyle, and prolonged periods of time standing or sitting.5
The lymphatic and venous systems are closely interconnected6 and many patients with advanced CVI develop lymphedema, a disease which results from lymphatic failure and manifests many of the same symptoms, i.e. swelling, skin fibrosis, and poor immune response, seen in CVI. Previously, it was thought that venous hypertension prevented reabsorption of capillary filtrate into the venous circulation, yet recent studies show that venous intraluminal glycocalyx lining severely limits fluid reabsorption, requiring the lymphatics to be responsible for fluid homeostasis under normal physiologic conditions.7
Traditionally, clinical lymphatic imaging is accomplished using lymphoscintigraphy, to visualize large lymphatic trunks and lymph nodes. While this technique readily images lymphatic blockage (i.e. lack of radiotracer moving through lymphatic trunks) and backflow (i.e. gross trace movement into skin lymphatics and/or interstitial space), it does not have sufficient spatial or temporal resolution to image fine superficial lymphatic vasculature nor lymphatic contractile propulsion. Advances in optical imaging enable the use of near-infrared fluorescent (NIRF) contrast agents, such as indocyanine green (ICG), for lymphatic imaging. While optical techniques are depth limited, owing to light scattering in tissues, they provide ample spatial and temporal opportunities to image superficial lymphatic anatomy and contractile function in real-time, as well as lymph nodes up to 3–4 cm deep8–10 and NIRF-LI may also more accurately detect early lymphedema than lymphoscintigraphy,11,12 as well as allow direct assessment of interventions on lymphatic function.
In a seminal study, Franzeck et al. used fluorescence microlymphangiography to assess the lymphatic capillary bed in the skin of persons with CVI disease and noted more extensive dermal lymphatic capillaries in early disease, as compared to normal subjects, which then degraded and were obliterated in advanced CVI.13 Previously, we used NIRF lymphatic imaging (NIRF-LI) to assess the lymphatics in patients with active C6 disease and reported impaired lymphatic function with abnormal lymphatic pooling and/or dermal lymphatic backflow in all limbs with C5 and C6 disease, with fewer functional lymphatic vessels observed in subjects with the longest duration of ulceration.14 Interestingly, we observed lymphatic pooling in a contralateral leg with no observable (C0) disease. In a case of C4 disease we observed dermal lymphatic backflow which corresponded to the hemosiderin stain in the leg. As a result of these observations of abnormal lymphatics in early, non-ulcerated stages of CVI, we commenced a pilot study with NIRF-LI to assess the lymphatics in pre-ulcerative (C2-C4) CVI to determine whether involvement and/or degradation of lymphatic anatomy or function could play a role in the progression of CVI. We additionally imaged a subject with peripheral arterial disease (PAD) and include those results in the supplemental materials.
Methods
Subjects 18 years or older and diagnosed with venous insufficiency, were recruited for this study, which was approved by our local Institutional Review Board and the Food and Drug Administration (IND 102,765) for the intradermal, off-label use of indocyanine green (ICG) and the investigational use of a custom NIRF-LI system15. Subjects with active ulceration, clinically diagnosed lymphatic disorders, such as clinically overt lymphedema, or history of venous surgery, including vein stripping and ligation, phlebectomy, and/or venous bypass surgery were excluded. However, participants with history of less invasive sclerotherapy and/or endovascular procedures, such as radiofrequency ablation, were included to increase the size of the potentially eligible population.
After informed consent, subjects received 6–12 intradermal injections, each containing 25 μg ICG in 0.1 ml of saline, for a total dose of 150–300 μg ICG. Normal subjects received 12 standard injections, including two injections in the top of each foot, one injection in each medial ankle, and one injection in each lateral and medial calf, and one on the upper thigh. Most CVI subjects received eight of the standard injections excluding the medial calf and thigh injections. Subjects S01 and S02 also received the medial calf injections, for a total of 10 injections. After S02, the medial calf injections were discontinued, as, owing to their location immediately over the foot-draining lymphatics, they did not provide additional information. Subject S09 received the foot and ankle injections, but not the lateral calf injections, for a total of 6 injections.
Immediately after injection, NIRF-LI was accomplished by illuminating the subjects’ legs with diffuse 785 nm excitation light and collecting the resultant 830 nm fluorescent signal emanating from ICG-laden lymphatics using an image-intensified, scientific complementary metal–oxide–semiconductor camera system. We acquired approximately 3 frames per second with exposure times of 200 ms per image. This acquisition rate was sufficient to capture the movement of ICG-laden lymph through the lymphatic vessels, and image sequences were compiled to produce movies of lymphatic contractile pumping. For the first 45 minutes following injection in CVI subjects, NIRF-LI images were acquired in the supine position, after which, imaging was conducted while standing, both before and after 2 minutes of walking. Normal subjects were not imaged in the standing position.
Images and movies were assessed to identify abnormal lymphatic architectural characteristics, including interstitial backflow at the injection sites, dermal lymphatic backflow, dilated and/or segmented lymphatic vasculature, and vessels radiating from injection sites but not connecting to main collecting vessels. Dermal backflow is commonly observed in lymphatic failure, and results from the retrograde movement of lCG-laden lymph from collecting lymphatic vessels into the lymphatic capillaries and/or into the interstitial space. Interstitial backflow is similar to dermal backflow, but results from the spreading of ICG from the injection site into the surrounding interstitial tissues, most commonly without apparent backflow of ICG into lymphatic capillaries. The extent of dermal and interstitial backflow was quantified by adjusting the brightness of the image to 20% of the maximum pixel value and drawing a region of interest around the corresponding “bright” fluorescent area. The area of each region of interest was computed using ImageJ (NIH). Areas of interstitial backflow less than 5.0 cm2 were noted but not quantified, as they were generally obscured by the round bandages and black vinyl tape placed over injection sites. Segmented lymphatics had alternating sections of bright and dark vasculature that may be indicative of a corkscrew geometry similar to varicosities. Qualitative evaluation, i.e. absence or presence, of tortuosity, vessel dilation, and segmented appearance of lymphatic vessels was made by comparing images from previous studies of normal subjects. NIRF-LI images in the figures are presented in pseudo color and have been adjusted for brightness and contrast to enhance visualization of the full 16-bit image depth information.
Active lymphatic function or pumping was assessed as previously described9 by quantifying the lymphatic propulsion rates observed in the limbs while the subject was supine and standing, both before and after brief walking. Propulsion events were typically counted within 3–5 centimeters of the injection sites. As appropriate, paired and unpaired Student’s t-tests (α=0.05), were used to determine statistical significance in the quantified parameters across position (paired) and disease classification (unpaired).
Results
Eleven subjects, ten with CVI (S02-S11) were enrolled into this pilot study and one (S01) with PAD (see supplemental material) was enrolled into the exploratory PAD study. Nine subjects (N01-N09), with no known history of lymphatic or venous disease, were identified from an ongoing study (NCT000833599) and serve as a control group. After informed consent, subjects were imaged in accordance to institutional guidelines. Table I presents the demographic information for each subject. The median age of the CVI subjects was 53.5 years (range, 38–70 years; mean, 53 years) and most were severely obese with a median BMI of 38.9 (range 28.2–47.2; mean, 37.8). The median age of the normal subjects was 43.0 years (range, 30–58; mean, 44.9) with a median BMI of 30.3 (range, 23.5–37.6; mean, 29.6). Table II reports the abnormal lymphatic anatomical observations and the propulsion rates for each normal and CVI leg. The PAD subject’s imaging data is included in the supplemental material, including Online Figure 1 and Video 1 showing lymphatic reflux, but is not included in the associated analysis for evaluating the role of lymphatics in CVI. For analysis, the limbs were assigned to groups based on reported disease classification and included eighteen normal limbs, two C2 limbs, nine C3 limbs, eight C4 limbs, and one C5 limb. When identifying subjects below, the Subject ID is hyphenated with the referenced limb, for example, S09-R refers to the right leg of subject S09.
Table I:
Disease classification and demographics of study subjects.*
| Subject ID | Classification | Age | Gender | Race (Ethnicity) | Height [m] | Weight [kg] | BMI [kg/m2] | Primary Diagnosis for Participation | Prior Leg Surgery and Venous Intervention | Number of Injections (Total Dose ICG fog]) |
|---|---|---|---|---|---|---|---|---|---|---|
| S01 | Rutherford II - R, L | 61 | M | AA (NHL) | 1.70 | 124.74 | 43.1 | PAD | Bilateral bypass surgery | 10 (250) |
| S02 | C3 - R; C5 - L | 64 | M | C (NHL) | 1.88 | 106.59 | 30.2 | Bilateral venous insufficiency with edema | Possible history of venous ablation | 10 (250) |
| S03 | C4 - R, L | 54 | F | C (NHL) | 1.63 | 124.74 | 47.2 | VV with stasis changes | -- | 8 (200) |
| S04 | C2 - R, L | 44 | F | AA (NHL) | 1.68 | 79.37 | 28.2 | VV with edema | -- | 8 (200) |
| S05 | C4 - R, L | 57 | F | C (HL) | 1.57 | 104.32 | 42.1 | Bilateral VV | 8 (200) | |
| S06 | C3 - R, L | 70 | M | C (NHL) | 1.83 | 98.88 | 29.6 | Persistent edema & numbness | Endovenous RF ablation of GSV | 8 (200) |
| S07 | C3 - R, L | 63 | F | C (NHL) | 1.75 | 131.54 | 42.8 | Persistent ankle & foot edema | Tarsal tunnel stretch | 8 (200) |
| S08 | C4 - R, L | 38 | F | C (NHL) | 1.57 | 86.18 | 34.8 | Bilateral VV with complications | -- | 8 (200) |
| S09 | C3 – L; C4 – R | 38 | F | C (NHL) | 1.68 | 124.74 | 44.4 | Bilateral VV | -- | 8 (200) |
| S10 | C3 - R, L | 50 | F | NR (NHL) | 1.52 | 100.24 | 43.2 | Venous insufficiency, leg pain, edema | -- | 6 (150) |
| S11 | C3 – L; C4 – R | 53 | M | C (NHL) | 1.85 | 122.47 | 35.6 | VV, leg pain, edema History of DVT | -- | 8 (200) |
| N01 | Normal | 38 | M | A (NHL) | 1.73 | 70.31 | 23.5 | -- | -- | 12 (300) |
| N02 | Normal | 43 | F | C (NHL) | 1.65 | 85.28 | 31.3 | -- | -- | 12 (300) |
| N03 | Normal | 30 | F | C (NHL) | 1.52 | 78.02 | 33.8 | -- | -- | 12 (300) |
| N04 | Normal | 53 | F | C (NHL) | 1.63 | 99.79 | 37.6 | -- | -- | 12 (300) |
| N05 | Normal | 47 | M | C (NHL) | 1.78 | 97.52 | 30.8 | -- | -- | 12 (300) |
| N06 | Normal | 35 | M | C (NHL) | 1.70 | 79.38 | 27.5 | -- | -- | 12 (300) |
| N07 | Normal | 58 | M | C (NHL) | 1.80 | 81.65 | 25.2 | -- | Bilateral knee surgeries | 12 (300) |
| N08 | Normal | 43 | M | C (HL) | 1.57 | 74.84 | 30.3 | -- | -- | 12 (300) |
| N09 | Normal | 57 | M | C (HL) | 1.65 | 72.57 | 26.7 | -- | -- | 12 (300) |
Abbreviations: A – Asian; AA – African American; C – Caucasian; DVT – Deep Venous Thrombosis; GSV – Great Saphenous Vein; F – Female; ICG – Indocyanine Green; HL – Hispanic/Latino; L – Left; M – Male; NHL – Non-Hispanic/Latino; NR – Not Reported; R – Right; RF – Radiofrequency; VV – Varicose Veins
Table II:
Summary of observed lymphatic phenotypes and propulsion rates for limbs with CVI.
| CEAP Classification | Subject ID - Limb | Abnormal Anatomical Observations† | Injection-Associated Backflow (Area [cm2])‡ | Proximal Dermal Backflow (Area [cm2]) | Propulsion Rate [events/min] | Comments | |||
|---|---|---|---|---|---|---|---|---|---|
| Supine | Standing | Post-Walking | Overall | ||||||
| C2 | S04-L | U; Signs of S and D | IB-F (<5.0) | N (-) | 0.99 | 1.53 | 2.15 | 1.27 | - |
| S04-R | U; Signs of S and D | IB-A (<5.0) | N (-) | 0.74 | 1.02 | 1.41 | 0.94 | Relatively poor ICG uptake from foot and ankle injections as compared to contralateral limb | |
| C3 | S02-R | R; S; D; TF | IB-A (<5.0) | N (-) | 0.82 | - | 1.05 | 0.87 | - |
| S06-L | S; D | DB-A (34.9) IB-C (<5.0) IB-F (<5.0) |
N (-) | 0.80 | 1.24 | 1.23 | 1.07 | - | |
| S06-R | S; D | DB-A (26.9) | N (-) | 1.04 | 1.94 | 1.72 | 1.34 | - | |
| S07-L | S; D | DB-A (63.0) | N (-) | 0.37 | 1.17 | 0.00 | 0.39 | DB wraps around back of ankle | |
| S07-R | S; D | DB-A (51.0) | N (-) | 0.78 | 1.76 | 1.53 | 0.93 | - | |
| S09-L | Signs of S and D | IB-A (18.5) | N (-) | 1.55 | 2.42 | 1.44 | 1.68 | Relatively poor ICG uptake as compared to contralateral limb | |
| S10-L | S; D | IB-F (<5.0) | N (-) | 0.79 | 3.74 | - | 0.94 | - | |
| S10-R | S; D | IB-A (<5.0) | N (-) | 2.40 | 3.74 | - | 2.48 | - | |
| S11-L | U; S; D | IB-F (<5.0) | N (-) | 1.42 | - | - | 1.42 | - | |
| C4 | S03-L | S; D | IB-A (<5.0) | N (-) | 0.27 | 1.91 | 2.02 | 0.90 | - |
| S03-R | U; S; D | IB-F (<5.0) | N (-) | 1.12 | 4.01 | 1.84 | 1.89 | - | |
| S05-L | DB-C (25.6) IB-F (<5.0) |
DB-F (9.2) | 0.27 | 0.00 | 0.00 | 0.21 | Limited vasculature observed above ankle injection | ||
| S05-R | S | DB-A (>5.0) IB-C (<5.0) |
N (-) | 0.31 | 0.74 | 0.98 | 0.41 | DB wraps around back of ankle; Limited vasculature observed above ankle injection |
|
| S08-L | S; D | DB-A (27.1) DB-C (48.1) |
N (-) | 0.75 | 0.46 | 0.84 | 0.73 | - | |
| S08-R | S; D | IB-A (8.4) IB-C (<5.0) |
N (-) | 0.42 | 0.46 | 0.42 | 0.43 | - | |
| S09-R | S; D | DB-F (<5.0) | DB-S (47.0) | 1.20 | 1.07 | 1.15 | 1.18 | - | |
| S11-R | U; S; D, T | IB-A (<5.0) IB-F (<5.0) |
N (-) | 0.51 | - | - | 0.51 | - | |
| C5 | S02-L | R; S; D | DB-A (9.1) | DB-I (97.4) | 0.90 | 0.00 | 1.06 | 0.92 | - |
| Normal | |||||||||
L – Left; R – Right;
D – Dilated vessel; N – None; R – Vessels radiating from injection site; S – Segmented vessel; T – Tortuous vessel; TF – Tortuous on Foot; U – Unusual drainage pattern
A – Ankle; C – Calf; DB – Dermal backflow; F – Foot; I-Inguinal; IB – Interstitial Backflow; N – None; S-Shin
Lymphatic Architecture
Normal lymphatic vessels are generally well-formed and linear in nature and present with pulsatile, unidirectional flow from injection sites towards the regional nodal basins as shown in Figure 1A–1C. Radiating vessels were observed in four (N01-R, N06-L, N06-R, and N08-R) of eighteen normal limbs. One limb (N04-L) had a tortuous vessel that drained from the medial ankle across to the top of the foot where it made a small loop before continuing up the shin in an otherwise normal appearing vessel. Signs of segmentation were observed in two limbs (N07-L and N07-R) and an unusual drainage pattern was observed in the shin of N07-R. One of eighteen limbs (N02-R) had a single injection site with a small area (<5.0 cm2) of interstitial backflow and initially exhibited poor ICG uptake.
Figure 1:
NIRFLI images, with corresponding color images inset, of a normal subject (N09, A-C) and subjects with C2 disease (D-E). Images of normal lymphatics show typical lymphatic drainage patterns in the (A) lower legs, (B) thighs and inguinal regions, and (C) lateral calf. Images of C2 disease show the unusual lymphatic drainage patterns observed in the (D) right lateral calf and (E) left medial knee of S04. Red circles indicate the NIRFLI’s field of view within the color images. Injection sites are covered with round bandages and/or black vinyl tape.
Both C2 limbs presented with well-formed lymphatic vessels and active lymphatic propulsion, however, both legs presented with unusual lymphatic drainage patterns in both lateral calves and the left medial knee (Figures 1D–1E) and appeared to show early signs of vessel dilation and segmentation. In addition, both legs presented with a small area (<5.0 cm2) of interstitial backflow near an injection site, although, dermal backflow was not observed in either limb. While all nine C3 legs presented with increased dilation in some vessels, the most distinguishing abnormal anatomical feature was the distinct segmentation of lymphatic vessels as shown in Figure 2. Figure 2A shows the lower legs of S02 who presented with C3 and C5 disease on the right and left legs, respectively. Radiating vessels were observed in one (S02-R) of nine legs. S11-L had an unusual drainage pattern in the shin (Figure 2C). Interstitial backflow was observed at seven injection sites in six of nine legs, with six injections, three being in the foot (S06-L S10-L, and S11-L), two in the ankle (S02-R and S10-R), and one in the calf (S06-L) having an area <5.0 cm2, and only one ankle (S09-L) having more extensive interstitial backflow (18.5 cm2). Dermal backflow around the ankle injection sites (Figure 2D) was also observed in both legs of S06 and S07; however, both subjects had histories of ankle incisions (see Table II) which may contribute to the appearance and extent of dermal backflow at these injection sites. Only one subject (S06-L) presented with multiple (3) injection sites exhibiting dermal and/or interstitial backflow. Dermal backflow was not observed proximal to the injection sites in C3 limbs. One subject (S02-R) had a tortuous lymphatic vessel in the foot that actively propelled fluorescent lymph through a vessel which looped from the injection site down to the toes and back towards the ankle.
Fig 2.
Near-infrared fluorescent lymphatic imaging (NIRF-LI) studies, with corresponding color images inset, of limbs classified with C3 venous insufficiency. A, Distinct lymphatic vessel segmentation common in both C3 (right) and C5 (left) disease seen in the lower legs of S02. The arrow indicates the location of the healed venous ulcer. B, Drainage pattern in the left, medial knee draining into dilated vessels in the thigh of S06. C, An unusual drainage pattern with tortuous lymphatics draining the left upper shin of S11. D, Dermal backflow was associated with both medial ankle injections in S06. Red circles indicate the NIRF-LI field of view within the color images. Injection sites were covered with round bandages and/or black vinyl tape.
In C4 disease, the lymphatics possessed the same, and often more exaggerated, abnormal anatomical features observed in C3 disease. In these limbs, segmentation was observed in 7 of 8 legs and vessel dilation was observed in 6 of 8 legs (Figure 3A). The limbs lacking segmentation and/or dilation belonged to the same subject (S05), who had limited lymphatic drainage observed above the ankles of both legs. The lack of ICG proximal to the injection sites in this subject could be due to intradermal injection technique, as poor/delayed uptake occurs when ICG is inadvertently administered slightly below the dermis. Despite the poor ICG uptake, dermal backflow was observed around the injection sites in the left ankle and right calf, as well as proximal to the injections in the left foot. Overall, interstitial backflow and/or dermal backflow was observed around at least one injection site in all eight C4 legs, with interstitial backflow around the injection sites in three ankles (S03-L, S08-R, and S11-R), three feet (S03-R, S05-L, and S11-R), and one calf (S05-R). Dermal backflow was observed around injection sites in two ankles (S05-R and S08-L), one foot (S09-R), and two calves (S05-L and S08-L). Dermal backflow proximal to the injections sites was observed in the foot of S05-L and the upper shin of S09-R (Figure 3B). Two limbs (S03-R and S11-R) also presented with indistinct and/or unusual lymphatic drainage patterns as shown in Figures 3C and 3D.
Figure 3:
NIRFLI images, with corresponding color images inset, of limbs classified with C4 (A-D) and C5 (E-F) venous insufficiency. Images of C4 disease illustrate the (A) segmented and highly dilated lymphatics in the medial right leg of S03; (B) injection-associated dermal backflow in the medial right ankle and proximal dermal backflow in the upper shin of S04; (C) dilated, segmented, and poorly defined lymphatic vessels draining the right lateral calf of S03; and (D) unusual drainage pattern with dilated and tortuous lymphatics draining the right calf injection of S11. Images of C5 disease show (E) lymphatics draining the ankle injection around the healed wound bed (arrow) and (F) proximal dermal backflow in the upper thigh of the C5 limb (S02-L). Red circles indicate the NIRFLI’s field of view within the color images. Injection sites are covered with round bandages and/or black vinyl tape.
One limb (S02-L) presented with C5 disease and a history of venous ulceration that healed 8 months prior to imaging. The lymphatics in this limb were not as bright as in the contralateral C3 limb (S02-R, see Figure 2A) suggesting that ICG uptake from the injection sites might be impaired. The vessels were segmented and dilated with dermal backflow around the ankle injection (Figure 3E) and, most strikingly, in the inguinal region (Figure 3F), indicating that lymphatic congestion in the nodal basin may contribute to the reduced uptake from the injection sites and the worsening symptoms of CVI in advanced disease.
Quantitative Measures
Interstitial and Dermal Backflow
To assess the frequency of interstitial and dermal backflow, the number of injection sites with these abnormalities on each limb were counted and divided by the number of injections on that limb. On average, interstitial and/or dermal backflow was observed around 0.46±0.02%, 25.0±0.0%, 31.9±0.2%, 40.6±0.1% and 20% of the injection sites in the normal, C2, C3, C4, and C5 limbs respectively. While we had insufficient C2 and C5 limbs for statistical analysis, Student’s t-tests indicate that the average frequencies of interstitial/dermal backflow were not significantly different between C3 and C4 legs (p=0.1216) but were significant between the normal legs and both C3 and C4 legs (p-values <0.0005). Because the number of sites with proximal dermal backflow were limited, statistical analyses on the frequency of these abnormalities were not performed. However, it is noted that proximal dermal backflow was not observed in any C2 or C3 limbs, while two areas were observed in the C4 limbs and one area was observed in the single C5 limb. Given this limited dataset, it is likely that the frequency of proximal dermal backflow increases with disease classification as well.
Lymphatic pumping
Active lymphatic propulsion or pumping was observed in all limbs regardless of disease classification. As shown in Table II, for analytic purposes we computed the propulsion rates for each limb in the supine position and for the CVI limbs standing or pre-walking, and post-walking positions. The position rates for CVI limbs were also cumulated to produce an overall propulsion rate. Owing to the limited imaging time, approximately 2 minutes, in the standing and post-walking positions, the number of propulsion events observed limited the statistical power of these rates. However, as shown in Figure 4, the average propulsion rates were generally higher in the standing and post-walking positions as compared to the supine position for each disease classification, and paired t-tests, including all CVI limbs, indicated significant rate increases between the supine and the standing (p=0.0037) positions and the supine and post-walking (p=0.0125) positions. In the supine position, the average propulsion rates were 0.9±0.4, 0.9±0.2, 1.1±0.6, and 0.6±0.4 events/min for Normal, C2, C3, and C4 limbs respectively. While the supine rate in C4 disease was nearly half the value of Normal and C3 rates, it did not obtain statistical significance, as the p-values were 0.0609 and 0.0614 respectively. The overall average propulsion rates were 1.3±0.4 events/min, 1.2±0.7 events/min, and 0.8±0.5 events/min for C2, C3, and C4 disease respectively, and although a reduction is evident, statistical significance was not achieved between disease classifications. Because only two C2 limbs and one C5 limb, with an average rate of 0.9 events/min in both the supine position and overall, were reported, they were not included in the statistical analysis.
Figure 4:
Plot of the average lymphatic propulsion rates observed in the normal and CVI limbs as a function of position and disease classification. While the only significance difference in the intra-CEAP classification rates was between the C3 supine and standing positions (p=0.0148), paired t-tests of the position data across all CVI limbs indicated a significant difference between supine and standing (p=0.0037) and supine and post-walking (p=0.0125) but not standing and post-walking (p=0.3044). Unpaired t-tests did not find statistical significance in the rates between classes.
Discussion
This pilot study demonstrates lymphatic involvement in mild and moderate (C2-C4) venous insufficiency and demonstrates degradation of lymphatic anatomy and function with the progression of CVI and as compared to normal subjects. Abnormal lymphatic architecture, including interstitial backflow, dermal backflow, vessel segmentation, dilation, and/or unusual drainage patterns, were observed in all CVI limbs, with severity generally progressing with venous insufficiency classification. While tortuous or radiating vessels were observed in five normal limbs, only two limbs exhibited signs of segmentation, with one of those having an unusual pattern, and only one presented with interstitial backflow and delayed lymphatic uptake.
Interstitial backflow near the injection sites appears to be the earliest sign of reduced lymphatic uptake that may indicate higher, if clinically indistinguishable, interstitial fluid volumes and pressures forcing intradermally injected boluses to spread throughout the skin prior to lymphatic uptake. Microlymphangiography measures dermal lymphatic capillary backflow,13 and our results are consistent with past studies. Dermal backflow, a feature in which lymph abnormally backfills the capillary lymphatic network, has not been observed in normal subjects, but is common in pre-symptomatic and early stages of lymphedema16–18. While qualitatively assessed herein, these abnormal lymphatic findings may be associated with the clinical symptom of edema that identifies C3 disease classification and may indicate the beginning of lymphedema that exacerbates the venous symptoms of many advanced CVI patients. Whether the lymphatic insufficiency indicated by the presence of these abnormalities is causative, as suggested by Tanaka and colleagues19,20, or resultant of venous degradation, or provide evidence of truncular lymphatic malformations independent of CVI remains to be elucidated. In any event, these abnormalities, particularly backflow and vessel segmentation, are likely harbingers of subclinical lymphedema and may portend development of clinically significant lymphedema.
While the cause of lymphatic vessel segmentation has not been definitively identified, in a past study using iodinated contrast lymphangiography we observed varicose lymphatic vessels,21 and hypothesize that the appearance of segmented lymphatic vessels in NIRF-LI may result from the varying depths of varicose lymphatic vessels as they ‘corkscrew’ through the legs towards the inguinal region22. If accurate, this further suggests that lymphatic varicosities, and resulting lymphatic insufficiency, may accompany the development of venous varicosities in C2 disease and are especially prevalent in C3 disease where limb swelling is the distinguishing clinical feature. Interestingly, subject N07, who presented with bilateral signs of segmentation, had a history of multiple, bilateral knee surgeries. Whether lymphatic varicosities are a primary disease/condition or are secondary to CVI and/or other disease/interventions, needs further investigation.
Despite the C4 rate being nearly half the normal rate, the C5 rate was equivalent to the normal rate, possibly indicating an accompanying recovery of lymphatic function with improved venous function. We observed a significant increase in the propulsion rates, across all legs, when the subjects were erect, consistent with our previous observation of increased propulsion rates in healthy subjects when moved from the gravity neutral, supine position to a positon influenced by gravity, whether sitting or in a head-down-tilt position.23 In addition, the trend of reduced propulsion rates with the progression of CVI is consistent with our observation of little or no active propulsion in the limbs of subjects with histories of advanced (C6) disease.14 Increased lymphatic contractile activity has been postulated to be caused in part by the gravity-induced filling of lymphangions, increased circumferential stresses, and stimulation of lymphatic smooth muscle contraction.24 Because standing moves a large blood volume to the lower extremities resulting in increased capillary filtration into the interstitium, lymphatic pumping helps to restore blood volume and interstitial pressures25 while standing. Given that sympathetic and parasympathetic nerve systems have been shown to affect lymphatics in humans,26 autonomic control through the arterial baroflex system may be responsible for mobilizing lymphatic pumping in the lower extremities27. Whether the compensatory mechanism of the baroflex system is impaired in peripheral vascular disease remains to be more fully tested in appropriately designed studies.
Conclusion
In conclusion, this pilot study demonstrates the use of NIRF-LI to image lymphatic anatomical structure and propulsatile function in early (C2-C4) venous insufficiency. As imaged, the anatomy and function generally degrade with severity of venous insufficiency. Whether lymphatic insufficiency is causative or resultant of venous insufficiency or develops from other mechanisms remains to be elucidated in future studies.
Supplementary Material
Supplemental Movie 1: Lymphatic reflux in the leg of a subject with peripheral arterial disease.
Supplemental Figure 1: NIRFLI images, with corresponding color images inset, of the legs of S01 diagnosed with PAD. (A) Intact lymphatic vessel on the right shin that drains from an area of injection-associated dermal back in the medial ankle to another area of injection-associated dermal backflow in the medial right calf. (B) Unusual drainage pattern with tortuous lymphatics in the right lateral calf soon after injection. (C) Extensive dermal backflow throughout the entire shins of both legs as observed towards the end of imaging. Note that the intact lymphatic vessel observed in (A) is no long clearly visible. (D) Image of a vessel (arrow) with lymphatic reflux draining from the foot to the shin (See Supplemental Movie 1). The dermal backflow from the foot injections had drained into the second and third toes (double arrow) by the end of imaging. Red circles indicate the NIRFLI’s field of view within the color images. Injection sites are covered with round bandages and/or black vinyl tape.
Article Highlights.
Type of Research:
Single-center, observational cross-sectional study
Key Findings:
Near-infrared fluorescence lymphatic imaging of the legs of 10 patients with early venous insufficiency indicated increased prevalence of lymphatic abnormalities and reduced lymphatic propulsion rates with disease progression, suggesting a lymphatic component to progressive venous insufficiency.
Take Home Message:
Treating lymphatic dysfunction in early venous insufficiency could potentially slow progression and improve outcomes.
Funding:
This study was funded by the National Institutes of Health (R21 HL132598).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Early results of this study were presented at the 30th Annual Meeting of the American Venous Forum, Tucson, AZ February 20–23, 2018.
Conflicts of Interest: CEF, JCR and EMS-M are listed as inventors on patents related to NIRF-LI and may receive future financial benefit from its commercialization. CEF, JCR, EMS-M and The University of Texas Health Science Center at Houston have research-related financial interests in Lymphatic Science, Inc. The remaining authors have no financial relationships relevant to this article to disclose.
References
- 1.O’Donnell TF, Browse NL, Burnand KG, Thomas ML. The socioeconomic effects of an iliofemoral venous thrombosis. J Surg Res. 1977. May;22(5):483–8. [DOI] [PubMed] [Google Scholar]
- 2.Eklöf B, Rutherford RB, Bergan JJ, Carpentier PH, Gloviczki P, Kistner RL, et al. Revision of the CEAP classification for chronic venous disorders: consensus statement. J Vasc Surg. 2004. December;40(6):1248–52. [DOI] [PubMed] [Google Scholar]
- 3.Lurie F, Passman M, Meisner M, Dalsing M, Masuda E, Welch H, et al. CEAP classification system and reporting standard, revision 2020. Journal of Vascular Surgery: Venous and Lymphatic Disorders [Internet]. 2020. February 27 [cited 2020 Apr 10];0(0). Available from: https://www.jvsvenous.org/article/S2213-333X(20)30063-9/abstract [DOI] [PubMed] [Google Scholar]
- 4.Beebe-Dimmer JL, Pfeifer JR, Engle JS, Schottenfeld D. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol. 2005. March;15(3):175–84. [DOI] [PubMed] [Google Scholar]
- 5.Eberhardt Robert T, Raffetto Joseph D Chronic Venous Insufficiency. Circulation. 2014. July 22;130(4):333–46. [DOI] [PubMed] [Google Scholar]
- 6.Partsch H, Lee B. Phlebology and lymphology--a family affair. Phlebology. 2014. December;29(10):645–7. [DOI] [PubMed] [Google Scholar]
- 7.Levick JR, Michel CC. Microvascular fluid exchange and the revised Starling principle. Cardiovasc Res. 2010. July 15;87(2):198–210. [DOI] [PubMed] [Google Scholar]
- 8.Sevick-Muraca EM, Sharma R, Rasmussen JC, Marshall MV, Wendt JA, Pham HQ, et al. Imaging of lymph flow in breast cancer patients after microdose administration of a near-infrared fluorophore: feasibility study. Radiology. 2008. March;246(3):734–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Rasmussen JC, Tan IC, Marshall MV, Adams KE, Kwon S, Fife CE, et al. Human lymphatic architecture and dynamic transport imaged using near-infrared fluorescence. Translational Oncology. 2010;3(6):362–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sevick-Muraca EM. Translation of Near-Infrared Fluorescence Imaging Technologies: Emerging Clinical Applications. Annual Review of Medicine. 2012;63(1):217–31. [DOI] [PubMed] [Google Scholar]
- 11.Mihara M, Hara H, Narushima M, Todokoro T, Iida T, Ohtsu H, et al. Indocyanine green lymphography is superior to lymphoscintigraphy in imaging diagnosis of secondary lymphedema of the lower limbs. J Vasc Surg Venous Lymphat Disord. 2013. April;1(2):194–201. [DOI] [PubMed] [Google Scholar]
- 12.Mihara M, Hara H, Araki J, Kikuchi K, Narushima M, Yamamoto T, et al. Indocyanine Green (ICG) Lymphography Is Superior to Lymphoscintigraphy for Diagnostic Imaging of Early Lymphedema of the Upper Limbs. PLoS ONE. 2012;7(6):e38182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Franzeck UK, Haselbach P, Speiser D, Bollinger A. Microangiopathy of cutaneous blood and lymphatic capillaries in chronic venous insufficiency (CVI). Yale J Biol Med. 1993;66(1):37–46. [PMC free article] [PubMed] [Google Scholar]
- 14.Rasmussen JC, Aldrich MB, Tan I-C, Darne C, Zhu B, O’Donnell TF, et al. Lymphatic transport in patients with chronic venous insufficiency and venous leg ulcers following sequential pneumatic compression. Journal of Vascular Surgery: Venous and Lymphatic Disorders. 2016. January 1;4(1):9–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rasmussen J, Kwon S, Sevick-Muraca E, Cormier J. The Role of Lymphatics in Cancer as Assessed by Near-Infrared Fluorescence Imaging. Annals of Biomedical Engineering. 2012;40(2):408–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Yamamoto T, Yamamoto N, Doi K, Oshima A, Yoshimatsu H, Todokoro T, et al. Indocyanine Green Enhanced Lymphography for Upper Extremity Lymphedema: A Novel Severity Staging System Using Dermal Backflow Patterns. Plastic and Reconstructive Surgery. 2011;128(4):941–947 10.1097/PRS.0b013e3182268cd9. [DOI] [PubMed] [Google Scholar]
- 17.Rasmussen JC, Tan I-C, Naqvi S, Aldrich MB, Maus EA, Blanco AI, et al. Longitudinal monitoring of the head and neck lymphatics in response to surgery and radiation. Head Neck. 2017. June;39(6):1177–88. [DOI] [PubMed] [Google Scholar]
- 18.Yoon JA, Shin MJ, Choi YJ, Kim JH, Kang T, Park H. Early Stage Lymphedema in Breast Cancer Patient Detected by Indocyanine Green Lymphography but not by Lymphoscintigraphy: A Case Report. Journal of Breast Disease. 2019;7(2):117–20. [Google Scholar]
- 19.Tanaka H, Zaima N, Sasaki T, Yamamoto N, Sano M, Konno H, et al. Loss of lymphatic vessels and regional lipid accumulation is associated with great saphenous vein incompetence. J Vasc Surg. 2012. May;55(5):1440–8. [DOI] [PubMed] [Google Scholar]
- 20.Tanaka H, Yamamoto N, Suzuki M, Mano Y, Sano M, Zaima N, et al. Insufficient Lymph Drainage Causes Abnormal Lipid Accumulation and Vein Wall Degeneration. Ann Vasc Dis. 2016/December/01 ed. 2016;9(4):277–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Burrows PE, Gonzalez-Garay ML, Rasmussen JC, Aldrich MB, Guilliod R, Maus EA, et al. Lymphatic abnormalities are associated with RASA1 gene mutations in mouse and man. Proceedings of the National Academy of Sciences. 2013. May 21;110(21):8621–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Rasmussen JC, Zvavanjanja RC, Aldrich MB, Greives MR, Sevick-Muraca EM. Near-infrared fluorescence lymphatic imaging of Klippel-Trénaunay syndrome. J Vasc Surg Venous Lymphat Disord. 2017. July;5(4):533–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Rasmussen JC, Kwon S, Pinal A, Bareis A, Velasquez FC, Janssen CF, et al. Assessing lymphatic route of CSF outflow and peripheral lymphatic contractile activity during head-down tilt using near-infrared fluorescence imaging. Physiological Reports. 2020;8(4):e14375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zawieja DC. Contractile Physiology of Lymphatics. Lymphatic Research and Biology. 2009;7(2):87–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Huxley VH, Scallan J. Lymphatic fluid: exchange mechanisms and regulation. J Physiol (Lond). 2011. June 15;589(Pt 12):2935–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mignini F, Sabbatini M, Coppola L, Cavallotti C. Analysis of nerve supply pattern in human lymphatic vessels of young and old men. Lymphat Res Biol. 2012. December;10(4):189–97. [DOI] [PubMed] [Google Scholar]
- 27.Hedrick MS, McNew KA, Crossley DA. Baroreflex function in anurans from different environments. Comp Biochem Physiol, Part A Mol Integr Physiol. 2015. January;179:144–8. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Movie 1: Lymphatic reflux in the leg of a subject with peripheral arterial disease.
Supplemental Figure 1: NIRFLI images, with corresponding color images inset, of the legs of S01 diagnosed with PAD. (A) Intact lymphatic vessel on the right shin that drains from an area of injection-associated dermal back in the medial ankle to another area of injection-associated dermal backflow in the medial right calf. (B) Unusual drainage pattern with tortuous lymphatics in the right lateral calf soon after injection. (C) Extensive dermal backflow throughout the entire shins of both legs as observed towards the end of imaging. Note that the intact lymphatic vessel observed in (A) is no long clearly visible. (D) Image of a vessel (arrow) with lymphatic reflux draining from the foot to the shin (See Supplemental Movie 1). The dermal backflow from the foot injections had drained into the second and third toes (double arrow) by the end of imaging. Red circles indicate the NIRFLI’s field of view within the color images. Injection sites are covered with round bandages and/or black vinyl tape.