If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: Australian Lymphoedema Education, Research and Treatment (ALERT), Faculty of Medicine, Health and Human Sciences, Level 1, 75 Talavera Rd, Macquarie University, NSW 2109, Australia.
Australian Lymphoedema Education, Research and Treatment (ALERT), Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, AustraliaMt Wilga Private Hospital, Hornsby, NSW, Australia
Australian Lymphoedema Education, Research and Treatment (ALERT), Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, AustraliaICON Cancer Centre, NSW, Sydney Adventist Hospital, Sydney, NSW, Australia
The lower limbs are a common body site affected by chronic edema. Imaging examination of the lymphatic system is useful to diagnose lymphoedema, identify structural changes in individuals, and guide interventional strategies. In this study, we used a protocol combining indocyanine green (ICG) lymphography and ICG-guided manual lymphatic drainage (MLD) for the diagnostic assessment of lower limb lymphoedema.
Materials and methods
Patients with lower limb lymphoedema were divided into three groups by their medical history: primary, secondary cancer-related, or secondary non-cancer-related. ICG lymphography was conducted in three phases: initial observation, MLD to accelerate ICG dye transit and reduce imaging time, and imaging data collection. Lymphatic drainage regions were recorded, and the MD Anderson Cancer Center ICG staging was applied. We collected routine lymphoedema assessment data, including limb volume and bioimpedance spectroscopy measurements.
Three hundred and twenty-six lower limbs that underwent ICG lymphography were analyzed. Eight drainage regions were identified. The ipsilateral inguinal and popliteal were recognized as the original regions, and the remaining six regions were considered compensatory regions that occur only in lymphoedema. More than half of the secondary cancer-related lower limb lymphoedema (57.6%) continued to drain to the ipsilateral inguinal region. The incidence of drainage to the ipsilateral inguinal region was even higher for the primary (82.8%) and secondary non-cancer-related (87.1%) groups. Significant associations were observed between cancer-related lymphoedema and the presence of compensatory drainage regions.
We proposed a prospective ICG lymphography protocol for the diagnostic assessment of lower limb lymphoedema in combination with MLD. Eight drainage regions were identified, including two original and six compensatory regions.
Lower limb lymphoedema is a debilitating condition characterized by increased volume, skin induration, gait disturbance, and recurrent cellulitis which may lead to a physical, psychosocial, and financial burden for individuals.
Lower limb lymphoedema is conventionally classified into two types depending on the patient's medical history: primary lymphoedema with an unknown physical cause and secondary lymphoedema with a known cause.
Secondary lymphoedema can be further subclassified into two additional types: cancer-related and non-cancer-related lymphoedema. Secondary cancer-related lower limb lymphoedema develops commonly after gynecological, urological, or skin cancer treatment associated with inguinal and/or intrapelvic lymph node dissection and/or irradiation.
Secondary non-cancer-related lower limb lymphoedema is caused by traumatic injury, including following reconstructive leg surgery and resection of benign tumors, or parasitic infection, such as filariasis.
The lower limbs are a common body site affected by chronic edema or swelling because of the effects of gravity. Therefore, lower limb lymphoedema should be distinguished from other diseases causing edema or swelling; venous insufficiencies including May-Thurner syndrome and deep vein thrombosis, dependent edema, chronic heart failure, and lipedema.
Given that various diseases result in increasing lower limb volume, lymphatic imaging examination can assist in the accurate differential diagnosis of lymphoedema. Lymphoscintigraphy is the current gold standard lymphatic imaging technique to diagnose lymphoedema.
Although lymphoscintigraphy has been widely available for the diagnostic imaging of lower limb lymphoedema, the viscous radioactive tracer often takes a long time to reach the lymph nodes in the affected limb, thereby increasing imaging time.
Additionally, there is no standardized protocol for lymphoscintigraphy and differences in the choice of radioactive tracer, sites of tracer injection, and duration of study all affect diagnostic outcomes.
Indocyanine green (ICG) lymphography has gained popularity as an imaging device for lymphovenous anastomosis (LVA) as it can map lymphatic vessels in real time.
Despite the increasing various uses of ICG lymphography for lymphoedema, no standard protocol has been developed for the lower limb. Therefore, in a previous study, we developed a cadaver model to determine the most appropriate ICG injection sites and number of injections to comprehensively assess the lower limb lymphatic system.
We elucidated that four circumferential injection sites around the foot are recommended because the lymphatics in the lower leg are divided into four distinct lymphatic pathways: anteromedial, anterolateral, posteromedial, and posterolateral.
The ICG lymphography protocol to assess upper limb lymphoedema was previously developed by the authors.
ICG lymphography was undertaken simultaneously with ICG-guided manual lymphatic drainage (MLD) performed by an accredited lymphoedema therapist. We found that MLD facilitated dye movement more efficiently compared to post-injection exercise and delayed scanning.
Continuous lymphatic imaging of the upper limb enabled us to specify drainage region/s and dermal backflow areas in the affected limb with the procedure taking approximately 1 h. The imaging results assisted the selection of surgical options and helped to guide the application of MLD for conservative management for the individual patient.
In this study, we initiated the same approach of a combination of ICG lymphography and MLD to develop a new ICG lymphography protocol for the diagnostic assessment of lower limb lymphoedema.
Materials and methods
A retrospective cohort audit was conducted by reviewing prospectively collected data from 376 patients reporting lower limb swelling who attended the ALERT clinic at Macquarie University between August 2017 and October 2020. Ninety-eight patients diagnosed with venous edema or lipedema by medical history and physical examination after the initial consultation and/or who did not have evidence of lymphoedema on ICG lymphography were excluded from this study. However, 351 lower limbs from the remaining 278 patients were examined by ICG lymphography. Twenty-five limbs (6.6%) with undetermined drainage regions were excluded from the imaging data analysis, and the remaining 326 lower limbs were included in this study. The patients were divided into three groups by their medical history: primary, secondary cancer-related, and secondary non-cancer-related. In the secondary non-cancer-related group, the evident episode triggering lymphoedema in the affected limb was known, but cancer treatment, including radiation, lymph node dissection, and chemotherapy, was not involved. We simultaneously collected routine lymphoedema assessment data including limb volume and bioimpedance spectroscopy (BIS) measurements of the lower limbs.
ICG lymphography imaging protocol
ICG (Verdye 25 mg; Diagnostic Green GmbH, Aschheim-Dornach, Germany) was mixed with 10 ml of saline. Four injection sites were used in the foot on the affected side: first web space, below medial malleoli, midpoint between below lateral malleoli and the Achilles tendon, and mid-point between the fifth metatarsophalangeal joint and below lateral malleoli (Supplemental Figure 1). These circumferential four injection sites were selected based on our previous results in anatomical studies of the lymphatics in a cadaveric model.
Intradermal injections were performed with a 30-gage needle and a 1 ml syringe. At each injection site, 0.05–0.1 ml (0.125–0.25 mg) of ICG solution was administered. A cryogenic numbing device (CoolSense; CoolSense Medical Ltd., Tel Aviv, Israel) was applied for 3 s, immediately before each injection to reduce needle discomfort.
Lymphatic scanning using the near-infrared camera system (PDE Neo II; Hamamatsu Photonics K.K., Hamamatsu, Japan) commenced immediately after the injections, and imaging data were recorded using a digital video recorder (MDR-600HD: Ikegami Tsushinki Co., Ltd., Tokyo, Japan). Lymphatic imaging of the lower limb was continuously conducted for approximately 1 h.
Lymphatic scanning was conducted in three phases. Throughout the scanning, lymphatic vessels, direction of ICG movement, and demarcation of dermal backflow were marked on the patient's skin using colored pens (Figure 1, right).
Observation of any spontaneous movement of ICG via the lymphatics for approximately 10 min. Gentle pressure was applied at the injection sites if ICG did not enter the lymphatics. Phase one identified the remaining lymphatic vessels in the dorsum of the foot.
Scanning continued while ICG-guided MLD was performed to facilitate ICG transit via the lymphatics particularly through the areas of dermal backflow. The visualized movement of ICG tracer informed the pressure, speed, and direction for the MLD. In contrast to conventional MLD, ICG-guided MLD is applied in a distal to proximal direction and follows the ICG extension to the drainage region. We have previously demonstrated that while the application of ICG-guided MLD reduces scanning time, delayed imaging 24 h post-injection did not alter the investigation findings.
Continuous scanning enabled the operator to identify lymphatic vessels before being masked by dermal backflow, as well as the direction of ICG movement, demarcation of dermal backflow, and drainage regions (original and/or compensatory). The patient moved from supine to lateral and prone positions. Phase two continued until the dissemination of ICG stabilized, usually between 30 and 45 min.
The collection of imaging data through still photography with both near-infrared and digital cameras was taken in the supine and prone positions (Figure 1). Phase three took 15 min.
Imaging data analysis
Lymphatic images were analyzed for (i) lymphatic vessels which are linear structures in the subcutaneous tissue, (ii) dermal backflow which is reflux of lymph fluid into the dermal lymphatics and is diagnostic criterion of lymphoedema, and (iii) lymphatic drainage regions. These regions were determined by the location of identified lymph nodes or extension of ICG via dermal backflow or lymphatic vessels to the skin regions over known lymph nodes or perforating lymphatic vessels. Captured images were interpreted by a single evaluator (HS) to avoid interobserver variability.
Lymphoedematous lower limbs were also classified by the MDACC ICG staging scale that was originally developed to assess upper limb lymphoedema severity.
Stage 0: normal lymphatics, Stage 1: many patent lymphatic vessels with minimal patchy dermal backflow, Stage 2: moderate number of patent lymphatic vessels with segmental dermal backflow, Stage 3: few patent lymphatic vessels with extensive dermal backflow involving the entire leg, Stage 4: no patent lymphatic vessels seen with dermal backflow involving the entire leg extending to the dorsum of the foot, and Stage 5: ICG does not move from injection sites (Supplemental Figure 2). Correspondence of the ICG lower limb staging with other measures is under validation.
All scanning procedures were performed by one technician (HS) following our previously published standardized protocol in the upper limb.
All therapists involved in performing ICG-guided MLD were trained in the techniques and assessed for competency against our standardized criteria.
Limb volume and extracellular fluid measures
Limb volume was determined by Perometry (350NT Model, Pero-System, Wuppertal, Germany) or circumferential limb measurements were taken at 4 cm intervals using a tape measure and calculated using the truncated cone formula.
Percentage volume difference was calculated only for unilateral lymphoedema, by comparing the affected limb with the unaffected limb. BIS was used to measure extracellular fluid in the affected limb as a ratio compared to the unaffected limb for unilateral lymphoedema and the ipsilateral arm for bilateral lymphoedema. BIS measurements were recorded in l-Dex units (normal range is −10 to +10).
Two BIS devices were used (L-Dex® U400 or SOZO®: ImpediMed, Brisbane, Australia) following standard procedures.
Statistical analysis was conducted using the software package IBM SPSS Statistics for Windows (Version 26; IBM Corp., NY, USA). The alpha level of significance was set at p ≤ 0.05. Data were tested for normality with the Shapiro–Wilk test. As data did not have a normal distribution, Kruskal–Wallis ANOVA was used to compare the number of drainage regions between the three groups, and pairwise comparisons were performed with a Bonferroni correction to determine which groups were different. Data are presented as median and interquartile range (IQR). Pearson chi-square and odds ratio were used to examine associations between lymphoedema group and the presence of compensatory drainage regions. Spearman correlation was performed to investigate associations between the ICG stage and limb volume and ICG stage and BIS.
The patient's characteristics are summarized in Table 1. Eight distinct drainage regions were identified in lower limb lymphoedema, and they were categorized into two groups by considering their anatomical characteristics: two original and six compensatory regions (Figure 2). Original lymphatic drainage occurred to the ipsilateral inguinal and popliteal regions. Compensatory drainage regions were the contralateral inguinal, posterior thigh, gluteal, upper lateral thigh, lower abdominal, and axilla or parasternal regions (Table 2). Drainage to the axilla or parasternal regions was found in nine out of 99 lower limbs (9.1%) in the secondary cancer-related group. Only one out of 134 (0.7%) in the primary group demonstrated drainage to the parasternal region.
Moderate to strong positive associations were identified between the MDACC ICG stage and limb volume (r = 0.68, p < 0.001) and the ICG stage and BIS (r = 0.58, p < 0.001). Three key differences were identified between the secondary cancer-related group and primary and secondary non-cancer-related groups. Firstly, significant associations were observed between cancer-related lymphoedema and the presence of the compensatory drainage pathways (Χ2 = 63.863(1), p<0.001), and the odds ratio revealed that individuals with cancer-related lymphoedema were 7.83 times more likely to have compensatory pathways than individuals with lymphoedema due to causes other than cancer. Secondly, the prevalence of drainage to the ipsilateral inguinal drainage region was higher in secondary non-cancer-related (87.1%) and primary (82.8%) groups than in the cancer-related group (57.6%) (Table 2). Finally, the cancer-related group demonstrated a higher likelihood of drainage to multiple drainage regions (Med = 2, IQR = 1 - 2) compared to the secondary non-cancer-related (Med = 1, IQR = 1 – 2, p = 0.002) and primary groups (Med = 1, IQR = 1 – 2, p = 0.001).
As the MDACC stage increased due to increased lymphatic dysfunction, there was less drainage to the ipsilateral inguinal and popliteal nodes, and more compensatory regions were demonstrated for all three groups. For example, in the MDACC Stage 4 in each group, drainage to the compensatory regions occurred in the secondary cancer-related (67.6%), secondary non-cancer-related (100%), and primary groups (62.1%) (Table 2).
Our new protocol combining ICG lymphography and ICG-guided MLD can provide a diagnostic assessment of lower limb lymphoedema based on the presence of dermal backflow. Additionally, it can be used to stage lower limb lymphoedema severity and identify the drainage regions for each affected limb. Compared to lymphoscintigraphy, ICG lymphography has the advantage of being able to be combined with MLD to facilitate the tracer transit without the lymphoedema therapist being exposed to radiation. This has the potential to reduce examination time and enhance visualization of lymphatic drainage regions.
The use of circumferential four injection sites in the foot, based on prior studies demonstrating four normal lymphatic pathways of the lower limb,
enables assessment of the patency of each of these pathways and provides a comprehensive assessment of lower limb lymphoedema. Lymphoscintigraphy, where one or two injections are applied in the foot, may only demonstrate the anteromedial pathway. As this pathway is often compromised in lymphoedema, this can lead to the radioactive tracer remaining in the foot losing the opportunity to visualize the compensatory drainage patterns more proximally in the affected limb.
Identification of compensatory drainage regions in lower limb lymphoedema was one of the notable findings of this study (Figures 1 and 2, Table 2). Imaging information about the drainage regions could contribute to an understanding of the pathophysiology of lymphoedema in an individual. If the original drainage regions of the lower limb are partially or completely obstructed, the lymphatic system may form compensatory drainage to other regions to bypass this obstruction. Compensatory drainage via the superficial lymphatic system to residual lymph nodes, such as the contralateral inguinal, lower abdomen, and axilla, was observed in this study. These findings are supported by our previous studies in a canine model that demonstrated that lymphatic vessels severed by lymph node dissection were able to connect to residual lymph nodes.
Our study also identified drainage regions that we postulate connect to the deep lymphatic system via perforating lymphatic vessels. Drainage regions were identified in the gluteal, lateral upper thigh, and posterior thigh. Dermal backflow was observed to directionally extend to these areas and not beyond, suggesting drainage to deeper lymphatic structures. A vertical vessel connecting the superficial and deep lymphatic vessels in lymphoedema after lymph node dissection which was not observed in normal condition has been reported with lymphangiography and magnetic resonance lymphography.
These results demonstrated the unusual lymphatic drainage to the intrapelvic lymph node via dermal backflow in the gluteal region. The cross-sectional SPECT/CT image of this study identified a radioactive signal beneath the gluteus maxima to demonstrate the connection between the superficial and deep lymphatics (Figure 3). These findings suggest that lymph fluid in the superficial lymphatic system could drain to the deep lymphatics via a perforating lymphatic vessel in lower limb lymphoedema. An anatomic study performed by Rouvière depicted the perforating lymphatic vessels along branches of the gluteal arteries (Supplemental Figure 3).
Our study results also demonstrated that dermal backflow extended to the gluteal region and then disappeared in 23.2% of cancer-related lymphoedema cases (Figure 1, Table 2). Most patients with drainage to the gluteal region were the MDACC ICG Stages 3 and 4, and this suggested that compensatory drainage toward the gluteal region was more likely to occur in advanced lymphoedema cases. These findings speculate that in advanced Stages 3 and 4 lymphoedema, lower limb drainage through the intrapelvic region via the inguinal region may be blocked by oncologic surgery and/or radiation. Subsequent compensatory drainage develops, whereby dermal backflow extends from the thigh to the gluteal region, where perforating lymphatic vessels along the gluteal arteries can drain lymph to the lumbar lymph nodes bypassing the intrapelvic pathway (Figure 4). The other compensatory drainage regions: lateral upper thigh and posterior thigh regions are also common sites to identify the perforating vessels of the descending femoral artery and the deep femoral artery, respectively.
However, the drainage pathways in the affected limb have not been routinely assessed. Our new ICG lymphography protocol can identify which lymphatic vessels functionally drain lymphatic fluid toward drainage regions. This information helps to select which vessels are suitable for LVA and which vessels should be protected.
MLD is an important component of complex decongestive therapy (CDT). The rationale of MLD is to shift extracellular lymph fluid manually toward the intact lymph node drainage regions.
The ipsilateral axilla is a routine drainage region targeted by therapists treating lower limb lymphoedema. However, our results indicate that lower limb lymphoedema rarely drains to the axillary region (3.1%) and is more likely to drain to the contralateral inguinal region (14.4%) or other regions within the affected limb (Table 2). Furthermore, certain drainage regions were more commonly observed in the different groups of lymphoedema. Primary lymphoedema and secondary non-cancer-related lymphoedema were more likely to drain to the original drainage regions, while secondary cancer-related lymphoedema was associated with multiple compensatory drainage regions. Understanding the frequency of drainage regions in different lymphoedema groups means that valuable therapy time can be spent directing MLD on the affected limb to more likely drainage regions, improving patient outcomes. Finally, consideration of severity in patients with lower limb lymphoedema could also assist in making decisions about which lymphatic drainage regions to direct MLD (Table 2).
Our results have demonstrated that patients, in all three groups, with milder lymphoedema were more likely to drain to the original drainage regions, while those with increased severity utilized compensatory drainage consistent with our findings in upper limb lymphoedema. For example, in patients with MDACC Stage 1, 94.4% and 0% drained to the ipsilateral and contralateral axillary nodes, respectively, compared to 57.8% and 15.6% for MDACC Stage 4.
However, ICG has advantages of fast tracer movement to reduce examination and constant imaging to examine lymphatic drainage from the limb in real time. Additional limitations of ICG lymphography include the requirement of dark room space at the clinic, allocation of two expert staff members who understand the procedure, training of lymphoedema therapists for performing competent ICG-guided MLD, and knowledge of normal lymphatic anatomy and how it changes in lymphoedema for interpretating the images. ALERT is a multidisciplinary program for helping predominantly lymphoedema patients, and ICG lymphography has been a part of our clinical services. This ICG imaging protocol may be a little challenging in terms of human resources and expertise; however, our new ICG protocol would be a considerable option where plastic surgeons and lymphoedema therapists are treating lymphoedema patients within the same institution.
We developed a new protocol combining ICG lymphography and ICG-guided MLD for diagnostic assessment of lower limb lymphoedema. Eight distinct drainage regions were identified, including two original regions and six compensatory regions. The MDACC ICG stage and location and number of drainage regions were considered as reliable parameters to represent lymphoedema status and enhance understanding of the pathophysiology in each patient. The obtained imaging information can help in selecting lymphatic vessels for LVA surgery and inform ICG-guided personalized MLD for conservative management.
The authors thank Emma Moloney, Katrina Gaitatzis, Catherine Riley, and Lori Lewis, research officers at ALERT for their tremendous help to collect the patients’ demographic and physical data and coordinate the institutional ethic approval process.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of Competing Interest
The authors declare that they have no conflicts of interest.
This study was approved by the institutional ethic committee (MQCIAC2018017A), and informed consent was obtained from all patients.
Supplemental Video. ICG lymphography procedure in the patient with left lower limb lymphoedema in Figure 1. Compensatory drainage regions were identified toward the contralateral inguinal and gluteal regions (Figure 2 F&G).
O Brien P.J.
Lymphedema prevalence and treatment benefits in cancer: impact of a therapeutic intervention on health outcomes and costs.