The lymphatic system plays several varied, important roles in the function of the human body. On a daily basis, it takes up several liters of fluid that extravasates from the capillaries into the interstitial space and returns it to the circulation [1]. Lymphatic vasculature plays a key role in the immune system, transporting antigens and antigen-presenting cells to lymph nodes, resulting in detection and response to threats [1, 2]. Lacteals, a set of lymphatic vessels in the microvilli of the intestine, take up ingested lipids that cannot enter the bloodstream directly, providing a path for them to ultimately enter circulation [3, 4]. Due to their vital role as a trafficking system, the lymphatics can cause or exacerbate several debilitating diseases when a physiological irregularity arises in the body. For instance, lymphedema, a painful condition resulting in swelling of the arms and legs, is directly caused by an inability of the lymphatic vasculature to adequately drain interstitial fluid [5]. Lymphatic vasculature, in addition to the cardiovascular system, also provides routes for cancer to metastasize, a critical event resulting in much higher mortality rates for patients [1, 6].

Despite the significance of the lymphatic system in both day-to-day function and disease progression, knowledge regarding its inner workings has historically lagged behind that of its sanguine counterpart. This disconnect is perhaps most striking when considering soon-to-be clinicians’ level of exposure to the topic during their training; a 2004 article by Rockson et al. reported that the average US medical student received less than 30 min of instruction regarding the lymphatic system during the course of their studies [7]. In recent decades, a surge of academic interest in the lymphatics has helped fill this knowledge gap. Interest in lymphatic imaging has helped lead the way, with over 27,000 new entries into the PubMed database since the turn of the millennium (Fig. 1). Despite this recent interest, lymphatic imaging remains as a relatively small proportion of all conducted lymphatic research. Additional work on the imaging front needs to be done to better understand the lymphatic system and its role in various disease states. In this review, we discuss several current and developing lymphatic imaging approaches for assessment of lymphatic function and development, as well as imaging methods used for diagnosing, staging, and planning the treatment of associated diseases like cancer and lymphedema. We also discuss the current limitations of some of these approaches and areas of improvement for future techniques to ultimately improve patient outcomes.

Academic Interest in lymphatics and lymphatic imaging from 1945 to 2021 in the PubMed Database. A) New entries for “lymphatic” (plotted in grey) and “lymphatic imaging” (plotted in black)

The most common method for observing the lymphatics, particularly drainage patterns, is lymphoscintigraphy. In lymphoscintigraphy, a radiotracer (most commonly 99mTc-sulfur colloid) is injected intradermally into the affected regions, typically the arms or the legs. The colloid emits gamma rays which can be detected using a gamma camera, which uses the detected radiation to produce an image [8]. Several other techniques similar to lymphoscintigraphy also exist, including near-infrared fluorescence (NIRF) lymphangiography and MRI lymphangiography. Additional information about some of these techniques is included in the “Imaging Techniques for Lymphedema Diagnosis and Treatment” section.

Visualization of the truncal lymphatics specifically can also leverage one of the intrinsic functions of the lymphatic system. The lymphatics play a key role in transporting re-packaged dietary lipids in the form of chylomicrons from the small intestine to the heart, where they enter systemic circulation [9]. The use of radioactive lipid analogs have been administered via oral gavage in animal models to better observe mesenteric and truncal lymphatics for varied applications, such as studying ovarian cancer [4].

The use of modern imaging techniques, such as intravital microscopy (particularly using multiphoton microscopy), have recently helped shed additional light on the functionality of the lymphatic system. Multiphoton microscopy involves using a pulsed laser to shoot two or more lower-energy photons near-simultaneously at a fluorophore, exciting it to a higher-energy state. The released photon following the transition of the fluorophore to its ground state is detected and used to form the microscopy images. This technique allows for high-resolution real-time imaging at greater depths, making it ideal for intravital imaging applications [10].

The use of multiphoton imaging has been especially beneficial for observing the role of the lymphatics in the immune response. Several studies have demonstrated important lymphatic phenomena occurring in vivo, such as the intralymphatic “crawling” mechanism mediating the movement of T cells from inflammation sites to draining nodes and macrophage presentation of antigens to B cells [11, 12]. Steven et al. used suture placement in BALB/c mice corneas to induce neovascularization and inflammation and subsequently observed migration of immune cells such as T cells and macrophages into lymphatic vasculature [13]. Another intravital study using a similar corneal implant suture methodology on transgenic mice expressing green fluorescent protein (GFP) under the Prox-1 promoter sought to observe more fundamental lymphatic processes. In this study, researchers were able to observe lymphatic angiogenesis and valvulogenesis, which they noted originated in existing limbal vessels. Additionally, they were able to observe that lymphatic elongation was a result of stalk cell migration, a phenomenon that is impossible to detect using traditional methods. These observations all offer important glimpses into lymphatic propagation, particularly in inflammatory/diseased states which could ultimately prove useful in therapeutic approaches with further study [14].

These intravital techniques have also lent themselves to enhancing our foundational understanding of lymphatic anatomy. Previously, it had long been thought that the central nervous system lacked lymphatic vessels. However, following drainage comparison studies using Qdot655 (a fluorescent nanoparticle quantum dot) in the meninges of mice, Louveau et al. were able to conclude that non-cardiac vessels lining the dural sinuses did in fact drain the cerebrospinal fluid (Fig. 2). These vessels were subsequently affirmed to be lymphatic channels based on histologic analysis and identification of various lymphatic biomarkers. Immunohistochemical analysis also revealed that these vessels could carry leukocytes such as T lymphocytes and MHCII+ cells [15].

Initial lymphatic features of meningeal lymphatic vessels. a Representative images of CCL21 and Lyve-1 labeling of the meningeal lymphatic vessels (scale bars, 10 μm). b, c Representative images of VE-Cadherin and Lyve-1 staining on meningeal blood vessels (b) and meningeal lymphatic vessels (c), arrowheads point to the VE-Cadherin aggregates; scale bars, 10 μm. (Adapted with permission from Louveau et al., 2015, ©2015, Nature published by Springer Nature)

A subsequent 2018 study by Louveau et al. further investigated these newly discovered lymphatic vessels. Using transgenic mice models, authors were able to demonstrate that the lymphatic vasculature played a key role in T cell drainage. In the same study, the authors also conducted an investigation using experimental autoimmune encephalomyelitis (EAE) in mice as a model for multiple sclerosis (MS) [16]. When they surgically ablated the lymphatic vessels, the authors found a reduction in the EAE pathology (Fig. 3). These results indicated that these meningeal lymphatics played an important role in the activation of encephalitogenic T cells, thereby playing a key role in regulating neuroinflammatory responses and subsequently serving as a target for future therapeutic interventions [17]. The significance of works investigating the meningeal lymphatics in mice models was further amplified when Absinta et al. were able to image meningeal lymphatic vessels in marmoset monkeys as well as humans using MRI with gadolinium-based contrast agents. More specifically, they were able to determine that the topographies of these visualized vessels matched those described in the prior mouse studies [18]. This breakthrough observation via noninvasive methods opens the door to further study of CNS lymphatics and investigation of how they might be involved in the manifestation of various neurological disorders.

a Representative images of MOG-specific T cells (2D2) and OVA-specific T cells (OTII) in the dCLNs of mice treated with laser and Visudyne (i.c.m.) + laser at day 8 after EAE induction. Green arrowheads, OTII T cells. Yellow arrowheads, 2D2 cells not in contact with CD11c+ cells; white arrowheads, 2D2 cells in contact with CD11c+ cells. Scale bar, 150 μm; insets, 25 μm. Representative of 5 independent mice per group. b Representative images and associated profile plots of MOG-specific T cells (2D2, red) in close contact (top) or not in contact (bottom) with CD11c+ cells (cyan) in the dCLNs of mice treated with laser and Visudyne (i.c.m.) + laser 8 d after EAE induction. Scale bar, 10 μm. Representative of 5 independent mice per group. (Adapted with permission from Louveau et al., 2018, ©2018, Nature Neuroscience published by Springer Nature)

Lymphedema is a chronic, debilitating disease estimated to affect approximately 200 million people around the world [19]. It can be categorized into two main types. Primary lymphedema is genetic and quite rare. Onset typically occurs during childhood or adolescence and approximately 1 in 100,000 are affected by the condition [20]. Secondary lymphedema, in contrast, is much more common and is estimated to effect 2 to 3 million patients in the USA alone [21, 22]. Also, unlike primary lymphedema, it is an acquired condition, with onset tending to occur in adulthood [5, 23]. It is the direct result of significant damage to lymphatic structures, resulting in their inability to maintain fluid homeostasis in tissues. In developing nations, the primary causes of secondary lymphedema are infections resulting in inflammation and fibrosis of lymphatic channels and nodes. This is not the case in developed nations, where the requisite lymphatic trauma is often the result of medical intervention, particularly surgical and radiation-based treatments of cancer [5, 21].

In addition to these primary types, lymphedema can be further sub-categorized into different stages. The International Society of Lymphology categorizes these stages as follows. In Stage 0 lymphedema, there are no signs readily apparent in a clinical examination, although there is still abnormal lymphatic flow and drainage. Stage 1 is characterized by mild swelling which can be mitigated by limb elevation; pitting may also be present at this stage. Stage 2 lymphedema can no longer be alleviated by limb elevation. Early phases of Stage 2 are also characterized by pitting edema, while late phases of Stage 2 may or may not involve pitting but do result in fibrosis and irreversible damage to the tissue. Stage 3 consists of lymphostatic elephantiasis, the most severe form of lymphedema. At this point, swelling is extreme, resulting in large limbs and trophic skin changes, such as leathery, and potentially warty, skin [5, 23,24,25].

Although, in many instances, lymphedema can be diagnosed through physical examination, its manifestation cannot be confirmed without the use of a more specific diagnostic test, typically one that is imaging-based [23]. This confirmation can be especially important in improving patient outcomes because lymphedema is often misdiagnosed in initial screenings. In fact, a 2011 study by Schook et al. found that approximately one-fourth of all pediatric patients referred to their clinic with “lymphedema” were suffering from a different condition [26]. A 2015 study by Maclellan et al. noted a similar prevalence of initial misdiagnosis in a pediatric and adult patient population [27]. The use of diagnostic tests can also help illustrate the pathophysiology of each patient’s lymphedema, helping stage the disease and affecting treatment choice.

The current “gold-standard” for diagnosis confirmation is lymphoscintigraphy. If the lymphatics are functioning normally, then the colloid will be taken up by lymphatic vessels and drain to downstream nodes. Lymphoscintigraphy can be used qualitatively or quantitatively to assess lymphatic function. Qualitative approaches rely on making a diagnosis based on visual assessment of lymphatic vasculature and nodes, and their ability to uptake the radiotracer. In contrast, quantitative approaches utilize the use of more defined metrics such as time of uptake and clearance of the radiotracer from the limb of interest or from the injection site [24, 28].

Lymphoscintigraphy’s status as the go-to imaging approach for lymphedema diagnosis has been well-earned due to its consistent accuracy in detecting lymphedema for decades. A 1988 study using lymphoscintigraphy to assess lymphatic function in a group of 238 patients demonstrated that qualitative analysis of the image alone allowed researchers to identify flow abnormalities in 70.1% of studied extremities and subsequent quantitative analysis allowed for identification in all extremities [29]. Similarly, a 1989 study conducted by Gloviczki et al., in which the authors performed lymphoscintigraphy on 190 patients, found that semiquantitative evaluation of the lymphatics allowed them to detect lymphedema with a sensitivity of 92% and a specificity of 100% [30]. A more recent 2017 study with 227 patients found that lymphoscintigraphy had a 96% sensitivity and 100% specificity, reaffirming and marking a slight improvement on the historical trend [31]. In 2018, a study used traditional lymphoscintigraphy to differentiate primary versus secondary lower extremity lymphedema after surgical lymphadenectomy. The retrospective analysis revealed that the appearance of lower limb lymphedema was not related to cancer therapeutic intervention but points to a primary lymphatic disease prior to treatment (Fig. 4a) [32].

a Traditional lymphoscintigraphy shows the consequences of extensive inguinoiliac lymphadenodysplasia (the inguino-iliac lymph nodes are not visible, arrow 5 on the right side with the tracer flowing into lymphatic vessels and the superficial lymphatic collateralization network up to the root of the limb. Arrows with L indicate the liver, wherein radiocolloids are taken up when they have reached the systemic circulation. (Adapted with permission from Roman et al., 2018, ©2018, World Journal of Surgical Oncology published by Springer Nature) b T1-weighted gradient echo MRI maximum intensity projection image of a 55-year-old woman with a history of ovarian cancer and pelvic lymphadenectomy who presented with right leg edema. After intracutaneous administration of a contrast agent into the web spaces of the right foot MRI imaging reveals a large lymphatic channel ascending the anterior calf and dermal backflow to the medial upper calf (arrowhead). Dermal backflow is intermittently identified with MRI in some patients with secondary lymphedema. Note the increase in spatial resolution as compared with lymphoscintigraphy. (Adapted with permission from Lee et al., 2022, ©2022, RadioGraphics published by RSNA) c Near infrared fluorescence images of diseased lymphatics (top) showing retrograde flow in symptomatic hand, and (bottom) tortuous vessels in symptomatic leg. Black spots are covered injection sites. (Adapted with permission from Rasmussen et al., 2009, ©2009, Current Opinion Biotechnology published by Elsevier)

Despite this great accuracy in detecting true negatives and positives, lymphoscintigraphy does have some drawbacks. First and foremost, lymphoscintigraphy has poor spatial resolution [33]. This may not have a significant impact on its ability to detect the presence of lymphedema, but it does mean that lymphoscintigraphy cannot provide much anatomical information, placing limitations on how much information it can provide clinicians about the specific disease states of individual cases. Lymphoscintigraphy also lacks standardization, particularly regarding quantitative analysis of results. There are no “standardized” sets of drainage times or even acquisition time points by which clinicians can conclusively diagnose lymphedema [34]. Instead, they largely rely on conducting bilateral lymphoscintigraphy, and compare drainage times and patterns between the suspected diseased and healthy limbs. This introduces another variable into the diagnosis process and can open the door to inconsistency in diagnosis from center to center and clinician to clinician. Lastly, although quantitative lymphoscintigraphy does allow for increased sensitivity relative to qualitative methods, lymphoscintigraphy images can still be “normal” for patients in very early stages of the disease process [35]. This means that lymphoscintigraphy may be unable to detect lymphedema early enough, resulting in detection only when more extensive physiological abnormalities such as dermal backflow are manifested more thoroughly.

The use of techniques such as MR lymphangiography instead of lymphoscintigraphy represents a paradigm shift in philosophy for lymphedema diagnosis. With its poor resolution, lymphoscintigraphy provides diagnosis primarily based on lymphatic functionality. MR lymphangiography, on the other hand, uses direct visualization of the lymphatic vasculature to determine whether there are issues in lymphatic drainage that could be attributed to lymphedema. Greater anatomical resolution makes MR lymphangiography useful for gaining a better understanding of pathophysiology of disease (Fig. 4b) [36, 37]. This improved resolution, also has, in several recent studies, been used in planning clinical procedures involving lymphatic vasculature such as lymphaticovenous anastomosis, a surgical technique used in the treatment of lymphedema [38, 39]. For the purposes of lymphedema diagnosis, MR lymphangiography is typically performed using a 1.5 T magnet (although, more recently, 3.0 T magnets have also been used) and injection of gadolinium-based contrast agents into the interdigital webbings of the hand and the foot [36, 40]. More recently, various studies are also beginning to apply non-contrast approaches. The subsequent images allow for visualization of lymphatic vessels, more specifically delayed lymphatic drainage, lymphatic dilation as a result of potential obstruction, and tortuous vessels [36].

Given this high degree of anatomic detail, the question then becomes whether this additional information proves useful for MR lymphangiography to match or surpass lymphoscintigraphy in lymphedema detection sensitivity and specificity. To this point, the answer appears to be no. While some recent studies have indicated that MR lymphangiography’s sensitivity could match that of lymphoscintigraphy, this is often not the case. In general, lymphoscintigraphy appears to consistently have higher sensitivity and specificity than MR lymphangiography. A representative, comparative study conducted by Weiss et al., published in 2014, found that while the modalities did have some association (correlation coefficient = 0.62), MR lymphangiography only achieved a sensitivity of 68% and a specificity of 91% [40]. Compared to several large lymphoscintigraphy studies conducted over the past several decades that achieved sensitivities above 90% and specificities close to 100%, the diagnostic performance of MR is lacking [30, 31]. Despite this, MR lymphangiography still remains an extremely effective tool for visualizing lymphatic anatomy, and, in the future, could prove to be a fantastic supplementary tool for planning lymphatic-based procedures.

In keeping with the same paradigm of anatomy-based diagnosis of lymphedema, ultrasound techniques also present a potential alternative to lymphoscintigraphy. Like MR lymphangiography, ultrasound relies primarily on imaging soft tissue to make lymphedema diagnoses. Unlike MR techniques (and lymphoscintigraphy, for that matter) ultrasound is quite cheap, quick, and readily available. It also does not require the injection of any contrast agents. This accessibility, combined with the relative simplicity of the actual image acquisition process, has made ultrasonography a mode of interest for lymphedema diagnosis [41]. Ultrasound is particularly useful for identifying volumetric changes, and so clinicians typically look for thickening of the dermis, subcutaneous layer, and muscle. The specific changes in these different layers varies between cases of primary and secondary lymphedema, indicating that ultrasound can serve as an effective tool for further characterization of patient disease states [42]. To this point, the use of duplex ultrasound techniques have been shown to assist in staging lymphedema. A 2013 study by Suehiro et al. showed that lymphedema could be reliably staged by measuring skin thickness, subcutaneous tissue thickness, and subcutaneous tissue echogenicity. However, the authors of the study noted difficulty in measuring subcutaneous tissue and skin thickness for later stage patients, and concluded that tissue echogenicity should be the primary ultrasound measurement used for diagnostic purposes [43]. Previous studies have also compared the efficacy of lymphoscintigraphy with that of ultrasound for lymphedema diagnosis. A 2021 paper with 14 enrolled lymphaticovenous anastomosis patients found that ultrasound had a 94.6% diagnostic accuracy and actually correctly diagnosed lymphedema in 39 of 54 leg areas where lymphoscintigraphy failed to detect it [44]. A 2010 study utilized both lymphoscintigraphy and duplex ultrasound to effectively distinguish between leg lymphedema and other edema abnormalities [45]. At the very least, the current literature warrants further exploration of using ultrasound and lymphoscintigraphy in concert to diagnose and characterize lymphedema.

One of the most recent lymphedema diagnosis methods is the use of near-infrared fluorescence (NIRF) imaging. As the name suggests, NIRF imaging utilizes fluorophores (the most common one being indocyanine green, or ICG) that are readily excited in the near infrared range to generate fluorescent signals that are then detected and used to construct images of the tissue containing the imaging agents [46]. While this technique itself, and particularly its clinical application to lymphedema diagnosis, is relatively modern, ICG has a long history of use in clinical diagnostics, dating back to 1956 [47]. NIRF imaging, although depth-limited at about 2–4 cm, provides improved spatial and temporal resolution relative to many common lymphatic imaging techniques including lymphoscintigraphy and MR lymphangiography [4, 46]. This allows for enhanced visualization of the lymphatic vessel anatomy. In fact, NIRF can allow for non-invasive observation of lymphatic phenomena such as pumping in both pre-clinical and clinical settings, making it particularly useful for readily identifying issues with lymphatic transport [48]. In regards to lymphedema, signs of lymphatic irregularities such as dermal backflow and tortuous vessels can then be used to make a diagnosis of lymphedema in the limbs (Fig. 4c) [24]. Additionally, NIRF imaging is quicker to perform and cheaper to execute in comparison to other techniques such as MR and PET, making it a particularly appealing option in clinical settings. While some studies have indicated that NIRF imaging could be comparable or even represent an improvement in some instances to lymphoscintigraphy in lymphedema diagnosis, given the decades-long track history of lymphoscintigraphy as a very effective tool for this purpose, additional work is required to draw any definitive conclusions about the comparability of these techniques [35, 49].

Several of these (and other) lymphatic imaging techniques have also been utilized for planning surgical treatments of lymphedema, such as lymphaticovenous anastomosis, or LVA.