For the past five years, we have prospectively evaluated patients with secondary lymphedema of the upper extremity who presented for surgical consultation at Memorial Sloan Kettering Cancer Center (MSK).
In our paper, published recently in the journal Cancers, we shared the results and key learning from our analysis of the data. Overall, we confirmed that preoperative assessment of upper extremity lymphedema is complex and requires multimodal assessment. All of the techniques we reviewed played a role, but the most valuable tools were magnetic resonance angiography (MRA), bioimpedance spectroscopy (L-Dex scores), indocyanine green (ICG) lymphography, and patient-reported outcome measures. (1)
Our goal was to review our preoperative assessment protocol and inform the development of a more streamlined template for patient evaluation. We also correlated preoperative measures of lymphedema and validated patient-reported outcome surveys with lymphedema stage to understand the contribution of clinical findings and patients’ reported symptoms to the nature of the disease.
At MSK, in addition to improving patient assessment methods, we are dedicated to investigating the mechanisms that lead to the development of lymphedema and pioneering microsurgical techniques to treat lymphedema and reduce its burden on patients. There are no other cancer centers with an active lymphedema surgical program supported by a dedicated lymphedema research laboratory.
Assessing Upper Extremity Lymphedema
Lymphedema is a chronic condition that develops after lymph node dissection for cancer treatment. It involves the accumulation of interstitial fluid in tissues and the deposition of fibro-adipose tissue in the affected limb. Lymphedema is a disabling condition that causes pain, stiffness, emotional issues, and a lower quality of life.
An improved understanding of the underlying biology and advances in microsurgical techniques have led to the development of a variety of options for the surgical treatment of secondary lymphedema. (2), (3), (4), (5), (6) Numerous investigators have published outcomes for a variety of surgical techniques. However, a detailed review of assessment methods comparing the rationale, utility and comparative efficacy for different modalities has not been reported. Also, an evaluation of patient-reported outcomes has not been performed using validated questionnaires. (1)
We identified 118 patients who had been evaluated for surgical management of unilateral upper extremity lymphedema following axillary surgery between January 2015 and July 2018.
The average age of patients was 54 ± 11 years, and the average BMI was 26.1 ± 3.9 kg/m.2 Most patients, 98 percent, were women, and breast cancer treatment was the most common cause of lymphedema for 86 percent. The average duration of lymphedema was 41 ± 54 months, and the average time to develop lymphedema after surgery was 23.8 ± 57.1 months. The vast majority of patients had stage 1 or 2 disease as defined by the International Society of Lymphology (ISL). (1)
A total of 98 percent of patients reported swelling at the time of evaluation in various locations: the hand (58 percent), forearm and elbow (84 percent), and upper arm (84 percent), and 48 percent of patients reported swelling of the entire upper extremity. Other symptoms included heaviness (71 percent), pitting edema (71 percent), history of infections (34 percent), pain (25 percent), and anxiety (24 percent). Common findings on clinical exam other than swelling and pitting edema included sensory impairment (16 percent), motor impairment (9 percent), and limited range of motion of the shoulder joint (38 percent). (1)
Lymphedema Assessment Tools
We reviewed the following assessment tools in our study: (1)
- Limb volume measurements. These are obtained using an automated perometer, or by manually measuring limb circumference and calculating limb volume using the truncated cone formula. A difference of more than 10 percent compared to the contralateral limb is considered diagnostic of lymphedema.
- Bioimpedance spectroscopy. This technique measures electrical impedance in the limb using the L-Dex model U400 (Impedimed, Brisbane, Australia). Extracellular fluid content measurements are extrapolated and applied to a normative dataset to derive an L-Dex score. A score greater than 10 is considered diagnostic of lymphedema.
- Patient-reported quality of life assessments. The Lymphedema Life Impact Scale (version 2) (LLIS) (7) includes 18 questions across physical, functional, and psychological domains. The upper limb lymphedema 27 (ULL-27) (8) has 27 questions across physical, emotional, and social domains. Both tools generate an impairment score as a percentage: zero percent means no impairment, and 100 percent represents severe impairment. However, the period differs: the LLIS asks patients about symptoms over the past week, whereas the ULL-27 asks about symptoms over the past month.
- MRA imaging. Magnetic resonance angiography of the chest and upper extremities assesses fluid and fat composition in the limb, evaluates for venous stenosis, and rules out occult malignancy. It also reveals the degree of lymphedema-related fat hypertrophy, which is a significant confounding variable during treatment. (9), (10) Using contrast agents, patients are imaged in the supine position using a 1.5 Tesla MRI (GE Healthcare).
- Lymphoscintigraphy. This technique involves injecting lidocaine, followed by a radiotracer into the first and third webspaces. Images are acquired every 30 minutes for up to three hours. Dermal reflux or lack of radiotracer uptake in the axilla after three hours is considered indicative of lymphedema. Many clinicians consider lymphoscintigraphy to be the gold standard imaging modality for lymphedema. (11)
- ICG lymphography. This method allows for preoperative staging and direct visualization of the lymphatic channels. It involves injecting lidocaine followed by indocyanine green dye into the first and third webspaces. Images are taken 30 minutes post-injection. Pathological changes are classified using a staging system where a linear pattern is considered normal, and splash, stardust, and diffuse patterns represent increasing levels of deterioration in lymphatic function. (12)
Key Learning from Our Review of Lymphedema Assessment Tools
ISL staging is subjective and not useful for preoperative classification. We found only loose correlations between ISL stage and limb volume difference, L-Dex scores, patient-reported outcome measures, and ICG stage. (1)For reference, ISL staging has long been used to classify the severity of lymphedema as follows: stage 0 is subclinical lymphedema; stage 1 is mild lymphedema that improves with elevation; stage 2a is moderate pitting edema; stage 2b is non-pitting lymphedema; and stage 3 is elephantiasis and irreversible skin changes. There is a definite need for a more sophisticated staging system in the future!
Limb volume measurements are more sensitive than assessing changes in limb circumference. Circumferential measurements had a relatively low sensitivity (83 percent) and specificity (85 percent) compared to limb volume measurements and tended to under-diagnose and under-estimate the degree of lymphedema. We also learned that volume measurements obtained by perometer were not interchangeable with manual measurements. The same technique should be used consistently when assessing changes in limb volume over time. (1)
Bioimpedance is sensitive and specific for measuring early-stage disease but requires optimization for reproducible results. Abnormal L-Dex scores were recorded in 71 patients (66 percent). We found that the L-Dex score was the most rapid, reliable, and non-invasive method for detecting early-stage lymphedema. L-Dex scores were significantly correlated with both ISL stage (r2 = 0.521, p < 0.001) and the percentage difference in limb volume (r2 = 0.714, p < 0.001). Compared to limb circumference measurements, L-Dex scores were more sensitive in diagnosing lymphedema (91 percent vs. 84 percent). They also had a higher predictive value (87 percent vs. 76 percent) when using a diagnostic threshold of a limb volume difference of greater than 10 percent. (1)
However, using the U400 device required a significant amount of training and optimization for consistent results. Changes in temperature, humidity, sweating, patient preparation, activity, and the use of compression garments had significant impacts on results. The device required the use of adhesive pads and leads, which may have caused the variation in results. (1) We have largely addressed these issues by moving to the newer SOZO device (Impedimed), which we plan to study in the future.
Even minor changes in limb volume can have significant effects on patient-reported quality of life. A substantial number of patients had only a minimal difference in limb volume and early stage of lymphedema yet reported a high degree of impairment in quality of life, especially in the physical/functional and psychological domains. These discrepancies suggest that lymphedema symptoms are a major cause of morbidity and that patient-reported outcome measures are essential to gain a complete picture. We also found that the LLIS patient-reported outcome measurement tool was more sensitive for measuring the degree of physical and functional disability compared to the ULL-27. (1)
MRA imaging is a highly sensitive tool for detecting fluid in subcutaneous tissues that is not normally present. MRA imaging identified that 15 percent of patients had venous stenosis, an important finding since venous hypertension can contribute to swelling and also compromise the effects of lymph node transplant or lymphovenous bypass. (13)
Lymphoscintigraphy had some limitations but was useful for preoperative assessment. This method had limitations as it only reported dermal backflow in a minority of patients. However, it was helpful as a preoperative assessment tool to identify the presence of any functional nodes that needed to be protected during lymph node transplant. It also provided a preoperative baseline to compare uptake by transplanted lymph nodes postoperatively. (1)
ICG lymphography was the most sensitive test for diagnosing lymphedema. Ninety-seven patients (82 percent) underwent ICG lymphography as part of their initial evaluation at MSK. All of these patients had a pathologic ICG pattern in the limb with lymphedema, and all healthy limbs had a regular linear pattern. Patients with early-stage splash patterns on ICG tended to have lower L-Dex scores and lower limb volumes. However, there was no tight correlation, reflecting that limb volume and extracellular fluid content do not necessarily reflect the underlying physiology of an abnormal lymphatic system. (1) Several subjective, pattern-based staging systems have been proposed based on recurrent patterns of backflow, (14), (15), (16), (17), (18) In the future, a quantitative ICG staging system would be more meaningful for determining the degree of lymphatic impairment.
Advancing Lymphedema Care Through Research
At MSK, we continue to look for ways to maximize outcomes for patients who develop lymphedema as a consequence of cancer treatment. We opened a phase III randomized controlled clinical trial in January 2020 to investigate whether immediate lymphatic reconstruction decreases the risk of lymphedema after axillary lymph node removal during breast cancer surgery and improves patients’ quality of life. Our goal is to recruit 184 patients and complete the study within two years.
Dr. Dayan is a paid consultant for Stryker Corporation, which manufactures a near-infrared camera that is commonly used off-label for lymphatic imaging in the microsurgical community. Near-infrared imaging is discussed in the paper. Dr. Mehrara and Dr. Coriddi declare no conflicts of interest related to the study.