The Prevalence of Lower Limb and Genital Lymphedema after Prostate Cancer Treatment: A Systematic Review

Simple Summary Prostate cancer patients that receive treatment (surgery of radiation therapy) directed to the pelvic lymph nodes may suffer from secondary lymphedema in the lower limbs and/or the genital area. Despite its potential impact on quality of life, reports on secondary lymphedema after prostate cancer therapy are scarce and prevalence rates vary between different studies. Here we perform a systematic literature search to estimate the prevalence of lymphedema after surgery, radiation therapy, or both, to the pelvic lymph nodes in men with prostate cancer. Abstract (1) Background: Secondary lymphedema is a chronic, progressive, and debilitating condition with an important impact on quality of life. Lymphedema is a frequently reported complication in oncological surgery but has not been systematically studied in the setting of prostate cancer. (2) Methods: Pubmed/MEDLINE and Embase were systematically searched to identify articles reporting on lower limb or genital lymphedema after primary treatment (surgery of radiation therapy) of the prostate and the pelvic lymph nodes in men with prostate cancer. Primary outcome was the prevalence of lower limb and genital lymphedema. (3) Results: Eighteen articles were eligible for qualitative synthesis. Risk of bias was high in all included studies, with only one study providing a prespecified definition of secondary lymphedema. Eleven studies report the prevalence of lower limb (0–14%) and genital (0–1%) lymphedema after radical prostatectomy with pelvic lymph node dissection (PLND) Seven studies report a low prevalence of lower limb (0–9%) and genital (0–8%) lymphedema after irradiation of the pelvic lymph nodes. However, in the patient subgroups that underwent pelvic irradiation after staging pelvic lymph node dissections, the prevalence of lower limb (18–29%) and genital (2–22%) lymphedema is substantially elevated. (4) Conclusion: Prostate cancer patients undergoing surgery or irradiation of the pelvic lymph nodes are at risk of developing secondary lymphedema in the lower limbs and the genital region. Patients receiving pelvic radiation after pelvic lymph node dissection have the highest prevalence of lymphedema. The lack of a uniform definition and standardized diagnostic criteria for lower limb and genital lymphedema hampers the accurate estimation of their true prevalence. Future clinicals trials are needed to specifically evaluate secondary lymphedema in patients undergoing prostate cancer treatments, to identify potential risk factors and to determine the impact on quality of life.


Introduction
Secondary lymphedema is a well-known complication of cancer therapy. In men undergoing prostate cancer treatment, surgical resection or irradiation of the pelvic lymph nodes can result in lymphedema (LE) of the lower limbs and the scrotal and suprapubic regions.

Outcome Measurement
The primary outcome measurement is the prevalence of lower limb, genital or suprapubic LE. The definitions of LE provided by the authors were used but LE needed to be reported as a separate entity. A secondary outcome is to evaluate potential risk factors for secondary lymphedema (if described).

Risk of Bias and Study Quality Assessment
To assess the validity of the included studies we used The Cochrane Handbook for Systematic Reviews [11]. We judged the risk of bias (RoB) from each included study as 'high', 'low', or 'unclear' for the following seven individual items: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants (performance bias), blinding of outcome assessor (detection bias), completeness of outcome data reporting (attrition bias), selective reporting (reporting bias), and other possible sources of bias such as conflicts of interest.

Study Selection
Our initial electronic database search identified 10,561 records ( Figure 1). After removing duplicates and screening all titles and abstracts, 142 trials were scrutinized for further eligibility. Of those, eighteen articles met our eligibility criteria and were consequently included in our qualitative analysis. Most records were excluded because they did not report lower limb or genital LE as a separate outcome. Table 1 shows baseline study characteristics from each included study, organized by primary intervention (radical prostatectomy versus external beam radiotherapy (EBRT). All studies were published between 1980 and 2022. Of the nineteen studies (in eighteen papers) included, three were randomized controlled trials (RCT), 4 were prospective comparative studies, 4 were prospective observational studies, and 8 were retrospective non-randomized trials. Sample size ranged from 99 to 3675 with a total of 9223 participants included in this qualitative analysis. Median age ranged from 61 to 68 years of age. Overall, the RoB within included studies was considered very high. Since only three RCTs were included, there was a high risk of selection, detection and performance bias. Most studies had a low or moderate RoB regarding attrition bias. Reporting bias was rated as high, with only one study that predefined lymphedema in its methods [13]. Other sources of bias were often unclear. Cancers 2022, 14, Table 1 shows baseline study characteristics from each included study, organized by primary intervention (radical prostatectomy versus external beam radiotherapy (EBRT). All studies were published between 1980 and 2022. Of the nineteen studies (in eighteen papers) included, three were randomized controlled trials (RCT), 4 were prospective comparative studies, 4 were prospective observational studies, and 8 were retrospective nonrandomized trials. Sample size ranged from 99 to 3675 with a total of 9223 participants included in this qualitative analysis. Median age ranged from 61 to 68 years of age.  NR NR n = number of patients; FU = follow-up; IQR = interquartile range; iPSA = initial Prostate-Specific Antigen; bGS = biopsy Gleason Score; cT = clinical T stage; pN = pathological N stage; RT = radiotherapy; EBRT = external beam radiation therapy; RP = radical prostatectomy; PLND = pelvic lymph node dissection; LN = lymph node; RRP = radical retropubic prostatectomy; NR = not reported; NA = not assessed; RARP = robot-assisted radical prostatectomy; BMI = Body Mass Index; ADT = androgen deprivation therapy. Figures 2 and 3 outline the Risk of Bias (RoB) assessment of all the included studies. Overall, the RoB within included studies was considered very high. Since only three RCTs were included, there was a high risk of selection, detection and performance bias. Most studies had a low or moderate RoB regarding attrition bias. Reporting bias was rated as high, with only one study that predefined lymphedema in its methods [13]. Other sources of bias were often unclear.    Overall, the RoB within included studies was considered very high. Since only three RCTs were included, there was a high risk of selection, detection and performance bias. Most studies had a low or moderate RoB regarding attrition bias. Reporting bias was rated as high, with only one study that predefined lymphedema in its methods [13]. Other sources of bias were often unclear.

Lower Limb Lymphedema
All the included studies report the prevalence of lower limb LE, with a prevalence ranging from 0% and 14% (Table 2). Importantly, only the LAPPRO study provides a prespecified definition of lower limb LE and the methodology for assessment of LE [13]. In this study, the authors use a standardized questionnaire with two specific questions to determine patient-reported "swelling in the left/right groin" and "swelling in the left/right leg" at three months after surgery. In addition, they also describe staff-reported LE at different time points after surgery. It is unclear how lower limb LE is determined in the other included studies. The bubble graph in Figure 4 depicts the prevalence of lower limb lymphedema in the included studies from 1980 to 2022.

Lower Limb Lymphedema
All the included studies report the prevalence of lower limb LE, with a prevalence ranging from 0% and 14% (Table 2). Importantly, only the LAPPRO study provides a prespecified definition of lower limb LE and the methodology for assessment of LE [13]. In this study, the authors use a standardized questionnaire with two specific questions to determine patient-reported "swelling in the left/right groin" and "swelling in the left/right leg" at three months after surgery. In addition, they also describe staff-reported LE at different time points after surgery. It is unclear how lower limb LE is determined in the other included studies. The bubble graph in Figure 4 depicts the prevalence of lower limb lymphedema in the included studies from 1980 to 2022.

Surgery
The prevalence of secondary lower limb LE after pelvic lymph node dissection ranged from 0 to 14% (Table 2). Five studies compare LE after RP with extended PLND versus RP with limited PLND [13,15,16,20,22]. Only Morizane et al., found a statistically significant difference in the rate of LE with 6% (28/431 patients) LE in patients undergoing extended PLND versus 1% (7/902 patients) in the limited PLND group (p < 0.001) [20]. Four studies without comparator group evaluate the prevalence of lower limb LE after RP with extended PLND [17][18][19]21]. In these studies, lymphedema is observed in 2-10% of patients. The highest prevalence of lower limb LE is reported in the LAPPRO trial, which reports patient-reported outcomes. Importantly, patient-reported prevalence (14%, 85/621 patients) of lower limb LE in this study is considerably higher than staff-reported LE rates (5%, 32/616 patients).

External Beam Radiotherapy with or without Staging PLND
Seven manuscripts (reporting on eight trials) report the prevalence of lower limb LE after RT to the prostate and the pelvic lymph node regions, with lymphedema rates ranging from 0% to 9% (Table 2) [23][24][25][26][27][28][29]. Four studies specifically report the prevalence of LE in subgroups that underwent staging PLND followed by irradiation of the pelvic lymph nodes in case of pathological lymph node involvement [26][27][28][29]. In these subgroups, the prevalence of secondary lymphedema (18-29%) is considerably higher than in subgroups that did not undergo staging PLND (0-8%).

Surgery
The prevalence of secondary lower limb LE after pelvic lymph node dissection ranged from 0 to 14% (Table 2). Five studies compare LE after RP with extended PLND versus RP with limited PLND [13,15,16,20,22]. Only Morizane et al., found a statistically significant difference in the rate of LE with 6% (28/431 patients) LE in patients undergoing extended PLND versus 1% (7/902 patients) in the limited PLND group (p < 0.001) [20]. Four studies without comparator group evaluate the prevalence of lower limb LE after RP with extended PLND [17][18][19]21]. In these studies, lymphedema is observed in 2-10% of patients. The highest prevalence of lower limb LE is reported in the LAPPRO trial, which reports patient-reported outcomes. Importantly, patient-reported prevalence (14%, 85/621 patients) of lower limb LE in this study is considerably higher than staff-reported LE rates (5%, 32/616 patients).

External Beam Radiotherapy with or without Staging PLND
Seven manuscripts (reporting on eight trials) report the prevalence of lower limb LE after RT to the prostate and the pelvic lymph node regions, with lymphedema rates ranging from 0% to 9% (Table 2) [23][24][25][26][27][28][29]. Four studies specifically report the prevalence of LE in subgroups that underwent staging PLND followed by irradiation of the pelvic lymph nodes in case of pathological lymph node involvement [26][27][28][29]. In these subgroups, the prevalence of secondary lymphedema (18-29%) is considerably higher than in subgroups that did not undergo staging PLND (0-8%).

Genital Lymphedema
Only a few studies make a distinction between lower limb and genital LE ( Table 2). A description of the methodology to assess genital LE is lacking in all included studies. Genital LE as a separate entity is reported in 0% to 22% of patients [21,22,[26][27][28][29].   Porcaro et al., reports only one out of 211 patients (0.5%) suffering from scrotal edema after RP with ePLND [21]. In a prospective observational study, Yuh et al., describe scrotal edema in 1.5% (3/204) of patients undergoing RP with extended PLND, and 0.5% (1/202) of patients undergoing RP with limited PLND [22].

External Beam Radiotherapy with or without Staging PLND
Five radiotherapy studies report the prevalence of genital LE (Table 2) [24,[26][27][28][29]. Aristizabal et al., report scrotal or penile LE in 2% (4/218) of patients treated with external beam radiotherapy only [24]. In Perez et al., genital LE is observed in 4 of 195 patients (2%) of which 14 patients received a staging laparotomy [27]. Scrotal and penile LE was observed by Pilepich et al., in 6 of 267 patients (2%), all of which underwent a staging PLND before whole pelvis irradiation [28]. In the RTOG75-06 and RTOG-77 trials, genital LE is reported in 0 to 6% of patients; with higher lymphedema rates in the subgroup that underwent staging PLND [29]. The highest prevalence of genital LE is reported by Forman et al., in 22% (9/41) of patients that underwent pelvic EBRT following a staging PLND versus only 1% (2/199) in patients who did not undergo staging PLND [26].

Discussion
Secondary lymphedema can be a major concern for patients undergoing oncological therapy, causing discomfort, functional impairment, and even psychosocial distress [4].
Most data from quality of life and medical costs are derived from upper limb lymphedema in women undergoing breast cancer treatment [4], whereas data from prostate cancer patients are sparse. Here, we performed a systematic literature review to determine the prevalence of secondary lymphedema in prostate cancer patients undergoing primary treatment of the prostate and the pelvic lymph nodes with surgery and/or radiation therapy.
In this systematic review, we found the rate of secondary LE ranging from zero to fourteen percent in patients undergoing PLND and from zero to eight percent in patients undergoing pelvic nodal irradiation. Importantly the prevalence of secondary LE is much higher in the subgroups that underwent pelvic nodal irradiation after staging PLND (between 18 and 29%) suggesting that the cumulative effect of surgery and irradiation results in substantially higher LE rates. PLND is considered the most sensitive technique to determine microscopic lymph node involvement, but the oncological benefits of this procedure remain elusive [30,31]. Since performing a PLND is not only associated with potential peri-and postoperative complications, including lymphoceles, thromboembolic events and neurovascular injuries [32], but also with the long-term risk of lower limb and genital edema, careful preoperative patient selection and counseling are crucial.
In this review, the reported LE prevalence varies considerably between different studies. These differences depend on differences in patient selection, differences in technique (e.g., extend of PLND) as well as differences in lymphedema assessment between different studies. The International Society of Lymphology defines Lymphedema as the 'external manifestation of lymphatic system insufficiency and deranged lymph transport.' The detection of lymphedema can be clinically evident in patients with clinically measurable swelling but can be more tedious in patients with subjective perceptions of swelling and/or limb heaviness without a clinically detectable swelling. Therefore, the diagnosis of lymphedema depends on patient-reported symptoms, visual inspection, skin palpation and measurements of volume differences between both limbs [1,[33][34][35].
The LAPPRO trial [13] was the only included study that performed a standardized assessment of postoperative LE. Lymphedema was defined as patient-reported "swelling in the left/right groin" and "swelling in the left/right leg" using a standardized questionnaire at 3 months after surgery. The authors also recorded staff-reported lymphedema, but no objective measurements were performed. Interestingly, the rate of patient-reported swelling (14%) at 3 months was almost threefold higher than staff-reported swelling (4%), suggesting an underreporting on staff reports. In all other studies a clear definition of LE or the methodology of how LE was determined is completely lacking. Therefore, the reported rates of secondary lymphedema might represent an underestimation of the true prevalence.
In the context of breast cancer treatments, LE is a well-known complication [4]. Several risk factors have been identified, including axillary lymph-node dissection, adjuvant RT, and high BMI, and several risk models have been developed to predict upper limb LE [35,36]. Moreover, there is a remarkable awareness for health-related QoL in these patients with routine use of patient-reported outcome measurements [33]. In contrast, no risk factors, other than performing a PLND have been identified as a risk factor for lower limb LE in PCa patients [13]. Although Morizane et al. [20] found a significantly higher prevalence of lower limb LE in patients undergoing extended versus limited PLND, Carlsson et al., did not find a correlation between the number of lymph nodes removed and the prevalence of secondary LE [13].
It is remarkable that, compared to breast cancer, secondary LE in prostate cancer patients has received little attention. A possible explanation could be the lower prevalence of lower limb LE in men undergoing prostate cancer treatments (0-14%) compared to upper limb LE in women undergoing breast cancer therapies (14-40%) [1]. Moreover, the functional and cosmetic aspects of LE may receive more attention in breast cancer, whereas sexual and urinary function are the main focus of attention in PCa patients [37]. Another reason could be the difficulty to objectivize lower limb LE when both limbs are affected. In patients with unilateral breast cancer, volume and circumference measurements of the affected limb, can easily be compared to the limb on the untreated side. In contract, PCa patients usually undergo bilateral PLND hereby affecting lymphatic transport in both limbs. Moreover, bilateral measurements can be biased by muscle hypertrophy or weight gain, equally affecting both limbs. The use of techniques that evaluate edema in a direct way, such a bio-impedance spectroscopy and tissue dielectric constant measurements, can assist in the diagnosis of LE, but these techniques have not been validated in the setting of lower limb or genital LE [38][39][40].

Limitations of This Study
Despite our systematic methodology, this review has several limitations. First, only a limited number of studies report on our outcomes of interest. Second, there is a lack of standardization in the definitions of LE and the methodology to determine the presence of lower limb and genital LE. Moreover, details about the time course of lymphedema are lacking in all but one study. As such, most included studies had a high RoB. Third, there is substantial heterogeneity between studies considering the proportion of patients undergoing staging PLND, surgical (open versus robot-assisted, extend of PLND) and radiation techniques (the template, duration, total dose). Moreover, outcomes of pelvic irradiation were published between 1980 and 1997, which may limit the translation to modern radiotherapy techniques [41].The lack of a unified definition of LE and the heterogeneity of the included studies withheld us from performing a meta-analysis.

Conclusions
This review systematically analyzes the published literature to determine the prevalence of lower limb and genital LE in PCa patients undergoing surgery or irradiation of the pelvic lymph nodes. The prevalence of lymphedema in the lower limbs and genital regions range from 0-14% and 0-1% after surgery, and 0-9% and 0-8% after pelvic radiation respectively, with a much higher prevalence in patients that underwent PLND followed by pelvic radiotherapy (18-29% and 2-22%). The great heterogeneity between different studies can be attributed to a lack of a standardized definition, a lack of standardized assessment tools and the absence of well-designed prospective studies to assess secondary lymphedema and its impact on quality of life. For PCa patients, LE is still the 'forgotten vascular disease' [42].