Advertisement

Risk factors for late-onset lower limb lymphedema after gynecological cancer treatment: A multi-institutional retrospective study

Published:January 31, 2020DOI:https://doi.org/10.1016/j.ejso.2020.01.033

      Abstract

      Introduction

      Late-onset lower limb lymphedema (LLL) is a significant clinical challenge for physicians dealing with patients that undergo treatment involving the pelvic cavity. We aimed to clarify the prevalence of and risk factors for late-onset LLL after treatment for gynecological cancer.

      Methods

      We conducted a multicenter retrospective study using records of cases in which LLL diagnosed by physical findings and measurement of limbs girths. Patients with LLL after treatment for uterine cervical, endometrial, and ovarian cancer were sequentially enrolled. We examined the timing of LLL onset and the associations between the time to onset and clinical characteristics, including age, type of cancer, lymphadenectomy sites, and performance of radiotherapy. We also investigated the risk factors for late-onset LLL and their effects on the cumulative incidence of late-onset LLL.

      Results

      In total, 711 patients fulfilled the required criteria. Mean age of was 50.2 years old and median follow-up period was 5.05 years. More than half of them (50.5%) presented with LLL ≥5 years after undergoing treatment for gynecological cancer. A substantial number of patients (29.4%) developed LLL ≥10 years after undergoing treatment for gynecological cancer. Being aged <50 years [(odds ratio (OR): 1.919, P = 0.001), cervical cancer (OR: 1.912, P = 0.001), and radiotherapy (OR: 1.664, P = 0.017) were identified as significant risk factors for late-onset LLL in multivariate logistic regression analysis.

      Conclusions

      A substantial number of patients present with LLL ≥5 years after receiving treatment for gynecological malignancies. Clinicians are required to identify high-risk patients and inform them of the risk of late-onset LLL.

      Keywords

      Abbreviations:

      LLL (lower limb lymphedema), OR (odds ratio), CI (confidence interval), HR (hazard ratio)

      Introduction

      Lower limb lymphedema (LLL) is one of the most disabling and debilitating complications of treatment for gynecological cancer [
      • Beesley V.
      • Janda M.
      • Eakin E.
      • Obermair A.
      • Battistutta D.
      Lymphedema after gynecological cancer treatment: prevalence, correlates, and supportive care needs.
      ]. It is characterized by lymphatic fluid accumulation in the interstitium, resulting from an injury to the lymphatic vessels or the obstruction of lymphatic channels or nodes caused by treatment, such as surgical procedures involving lymphadenectomy [
      • Hareyama H.
      • Hada K.
      • Goto K.
      • et al.
      Prevalence, classification, and risk factors for postoperative lower extremity lymphedema in women with gynecologic malignancies: a retrospective study.
      ]. The prevalence of LLL after gynecological cancer treatment reportedly ranges from 5% to 58% [
      • Beesley V.
      • Janda M.
      • Eakin E.
      • Obermair A.
      • Battistutta D.
      Lymphedema after gynecological cancer treatment: prevalence, correlates, and supportive care needs.
      ,
      • Hareyama H.
      • Hada K.
      • Goto K.
      • et al.
      Prevalence, classification, and risk factors for postoperative lower extremity lymphedema in women with gynecologic malignancies: a retrospective study.
      ,
      • Kuroda K.
      • Yamamoto Y.
      • Yanagisawa M.
      • et al.
      Risk factors and a prediction model for lower limb lymphedema following lymphadenectomy in gynecologic cancer: a hospital-based retrospective cohort study.
      ,
      • Graf N.
      • Rufibach K.
      • Schmidt A.M.
      • Fehr M.
      • Fink D.
      • Baege A.C.
      Frequency and risk factors of lower limb lymphedema following lymphadenectomy in patients with gynecological malignancies.
      ,
      • Ohba Y.
      • Todo Y.
      • Kobayashi N.
      • et al.
      Risk factors for lower-limb lymphedema after surgery for cervical cancer.
      ,
      • Hayes S.C.
      • Janda M.
      • Ward L.C.
      • et al.
      Lymphedema following gynecological cancer: results from a prospective, longitudinal cohort study on prevalence, incidence and risk factors.
      ,
      • Kim J.H.
      • Choi J.H.
      • Ki E.Y.
      • et al.
      Incidence and risk factors of lower-extremity lymphedema after radical surgery with or without adjuvant radiotherapy in patients with FIGO stage I to stage IIA cervical cancer.
      ,
      • Ki E.Y.
      • Park J.S.
      • Lee K.H.
      • Hur S.Y.
      Incidence and risk factors of lower extremity lymphedema after gynecologic surgery in ovarian cancer.
      ,
      • Beesley V.L.
      • Rowlands I.J.
      • Hayes S.C.
      • et al.
      Australian National Endometrial Cancer Study Group. Incidence, risk factors and estimates of a woman's risk of developing secondary lower limb lymphedema and lymphedema-specific supportive care needs in women treated for endometrial cancer.
      ,
      • Hopp E.E.
      • Osborne J.L.
      • Schneider D.K.
      • Bojar C.J.
      • Uyar D.S.
      A prospective pilot study on the incidence of post-operative lymphedema in women with endometrial cancer.
      ,
      • Abu-Rustum N.R.
      • Alektiar K.
      • Iasonos A.
      • et al.
      The incidence of symptomatic lower-extremity lymphedema following treatment of uterine corpus malignancies: a 12-year experience at Memorial Sloan-Kettering Cancer Center.
      ]. The marked variation in the prevalence of LLL after gynecological cancer treatment is caused by differences in the observational period among studies and the absence of standard diagnostic criteria. The clinical presentations of LLL are usually chronic and progressive, and LLL ultimately leads to a reduction in quality of life [
      • Yost K.J.
      • Cheville A.L.
      • Al-Hilli M.M.
      • et al.
      Lymphedema after surgery for endometrial cancer: prevalence, risk factors, and quality of life.
      ,
      • Kim S.I.
      • Lim M.C.
      • Lee J.S.
      • et al.
      Impact of lower limb lymphedema on quality of life in gynecologic cancer survivors after pelvic lymph node dissection.
      ,
      • Rowlands I.J.
      • Beesley V.L.
      • Janda M.
      • et al.
      Australian National Endometrial Cancer Study Group. Quality of life of women with lower limb swelling or lymphedema 3-5 years following endometrial cancer.
      ].
      Focusing on the period from treatment to the onset of LLL, it has been reported that some patients developed LLL ≥10 years after receiving treatment for gynecological cancer [
      • Ohba Y.
      • Todo Y.
      • Kobayashi N.
      • et al.
      Risk factors for lower-limb lymphedema after surgery for cervical cancer.
      ,
      • Kim J.H.
      • Choi J.H.
      • Ki E.Y.
      • et al.
      Incidence and risk factors of lower-extremity lymphedema after radical surgery with or without adjuvant radiotherapy in patients with FIGO stage I to stage IIA cervical cancer.
      ]. In contrast, most of the literature focuses on LLL that develops early in the follow-up period; however, late-onset LLL is a significant clinical challenge for physicians that deal with patients that undergo surgery involving the pelvic cavity. In ordinary clinical practice, most patients that do not exhibit any evidence of cancer recurrence stop visiting healthcare institutions after 5–10 years of follow-up. Therefore, a substantial number of cases of late-onset LLL might be missed due to the current clinical situation. To detect as many cases of LLL as possible, even after post-cancer-treatment follow-up, we need to be able to identify high-risk patients and inform them of the possibility of late-onset LLL to implement possible interventions such as lymphaticovenular anastomosis [
      • Takeishi M.
      • Kojima M.
      • Mori K.
      • Kurihara K.
      • Sasaki H.
      Primary intrapelvic lymphaticovenular anastomosis following lymph node dissection.
      ,
      • Gentileschi S.
      • Albanese R.
      • Pino V.
      • et al.
      SPECT/CT and fusion ultrasound to target the efferent groin lymph node for lymphatic surgery.
      ], and creating perforator flap [
      • Gentileschi S.
      • Servillo M.
      • Garganese G.
      • et al.
      The lymphatic superficial circumflex iliac vessels deep branch perforator flap: a new preventive approach to lower limb lymphedema after groin dissection-preliminary evidence.
      ]. In this study, we aimed to evaluate the prevalence of and risk factors for late-onset LLL after gynecological cancer treatment and to assess the associations between clinical variables and the timing of LLL onset.

      Materials and methods

      We conducted a multicenter retrospective study using the records of 7 medical institutions (Toyota Memorial Hospital, Hirota Internal Medicine Clinic, Nagumo Clinic Fukuoka, Northern Fukushima Medical Center, Limbs Tokushima Clinic, Iwate Prefectural Miyako Hospital, and the Cancer Institute Hospital of the Japanese Foundation for Cancer Research) for the period from April 2002 to November 2014. Every institution follows the recommendations of the National Lymphedema Network who require a physician, nurse, physical therapist, or occupational therapist to complete a 135 h coursework before treating patients [
      • National Lymphedema Network
      Position paper: training of lymphedema therapists.
      ]. This study was authorized and approved by the ethical committee of School of Engineering, the University of Tokyo, and conducted in accordance with the principles of the Declaration of Helsinki and all the data were anonymously analyzed. As this was an observational retrospective study, the institutional ethics committee stated that it was not necessary to obtain informed consent from each patient.
      Patients who developed LLL after undergoing treatment for gynecological cancer and referred to the institutions were sequentially enrolled for this study during the study period. The diagnosis of LLL was made based on physical findings according to the International Society of Lymphology staging system [
      • Executive Committee
      The diagnosis and treatment of peripheral lymphedema: 2016 consensus document of the international society of Lymphology.
      ]. Principally, clinicians and therapists assessed the difference of limbs girths and performed physical examination for accurate diagnosis and staging. A few patients were affected by bilateral LLL, but the onset time and degree of LLL differed between each leg in these patients; thus, every affected limb was counted as an individual sample; i.e., bilateral LLL was counted as two samples. We focused on the three most common types of gynecological cancer (cervical, endometrial, and ovarian cancer) and excluded other rare types of cancer, such as vaginal cancer, vulvar cancer, and Paget's disease because surgical approach for these cancers were different from the targeted diseases. We also excluded cases of LLL caused by definitively diagnosed primary heart failure, liver failure, or renal failure.
      We collected data on the patients’ baseline characteristics, including age at surgery, the type of cancer, the site of lymphadenectomy (pelvic region, or the pelvic and para-aortic and/or inguinal regions), and the presence/absence of radiotherapy. We also recorded the date of LLL onset and calculated the time from surgery to the onset of LLL. All of the data were obtained via a questionnaire, which was completed by the treating therapists, or through a chart review. The primary outcome of this study was to determine the prevalence of and identify risk factors for late-onset LLL after treatment for gynecological cancer. In addition to evaluating the frequency of LLL onset in each year after treatment for gynecological cancer, we investigated the associations between each examined variable and the time until LLL onset using univariate and multivariate analyses. All of the patients had undergone surgery for gynecological malignancies. The lymphadenectomy sites were classified into the pelvic region (one region) or the pelvic and para-aortic and/or inguinal regions (two or more regions). We determined late-onset LLL as LLL that was diagnosed ≥5 years after the initial treatment for gynecological malignancy.
      The patients’ baseline characteristics are presented as the mean and SD or as absolute values and percentages (%). In addition, each characteristic was stratified into three LLL onset periods: <1 year, 1 year to <5 years, and ≥5 years. Comparisons among the groups were performed via one-way analysis of variance or the chi-squared test, as appropriate. The variations in the time to LLL onset between different age groups (<50 vs. ≥50), types of cancer (cervical vs. endometrial or ovarian cancer), or lymphadenectomy sites (pelvic region alone vs. pelvic plus para-aortic and/or inguinal regions) or between the presence/absence of radiotherapy (received vs. not received) were evaluated with the Mann-Whitney U test. The risk factors for late-onset LLL, and their predictive powers were quantified with univariate and multivariate-adjusted OR, which were estimated by a logistic regression model using four variables: age, type of cancer, lymphadenectomy sites, and performance of radiotherapy. The cumulative incidence of LLL and how it differed among the identified risk factors were evaluated with Kaplan-Meier curves, and the log-rank test was used to assess the significance of any differences. Cox regression analysis was also used to estimate HR and 95% confidence intervals (95%CI). Statistical significance was determined using two-tailed p-values, and the significance level was set at P < 0.05. All of the statistical analyses were performed using IBM SPSS Statistics, version 24.0 (IBM Corp., Armonk, NY, USA).

      Results

      In total, 711 patients with LLL who visited one of the 7 institutions during the study period fulfilled the required criteria. The mean age of the patients was 50.2 years old. ISL stages I, II, late in II, and III were noted in 6 (0.8%), 634 (89.2%), 65 (9.1%) and 6 (0.8%) patients, respectively. One hundred and ninety-five (27.4%) patients received radiotherapy. The baseline characteristics of the eligible cases are listed in Table 1. The patients’ characteristics were also stratified into three LLL onset period categories (<1, 1 to <5, and ≥5 years). Among the baseline data, there were significant differences in the distributions of age, the type of cancer, and the presence/absence of radiotherapy. Young patients, patients with cervical cancer, and patients that underwent radiotherapy tended to present with late-onset LLL more often than the patients they were compared with (all P < 0.001). The lymphadenectomy site had no effect on the prevalence of late-onset LLL (P = 0.472).
      Table 1Baseline characteristics of the patients with lower limb lymphedema (n = 711).
      Time to onsetP-value
      CharacteristicsCategoriesTotal<1 year (n = 143)1 to < 5 years (n = 209)≥5 years (n = 359)
      Age, years (SD)50.2 (11.3)53.1 (11.2)53.7 (11.5)47.0 (10.3)<0.001
      Type of cancer, n (%)Cervical cancer335 (47.1)55 (38.5)75 (35.9)205 (57.1)<0.001
      Endometrial cancer225 (31.6)58 (40.5)69 (33.0)98 (27.3)
      Ovarian cancer131 (18.4)28 (19.6)63 (30.1)40 (11.1)
      Unknown20 (2.8)2 (1.4)2 (1.0)16 (4.5)
      Lymphadenectomy, n (%)Pelvic region436 (61.3)88 (61.5)139 (66.5)209 (58.2)0.472
      Pelvic + para-aortic and/or inguinal regions116 (16.3)28 (19.6)31 (14.8)57 (15.9)
      Unknown159 (22.4)27 (18.9)39 (18.7)93 (25.9)
      Radiotherapy, n (%)Received195 (27.4)20 (14.0)53 (25.4)122 (34.0)<0.001
      Unknown29 (4.1)4 (2.8)9 (4.3)16 (4.5)
      Data are expressed as group means (standard deviation) or absolute values (and percentages).
      The p-values were obtained using one-way analysis of variance for continuous variables or the chi-square test as appropriate.
      Abbreviation: SD, standard deviation.
      The median duration of the follow-up period was 5.05 years. Late-onset LLL was defined as LLL that occurred ≥5 years after treatment for gynecological cancer, and more than half of the LLL patients (50.5%) presented with late-onset LLL. There were a substantial number of patients that developed LLL ≥10 years (29.4%) after undergoing treatment for gynecological cancer. The frequency of LLL onset was highest in the first year after surgery (20.1%). However, some patients developed LLL ≥20 years after undergoing treatment for gynecological malignancies.
      The results of the univariate and multivariate analyses aimed at identifying risk factors for late-onset LLL are shown in Table 2. The time to onset differed significantly between the two age groups (<50 years and ≥50 years). Compared with the LLL patients aged ≥50 years, the younger patients experienced late-onset LLL more frequently [median time to onset: 8.38 vs. 3.47 years (range: 0.05–51.08 vs. 0.05–32.10 years), P < 0.001]. Regarding the types of gynecological malignancies, we divided the type of cancer into two categories: cervical cancer, and endometrial or ovarian cancer. The patients with cervical cancer presented with late-onset LLL more frequently than those with endometrial or ovarian cancer [median time to onset: 7.24 vs. 3.19 years (range: 0.07–47.97 vs. 0.05–51.08 years), P < 0.001]. There was no significant difference in the time to LLL onset between the two types of lymphadenectomy site (pelvic region and pelvic with para-aortic and/or inguinal regions) [median time to onset: 4.72 vs. 4.72 years (range: 0.05–51.08 vs. 0.07–31.61 years), P = 0.449]. As for radiotherapy, the patients that received radiotherapy presented with late-onset LLL more frequently than the patients that did not receive such treatment [median: 4.72 vs. 7.44 years (range: 0.05–51.08 vs. 0.07–47.97 years), P < 0.001]. The following variables were identified as significant risk factors for late-onset LLL in the univariate logistic regression analyses: being aged under 50 years [odds ratio (OR): 2.449, 95%CI: 1.809–3.315, P < 0.001], cervical cancer (OR: 2.491, 95%CI: 1.834–3.383, P < 0.001), and radiotherapy (OR: 2.012, 95%CI: 1.431–2.827, P < 0.001). In the multivariate analysis, all three of the abovementioned factors were identified as independent risk factors for late-onset LLL: being aged <50 years (OR: 1.919, 95%CI: 1.329–2.770, P = 0.001), cervical cancer (OR: 1.912, 95%CI: 1.296–2.820, P = 0.001), and radiotherapy (OR: 1.664, 95%CI: 1.096–2.527, P = 0.017). Super-late-onset LLL was defined as LLL that occurred ≥10 years after treatment for gynecological cancer, and the following variables were found to be significant risk factors for super-late-onset LLL in the univariate logistic regression analyses: being aged <50 years (OR: 3.418, 95%CI: 2.419–4.830, P < 0.001), cervical cancer (OR: 2.259, 95%CI: 1.612–3.165, P < 0.001), and radiotherapy (OR: 1.830, 95%CI: 1.284–2.608, P = 0.001). In the multivariate analysis, the following two variables were identified as independent risk factors for super-late-onset LLL: being aged <50 years (OR: 3.657, 95%CI: 2.353–5.682, P < 0.001) and radiotherapy (OR: 1.801, 95%CI: 1.139–2.847, P = 0.012). The lymphadenectomy site was not found to be significantly associated with late-onset LLL in any of the analyses.
      Table 2Univariate and multivariate analyses of the factors related to late-onset lymphedema.
      Time to onset: ≥5 yearsTime to onset: ≥10 years
      Univariate analysisMultivariate analysisUnivariate analysisMultivariate analysis
      CharacteristicsCategoriesOR (95%CI)P-valueOR (95%CI)P-valueOR (95%CI)P-valueOR (95%CI)P-value
      Age<50 yearsreference<0.001reference0.001reference<0.001reference<0.001
      ≥50 years2.449 (1.809–3.315)1.919 (1.329–2.770)3.418 (2.419–4.830)3.657 (2.353–5.682)
      Type of cancerEndometrial or ovarian cancerreference<0.001reference0.001reference<0.001reference0.238
      Ovarian cancer2.491 (1.834–3.383)1.912 (1.296–2.820)2.259 (1.612–3.165)1.313 (0.835–2.066)
      LymphadenectomyPelvic regionreference0.818reference0.858reference0.384reference0.461
      Pelvic + para-aortic and/or inguinal regions1.049 (0.697–1.581)1.041 (0.673–1.610)0.808 (0.500–1.305)0.824 (0.492–1.379)
      RadiotherapyNot receivedreference<0.001reference0.017reference0.001reference0.012
      Received2.012 (1.431–2.827)1.664 (1.096–2.527)1.830 (1.284–2.608)1.801 (1.139–2.847)
      Abbreviations: OR, odds ratio; CI, confidence interval.
      Fig. 1 illustrates the cumulative incidence of LLL stratified according to the identified risk factors. The incidence of LLL rose sharply in the first year after treatment and was still increasing after 10 years. It reached a plateau after around 25 years, but some patients developed LLL after >25 years (Fig. 1A). Clear differences in the incidence of LLL were seen in the comparisons involving age [hazard ratio (HR): 1.903, 95%CI: 1.630–2.222, P < 0.001], the presence/absence of cervical cancer (HR: 0.644, 95%CI: 0.553–0.749, log-rank test, P < 0.001), and the presence/absence of radiotherapy (HR: 0.658, 95%CI: 0.556–0.778, P < 0.001) (Fig. 1B and C).
      Fig. 1
      Fig. 1(A) Cumulative incidence of lower limb lymphedema according to the time (years) since treatment for gynecological cancer. The cumulative incidence of lower limb lymphedema was also plotted according to (B) age (<50 or ≥50 years), (C) the presence/absence of cervical cancer, and (D) the presence/absence of radiotherapy (RT). The hazard ratios (HR) and 95% confidence intervals (CI) were estimated using an age of <50, endometrial or ovarian cancer, and not receiving RT as references. All p-values were calculated using the log-rank test.

      Discussion

      In the present study, more than half of the examined cases of LLL were categorized as late-onset LLL. Moreover, approximately 30% of LLL developed ≥10 years after treatment for gynecological cancer. This indicates that a substantial number of patients develop LLL 5–10 years after undergoing treatment for gynecological cancer. We also confirmed that being aged <50 years, having a history of cervical cancer rather than endometrial or ovarian cancer, and receiving radiotherapy were independent risk factors for late-onset LLL, and so patients’ clinical histories can be used to identify patients that are at high-risk of late-onset LLL after treatment for gynecological cancer. Moreover, a younger age and radiotherapy were found to be strongly associated with the development of LLL ≥10 years after treatment for gynecological cancer. Although this study had a retrospective design, it contained a large sample and identified novel risk factors for late-onset LLL after treatment for gynecological cancer. These findings will be beneficial not only for gynecologists, but also for surgeons that perform surgery involving the organs or the pelvic cavity, including those involved in gastroenterology or urology.
      There might be pathophysiological differences between early and late-onset LLL. LLL that develops within 1 year of treatment for gynecological cancer includes acute LLL due to the direct obstruction of lymphatic vessels caused by surgical damage. In our cohort, LLL onset most frequently occurred during the first year after treatment, which is consistent with previous reports [
      • Beesley V.
      • Janda M.
      • Eakin E.
      • Obermair A.
      • Battistutta D.
      Lymphedema after gynecological cancer treatment: prevalence, correlates, and supportive care needs.
      ,
      • Hareyama H.
      • Hada K.
      • Goto K.
      • et al.
      Prevalence, classification, and risk factors for postoperative lower extremity lymphedema in women with gynecologic malignancies: a retrospective study.
      ,
      • Kuroda K.
      • Yamamoto Y.
      • Yanagisawa M.
      • et al.
      Risk factors and a prediction model for lower limb lymphedema following lymphadenectomy in gynecologic cancer: a hospital-based retrospective cohort study.
      ,
      • Hayes S.C.
      • Janda M.
      • Ward L.C.
      • et al.
      Lymphedema following gynecological cancer: results from a prospective, longitudinal cohort study on prevalence, incidence and risk factors.
      ,
      • Kim J.H.
      • Choi J.H.
      • Ki E.Y.
      • et al.
      Incidence and risk factors of lower-extremity lymphedema after radical surgery with or without adjuvant radiotherapy in patients with FIGO stage I to stage IIA cervical cancer.
      ,
      • Ki E.Y.
      • Park J.S.
      • Lee K.H.
      • Hur S.Y.
      Incidence and risk factors of lower extremity lymphedema after gynecologic surgery in ovarian cancer.
      ]. This result supports the theory that the main cause of LLL is physical damage that affects lymphatic flow from the lower limbs to the trunk. On the other hand, it is difficult to attribute cases of LLL that emerge ≥5 years after treatment solely to the direct effects of surgery. Some cases must include indirect effects of surgery and/or factors associated with the lymphatic regeneration process. This is why the baseline characteristics of the patients that developed LLL within the first year after treatment differed from those of the patients that developed LLL ≥5 years after treatment.
      We identified the following significant risk factors for late-onset LLL after treatment for gynecological cancer: being aged <50, having cervical cancer, and receiving radiotherapy. As for being aged <50, the superior pumping function of the limbs found in such patients might help to delay the onset of LLL, but this needs to be confirmed in a future study. Few previous studies have mentioned the effects of age on the development of LLL. However, clinicians need to be aware that young gynecological cancer survivors are at risk of LLL even 10 years after being treated for gynecological malignancies. Compared with endometrial and ovarian cancer, cervical cancer was found to be an independent risk factor for late-onset LLL after treatment for gynecological cancer. In other words, endometrial and ovarian cancer cause LLL to occur earlier than cervical cancer. The recent study by Cariati M et al. reported that the adjuvant use of taxane is associated with breast cancer-related arm lymphedema [
      • Cariati M.
      • Bains S.K.
      • Grootendorst M.R.
      • et al.
      Adjuvant taxanes and the development of breast cancer-related arm lymphoedema.
      ]. Kilbreath SL et al. also showed that arm swelling at 6 and 12 months after surgery for breast cancer was associated with taxane-based chemotherapy [
      • Kilbreath S.L.
      • Refshauge K.M.
      • Beith J.M.
      • et al.
      Risk factors for lymphoedema in women with breast cancer: a large prospective cohort.
      ]. Although gynecological cancer-related LLL is not the same entity as breast cancer-related lymphedema, taxane-based chemotherapy, which is mainly used to treat endometrial and ovarian cancer patients, might increase the risk of early onset LLL. On the other hand, in the current study the lymphadenectomy site did not have significant effects on the risk of late-onset LLL. Trials focusing on the specific impact of lymphadenectomy on the risk of late-onset LLL are needed. Radiotherapy also significantly contributed to the late onset of LLL, possibly because of late damage to lymphatic tissue. There have been several reports about the relationship between radiotherapy and the development of LLL in patients with gynecological malignancies [
      • Beesley V.
      • Janda M.
      • Eakin E.
      • Obermair A.
      • Battistutta D.
      Lymphedema after gynecological cancer treatment: prevalence, correlates, and supportive care needs.
      ,
      • Kuroda K.
      • Yamamoto Y.
      • Yanagisawa M.
      • et al.
      Risk factors and a prediction model for lower limb lymphedema following lymphadenectomy in gynecologic cancer: a hospital-based retrospective cohort study.
      ,
      • Ohba Y.
      • Todo Y.
      • Kobayashi N.
      • et al.
      Risk factors for lower-limb lymphedema after surgery for cervical cancer.
      ,
      • Hayes S.C.
      • Janda M.
      • Ward L.C.
      • et al.
      Lymphedema following gynecological cancer: results from a prospective, longitudinal cohort study on prevalence, incidence and risk factors.
      ,
      • Beesley V.L.
      • Rowlands I.J.
      • Hayes S.C.
      • et al.
      Australian National Endometrial Cancer Study Group. Incidence, risk factors and estimates of a woman's risk of developing secondary lower limb lymphedema and lymphedema-specific supportive care needs in women treated for endometrial cancer.
      ]. In this study, we additionally revealed that the effect of radiotherapy can persist for quite a long time, even after gynecological cancer has been completely cured.
      The main strength of this study is that it involved a large sample of LLL patients from multiple institutions, which aided the identification of novel risk factors for late-onset LLL after gynecological cancer treatment. The limitations of this study include the possibility that cases of mild LLL might have been excluded because the institutions that participated in the present study were not primary clinics, but rather were specialized institutions for the treatment of lymphedema. Also, obesity or overweight might influence the onset of LLL, which were reported to be associated with development of lymphedema [
      • D'Ettorre M.
      • Gniuli D.
      • Bracaglia R.
      • et al.
      Micro- and macroscopic structural modification of subcutaneous adipose tissue after bariatric surgery.
      ,
      • Bracaglia R.
      • D'Ettorre M.
      • Gentileschi S.
      • Mingrone G.
      • Tambasco D.
      Kris knife" brachioplasty after bariatric surgery and massive weight loss.
      ,
      • Bracaglia R.
      • Tambasco D.
      • Gentileschi S.
      • D'Ettorre M.
      Abdominoplasty in patient with penile implant.
      ]. In addition, due to the study's retrospective design, the day of diagnosis did not always match the day on which the initial symptoms of LLL arose. Also, the institutions did not use diagnostic modality, including volumetry, other than measurement of girths, which was one of the limitations for design of the study. Ideally, final diagnosis of lymphedema should be obtained by lymphoscintigraphy and indocyanine green lymphography [
      • Gentileschi S.
      • Servillo M.
      • Ferrandina G.
      • Salgarello M.
      Lymphatic and sensory function of the upper limb after brachioplasty in post-bariatric massive weight loss patients.
      ,
      • Gentileschi S.
      • Servillo M.
      • Albanese R.
      • De Bonis F.
      • Tartaglione G.
      • Salgarello M.
      Lymphatic mapping of the upper limb with lymphedema before lymphatic supermicrosurgery by mirroring of the healthy limb.
      ]. Regarding these aspects, the results of this study should primarily be used as a basis for additional prospective studies.
      In conclusion, a substantial number of patients present with LLL ≥5 years after receiving treatment for gynecological malignancies. Clinicians are required to identify high-risk patients and inform them of not only the risk of late-onset LLL but also possible primary or secondary preventive approaches [
      • Bracaglia R.
      • D'Ettorre M.
      • Gentileschi S.
      • Tambasco D.
      Was the surgeon a satisfactory informant? How to minimize room for claims.
      ].

      Funding sources

      This study was supported by JSPS KAKENHI, Grant Number 26293463.

      Declaration of competing interest

      None declared.

      Acknowledgment

      We express our appreciation to the seven medical institutions that contributed to our study. We also express our sincere gratitude to Dr. Koji Tamakoshi, who contributed to the design of this study and the statistical analysis with patience and knowledge.

      References

        • Beesley V.
        • Janda M.
        • Eakin E.
        • Obermair A.
        • Battistutta D.
        Lymphedema after gynecological cancer treatment: prevalence, correlates, and supportive care needs.
        Cancer. 2007; 109: 2607-2614
        • Hareyama H.
        • Hada K.
        • Goto K.
        • et al.
        Prevalence, classification, and risk factors for postoperative lower extremity lymphedema in women with gynecologic malignancies: a retrospective study.
        Int J Gynecol Canc. 2015; 25: 751-757
        • Kuroda K.
        • Yamamoto Y.
        • Yanagisawa M.
        • et al.
        Risk factors and a prediction model for lower limb lymphedema following lymphadenectomy in gynecologic cancer: a hospital-based retrospective cohort study.
        BMC Wom Health. 2017; 17: 50
        • Graf N.
        • Rufibach K.
        • Schmidt A.M.
        • Fehr M.
        • Fink D.
        • Baege A.C.
        Frequency and risk factors of lower limb lymphedema following lymphadenectomy in patients with gynecological malignancies.
        Eur J Gynaecol Oncol. 2013; 34: 23-27
        • Ohba Y.
        • Todo Y.
        • Kobayashi N.
        • et al.
        Risk factors for lower-limb lymphedema after surgery for cervical cancer.
        Int J Clin Oncol. 2011; 16: 238-243
        • Hayes S.C.
        • Janda M.
        • Ward L.C.
        • et al.
        Lymphedema following gynecological cancer: results from a prospective, longitudinal cohort study on prevalence, incidence and risk factors.
        Gynecol Oncol. 2017; 146: 623-629
        • Kim J.H.
        • Choi J.H.
        • Ki E.Y.
        • et al.
        Incidence and risk factors of lower-extremity lymphedema after radical surgery with or without adjuvant radiotherapy in patients with FIGO stage I to stage IIA cervical cancer.
        Int J Gynecol Canc. 2012; 22: 686-691
        • Ki E.Y.
        • Park J.S.
        • Lee K.H.
        • Hur S.Y.
        Incidence and risk factors of lower extremity lymphedema after gynecologic surgery in ovarian cancer.
        Int J Gynecol Canc. 2016 Sep; 26: 1327-1332
        • Beesley V.L.
        • Rowlands I.J.
        • Hayes S.C.
        • et al.
        Australian National Endometrial Cancer Study Group. Incidence, risk factors and estimates of a woman's risk of developing secondary lower limb lymphedema and lymphedema-specific supportive care needs in women treated for endometrial cancer.
        Gynecol Oncol. 2015; 136: 87-93
        • Hopp E.E.
        • Osborne J.L.
        • Schneider D.K.
        • Bojar C.J.
        • Uyar D.S.
        A prospective pilot study on the incidence of post-operative lymphedema in women with endometrial cancer.
        Gynecol Oncol Rep. 2015; 15: 25-28
        • Abu-Rustum N.R.
        • Alektiar K.
        • Iasonos A.
        • et al.
        The incidence of symptomatic lower-extremity lymphedema following treatment of uterine corpus malignancies: a 12-year experience at Memorial Sloan-Kettering Cancer Center.
        Gynecol Oncol. 2006; 103: 714-718
        • Yost K.J.
        • Cheville A.L.
        • Al-Hilli M.M.
        • et al.
        Lymphedema after surgery for endometrial cancer: prevalence, risk factors, and quality of life.
        Obstet Gynecol. 2014; 124: 307-315
        • Kim S.I.
        • Lim M.C.
        • Lee J.S.
        • et al.
        Impact of lower limb lymphedema on quality of life in gynecologic cancer survivors after pelvic lymph node dissection.
        Eur J Obstet Gynecol Reprod Biol. 2015; 192: 31-36
        • Rowlands I.J.
        • Beesley V.L.
        • Janda M.
        • et al.
        Australian National Endometrial Cancer Study Group. Quality of life of women with lower limb swelling or lymphedema 3-5 years following endometrial cancer.
        Gynecol Oncol. 2014; 133: 314-318
        • Takeishi M.
        • Kojima M.
        • Mori K.
        • Kurihara K.
        • Sasaki H.
        Primary intrapelvic lymphaticovenular anastomosis following lymph node dissection.
        Ann Plast Surg. 2006; 57: 300-304
        • Gentileschi S.
        • Albanese R.
        • Pino V.
        • et al.
        SPECT/CT and fusion ultrasound to target the efferent groin lymph node for lymphatic surgery.
        Microsurgery. 2019; 39: 605-612
        • Gentileschi S.
        • Servillo M.
        • Garganese G.
        • et al.
        The lymphatic superficial circumflex iliac vessels deep branch perforator flap: a new preventive approach to lower limb lymphedema after groin dissection-preliminary evidence.
        Microsurgery. 2017; 37: 564-573
        • National Lymphedema Network
        Position paper: training of lymphedema therapists.
        (Available from URL)
        • Executive Committee
        The diagnosis and treatment of peripheral lymphedema: 2016 consensus document of the international society of Lymphology.
        Lymphology. 2016; 49: 170-184
        • Cariati M.
        • Bains S.K.
        • Grootendorst M.R.
        • et al.
        Adjuvant taxanes and the development of breast cancer-related arm lymphoedema.
        Br J Surg. 2015; 102: 1071-1078
        • Kilbreath S.L.
        • Refshauge K.M.
        • Beith J.M.
        • et al.
        Risk factors for lymphoedema in women with breast cancer: a large prospective cohort.
        Breast. 2016; 28: 29-36
        • D'Ettorre M.
        • Gniuli D.
        • Bracaglia R.
        • et al.
        Micro- and macroscopic structural modification of subcutaneous adipose tissue after bariatric surgery.
        Aesthetic Plast Surg. 2012; 36: 213-214
        • Bracaglia R.
        • D'Ettorre M.
        • Gentileschi S.
        • Mingrone G.
        • Tambasco D.
        Kris knife" brachioplasty after bariatric surgery and massive weight loss.
        Aesthetic Plast Surg. 2013; 37: 640-642
        • Bracaglia R.
        • Tambasco D.
        • Gentileschi S.
        • D'Ettorre M.
        Abdominoplasty in patient with penile implant.
        Aesthetic Plast Surg. 2011; 35: 933-935
        • Gentileschi S.
        • Servillo M.
        • Ferrandina G.
        • Salgarello M.
        Lymphatic and sensory function of the upper limb after brachioplasty in post-bariatric massive weight loss patients.
        Aesthetic Surg J. 2017; 37: 1022-1031
        • Gentileschi S.
        • Servillo M.
        • Albanese R.
        • De Bonis F.
        • Tartaglione G.
        • Salgarello M.
        Lymphatic mapping of the upper limb with lymphedema before lymphatic supermicrosurgery by mirroring of the healthy limb.
        Microsurgery. 2017; 37: 881-889
        • Bracaglia R.
        • D'Ettorre M.
        • Gentileschi S.
        • Tambasco D.
        Was the surgeon a satisfactory informant? How to minimize room for claims.
        Aesthetic Surg J. 2014; 34: 632-635