Key words
lymphocysts - gynecologic malignancies - lymphadenectomy
Schlüsselwörter
Lymphzyste - gynäkologisches Malignom - Lymphadenektomie
Introduction
Pelvic lymphadenectomy is a common procedure for evaluating lymph node involvement
in patients with gynecological cancer. Although this procedure provides important
information, various complications, such as hemorrhage, increased operative time,
and lymphocysts can occur [1]. Formation of a lymphocyst is one of the most common complications of lymphadenectomy
[1]. A lymphocyst is a thick-walled cystic mass filled with lymphoid fluid that results
from disruption of lymphatic channels and is generally encountered during uro-gynecological
oncology surgeries or during the post-renal transplantation period [2], [3], [4]. Mori et al. reported formation of lymphocysts in 68 patients with cervical cancer
who underwent lymphadenectomy in 1955 [5].
Although lymphocysts are generally asymptomatic and are found incidentally during
postoperative follow-up, they may severely affect a patientʼs life by obstructing
the ureter (hydronephrosis), bowel (ileus), or vessels (thrombosis), and they may
become infected [3], [6], [7], [8]. Lymphocysts may also adversely delay treatment of the primary disease. The reported
incidence of lymphocysts in patients undergoing gynecological cancer surgery is 1–58 %
[3], [5], [6], [7], [8], [9], [10]. The wide variation of the reported incidence might be explained partially by the
presence of symptoms, type of surgical techniques used in treatment (laparatomy, laparoscopy,
or robotic surgery) and diagnostic tools used for detection (ultrasonography, CT,
or MRI). The data about the highest lymphocyst formation in which gynecological cancer
occurs is scant and controversial. In a study by Kim et al. [8], the highest risk was reported to be in cervical cancer, whereas the highest rate
was reported in ovarian cancer by Zikan et al. [10].
Various factors are associated with lymphocyst formation, such as the number of lymph
nodes excised, lymph node involvement, surgical technique, anti-coagulant use, and
postoperative radiotherapy (RT) and chemotherapy (CHT) [3], [7], [8], [9], [10], [11]. Although lymphocysts are the most common complication of lymphadenectomy, a limited
number of studies have evaluated lymphocysts discovered during gynecological surgery.
Most studies have been performed by urological and transplant surgeons, and the rest
of the studies have controversial data about the risk factors. In order to prevent
lymphocyst formation, several recommendations have been proposed such as drainage
of resected sites, use of different energy sources for lymphadenectomy, non-closure
of the peritoneum [3], [11].
In the present study, we evaluated and defined the risk factors for formation of a
lymphocyst in patients with various gynecological cancer types in whom a lymphadenectomy
was performed.
Materials and Methods
Study protocol
This retrospective case–control study was performed on 206 patients, who underwent
surgery and were followed by the department of gynecological oncology at Izmir Tepecik
Research and Training Hospital between January 2011 and March 2015 after Institutional
Review Board approval. Of the 206 patients, 100 were diagnosed with a lymphocyst,
and 106 were assigned to a control group. The control group patients were selected
based on the ratio of retroperitoneal lymphadenectomy, number of pelvic (P) and para-aortic
lymph (PA) nodes, and tumor histology type. Laboratory findings and surgical characteristics
of the patients were obtained from medical records.
Management and follow-up of gynecological cancer in the local institution
Diagnosis and treatment of the gynecological cancers (endometrium cancer, cervix cancer,
ovarian tumor, etc.) were performed according to our clinic guidelines. Management
of endometrial cancer varied among practitioners, particularly with respect to the
role of lymphadenectomy; no lymph nodes were sampled in some patients, only the P
or PA nodes were sampled in other patients, complete staging with bilateral P lymph
node dissection (LND) was applied in some patients, and some patients underwent complete
staging with bilateral P and PA LND. The practitioners were responsible for these
differences during the study. We performed retroperitoneal lymphadenectomy up to the
renal vein or only harvested bulky lymph nodes in patients with ovarian cancer who
underwent neoadjuvant CHT. We harvested P and PA lymph nodes to the level of the inferior
mesenteric artery or common iliac bifurcation in patients with cervical cancer. Unipolar
cautery and/or ligature were used to dissect the lymphatic tissues. Additionally,
absorbable sutures or hemoclips were used to ligate lymphatic channels during dissection,
if necessary. Intra-abdominal drains were placed if necessary to obtain early data
about postoperative bleeding or leakage. Blood stopper was used in patients who underwent
removal of P lymph nodes up to the left renal vein to evaluate whether blood stopper
reduced the incidence of chylous acid (unpublished case–control study). No hemostatic
or sealant agent, such as fibrin glue, was used after LND. Intra-abdominal drains
were placed if necessary to obtain early data about postoperative bleeding or leakage.
Low-molecular weight heparin (LMWH) prophylaxis is started in our hospital 12 hours
prior to surgery and continues until the patient is discharged or at postoperative
week 3–4. The dose and duration of LMWH were calculated according to body mass index
(BMI) and any patient risk factors (history of emboli, atrial fibrillation, or cardiovascular
disease). We used dose (mg/kg) × duration (days) of prophylaxis to estimate the total
dose administered, as this method provides more precise information than either dose
or duration alone.
Patients with cervical cancer are typically seen in our clinic every 3 months for
the first 2 years, every 6 months for the next 3 years, and yearly thereafter. Visits
for patients with endometrial, ovarian, and fallopian tube cancers are scheduled every
3 months during the first year, every 4 months during the second year, every 6 months
during years 3–5, and yearly thereafter.
Some of the patients were managed postoperatively with CHT alone, radiotherapy (RT)
alone, or with both CHT and RT. RT was given as either external radiotherapy (ERT)
alone, brachytherapy (IRT) alone, or ERT with IRT.
Diagnosis and diagnostic tools
The lymphocyst diagnosis was made based on findings of a fluid-filled cystic structure
of varying shape, structure, and echogenicity (uni- or multi-locular, septated, etc.)
as detected on trans-vaginal, trans-abdominal ultrasonography. The location, size
in three dimensions (anteroposterior × transverse × caudal), shape (round or oval),
echogenicity (hyper or hypo), and proximity to internal organs were evaluated by a
radiologist who was an expert in gynecological ultrasonography. In cases of doubt
or need for further details, computed tomography (CT) or magnetic resonance imaging (MRI) was ordered.
Statistical analysis
The χ2 test and Studentʼs t-test were used for the statistical analysis of unpaired data.
A logistic regression analysis was conducted to determine factors affecting formation
of a lymphocyst, and the results are presented as odds ratios (ORs) and 95 % confidence
intervals (CIs). All statistical analyses were performed using MedCalc software ver.
11.5 for Windows (MedCalc Software, Inc., Ostend, Belgium). A p-value < 0.05 was considered
significant.
Results
Demographic characteristics and laboratory findings of the groups
A total of 206 patients were analyzed, and 106 patients without lymphocysts were selected
as the control group. The demographic and surgical characteristics of the groups are
shown in [Table 1]. No differences were observed in age, pre-operative hemoglobin; platelet, white
blood cell, and lymphocyte counts; or preoperative albumin level (p = 0.315, 0.500,
0.525, 0.683, 0.740, and 0.97, respectively). A significant effect of the heparin
dose × heparin days interaction and lymphocyst formation (5.8 ± 5.58 in the control
group, 10.3 ± 7.2 in the lymphocyst group) was observed (p = 0.002).
Table 1 Demographic characteristics and laboratory findings of the control and lymphocyst
groups.
|
Variable
|
Control group
|
Lymphocyst group
|
p-value
|
|
* Preoperative hemoglobin, preoperative white blood cell count, § preoperative platelet count, € low molecular weight heparin
|
|
Age (years)
|
54.9 ± 10.6
|
55.9 ± 9.6
|
0.315
|
|
Preop. Hg*
|
11.8 ± 1.4
|
12.1 ± 1.3
|
0.500
|
|
Preop. WBC
|
7 947 ± 2 398
|
8 091 ± 2 259
|
0.525
|
|
Preop. Lymphocytes
|
2 001 ± 738
|
2 104 ± 870
|
0.683
|
|
Preop. Plt§
|
302 600 ± 114 602
|
301 150 ± 110 226
|
0.740
|
|
Preop. Albumin
|
2.8 ± 0.4
|
2.8 ± 0.4
|
0.197
|
|
LMWH€ (dose) × days
|
5.8 ± 5.58
|
10.3 ± 7.2
|
0.0001
|
Results about the cancer types
Cancer types and their percentages are given in [Table 2]. No differences were found in cancer type between the lymphocyst and control groups
(p = 0.058). Of the 100 patients with lymphocysts, 49 (49 %) had ovarian cancer, 34
(34 %) had endometrial cancer, 12 (12 %) had cervical cancer, and 5 (5 %) had uterine
sarcoma. Lymphocysts were most frequently detected in the ovarian cancer subgroup
(49 %). The number of resected lymph nodes and positive lymph node counts are shown
in [Table 3]. The numbers of P (22.2 ± 10.2 in the control group, 24.1 ± 11.3 in the lymphocyst
group) and PA lymph nodes resected (15.0 ± 10.8 in the control group, 16.2 ± 11.1
in the lymphocyst group) and the numbers of positive P (0.8 ± 2.8 in the control,
1.0 ± 3.0 in the lymphocyst group) and PA lymph nodes (0.9 ± 2.9 in the control, 0.7 ± 2.1
in the lymphocyst group) were similar between the two groups (p = 0.228, 0.674, 0.416,
and 0.584, respectively).
Table 2 Cancer types and ratios in the control group and lymphocyst group.
|
Cancer type
|
Control group
|
Lymphocyst group
|
p-value
|
|
Cervical
|
12 (11.3 %)
|
12 (12 %)
|
0.058
|
|
Endometrial
|
55 (51.9 %)
|
34 (34 %)
|
|
|
Ovarian
|
34 (32.1 %)
|
49 (49 %)
|
|
|
Sarcoma
|
5 (4.7 %)
|
5 (5 %)
|
|
Table 3 Resected numbers (#) of pelvic lymph nodes (PLN)/para-aortic lymph nodes (PALN),
and number of positive pelvic lymph nodes (PPLN)/positive para-aortic lymph nodes
(PPALN) in the control group and lymphocyst group.
|
Lymph node type
|
Control group
|
Lymphocyst group
|
p-value
|
|
PLN
|
22.2 ± 10.2
|
24.1 ± 11.3
|
0.774
|
|
PALN
|
15.0 ± 10.8
|
16.2 ± 11.1
|
0.890
|
|
PPLN
|
0.8 ± 2.8
|
1.0 ± 3.0
|
0.514
|
|
PPALN
|
0.9 ± 2.9
|
0.7 ± 2.1
|
0.375
|
Results about the surgical techniques and other treatment modalities (CT & RT)
The types and statuses of the patientʼs surgeries and treatments (omental resection,
pelvic drain, and administration of CHT only, RT only, or CHT and RT) in the two groups
are shown in [Table 4]. Significant differences were detected between the groups in the percentages of
patients who underwent CT only and RT only treatments (p = 0.001 and 0.002, respectively).
Table 4 Relationships between various factors (status of omentum resection, pelvic drains,
and chemo/radiotherapy [CHT/RT] administration) and lymphocyst formation.
|
|
Control group
n (%)
|
Lymphocyst group
n (%)
|
p-value
|
|
Omentum resection
|
48 (45.3 %)
|
52 (52 %)
|
0.403
|
|
Drain
|
92 (86.8 %)
|
95 (95 %)
|
0.110
|
|
CHT only
|
25 (23.6 %)
|
46 (46 %)
|
0.001
|
|
RT only
|
45 (42.5 %)
|
23 (23 %)
|
0.002
|
|
CHT and RT
|
23 (21.7 %)
|
20 (20 %)
|
0.449
|
The logistic regression analysis
The logistic regression analysis revealed a relationship between the LMWH dose × days
interaction and formation of a lymphocyst (OR, 1.10; 95 % CI, 1.0–1.13; p = 0.01)
([Table 5]). However, the significant relationship between the RT only and CT only treatments
and lymphocyst formation disappeared in the logistic regression analysis (OR, 2.1;
95 % CI, 0.6–7.1; p = 0.32; OR, 1.7; 95 % CI, 0.5–5.7; p = 0.32, respectively).
Table 5 Logistic regression analysis of the risk factors.
|
Variables
|
Odds ratio
|
95 % confidence interval
|
p-value
|
|
LMWH, low molecular weight heparin; PLN, pelvic lymph node; PALN, para-aortic lymph
node; PPLN, positive pelvic lymph node; PPALN, positive para-aortic lymph node; CHT/RT,
chemo and radiotherapy; CHT, chemotherapy alone; RT, radiotherapy alone; lymph, lymphocyte
count; plt, platelet count.
|
|
Albumin
|
1.5
|
0.6–4.0
|
0.32
|
|
LMWH dose × days
|
1.10
|
1.0–1.13
|
0.01
|
|
Cancer type
|
1.3
|
0.8–2.0
|
0.26
|
|
Omentum resection
|
1.15
|
0.5–2.3
|
0.68
|
|
# PLN
|
1.0
|
0.9–1.0
|
0.48
|
|
# PALN
|
1.0
|
0.95–1.0
|
0.54
|
|
# PPLN
|
1.0
|
0.8–1.2
|
0.66
|
|
# PPALN
|
1.1
|
1.0–1.4
|
0.36
|
|
CHT/RT
|
1.1
|
0.3–3.8
|
0.90
|
|
CHT only
|
1.7
|
0.5–5.7
|
0.32
|
|
RT only
|
2.1
|
0.6–7.1
|
0.32
|
|
Preop. lymph
|
1.5
|
0.8–1.5
|
0.35
|
|
Preop. plt
|
1.0
|
0.9–1.0
|
0.97
|
Discussion
Lymphocysts, which are also called lymphoceles or cystic masses filled with lymph,
are primarily detected after a renal transplantation or urological and gynecological
surgery that includes excision of lymph nodes [3], [11]. It is one of the most common complications of lymphadenectomy. Although lymphocysts
are generally asymptomatic, they adversely affect treatment and may make patients
and surgeons anxious.
The incidence of lymphocysts varies between 1 and 58 % [3], [5], [6], [7], [8], [9], [10]. The wide reported variation is probably due to different surgical techniques (laparatomy,
laparoscopy, or robotic surgery) and diagnostic modalities used for detection (ultrasonography,
CT, or MRI).
Several risk factors have been associated with lymphocyst formation, including BMI,
gynecological cancer type, lymphadenectomy type (P or PA), number of positive lymph
nodes, and surgery type (laparatomy, laparoscopy, or robotic surgery) [3], [8], [10], [11]. However, data regarding these risk factors are controversial, and a very limited
number of prospective clinical trials have evaluated them.
Kim et al. [8] reported that the highest lymphocyst formation rate was found in patients with cervical
cancer, whereas Zikan et al. [10] reported the highest incidence in patients with ovarian cancer. Zikan et al. indicated
that the difference is due to the complexity of the surgery performed to treat ovarian
cancer. Although a difference was detected among the ovarian, endometrial, and cervical
cancer subgroups in the incidence of lymphocyst formation, it was not statistically
significant. As in the study by Zikan et al., the highest incidence we found was in
patients with ovarian cancer, which may have been due to the greater number of lymph
nodes removed and the difficulty of the surgery in these patients.
Data about the relationship between RT and lymphocyst formation are controversial
[3], [8], [10], [11]. Kim et al. [8] reported a higher incidence of lymphocysts in patients who received RT, whereas
Zikan et al. [10] and Achouri et al. [11] reported no association between RT and an increased incidence of lymphocysts. In
the present study, a significant relationship was found between RT and the development
of lymphocysts, but the significance disappeared in the logistic regression analysis.
Thus, controversy continues with regard to positive lymph nodes removed and the formation
of lymphocysts. Although Zikan et al. [10] and Petru et al. [9] reported a significant relationship between positive lymph nodes and formation of
lymphocysts, Achouri et al. [11] found no relationship between the number of positive lymph nodes and formation of
lymphocysts. We also did not detect any relationship.
Several recommendations have been proposed to prevent lymphocysts, including non-closure
of the peritoneum, an open-vaginal vault, drainage of resected sites, use of different
energy sources for lymphadenectomy, and postoperative use of octreotide [3], [11].
The omentum protects the peritoneal cavity and plays a role in the turnover of peritoneal
fluid, which has been assumed to be effective for preventing lymphatic complications
and formation of lymphocysts. Several studies have evaluated the role of the omentum
in preventing the formation of lymphocysts in patients with endometrial cancer. However,
the sample sizes in these studies were very small (n = 22 and 64) [12], [13]. We found no difference in the rate of lymphocyst formation in patients who did
and did not undergo omentectomy.
Although drains have been recommended, despite the lack of evidence for their role
in preventing formation of lymphocysts, and data about their effects are controversial
[3], [8], [11]. In the present study, we found no association between drain placement or the number
of drains and lymphocyst formation.
Most studies about lymphocyst formation and risk factors for their formation have
been performed by urologists, and some possible risk factors have not been investigated
in gynecological cases, such as heparin use. The association between use of LMWH and
formation of lymphocysts has been investigated in several urological studies [14], [15]. Sogani et al. reported that the incidence of lymphocysts was 19 times higher in
patients who used prophylactic heparin compared with that in a control group [15]. However, Schmitges et al. evaluated the effect of LMWH on lymphocyst formation
in patients who underwent prostatectomy and found no association between lymphocyst
formation and heparin use [14]. Heinz et al. investigated the role of fibrin glue on the formation of lymphocysts
in 47 patients with gynecological cancers [16], and reported no association between duration of heparin administration and lymphocyst
formation. However, their study was designed to investigate the effect of fibrin on
lymphocyst formation and did not report the dose or type of heparin used. In the present
study, we found a significant relationship between heparin dose × days and lymphocyst
formation. LMWH probably increases the risk for lymphocysts by altering tissue hemostasis
after removing lymph nodes. The combination of the dose selected and duration of LMWH
use may have an effect on lymphocyst formation.
The retrospective nature of the present study was the biggest limitation. Although,
control group patients were selected randomly in order to prevent any bias, there
are still differences within and between the groups with regards to cancer types,
surgical and treatment types. Despite to these limitations, we investigated various
risk factors that have not been fully evaluated previously, such as drains, omental
resection, and CHT/RT, as well as factors that have not been investigated in patients
with gynecological cancers, such as use of LMWH.
Conclusion
In the present study, the association between total LMWH dose administered and the
formation of lymphocysts in patients with gynecological pelvic cancer was investigated
for the first time. Although anticoagulation with LMWH is essential for preventing
thromboembolism, it should be used appropriately to prevent other complications, such
as bleeding and lymphocysts. A risk estimate analysis should be performed on prophylactic
heparin use and the formation of lymphocysts in further randomized multicenter clinical
studies.
