Microbiology of Pelvic Lymphocyst Infection after Lymphadenectomy for Malignant Gynecologic Tumors

Abstract

Background:

Pelvic lymphocyst infection is a rare complication after lymphadenectomy for malignant gynecologic tumors. Although medical therapy is a useful addition to surgical drainage, the appropriate antibiotic regimen is unknown because few studies have examined the causative organisms. The purpose of this case series was to identify the micro-organisms infecting pelvic lymphocysts.

Methods:

This was a single-center, retrospective, case-series review conducted at a tertiary-care cancer center between October 2002 and March 2013. The participants included all patients who experienced their first pelvic lymphocyst infection after undergoing pelvic lymphadenectomy for cervical, endometrial, or ovarian cancer and exhibited positive lymphocyst fluid culture. Computed tomography- or sonography-guided percutaneous aspiration procedures were performed to obtain lymphocyst fluid for culture.

Results:

During the study period, 878 patients underwent lymphadenectomy for gynecologic malignant tumors, and 13 developed a pelvic lymphocyst infection documented microbiologically. Cultures identified Staphylococcus aureus (three patients), S. epidermidis (one patient), Streptococcus agalactiae (three patients), Enterococcus (two patients), Escherichia coli (one patient), and anaerobic bacteria (three patients). They were all monomicrobial infections.

Conclusions:

Our study and other smaller ones suggest that lymphocyst infections following pelvic lymphadenectomy for malignant gynecologic tumors usually are monomicrobial and caused by gram-positive cocci, including Staphylococcus, Streptococcus, and Enterococcus, and anaerobes such as Bacteroides fragilis. These bacteria should be considered when selecting empiric antibiotic therapy.

Pelvic and para-aortic lymphadenectomies are performed as staging and therapeutic procedures in patients with malignant gynecologic tumors. A pelvic lymphocyst, defined as a collection of lymphatic fluid in the retroperitoneal space, is a well-known complication of pelvic lymphadenectomy [1]. Most lymphocysts are believed to result from operative disruption and inadequate ligation of lymphatic trunks [2].

Although lymphocysts are considered widely to be of little clinical importance, the contents can become infected via hematogenous or lymphatic dissemination, contiguous spread from adjacent tissues, or direct inoculation as a result of surgery [3,4]. Two studies have found that the infection incidence rate is less than 1% [5,6]. This infection typically is not life-threatening, but it causes symptoms such as fever and lower abdominal pain. Medical therapy is useful when combined with surgical drainage; however, the appropriate antibiotic regimen is unknown because few data are available on the causative organisms. Case reports suggest the causative organisms are primarily gram-positive cocci and anaerobic bacteria [7 –10]. The purpose of this case series was to describe the microbiologic causes of pelvic lymphocyst infections.

Patients and Methods

The study was a single-center, retrospective, case-series review of patients who developed a culture-documented pelvic lymphocyst infection after undergoing lymphadenectomy for malignant gynecologic tumors at the Shizuoka Cancer Center Hospital between October 2002 and March 2013. All study participants provided informed consent, and the study design was approved by the hospital's ethics review board.

The participants consisted of all patients who experienced their first pelvic lymphocyst infection after undergoing pelvic lymphadenectomy with or without para-aortic lymphadenectomy for cervical, endometrial, or ovarian cancer and who exhibited a positive lymphocyst fluid culture. Regarding the surgical procedure, the pelvic peritoneum was left open, and no drains were placed retroperitoneally. The diagnosis of pelvic lymphocyst infection was based on physical findings (tenderness at the lymphocyst site) and the results of sonography or computed tomography (CT), which showed a low-attenuation lesion in the pelvic spaces with a thick capsule (Fig. 1). Computed tomography- or sonography-guided percutaneous aspiration was performed for diagnosis and therapy. Lymphocyst fluid specimens for culture were obtained during the aspiration procedures. Patients were excluded if the fluid culture was negative or not performed. Data collected from the medical records included demographic information, microbiologic data, medical and surgical management, and outcomes for an epidemiologic description of the infection.

FIG. 1.

Computed tomography scan of pelvis of patient with post-operative lymphocyst infection. Note wall thickening at right (arrow). This was one of the criteria used to define this lymphocyst as infected.

All the cultures used to evaluate lymphocyst infection were performed by the Microbiology Laboratory at the Shizuoka Cancer Center. The Vitek 2 (bioMérieux, Marcy l'Etoile, France) and the MicroScan WalkAway (Siemens, Germany) instruments were used for automated bacterial identification and susceptibility testing. The Streptococcal Grouping Kit (Oxoid, Thermo Scientific, Waltham, MA), which is a rapid latex agglutination test, was performed to identify streptococcal groups. Identification of anaerobic bacteria was performed using a gram-stained smear, colony properties, and biochemical characteristics.

Results

During the study period, 878 patients underwent lymphadenectomy for malignant gynecologic tumors, and 13 had a documented pelvic lymphocyst infection. The median age of these patients was 57 y (range 29–64 y). Six patients had cervical cancer, four had endometrial cancer, and three had ovarian cancer. Three patients underwent pelvic and para-aortic lymphadenectomy, and 10 patients underwent pelvic lymphadenectomy alone. Four patients had received chemotherapy within the previous 30 d. The median time from surgery to the onset of infection symptoms was 72 d (range 12–1,120 d).

Culture of the infected lymphocyst fluid revealed Staphylococcus aureus (three patients), S. epidermidis (one patient), Streptococcus agalactiae (three patients), Enterococcus spp. (two patients), Escherichia coli (one patient), Bacteroides fragilis (one patient), and Peptostreptococcus spp. (two patients). They were all monomicrobial infections. All S. aureus isolates were sensitive to methicillin. In one patient, S. aureus was isolated from both the lymphocyst fluid and the blood.

The 13 patients received a spectrum of antibiotics, and the median duration of antibiotic therapy was 24 d (range 12–56 d). In 10 patients, antibiotic administration was initiated before the lymphocyst fluid was sampled. Two patients had concurrent febrile neutropenia and received antipseudomonal β-lactam agents as empiric therapy. All patients were switched from intravenous to oral antibiotics once the symptoms, such as fever and right or left lower quadrant abdominal pain, had subsided. Twelve patients had continuous drainage, and one patient underwent a one-time drainage procedure because the contents of the lymphocyst were viscous and thus would not drain well. The median duration of drainage was 6 d (range 0–18 d). All 13 patients showed complete recovery.

Discussion

Pelvic lymphocyst infection is a rare complication after lymphadenectomy for malignant gynecologic tumors. Although medical therapy and surgical drainage are effective, the appropriate antibiotic therapy remains uncertain because of a lack of microbiologic data. To our knowledge, this series of 13 patients is the largest published case series. Our findings demonstrated that this infection is monomicrobial and is caused predominantly by gram-positive cocci and anaerobic bacteria.

Few reports have illustrated the bacterial etiology of pelvic lymphocyst infections in these patients. Takeuchi et al. reported the isolation of S. aureus (two patients), S. agalactiae (one patient), S. anginosus (one patient), B. fragilis (one patient), and unidentified anaerobes (two patients) from lymphocyst fluid [7]. One patient had a polymicrobial infection consisting of S. agalactiae and B. fragilis. In other case reports, lymphocyst culture showed S. aureus, coagulase-negative Staphylococcus, and E. faecium [8 –10]. These microbiologic findings are similar to our results and indicate that most of the microbes implicated in pelvic lymphocyst infection originate from the normal skin or urogenital flora.

As ubiquitous skin commensals, coagulase-negative staphylococci such as S. epidermidis are common contaminants found after percutaneous procedures. In clinical practice, it is important to distinguish between true infection and contamination to prevent unnecessary prescription of antimicrobial agents. However, there is no diagnostic reference standard for making this discrimination. In our case, S. epidermidis was considered to be causing a true infection because of the presence of numerous gram-positive cocci with neutrophils in a gram-stained smear.

Antibiotics are useful in the treatment of infected lymphocysts that follow pelvic lymphadenectomy, particularly in cases where surgical drainage is insufficient because of cysts with septae or high-viscosity fluid. Although the most efficacious regimen is unknown, the microbiologic etiology supports the value of antibiotic therapy. Because gram-positive cocci and anaerobes are the most common causes of pelvic lymphocyst infection, we consider monotherapy with either ampicillin-sulbactam or amoxicillin-clavulanic acid to be an appropriate empiric antibiotic choice. Antipseudomonal treatment should be considered in patients with concurrent chemotherapy-induced neutropenia [11]. The appropriate duration of therapy is unknown, but almost certainly is shorter than that reported herein if drainage is adequate.

Material obtained from CT- or sonograhy-guided aspiration should be sent to the laboratory for both aerobic and anaerobic culture. Anaerobic cultures should be requested specifically. Detecting organisms and their antibiotic sensitivity is useful for de-escalating from empiric to targeted therapy, which helps avoid toxicity, treatment failure, and the emergence of antimicrobial resistance.

A limitation of our study is the relatively small sample for describing the microbial etiology. However, the incidence of this infection is too low for one medical center to collect a large data set. Multicenter trials would be a valid approach to resolving this problem. Another limitation of our study is the administration of antibiotics in most patients before the lymphocyst fluid samples were obtained. Some of the infecting microbes may not have been isolated because of the these antibiotics. To exclude the possibility of the presence of more bacteria, future prospective studies need to consider the timing of antibiotic therapy.

In conclusion, our study and other smaller studies suggest that lymphocyst infection following pelvic lymphadenectomy for malignant gynecologic tumors usually is monomicrobial and caused by gram-positive cocci, such as Staphylococcus, Streptococcus, Enterococcus, or anaerobes such as B. fragilis. These micro-organisms should be considered when designing empiric antibiotic therapy. More data need to be collected, and the impact of empiric therapy should be examined.

Footnotes

Acknowledgments

The authors thank Munetaka Takekuma and Akira Tanaka, Division of Gynecology, Shizuoka Cancer Center Hospital, for their assistance in the collection of data.

Author Disclosure Statement

No competing financial interests exist.

References

Charoenkwan

, Kietpeerakool

. Retroperitoneal drainage versus no drainage after pelvic lymphadenectomy for the prevention of lymphocyst formation in patients with gynaecological malignancies. Cochrane Database Syst Rev, 2010:CD007387.

White

, Mueller

, Ferrucci

Jr , et al. Percutaneous drainage of postoperative abdominal and pelvic lymphoceles. AJR Am J Roentgenol, 1985; 145:1065–1069.

Achouri

, Huchon

, Bats

, et al. Postoperative lymphocysts after lymphadenectomy for gynaecological malignancies: Preventive techniques and prospects. Eur J Obstet Gynaecol Reprod Biol, 2012; 161:125–129.

Lentz

, Axtell

, Vasilev

. Gynecologic Oncology: Evidence-Based Perioperative and Supportive Care. Second Edition. Hoboken, NJ. Wiley-Blackwell, 2011.

Kavallaris

. Management of symptomatic pelvic lymphocyst after radical pelvic or pelvic and paraaortic lymphadenectomy for cervical and endometrial cancer. Gynecol Surg, 2009; 6:345–349.

Tam

, Lam

, Chan

, et al. Natural history of pelvic lymphocysts as observed by ultrasonography after bilateral pelvic lymphadenectomy. Ultrasound Obstet Gynecol, 2008; 32:87–90.

Takeuchi

, Toyota

, Nakamura

, et al. Percutaneous catheter drainage of postoperative pelvic lymphoceles. JIVR, 2010; 25:476–482.

Kew

, Noble

, Sandhu

. Infection of a lymphocyst after pelvic lymph node dissection for endocervical adenocarcinoma. J Obstet Gynaecol, 2012; 32:606.

Imafuku

, Yoshida

, Sato

, et al. A case of lymphocyst infection caused by vanB type VRE. Kansen Zasshi, 1999; 73:473–476.

10.

Saunders

, Paterson

MEL

. Infected lymphocyst following pelvic lymphadenectomy. J Obstet Gynaecol, 1988; 9:159–160.

11.

Freifeld

, Bow

, Sepkowitz

, et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis, 2011; 52:e56–e93.