Consensus on definition and severity grading of lymphatic complications after kidney transplantation

The incidence of lymphatic complications after kidney transplantation varies considerably in the literature. This is partly because a universally accepted definition has not been established. This study aimed to propose an acceptable definition and severity grading system for lymphatic complications based on their management strategy.


Introduction
Lymphatic complications after kidney transplantation, such as lymphorrhoea and lymphocele, can be challenging.
Lymphocele is associated with morbidities such as abdominal discomfort, impaired wound healing, thrombosis and organ failure 1 . Lymphorrhoea/lymphoceles are usually diagnosed in retroperitoneal kidney grafts. Lymphoceles rarely develop in intraperitoneal kidney grafts 2 . The reported incidence of lymphatic complications, including lymphorrhoea and lymphocele, ranges between 0⋅6 and 51 per cent 2 -8 ; the peak incidence of lymphocele is during the sixth postoperative week (range 2 weeks to 6 months) 9 . This big difference in reported rates can be attributed to the lack of a standard definition of lymphatic complications after kidney transplantation, although postoperative diagnosis and follow-up protocols are available 10,11 .
Varying definitions of lymphatic complications 3,12 -14 mean that results from different studies cannot be compared. In the past decade, international study groups have introduced classifications of different postoperative complications in hepatopancreatobiliary surgery, such as bile leakage and pancreatic fistula 15 -19 . A standardized definition and grading system for lymphatic complications after kidney transplantation would allow multicentre clinical trials of their management and treatment.
The aim of this study was to review the literature for definitions of lymphatic complications, and for information about their incidence, clinical manifestation, diagnosis and management. The results of this literature review were used to propose a definition and practical severity grading system, based on management strategy, that could then be reviewed by a European consensus team from high-volume transplantation centres.

Literature search
The literature was reviewed systematically to identify articles that reported lymphatic complications after kidney transplantation, in accordance with PRISMA guidelines 20 . Two databases (MEDLINE and Web of Science) were searched systematically to identify relevant articles published between January 1985 and December 2018. Search terms included (kidney OR renal) AND (transplantation) AND (lymphocele OR lymph leakage OR lymphorrhoea OR lymph fistula OR lymphorragia). Reference lists of the retrieved articles were also searched for additional relevant publications.
Relevant studies and the definitions were identified and extracted. The abstracts were screened and the full texts of potentially relevant studies were obtained. The identified articles were double-checked independently by two authors, and any disagreements during selection, extraction and assessment resolved with other authors. All titles and abstracts were selected based on the predefined PICo (population, interest, context) eligibility criteria for qualitative studies. All studies reporting kidney transplantation in human subjects that reported lymphorrhoea/lymphocele, and including at least 150 procedures, were included in the present review. Any study design was included, except case reports, narrative or systematic reviews, study protocols, experimental studies, conference abstracts, letters and common overviews. Studies on lymphorrhoea/lymphocele after other urological, vascular or gynaecological procedures were also excluded. All articles were also screened to ascertain whether standard definitions of lymphorrhoea and lymphocele had already been settled by the authors.

Data extraction
Definitions of lymphorrhoea and lymphocele were extracted. The parameters used to define lymphorrhoea and lymphocele included: fluid collection, lymph content, histology, location, size and timing. The sample size, incidence of lymphatic complications, and volume or size cut-off of asymptomatic and symptomatic lymphoceles were recorded. In addition, the symptoms of patients with symptomatic lymphoceles, and information regarding the diagnosis (ultrasound imaging, CT, etc.) and treatment of lymphorrhoea/lymphocele (aspiration, external drainage, sclerotherapy and surgery) were recorded.

Assessment of methodological quality
The quality of studies was assessed independently by two authors using the methodological index for non-randomized studies (MINORS) 21 . Disagreements were resolved by consensus. MINORS includes a total of 12 items; the first eight are related to non-comparative studies and a further four items are applicable only to comparative studies. The items are categorized and scored as 0 (not reported), 1 (reported but inadequate) or 2 (reported and adequate). In assessment of non-comparative studies, the highest possible score is 16; studies with fewer than 8 points were deemed to be of low quality, studies with 8-12 points of intermediate quality and those with more than 12 points of high quality. In assessment of comparative studies, the highest possible score is 24; studies with fewer than 12 points were considered to be of low quality, those with 12-18 points of intermediate quality, and those with more than 18 points of high quality.

Final agreement
The extracted data were used to create a comprehensive definition of lymphorrhoea and lymphocele. In addition, a grading system that defined the severity of lymphorrhoea/lymphocele based on the required management strategies was suggested. Next, members of European high-volume transplantation centres (performing more than 100 kidney transplants per year) in Germany, Austria and Switzerland were invited to join a consensus team. The proposed definitions and severity grading system were reviewed by the consensus team, and a revised version was recirculated for further comments or final approval. After e-mail discussions, the definitions and grading system were approved by all members of the consensus team. As recommended by some consensus team members, ten clinical case examples with various clinical scenarios are presented to clarify uncertainties regarding application of the severity grading system.

Results
The literature search yielded 910 articles after removal of duplicates ( Fig. 1). Eighty-seven articles 4,5,9,11 -14,22-101 met the inclusion criteria and were included in the report. Seventy-six articles reported on lymphoceles only, four articles on lymphorrhoea, and seven articles on lymphocele and lymphorrhoea. Forty-eight studies had a sample size of 150-500 procedures, 27 studies a sample size of 500-1000 and 12 studies included more than 1000 kidney transplants.

Qualitative analysis
Sixteen studies were of intermediate quality and the remaining 71 of poor quality (Table S1, supporting information). The quality of the included studies was poor because of retrospective design, lack of control group and inadequate follow-up.

Common terminology used in the literature
'Post-renal/kidney transplantation lymphocele', 'lymphocele after renal/kidney transplantation' and 'lymph/ lymphatic leakage/lymphorrhoea/lymph fistula following renal/kidney transplantation' were the terms most commonly used to describe lymphatic complications after kidney transplantation. To standardize the terminology of these surgical complications, the consensus team proposes the terms 'post-kidney transplantation lymphorrhoea' and 'post-kidney transplantation lymphocele'.

Components used to define lymphatic complications after kidney transplantation
The proposed definitions from published papers are listed in Table 1. The systematic review confirmed that Fluid collection with electrolyte and creatinine levels similar to those of the patient's plasma no accepted definition exists for lymphorrhoea and lymphocele. Lymphorrhoea/lymphocele was not defined in more than 60 per cent of included articles. Thirty-two articles defined lymphorrhoea or lymphocele. The main indicators were: fluid collection in 20 articles (67 per cent), lymph content in 18, histology of the membrane in ten, location of lymphocele in 16, size of lymphocele in seven and timing of diagnosis in two articles (7 per cent). The reported rates of lymphocele ranged from 0⋅6 to 33⋅9 per cent (Table S2, supporting information).

Symptoms of lymphatic complications after kidney transplantation
Fifty-five of 87 articles (63 per cent) reported the incidence of symptomatic lymphoceles, and 20 (23 per cent) reported symptomatic and asymptomatic lymphoceles separately. They included general, visceral, vascular and renal symptoms (Table S3,

Radiological assessment
The method of radiological assessment was documented in 65 articles (75 per cent) (  22,23 .

Agreed definition of lymphatic complications after kidney transplantation
Fluid content, fluid volume and duration of leakage were included in the published definitions. Leakage was usually analysed biochemically to rule out urine, blood or pus when more than 30-50 ml of fluid leaked per day. However, some authors analysed the leakage only after 30-50 ml of fluid continued to leak daily after postoperative days 7-10. The consensus team proposed the following definition of lymphorrhoea after kidney transplantation: leakage of more than 50 ml fluid (not urine, blood or pus) per day from the drain, or the drain site after removal of the drain, for more than 1 week after kidney transplantation. Most published definitions of lymphocele included content of the fluid, histology of the surrounding membrane, location of the lymphocele and size of accumulated fluid. Terms frequently used to describe the surrounding membrane of a lymphocele included pseudocyst, cavity surrounded by pseudomembrane, or non-epithelialized cavity. There was no cut-off size that defined a fluid collection as a lymphocele. Fluid that accumulates after a kidney transplant may have different contents. It is generally agreed that the term lymphocele is used only when urinoma, haematoma and abscess have been ruled out. The consensus team defined lymphocele as: fluid collection of any size near to the transplanted kidney, after urinoma, haematoma and abscess have been excluded. This definition applies to both asymptomatic and symptomatic lymphoceles.

Proposed grading of severity of lymphatic complications after kidney transplantation
The severity of lymphorrhoea/lymphocele has not been graded consistently in the literature (Table S3, supporting information). One study 24 reported persistent lymphatic fluid leakage as removal of the drain delayed for more than 15 postoperative days. Focusing on lymphocele, most authors (20 articles) categorized lymphoceles as asymptomatic or symptomatic, according to the clinical symptoms. Presser and colleagues 25 classified lymphoceles into three groups according to treatment: open surgery, laparoscopic surgery and fibrin glue treatment. Only symptomatic lymphoceles were reported in 35 studies; lymphoceles were classified by size in three studies, and lymphoceles were not classified either by symptoms or size in 25 studies. The consensus team recommended a simple grading system to classify lymphatic complications based on the severity and invasiveness of the management strategy: grades A, B and C, where grade A requires the least invasive and grade C the most invasive management strategy ( Table 2).

Grade A lymphatic complications
Grade A complications need no treatment or are treated with aspiration. They have a minor and/or non-invasive impact on clinical management. Grade A lymphorrhoea may prolong the hospital stay. Drain removal may need to be postponed for more than 7 days because of increasing or constant drain outflow. Fluid may continue to leak after drain removal. Grade A lymphorrhoea should cease spontaneously. Grade A lymphoceles are usually diagnosed incidentally during routine ultrasound examination. Grade A lymphoceles do not progress and usually resolve after aspiration.

Grade B lymphatic complications
Grade B complications need non-surgical intervention. Leakage in grade B lymphorrhoea does not resolve spontaneously and requires further intervention, such as medical therapy (for example somatostatin), sclerosing agents or radiotherapy. Infected or complicated lymphorrhoea that needs antibiotic treatment is also classified as grade B. Patients with a grade B lymphocele usually present with clinically relevant symptoms, which are characterized as general, visceral, renal and/or vascular (Table S3, supporting information). Creatinine levels may rise in patients with grade B lymphocele. Grade B lymphoceles are managed non-surgically with percutaneous drainage with or without sclerotherapy, sclerotherapy alone, or radiation. Lymphoceles that are treated by repeated non-surgical intervention, and those that necessitate decompression of the ureter by insertion of a double-J stent, are also classified as grade B.

Grade C lymphatic complications
Grade C complications require invasive surgical intervention (open or laparoscopic). They include: persistent lymphorrhoea that needs surgical treatment; lymphoceles that are loculated or inappropriately located for non-surgical intervention; recur after drainage and/or sclerotherapy and need to be managed surgically; were not treated successfully by, or recurred after, surgical intervention; and are managed surgically during an operation for other co-morbidities or conditions (for example, incisional/umbilical hernia, fascial dehiscence, acute appendicitis). When non-surgical treatment of a grade A/B lymphocele leads to a complication that has to be treated surgically (such as bleeding or intestinal perforation) it should be also graded as C.
Examples A new grading system needs to be explained using clinical examples to facilitate its use in routine clinical practice.
To clarify the application of the proposed severity grading system, ten clinical examples are described in Table S5 (supporting information).

Discussion
Lymphatic complications, including lymphorrhoea and lymphocele, are common after kidney transplantation. Despite significant improvements in prevention, diagnosis and management, standard definitions and a severity grading system have not yet been established 5,10,11,13,102 . The aim of this report was to suggest standard definitions and a severity grading system for lymphatic complications based on management strategy, to facilitate comparison of results across studies. The goal was to propose neither clinical guidelines (treatment action plan), nor a management strategy.
Intraoperative drainage is commonly used after kidney transplantaion 2,26,103 . Although drainage decreases the incidence of lymphocele 10,26 , it can lead to lymphorrhoea. Lymph leakage can occur if a drain is inserted, or if a drain is not used. As a result, the peak incidence of lymphorrhoea and lymphocele is heterogeneous, and may be up to 6 months after kidney transplantation.
The proposed definitions and severity grading system are based on fluid content, fluid volume, duration of leakage and management, but not size. Lymphocele size can affect the development of symptoms, but these depend on the location of the lymphocele and size of the patient. Some small lymphoceles can disturb graft function; in contrast, some large ones exert no pressure on the graft and are asymptomatic. Therefore, size is not a reliable parameter for grading the severity of lymphoceles after transplantation. Most fluid collections are asymptomatic 27 , and in almost all instances asymptomatic collections require no treatment 5,9,13,25,28,29 . A rise in serum creatinine levels may not be associated with clinical symptoms. Atray and colleagues 3 reported raised creatinine levels in 60 per cent of patients with a lymphocele but without any symptoms. These patients normally required invasive diagnostic and therapeutic procedures 3 , so the lymphatic complications would be graded as B or C according to the proposed system, even though the patients were asymptomatic.
Lymphatic complications after kidney transplantation have not been graded before, except into asymptomatic and symptomatic categories 11,26,34,36 . Lymphoceles are not usually managed according to their size 37 , although Hamza and co-workers 14 treated all lymphoceles greater than 80 ml, irrespective of the symptoms, and Dubeaux et al. 38 operated on asymptomatic lymphoceles based on their size. Ulrich and colleagues 11 drained symptomatic lymphoceles between 3 and 5 cm, and operated on those with a diameter greater than 5 cm. Location and accessibility are also important factors. If the lymphocele is located deep within the pelvic cavity adjacent to the iliac vessels 5 or inferomedially 37 , aspiration may be difficult or impossible 40 , and surgery is the only possibility. Septation of the lymphocele also influences the therapeutic approach. Zargar-Shoshtari and co-workers 40 demonstrated that aspiration was possible for non-loculated lymphoceles, whereas loculated ones should be treated surgically 37,40,41 .
The severity of lymphatic complications should be graded according to required clinical management rather than size, symptoms, location or loculation. This is in accordance with suggested grading systems for other surgical complications 17,18,104 . No treatment is required for grade A lymphorrhoea, but lymphatic leak or hospitalization may be prolonged. After aspiration, grade A lymphoceles do not persist or recur. In contrast, grade B lymphatic complications affect the clinical course if drain insertion, drain replacement, sclerotherapy or radiotherapy is required. Surgically managed patients are exposed to anaesthetic and procedure-related complications; therefore, these should be classified as grade C.
Although the definitions and severity grading system for lymphatic complications after kidney transplantation presented here should minimize the discrepancy in reporting between centres, the lack of standards in radiological assessment should be noted as a potential limitation. The lymphocele rate may vary between centres because of differences in diagnostic tools and radiological assessments, and may be underdiagnosed or overdiagnosed as a result.
The proposed definitions and severity grading of lymphatic complications after kidney transplantation are the result of a consensus between large-volume European centres. The severity grading approach is based on management strategy, and does not consider size, symptoms or location, while treatment selection mainly depends on the patient's clinical condition, the surgeon's preference and skills and centers' policies. The proposed definition and grading system does not aim to suggest a treatment action plan, or a management strategy, but represents the effect of chosen treatment strategy, regardless of why it was selected. The aim was to standardize reported results and to allow comparison between studies. The proposed definitions and grading system should be validated in future studies.