Objectives: Drains are used routinely in many centers at the conclusion of kidney transplant, despite a paucity of evidence to guide practice in kidney transplant. Studies have not shown benefit from prophylactic drain placement following other major abdominal and vascular operations, and usage is consequently declining. Our aim was to understand practice patterns and rationale for behavior in drain placement and management in kidney transplant.
Materials and Methods: We conducted an online survey of surgeons who routinely perform kidney transplants across Australia and New Zealand.
Results: The response rate was 66% (43/66). Of respondents, 61% reported routine drain insertion, whereas 21% seldom inserted drains. Concerns about bleeding and anticoagulation (63%) and routine practice (58%) were the dominant reasons for drain insertion. The factors selected as most significant in determining drain removal were both volume and time (44%) and volume alone (33%). A volume of < 50 mL/day (51%) was the most commonly reported threshold for removal. The postoperative period of days 3 to 5 was the most commonly selected time point for drain removal (63%). Seventy-four percent of respondents would consider enrolling their patients in a randomized controlled trial to determine the benefits and harms of drain insertion.
Conclusions: Although drain insertion is a common practice, transplant surgeons in Australia and New Zealand reported sufficient uncertainty concerning the potential benefits and harms to warrant design and conduct of a randomized controlled trial.
Key words : Drain insertion, Kidney transplant, Surgeon preference
Kidney transplant is conducted on an increasingly elderly and multimorbid population of people with end-stage kidney disease. Outcomes have remained stable or have improved,1 and this is attributed to factors that include improvements in recipient and donor selection and perioperative care and better immunosuppressive regimens. Although vascular or urologic surgical complications are rare in kidney transplant recipients, surgical site complications such as infection, dehiscence, or the development of perigraft fluid collections are common. Although wound complications do not seem to have a strong negative effect on long-term patient or graft survival rates, these do contribute considerably to postoperative morbidity, expenditure, and patient-reported quality-of-life outcomes.1-4
Drains are frequently inserted by surgeons in the extraperitoneal space surrounding the graft.5 The rationale for this practice is unclear; however, given that the incidence of clinically significant hemorrhage or urine leak is low, drainage of lymph seems the most reasonable rationale.5-7 Perigraft fluid collections are reported to occur in 0.6% to 51% of patients after kidney transplant. Most perigraft collections are not associated with graft dysfunction or symptoms and are detected incidentally on routine ultrasonographic imaging.8 The substantial variations of the incidence reported in the literature regarding perigraft collections are due to variable posttransplant surveillance imaging protocols and the lack of any specific definition of such collections considered significant, particularly in terms of volume.9-17
Data on prophylactic drainage following vascular,18 thyroid,19 gastrointestinal, and breast cancer surgery20-22 have not substantiated a benefit to routine drainage. Moreover, prophylactic drain placement may increase the risks of surgical site infection, cause postoperative pain, and delay discharge.20,21,23,24 In other areas of surgery, such as reduction mammoplasty,25 and even oncological breast surgery and axillary dissection,26 surgical practice patterns with respect to drains are often heterogeneous, and the practice of drain insertion persists, despite substantial evidence of lack of benefit associated with drain insertion. In kidney transplant, only a small number of comparative studies of drain insertion and omission have been reported, and these studies have provided conflicting results.1,5,27-29
Notwithstanding these data on the use of prophylactic drains in other contexts, the immunosuppressed state in transplant may promote the incidence of surgical site complications such as fluid accumulation,1,3,4,7,30-33 and so extrapolating data from other surgery contexts to the context of the kidney transplant population may be misleading. There is no internationally accepted guideline or established consensus regarding the use of prophylactic drains in kidney transplants. However, as was seen recently in a case presented to the Supreme Court of New South Wales, “expert” witnesses frequently make claims regarding what is considered “standard of care” by a surgical community, even when this standard seems contrary to the evidence.34,35 As a consequence, we wished to clarify what practice with respect to prophylactic drain insertion was considered standard in Australia and New Zealand (ANZ) for surgeons performing kidney transplant procedures. We used this as an opportunity to garner further information about attitudes, rationale for drain usage, and patterns of management.
Materials and Methods
A web-based online survey was created via a secure online platform, known as Research Electronic Data Capture (REDCap).36 Ethics approval for this study was obtained from the Research Ethics and Governance Office of the New South Wales Sydney Local Health District. The survey was designed by practicing transplant surgeons, and the questions included information about their training background, transplant experience, the routine use of drains in kidney transplant, and the rationale for drain usage. The survey was designed to determine the demographics of the surgeons, including their field of subspecialty and breadth of experience. The practice of prophylactic drain insertion following renal transplant and the reasons for the choice to use (or not use) prophylactic drains were investigated. The online survey was tested by a group of selected transplant surgeons, and their feedback was incorporated into the final version of the survey.
Consultant surgeons who regularly perform kidney transplant at transplant centers across ANZ were identified through the Transplantation Society of Australia and New Zealand. An electronic link to the survey was distributed via electronic mail, individually or via their respective departments. Respondents who electronically consented to participating in this study completed the survey and identified themselves as one or more of the following: general surgeon (upper gastrointestinal surgery, transplant surgery, others), vascular surgeon, urologist, or other.
Between November 2 and December 2, 2017, we identified 66 surgeons who perform renal transplants. An electronic link to the REDCap survey was distributed via electronic mail to each surgeon or through administrators of their department. Reminder emails were sent weekly until a survey response was obtained. Five unsuccessful attempts at reminder emails led to a “no response” from a surgeon. Respondents who completed the questionnaire were deidentified according to the name and transplant unit. No financial incentives were provided.
Data were exported from the REDCap program, and statistical analyses were conducted with the SPSS Statistics package (IBM SPSS Statistics version 24). The Somers D test was used to test the association between an ordinal dependent variable and an ordinal independent variable (surgeon experience and drain insertion practice). A one-way ANOVA was conducted to determine whether general surgeons were more likely to insert a drain than non-general surgeons. Respondents were classified as surgeons trained in general surgery (n = 21) or surgeons trained in non-general surgery (vascular surgeons or urologists, n = 22), and we surveyed the likelihood of each group to insert a drain: > 90% of the time was expressed as a score of 1; > 10% of the time expressed as a score of 2; < 10% of the time expressed as a score of 3. Statistical significance was defined as P < .05.
Respondents and demographics
The response rate to the survey was 66% (43/66). The responses to questions about the length of consultant practice, transplant volume, and specialty training background are summarized in Figure 1.
Drain insertion practice
The responses to questions about the estimated frequency of drain insertion and the number of drains placed are summarized in Figure 2. The vast majority of surgeons (40/43; 93%) reported placing drains (when drains were used) only in the perigraft extraperitoneal space, and 1 surgeon specifically commented that an additional drain was placed in the subcutaneous space in “obese” patients.
Respondents were asked to identify any and all factors that may lead to a decision to place a prophylactic drain from 4 options; also, respondents were given the opportunity to provide specific personal reasons for the decision as free text, if desired. Multiple responses were permitted. The most common responses are summarized in Figure 3A. Concerns about bleeding and anticoagulation (27/43; 63%) and routine practice (25/43; 58%) were the dominant reasons cited by respondents. Five respondents provided comments indicating that the choice to insert a prophylactic drain could arise in circumstances of “difficult bladder,” “large amount of lymphatic tissue,” “technically difficult operation,” “unit policy,” and “I never put drains unless after take back for bleeding.” Respondents were also asked to select what they perceived as the benefits of prophylactic drain insertion from 5 options (multiple responses permitted), as well as provide additional comments as free text, if desired. These responses are shown in Figure 3B. The 2 most common responses were that drain insertion reduces symptomatic perigraft fluid collections (21/43; 49%) and that the respondents were uncertain about the benefit of drain insertion (13/43; 30%). Two respondents commented in free-text format that “bleeding is better predicted with drain inserted” and “I use drain to detect lymphocele postoperatively.”
Respondents were asked about the management of the drain once inserted. Responses regarding factors such as the importance of time, volume permitting safe removal, and the preferred number of posttransplant days for removal are shown in Figure 4. The factors selected as most significant in determining drain removal were both volume and time (19/44; 44%) and only volume (14/43; 33%). A volume of “< 50 mL/day” (22/43; 51%) was the most common response. Some respondents specified further (eg, “depends on the size of the patient” and “depends on fluid composition”), and 1 commented that it was decided to “remove the drain at day 5 even if volume is > 50 mL/day.” When asked whether they would discharge patients with high-output drains in situ, 16/43 (37%) responded never, 25/43 (58%) responded infrequently, and 2/43 (5%) responded frequently. Twenty-five respondents (58%) believed that fluid composition was important for the decision to remove a drain, whereas 18 respondents (42%) did not think it was relevant. Thirty-five percent (15/43) of respondents indicated that they routinely test drain fluid for creatinine before removing the drain. Twenty-one percent (9/43) responded that if the fluid consisted of hemoserous fluid, then this was an indicator for safe drain removal. The number of days posttransplant was a factor for drain removal for 27/43 respondents (63%). The postoperative period of days 3 to 5 was the most commonly selected time point for drain removal (27/43; 63%). One respondent specified that drain removal was appropriate at “24 hours post-IDC removal.” One respondent commented on the choice to “shorten tract if ongoing volume > 100 mL/day, aiming to get tissue sealing before removing to prevent expanding lymphocele.”
Association of experience and training background and drain insertion practices
There was no association between level of experience of surgeons and drain insertion practice (Somers D; d = .051, P = .673). General surgeons were more likely to insert drains than non-general surgeons (F = 6.121, P = .019; general surgeons, mean = 1.95 ± 0.865; non-general surgeons, mean = 1.27 ± 0.631). There was no association between the practice of drain insertion and the number of kidney transplants per year (F = 1.44, P = .251).
Willingness to participate in further study
Thirty-two respondents (74%) would consider enrolling their patients in a randomized controlled trial to determine the benefits and harms of drain insertion.
This study reports the results of a survey to assess the attitudes and preferences of surgeons in ANZ toward the use of prophylactic drains after kidney transplant. We have demonstrated that there is substantial variation in practice and rationale for the use of drains, duration of drainage, and indications for removal of the drains. More than half of respondents use drains routinely (> 90% of the time). The majority of routine drain placement was a consequence of “unit policy or routine practice,” suggesting that it is a consequence of established tradition. It must be acknowledged that drain insertion has been a traditional, standard procedure in most major surgical operations. It is not surprising, given the lack of any strong evidence to the contrary, that the practice persists. Even in the face of substantial evidence, traditional practices are often defended and maintained.25 This is likely a consequence of the conservatism among members of a highly regulated profession for which there are barriers in the process of changing/ceasing previous behaviors.25,37-39 A substantial minority of respondents seldom or never insert drains, and the basis of this perspective may be evidence gleaned from other areas of surgery for which the benefits of routine prophylactic drainage seem questionable.
This survey could not resolve the clinical question of prophylactic drain usage because the evidence is insubstantial. We have recently conducted a systematic review, and the results suggest that drains may be associated with a lower rate of perigraft fluid collection. Notwithstanding the significant bias in the design of the studies analyzed, even if correct, we do not know whether this would translate into any significant clinical outcome (such as graft loss, wound infection, requirement for interventions) or whether the complications of drains outweigh the benefits of reducing collections that do not have any clinical impact. Resolution of the issue would require a multicenter, randomized controlled trial, adequately powered to detect relatively rare but clinically significant outcomes. The large sample size required for such a study may be prohibitive. Therefore, surgeons will likely remain in an environment of clinical uncertainty with regard to the appropriate use of drains.
This survey had a response rate of 65% of practicing kidney transplant surgeons in ANZ. There is a large variation in the number of transplants performed by each survey participant. As a result, it is difficult to determine the proportion of the approximately 1200 kidney transplants performed annually in ANZ40 that contributed to the career experience of the respondents. Therefore, it is difficult to comment on what is actually a routine practice for all kidney transplants across Australia and New Zealand. A limitation of this study is that it represents a snapshot in time, whereas actual patterns of practice are in constant flux. It will be interesting to repeat the survey in the future to ascertain the trend in drain insertion.
Thirty percent of surgeons were uncertain with regard to the level of prophylaxis that could be provided by drain insertion for surgical complications. On the other hand, 48% of surgeons believed that the surgical drain was helpful in reducing the compressive effect of a perigraft fluid collection, and 20% thought it would reduce the rate of graft loss due to the mass effect of a collection. All of these views may be correct. Clearly, practices are driven by the manner in which the surgeons view the competing risks and potential benefits of drain insertion and the relative importance assigned to those outcomes. This survey highlights the clinically relevant application of surgical experience and heuristics in managing these complex cases. Although approximately 75% of respondents indicated willingness to participate in a randomized study, a significant minority of respondents are not willing. This likely reflects the lack of objective evidence against or for the insertion of a drain among practitioners. However, both groups are likely to believe that their side of the argument would be correct. We need to accept as a surgical community that the outcome the practitioner considers to be the most significant will determine which approach the practitioner may take and that there is no definite right or wrong answer at this stage.
Volume : 18
Issue : 7
Pages : 771 - 777
DOI : 10.6002/ect.2020.0071
From the 1Royal Prince Alfred Hospital Institute of Academic Surgery, University
of Sydney, Camperdown, New South Wales, Australia; the 2Department of Surgery,
Westmead Hospital, University of Sydney, Westmead, New South Wales, Australia;
and the 3Department of Renal Medicine, Royal Prince Alfred Hospital, University
of Sydney, Camperdown, New South Wales, Australia
Acknowledgements: This project was completely supported with timely donations in kind from the contributing authors. The authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no potential declarations of interest. Author contributions are as follows: Mugino, Lam, Hameed, Sandroussi, Chadban, Pleass, and Laurence performed the research; Mugino, Lee, Lam, Hameed, Pleass, and Laurence analyzed the data; all authors drafted the manuscript.
Corresponding author: Jerome Laurence, Royal Prince Alfred Hospital, Institute of Academic Surgery, University of Sydney, NSW 2006, Australia
Figure 1. Responses to Questions About the Length of Consultant Practice, Transplant Volume, and Specialty Training Background
Figure 2. Responses to Questions About the Estimated Frequency of Drain Insertion and the Number of Drains Placed
Figure 3. Responses to Questions About Factors Contributing to Prophylactic Drain Insertion
Figure 4. Responses Regarding Factors Affecting Drain Removal