Objectives: The influence of peritoneal dialysis on outcomes after simultaneous pancreas and kidney transplant is still vague. In addition, whether peritoneal dialysis leads to a higher risk of infectious complications and higher mortality rates in these transplant patients has not been unambiguously confirmed. In this study, our aim was to verify whether dialysis type determined outcomes on the pancreas graft and whether dialysis type was a risk factor for graftectomy or recipient death.
Materials and Methods: Our study group included 44 simultaneous pancreas and kidney transplant patients. Analyzed parameters included type and duration of dialysis treatment, age, sex, long-term pancreas graft survival and patient survival, overall mortality, and number of graftectomies.
Results: Of 44 patients, 3 (7%) required a graftectomy. Mortality rate of the group was 5%. Of 44 patients, 33 had hemodialysis and 11 had peritoneal dialysis. In those who had hemodialysis, the mean duration of renal replacement therapy was 30.5 months, which was significantly longer than duration for those who had peritoneal dialysis (20.4 mo; P < .01). There were 3 graftectomies and 1 death in the hemodialysis group. In the peritoneal dialysis group, there were no graftectomies and 1 death, with no significant differences in the number of graftectomies and mortality rates between the groups. Long-term survival also did not differ between the groups.
Conclusions: We found that type of dialysis did not affect outcomes in our group of simultaneous pancreas and kidney transplant patients. Before transplant, each patient requires an individual approach to treatment. The type of dialysis performed should not be viewed as a contradiction for transplant.
Key words : End-stage renal disease, Graft survival, Mortality, Renal transplant
Simultaneous pancreas and kidney (SPK) transplant is a therapeutic option for patients with diabetes and end-stage renal disease (ESRD).1 Increasing experience in the field, including the advancement of surgical techniques and immunosuppression, has resulted in SPK transplant becoming the standard method for this select group of patients.1,2 Simultaneous pancreas and kidney transplant can effectively ensure normoglycemia for several years posttransplant. After SPK transplant, patients have better survival rates than patients who undergo kidney transplant alone.2,3 Pancreas transplant has favorable effects on nephropathy, neuropathy, and cardiovascular complications.2,4 Despite these effects, the number of pancreas transplants globally has recently decreased.5
Duration of pretransplant dialysis may influence outcomes of SPK transplant.6 Data have suggested that patients without dialysis benefit from SPK more and have fewer perioperative complications.7 In addition, peritoneal dialysis (PD) has been shown to be a risk factor for infectious complications, due to peritonitis, which is the most common complication of PD. In SPK transplant patients, enteric drainage and long history of diabetes remain likely risk factors for intra-abdominal infections.8 Peritoneal dialysis is often administered to patients who had vascular access problems, often due to preexisting thrombosis,9 which may cause severe SPK transplant complications.
In this study, our aim was to verify whether the duration and type of dialysis affected outcomes of SPK transplant and whether these factors influenced long-term results.
Materials and Methods
We performed a retrospective analysis of the clinical data of 200 patients who underwent pancreas transplant at our center over the past 14 years. From this group, we included 44 SPK transplant patients who fulfilled our study criteria. The analyzed parameters included type and duration of dialysis treatment, age, sex, long-term pancreas graft survival and patient survival, overall mortality, and number of graftectomies. We also analyzed laboratory results, which included serum levels of amylase, lipase, C-peptide, C-reactive protein, and procalcitonin. Imaging examination included ultrasonography and computed tomography (CT) with administration of an intravenous contrast medium.
Poland has a centralized recipient allocation system, with the qualification process for transplant occurring in transplant centers. For SPK transplant, our center includes patients with complicated type 1 diabetes and ESRD. We also have experience with type 3 diabetes (postinflammatory or after total pancreatectomy). These patients are more commonly candidates for pancreas transplant alone. Potential donors are assessed with the use of the Pancreas Suitability Score before organ retrieval.
Arterial blood supply to the pancreatic graft is reconstructed into a common trunk on a back table, with a Y-graft from the iliac artery of the donor anastomosed to the superior mesenteric and splenic arteries. If the portal vein of the graft is sufficiently long, we prefer to omit the reconstruction. We perform enteric drainage of the pancreatic graft, with duodeno-duodenal anastomosis or duodeno-intestinal anastomosis. We transplant the kidney from the same median incision. We perform end-to-side anastomoses between the graft vessels and the recipient’s left external iliac vein and artery. We do not routinely implement double J stents.
From 2017 onward, we have routinely examined cultures from peritoneal fluid and from the peritoneal catheter in every PD patient. Previously, microbiologic samples had been taken only in cases suspected of infectious complications from wound, blood, and urine. The primary indications were high fever and elevated C-reactive protein or procalcitonin levels.
Ultrasonographic and CT images were reviewed for the following findings: abnormal graft structure, peripancreatic fluid collections, peripancreatic fat infiltration, abdominal fluid collection, enlarged mesenteric lymph nodes, total venous thrombosis, partial venous thrombosis, total arterial thrombosis, partial arterial thrombosis, and formation of aneurysms or stenosis at the site of arterial anastomosis. Peripancreatic fluid collections, peripancreatic fat infiltration, enlarged mesenteric lymph nodes, and free fluid in the abdominal cavity were considered indicative for presence of intra-abdominal inflammation.
This study was a retrospective analysis of clinical data. There was no risk to patients during the study, and only nonidentifiable data were used for publication.
For statistical analysis, we used Statistica 12 software (StatSoft, Krakow, Poland). A single- and multifactor statistical analysis was performed. We used nonparametric tests for 2 independent value Mann-Whitney U tests and Pearson correlations. Statistical significance was estimated at P < .05.
Table 1 presents characteristics of our study patients who underwent SPK transplant. Of total study group patients, 3 (7%) required graftectomies. The mortality rate of the group was 5%. Of 44 patients, 33 received hemodialysis and 11 received PD. In the subgroup with hemodialysis, the mean duration of renal replacement therapy was 30.5 months, which was significantly longer than duration in the PD group (20.4 mo; P < .01). There were 3 graftectomies and 1 death in this group. In the PD group, no patients received a graftectomy and there was 1 death. Indications for graftectomy are shown in Table 1. In the hemodialysis group, 7% of the patients were insulin dependent, whereas in the PD group, all grafts had sufficient function through our observation period.
The mean long-term survival of the pancreas graft was 4.5 years (P = .3). Mean patient survival was 5 years (P = .1), with no significant differences shown between the groups.
Serum amylase levels were significantly higher in the PD subgroup (P < .05). We observed no significant differences between groups for the remaining analyzed parameters.
During surgery, patients in the PD group had microbiologic analyses. In 2 patients, although no pathogens were found in the peritoneal fluid or catheter, they subsequently developed wound infections with multidrug-resistant pathogens. Two other patients were found to have microbiota that later caused wound infections (Table 2). None of these patients lost their pancreas graft, and all grafts maintained good function. One patient, number 4 (Table 2), died of a duodenal fistula and sepsis after 74 days.
We did not observe differences between groups in the number of peripancreatic fluid collections shown with ultrasonography and CT scans. In the PD group, however, arterial thrombosis was present in ultrasonography scans more frequently, but this finding did not correlate with frequency of graftectomy or deaths.
Pretransplant dialysis has a negative impact on both patient survival and graft function in patients who have kidney transplant alone. In patients who undergo SPK transplant, these data are unequivocal. Wiseman and associates observed better recipient survival after 7 years in those who underwent preemptive dialysis before SPK transplant.10 With regard to preemptive SPK transplant, data are limited. Most facilities tend to have more experience with preemptive kidney transplant. Although higher estimated glomerular filtration rate pretransplant is not associated with graft function or graft loss, early graft function is slightly more favorable with preemptive kidney transplant, although this favorable kidney graft function diminishes after 1 year.11 Preemptive SPK transplant can involve a higher risk of kidney allograft failure, but it can also be associated with lower risk of pancreas allograft failure compared with pancreas transplant after kidney transplant.12 On the other hand, preemptive SPK is reported to be affected by fewer perioperative complications and to result in lower mortality.7
In our study, we found that the type of dialysis does not affect the long-term outcomes of SPK transplant. Since 2017, we have performed microbiologic tests of peritoneal fluid and the peritoneal catheter in every patient who receives PD because we observed that peripancreatic fluid collections and wound infections occurred more frequently in those who receive PD. The aim was to have an antibiogram ready when the first infectious symptoms present. Our experience shows that, when peritoneal fluid or the catheter is contaminated, these pathogens are the first to cause wound infections and the potentially serious septic complications. However, aseptic fluid does not prevent later wound infections, and these patients are more likely to be colonized with multidrug-resistant pathogens.
Padillo-Ruiz and associates13 proposed a classification for intra-abdominal posttransplant-complicated fluid collections. Fluid collections classified as grades 1 and 2 respond to conservative treatment, antibiotics, or percutaneous drainage. For grade 3, surgical intervention is required. Graftectomy represents grade 4, and patient death is classified as grade 5.13 In their retrospective observations, the authors found no differences in the rate of intra-abdominal fluid collections, outcomes, and complication rates.13 In their study, Kim and associates did not observe increased intra-abdominal infections in patients who received PD.8
In our patient group, there were no differences in the frequency of diagnosis of fluid collections with both ultrasonography and CT. Therefore, we suggest that ultrasonography is an effective method for diagnosis. However, CT and magnetic resonance imaging examinations are often considered superior to ultrasonography in terms of diagnosing fluid collections, as these methods are not compromised by bowel gases or affected by the ultrasonographer’s skill level.14 The differential diagnosis of anechogenic or heterogenic peripancreatic fluid collections based solely on imaging is broad and includes seromas, pseudocysts, hematomas, and abscesses or can be indicative of fistula. Percutaneous image-guided aspiration is sometimes necessary to precisely assess the fluid collection content, and ultrasonographic-guided biopsy can be more feasible than CT or magnetic resonance imaging-guided biopsies in most instances.15 Fluid collections are believed to increase mortality and likelihood of graft loss and can also be a manifestation of exocrine leakage.15 The nature of the secretion from the wound or the drain is commonly indicative of the nature of the fistula. Elevated infectious parameters (C-reactive protein and procalcitonin), significant fluid peripancreatic collections, severe conditions of the patient, and the nature of the secretion are primary indications for surgery.
In our study, duration of pancreas graft function and patient survival did not vary between the hemodialysis and PD groups. Some studies suggest that PD is a risk factor for vascular thrombosis, relaparotomy, pancreas graft loss, and higher mortality.9 We found that, although arterial thrombosis was more commonly visualized with ultrasonography in the PD group, it did not influence graft loss. Type and duration of dialysis did not influence radiologic findings and long-term results.
The type of exocrine drainage performed differs between centers. All of our study patients had enteric drainage. Jimenez and associates16 previously compared outcomes of SPK transplant in hemodialysis patients with enteric drainage and PD patients with bladder drainage. Although the rate of infectious complications was similar, duodenal leakages were more common in the PD group. The long-term results were also similar.16 Enteric drainage improves quality of life and is less commonly affected by duodenal fistulas.3
In a large study of 1194 transplant patients, the investigators found that type of dialysis did not affect the 1-year patient survival or loss of pancreas and kidney grafts.17 Frequency of technical failures also was not related to type of dialysis administered.17
Hemodialysis is also a risk factor for vascular and cardiovascular complications. Patients who receive hemodialysis are more likely to develop myocardiac and cerebrovascular incidents and to require small amputations.18 In a study that included long-term observations, cardiovascular events and infections were the primary causes of death, although the number of cardiac deaths was lower in patients who received preemptive transplant.18 Kidney and pancreas allograft survival also did not differ between those who received or did not receive preemptive transplant. The number of small and large amputations was also slightly lower in patients who had preemptive hemodialysis.18 These results suggest that dialysis in pancreatic transplant patients led to development of micro- and macroangiopathies. One of the benefits of normoglycemia is regression of coronary atherosclerosis.19
Some data have suggested that a dialysis treatment period of over 9 months is a potential risk factor for graftectomy.6 In our study, in which the mean dialysis period was 23.5 months in the hemodialysis group and 19 months in the PD group, duration of dialysis treatment did not affect transplant outcomes or patient survival.
Our study showed that type of dialysis does not affect the outcome of SPK transplant. We also showed that patients who receive hemodialysis do not develop vascular complications more commonly than PD patients or that patients who receive PD are more likely to present with abdominal inflammations due to chronic bacterial contamination.
Every patient after SPK transplant requires an individual approach to treatment. The type of dialysis administered should not be viewed as a contradiction for transplant. Indeed, patients who undergo SPK transplant have fewer complications and better long-term results. We suggest that patients with diabetes and ESRD who are candidates for SPK transplant should be added to wait lists before they begin receiving dialysis treatment.
DOI : 10.6002/ect.2019.0204
From the 1Department of Gastroenterological Surgery and Transplantation of
Medical Centre of Postgraduate Education at the Central Clinical Hospital of the
Ministry of the Interior in Warsaw, Warsaw, Poland; the 2Mossakowski Medical
Research Centre of the Polish Academy of Sciences, Department of Surgical
Research and Transplantology, Warsaw, Poland; and the 3Medical Centre of
Postgraduate Education, Warsaw, Poland
Acknowledgements: The authors have no conflict of interests to declare. The study did not receive any external funding. All authors read and approved the final manuscript. The manuscript and all of its contents have not been published and have not been submitted elsewhere.
Corresponding author: Agnieszka Surowiecka, CSK MSWiA Woloska 137 street, 02-507 Warsaw, Poland
Phone: +48 22 508 12 17
Table 1. Characteristics of the Simultaneous Pancreas and Kidney Transplant Patients (N = 44)
Table 2. Result of Microbiologic Tests of Peritoneal Dialysis Group