- Email: [email protected]
Multivariate Analysis of Complications After Simultaneous Pancreas and Kidney Transplantation
Multivariate Analysis of Complications After Simultaneous Pancreas and Kidney Transplantation T. Grochowieckia,*, Z. Gałązkaa, K. Madeja, S. Frunzea, S. Nazarewskia, T. Jakimowicza, L. Paczekb, M. Durlikc, and J. Szmidta a Department of General, Vascular and Transplant Surgery, Medical University of Warsaw, Poland; bDepartment of Immunology, Transplantology and Internal Diseases, Medical University of Warsaw, Poland; and cDepartment of Transplantation Medicine and Nephrology, Medical University of Warsaw, Poland
ABSTRACT Objective. Identification of factors that have an impact on postoperative complications after simultaneous pancreas and kidney transplantation (SPKTx) could help overcome limitations of this kind of treatment. Methods. Postoperative complications among 112 SPKTx recipients were divided into 3 groups: related to transplanted pancreas (n ¼ 66), related to transplanted kidney (n ¼ 23) and general surgical complications (n ¼ 31) 120 refers to complications among 112 recipients. According to the modified Clavien-Dindo scale, complications were classified according to their severity for each group. Risk factors for complication development related to donor, recipient, surgical technique, and immunosuppression were included to establish the multivariable model using logistic regression. Results. Multiple regression analysis showed the following independent factors influenced mortal complications due to transplanted pancreas: age of donor (OR, 1.07; P < .04), duration of vascular pancreas anastomosis above 35 minutes (OR, 3.94; P < .04) and duration of recipient dialysis above 24 months before transplantation (OR, 0.14; P < .01). Area under receiver operating characteristic curve for this model was 0.8. Conclusion. To improve results, the following modification of identified risk factors should be assumed: selection of donor in term of age, shortening of the second warm ischemia time, and adjustment of the waiting list to avoid prolongation of recipient dialysis before SPKTx.
A
CCORDING to UNOS/IPTR, a 20% decrease in pancreas transplantation was observed in 2010 compared with 2004 [1]. The declining trend also was noticed in 2011 [2]. One possible explanation of this tendency could be the cost of surgical complications after this type of transplantation. It was estimated that a patient with a surgical complication after pancreas transplantation would cost an additional $17,363 [3,4]. Pancreas transplantation was characterized as having the highest complication and reoperation rates compared with the other solid organ transplants [3,5]. The reoperation rate has been reported from 11.6% to 43%, in contrast with a 2% to 5% relaparotomy rate after general surgical operations [6e10]. When a recipient requires relaparotomy, the cost of pancreas transplantation almost doubled [5]. Within this context, the identification of risk factors that influence the development of surgical complications was crucial. This knowledge could help
to modify donor and recipient selection criteria to avoid the high-risk situations and change post-transplant management in these cases. The aim of this study was to identify factors that have influence on postoperative complications after simultaneous pancreas and kidney transplantation (SPKTx). MATERIALS AND METHODS Postoperative complications related to transplanted kidney among 112 SPKTx recipients who received their grafts from February 1988 to
*Address correspondence to Tadeusz Grochowiecki, Department of General, Vascular and Transplant Surgery, Medical University of Warsaw, 1a Banacha Street, 02-097, Warsaw, Poland. E-mail: [email protected]
0041-1345/14 http://dx.doi.org/10.1016/j.transproceed.2014.08.010
ª 2014 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
2806
Transplantation Proceedings, 46, 2806e2809 (2014)
COMPLICATIONS AFTER SIMULTANEOUS SPKTX July 2010 were analyzed. The indications for transplantation were endstage renal disease and diabetes type 1. Characteristics of donors and SPKTx recipients are shown in Table 1. For all recipients, the transplanted pancreas was placed at the right iliac fossa intraperitoneally with vascular anastomoses to the iliac vessels. Twelve patients (19%) had portoenteric drainage. Among 100 recipients with systemicenteric drainage, 24 had transplanted pancreatic segment using fouranastomoses technique [11]. Kidney grafts were placed at the left iliac fossa extraperitoneally. Vascular anastomoses were performed on the iliac vessels with subsequent anastomosis between the end ureter to urinary bladder. Twenty-five (22.3%) recipients received segmental transplant, and immunosuppression consisted of ATG, azathioprine, cyclosporine, and steroids. Twenty-five of 112 (22.3%) recipients were prescribed daclizimab, cyclosporine, and mycophenolate mofetil. ATG, mycophenolate mofetil, tacrolimus, and steroids were administered to 64 of 112 patients (57.1%). Severity of complication was classified according to the Clavien-Dindo scale, adjusted for SPKTx transplant [12,13]. The suffix “g” was introduced and it refers to graft removal (Table 2). Postoperative complications among 112 SPKTx recipients were divided into 3 groups: related to transplanted pancreas (66 complications), related to transplanted kidney (23 complications), and related to neither transplanted pancreas nor kidney (renamed “general surgical,” 31 complications). If the recipient developed more than 1 complication, the higher-grade complication was used. According to the modified Clavien-Dindo scale, each recipient with a complication was classified according to his/her severity and for each complication group (Table 3). There were 4 groups of risk factors related to donor (age, gender, cause of death [traumatic vs nontraumatic]), recipient (age, gender, duration of diabetes, type and duration of renal replacement therapy), perioperative (type and time of pancreatic and kidney graft anastomosis, type of enteric anastomosis, total ischemia time), and immunosuppression (tacrolimus vs cyclosporine; monoclonal vs polyclonal antibodies). Body index mass (BMI) of donor and recipients was not incorporated because BMI values above 25 kg/m2 were not considered for pancreas harvesting
Table 1. Characteristics of Organ Donors and SPKTx Recipients Gender of recipients (male/female) Age of recipient at SPKTx (y) Duration of diabetes before transplantation (y) Duration of renal replacement therapy before transplantation (mo) Renal replacement therapy before SPKTx Hemodialysis Peritoneal dialysis Preemptive kidney transplantation Mean vascular anastomosis time of transplanted pancreas (min) Mean vascular anastomosis time of transplanted kidney (min) Mean total ischemia time of transplanted pancreas (min) Mean total ischemia time of transplanted kidney (min) Cause of donor death Traumatic Nontraumatic Gender of donor (male/female) Age of donors (y) *Values given as median (minimumemaximum).
69/43 (61.6%/38.4%) 38 (24e59)* 24 (9e44)*
2807 Table 2. Modified Clavien-Dindo Scale of Complication Severity After Simultaneous Pancreas and Kidney Transplantation Severity of Complication
I II III IIIA IIIB IV IVA IVB
Definition
No complication Pharmacological treatment Invasive intervention (radiological, endoscopic, operative) invasive intervention general anesthesia not required invasive intervention general anesthesia required Life- or graft-threatening complication failure of the graft death
If the graft was removed the suffix “g” was added to the grade of complication severity.
and transplantation. Impacts of risk factors (independent variables) on severity of complication grades (dependent variables) were analyzed. In the case of missing data, variable discretization was performed. Separate categories were created for missing values A risk-adjusted model using multivariable logistic regression was built in backward selection. The area under the receiver operating characteristic curve for the model was calculated for internal validation according to recommended standards [14]. All statistical analysis was performed using STATISTICA 12 PL (Statsoft Polska, Kraków, Polska).
RESULTS
Mortality due to transplanted pancreas, kidney, and general surgery complications were 12,5%, 1,7%, and 1,7% respectively. It was possible to build the model only for mortal complications related to transplanted pancreas (grade IVB). The constructed model included: donor age (odds ratio [OR], 1.07; confidence interval [CI], 1.0e1.13; P < .04), duration of vascular pancreas anastomosis above 35 minutes (OR, 3.94; CI, 1.08e14.36; P < .04) and duration of recipient dialysis above 24 months prior to SPKTx (OR, 0.14; CI 0.03e0.62; P < .01). The following model was built: Logit P ¼ 4; 08 þ 0; 064 donor age þ 1:37 time of transplanted pancreas vascular anastomosis above 35 minutes 1:93
14 (0e102)*
duration of dialysis therapy below 69 18 25 39
(61.6%) (16.1%) (22.3%) (20e95)
24 months: The area under receiver operating characteristic curve for this model was 0.8 (Fig 1).
33 (15e72) 517 (183e860) 576 (165e1099)
53 41 67/27 (57.1%/42.9%) 28.5 (17e53)
DISCUSSION
SPKTx is the most commonly performed multiorgan transplantation, however it is characterized by a high rate of surgical complications. To describe problem of surgical complications variuos methodological approaches were applied [15e17]. In this study, the modified Claven-Dindo scale was used. In this scale the severity of complication was described by the most efficient treatment of particular complication. This scale has been widely known but rarely
2808
GROCHOWIECKI, GAŁĄZKA, MADEJ ET AL
Table 3. Prevalence of Complications According to the Severity Among SPKTx Recipients
Grade of the Complication Severity
No. of Recipients Who Developed Complications Due to Pancreatic Graft (%)
No. of Recipients Who Developed Complications Due to Kidney Graft (%)
I 49 (44.6%) 90 (80.3%) II 7 (6.25%) 7 (6.2%) IIIA 3 (2.6 %) 1 (0.9%) IIIB 10 (8.9 %) 6 (5.4%) IVA/IVA (g) 29 [1 (0.9%)/28 (25%)] 6 [0/6] (5.4%) IVB/IVB (g) 14 [3 (2.6%)/11 (9.8%)] 2 [1/1] (1.8%)
No. of Recipients Who Developed General Surgical Complications (%)
87 (77.8%) 4 (3.6%) 0 19 (16.9%) 0 2 (1.7%)
employed for SPKTx [18,19]. Using this scale, it was easily shown that graft thrombosis or pancreatitis did not always lead to graft loss [16,20,21]. Thus, we believed that this methodological approach provided broader insight regarding the underlying pathology than simply assessing pancreatic graft technical failure. Fellmer et al approached to idea of Clavien scale. However, he distinguished only surgical complications which required surgical or radiological interventions [16]. In this study, complications were divided into those related to (1) pancreatic graft, (2) renal graft, and (3) general surgical complications, which were due to neither pancreatic graft nor kidney graft. Such an approach is rational and used in many publications [16,22,23]. Consistent with other studies, it was found that the most numerous and serious complications were related to pancreatic graft [5,16,22,23]. In contrast to other studies, the lethal complications related to pancreatic graft (grade IVB) but not the technical failure was used as a analysed factor in the multivariate model. In this study, we was found that the age of donor, time of performing pancreatic graft anastomosis, and duration of dialysis treatment prior to SPKTx have significant impact on development of mortal complications relating to pancreatic graft. The area under the receiver operating characteristic curve was 0.8, however
AUC: 0.8
1,2
1,0
sensitivity
0,8
the prospective study to validate the model should be necesary. Modification of the indentified factors could decrease the risk of lethal complications related to pancreatic graft. In our model, all variables would be modifiable. Many authors identified donor age as a significant factor that influences pancreas-related complications [5,15,16]. The pancreas donor risk index and preprocurement pancreas allocation suitability score included donor age as a risk factor of pancreas graft failure [24,25]. Thus, selection of donors in terms of age is essential. The next significant factor was the time of pancreatic vessel graft anastomosis. Fellmer et al found that not cold but the second warm ischemic time (anastomosis time) was related to the prevalence of pancreas-specific complications [16]. Others showed that preservation time over 20 or 24 hours was the risk factor for technical failure [15,17,24]. In our study, the mean of the total ischemia time was below 9 hours; thus, it could be the reason that preservation time became not significant in analysis. The duration of recipient dialysis above 24 months prior to SPKTx was identified as the third risk factor. There were no fatal complications among simultaneous pancreas and preemptive kidney transplant recipients; therefore, variable preemptive vs nonpreemptive kidney transplantation could not be considered for methodological reasons. Nevertheless, duration of dialysis less than 24 months prior to SPKTx decreased the risk of lethal complications by 7-fold. Based on International Pancreas Transplant Registry data, Gruessner found that relative risk of death was significantly increased in SPK recipients who were on pretransplant dialysis (RR, 1.49; P < .017) [1]. Ojo showed that, compared with patients who underwent preemptive transplantation, dialysis duration of 1e6 months was associated with 15% higher mortality risk (RR, 1.15, P < .001) and 24 months of dialysis time was associated with 75% higher risk of death (RR, 1.75, P < .001) [26]. Interestingly, as far as only the technical failure or surgical complication development were concerned, the duration of the pretransplant dialysis had no importance [16,17]. Some authors did not even consider duration of dialysis as a risk factor in multivariate analysis [15,24]. From the practical point of view, the shorter duration time of dialysis, the fewer mortal complications that were developed by the recipient. Therefore SPKTx should be performed as fast as possible. This should have an influence on construction of the waiting list.
0,6
ACKNOWLEDGMENTS
0,4
The authors thank Mr Grzegorz Migut and Mr Janusz Wa˛troba, PhD, from StatSoft Polska for excellent help in statistical analysis.
0,2
0,0
-0,2 -0,2
REFERENCES 0,0
0,2
0,4
0,6
0,8
1,0
1,2
1-specificity
Fig 1. Area under the receiver operating characteristic curve (AUC) for described model.
[1] Gruessner AC. Update on pancreas transplantation: comprehensive trend analysis of 25,000 cases followed up over the course of twenty-four years at the International Pancreas Transplant Registry (IPTR). Rev Diabet Stud 2011;8:6e16.
COMPLICATIONS AFTER SIMULTANEOUS SPKTX [2] Kandaswamy R, Stocka PG, Skeansa MA, et al. OPTN/ SRTR 2011 Annual Data Report: pancreas. AJT 2012;12:47e70. [3] Cohn JA, Englesbe MJ. The financial implications of pancreas transplant complications. In: Preedy VR, Watson RR, editors. Handbook of disease burdens and quality of life measures. New York, N.Y: Springer; 2010. pp. 661e2. [4] Cohn JA, Englesbe MJ, Ads JL. Financial implication of pancreas transplant complications: a business case for quality improvement. Am J Transplant 2007;7:1656e60. [5] Troppmann C. Surgical complication requiring early relaparotomy after pancreas transplantation. Ann Surg 1998;227:255e68. [6] Boggi U, Vistoli F, Del Chiaro M, et al. Surgical complications in the modern era of pancreas transplantation. J Pancreas 2006;7:522. [7] Steuer W, Malaise J, Mark W, et al. Euro-SPK Study Group. Spectrum of surgical complications after simultaneous pancreaskidney transplantation in a prospectively randomized study of two immunosuppressive protocols. Nephrol Dial Transplant 2005;20:ii54. [8] Zer M, Dux S, Dintsman M, et al. The timing of relaparatomy and its influence on prognosis. A 10-year survey. Am J Surg 1980;139:338e43. [9] Kirk RM. Reoperative surgery for early complication following abdominal and abdominothoracic operations. J R Soc Med 1988;81:7e9. [10] Hinsdale JG, Jaffe BM. Reoperation for intraabdominal sepsis. Indications and results in modern critical-care setting. Ann Surg 1984;199:31e6. [11] Szmidt J, Lao M, Grochowiecki T, et al. Pancreas transplantation: four vascular anastomoses. Transplant Proc 1996;28: 3511e3. [12] Dindo D, Demartines N, Clavien PA, et al. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205e13. [13] Clavien PA, Barkun J, Oliviera ML, et al. The ClavienDindo classification of surgical complications. Ann Surg 2009;250: 187e96. [14] Kalil AC, Mattei J, Florescua DF, et al. Recommendations for the assessment and reporting of multivariable logistic regression in transplantation literature. Am J Transplant 2010;10:1695e703.
2809 [15] Humar A, Ramcharan T, Kandaswamy R, Gruessner RW, Gruessner AC, Sutherland DE. Technical failures after pancreas transplants: why grafts fail and the risk factorsda multivariate analysis. Transplantation 2004;78:1188e92. [16] Fellmer PT, Pascher A, Kahl A, et al. Influence of donorand recipient-specific factors on the postoperative course after combined pancreasekidney transplantation. Langenbecks Arch Surg 2010;395:19e25. [17] Finger EB, Radosevich DM, Dunn TB, et al. A composite risk model for predicting technical failure in pancreas transplantation. Am J Transplant 2013;13:1840e9. [18] Clavien PA, Strasberg SM. Severity grading of surgical complications. Ann Surg 2009;250:197e8. [19] Castoldi R, Ferrari G, Staudacher C, et al. Segmental ductinjected versus whole-bladder drained pancreas transplantation: the San Raffaele Hospital experience. Transplant Proc 1997;26:450. [20] Kim YH, Park JB, Lee SS, et al. How to avoid graft thrombosis requiring graftectomy immediate post-transplant CT angiography in pancreas transplantation. Transplantation 2012;94: 925e30. [21] Grabowska-Derlatka L, Pacho R, Grochowiecki T, Jakimowicz T, Rowi nski O. Imaging of pancreatic transplant vessels and anastomoses with 16-row multidetector computed tomography. Transplant Proc 2006;38:266e8. [22] Sutherland DER, et al. Lessons learned from more than 1000 pancreas transplants at a single institution. Ann Surg 2001;233:463e501. [23] Sollinger HW, Odorico JS, Becker YT, D’Alessandro AM, Pirsch JD. One thousand simultaneous pancreas-kidney transplants at a single center with 22-year follow-up. Ann Surg 2009;250:618. [24] Axelorod DA, Sung RS, Meyer KH. Systemic evaluation of pancreas allograft quality, outcomes and geographic variation in utilization. Am J Transplant 2010;10:837e45. [25] Vinkers MT, Rahmel AO, Slot MC, et al. How to recognize suitable pancreas donor: a Eurotransplant study of preprocurement factors. Transplant Proc 2008;40:1275. [26] Ojo AO, Meier-Kriesche HU, Hanson JA, et al. The impact of simultaneous pancreas-kidney transplantation on long-term patient survival. Transplantation 2001;71:82e90.
ABSTRACT Objective. Identification of factors that have an impact on postoperative complications after simultaneous pancreas and kidney transplantation (SPKTx) could help overcome limitations of this kind of treatment. Methods. Postoperative complications among 112 SPKTx recipients were divided into 3 groups: related to transplanted pancreas (n ¼ 66), related to transplanted kidney (n ¼ 23) and general surgical complications (n ¼ 31) 120 refers to complications among 112 recipients. According to the modified Clavien-Dindo scale, complications were classified according to their severity for each group. Risk factors for complication development related to donor, recipient, surgical technique, and immunosuppression were included to establish the multivariable model using logistic regression. Results. Multiple regression analysis showed the following independent factors influenced mortal complications due to transplanted pancreas: age of donor (OR, 1.07; P < .04), duration of vascular pancreas anastomosis above 35 minutes (OR, 3.94; P < .04) and duration of recipient dialysis above 24 months before transplantation (OR, 0.14; P < .01). Area under receiver operating characteristic curve for this model was 0.8. Conclusion. To improve results, the following modification of identified risk factors should be assumed: selection of donor in term of age, shortening of the second warm ischemia time, and adjustment of the waiting list to avoid prolongation of recipient dialysis before SPKTx.
A
CCORDING to UNOS/IPTR, a 20% decrease in pancreas transplantation was observed in 2010 compared with 2004 [1]. The declining trend also was noticed in 2011 [2]. One possible explanation of this tendency could be the cost of surgical complications after this type of transplantation. It was estimated that a patient with a surgical complication after pancreas transplantation would cost an additional $17,363 [3,4]. Pancreas transplantation was characterized as having the highest complication and reoperation rates compared with the other solid organ transplants [3,5]. The reoperation rate has been reported from 11.6% to 43%, in contrast with a 2% to 5% relaparotomy rate after general surgical operations [6e10]. When a recipient requires relaparotomy, the cost of pancreas transplantation almost doubled [5]. Within this context, the identification of risk factors that influence the development of surgical complications was crucial. This knowledge could help
to modify donor and recipient selection criteria to avoid the high-risk situations and change post-transplant management in these cases. The aim of this study was to identify factors that have influence on postoperative complications after simultaneous pancreas and kidney transplantation (SPKTx). MATERIALS AND METHODS Postoperative complications related to transplanted kidney among 112 SPKTx recipients who received their grafts from February 1988 to
*Address correspondence to Tadeusz Grochowiecki, Department of General, Vascular and Transplant Surgery, Medical University of Warsaw, 1a Banacha Street, 02-097, Warsaw, Poland. E-mail: [email protected]
0041-1345/14 http://dx.doi.org/10.1016/j.transproceed.2014.08.010
ª 2014 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
2806
Transplantation Proceedings, 46, 2806e2809 (2014)
COMPLICATIONS AFTER SIMULTANEOUS SPKTX July 2010 were analyzed. The indications for transplantation were endstage renal disease and diabetes type 1. Characteristics of donors and SPKTx recipients are shown in Table 1. For all recipients, the transplanted pancreas was placed at the right iliac fossa intraperitoneally with vascular anastomoses to the iliac vessels. Twelve patients (19%) had portoenteric drainage. Among 100 recipients with systemicenteric drainage, 24 had transplanted pancreatic segment using fouranastomoses technique [11]. Kidney grafts were placed at the left iliac fossa extraperitoneally. Vascular anastomoses were performed on the iliac vessels with subsequent anastomosis between the end ureter to urinary bladder. Twenty-five (22.3%) recipients received segmental transplant, and immunosuppression consisted of ATG, azathioprine, cyclosporine, and steroids. Twenty-five of 112 (22.3%) recipients were prescribed daclizimab, cyclosporine, and mycophenolate mofetil. ATG, mycophenolate mofetil, tacrolimus, and steroids were administered to 64 of 112 patients (57.1%). Severity of complication was classified according to the Clavien-Dindo scale, adjusted for SPKTx transplant [12,13]. The suffix “g” was introduced and it refers to graft removal (Table 2). Postoperative complications among 112 SPKTx recipients were divided into 3 groups: related to transplanted pancreas (66 complications), related to transplanted kidney (23 complications), and related to neither transplanted pancreas nor kidney (renamed “general surgical,” 31 complications). If the recipient developed more than 1 complication, the higher-grade complication was used. According to the modified Clavien-Dindo scale, each recipient with a complication was classified according to his/her severity and for each complication group (Table 3). There were 4 groups of risk factors related to donor (age, gender, cause of death [traumatic vs nontraumatic]), recipient (age, gender, duration of diabetes, type and duration of renal replacement therapy), perioperative (type and time of pancreatic and kidney graft anastomosis, type of enteric anastomosis, total ischemia time), and immunosuppression (tacrolimus vs cyclosporine; monoclonal vs polyclonal antibodies). Body index mass (BMI) of donor and recipients was not incorporated because BMI values above 25 kg/m2 were not considered for pancreas harvesting
Table 1. Characteristics of Organ Donors and SPKTx Recipients Gender of recipients (male/female) Age of recipient at SPKTx (y) Duration of diabetes before transplantation (y) Duration of renal replacement therapy before transplantation (mo) Renal replacement therapy before SPKTx Hemodialysis Peritoneal dialysis Preemptive kidney transplantation Mean vascular anastomosis time of transplanted pancreas (min) Mean vascular anastomosis time of transplanted kidney (min) Mean total ischemia time of transplanted pancreas (min) Mean total ischemia time of transplanted kidney (min) Cause of donor death Traumatic Nontraumatic Gender of donor (male/female) Age of donors (y) *Values given as median (minimumemaximum).
69/43 (61.6%/38.4%) 38 (24e59)* 24 (9e44)*
2807 Table 2. Modified Clavien-Dindo Scale of Complication Severity After Simultaneous Pancreas and Kidney Transplantation Severity of Complication
I II III IIIA IIIB IV IVA IVB
Definition
No complication Pharmacological treatment Invasive intervention (radiological, endoscopic, operative) invasive intervention general anesthesia not required invasive intervention general anesthesia required Life- or graft-threatening complication failure of the graft death
If the graft was removed the suffix “g” was added to the grade of complication severity.
and transplantation. Impacts of risk factors (independent variables) on severity of complication grades (dependent variables) were analyzed. In the case of missing data, variable discretization was performed. Separate categories were created for missing values A risk-adjusted model using multivariable logistic regression was built in backward selection. The area under the receiver operating characteristic curve for the model was calculated for internal validation according to recommended standards [14]. All statistical analysis was performed using STATISTICA 12 PL (Statsoft Polska, Kraków, Polska).
RESULTS
Mortality due to transplanted pancreas, kidney, and general surgery complications were 12,5%, 1,7%, and 1,7% respectively. It was possible to build the model only for mortal complications related to transplanted pancreas (grade IVB). The constructed model included: donor age (odds ratio [OR], 1.07; confidence interval [CI], 1.0e1.13; P < .04), duration of vascular pancreas anastomosis above 35 minutes (OR, 3.94; CI, 1.08e14.36; P < .04) and duration of recipient dialysis above 24 months prior to SPKTx (OR, 0.14; CI 0.03e0.62; P < .01). The following model was built: Logit P ¼ 4; 08 þ 0; 064 donor age þ 1:37 time of transplanted pancreas vascular anastomosis above 35 minutes 1:93
14 (0e102)*
duration of dialysis therapy below 69 18 25 39
(61.6%) (16.1%) (22.3%) (20e95)
24 months: The area under receiver operating characteristic curve for this model was 0.8 (Fig 1).
33 (15e72) 517 (183e860) 576 (165e1099)
53 41 67/27 (57.1%/42.9%) 28.5 (17e53)
DISCUSSION
SPKTx is the most commonly performed multiorgan transplantation, however it is characterized by a high rate of surgical complications. To describe problem of surgical complications variuos methodological approaches were applied [15e17]. In this study, the modified Claven-Dindo scale was used. In this scale the severity of complication was described by the most efficient treatment of particular complication. This scale has been widely known but rarely
2808
GROCHOWIECKI, GAŁĄZKA, MADEJ ET AL
Table 3. Prevalence of Complications According to the Severity Among SPKTx Recipients
Grade of the Complication Severity
No. of Recipients Who Developed Complications Due to Pancreatic Graft (%)
No. of Recipients Who Developed Complications Due to Kidney Graft (%)
I 49 (44.6%) 90 (80.3%) II 7 (6.25%) 7 (6.2%) IIIA 3 (2.6 %) 1 (0.9%) IIIB 10 (8.9 %) 6 (5.4%) IVA/IVA (g) 29 [1 (0.9%)/28 (25%)] 6 [0/6] (5.4%) IVB/IVB (g) 14 [3 (2.6%)/11 (9.8%)] 2 [1/1] (1.8%)
No. of Recipients Who Developed General Surgical Complications (%)
87 (77.8%) 4 (3.6%) 0 19 (16.9%) 0 2 (1.7%)
employed for SPKTx [18,19]. Using this scale, it was easily shown that graft thrombosis or pancreatitis did not always lead to graft loss [16,20,21]. Thus, we believed that this methodological approach provided broader insight regarding the underlying pathology than simply assessing pancreatic graft technical failure. Fellmer et al approached to idea of Clavien scale. However, he distinguished only surgical complications which required surgical or radiological interventions [16]. In this study, complications were divided into those related to (1) pancreatic graft, (2) renal graft, and (3) general surgical complications, which were due to neither pancreatic graft nor kidney graft. Such an approach is rational and used in many publications [16,22,23]. Consistent with other studies, it was found that the most numerous and serious complications were related to pancreatic graft [5,16,22,23]. In contrast to other studies, the lethal complications related to pancreatic graft (grade IVB) but not the technical failure was used as a analysed factor in the multivariate model. In this study, we was found that the age of donor, time of performing pancreatic graft anastomosis, and duration of dialysis treatment prior to SPKTx have significant impact on development of mortal complications relating to pancreatic graft. The area under the receiver operating characteristic curve was 0.8, however
AUC: 0.8
1,2
1,0
sensitivity
0,8
the prospective study to validate the model should be necesary. Modification of the indentified factors could decrease the risk of lethal complications related to pancreatic graft. In our model, all variables would be modifiable. Many authors identified donor age as a significant factor that influences pancreas-related complications [5,15,16]. The pancreas donor risk index and preprocurement pancreas allocation suitability score included donor age as a risk factor of pancreas graft failure [24,25]. Thus, selection of donors in terms of age is essential. The next significant factor was the time of pancreatic vessel graft anastomosis. Fellmer et al found that not cold but the second warm ischemic time (anastomosis time) was related to the prevalence of pancreas-specific complications [16]. Others showed that preservation time over 20 or 24 hours was the risk factor for technical failure [15,17,24]. In our study, the mean of the total ischemia time was below 9 hours; thus, it could be the reason that preservation time became not significant in analysis. The duration of recipient dialysis above 24 months prior to SPKTx was identified as the third risk factor. There were no fatal complications among simultaneous pancreas and preemptive kidney transplant recipients; therefore, variable preemptive vs nonpreemptive kidney transplantation could not be considered for methodological reasons. Nevertheless, duration of dialysis less than 24 months prior to SPKTx decreased the risk of lethal complications by 7-fold. Based on International Pancreas Transplant Registry data, Gruessner found that relative risk of death was significantly increased in SPK recipients who were on pretransplant dialysis (RR, 1.49; P < .017) [1]. Ojo showed that, compared with patients who underwent preemptive transplantation, dialysis duration of 1e6 months was associated with 15% higher mortality risk (RR, 1.15, P < .001) and 24 months of dialysis time was associated with 75% higher risk of death (RR, 1.75, P < .001) [26]. Interestingly, as far as only the technical failure or surgical complication development were concerned, the duration of the pretransplant dialysis had no importance [16,17]. Some authors did not even consider duration of dialysis as a risk factor in multivariate analysis [15,24]. From the practical point of view, the shorter duration time of dialysis, the fewer mortal complications that were developed by the recipient. Therefore SPKTx should be performed as fast as possible. This should have an influence on construction of the waiting list.
0,6
ACKNOWLEDGMENTS
0,4
The authors thank Mr Grzegorz Migut and Mr Janusz Wa˛troba, PhD, from StatSoft Polska for excellent help in statistical analysis.
0,2
0,0
-0,2 -0,2
REFERENCES 0,0
0,2
0,4
0,6
0,8
1,0
1,2
1-specificity
Fig 1. Area under the receiver operating characteristic curve (AUC) for described model.
[1] Gruessner AC. Update on pancreas transplantation: comprehensive trend analysis of 25,000 cases followed up over the course of twenty-four years at the International Pancreas Transplant Registry (IPTR). Rev Diabet Stud 2011;8:6e16.
COMPLICATIONS AFTER SIMULTANEOUS SPKTX [2] Kandaswamy R, Stocka PG, Skeansa MA, et al. OPTN/ SRTR 2011 Annual Data Report: pancreas. AJT 2012;12:47e70. [3] Cohn JA, Englesbe MJ. The financial implications of pancreas transplant complications. In: Preedy VR, Watson RR, editors. Handbook of disease burdens and quality of life measures. New York, N.Y: Springer; 2010. pp. 661e2. [4] Cohn JA, Englesbe MJ, Ads JL. Financial implication of pancreas transplant complications: a business case for quality improvement. Am J Transplant 2007;7:1656e60. [5] Troppmann C. Surgical complication requiring early relaparotomy after pancreas transplantation. Ann Surg 1998;227:255e68. [6] Boggi U, Vistoli F, Del Chiaro M, et al. Surgical complications in the modern era of pancreas transplantation. J Pancreas 2006;7:522. [7] Steuer W, Malaise J, Mark W, et al. Euro-SPK Study Group. Spectrum of surgical complications after simultaneous pancreaskidney transplantation in a prospectively randomized study of two immunosuppressive protocols. Nephrol Dial Transplant 2005;20:ii54. [8] Zer M, Dux S, Dintsman M, et al. The timing of relaparatomy and its influence on prognosis. A 10-year survey. Am J Surg 1980;139:338e43. [9] Kirk RM. Reoperative surgery for early complication following abdominal and abdominothoracic operations. J R Soc Med 1988;81:7e9. [10] Hinsdale JG, Jaffe BM. Reoperation for intraabdominal sepsis. Indications and results in modern critical-care setting. Ann Surg 1984;199:31e6. [11] Szmidt J, Lao M, Grochowiecki T, et al. Pancreas transplantation: four vascular anastomoses. Transplant Proc 1996;28: 3511e3. [12] Dindo D, Demartines N, Clavien PA, et al. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205e13. [13] Clavien PA, Barkun J, Oliviera ML, et al. The ClavienDindo classification of surgical complications. Ann Surg 2009;250: 187e96. [14] Kalil AC, Mattei J, Florescua DF, et al. Recommendations for the assessment and reporting of multivariable logistic regression in transplantation literature. Am J Transplant 2010;10:1695e703.
2809 [15] Humar A, Ramcharan T, Kandaswamy R, Gruessner RW, Gruessner AC, Sutherland DE. Technical failures after pancreas transplants: why grafts fail and the risk factorsda multivariate analysis. Transplantation 2004;78:1188e92. [16] Fellmer PT, Pascher A, Kahl A, et al. Influence of donorand recipient-specific factors on the postoperative course after combined pancreasekidney transplantation. Langenbecks Arch Surg 2010;395:19e25. [17] Finger EB, Radosevich DM, Dunn TB, et al. A composite risk model for predicting technical failure in pancreas transplantation. Am J Transplant 2013;13:1840e9. [18] Clavien PA, Strasberg SM. Severity grading of surgical complications. Ann Surg 2009;250:197e8. [19] Castoldi R, Ferrari G, Staudacher C, et al. Segmental ductinjected versus whole-bladder drained pancreas transplantation: the San Raffaele Hospital experience. Transplant Proc 1997;26:450. [20] Kim YH, Park JB, Lee SS, et al. How to avoid graft thrombosis requiring graftectomy immediate post-transplant CT angiography in pancreas transplantation. Transplantation 2012;94: 925e30. [21] Grabowska-Derlatka L, Pacho R, Grochowiecki T, Jakimowicz T, Rowi nski O. Imaging of pancreatic transplant vessels and anastomoses with 16-row multidetector computed tomography. Transplant Proc 2006;38:266e8. [22] Sutherland DER, et al. Lessons learned from more than 1000 pancreas transplants at a single institution. Ann Surg 2001;233:463e501. [23] Sollinger HW, Odorico JS, Becker YT, D’Alessandro AM, Pirsch JD. One thousand simultaneous pancreas-kidney transplants at a single center with 22-year follow-up. Ann Surg 2009;250:618. [24] Axelorod DA, Sung RS, Meyer KH. Systemic evaluation of pancreas allograft quality, outcomes and geographic variation in utilization. Am J Transplant 2010;10:837e45. [25] Vinkers MT, Rahmel AO, Slot MC, et al. How to recognize suitable pancreas donor: a Eurotransplant study of preprocurement factors. Transplant Proc 2008;40:1275. [26] Ojo AO, Meier-Kriesche HU, Hanson JA, et al. The impact of simultaneous pancreas-kidney transplantation on long-term patient survival. Transplantation 2001;71:82e90.