The Influence of Intraoperative Central Venous Pressure on Delayed Graft Function in Renal Transplantation: A Single-Center Experience G. Bacchi, A. Buscaroli, M. Fusari, L. Neri, M.L. Cappuccilli, E. Carretta, and S. Stefoni ABSTRACT Introduction. Delayed graft function (DGF) is a common complication in kidney transplantation. We sought to evaluate possible correlates for DGF including intraoperative parameters, focusing on fluid replacement and central venous pressure (CVP) values among patients undergoing kidney transplantation at our center. Methods. One hundred fifty-five cadaveric donor transplantations performed at our center between 2001 and 2005 were selected for the study. We compared intraoperative parameters together with 15 other clinical and socio-demographic recipient and donor variables among patients experiencing DGF (n ⫽ 58) versus those with immediate graft function (IGF; n ⫽ 97). All significant variables at P ⬍ .05 upon univariate analysis were entered into a multivariate logistic regression model to identify risk factors for DGF. Results. CVP at awakening of ⱕ8 mm Hg (odds ratio [OR] ⫽ 3.53; 95% confidence interval [CI], 1.63–7.63), fluid input during surgery ⱕ2.250 mL (OR ⫽ 2.12; 95% CI, 1.00 – 4.51), and recipient age ⱖ50 years (OR ⫽ 2.72; 95% CI, 1.11– 6.68) were the strongest correlates of DGF. Conclusions. Our data suggested that reduced intraoperative perfusion as measured using CVP monitoring might increase DGF risk. This study provides the rationale to further investigate the optimal CVP target during this surgery. ELAYED graft function (DGF) is a common complication in the immediate post–renal transplantation period. Although the reported incidence of DGF varies widely, depending on the study sample and the definition,1–9 its negative consequences are well known. Patients experiencing DGF face reduced graft and patient survivals,10 –15 decreased long-term function,1 and greater rates of acute rejection episodes.16,17 Several studies have investigated the clinical correlates of DGF.17–19 Risk index tools have been developed to identify patients at higher risk using United States Renal Data System registry data.20 However, the contribution of intraoperative factors toward DGF risk are less well investigated. Early studies showed that maximal intraoperative hydration with the administration of crystalloid and colloid infusion21,22 or volume expansion with albumin23 have led to improved posttransplantation function and reduced need for dialysis over the first week postsurgery. More recent evidence has suggested that several intraoperative hemodynamic parameters are associated with primary nonfunction (PNF),24 slow graft function,6 or DGF.25 In this study we evaluated possible risk factors for D DGF including intraoperative variables focusing on fluid replacement and central venous pressure (CVP) values among patients undergoing kidney transplantation at our center. METHODS Data Source and Study Sample The study population included adult patients who underwent kidney transplantations from January 1, 2001 to December 31, From the Department of Internal Medicine, Aging and Renal Disease (G.B., A.B., M.L.C., E.C., S.S.), University of Bologna, Department of Anaesthesiology (M.F.), S. Orsola-Malpighi University Hospital, Bologna, Dipartimento di Medicina del Lavoro “Clinica L. Devoto” dell’Università degli Studi di Milano (L.N.), Milan, and Department of Medicine and Public Health (E.C.), University of Bologna, Bologna, Italy. Address reprint requests to Giuliana Bacchi, BSc, PhD, Department of Internal Medicine, Aging and Renal Diseases, Section of Nephrology, S. Orsola University Hospital (Pad. 15), Via G. Massarenti 9, 40138 Bologna, Italy. E-mail: giuliana.bacchi@ unibo.it © 2010 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710 0041-1345/–see front matter doi:10.1016/j.transproceed.2010.08.042 Transplantation Proceedings, 42, 3387–3391 (2010) 3387 3388 2005. Donor, graft, and recipient characteristics as well as clinical features were retrieved from electronic charts with follow-up data extracted from our recorded database. All donors and recipients were white; there were no non– heart-beating donors (NHBD). We excluded cases of PNF, those with multiple or multi-organ transplants, living donor recipients or those experiencing intraoperative surgical complications. We also excluded subjects who experienced surgery complications within the first week after transplantation or had incomplete or illegible CVP data. The 155 who satisfied these inclusion/exclusion criteria were selected for the study. There were 58 patients who experienced DGF, defined as the need for dialysis in the first week posttransplantation. The control group consisted of 97 subjects with immediate graft function (IGF). Perioperative Kidney Transplant Management The patients were usually admitted to the hospital 3– 8 hours prior to cadaveric kidney transplantation. Hemodialysis was performed if the candidate had not been dialyzed for more than 24 hours or if fluid overload or hyperkalemia was present; patients on peritoneal dialysis continued to respect their dialysis schedule. All patients received similar general anesthesia and transplantation procedures performed by the same surgical team using standard intraoperative protocols. In the operating room a bladder catheter was inserted into all patients to monitor urine output; another catheter was placed into a peripheral artery opposite the site of the arterovenous fistula to monitor the blood pressure. CVP was measured using a catheter placed into the superior vena cava through the internal jugular vein. Blood pressure and CVP were recorded on the clinical chart every 7 minutes. Electrocardiograms, blood pressure, heart rate, blood gases, hemoglobin, hematocrit, potassium, and glucose levels were regularly monitored during the operative procedure. Blood transfusions were given for specific indications: significant blood loss and/or hematocrit ⬍25%. Volume management included fluid load with crystalloid, mainly normal saline, and less frequently synthetic colloids, mainly hydroxyethyl starch. Albumin was never used. Before releasing the clamps on the transplant vessels, all patients received 0.5 g hydrocortisone and 250 mg furosemide. When dopamine was used the rate of infusion was from 1–3 ␮g/kg/min. During surgery the antihypertensive of choice was clonidine as needed for severe hypertension and/or hypertensive crisis. Immunosuppression was mainly based on a combination of corticosteroids, and cyclosporine or tacrolimus (depending on the protocol at the time) combined with mycophenolate mofetil (MMF) or sirolimus. Some patients additionally received induction therapy with an interlukin-2 receptor antibody. Study Measures CVP was recorded at surgery start time (CVP1), at graft reperfusion (CVP2), and at awakening (CVP3). Hemodilution during surgery was approximated as the difference between hemoglobin concentration (⌬Hb) at surgery start (Hb1) versus awakening (Hb2). Hypotension was defined as mean arterial pressure (MAP) ⱕ60 mm Hg for 2 data points captured consecutively. MAP was calculated using the formula: diastolic pressure ⫹ 1/3 (systolic pressure ⫺ diastolic pressure). Also recorded were the following: crystalloid infusion volume, colloidal infusion solution, dopamine, antihypertensive therapy, and urine output. Statistical Analysis Statistical analysis was performed using the SAS software package 8.02 (SAS Institute, Cary, NC, USA). Continuous data are pre- BACCHI, BUSCAROLI, FUSARI ET AL sented as mean values ⫾ standard deviation (SD) and categorical data as percentages. Univariate differences between DGF and IGF groups were tested using Wilcoxon-Mann-Whitney tests for continuous variables and chi-square tests for categorical variables. All significant variables upon univariate analysis were entered into a multivariate logistic regression model. The cut-off for dichotomization of CVP and cristalloid volume infusion was based on the median value of their distribution among the whole study sample. The distribution of CVP1 was used for categorization. The risk factors for DGF were expressed as odds ratios (OR) ⫹ 95% confidence interval (CI). P values less than .05 were regarded as statistically significant for all analyses. Statistical power analysis was assessed using Power and Sample Size Calculation software (http://biostat.mc.vanderbilt.edu/twiki/ bin/view/Main/PowerSampleSize) assuming a significance level of .05. The present study achieved a power of 96% to detect an OR of 3.53 for patients with CVP3 ⱕ8 mm Hg compared with patients with CVP3 ⬎8 mm Hg. RESULTS Of 155 recipients, 58 (37%) experienced DGF and 97 (63%) had IGF. Upon univariate analysis (Table 1), older recipients and donors, a donor history of hypertension, and head trauma as the cause of donor death were all more prevalent among DGF compared with IGF patients. The time on dialysis was significantly longer among the DGF group, whereas the dialysis modality was similar between the groups. Classical DGF risk factors such as cold ischemia time (CIT), human leukocyte antigens HLA, and panelreactive antibodies (PRA) were not significantly different between the DGF and IGF groups. Among intraoperative parameters, CVP3 and the quantity of administered fluid infusion were significantly lower in the DGF compared with the IGF group. In addition, a lower CVP at reperfusion was prevalent in the DGF group, but it did not reach significance (P ⫽ .055). Also, receipt of blood transfusions was not associated with increased DGF risk; there were 2 patients in each group who received blood transfusions (not shown). After entering all univariate correlates of DGF in the multivariate logistic regression model (Table 2), a low CVP3 showed the strongest association with the endpoint of interest (OR 3.53; 95% CI, 1.63–7.63). In addition, crystalloid infusion volume ⬍2250 mL resulted in a 2-fold increase in the risk of DGF (OR 2.12; 95% CI, 1.00 – 4.51). Recipients older than 50 years showed a higher risk of DGF compared with their younger counterparts (OR 2.72; 95% CI, 1.11– 6.69). Figure 1 compares the trend in CVP at the 3 study times during surgery. DISCUSSION For several years, CVP monitoring has been the option of choice for intraoperative fluid management based on the sparse evidence that associated it with better early postoperative courses. Today a CVP from 10 –15 mm Hg is recommended in anesthetic management to maintain optimal intravascular volume.26 However, a recent systematic review of literature has challenged the usefulness of CVP to INFLUENCE OF CVP 3389 Table 1. Baseline Characteristics: Comparison of IGF and DGF Groups Intraoperative parameters MAP ⱕ60 mm Hg (%) Antihypertensive therapy (%) Dopamine infusion (%) CVP1 mm Hg (mean ⫾ SD) CVP2 mm Hg (mean ⫾ SD) CVP3 mm Hg (mean ⫾ SD) Hb1, g/dL (mean ⫾ SD) Hb2, g/dL (mean ⫾ SD) Hb ⌬, g/dL (mean ⫾ SD) Liquid infusion mL (mean ⫾ SD) Colloid infusion (%) Urine output, mL (mean ⫾ SD) Recipient Male (%) Age y (mean ⫾ SD) Hemodialysis (%) Time on dialysis mo (mean ⫾ SD) HLA matches n ⬎3 (%) Last PRA ⬎50% (%) Donor Male (%) Age y (mean ⫾ SD) Smoking (%) Hypertension (%) Diabetes (%) Cardiovascular disease (%) CIT h (mean ⫾ SD) Serum creatinine mg/dL (mean ⫾ SD) Cause of death: head injury (%) IGF (n ⫽ 97; 63%) DGF (n ⫽ 58; 37%) P 19.59 22.68 71.13 8.49 ⫾ 3.65 9.41 ⫾ 3.82 9.71 ⴞ 2.94 11.85 ⫾ 1.76 11.04 ⫾ 1.61 1.06 ⫾ 0.89 2400.52 ⴞ 792.46 35.05 283.54 ⫾ 285.14 20.60 17.24 79.31 7.83 ⫾ 3.30 8.90 ⫾ 3.74 8.43 ⴞ 3.28 11.93 ⫾ 1.61 11.24 ⫾ 1.12 1.07 ⫾ 1.06 2161.21 ⴞ 726.57 34.48 222.43 ⫾ 232.53 .8682 .4182 .2603 .2116 .0553 .0037 .7302 .0939 .6914 .0428 .9428 .3883 62.89 44.21 ⴞ 12.20 73.96 33.95 ⴞ 25.16 44.33 4.12 65.52 53.51 ⴞ 11.27 79.31 47 ⴞ 28.42 41.38 8.62 .7415 .0001 .4513 .0011 .7197 .4783 52.58 44.71 ⴞ 17.17 22.68 20.62 6.19 8.22 16.45 ⫾ 4.61 0.90 ⫾ 0.22 40.21 51.72 55.90 ⴞ 14.75 25.86 41.38 10.34 14.52 17.84 ⫾ 4.93 0.97 ⫾ 0.24 24.14 .9180 .0002 .6530 .0055 .3484 .1002 .0988 .1020 .0413 Abbreviation: MAP, mean arterial pressure; CVP1, CVP2, CVP3, central venous pressure at the induction time, at the graft reperfusion, and at the awakening time respectively; Hb1, Hb2, hemoglobin levels at surgery start and awakening time respectively; HLA, human leukocyte antigens; PRA, panel reactive antibodies. Note: Statistically significant differences in bold font. monitor hemodynamic balance and fluid responsiveness during surgery.27 In patients with chronic kidney disease who are generally at high risk for cardiovascular disease,28 volume expansion should be even more carefully monitored to avoid cardiac ischemia and pulmonary edema. Because blood pressure monitoring cannot be used as a reliable criterion for hemodynamic control and pulmonary arterial pressure monitoring is a more invasive technique, CVP still remains the most common guide for volume replacement therapy during transplantation. In agreement with other reports, our data suggested that better early postoperative transplant function could be obtained with CVP monitoring during surgery. Thomsen et al29 indicated a relationship between the onset of renal function and the vascular state among transplant recipients Table 2. Risk Factors for DGF Obtained Using Logistic Regression OR (95% CI) CVP3 ⱕ8 mm Hg Liquids input ⱕ2250 mL Recipient age ⬎50 y 3.53 (1.63–7.63) 2.12 (1.00–4.51) 2.72 (1.11–6.69) immediately after renal transplantation. CVP ⬎5 cm H20 during the perioperative period reduced the number of kidneys with delayed function. Snoeijs et al24 have shown that hemodynamic status during surgery is a major predictor of PNF among NHBD; low systolic blood pressure, low preoperative diastolic blood pressure, and low CVP increased the odds of PNF. More specifically, they reported that an average intraoperative CVP ⬍6 cm H2O increased the odds of PNF by 3.1-fold compared with higher values. In another single-center study of 90 consecutive recipients, good graft function was obtained in 94% of patients using a CVP target of 7–9 mm Hg.25 In our study, patients classified in the low-CVP group based on an end of surgery assessment displayed a higher DGF risk. In contrast, CVP measurements at the start of surgery and at reperfusion were not significantly associated with DGF risk, results that were different from other studies.22,24 We did not investigate the CVP in the intensive care unit (ICU), although Ferris et al30 reported a severe CVP decrease occurring upon arrival in the ICU, which continued for the next 4 – 8 hours postoperatively despite aggressive fluid management. Nevertheless, they noted that this CVP decrease did not have a significant effect on early function. 3390 Fig 1. Trend in CVP: IGF vs DGF. Baseline CVP at the surgery start time was similar in terms of medians in both groups. Then, whereas in patients with DGF the median value of CVP remained unchanged, patients with IGF had an increase in median from 8.2–9 –9.9 mm Hg, which was statistically significant. It is interesting that our patients generally received an average of 2300 ⫾ 770 mL crystalloid infusion volumes, smaller than those generally reported. Additionally we prescribed colloid products in a minority of cases. Our low number of volume infusions and the prevalent use of crystalloid may be explained by our relatively “healthy” patients, due to donor and recipient selection criteria, associated with a little change in Hb during surgery. Despite several strengths, including the serial recording of a number of intraoperative parameters, this study has some limitations. First, our analysis was restricted to a single center, thus questioning the generalizability of the results. Second, all surgeries were assisted by the same anesthetic team, which may lead to a restricted range of CVP values, which may underestimate the association between the predictor and the outcome. Third, our retrospective design cannot test causality hypotheses. In conclusion, although our study had several limitations, we observed that CVP during transplantation showed predictive ability in terms of DGF risk. This study provides the rationale to further investigate the association of intraoperative CVP with fluid volume and hemodynamic responsiveness in end-stage renal disease to derive the optimal CVP target for this type of surgery. REFERENCES 1. Boom H, Mallat MJ, de Fijter JW, et al: Delayed graft function influences renal function, but not survival. Kidney Int 58:859, 2000 2. Giral-Classe M, Hourmant M, Cantarovich D, et al: Delayed graft function of more than six days strongly decreases long-term survival of transplanted kidneys. Kidney Int 54:972, 1998 3. Gonwa TA, Mai ML, Smith LB, et al: Immunosuppression for delayed or slow graft function in primary cadaveric renal transplantation: use of low dose tacrolimus therapy with post-operative BACCHI, BUSCAROLI, FUSARI ET AL administration of anti-CD25 monoclonal antibody. Clin Transplant 16:144, 2002 4. Hetzel GR, Klein B, Brause M, et al: Risk factors for delayed graft function after renal transplantation and their significance for long-term clinical outcome. Transpl Int 15:10, 2002 5. Humar A, Johnson EM, Payne WD, et al: Effect of initial slow graft function on renal allograft rejection and survival. Clin Transplant 11:623, 1997 6. Sandid MS, Assi MA, Hall S: Intraoperative hypotension and prolonged operative time as risk factors for slow graft function in kidney transplant recipients. Clin Transplant 20:762, 2006 7. Johnston O, O’Kelly P, Spencer S, et al: Reduced graft function (with or without dialysis) vs immediate graft function–a comparison of long-term renal allograft survival. Nephrol Dial Transplant 21:2270, 2006 8. Rodrigo E, Ruiz JC, Piñera C, et al: Creatinine reduction ratio on post-transplant day two as criterion in defining delayed graft function. Am J Transplant 4:1163, 2004 9. Perico N, Cattaneo D, Sayegh MH, et al: Delayed graft function in kidney transplantation. Lancet 364:1814, 2004 10. Halloran PF, Hunsicker LG: Delayed graft function: state of the art, November 10 –11, 2000. Summit meeting, Scottsdale, Arizona, USA. Am J Transplant 1:115, 2001 11. Shoskes DA, Halloran PF: Delayed graft function in renal transplantation: etiology, management and long- term significance. J Urol 155:1831, 1996 12. Pérez Fontán M, Rodríquez-Carmona A, Bouza P, et al: Outcome of grafts with long-lasting delayed function after renal transplantation. Transplantation 62:42, 1996 13. Ojo AO, Hanson JA, Wolfe RA, et al: Long-term survival in renal transplant recipients with graft function. Kidney Int 57:307, 2000 14. Bronzatto EJ, da Silva Quadros KR, Santos RL, et al: Delayed graft function in renal transplant recipients: risk factors and impact on 1-year graft function: a single center analysis. Transplant Proc 41:849, 2009 15. Shoskes DA, Cecka JM: Deleterious effects of delayed graft function in cadaveric renal transplant recipients independent of acute rejection. Transplantation 66:1697, 1998 16. Cecka JM, Cho YW, Terasaki PI: Analyses of the UNOS Scientific Renal Transplant Registry at three years-early events affecting transplant success. Transplantation 53:59, 1992 17. Ojo AO, Wolfe RA, Held PJ, et al: Delayed graft function: risk factors and implications for renal allograft survival. Transplantation 63:968, 1997 18. Figueiredo A, Moreira P, Parada B, et al: Risk factors for delayed graft function and their impact on renal transplantation outcome. Transplant Proc 39:2473, 2007 19. McLaren AJ, Jassem W, Gray DW, et al: Delayed graft function: risk factors and the relative effects of early function and acute rejection on long-term survival in cadaveric renal transplantation. Clin Transplant 13:266, 1999 20. Irish WD, McCollum DA, Tesi RJ, et al: Nomogram for predicting the likelihood of delayed graft function in adult cadaveric renal transplant recipients. J Am Soc Nephrol 14:2967, 2003 21. Luciani J, Frantz P, Thibault P, et al: Early anuria prevention in human kidney transplantation. Advantage of fluid load under pulmonary arterial pressure monitoring during surgical period. Transplantation 28:308, 1979 22. Carlier M, Squifflet JP, Pirson Y, et al: Maximal hydration during anesthesia increases pulmonary arterial pressures and improves early function of human renal transplants. Transplantation 34:201, 1982 23. Dawidson IJ, Sandor ZF, Coorpender L, et al: Intraoperative albumin administration affects the outcome of cadaver renal transplantation. Transplantation 53:774, 1992 24. Snoeijs MG, Wiermans B, Christiaans MH, et al: Recipient hemodynamics during non-heart-beating donor kidney transplantation are major predictors of primary nonfunction. Am J Transplant 7:1158, 2007 INFLUENCE OF CVP 25. De Gasperi A, Narcisi S, Mazza E, et al: Perioperative fluid management in kidney transplantation: is volume overload still mandatory for graft function? Transplant Proc 38:807, 2006 26. Caldwell JE, Cook DR: Kidney transplantation. In Cook DR, Davis PJ (eds): Anesthetic Principles of Organ Transplantation. New York; Raven Press; 1994 27. Marik PE, Baram M, Vahid B: Does central venous pressure predict fluid responsiveness? A systematic review of the literature and the tale of seven mares. Chest 134:172, 2008 3391 28. Saran AM, DuBose TD Jr: Cardiovascular disease in chronic kidney disease. Ther Adv Cardiovasc Dis 2:452, 2008 29. Thomsen HS, Løkkegaard H, Munck O: Influence of normal central venous pressure on onset of function in renal allografts. Scand J Urol Nephrol 21:143, 1987 30. Ferris RL, Kittur DS, Wilasrusmee C, et al: Early hemodynamic changes after renal transplantation: determinants of low central venous pressure in the recipients and correlation with acute renal dysfunction. Med Sci Monit 9:CR61, 2003