Intrarenal Resistive Index after Renal Transplantation

Transcription

1 The new england journal of medicine original article Intrarenal Resistive Index after Renal Transplantation Maarten Naesens, M.D., Ph.D., Line Heylen, M.D., Evelyne Lerut, M.D., Ph.D., Kathleen Claes, M.D., Ph.D., Liesbeth De Wever, M.D., Filip Claus, M.D., Ph.D., Raymond Oyen, M.D., Ph.D., Dirk Kuypers, M.D., Ph.D., Pieter Evenepoel, M.D., Ph.D., Bert Bammens, M.D., Ph.D., Ben Sprangers, M.D., Ph.D., Björn Meijers, M.D., Ph.D., Jacques Pirenne, M.D., Ph.D., Diethard Monbaliu, M.D., Ph.D., Hylke de Jonge, M.D., Ph.D., Christoph Metalidis, M.D., Katrien De Vusser, M.D., and Yves Vanrenterghem, M.D., Ph.D.* Abstract Background The intrarenal resistive index is routinely measured in many renal-transplantation centers for assessment of renal-allograft status, although the value of the resistive index remains unclear. Methods In a single-center, prospective study involving 321 renal-allograft recipients, we measured the resistive index at baseline, at the time of protocol-specified renal-allograft biopsies (3, 12, and 24 months after transplantation), and at the time of biopsies performed because of graft dysfunction. A total of 1124 renal-allograft resistiveindex measurements were included in the analysis. All patients were followed for at least 4.5 years after transplantation. Results Allograft recipients with a resistive index of at least 0.80 had higher mortality than those with a resistive index of less than 0.80 at 3, 12, and 24 months after transplantation (hazard ratio, 5.20 [95% confidence interval {CI}, 2.14 to 12.64; P<0.001]; 3.46 [95% CI, 1.39 to 8.56; P = 0.007]; and 4.12 [95% CI, 1.26 to 13.45; P = 0.02], respectively). The need for dialysis did not differ significantly between patients with a resistive index of at least 0.80 and those with a resistive index of less than 0.80 at 3, 12, and 24 months after transplantation (hazard ratio, 1.95 [95% CI, 0.39 to 9.82; P = 0.42]; 0.44 [95% CI, 0.05 to 3.72; P = 0.45]; and 1.34 [95% CI, 0.20 to 8.82; P = 0.76], respectively). At protocol-specified biopsy time points, the resistive index was not associated with renalallograft histologic features. Older recipient age was the strongest determinant of a higher resistive index (P<0.001). At the time of biopsies performed because of graft dysfunction, antibody-mediated rejection or acute tubular necrosis, as compared with normal biopsy results, was associated with a higher resistive index (0.87±0.12 vs. 0.78±0.14 [P = 0.05], and 0.86±0.09 vs. 0.78±0.14 [P = 0.007], respectively). From the Departments of Nephrology and Renal Transplantation (M.N., L.H., K.C., D.K., P.E., B.B., B.S., B.M., H.J., C.M., K.D.V., Y.V.), Pathology (E.L.), Radiology (L.D.W., F.C., R.O.), and Abdominal Transplant Surgery (J.P., D.M.), University Hospitals Leuven, and the Departments of Microbiology and Immunology (M.N., K.C., D.K., P.E., B.B., B.S., B.M., J.P., D.M., K.D.V., Y.V.) and Imaging and Pathology (E.L., L.D.W., F.C., R.O.), KU Leuven both in Leuven, Belgium. Address reprint requests to Dr. Naesens at the Department of Nephrology and Renal Transplantation, University Hospitals Leuven, Department of Microbiology and Immunology, KU Leuven, Herestraat 49, B-3000 Leuven, Belgium, or at maarten.naesens@uzleuven.be. Drs. Naesens and Heylen contributed equally to this article. * Deceased. N Engl J Med 2013;369: DOI: /NEJMoa Copyright 2013 Massachusetts Medical Society. Conclusions The resistive index, routinely measured at predefined time points after transplantation, reflects characteristics of the recipient but not those of the graft. (ClinicalTrials.gov number, NCT ) n engl j med 369;19 nejm.org november 7,

2 The new england journal of medicine In many renal-transplantation centers, measurement of the intrarenal resistive index by means of Doppler ultrasonography is routinely used to evaluate renal allografts. 1,2 The resistive index is derived from the pulsatile flow-velocity waveform. In native kidneys, a higher resistive index, as compared with a lower resistive index, is a significant predictor of progressive renal dysfunction and adverse cardiovascular events. 3-9 A previous cross-sectional study linked an increased intrarenal resistive index after kidney transplantation with an increased risk of graft loss or recipient death. 10 Although these data suggest that the intrarenal resistive index reflects the intrinsic state of the allograft, it seems evident that the resistive index strongly depends on the aortic pulse pressure and aortic stiffness and thus on central hemodynamic factors Apart from one study that showed a significant but weak correlation between the resistive index and chronic histologic damage in renal allografts 10 and a smaller study in which the resistive index was associated with arteriolosclerosis in native kidneys, 15 a conclusive assessment of the role of renal histologic features in the intrarenal resistive index has been lacking. In native kidneys, it is not possible to dissect the contribution of renal histologic features from the contribution of central hemodynamics, because they are closely related and often involve the same pathophysiological processes and because systematic renal biopsies are not routinely performed in native kidneys. Thus, the relative contributions of renal histologic features and central hemodynamic factors to the intrarenal resistive index remain unclear. 16 We performed a prospective study to evaluate the prognostic performance of the intrarenal resistive index with regard to graft function and patient and graft survival in the first years after kidney transplantation. We also performed a detailed analysis of the relation between graft histologic features and the intrarenal resistive index. Methods Inclusion and Exclusion Criteria All patients who received a single kidney transplant at University Hospitals Leuven, in Leuven, Belgium, between March 2004 and October 2007 were included in this prospective cohort study. The intrarenal resistive index was measured at predetermined time points after transplantation: at baseline (once renal function had stabilized during the first hospitalization) and at 3, 12, and 24 months after transplantation (see the Supplementary Appendix, available with the full text of this article at NEJM.org). At 3, 12, and 24 months after transplantation, concomitant protocol-specified renalallograft biopsies were performed. The resistive index was also measured when transplant recipients underwent biopsies because of acute or subacute graft dysfunction. Pretransplantation biopsies were performed in a subgroup of patients. All data were collected during the routine clinical follow-up of the transplant recipients. All patients provided written informed consent to use their clinical data for study purposes. Resistive-index measurements and concomitant biopsies were not performed, or were excluded from the analyses, when factors were present on the day of ultrasonography that could influence the resistive-index measurement; these included radiographically significant renal-artery stenosis (confirmed by means of magnetic resonance angiography), 17 hydronephrosis, and perirenal-fluid collections with marked compression. Primary and Secondary End Points The composite primary end point was a reduction of 50% or more in the estimated glomerular filtration rate (GFR, calculated with the use of the Modification of Diet in Renal Disease equation 18,19 at three consecutive measurements at least 1 day apart) from the value at the time of ultrasonography, end-stage renal failure requiring the reinstitution of dialysis, or recipient death with a functioning graft. 10 Secondary end points were a composite of end-stage renal failure requiring the reinstitution of dialysis or recipient death with a functioning graft (for analysis of overall graft survival), end-stage renal failure requiring the reinstitution of dialysis (for the analysis of graft survival censored at the time of recipient death), a reduction of 50% or more in the estimated GFR from the value at the time of ultrasonography, and recipient death with a functioning graft (for analysis of patient survival). Intrarenal Resistive Index as Determined by Doppler Ultrasonography The peak systolic velocity (V max ) and the minimal diastolic velocity (V min ) were determined in each 1798 n engl j med 369;19 nejm.org november 7, 2013

3 Intrarenal Resistive Index in Renal Transplantation patient in two or three representative interlobar arteries with a multifrequency (2 to 6 MHz) convex transducer (ProSound ALPHA 10, Hitachi Aloka Medical). The resistive index was calculated as [1 (V min V max )]. Results of three measurements at three places in the kidney (lower pole, between poles, and upper pole) were averaged (see the Supplementary Appendix). 20 Intraobserver, intrasession variation and interobserver, intrasession variation were 4.3% and 3.3%, respectively. Resistive-index values were not used for clinical management. Allograft Biopsies After transplantation, percutaneous, ultrasoundguided core needle biopsies (2 14-gauge needle) of the allografts were performed immediately after the measurement of the resistive index. Pretransplantation specimens were obtained by means of wedge biopsy. All biopsy specimens included in this study were reviewed by one pathologist, according to the revised Banff criteria (see the Supplementary Appendix). 21,22 The score on the Chronic Allograft Damage Index (CADI; range, 0 to 18, with higher scores indicating more damage) was calculated for all post-transplantation biopsy specimens as the sum of the histologic scores for interstitial inflammation, interstitial fibrosis, tubular atrophy, increase in mesangial matrix, glomerulosclerosis, and vascular intimal proliferation. 23 The CADI, calculated for early protocol-specified biopsies, correlates with graft survival and is a surrogate end point for transplantation outcome. 23 Statistical Analysis To assess the clinical determinants and the temporal evolution of the resistive index, we used a mixed-model repeated-measures analysis. Correlations were assessed by means of Pearson or Spearman analysis, as appropriate. Univariate and multivariate Cox proportional-hazards analy sis, Kaplan Meier analysis, and log-rank tests were used to examine the association between the resistive index and the primary and secondary end points (see the Supplementary Appendix). On the basis of the previously reported difference in outcomes between a resistive index of less than 0.80 and a resistive index of 0.80 or higher, 10 we divided the patients into two groups and estimated that 12 patients in each group would be sufficient to show a significant difference in the primary end point, with 80% power and an alpha level of Survival analyses were confined to the first 7 years after transplantation. Details of power analyses for the other end points are provided in the Supplementary Appendix. Two-sided P values of less than 0.05 were considered to indicate statistical significance. We used SAS software, version 9.1 (SAS Institute), for all statistical analyses and GraphPad Prism, version 5.0 (Graph- Pad Software), for data presentation. Results Patients A total of 346 patients received transplants during the study period (Fig. 1). A total of 1124 data points were included in this study 203 at the time of biopsies performed because of graft dysfunction and 921 at predetermined time points. In 25 patients, no resistive-index measurements were performed. A total of 321 patients had at least one resistive-index measurement; all reported data refer to these patients (Tables 1 and 2, and Table S1 in the Supplementary Appendix). The mean duration of follow-up for patients with a functioning graft at the time of data extraction from the clinical follow-up database (June 1, 2012) was 6.3±1.0 years (range, 4.7 to 8.2). During this period, 28 patients died with a functioning graft, and 30 patients had graft loss. Resistive Index and Graft Outcome The intrarenal resistive index remained stable over time after transplantation (Fig. S1 and Table S2 in the Supplementary Appendix); the coefficient of variation for the individual patients was 4.7% at the protocol-specified biopsy time points. Of 260 patients with a resistive-index measurement at 3 months, 46 had a resistive index of 0.80 or higher and had a resistive index of less than During follow-up after the resistive-index measurement at 3 months, 15 patients with a resistive index of 0.80 or higher and 12 patients with a resistive index of less than 0.80 died (33% and 6%, respectively); and 2 patients with a resistive index of 0.80 or higher and 11 patients with a resistive index of less than 0.80 lost graft function (4% and 5%, respectively). In a multivariate Cox proportional-hazards analysis, the combined primary end point of a decrease in the estimated GFR of at least 50%, the need for dialysis, or recipient death did not differ significantly between patients with an intrarenal re- n engl j med 369;19 nejm.org november 7,

4 The new england journal of medicine 346 Patients were eligible 25 Were excluded 10 Declined to participate 8 Had graft loss 4 Had medical contraindication for post-transplantation biopsies 3 Had other reasons 321 Were enrolled 295 Underwent resistive-index measurements at predetermined time points after transplantation (921 measurements) 136 Underwent resistive-index measurements at the time of biopsies performed because of graft dysfunction (203 measurements) 202 Underwent resistive-index measurements at 0 mo 110 Had undergone pretransplantation biopsies 119 Did not undergo resistiveindex measurements owing to unstable graft function 260 Underwent resistive-index measurements at 3 mo with concomitant protocol-specified biopsies 61 Did not undergo resistive-index measurements at 3 mo 12 Declined to participate 16 Had undergone recent biopsy because of graft dysfunction 8 Had graft loss or died 1 Was lost to follow-up 2 Had concurrent medical problems 22 Had other or unknown reason 257 Underwent resistive-index measurements at 12 mo with concomitant protocol-specified biopsies 64 Did not undergo resistive-index measurements at 12 mo 27 Declined to participate 4 Had undergone recent biopsy because of graft dysfunction 12 Had graft loss or died 3 Were lost to follow-up 7 Had concurrent medical problems 11 Had other or unknown reason 202 Underwent resistive-index measurements at 24 mo with concomitant protocol-specified biopsies 119 Did not undergo resistive-index measurements at 24 mo 63 Declined to participate 3 Had undergone recent biopsy because of graft dysfunction 18 Had graft loss or died 7 Were lost to follow-up 13 Had concurrent medical problems 15 Had other or unknown reason Figure 1. Enrollment and Resistive-Index Measurements. Resistive-index measurements were performed at predetermined time points after transplantation: at baseline (0 months) and at 3, 12, and 24 months after transplantation. At 3, 12, and 24 months after transplantation, concomitant protocol-specified renal-allograft biopsies were performed. Resistive-index measurements were also performed when transplant recipients underwent biopsies because of acute or subacute graft dysfunction. Pretransplantation biopsies were performed in a subgroup of patients. sistive index of 0.80 or higher and those with a resistive index of less than 0.80 at 3 months after transplantation (P = 0.12) but did differ significantly at 12 and 24 months after transplantation (P = 0.02 and P = 0.04, respectively) (Table 3 and Fig. 2, and Fig. S2 and S3 in the Supplementary Appendix), with adjustment for the relevant clinical variables (Tables S3, S4, and S5 in the Supplementary Appendix). The risk of the combined secondary end point the need for dialysis or recipient death (for analysis of overall graft survival) was significantly higher among patients with a resistive index of 0.80 or higher than among those with a resistive index of less than 0.80 at 3 and 12 months after transplantation, and the difference had borderline significance at 24 months (P = 0.07) (Table 3 and Fig. 2, and Fig. S2 and S3 in the Supplementary Appendix). This association between overall graft survival and the intrarenal resistive index appeared uniquely attributable to the consistent association between the resistive index and recipient death (i.e., graft loss due to death with a functioning graft). The need for dialysis did not differ significantly between patients with a higher resistive index and those with a lower resistive index at all time points after transplantation. Similarly, there was no significant difference in the risk of a 50% reduction in the estimated GFR between patients with a higher resistive 1800 n engl j med 369;19 nejm.org november 7, 2013

5 Intrarenal Resistive Index in Renal Transplantation index and those with a lower resistive index, at all time points after transplantation (Table 3 and Fig. 2, and Fig. S2 and S3 in the Supplementary Appendix). Resistive Index and Graft Function There was no correlation between the estimated GFR and the intrarenal resistive index at 3 months (r = 0.08, P = 0.21) or at 24 months (r = 0.11, P = 0.16). At 12 months, there was a significant but weak correlation (r = 0.17, P = 0.01). The resistive index measured concomitantly with biopsies performed because of graft dysfunction during the first 3 months after transplantation was significantly higher than the resistive index measured concomitantly with the protocol-specified biopsy at 3 months (0.82±0.13 vs. 0.74±0.08, P<0.001). For biopsies performed because of graft dysfunction, there was an inverse correlation between the estimated GFR and the intrarenal resistive index (r = 0.33, P<0.001). Clinical Associations with the Resistive Index At the protocol-specified biopsy time points, older recipient age was the strongest independent factor associated with an increased intrarenal resistive index (P<0.001) (Fig. 3, and Table S2 in the Supplementary Appendix). Other independent factors associated with a higher resistive index were a higher pulse pressure and, to a lesser extent, a lower mean blood pressure, beta-blocker use, and diuretic use. In addition, male donor status was significantly associated with an increased resistive index. Delayed graft function, cold-ischemia time, and donor age were eliminated from the final multivariate model (Table S2 in the Supplementary Appendix). Resistive Index and Histologic Findings Protocol-Specified Biopsies For protocol-specified biopsies at 0, 3, 12, and 24 months, there was no significant or consistent correlation between any histologic finding, or the CADI, and the intrarenal resistive index (Table S6 in the Supplementary Appendix). Similarly, there was no significant or consistent correlation between the histologic features or change in histologic features and the resistive index or change in the resistive index (Tables S7, S8, and S9 in the Supplementary Appendix). When diagnostic categories 22 were analyzed, there Table 1. Demographic and Clinical Characteristics of 321 Renal-Transplant Donors and Recipients at Baseline.* Characteristic Value Donors Age yr 46±15 Male sex no. (%) 184 (57) Deceased no. (%) 302 (94) Cold-ischemia time hr 14.9±5.8 Recipients Age yr 54±15 Male sex no. (%) 197 (61) Repeat transplantation no. (%) 45 (14) No. of HLA mismatches 2.5±1.3 Additional induction therapy no. (%) 135 (42) Other therapy no. (%) Glucocorticoids 321 (100) Calcineurin inhibitors 308 (96) Tacrolimus 263 (82) Mycophenolate 287 (89) Delayed graft function no. (%) 47 (15) * Plus minus values are means ±SD. Patients with increased immunologic risk received interleukin-2 receptor blockers (122 patients [38%]) or antithymocyte globulins (13 patients [4%]), in addition to 3-drug immunosuppressive therapy. Calcineurin inhibitors were tacrolimus and cyclosporine. Delayed graft function was defined as the need for dialysis in the first week after transplantation. were no significant associations between these categories and the resistive index. The mean resistive-index values according to histologic findings were as follows: normal findings at 3, 12, and 24 months, 0.74±0.08, 0.74±0.07, and 0.73±0.08, respectively; subclinical acute T-cell mediated rejection at 3, 12, and 24 months, 0.73±0.07, 0.74±0.16, and 0.75±0.07, respectively; borderline changes at 3, 12, and 24 months, 0.72±0.06, 0.73±0.05, and 0.75±0.09, respectively; acute antibody-mediated rejection at 3, 12, and 24 months, 0.73±0.12, 0.78±0.07, and 0.76±0.07, respectively; and interstitial fibrosis or tubular atrophy (not otherwise specified) at 3, 12, and 24 months, 0.75±0.07, 0.74±0.078, and 0.74±0.07, respectively. None of the comparisons between the histologic diagnostic groups showed significant differences. n engl j med 369;19 nejm.org november 7,

6 The new england journal of medicine Table 2. Post-Transplantation Clinical Characteristics and Histologic Findings at Protocol-Specified Biopsy Time Points and at the Time of Biopsies Performed because of Graft Dysfunction.* Variable 3 Mo (N = 260) Protocol-Specified Biopsies 12 Mo (N = 257) 24 Mo (N = 202) Biopsies Performed because of Graft Dysfunction (N = 203) Clinical characteristics Estimated GFR ml/min/1.73 m 2 49±15 55±17 57±19 21±10 Resistive index 0.74± ± ± ±0.12 Pulse pressure mm Hg 56±12 56±13 56±14 59±14 Blood pressure mm Hg Systolic 133±13 133±13 134±14 137±15 Diastolic 77±8 77±8 78±9 77±9 Histologic findings no. (%) Normal findings or nonspecific changes 76 (29) 102 (40) 81 (40) 26 (13) Acute tubular necrosis of grade 2 or higher 44 (17) 29 (11) 18 (9) 90 (44) Interstitial fibrosis or tubular atrophy of grade 7 (3) 34 (13) 31 (15) 21 (10) 2 or 3 Borderline changes 42 (16) 35 (14) 21 (10) 18 (9) Acute T-cell mediated rejection 32 (12) 12 (5) 5 (2) 83 (41) Acute antibody-mediated rejection 6 (2) 5 (2) 2 (1) 24 (12) * Plus minus values are means ±SD. GFR denotes glomerular filtration rate. Histologic lesions were scored according to the Banff classification. 21,22 Detailed data on the individual histologic lesions are provided in Table S1 in the Supplementary Appendix. Biopsies Performed because of Graft Dysfunction For biopsies performed because of graft dysfunction, there was a significant correlation between the intrarenal resistive index and glomerulitis, C4d deposition in peritubular capillaries, and more severe acute tubular necrosis. Other types of histologic lesions and the CADI were not correlated with the intrarenal resistive index (Table S6 in the Supplementary Appendix). When diagnostic categories 22 were analyzed, there was a significant difference in the resistive index between normal findings and acute antibody-mediated rejection for biopsies performed because of graft dysfunction (0.78±0.14 vs. 0.87±0.12, P = 0.05). Patients with a histologic diagnosis of acute tubular necrosis on the basis of a biopsy performed because of graft dysfunction had a significantly higher resistive index than patients with normal histologic findings at the time of such a biopsy (0.86±0.09 vs. 0.78±0.14, P = 0.007). These differences were no longer significant after correction for multiple comparisons. There were no significant differences in the intrarenal resistive index between normal findings (0.78±0.14) and borderline findings (0.80±0.11), acute T-cell mediated rejection (0.80±0.13), or interstitial fibrosis or tubular atrophy (0.78±0.08). A total of 42 patients had a resistive index of 1.00 at the time of a biopsy performed because of graft dysfunction. Assessment of the biopsy specimen showed acute T-cell mediated rejection in 17 of these patients, acute antibody-mediated rejection in 10 (4 of whom also had T-cell mediated rejection), acute tubular necrosis in 10, borderline changes in 2, nonspecific interstitial fibrosis or tubular atrophy in 1, and normal histologic findings in 2. Discussion Our analysis of systematic, protocol-specified renal-allograft biopsies in a large patient cohort indicated that the intrarenal resistive index is not associated with graft function, with acute inflammatory lesions, or with chronic histologic damage. Instead of reflecting intrarenal disease 1802 n engl j med 369;19 nejm.org november 7, 2013

7 Intrarenal Resistive Index in Renal Transplantation Table 3. Hazard Ratio for Increased Intrarenal Resistive Index According to Primary and Secondary End Points.* End Point 3 Mo after Transplantation (N = 260) Adjusted Hazard Ratio for Resistive Index 0.80 (95% CI) P Value 12 Mo after Transplantation (N = 257) Adjusted Hazard Ratio for Resistive Index 0.80 (95% CI) P Value 24 Mo after Transplantation (N = 202) Adjusted Hazard Ratio for Resistive Index 0.80 (95% CI) P Value Composite primary end point 1.69 ( ) ( ) ( ) 0.04 Secondary end points Need for dialysis or recipient death 3.93 ( ) < ( ) ( ) 0.07 Need for dialysis 1.95 ( ) ( ) ( ) % Reduction in estimated GFR 0.81 ( ) ( ) ( ) 0.29 Recipient death 5.20 ( ) < ( ) ( ) 0.02 * Hazard ratios are for a resistive index of 0.80 or higher as compared with an index of less than The data are based on multivariate Cox proportional-hazards models adjusted for recipient age, donor age, status with respect to delayed graft function, and first or repeat transplantation. The composite primary end point was a reduction of 50% or more in the estimated GFR, end-stage renal failure requiring the reinstitution of dialysis, or recipient death with a functioning graft. processes, the intrarenal resistive index was related to recipient age and central hemodynamic factors; thus, the resistive index was associated with recipient survival but not with graft survival. Taken together with previously reported findings, 10,24-26 our results suggest that the reference range of the resistive index in transplant recipients is dependent on recipient age. Although our study confirms the finding in the previous study, reported by Radermacher et al., 10 that the resistive index is associated with the primary composite end point of a 50% decrease in graft function, the need for dialysis, or recipient death and with the secondary end point of recipient death with a functioning graft, we could not validate the previously described association of a resistive index of 0.80 or higher with graft survival censored at the time of recipient death or a 50% decrease in graft function. This apparent discrepancy may be explained by the cross-sectional design of the study by Radermacher et al., in contrast to our prospective design. In addition, in their study, the ultrasonographic measurements were performed substantially later after transplantation in the group of patients with an intrarenal resistive index of 0.80 or higher than in the group of patients with a lower intrarenal resistive index. The time elapsed after transplantation, which is a major determinant of graft outcome, 27 may have affected the results of that study. 10 The different eras in which the current study and the study by Radermacher et al. were performed, with different immunosuppressive therapy (82% of patients received tacrolimus in the current study and 83% of patients received cyclosporine in the previous study), 10 as well as changes in kidney-allocation procedures and HLA matching in the Eurotransplant region, 28,29 may be factors that confound the comparisons between the two studies. 27 Finally, Radermacher et al. measured the resistive index in segmental renal arteries, whereas we measured it in the interlobar arteries. Although the Doppler signals measured in segmental arteries are not substantially different from those measured in interlobar arteries, 30 further investigation of the potential effect of the technique used to measure the resistive index is important. In our study, the intrarenal resistive index was significantly higher at the time of biopsies performed because of graft dysfunction than at protocol-specified biopsy time points, a finding that is consistent with the observed association between allograft dysfunction and an increased resistive index in previous studies. 31,32 The intrarenal resistive index was associated with intrarenal histologic findings only when the biopsy was performed because of graft dysfunction, with an increased resistive index associated with acute tubular necrosis and acute antibody-mediated rejection. At protocol-specified biopsy time points, changes in the resistive index did not reflect changes in histologic features, although n engl j med 369;19 nejm.org november 7,

8 The new england journal of medicine A Composite Primary End Point B Need for Dialysis or Recipient Death C Need for Dialysis P=0.03 RI <0.80 at 3 mo RI 0.80 at 3 mo Patients Not Reaching Composite Primary End Point Years since Transplantation Patients with Functioning Graft (%) 0 P<0.001 RI <0.80 at 3 mo RI 0.80 at 3 mo Years since Transplantation Patients without Need for Dialysis (%) 0 P=0.86 RI <0.80 at 3 mo RI 0.80 at 3 mo Years since Transplantation No. at Risk RI <0.80 RI No. at Risk RI <0.80 RI No. at Risk RI <0.80 RI D 50% Reduction in Estimated GFR E Recipient Death P=0.63 Patients without 50% Decrease in Estimated GFR (%) Years since Transplantation RI 0.80 at 3 mo RI <0.80 at 3 mo Patients Surviving (%) 0 P< Years since Transplantation RI <0.80 at 3 mo RI 0.80 at 3 mo No. at Risk RI <0.80 RI No. at Risk RI <0.80 RI Figure 2. Kaplan Meier Survival Curves. Kaplan Meier survival curves are shown for the composite primary end point of a reduction of 50% or more in the estimated glomerular filtration rate (GFR), the need for dialysis, or recipient death (Panel A) and for the secondary end points of the need for dialysis or recipient death (for analysis of overall graft survival) (Panel B), the need for dialysis (for analysis of graft survival censored at the time of recipient death) (Panel C), a reduction of 50% or more in the estimated GFR from the value at the time of ultrasonography (Panel D), and recipient death with a functioning graft (for analysis of patient survival) (Panel E), according to the resistive index (RI) at 3 months after transplantation. Data were censored at 7 years after transplantation, at the time of data extraction from the clinical follow-up database (June 1, 2012), and, for death-censored graft survival, at the time of death with a functioning graft. The survival curves according to the RI at 12 and 24 months after transplantation are provided in Figures S2 and S3, respectively, in the Supplementary Appendix. P values were calculated with the use of the log-rank test n engl j med 369;19 nejm.org november 7, 2013

9 Intrarenal Resistive Index in Renal Transplantation missing values were a limitation of this analysis. Whether changes in the resistive index reflect changes in the features of acute histologic lesions at a time of acute or subacute graft dysfunction could not be conclusively evaluated in the current study, because we had no systematic data on resistive-index values before acute graft dysfunction (i.e., before a biopsy performed because of graft dysfunction) or in the follow-up period after treatment of the acute disease. Although the resistive index at a time of acute or subacute graft dysfunction is associated with acute tubular necrosis and acute antibody-mediated rejection, Doppler ultrasonography does not appear to help differentiate between these very different causes of graft dysfunction. Thus, measurement of the resistive index cannot replace a renal-allograft biopsy to confirm the histologic diagnosis. Our single-center study has several limitations. Although it was adequately powered to detect a significant association between the resistive index and graft outcome, the number of patients was relatively small, and the number of recipients with loss of graft function was smaller than the number in the study by Radermacher et al. 10 Furthermore, the analyses in the current study were restricted to the first years after transplantation; therefore, the data cannot be directly extrapolated to a later stage after transplantation. Because we excluded patients with renal-artery stenosis, this study does not provide guidance for the identification of such patients. In addition, although the reproducibility of the resistive-index measurements in our study was high, ultrasound measurements may be operator-dependent. Finally, whether there is an association between graft outcome and the resistive index at the time of acute or subacute graft dysfunction remains to be evaluated, as well as whether our findings can be extrapolated to native kidneys. In conclusion, our study shows that the resistive index, measured at predefined time points in the first 2 years after transplantation, reflects Intrarenal Resistive Index <40 yr yr >60 yr Months since Transplantation Figure 3. Intrarenal Resistive Index over Time, According to Recipient Age. In a mixed-model repeated-measures analysis, older recipient age was significantly associated with a higher resistive index at all time points (P<0.001 for all comparisons) (Table S2 in the Supplementary Appendix). The horizontal lines within the boxes indicate means, the upper and lower ends of the boxes indicate standard deviations, and the whiskers indicate 95th percentiles. recipient factors, and age-specific reference values for the resistive index remain to be established. A resistive index of 0.80 or higher does not reflect the intrinsic characteristics of the graft, nor does a higher resistive index, as compared with a lower resistive index, predict renal-allograft outcome. According to our findings, routine resistiveindex values, measured at predefined time points after transplantation, do not appear to be sufficiently accurate for the management of renal allografts. Validation of our single-center findings is warranted in a larger prospective, multicenter cohort study with a longer follow-up period and biopsies at later time points. Supported by funding from the Clinical Research Foundation, University Hospitals Leuven, Leuven, Belgium (to Drs. Naesens and Lerut). Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. We thank the centers of the Leuven Collaborative Group for Renal Transplantation; their clinicians, surgeons, and nursing staff; and the patients who participated in the study. References 1. Danovitch GM, ed. Handbook of kidney transplantation. 5th ed. Philadelphia: Lippincott Williams & Wilkins, Nankivell BJ, Kuypers DR. Diagnosis and prevention of chronic kidney allograft loss. Lancet 2011;378: Petersen LJ, Petersen JR, Talleruphuus U, Ladefoged SD, Mehlsen J, Jensen HA. The pulsatility index and the resistive index in renal arteries: associations with long-term progression in chronic renal failure. Nephrol Dial Transplant 1997;12: Radermacher J, Ellis S, Haller H. Renal resistance index and progression of renal disease. Hypertension 2002;39: Nosadini R, Velussi M, Brocco E, et al. Increased renal arterial resistance predicts the course of renal function in type 2 diabetes with microalbuminuria. Diabetes 2006;55: Hamano K, Nitta A, Ohtake T, Kobayashi S. Associations of renal vascular resistance with albuminuria and other macroangiopathy in type 2 diabetic patients. Diabetes Care 2008;31: Parolini C, Noce A, Staffolani E, Giar- n engl j med 369;19 nejm.org november 7,

10 Intrarenal Resistive Index in Renal Transplantation rizzo GF, Costanzi S, Splendiani G. Renal resistive index and long-term outcome in chronic nephropathies. Radiology 2009; 252: Sugiura T, Wada A. Resistive index predicts renal prognosis in chronic kidney disease. Nephrol Dial Transplant 2009;24: Okura T, Kurata M, Irita J, et al. Renal resistance index is a marker of future renal dysfunction in patients with essential hypertension. J Nephrol 2010;23: Radermacher J, Mengel M, Ellis S, et al. The renal arterial resistance index and renal allograft survival. N Engl J Med 2003; 349: Ohta Y, Fujii K, Arima H, et al. Increased renal resistive index in atherosclerosis and diabetic nephropathy assessed by Doppler sonography. J Hypertens 2005;23: Hashimoto J, Ito S. Central pulse pressure and aortic stiffness determine renal hemodynamics: pathophysiological implication for microalbuminuria in hypertension. Hypertension 2011;58: Tublin ME, Tessler FN, Murphy ME. Correlation between renal vascular resistance, pulse pressure, and the resistive index in isolated perfused rabbit kidneys. Radiology 1999;213: Heine GH, Gerhart MK, Ulrich C, Köhler H, Girndt M. Renal Doppler resistance indices are associated with systemic atherosclerosis in kidney transplant recipients. Kidney Int 2005;68: Ikee R, Kobayashi S, Hemmi N, et al. Correlation between the resistive index by Doppler ultrasound and kidney function and histology. Am J Kidney Dis 2005;46: Hausberg M, Lang D, Barenbrock M, Kosch M. What do Doppler indices of renal perfusion tell us for the evaluation of renal disease? J Hypertens 2005;23: Gottlieb RH, Lieberman JL, Pabico RC, Waldman DL. Diagnosis of renal artery stenosis in transplanted kidneys: value of Doppler waveform analysis of the intrarenal arteries. AJR Am J Roentgenol 1995;165: Levey AS, Greene T, Kusek J, Beck GL, MDRD Study Group. A simplified equation to predict glomerular filtration rate from serum creatinine. J Am Soc Nephrol 2000;11:155A. abstract. 19. Poggio ED, Wang X, Weinstein DM, et al. Assessing glomerular filtration rate by estimation equations in kidney transplant recipients. Am J Transplant 2006;6: Keogan MT, Kliewer MA, Hertzberg BS, DeLong DM, Tupler RH, Carroll BA. Renal resistive indexes: variability in Doppler US measurement in a healthy population. Radiology 1996;199: Solez K, Axelsen RA, Benediktsson H, et al. International standardization of criteria for the histologic diagnosis of renal allograft rejection: the Banff working classification of kidney transplant pathology. Kidney Int 1993;44: Sis B, Mengel M, Haas M, et al. Banff 09 meeting report: antibody mediated graft deterioration and implementation of Banff working groups. Am J Transplant 2010;10: Yilmaz S, Tomlanovich S, Mathew T, et al. Protocol core needle biopsy and histologic Chronic Allograft Damage Index (CADI) as surrogate end point for longterm graft survival in multicenter studies. J Am Soc Nephrol 2003;14: Krumme B, Grotz W, Kirste G, Schollmeyer P, Rump LC. Determinants of intrarenal Doppler indices in stable renal allografts. J Am Soc Nephrol 1997;8: Vallejos A, Alperovich G, Moreso F, et al. Resistive index and chronic allograft nephropathy evaluated in protocol biopsies as predictors of graft outcome. Nephrol Dial Transplant 2005;20: Schwenger V, Keller T, Hofmann N, et al. Color Doppler indices of renal allografts depend on vascular stiffness of the transplant recipients. Am J Transplant 2006;6: Lamb KE, Lodhi S, Meier-Kriesche HU. Long-term renal allograft survival in the United States: a critical reappraisal. Am J Transplant 2011;11: Doxiadis II, Smits JM, Persijn GG, Frei U, Claas FH. It takes six to boogie: allocating cadaver kidneys in Eurotransplant. Transplantation 2004;77: Frei U, Noeldeke J, Machold-Fabrizii V, et al. Prospective age-matching in elderly kidney transplant recipients a 5-year analysis of the Eurotransplant Senior Program. Am J Transplant 2008;8: Buckley AR, Cooperberg PL, Reeve CE, Magil AB. The distinction between acute renal transplant rejection and cyclosporine nephrotoxicity: value of duplex sonography. AJR Am J Roentgenol 1987; 149: Rigsby CM, Taylor KJ, Weltin G, et al. Renal allografts in acute rejection: evaluation using duplex sonography. Radiology 1986;158: Rifkin MD, Needleman L, Pasto ME, et al. Evaluation of renal transplant rejection by duplex Doppler examination: value of the resistive index. AJR Am J Roentgenol 1987;148: Copyright 2013 Massachusetts Medical Society icmje recommendations The International Committee of Medical Journal Editors (ICMJE) has revised and renamed its Uniform Requirements. The new ICMJE Recommendations for the Conduct, Reporting, Editing and Publication of Scholarly Work in Medical Journals are available at n engl j med 369;19 nejm.org november 7, 2013