Contrast-Enhanced Ultrasonography in the Early Period After Kidney Transplantation Predicts Long-Term Allograft Function V. Schwengera, V. Hankelb, J. Seckingera, S. Macher-Göppingerc, C. Moratha, M. Zeisbricha, M. Zeiera, and L.P. Kihma,d,* a
Department of Nephrology, University Hospital, Heidelberg; bDepartment of Radiation Oncology, University of Munich, Munich; Department of Pathology, University Hospital, Heidelberg; and dDepartment of Endocrinology, University Hospital, Heidelberg, Germany c
ABSTRACT Introduction. Real-time contrast-enhanced sonography (CES) can assess microvascular tissue perfusion using gas-filled microbubbles. The purpose of the study was to evaluate the feasibility of early CES in predicting long-term kidney allograft function in comparison to color Doppler ultrasonography (CDUS). Methods. We prospectively studied 68 consecutive kidney transplant recipients using CES and conventional CDUS investigation 1 week after transplantation. Transplant tissue perfusion imaging was performed by low-power imaging during intravenous administration of the sonocontrast SonoVue. Renal tissue perfusion was assessed quantitatively using flash replenishment kinetics of microbubbles to estimate renal blood flow (RBF). The obtained sonography values were correlated with clinical data 1 week up to 1 year after transplantation. Results. In contrast with conventional CDUS resistive indices, RBF estimated by CES 1 week posttransplantation significantly correlated with kidney function after 1 year (r ¼ 0.67; P < .001). Determination of RBF by CES revealed a significant correlation with donor age but not recipient age, whereas conventional CDUS resistive index was significantly correlated to recipient age (r ¼ 0.54; P < .001) but not donor age. Furthermore RBF was associated with vascular fibrosis and intimal thickening of the engraftment biopsies. Conclusion. This is the first prospective study demonstrating the prognostic value of CES early after kidney transplantation. In contrast with CDUS, CES reveals information about kidney allograft perfusion independent of recipient vascular compliance.
A
T PRESENT, routine kidney allograft sonography is performed by measuring resistive and pulsatility indices in the larger intrarenal arteries [1e3]. The renal resistive index (RRI) has been proposed as a prognostic marker for allograft and patient survival; however, recent data suggest that RRI is related to the recipients pretransplant macrovascular compliance and therefore not specific for the kidney allograft itself [4e6]. Within this context, it is not surprising that a RRI of >0.8 was associated with recipient mortality as recently demonstrated [7]. In contrast with RRI, contrastenhanced ultrasonography (CES) allows imaging of the actual microvascular kidney perfusion and the quantification of renal blood flow (RBF) with the use of gas-filled microbubbles, and is able to detect early stages of chronic renal allograft injury [8]. Because the prognostic value of early
CES for long-term kidney allograft function is unclear, we performed a prospective study comparing CES with color Doppler ultrasonography (CDUS). MATERIALS AND METHODS Patients This prospective, explorative single-center clinical trial all patients >18 years who underwent kidney transplantation at University
V.S. and V.H. contributed equally to this work. *Address correspondence to Lars Philipp Kihm, MD, Department of Nephrology, University Hospital, Im Neuenheimer Feld 162, 69120 Heidelberg, Germany. E-mail: lars.kihm@med. uni-heidelberg.de
0041-1345/14 http://dx.doi.org/10.1016/j.transproceed.2014.04.013
ª 2014 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
3352
Transplantation Proceedings, 46, 3352e3357 (2014)
CONTRAST-ENHANCED ULTRASONOGRAPHY Hospital Heidelberg between January 2011 and September 2011 were eligible for the study. The local ethical committee approved the study protocol (S-497/2010), and all patients gave written informed consent. Exclusion criteria were defined as renal artery stenosis or arteriovenous fistula of the allograft, screened by CDUS, the presence of aortal synthetic vascular grafts and a lymphocele >150 mL. A total of 80 consecutive kidney allograft recipients were included. Ultrasonography was performed at day 7 after kidney transplantation and patients were prospectively followed for 1 year after kidney transplantation.
Assessment of Resistive Indices by CDUS Sonography was performed by an iU 22 (Philips Medical Systems, Bothell, WA), using a 2.5-MHz convex-array transducer. Two intrarenal Doppler signals were obtained from 3 representative arcuate and interlobar arteries each by a single investigator. The peak systolic velocity (Vmax), the minimal diastolic velocity (Vmin), and the mean velocity (Vmean) were determined; the renal segment arterial resistive index was calculated as 100 (1 [Vmin/Vmax]); the median RRI was assessed using 6 different measurements. RRI values were not used for clinical management.
Real-Time CES Baseline investigation of the allograft was performed in B-mode. After identification of the optimal long axis view of the allograft, the transducer was kept in a stable position, whereas the imaging mode was changed to low mechanical index (MI) contrast-specific imaging according to the international guidelines for the use of contrast agents in ultrasonography [9,10]. Gain settings were optimized for each study; imaging was performed with low MI (MI ¼ 0.1) in colorcoded harmonic power pulse inversion mode, with a low frame rate (7e10 images/sec), color gain of 70%, and dynamic range with a maximum of 170 dB. Five milliliters of SonoVue (Bracco Research SA, Geneva, Switzerland), a second-generation contrast agent, were applied by a perfusor (VueJect, syringe pump for ultrasound contrast agent; Bracco Research SA) over 90 seconds, for constant microvascular perfusion. Initial visualization of the contrast agent occurred after 15e25 seconds. When optimal renal contrast amplification (visualization) was achieved (after 35e45 seconds), a 300-msec pulse, with a high MI (MI ¼ 1.0; burst imaging), was transmitted. Thereafter, the imaging was reset to low power, and replenishment was visualized for 15 seconds. For quantitative analysis of RBF, a region of interest was placed in the renal cortex carefully excluding interlobar and arcuate arteries. The volume of the region of interest was kept at >300 mm2. Contrast intensity was measured after flash and spline curves of contrast intensity versus time were calculated according to an exponential function y ¼ A (1 ebt), as described previously [8,11]. The plateau of signal intensity (A) and the slope of maximal signal intensity rise (b) were measured and the product of A b was calculated in the renal cortex to estimate the blood flow. Images and perfusion were analyzed quantitatively, using a commercially available software tool (Q-Lab, Release 5.0; Philips). All measurements were performed by 2 investigators before the intake of immunosuppressive medication to avoid recently demonstrated effects on RBF [12]. The software analysis was performed by 2 blinded observers.
Laboratory Measurements eGFR was calculated according to the Modification of Diet in Renal Disease formula [13]. All other laboratory measurements were carried out using standard methods in a certified laboratory.
3353
Kidney Allograft Biopsies The implantation biopsy specimen obtained during engraftment of 39 patients were analyzed according to the Banff classification [14,15] with special regard to chronic/sclerosing lesions developing chronic tubulitis, chronic inflammation, and chronic vasculopathyscores for every specimen by a blinded pathologist.
Statistical Analysis For all statistical analysis, Prism Version 5 (GraphPad Software, La Jolla, CA) was used. Demographic and clinical variables are presented as mean values standard deviations. Correlations were analyzed using the Pearson correlation coefficient. For investigation of changes within the group, the 1-tailed Wilcoxon matched-pairs signed-rank test was used. For investigation of changes between groups, the Mann-Whitney U test was used. Receiver operating characteristic (ROC) curves were used to assess diagnostic characteristics CES and CDUS to discriminate kidney allograft function. Differences were considered significant at P < .05.
RESULTS Patients
A total of 80 patients were eligible for the study. Seventyone patients met inclusion criteria and were enrolled. Only 2 patients (2.94%) lost their graft during follow-up. Both patients suffered from refractory acute cellular rejection in the first month after kidney transplantation. There was no association to RRI; 1 patient had an RRI > 0.7 and 1 patient < 0.7, and RBF (1 patient > 12 dB/s, 1 patient < 12 dB/s), respectively. One patient died during the follow-up at month 10 after kidney transplantation owing to acute myocardial infarction. Because of the low incidences, a survival analysis of allografts and patients was not performed. Clinical characteristics and comorbidities of the 68 patients, who completed the follow-up are given in Table 1. The median age was 49.9 years; 41% of the patients were female, and 15% had diabetes mellitus. The immunosuppressive regimen consisted of tacrolimus in 31 patients and cyclosporin in 37. Further immunosuppressive medication included mycophenolate mofetil or enteric-coated mycophenolate sodium and corticosteroids in all patients. All patients underwent induction therapy with the interleukin-2 antagonist basiliximab. Antihypertensive treatment was prescribed in 78% of recipients. According to the median RRI (0.7) and the median RBF (12 dB/s) patients were divided into 2 groups, above and below the median value, respectively. Patients with an RBF > 12 dB/s had a significantly younger donor and a significantly better kidney allograft function 1 week and 1 year after transplantation. Patients with a high RRI (>0.7) were older, had longer cold ischemia times, and a higher prevalence of diabetes mellitus compared with patients with a low RRI ( 12 dB/s (n ¼ 34)
RBF < 12 dB/s (n ¼ 34)
P
RRI < 0.7 (n ¼ 33)
RRI > 0.7 (n ¼ 35)
P
Recipient age (y) Female gender Weight (kg) Diabetes mellitus Donor age (y) Donor female gender Living-related donor HLA matches Cold ischemia time (min) Cyclosporin Tacrolimus Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) RRI RBF (dB/s) eGFR day 7 (mL/min) eGFR day 365 (mL/min)
49.9 14.7 41.2% (28/68) 77.0 15.2 14.7% (10/68) 49.7 13.3 60.3% (41/68) 45.6% (31/68) 2.69 1.73 573 431 54.4% (37/68) 45.6% (31/68) 136 19.3 82.1 13.1 0.70 0.09 12.1 5.62 33.9 23.4 49.4 21.0
46.6 13.8 41.2% (14/34) 74.2 14.5 11.7% (4/34) 46.2 11.7 52.9% (18/34) 52.9% (18/34) 2.64 1.81 504 401 55.8% (19/34) 44.1% (15/34) 139 20.2 83.9 13.5 0.68 0.09 16.3 4.76 42.3 24.9 61.6 17.0
53.2 15.1 41.2% (14/34) 79.8 15.7 17.6% (6/34) 58.7 12.7 67.6% (23/34) 38.2% (13/34) 2.73 1.68 643 455 52.9% (18/34) 47.1% (16/34) 134 18.3 80.2 12.6 0.71 0.09 7.92 2.30 25.6 18.9 38.5 17.7
NS NS NS NS
Department of Nephrology, University Hospital, Heidelberg; bDepartment of Radiation Oncology, University of Munich, Munich; Department of Pathology, University Hospital, Heidelberg; and dDepartment of Endocrinology, University Hospital, Heidelberg, Germany c
ABSTRACT Introduction. Real-time contrast-enhanced sonography (CES) can assess microvascular tissue perfusion using gas-filled microbubbles. The purpose of the study was to evaluate the feasibility of early CES in predicting long-term kidney allograft function in comparison to color Doppler ultrasonography (CDUS). Methods. We prospectively studied 68 consecutive kidney transplant recipients using CES and conventional CDUS investigation 1 week after transplantation. Transplant tissue perfusion imaging was performed by low-power imaging during intravenous administration of the sonocontrast SonoVue. Renal tissue perfusion was assessed quantitatively using flash replenishment kinetics of microbubbles to estimate renal blood flow (RBF). The obtained sonography values were correlated with clinical data 1 week up to 1 year after transplantation. Results. In contrast with conventional CDUS resistive indices, RBF estimated by CES 1 week posttransplantation significantly correlated with kidney function after 1 year (r ¼ 0.67; P < .001). Determination of RBF by CES revealed a significant correlation with donor age but not recipient age, whereas conventional CDUS resistive index was significantly correlated to recipient age (r ¼ 0.54; P < .001) but not donor age. Furthermore RBF was associated with vascular fibrosis and intimal thickening of the engraftment biopsies. Conclusion. This is the first prospective study demonstrating the prognostic value of CES early after kidney transplantation. In contrast with CDUS, CES reveals information about kidney allograft perfusion independent of recipient vascular compliance.
A
T PRESENT, routine kidney allograft sonography is performed by measuring resistive and pulsatility indices in the larger intrarenal arteries [1e3]. The renal resistive index (RRI) has been proposed as a prognostic marker for allograft and patient survival; however, recent data suggest that RRI is related to the recipients pretransplant macrovascular compliance and therefore not specific for the kidney allograft itself [4e6]. Within this context, it is not surprising that a RRI of >0.8 was associated with recipient mortality as recently demonstrated [7]. In contrast with RRI, contrastenhanced ultrasonography (CES) allows imaging of the actual microvascular kidney perfusion and the quantification of renal blood flow (RBF) with the use of gas-filled microbubbles, and is able to detect early stages of chronic renal allograft injury [8]. Because the prognostic value of early
CES for long-term kidney allograft function is unclear, we performed a prospective study comparing CES with color Doppler ultrasonography (CDUS). MATERIALS AND METHODS Patients This prospective, explorative single-center clinical trial all patients >18 years who underwent kidney transplantation at University
V.S. and V.H. contributed equally to this work. *Address correspondence to Lars Philipp Kihm, MD, Department of Nephrology, University Hospital, Im Neuenheimer Feld 162, 69120 Heidelberg, Germany. E-mail: lars.kihm@med. uni-heidelberg.de
0041-1345/14 http://dx.doi.org/10.1016/j.transproceed.2014.04.013
ª 2014 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
3352
Transplantation Proceedings, 46, 3352e3357 (2014)
CONTRAST-ENHANCED ULTRASONOGRAPHY Hospital Heidelberg between January 2011 and September 2011 were eligible for the study. The local ethical committee approved the study protocol (S-497/2010), and all patients gave written informed consent. Exclusion criteria were defined as renal artery stenosis or arteriovenous fistula of the allograft, screened by CDUS, the presence of aortal synthetic vascular grafts and a lymphocele >150 mL. A total of 80 consecutive kidney allograft recipients were included. Ultrasonography was performed at day 7 after kidney transplantation and patients were prospectively followed for 1 year after kidney transplantation.
Assessment of Resistive Indices by CDUS Sonography was performed by an iU 22 (Philips Medical Systems, Bothell, WA), using a 2.5-MHz convex-array transducer. Two intrarenal Doppler signals were obtained from 3 representative arcuate and interlobar arteries each by a single investigator. The peak systolic velocity (Vmax), the minimal diastolic velocity (Vmin), and the mean velocity (Vmean) were determined; the renal segment arterial resistive index was calculated as 100 (1 [Vmin/Vmax]); the median RRI was assessed using 6 different measurements. RRI values were not used for clinical management.
Real-Time CES Baseline investigation of the allograft was performed in B-mode. After identification of the optimal long axis view of the allograft, the transducer was kept in a stable position, whereas the imaging mode was changed to low mechanical index (MI) contrast-specific imaging according to the international guidelines for the use of contrast agents in ultrasonography [9,10]. Gain settings were optimized for each study; imaging was performed with low MI (MI ¼ 0.1) in colorcoded harmonic power pulse inversion mode, with a low frame rate (7e10 images/sec), color gain of 70%, and dynamic range with a maximum of 170 dB. Five milliliters of SonoVue (Bracco Research SA, Geneva, Switzerland), a second-generation contrast agent, were applied by a perfusor (VueJect, syringe pump for ultrasound contrast agent; Bracco Research SA) over 90 seconds, for constant microvascular perfusion. Initial visualization of the contrast agent occurred after 15e25 seconds. When optimal renal contrast amplification (visualization) was achieved (after 35e45 seconds), a 300-msec pulse, with a high MI (MI ¼ 1.0; burst imaging), was transmitted. Thereafter, the imaging was reset to low power, and replenishment was visualized for 15 seconds. For quantitative analysis of RBF, a region of interest was placed in the renal cortex carefully excluding interlobar and arcuate arteries. The volume of the region of interest was kept at >300 mm2. Contrast intensity was measured after flash and spline curves of contrast intensity versus time were calculated according to an exponential function y ¼ A (1 ebt), as described previously [8,11]. The plateau of signal intensity (A) and the slope of maximal signal intensity rise (b) were measured and the product of A b was calculated in the renal cortex to estimate the blood flow. Images and perfusion were analyzed quantitatively, using a commercially available software tool (Q-Lab, Release 5.0; Philips). All measurements were performed by 2 investigators before the intake of immunosuppressive medication to avoid recently demonstrated effects on RBF [12]. The software analysis was performed by 2 blinded observers.
Laboratory Measurements eGFR was calculated according to the Modification of Diet in Renal Disease formula [13]. All other laboratory measurements were carried out using standard methods in a certified laboratory.
3353
Kidney Allograft Biopsies The implantation biopsy specimen obtained during engraftment of 39 patients were analyzed according to the Banff classification [14,15] with special regard to chronic/sclerosing lesions developing chronic tubulitis, chronic inflammation, and chronic vasculopathyscores for every specimen by a blinded pathologist.
Statistical Analysis For all statistical analysis, Prism Version 5 (GraphPad Software, La Jolla, CA) was used. Demographic and clinical variables are presented as mean values standard deviations. Correlations were analyzed using the Pearson correlation coefficient. For investigation of changes within the group, the 1-tailed Wilcoxon matched-pairs signed-rank test was used. For investigation of changes between groups, the Mann-Whitney U test was used. Receiver operating characteristic (ROC) curves were used to assess diagnostic characteristics CES and CDUS to discriminate kidney allograft function. Differences were considered significant at P < .05.
RESULTS Patients
A total of 80 patients were eligible for the study. Seventyone patients met inclusion criteria and were enrolled. Only 2 patients (2.94%) lost their graft during follow-up. Both patients suffered from refractory acute cellular rejection in the first month after kidney transplantation. There was no association to RRI; 1 patient had an RRI > 0.7 and 1 patient < 0.7, and RBF (1 patient > 12 dB/s, 1 patient < 12 dB/s), respectively. One patient died during the follow-up at month 10 after kidney transplantation owing to acute myocardial infarction. Because of the low incidences, a survival analysis of allografts and patients was not performed. Clinical characteristics and comorbidities of the 68 patients, who completed the follow-up are given in Table 1. The median age was 49.9 years; 41% of the patients were female, and 15% had diabetes mellitus. The immunosuppressive regimen consisted of tacrolimus in 31 patients and cyclosporin in 37. Further immunosuppressive medication included mycophenolate mofetil or enteric-coated mycophenolate sodium and corticosteroids in all patients. All patients underwent induction therapy with the interleukin-2 antagonist basiliximab. Antihypertensive treatment was prescribed in 78% of recipients. According to the median RRI (0.7) and the median RBF (12 dB/s) patients were divided into 2 groups, above and below the median value, respectively. Patients with an RBF > 12 dB/s had a significantly younger donor and a significantly better kidney allograft function 1 week and 1 year after transplantation. Patients with a high RRI (>0.7) were older, had longer cold ischemia times, and a higher prevalence of diabetes mellitus compared with patients with a low RRI ( 12 dB/s (n ¼ 34)
RBF < 12 dB/s (n ¼ 34)
P
RRI < 0.7 (n ¼ 33)
RRI > 0.7 (n ¼ 35)
P
Recipient age (y) Female gender Weight (kg) Diabetes mellitus Donor age (y) Donor female gender Living-related donor HLA matches Cold ischemia time (min) Cyclosporin Tacrolimus Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) RRI RBF (dB/s) eGFR day 7 (mL/min) eGFR day 365 (mL/min)
49.9 14.7 41.2% (28/68) 77.0 15.2 14.7% (10/68) 49.7 13.3 60.3% (41/68) 45.6% (31/68) 2.69 1.73 573 431 54.4% (37/68) 45.6% (31/68) 136 19.3 82.1 13.1 0.70 0.09 12.1 5.62 33.9 23.4 49.4 21.0
46.6 13.8 41.2% (14/34) 74.2 14.5 11.7% (4/34) 46.2 11.7 52.9% (18/34) 52.9% (18/34) 2.64 1.81 504 401 55.8% (19/34) 44.1% (15/34) 139 20.2 83.9 13.5 0.68 0.09 16.3 4.76 42.3 24.9 61.6 17.0
53.2 15.1 41.2% (14/34) 79.8 15.7 17.6% (6/34) 58.7 12.7 67.6% (23/34) 38.2% (13/34) 2.73 1.68 643 455 52.9% (18/34) 47.1% (16/34) 134 18.3 80.2 12.6 0.71 0.09 7.92 2.30 25.6 18.9 38.5 17.7
NS NS NS NS
