99mTc-DTPA dynamic SPECT/CT renogram in adults. Feasibility and diagnostic benefit

doi:10.21203/rs.3.rs-3799428/v1

Download PDF

Research Article

^99mTc-DTPA dynamic SPECT/CT renogram in adults. Feasibility and diagnostic benefit

https://doi.org/10.21203/rs.3.rs-3799428/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 23 May, 2024

Read the published version in Nuclear Medicine Communications →

Version 1

posted

You are reading this latest preprint version

Objective

To investigate the feasibility, quantitative parameters and potential advantages of dynamic SPECT/CT renogram with ^99mTc-DTPA in adults.

Methods

Fifty-five (55) patients aged 19–80 y.o. most of them with obstructive uropathy were enrolled. The imaging protocol included: Day 1: a) Conventional ^99mTc-DTPA planar renogram, followed by b) ^99mTc-DMSA injection and static planar imaging 3 hrs post injection. Day 3: c) Dynamic SPECT/CT renogram post injection of 333–444 MBq of ^99mTc-DTPA. This included an initial CT scan followed by 12 consecutive SPECT sessions acquired via continuous acquisition mode for a total of 24 minutes. CT-derived maps were used for attenuation compensation of SPECT data dSPECT(AC). Renal activity was measured in the total kidney volume by Regions of Interest (ROIs) drawn manually on consecutive transaxial CT slices of the 3rd SPECT, which were then copied on the whole 12-SPECT series. Corresponding time-activity curves were created. Renographic dSPECT(AC) parameters were compared to those of conventional renogram. The reference method for split renal function was ^99mTc-DMSA (geometrical mean of anterior and posterior projection counts) and for Glomerular Filtration Rate (GFR) the 2-blood sample clearance method with Cr-51 EDTA.

Results

dSPECT(AC) images were of good quality. There was good correlation between renogram parameters (time to peak activity and NORA20) comparing the two techniques (Pearson’s r = 0.959 and 0.933, respectively). In 21 cases with > 30% absolute difference between the two kidneys, spilt renal function calculation by dSPECT(AC) correlated perfectly (r = 0.968) with the reference method, whereas planar renogram was less accurate (r = 0.843). Anatomic information provided by non-enhanced CT offered an integrated structural-functional view valuable for final diagnosis. dSPECT(AC) early kidney uptake as a fraction of injected dose correlated better with reference GFR (r = 0.789) than that estimated by the Gates’ method (r = 0.642).

Conclusions

^99mTc-DTPA dynamic 3D renogram corrected for attenuation and fused with CT images is feasible with conventional gamma cameras incorporated into hybrid SPECT/CT systems. It allows accurate split renal function measurement, is enriched by useful anatomical information and can be used for closer approximation of GFR compared with Gates’ method.

Renal dynamic scintigraphy (radioisotopic renogram) is an established modality for the qualitative and quantitative estimation of renal function. It is performed by acquiring sequential abdominal images via a gamma-camera, immediately after the intravenous administration of a radioactive tracer. After drawing regions of interest (ROIs) over the kidneys, time-activity curves are created, which represent the radioactivity changes throughout time and reveal changes in physiological uptake, retention or excretion of radiotracers. The radioisotopic renogram is indicated and contributes to the diagnosis of congenital anomalies, vascular and other abnormalities, urinary tract obstruction and parenchymal lesions. It provides valuable qualitative and quantitative parameters with diagnostic and prognostic value, such as split renal function (SRF) and the excretion rate of injected activity. Various types of pharmacologic and physiologic interventions can be implemented (e.g. diuretic drugs for the evaluation of obstructive uropathy), depending on clinical indications [1, 2]. Moreover, glomerular filtration rate (GFR) or effective renal plasma flow (ERPF) can be approximated [3, 4]. Radiotracers most widely used are Tc-99m diethylene-triamine-pentaacetic acid (DTPA) and Tc-99m mercapto-acetyl-triglycine (MAG-3), with the latter being more suitable for children and adult patients with renal failure [4].

The main disadvantages of planar dynamic renal scintigraphy are: suboptimal spatial resolution, lack of anatomic information and potentially inaccurate quantitative results, particularly when the kidneys have marked size differences or ectopic location due to dissimilar gamma-ray tissue attenuation between kidneys. Moreover, the method is prone to pitfalls, which have been already reported [5].

Nowadays, Single Photon Emission Computed Tomography/Computed Tomography (SPECT/CT) is widely used in daily Nuclear Medicine practice, providing attenuation correction of gamma-rays, anatomical location of scintigraphic findings and additional diagnostic information from CT images [6]. Recently, quantitative SPECT/CT is gaining growing importance in radionuclide therapy dosimetry and other applications [7]. This technique has rarely been used before for the improvement of planar radioisotopic renogram [8–11].

The purpose of the present study was a) to address the feasibility of ^99mTc-DTPA dynamic SPECT/CT renogram in adults and b) to compare it with the conventional planar renogram in qualitative and quantitative terms.

1. Patients

Sixty-two (62) consecutive adult patients referred to our laboratory for a DTPA and a dimercapto-succinic acid (DMSA) scan initially entered the study after informed written consent. The study was approved by the ethical committee of our institution. The whole study protocol was completed in 55 patients. There were 4 rejections due to technical SPECT/CT errors and 3 patient no-shows on the 2nd day of the protocol (see below). Twenty-three (23) males and 32 females formed the final study population, aged 19–80 years old (56.3 ± 16.0), with a body surface area ranging from 1.40 to 2.26 Kg/m² (1.85 ± 0.22). Forty-three (43) patients presented with hydronephrosis, 2 had a small kidney on ultrasound examination, 2 had uncontrolled hypertension, 2 were kidney donor candidates, and 4 had various other renal diseases. There was one patient with an ectopic kidney and another with a single kidney.

2. Study protocol

The study protocol included radioisotopic procedures taking place on two different days, separated by 48 hours, as follows:

Day 1: a) Conventional ^99mTc-DTPA planar renogram for 30–40 min in posterior projection, with lasix given at 20 min, when necessary, followed in this case by another 20 min data acquisition; b) ^99mTc-DMSA planar scan in anterior and posterior projection.

Day 3: c) ^99mTc-DTPA dynamic SPECT renogram, corrected for attenuation via CT-derived attenuation maps [dSPECT(AC)]; d) ⁵¹Cr-ethylenediamine tetraacetic acid (EDTA) GFR measurement, by taking blood samples at 2 and 4 hours after tracer injection [12].

The radioisotopic procedures, radioactive agents used, initiation time of each procedure, injected tracer doses and procedures’ duration are presented in Table 1.

Table 1

Study protocol: radioisotopic procedures, time of initiation, radioactive agents administered, injected doses, elapsed time from injection and duration of each procedure.
	Time		Dose (MBq)	Elapsed time	Procedure
DAY 1	0	^99mTc-DTPA	111–148	0	Planar renogram (lasix was given at 20 min in most cases)	20–40 min
	1 hr	^99mTc-DMSA	111–148	3 hrs	Planar images (in anterior and posterior projection)	4 min
DAY 3	0	^99mTc-DTPA	333–444	0	CT + Dynamic SPECT	3 + 24 min
	0	⁵¹Cr-EDTA	1.1	2 and 4 hrs	Blood sample collection

Either the Millenium VG or the Infinia 2-head gamma cameras (GE Healthcare), equipped with low-energy all-purpose collimators, were used for planar DTPA renogram and with high-resolution collimators for DMSA studies. The Millenium VG Hawkeye SPECT/CT system, with low-energy all-purpose collimators, was used for the dynamic SPECT renogram. This gamma camera model permits continuous SPECT acquisition, without time losses from one angular step to the other during gamma camera rotation. A CT scan was performed at the beginning of the imaging series. A number of CT slices were generated over a user-specified path length, which encompassed the area of the kidneys. Acquisition settings were 216° rotation of the X-ray tube; 10 mm slice step; 14 seconds slice time; 140 kV voltage; 2.5 mA current; 128×128 matrix size. A series of SPECT acquisitions ensued, using 180⁰ rotation of each camera head (360⁰ total) with body contour detection, 6⁰ angular step, 64x64 pixels matrix size and energy window set at 140 kEV ± 30%, with various time per frame settings, for a total of 24 min. No lasix was given afterwards. As regards time per frame settings and duration of each SPECT, the dSPECT(AC) acquisition protocol is detailed in Fig. 1.

2. Data processing

Data processing of the conventional planar renogram included the construction of renographic curves and the calculation of various curve parameters. Among these and for the purpose of simplicity, a) the time to peak activity (TTP), b) split renal function (SRF), c) the normalized residual activity ratio at 20 minutes (NORA20) and d) GFR estimation were chosen for comparison with dSPECT(AC). Whenever TTP exceeded 20 minutes then the value 20 was assigned to this parameter. Renal activity at 2 to 3 minutes post injection was used for the calculation of SRF with both techniques. NORA20 is the ratio of renal activity at 20 minutes post injection divided by the 2-min activity [13]. GFR was calculated according to Gates’ method [3].

The tomographic reconstruction of raw dSPECT(AC) data was accomplished on a “Xeleris 3” workstation (GE Healthcare), by iterative algorithm (2 iterations, 10 subsets) and using Butterworth post-filtering (order 5, with the cut-off set at 0.35 of Nyquist frequency). The CT-created attenuation map was used for attenuation correction of the entire dynamic SPECT series. CT and SPECT slices in transaxial, coronal and sagittal planes were viewed together to determine adequate SPECT-CT co-registration. CT images were examined for potentially useful anatomic information (renal cysts, renal pelvis or ureter dilation, stones, etc.). Figure 2 presents an example.

The reconstructed data were converted into DICOM format and transferred to the “Image J” software [14], which was downloaded on a PC for further processing. Regions of interest (ROIs) surrounding renal parenchyma were manually drawn in consecutive transaxial SPECT slices on the 3rd SPECT, representing renal activity at 2 to 3 minutes post injection. The corresponding CT slices were inspected simultaneously, in order to aid ROI creation. Afterwards, this ROI set was copied on the whole SPECT series. Renographic curves were created in an xlsx worksheet, after inserting the numerical data obtained from “Image J”. TTP, SRF and NORA20 were calculated from the renographic curves. Similar to the method proposed be Gates [3], GFR estimation relied on renal activity at 2–3 minutes post-injection as a fraction of injected activity. For injected activity measurement, the syringe before and after tracer administration was imaged on the gamma camera by the SPECT/CT technique and the net activity was calculated.

The geometric mean of measured renal activity in anterior and posterior projection in ^99mTc-DMSA images served as the reference method for SRF. ⁵¹Cr-EDTA served as the reference method for GFR.

Four Nuclear Medicine physicians (MS, NP, TS and DJA of the authors) were involved in the interpretation of planar and dSPECT(AC) studies. Interpretation inconsistencies were resolved by consensus.

3. Statistical analysis

Numerical data were expressed as the mean ± standard deviation (sd). The paired student’s t-test or the Wilcoxon signed-rank test, as appropriate, was used for the comparison of numerical parameters. The level of significance (p) was set below 0.05. The Pearson’s coefficient was used to express the correlation of numerical data series. The Bland-Altman analysis was implemented for the GFR estimation by the tested techniques in comparison with the reference method.

1. Image quality

Serial images of dSPECT(AC) were of adequate quality in all cases, even those with renal insufficiency. An example is given in Fig. 3, where maximum intensity projection images frozen in the posterior view are compared with those of conventional planar renogram.

2. Renographic parameters

The renographic curves of planar and dSPECT(AC) renogram were generally similar on visual evaluation. Figure 4 presents an example. The comparison of quantitative parameters was as follows below:

a) TTP: In 109 renal units examined, there was excellent correlation between dSPECT(AC) and planar renogram, with Pearson’s r = 0.959 (r² = 0.920). However, dSPECT(AC) TTP values were lower compared with those of planar renogram (7.83 ± 7.06 vs. 8.52 ± 6.78, respectively, p = 0.0004).

b) NORA20: Regarding this parameter, there was also high correlation between the two techniques, with r = 0.933 (r² = 0.871). Again, dSPECT(AC) NORA20 values were lower than those of planar renogram (0.928 ± 0.687 vs. 1.050 ± 0.728, respectively, p < 0.0001). The final diagnosis was similar with the planar and dSPECT(AC) renogram. In 4 patients examined for obstructive uropathy, who had received 10 mgr of i.v. furosemide during their planar study and had a good response to the diuretic stimulus, the NORA20 index was normal with dSPECT(AC) renogram (obviating the need for diuretic administration).

c) SRF: In the whole series of 55 patients, dSPECT(AC) exhibited slightly better correlation with the reference method (r = 0.982) than that of planar renogram (r = 0.924). However, in examining the mean of the absolute difference between the left and the right kidney, dSPECT(AC) values did not differ from those obtained by the DMSA scan (32.9%±31.5%,vs. 33.0%±31.6%, respectively, p = 0.92), whereas planar renogram values were significantly different (26.7%±25.4%, vs. 33.0%±31.6%, respectively, p = 0.0005). In 21 cases with more than 30% absolute difference between the two kidneys, the correlation of the two tested techniques with the reference method was r = 0.968 and 0.843, for dSPECT(AC) and planar renogram, respectively (Fig. 5). Similarly, in this subgroup, dSPECT(AC) values did not differ from those of the DMSA scan (66.5 ± 24.4 vs. 67.1 ± 23.6, p = 0.66.) whereas planar renogram values were significantly different (52.8 ± 22.4 vs. 67.1 ± 23.6, p = 0.0001).

d) GFR: Eighteen patients (18/55, 32.7%) had ⁵¹Cr-EDTA GFR values normalized for body surface area below 60 ml/min/1.73m², ranging from 16.8–52.3 (mean ± sd = 38.3 ± 10.3). In the entire study population, GFR estimation by dSPECT(AC) exhibited better correlation with the reference method (r = 0.789, r² = 0.622) than the planar renogram (r = 0.642, r² = 0.411). The correlation and the Bland-Altman graphs of the two tested techniques are shown in Fig. 6.

3. Additional diagnostic information obtained by the CT scan

The non-enhanced CT image obtained by the Hawkeye PECT/CT system, despite its suboptimal resolution, provided useful additional diagnostic information (renal cysts, pelvic system and/or ureter dilation, extra-renal pelvis, radiopaque stones, etc) in certain patients and explained many scintigraphic findings. Illustrative cases are depicted in Figs. 2 and 7.

SPECT/CT has gained widespread acceptance for many diagnostic applications of Nuclear Medicine [6]. However, its implementation in dynamic studies has been limited, mainly due to the low sensitivity of conventional SPECT gamma cameras to perform fast data acquisition. Some investigators have explored the feasibility of dynamic SPECT (dSPECT) with conventional gamma cameras and proposed its application in Nuclear Medicine studies [11, 15, 16], yet this approach was not applied in routine clinical practice. The interest in dSPECT has been renewed lately with the introduction of modern cadmium zinc telluride (CZT) cameras, particularly for cardiac studies [18]. Regarding renal dynamic SPECT, only few reports exist in literature [8–11].

The present study proposes a dynamic SPECT/CT acquisition protocol for ^99mTc DTPA renogram, compatible with conventional gamma cameras, via continuous-mode SPECT data collection. Apart from few cases where technical errors occurred during the dynamic SPECT series, the whole dSPECT(AC) protocol was completed in the vast majority of the study cohort. In order to increase counting statistics, particularly throughout the first 6 to 8 fast acquisitions (Fig. 1), the injected tracer activities were higher than usual (Table 1), albeit within the acceptable dose range for adults.

The quality of SPECT slices was deemed acceptable providing adequate spatial resolution (Fig. 3) and reliable quantitative results. The shape of the renographic curves was similar between the planar and the tomographic technique in the vast majority of cases. There was also excellent correlation between techniques, regarding TTP and NORA20 indices. These results are in accordance with Humphries’ findings presented in abstract form [8]. In that study, οptimal dSPECT image acquisition protocols (370 MBq injected, 360 degrees, 20 sec per 64 stops) yielded renal functional information provided by time activity curves (TACs) with an overall mean difference within 7–10% of planar TACs. Quantitative dSPECT analysis of 1mL renal volumes demonstrated consistent time to peak and clearance characteristics of the TACs, with peak amplitudes varying for different kidney regions. In the present study, the dSPECT(AC) values of TTP and NORA20 parameters were lower, probably due to more accurate delineation of renal parenchyma (and greater exclusion of renal pelvic activity) in drawing corresponding ROIs. Nevertheless, the final diagnosis was similar with the two techniques in all cases. Possibly, the implementation of dSPECT(AC) could obviate the need of diuretic stimulus in a minority of cases.

There are three main points in which dSPECT(AC) seems to clearly improve planar DTPA renogram according to our results.

a) First, for more accurate calculation of split renal function. SFR measured by dSPECT(AC) exhibited higher correlation with the reference method than planar renogram, particularly in cases with marked functional differences between the two kidneys. In line with this observation, Miyazaki et al reported similar results [10]. In that study, the authors implemented dynamic SPECT for 5 minutes post injection, corrected by CT-derived attenuation maps. They found that SRF derived from DTPA SPECT showed a better correlation with DMSA SPECT/CT (r = 0.849) than DTPA planar renogram (r = 0.663, P < 0.0001).

b) Second, for closer approximation of GFR. In our study, GFR calculation by dSPECT(AC) correlated better with the reference method than planar renogram. The method of GFR estimation proposed by Gates [3] has been automated and incorporated in the processing software of the renogram by many gamma-camera vendors. However, this technique suffers from many inaccuracies. Most importantly, the correlation of early DTPA uptake with GFR is arbitrary and does not take into consideration the clearance kinetics of the radiotracer from the renal parenchyma. Second, gamma-ray attenuation compensation relies on historical normograms, relating body mass index with the depth of the kidneys inside the body. However, kidney size and location differences result in dissimilar attenuation of radioactivity. Consequently, it seems reasonable that dSPECT(AC), by applying attenuation correction and more accurate delineation of renal parenchyma yields better results than planar scintigraphy. The importance of attenuation correction in dSPECT studies has been underlined also by other investigators [17].

c) Third, CT images provide useful additional diagnostic information regarding renal and pelvic system anatomy and relative disorders. Aibing et al reported similar results [9]. These authors did not perform dynamic SPECT, but a single SPECT/CT session after the conclusion of planar renogram. In examining 47 patients with hydronephrosis, they reported that their technique had a clear benefit in terms of diagnostic confidence in daily clinical practice.

Recently introduced CZT gamma-cameras provide significantly improved sensitivity, as well as improved spatial and energy resolution, allowing better counting statistics and/or an important reduction in the administered dose of Tc-99m labelled tracers. The present study could serve as an incentive for subsequent dynamic SPECT renal studies with this modern technology.

LIMITATIONS

Even with the use of continuous-mode acquisition and increased tracer doses injected, the counting statistics of dSPECT(AC) slices were not optimal, particularly in the first 6 SPECT acquisitions, where the time per frame was only 1 second. We implemented that fast SPECT data collection, 1sec/frame for 6 minutes and 2sec/frame for the next 4 minutes (Fig. 1), because the first 10 minutes of the renogram are critical for the evaluation of glomerular filtration function. However, Nuclear Medicine physicians involved in the interpretation of the studies considered the quality of SPECT slices as adequate for drawing diagnostic information (Figs. 2 and 7).

Due to the need for increased tracer doses, we did not attempt to implement dSPECT(AC) in MAG-3 studies in children with obstructive uropathy. Novel CZT high-sensitivity gamma cameras may achieve this without additional radiation burden.

The Hawkey SPECT/CT system used in this study for dSPECT(AC) provided low-quality CT images, with low resolution and artifacts induced by respiratory and bowel motion during their slow acquisition procedure. Even though, the diagnostic information gained by those images was considered valuable. Nowadays, current SPECT/CT devices incorporate 6 to16-multiclice CT scanners, offering improved spatial resolution, at a cost of somewhat higher radiation doses to the patient.

Tc-99m DTPA dynamic SPECT renogram corrected for attenuation and fused with CT images is feasible with conventional gamma cameras incorporated into hybrid SPECT/CT systems. It allows accurate SRF measurement, is enriched by useful anatomical information and can be used for closer approximation of GFR compared to the conventional planar renogram.

ACKNOWLEDGEMENTS

No potential conflicts of interest were disclosed

Conflict of interest and funding

None

Blaufox MD, De Palma D, Taylor A, Szabo Z, Prigent A, Samal M, et al. The SNMMI and EANM practice guideline for renal scintigraphy in adults. Eur J Nucl Med Mol Imaging 2018;45:2218-28.
Taylor AT, Brandon DC, de Palma D, Blaufox MD, Durand E, Erbas B, et al. SNMMI Procedure Standard/EANM Practice Guideline for Diuretic Renal Scintigraphy in Adults With Suspected Upper Urinary Tract Obstruction 1.0. Semin Nucl Med 2018;48:377-90.
Gates GF: Measurement of glomerular filtration rate using technetium-99m-DTPA. J Nucl Med 1985;26:1348.
Taylor AT. Radionuclides in nephrourology, part 1: Radiopharmaceuticals, quality control, and quantitative indices. J Nucl Med 2014;55:608-15.
Keramida G, James JM, Prescott MC, Peters AM. Pitfalls and Limitations of Radionuclide Renal Imaging in Adults. Semin Nucl Med 2015;45:428-39.
Israel O, Pellet O, Biassoni L, De Palma D, Estrada-Lobato E, Gnanasegaran G, et al. Two decades of SPECT/CT - the coming of age of a technology: An updated review of literature evidence. Eur J Nucl Med Mol Imaging 2019;46(10):1990-2012.
Dickson JC, Armstrong IS, Gabiña PM, Denis-Bacelar AM, Krizsan AK, Gear JM, et al. EANM practice guideline for quantitative SPECT-CT. Eur J Nucl Med Mol Imaging 2023;50:980-995.
Humphries T, Celler A, Attariwala R. Preliminary evaluation of a quantitative regional dynamic SPECT renogram. J Nucl Med 2010;51(suppl 2):424.
Aibing JI, Jigang Yang, Yi Li. Added value of SPECT/CT for the evaluation of obstructive uropathy. J Nucl Med 2018;59 (suppl 1):376.
Miyazaki C, Harada H, Shuke N. Okizaki, Miura M, Hirano T. 99mTc-DTPA dynamic SPECT and CT volumetry for measuring split renal function in live kidney donors. Ann Nucl Med 2010;24:189-95.
Szlavecz A, Szabo B, Benyo B. Realistic kidney simulation for the development of renal function diagnostics for Dynamic SPECT imaging. IFAC PapersOnline 2021;54-15; 275-80.
Blaufox MD, Aurell M, Bubeck B, Fommei E, Piepsz A, Russell C, Taylor A, Thomsen HS, Volterrani D. Report of the Radionuclides in Nephrourology Committee on renal clearance. J Nucl Med 1996;37:1883-90.
Piepsz A, Kuyvenhoven JD, Tondeur M, Ham H. Normalized residual activity: usual values and robustness of the method. J Nucl Med 2002;43:33-8.
Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods 2012;9:671-5.
Farncombe T, Celler A, Noll D, Maeght J. Dynamic SPECT imaging using a single camera rotation (d SPECT). Nuclear Science, IEEE Transactions 1999;46:1055-61
Gullberg G, Reutter B, Sitek A, Maltz J, Bundinger T. Dynamic single photon emission computed tomography - basic principles and cardiac applications. Phys Med Biol 2010;55: R111–R191.
Humphries T, Cellar A, Trummer M. Effects of attenuation in single slow rotation dynamic SPECT. Phys Med Biol 2012;57:Ν253-65.
Mallet F, Poitrasson-Rivière A, Mariano-Goulart D, Agostini D, Manrique A. Measuring myocardial blood flow using dynamic myocardial perfusion SPECT: artifacts and pitfalls. J Nucl Cardiol 2023;30:2006-17.

Download PDF

Journal Publication

published 23 May, 2024

Read the published version in Nuclear Medicine Communications →

Version 1

posted

You are reading this latest preprint version

^99mTc-DTPA dynamic SPECT/CT renogram in adults. Feasibility and diagnostic benefit

Status:

Journal Publication

Version 1

Abstract

Objective

Methods

Results

Conclusions

Figures

INTRODUCTION

PATIENTS AND METHODS

1. Patients

2. Study protocol

2. Data processing

3. Statistical analysis

RESULTS

1. Image quality

2. Renographic parameters

3. Additional diagnostic information obtained by the CT scan

DISCUSSION

LIMITATIONS

CONCLUSION

DECLARATIONS

REFERENCES

Status:

Journal Publication

Version 1