Abstract | | |
To evaluate the accuracy of different single plasma sample methods (SPSM) ^{99m} Tc-DTPA clearances and to test whether the SPSM can replace the dual plasma samples method (DPSM) in the measurement of glomerular filtration rate (GFR), we studied 430 subjects counting renal patients and donors (240 male, 190 female; mean age 43.40 ± 16.30 years). All the subjects underwent dynamic renal scintigraphy after injection of ^{99m} Tc-DTPA; the GFR was calculated by seven SPSMs in addition to DPSM as a reference. Each of the SPSM clearance was compared with the DPSM measurement. There was a high correlation of all the SPSMs and the DPSM. The limits of agreement (95%) were found between the DPSM and all the SPSMs. Overall, the best method among the SPSMs, which is closest to the DPSM, is Fleming's single method as it has a statistically significant low mean difference (bias), low standard error, close mean ± SD to the reference method, good limits of agreement and high correlation co-efficient. This study concludes that, among the SPSMs, Fleming can reflect GFR more accurately than other methods, particularly when the expected serum creatinine is normal.
**How to cite this article:** Osman AO, Elmadani AE. Comparison of slope-intercept with single plasma sample methods in estimating glomerular filtration rate using radionuclides. Saudi J Kidney Dis Transpl 2014;25:321-5 |
**How to cite this URL:** Osman AO, Elmadani AE. Comparison of slope-intercept with single plasma sample methods in estimating glomerular filtration rate using radionuclides. Saudi J Kidney Dis Transpl [serial online] 2014 [cited 2020 Jan 21];25:321-5. Available from: http://www.sjkdt.org/text.asp?2014/25/2/321/128521 |
Introduction | | |
The glomerular filtration rate (GFR) is traditionally considered the best overall index of renal function in health and disease. ^{[1]} Because GFR is difficult to measure in clinical practice, most clinicians estimate the GFR from the serum creatinine concentration. However, the accuracy of this estimate is limited because the serum creatinine concentration is affected by factors other than creatinine filtration. ^{[2],[3]}
The purpose of this study was to evaluate the degree of accuracy and precision of a single plasma sample method (SPSM) versus a terminal slope-intercept (S-I) technique based on dual plasma samples (DPSM) in determining total ^{99m} Tc-DTPA plasma clearance and to test whether it can replace the S-I method in the measurement of GFR.
Materials and Methods | | |
A prospective study included donors and renal disease patients aged 18 years or more during 2008-2010. Patients with liver disease on diuretics, edema or extreme adiposity were excluded from the study. All blood samples were obtained after obtaining an informed consent.
All subjects underwent dynamic renal scan after injection of 111-185 MBq of ^{99m} Tc-DTPA. Following ^{99m} Tc-DTPA injection, venous blood samples (3 iriL) were collected from the contralateral arm at 60, 120 and 180 (for donors and other patients) or 240 min (for known cases of renal failure patients), blood samples were transferred to dry tubes, then centrifuged for 5 min, and 1 mL of each plasma sample was pipette in new tube and counted (in triple) in a well gamma counter for 1 min (cpm). Also, 1 mL of the standard solution (in duplicate) was counted in the same well counter for 1 min at the same time as plasma; decay correction was performed during the counting process of all samples.
*Slope-intercept method*
This is a standard method and is based on the determination of only the late exponential by means of at least two blood samples around 1-4 h ^{[12]} . The clearance can then be expressed as: Cl_{1} = D/A, where, D is the injected dose and A is the estimated area under the plasma curve. The area is obtained from the exponential fitting of the late plasma sample activities.
Area = Y_{0}/b, where, Y_{0} is the intercept of this late exponential at time 0 and b is the slope or rate constant of this exponential corrected for the first exponential.
*Measurement of plasma clearance by the SPSM*
Calculations and times were adopted according to the author's recommendations and the methods used were as follows:
- Christensen and Groth's method modified by Watson.
^{[8],[20]}
This method is designed for both children and adults; the GFR (mL/min) is given by:
Where, a = t (0.0000017t-0.0012), b = t (-0.000775t+1.31), t is the sampling time (180 min), c = ECV Ln (ECV/V _{t} ), ECV is the extracellular volume (in mL) = 8116.6 × BSA-28.2, V _{t} is the volume of distribution (in mL) at 180 min and BSA is the body surface area (in m ^{2}). - Constable's method.
^{[9]}
This method is designed for adults, and the GFR (mL/min) is given by:
GFR = 24.5 (V _{3} -6.2) ^{1/2} -67
Where, V _{3} is volume of distribution (in L) at 180 min, V _{3} = D/A _{t} , D is the injected dose and At is the plasma activity at time 180 min. - Dakubu's method.
^{[10]}
This method is designed for adults, and the GFR (mL/min/1.73 m ^{2}), is given by:
GFR = 95.33 LnV _{3} -270.99
Where, V _{3} is volume of distribution (in L) at 180 min. - Fleming's method.
^{[11]}
This method is designed for both adults and children; the GFR (mL/min) is given by the new general equation:
Where, BSA is the body surface area in square meters, t is the time of sample in minutes and V _{app} is the apparent volume of distribution in L/1.73 m ^{2}. - Groth and Aasted's method.
^{[4]}
This method is designed for children, and the GFR (mL/min/1.73 m ^{2}) is given by:
GFR = (0.213T-104) Ln (Y _{t} xA/Q _{0} )+1.88T-928
Where, T is the sampling time in minutes (180), Y _{t} is the activity counts of a 180-min sample (cpm/mL), A is the body surface area (m ^{2} ) and Q _{0} is the total injected dose counts (cpm). - Ham and Piepsz's method.
^{[17]}
This method is designed for children, and the GFR (mL/min) is given by:
CI (mL/min) = (2.602xV _{120} ) - 0.273
Where, V_{120} is the so-called "virtual distribution volume" (in L) at 120 min post-injection of the radiotracer, and it is the injected dose divided by the plasma concentration at 120 min (V _{120} = D/P). The final GFR result is corrected for body surface area. - Russell's method.
^{[18]}
This method is designed for both adults and children; GFR (in mL/min) is given by:
GFR = A Ln (D/P) + B
Where, A = -0.278T+119.1+2450/T, B = 2.886T-1222.9-16820/T, D is the total injected dose counts (cpm), P is plasma the activity (cpm/mL) and T is the sampling time (180 min).
Statistical Analysis | | |
All values were expressed as mean ± SD. Pair-wise comparison was performed by calculating the mean of differences (bias), standard error, limit of agreement and correlation coefficients (r).
Results | | |
Four hundred and thirty subjects were included in the study; of these, 52 (12%) were donors and 378 (88%) were patients covering a wide range of GFR values. The male to female ratio was 1.3:1. The mean age was 43.40 ± 16.30 years (age range: 18-88 years). The mean body surface area (BSA) (m ^{2} ) was 1.73 ± 0.22, and the BSA range was 1.23-2.47 m ^{2} .
For the assessment of the performance of the SPSM compared with the DPSM when considering the mean difference (bias), the limits of agreements and the standard error values (precision) of each SPSM, all were put into [Table 1] for simplicity and ease of comparison. The limits of agreement (95 %) ^{[5]} were found between the DPSM and all the SPSMs. | Table 1: Summary of the results analysis of all SPSM comparing each to DPSM.
**Click here to view** |
Overall, the best method among the SPSMs, which is closest to the DPSM, is the Fleming's single method as it has a statistically significant low mean difference (bias), low standard error, close mean ± SD to the reference method, good limits of agreement and high correlation coefficient.
Discussion | | |
The SPSM maintains consensus as the recommended first choice method for determining GFR, ^{[6]} and it can be used as an alternative to the DPSM. ^{[15],[16],[4]}
In the present study, the DPSM was used as the reference following a single injection of ^{99m} Tc-DTPA. While the S-I method was used with the DPSM, the distribution volume method was used in the SPSM, which correlated well with the DPSM
The results of the S-I with triple plasma samples measurement (TPSM) were compared with the S-I with the dual plasma samples in estimating the GFR, and it showed a very high correlation coefficient (0.998).
When the Fleming's SPSM was compared with the DPSM, the results of calculations showed a similar mean ± SD, a lower standard error of estimate (5.36 mL/min/1.73 m ^{2} ), small bias (-0.59) and a best limits of agreement (10-11 mL/min/1.73 m ^{2} ) compared with the reference of Fleming ^{[9]} (up to 10-15 mI7 min); a close correlation (0.96) was also noted.
There is slight underestimation for both very low and very high values of GFR, with a minimum occurring at intermediate values. Hence, the Fleming's method significantly correlated with the S-I method, particularly in the range of 40 to less 95 mL/min, but it was less precise to predict the GFR in patients with very low and very high clearance, and this agrees with previous studies. ^{[6],[13]}
In the present study, the Christensen and Groth SPSM, modified by Watson, was also compared with the DPSM, and the results showed a close correlation; however, there was a small degree of non-linearity apparent in both, low and high clearance values, which agree with the previous Itoh study. ^{[15]} This method is not applicable although it has relatively low bias and standard error when compared with the DPSM as the limit of agreement is comparatively high.
When the Groth and Aasted SPSM was compared with the DPSM, there was a close correlation but a significant difference in the mean ± SD values with relatively high limits of agreement (18 mL/min) and slightly increased above the acceptable limits. In this method, there were clear errors at low GFR values (below 45 mL/min/1.73 m ^{2} ) and at high GFR values (above 95 mL/min/1.73 m ^{2} ). Accordingly, it is not applicable as it has high limits of agreement around the mean difference (18 mL/min) compared with the clinically acceptable values.
In the present study, the Russell's SPSM was compared with the DSPM, and it showed a close correlation with a significant difference in the mean ± SD values, slightly high mean difference (6.62), relatively high standard error (7.46 mL/min/1.73 m ^{2} ) and high limits of agreement (21 mL/min), which is significantly more than the acceptable limits. Errors were obvious at GFR values below 60 mL/min/1.73 m ^{2} and at GFR values above 80 mL/min. Our results agree with the old study of Russell ^{[18]} that showed a standard error of 7.50 mL/min/ 1.73 m ^{2} . However, this method is invalid because it has large limits of agreement range, a high mean difference and a high standard error compared with the DPSM values.
The comparison of the Constable SPSM with the DPSM in our study showed a close correlation, but there was a significant difference in the mean ± SD values, with elevated mean difference (6.98), a high standard error (7.19 mL/min/1.73 m ^{2} ) and large limits of agreement (21 mL/min), which is significantly more than the acceptable reference limits. This method is illogical because it has a high mean difference, high standard error and a large limit of agreement with the DPSM measurements.
When the Dakubu's SPSM was compared with the DPSM, our results showed a close correlation, but there was a significant difference in the mean ± SD values, elevated bias (7.67), relatively high standard error (8.92 mL/min/1.73 m ^{2} ) and significantly high limits of agreement (25 mL/min). In addition, there were clear errors at GFR values below 45 mL/min and at GFR values above 75 mL/ min/1.73 m ^{2} . Thus, this method is unreasonable because it has a high mean difference, high standard error and huge limits of agreement compared with the DPSM values.
Finally, the Ham and Piepsz's SPSM was compared with the DPSM, and our results showed a relatively low correlation coefficient (R ^{2} = 0.84), relatively high mean difference (6.95), high standard error (8.37 mL/min/1.73 m ^{2} ) and large limits of agreement (23 mL/ min). The regression analysis curve disclosed a large degree of non-linearity, especially at GFR values below 60 mL/min/1.73 m ^{2} and at GFR values above 75 mL/min/1.73 m ^{2} . One should note that Ham's method ^{[7]} is only valid for children and young adults and in case of GFR more than 35 mL/min. Moreover, it has a large bias, high standard error and extra large limit of agreement range with the DPSM values.
In general, all the SPSM yielded a large mean difference in patients in whom the GFR values were 30 mL/min. The SPSM is more convenient than the DPSM, but in patients with advanced renal failure, the result of the SPSM becomes unreliable. ^{[19],[16],[18],[8]}
In summary, the Fleming's SPSM, among all other single methods used in the current study, had the highest correlation coefficient, the lowest mean difference (bias) and standard error, and the most acceptable limits of agreement when compared the DPSM measurements. Accordingly, the Fleming's SPSM could be the method of choice for the calculations of GFR, particularly when the serum creatinine level is normal and the sample is obtained at 3 h after the injection of the radiotracer.
References | | |
1. | Smith HW. Diseases of the kidney and urinary tract. In: The Kidney: Structure and Function in Health and Disease. New York: Oxford Univ Press; 1951. p. 836-87. |
2. | Levey AS. Measurement of renal function in chronic renal disease. Kidney Int 1990;38:167-84. [PUBMED] |
3. | Perrone RD, Madias NE, Levey AS. Serum creatinine as an index of renal function: New insights into old concepts. Clin Chem 1992;38: 1933-53. |
4. | Groth S, Aasted M. Calculation of 51Cr-EDTA clearance in children from the activity in one plasma sample by transformation of the bi-exponential plasma time-activity curve into a monoexponential with identical integral area below the time activity curve. Clin Physiol 1984;4:61-74. |
5. | Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10. [PUBMED] |
6. | Blaufox MD, Aurell M, Bubeck B, et al. Report of the radionuclides in nephrology committee on renal clearance. J Nucl Med 1996;37:1883-90. |
7. | Brochner-Mortensen J. Routine methods and their reliability for assessment of glomerular filtration rate in adults. Dan Med Bull 1978;25: 181-202. |
8. | Christensen AB, Groth S. Determination of ^{99m} Tc-DTPA clearance by a single plasma sample method. Clin Physiol 1986;6:579-88. [PUBMED] |
9. | Constable AR, Hussein MM, Albrecht MP, Joekes AM. Renal clearance determined from single plasma samples. In: Hollenbery NK, Lange S, editors. Radionuclides in nephrology. Amsterdam: Elsevier; 1978. p. 62-6. |
10. | Dakubu S, Adu D, Nkrumah KN, Anim-Addo Y, Belcher EH. Single blood sample estimation of glomerular filtration rate. Nucl Med Commun 1980;1:83-6. |
11. | Fleming JS, Persaud L, Zivanovic MA. A general equation for estimating glomerular filtration rate from a single plasma sample. Nucl Med Commun 2005;26:743-8. |
12. | Fleming JS, Zivanovic MA, Blake GM, Burniston M, Cosgriff PS; British Nuclear Medicine Society. Guidelines for the measurement of glomerular filtration rate using plasma sampling. Nucl Med Commun 2004;25:759-69. |
13. | Fleming JS, Persaud L, Ralls K, Burry S, Fleming PJ, Zivanovic M. Quality control of two sample GFR measurements using single sample estimates. Nucl Med Commun 2002; 23:389. |
14. | Itoh K, Tsushima S, Tsukamoto E, Tamaki N. Accuracy of plasma sample methods for determination of glomerular filtration rate with 99mTc-DTPA. Ann Nucl Med 2002;16:39-44. |
15. | Itoh K, Tsushima S, Tsukamoto E, Tamaki N. Reappraisal of single-sample and gamma camera methods for determination of the glomerular filtration rate with 99mTc-DTPA. Ann Nucl Med 2000;14:143-50. |
16. | Li Yi, Lee H, Blaufox MD. Single-sample methods to measure GFR with Technetium-99m-DTPA. J Nucl Med 1997;38:1290-5. |
17. | Piepsz A, Colarinha P, Gordon I, et al. Guidelines for glomerular filtration rate determination in children. Eur J Nucl Med 2001;28: BP31-6. |
18. | Russell CD, Bischoff PG, Kontzen FN, et al. Measurement of glomerular filtration rate: Single injection plasma clearance method without urine collection. J Nucl Med 1985;26: 1243-7. [PUBMED] |
19. | Waller DG, Keast CM, Fleming JS, Ackery DM. Measurement of glomerular filtration rate with technetium-99mDTPA: Comparison of plasma clearance techniques. J Nucl Med 1987;28:372-7. [PUBMED] |
20. | Watson WS. A simple method of estimating glomerular filtration rate. Eur J Nucl Med 1992;19:827. [PUBMED] |
**
**
**Correspondence Address**: Ahmed Elkhidir Elmadani National Cancer Institute, University of Gezira, P.O. Box 20, Wad Medani Sudan
| **Check** |
**DOI**: 10.4103/1319-2442.128521
**PMID:** 24625998
[Table 1] |