| Abstract|| |
Renal clearances of suitable clearance markers such as inulin and p-aminohippurate, determined before and after protein load, have been used for a long time to determine glomerular filtration rate (GFR), renal blood-flow, renal functional reserve as well as renal vascular resistance. Despite this long history, inconsistent results were achieved in determining renal functional reserve after protein ingestion or aminoacids infusion. Most authors found acute 'blunted increases' in GFR after dietary stresses, whereas others found acute decreases. In hypertensive patients showing hyperfiltration, a contradictory situation was observed; the traditionally observed 'blunted increase' in GFR and increase in the intraglomerular pressure were neither accompanied by a change in albumin excretion nor by an expected dynamic GFR response after administration of angiotensin converting enzyme (ACE)-inhibitors. We are of the opinion that the traditional clinical methods for assessing acutely provoked changes in renal function have been inappropriate and that an approach based on the two-compartment model of pharmacokinetics is more adequate for studying dynamic clearance changes within clinically feasible time horizons. Such test evaluations performed in patients within a washout period, and after long-term treatment with various drugs enable one to assess the efficacy of the therapeutic regimens especially in patients showing apparently normal or increased baseline clearances in hyper filtering kidneys.
Keywords: Dietary protein load, Dynamic renal function testing, Glomerular filtration rate, Effective renal plasma flow, Renal functional reserve.
|How to cite this article:|
Zitta S, Holzer H, Estelberger W. Assessment of Renal Functional Reserve by Dynamic Renal Function Testing. Saudi J Kidney Dis Transpl 2002;13:241-9
|How to cite this URL:|
Zitta S, Holzer H, Estelberger W. Assessment of Renal Functional Reserve by Dynamic Renal Function Testing. Saudi J Kidney Dis Transpl [serial online] 2002 [cited 2019 Dec 11];13:241-9. Available from: http://www.sjkdt.org/text.asp?2002/13/3/241/33112
| Conflicting data in the literature in relation to determination of renal functional reserve|| |
In early stages of renal impairment, hyperfiltration is frequently a disturbing condition disguising deteriorating renal function ,
and therefore, restricting the value of single clearance measurements for detecting renal vascular changes. By contrast, double determinations of renal clearances of inulin and p-amino-hippurate (PAH), determined before and after protein load, have been tried for a long time for assessing renal functional reserve and thus, acute renal hemodynamic changes, bearing additional information on renal vascular status. But these clinical studies have presented a rather ambiguous plethora of results. In healthy subjects, increases in glomerular filtration rate (GFR) after protein ingestion have been uniformly observed. However, in patients with essential hypertension or in renal transplant patients, most authors have only found 'blunted increases' in GFR after dietary stresses, whereas others have also found decreases in the same categories of patients. This contradictory situation in the literature has been exacerbated by studies in hypertensive patients where the traditionally observed 'blunted increase' in GFR was originally thought to be due to an increase in intra-glomerular pressure. However, the alleged increase in GFR in response to protein load was not accompanied by a change in albumin excretion, and there was lack of dynamic GFR response after administration of ACE-inhibitors.  A similar contradictory situation is found in the literature pertaining to renal functional reserve studies in renal transplant patients. 
| Traditional evaluation methods for determination of renal functional reserve|| |
Such inconsistent results arouse the suspicion that the traditional clinical methods for assessing acutely provoked changes in renal function have been inappropriate to date.  Evaluation of acute dynamic changes in kidney function by traditional steady state methods of GFR determination appears to be a mathematically too naive approach, despite its general acceptance as 'gold standard' for single clearance determination. ,,, By this technique, 'constant infusion' of exogenous markers and relating the urinary elimination rate to the corresponding plasma concentration level of an excreted marker were employed for clearance determinations. By their nature, however, these methods require equilibration of marker concentrations between the different compartments, and especially between marker influx and elimination. Since there is necessarily a generally unknown delay between the changes of plasma and urinary signals, constant infusion methods for calculation of renal clearance are correct only over a long experimental time horizon. However, in clinical experiments time horizons are necessarily limited for practical reasons; all the more so for the longer clinical protocols required for assessing experimentally induced changes in renal function. Thus, the earlier dynamic test methods, which described qualitatively normal increases in GFR after amino acid stimulation, even in hyperfiltering patients with early deterioration of renal function, probably were either not sensitive enough or perhaps inappropriate.
For studying experimentally induced acute changes in renal function, it is therefore necessary to employ a method for the description of time-dependent non-steady state processes which incorporates distribution effects on time-dependent marker profiles in addition to elimination effects within clinically tolerable time horizons.  A first mathematical approach to the description of non-equilibrium processes avoiding the stereotyped transference of the traditional renal clearance techniques is given by the superposition of exponential functions, a well-founded model technique described in the literature.  These so-called empirical models implicitly represent processes of marker distribution and elimination in one or more compartments. However, only the concentration profile in the so-called central compartment, i.e., the blood compartment, is studied after a single injection of a marker bolus. Furthermore, only situations with initial marker concentrations being zero, before marker application, are considered.
| Pharmacokinetic approach to dynamic renal function testing.|| |
In a more extended mathematical model for dynamic renal function testing, however, marker amounts remaining in the extracellular space from the first kinetic experiment have to be taken into account for evaluation in another experiment immediately following. In order to take into account non-zero initial marker concentrations as required for evaluation of consecutive kinetic experiments, a more flexible approach was employed by adapting the basic pharmacokinetic model , to experimental non-steady state data gained within feasible short experimental time horizons. The model describes the extracellular space as composed of two compartments: the well-perfused central volume and the less perfused peripheral volume with convective distribution between the two compartments and elimination solely from the central compartment. The system-constants of the model equations are identified in a search algorithm for the calculation of clearance and distribution parameters of an individual patient. The essence of the technique of pharmacokinetic system identification is given by the following model description.
The marker kinetics as given by the temporal courses of the marker amounts in the two compartments is the result of the infusion strategy, the exchange transports between the two compartments, and finally the renal elimination process.
The model can be formulated by a set of two simultaneous differential equations describing the rates of change of the marker amounts in the two respective compartments:
dx 1 /dt = f(t)-(k 01 +k 21 )x 1 +k 12 x 2 (Eq. 1)
dx 2 /dt = k 21 x 1 - k 12 x 2 (Eq. 2)
Eqs. 1 and 2 can be stated verbally in the following way: Firstly, the rate of change of the marker amount in the central compartment, dx1/dt, is determined by the input strategy chosen, the loss of marker from the central to the peripheral compartment, its gain by the central from the peripheral volume, and its elimination through the renal excretion mechanism. Secondly, the rate of change of the marker amount in the peripheral space, dx2/dt, is due to gain from and loss to the central pool. These transport processes are assumed to be proportional to the marker amounts momentarily contained in the respective distribution volumes. The input function of an experiment consisting of a bolus injection followed by constant infusion is given by equations (Eqs) 3 and 4:
f(t) = D/ τ, if 0 <<τ (Eq. 3)
f(t) = p, if τ < t c (Eq. 4)
The initial marker amounts are given by
x 1 (0) = c 1 (0)V 1 = x 10 (Eq. 5)
x 2 (0)=c 2 (0)V 2 =c 2 (0)V 1 (k 21 /k 12 )=x 20 (Eq. 6)
Fitting of the solution of the model defined by Eqs. 1 to 6 to the experimental plasma concentration data measured over a sufficiently long time horizon can be done by a method for the search of the minimum of a criterion of the sort:
E= ε(c 1 (t i ) - cexp(t i )) 2 , (i = 1…n) (Eq. 7)
f(t) is the input strategy as a function of time t,
x 1 is the amount of the marker in the central compartment,
x 2 is the amount of the marker in the peripheral compartment,
k 21 is the relative rate of transport from compartment 1 to 2,
k 12 is the relative rate of transport from compartment 2 to 1,
k 01 is the relative rate of elimination,
D is the priming dose,
τ is the injection duration,
ρ is the infusion rate,
T c is the duration of the constant-infusion experiment,
V 1 is the volume of the central compartment,
V 2 is the volume of the peripheral compartment,
E is the error function,
c 1 (t i ) is the theoretical plasma concentration values at times t i ,
c exp (t i ) is the experimental plasma concentration values at t i ,
n is the number of measurements,
n p is the no.of independent parameters (n p = 4).
A computer program implementing this mathematical pharmacokinetic technique will be offered via the Internet in the near future.
| Illustration of the use of dynamic renal function testing in essential hypertension.|| |
In order to illustrate the suitability of this technique of system identification applied to a two-compartment model for the quantitative assessment of the influence of protein-rich meals on renal function, we have shown that the dynamic test response observed in healthy controls and in patients with essential hypertension leads to results which can be understood both in pathophysiological and biochemical terms. The illustration given is part of a study published previously. 
For determination of GFR and effective renal plasma flow (ERPF), 2500 mg sinistrin (Inutest® , Fresenius Pharma Austria, Linz, Austria), an inulin-like poly fructosan, and PAH (Aminohippurate® , Merck & Co, West Point, PA, USA) in a dosage of 10 mg/kg body weight (minimal PAH dose 500 mg, maximal PAH dose 1000 mg) were given intravenously over three minutes. All clearance estimates were referred to 1.73 m 2 body surface area. Paired and unpaired Student's 2-tailed t-tests were used for comparison of group mean values. All means are given with their standard errors of the mean (S.E.M.).
The subjects comprised 15 healthy controls (4 male, 11 female, mean age 44.1 ± 2.3 years, mean arterial pressure (MAP) = 93.7 ± 3 mm Hg), and 16 hypertensive patients (5 male, 11 female, mean age 52.3 ± 1.6 years, MAP 112 ± 2 mm Hg, mean duration of hypertension 8.4 ± 3.1 years; data given as mean ± SEM). Eight hypertensive patients were randomly chosen for six months treatment with the β-adrenoceptor blocker carvedilol. The other eight patients were chosen for six months treatment with the ACE- inhibitor fosinopril. They were studied first after a two-week washout period without antihypertensive drugs, and again after six months treatment with either carvedilol or fosinopril. All subjects had normal serum creatinine levels, normal creatinine clearances, no proteinuria, and no history of any renal disease. The determination of renal functional reserve requires the stimulation of renal function by amino acid infusion or a protein rich meal.
We chose an artificial protein rich meal (1 g protein/kg body weight) as a practical way for protein supplementation in order to avoid the local pain and phlebitis found in preliminary experiments using peripheral intravenous administration of aminoacids. The time-span chosen for protein ingestion was motivated by previous studies in healthy humans showing that GFR changes from one constant level to another one about 90 minutes after ingestion of aminoacids. In those studies the new GFR remained constant for at least 3 hours.
Carvedilol was given in a dose of 25 to 50 mg per day. The dosage of fosinopril was 1 to 20 mg per day. Dosages of the drugs were then adjusted to achieve a desired MAP goal of 100 mm Hg.
[Figure - 1] shows the temporal concentration profile of sinistrin in a hypertensive patient during the initial dynamic study. It illustrates the studies and evaluations done in all tested subjects.
[Figure - 2] shows the mean arterial blood pressures in the normotensive controls, the hypertensive patients during the washout period, and the same hypertensive patients after antihypertensive treatment. There is a long-term decrease in blood pressure, but no acute change in MAP due to the dietary stimulus during the dynamic renal function studies.
The [Figure - 3] and [Figure - 4] serve to illustrate that the method of dynamic renal function testing allows one to monitor long-term pathophysiological and treatment dependent changes in renal functional responses to dietary stimuli.
Since dynamic processes are generally more appropriately characterized by fractional increases or decreases and, since GFR is determined more directly than ERPF which is only approximated by the clearance of PAH, [Figure - 3] summarizes and accentuates the findings by means of the relative change of GFR, i.e., the difference between the GFR before and after protein stimulation (∆GFR), divided by the GFR before protein stimulation. ∆GFR/GFR is referred to as renal functional reserve (RFR).
[Figure - 4] shows the renal vascular resistance (RVR) for the normotensive controls, the patients with essential hypertension, the patients treated with carvedilol and the patients treated with fosinopril. RVR was calculated by the formula of Gomez: RVR=[(MAP-10)/ERPF)] *60 *1322 *(1Hematocrit) (dyn/cm 2 )/(ml/sec).
Using the system identification method of adapting a two-compartment model to experimental concentration profiles as illustrated in [Figure - 1], we found increases of GFR in normal controls, but 'paradoxical' decreases of GFR in patients with essential hypertension by about 10% [Figure - 3]. These acute decreases in GFR following protein ingestion cannot be explained by systemic blood pressure effects, since there were no acute systemic changes in the MAP [Figure - 2]. Neither can these decreases in GFR be explained by acute mesangial contraction, since there were no concomitant decreases in ERPF.
Instead, the dynamic renal responses to protein ingestion can be understood in terms of a preferentially pre-glomerular vascular dysfunction in patients with essential hypertension. This situation resembles the one found in the so-called captopril test where vasodilation of the efferent arterioles is achieved by the administration of captopril in kidneys with stenosed arteries resulting in acutely decreased GFR and filtration fraction.  In a similar way, a consideration of the filtration process in relation to renal vessel resistances, differentially altered between the vasa afferentia and the vasa efferentia, should let one expect not only an absence of increases in GFR after protein ingestion, but even 'paradoxical' decreases in GFR concomitant with ERPFs remaining the same in dynamic tests. The interplay of the resistances of the vasa afferentia remaining high, and of the vasa efferentia when acutely reduced by a vasodilatory stimulus, leads to an acutely reduced GFR during preservation of the renal blood flow as observed in previous studies in patients suffering from diabetes mellitus.-  Since protein ingestion did not cause any acute reduction in systemic blood pressure [Figure - 2], it probably interferes with glomerular filtration by reducing efferent vascular resistance in kidneys with no significant change in reactivity of the afferent vascular resistance to the dietary signal.
Impaired reactivity of the afferent vascular resistance in patients suffering from long-term essential hypertension, as seen in [Figure - 3] and [Figure - 4], has been ascribed to decreased bioavailability of nitric oxide (NO) due to increased levels of reactive oxygen species.  Since L-arginine has been shown to have free radical scavenging properties  and to serve as a substrate for NO synthesis, L-arginine might increase the bioavailability of NO,  and therefore the dietary signal could be the L-arginine contained in the applied protein-rich meal.
In [Figure - 3] the negative value of ∆GFR/GFR in the untreated hypertensive patients, in contrast to the positive value of ∆GFR/GFR in the normotensive controls, indicates a pathological alteration in renal vascular reactivity.
[Figure - 4] shows that there is a significant acute decrease in RVR upon protein ingestion in the normotensive controls, whereas in the hypertensive patients during the washout period there is no acute change in RVR upon protein ingestion. However, while a tendency towards reconstitution of reactivity to protein stimulation was found in the patients treated with carvedilol, in the patients treated with fosinopril a tendency towards long-term lowering of the RVR was seen. This is in accordance with the finding in spontaneously hypertensive rats where fosinopril has been shown to reduce structural, but not functional alterations in small arteries by interfering with growth factors. 
The RVR determinations both supplement and corroborate the conclusions drawn from the determinations of renal functional reserves. Thus, the concept of decreased afferent reactivity and increased afferent resistance in the hypertensive patients is supported by the increased and unchanging RVR seen during the washout period. Carvedilol appears to restore afferent reactivity, while fosinopril appears to reduce afferent resistance.
The hypertensive patients studied had normal serum creatinine concentrations, normal creatinine clearance and no proteinuria. These patients revealed baseline GFRs in the normal range over the entire treatment period. Obviously single GFR determinations cannot detect early alterations in renal function and renal hemodynamics. These become visible only by evaluating the acute changes in GFR following a protein load.
In contrast to previous studies finding only 'blunted increases', the method presented in our study was sensitive enough to detect acute decreases in GFR. In determining PAH clearances, which are up to 5-times greater than inulin or sinistrin clearances, traditional constant infusion methods will attain qualitatively the same results as modern methods of system identification within the short time horizons needed for dynamic experiments.
The pharmacokinetic system identification method underlying the method of dynamic renal function testing enables one to determine not only the clearance estimates and the derived RFR and RVR values with the required precision, but also with the required error bounds for the system parameter estimates in individual patients ,,
Because of this, the method presented here allows one to assess RFRs and RVRs to monitor the acute renal vascular effects of vasoactive agents in vivo. The method should be useful in monitoring treatment of pathophysiological changes in early stages of alterations in renal vascular properties.
After a period of skepticism about the usefulness of dynamic renal function tests,  our study demonstrates that modern compartmental analysis of kinetic clearance experiments enables one to detect early stages of intrarenal hemodynamic alterations in the development of hypertension when there are pronounced disparities between pre- and post-glomerular vascular resistances, especially when hyperfiltration due to increased blood pressure may be present, suggesting a normal GFR despite a decreased RFR and increased RVR.
| Dynamic renal function testing in renal transplants patients|| |
Determination of renal functional reserve has also been tried extensively in renal transplant recipients. Renal transplant patients are at risk of renal damage caused by the nephrotoxic effects of immunosuppressants such as cyclosporine A (CsA). Again, conflicting test findings describing both increases and decreases in GFR following protein ingestion or amino-acid infusion have been given by different authors previously.  'Paradoxical' decreases in GFR following amino-acid ingestion in renal transplant recipients have also been found previously.  Cyclosporine therapy has been found to cause afferent arteriolar vasoconstriction. , The consequence of such a preferential afferent deterioration is an imbalance in pre- and postglomerular responses to vasodilatory stimuli. The consequential decrease in GFR after protein ingestion, and therefore the correct measure of RFR, are revealed reliably by the precise dynamic renal function tests reviewed here. ,
| References|| |
|1.||Zatz R, Brenner B. Pathogenesis of diabetic microangiopathy. The hemodynamic view. Am J Med 1986;80:443-53. |
|2.||Mackenzie HS, Brenner BM. Fewer nephrons at birth: a missing link in the etiology of essential hypertension. In: Avram MM, Klahr S, editors. Renal Disease Progression and Management. Philadelphia, WB Saunders 1996;12-9. |
|3.||Woods LL. Mechanisms of renal hemodynamic regulation in response to protein feeding. Kidney Int 1993; 44:659-75. [PUBMED] |
|4.||Fagugli RM, Buoncristiani U, Selvi A, et al. Reduction of renal functional reserve in kidney transplant recipients: a possible role of arachidonic acid metabolism alterations. Clin Nephrol 1998;49:349-55. [PUBMED] |
|5.||Buclin T, Pechere-Bertschi A, Sechaud R, et al. Sinistrin clearance for determination of glomerular filtration rate: a reappraisal of various approaches using a new analytical method. J Clin Pharmacol 1997; 37:679-92. |
|6.||Schuster VL, Seldin DW. Renal clearance. In: Seldin DW, Giebisch G, editors. The Kidney: Physiology and Pathophysiology. New York, Raven Press 1992;943-78. |
|7.||Van Acker BA, Koomen GC, Arisz L. Drawbacks of the constant-infusion technique for measurement of renal function. Am J Physiol 1995;268:F543-F52. [PUBMED] [FULLTEXT]|
|8.||Hellerstein S, Berenbom M, Alon U, Warady BA. The renal clearance and infusion clearance of insulin are similar, but not identical. Kidney Int 1993;44:1058-61. [PUBMED] |
|9.||Florijn KW, Barendregt JN, Lentjes EG, et al. Glomerular filtration rate measurement by 'single-shot' injection of inulin. Kidney Int 1994;46:252-9. [PUBMED] |
|10.||Norwich KH. Molecular dynamics in biosystems. The kinetics of tracers in intact organisms. Oxford, Pergamon Press 1977; 175-91. |
|11.||Valk6 P, Vajda S. Advanced scientific computing in BASIC with applications in chemistry, biology and pharmacology. Amsterdam, Elsevier 1989;161-73. |
|12.||Estelberger W, Petek W, Zitta S, et al. Determination of the glomerular filtration rate by identification of sinistrin kinetics. Eur J Clin Chem Clin Biochem 1995;33: 201-9. [PUBMED] |
|13.||Estelberger W, Zitta S, Lang T, et al. System identification of the low-dose kinetics of p-aminohippuric acid. Eur J Clin Chem Clin Biochem 1995;33:847-53. [PUBMED] |
|14.||Zitta S, Stoschitzky K, Zweiker R, et al. Dynamic renal function testing by compartmental analysis: assessment of renal functional reserve in essential hypertension. Nephrol Dial Transplant 2000;15:1162-9. [PUBMED] [FULLTEXT]|
|15.||Cuocolo A, Esposito S, Volpe M, Celentano L, Brunetti A, Salvatore M. Renal artery stenosis detection by combined Gates' technique and captopril test in hypertensive patients. J Nucl Med 1989;30:51-6. [PUBMED] [FULLTEXT]|
|16.||Bosch JP, Lew S, Glabman S, Lauer A. Renal hemodynamic changes in humans. Response in protein loading in normal and diseased kidneys. Am J Med 1986;81: 809-15. |
|17.||Gomez-Alamillo C, Sanchez-Casajus A, Sierra M, Huarte E, Diez J. Vasoconstriction of the afferent arteriole and defective renal synthesis of nitric oxide in essential hypertension. Kidney Int 1996;55: S129-S31 |
|18.||Wascher TC, Posch K, Wallner S, Hermetter A, Kostner GM, Graier WF. Vascular effects of L-arginine: anything beyond a substrate for the NO-synthase? Biochem Biophys Res Commun 1997;234:35-8. [PUBMED] [FULLTEXT]|
|19.||Yue TL, McKenna PJ, Gu JL, Cheng HY, Ruffulo RR, Feuerstein GZ. Carvedilol, a new antihypertensive agent, prevents lipid peroxidation and oxidative injury to endothelial cells. Hypertension 1993;22: 922-8. |
|20.||Rizzoni D, Castellano M, Porteri E, et al. Effects of low and high doses of fosinopril on the structure and function of resistance arteries. Hypertension 1995;26:118-23. [PUBMED] [FULLTEXT]|
|21.||Maschio G, Oldrizzi L, Rugiu C, De Biase V. Dynamic evaluation of renal function: a chimera for nephrologists? J Nephrol 1989; 2(3):157-64. |
|22.||Zhang X, Ghio L, Ardissino G, et al. Renal and metabolic effects of L-arginine infusion in kidney transplant recipients. Clin Nephrol 1999;52:37-43. |
|23.||Magalini SC, Nanni G, Agnes S, et al. Paradoxical effect of short-term protein loading on CsA-treated kidney transplant recipients. Transplant Proc 1989;21:1500-1. [PUBMED] |
|24.||English J, Evan A, Houghton D, Bennett WM. Cyclosporine-induced acute renal dysfunction in the rat. Evidence of arteriolar vasoconstriction with preservation of tubular function. Transplantation 1987;44:135-41. |
|25.||Myers BD, Newton L, Boshkos C, et al. Chronic injury of human renal microvessels with low-dose cyclosporine therapy. Transplantation 1988;46(1):694-703. |
|26.||Zitta S, Holzer H, Reibnegger G, Estelberger W. Dynamic renal function testing in renal transplant recipients. Clin Nephrol 2000(2);54:E15-E6. |
|27.||Zitta S, Auprich M, Holzer H, Reibnegger G. Cystatin C concentration and glomerular filtration rate. Lancet 2001;357:635 (letter). |
Department of Medicine, Division of Nephrology, University of Graz, Auenbruggerplatz 27, A-8036 Graz
[Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4]