Evidence for Activation of Tubuloglomerular Feedback Following Nephrectomy in Human Subjects Carol A. Pollock, MB, BS, and Michael J. Field, MD • The intrarenal regulation of glomerular filtration rate (GFR) following a reduction in renal mass was evaluated in six normal subjects, studied before and 4 to 6 weeks after undergoing unilateral nephrectomy, performed for the purpose of living-related transplantation. The role of the tubuloglomerular feedback (TGF) system in regulating the final GFR on both occasions was assessed by determining the increase in GFR over baseline levels following blockade of TGF by a single dose of bumetanide, care being taken to replace sodium and water losses. Before donor nephrectomy, baseline GFR was 115 ± 7 mL/min, which increased by 9.5% ± 2.5% to 126 ± 9 mL/min following the removal of TGF. Four to six weeks following donor nephrectomy, the baseline GFR was 83 ± 8 mL/min, confirming significant hyperfiltration in the remaining kidney. After administration of bumetanide, the GFR increased by 15.6% ± 3.1% to 96 ± 9 mL/min. This greater fractional increase (P < 0.025) indicates an increase in tonic TGF activity in the uninephrectomized state compared with that demonstrated before nephrectomy. The results suggest that the hyperfiltration response following unilateral nephrectomy is limited by an appropriately activated TGF system. This phenomenon may be of importance in the maintenance of Na and water homeostasis, as well as limiting the pathological consequences that may result from, or be exacerbated by, an excessively high filtration rate in the remaining nephrons. Thus, dietary or therapeutic maneuvers that impair the TGF response may be best avoided following a reduction in renal mass. © 1992 by the National Kidney Foundation, Inc. INDEX WORDS: Nephrectomy; tubuloglomerular feedback; glomerular filtration.
O
N THEORETICAL GROUNDS, there is reason to anticipate that the hyperfiltration state which prevails in the nephrons of the remaining kidney following nephrectomy may contribute to glomerular sclerosis and obliteration in the long term, with progressive proteinuria, hypertension, and a decrease in glomerular filtration rate (GFR).l However, this prediction of hemodynamically mediated injury, based largely on experimental work in rats, has received little clinical verification in studies reported to date, although the limited precision of physiological investigations makes definitive conclusions on this matter difficult. 2 In particular, little has been done to define the role of intrinsic renal mechanisms involved in GFR regulation in allowing the remaining kidney to adapt to its increased workload without undergoing the pathological consequences that might be expected to From the Department of Medicine, University of Sydney, Concord Hospital, New South Wales, Australia. Received February 12, 1992; accepted in revisedform June 16, 1992. Supported by a scholarship (to CA.P,) from the National Health and Medical Research Council of Australia, and by the Department of Veterans Affairs, Commonwealth of Australia. Address reprint requests to Carol Pollock, MB, BS, PhD, FRACP, Senior Lecturer in Medicine, Department of Medicine, Wallace Freeborn Building, Royal North Shore Hospital, NSW, 2065 Australia. © 1992 by the National Kidney Foundation, Inc. 0272-6386/92/2004-0005$3.00/0 348
result from an excessively high filtration rate in the remaining nephrons. Tubuloglomerular feedback (TGF) is a system that normally operates to control the GFR of individual nephrons in relation to the salt and water load delivered to the distal tubule,3 and thereby aids in the maintenance of extracellular fluid homeostasis. We have recently observed that TGF activity is increased in animal models of compensatory hypertrophy,4,5 and have proposed this to be a protective mechanism limiting hyperfiltration to the remaining nephron units. From this it might be surmised that manipulations which impair the TGF response may theoretically contribute to long-term deterioration of renal function, particularly in the presence of other factors that may themselves result in renal impairment. This study was therefore designed to test the hypothesis that enhanced TGF activity develops following a reduction in renal mass in man. Specifically, we wished to measure the GFR (as reflected in creatinine clearance) of normal human subjects before and 4 to 6 weeks after removal of one kidney, and to define and compare the contribution of intrarenal TGF to the control of GFR in the same subjects before and after nephrectomy. METHODS
Assessment ofTGF Activity Clearly, in the intact human subject, an indirect means for assessing the state of activation of TGF was required. The
American Journal of Kidney Diseases, Vol XX, No 4 (October), 1992: pp 348-353
349
TGF ACTIVATION AFTER NEPHRECTOMY IN MAN
technique we devised involved studying the effect of a single dose of the loop-acting diuretic bumetanide on the GFR in the donors before and after nephrectomy. This agent is suitable to define the extent of activation of TGF in human subjects, since it is known to bind specifically to the NajKjCI triple cotransporter in the thick ascending limb and macula densa cells within the nephron, and at the latter site acts to interrupt the afferent limb of the feedback response. 6 Thus, as long as Na and volume homeostasis are maintained during the study, the extent to which the GFR increases after acute administration of bumetan ide may be taken as a measure of the degree of preexisting feedback activation, which may be compared in the subjects in their preoperative and postnephrectomy state.
Experimental Protocol Four women and two men, aged 32 to 62 years, were included in the study. Each had been evaluated before this study by their attending physician as being suitable for donor nephrectomy for the purposes of transplantation. Thus, no significant other illnesses were present. Exclusion criteria from the study included a refusal to be studied on two occasions, or a known or suspected hypersensitivity to sulfonamide chemicals. One woman, aged 56 years, took a daily conjugated estrogen compound; otherwise, no regular medications were ingested by the patients before inclusion in the study. In one patient, a 55-year-old woman, an affective disorder became manifest subsequent to the kidney donation, requiring instigation of therapy with lithium carbonate. It was elected to continue with the second assessment in this patient as any tubular effects induced by the lithium are accounted for in the study design. Patients reported for assessment between 8 and lOAM after consuming a light breakfast, which was similar in each study for each patient. Baseline blood and urine samples were collected, after which a moderate water diuresis was initiated and maintained by the ingestion of 10 mLjkg water, followed by further volumes of5 mLjkg half hourly throughout the study. Blood and urine specimens were collected at 30-minute intervals. After a I-hour equilibration period, three consecutive 30-minute blood and urine specimens were obtained, following which the subjects were given a I-mg oral dose of burnetanide, and oral hydration continued. Six further half-hourly urine collections were made by spontaneous voiding, with blood samples obtained on each occasion. Immediate measurements of urine volume and Na concentration were made, which allowed for replacement of extra Na and volume losses induced by the diuretic agent above the baseline excretion levels. Such losses were replaced as extra volumes of normal saline in the 30-minute period subsequent to that in which the loss was measured. Therefore, this protocol maintained an approximately stable extracellular volume state in the prediuretic and postdiuretic periods. The subject was weighed before and at the conclusion of the study. For each blood sample, Na, K, urea, creatinine, hematocrit, and osmolality were determined, and urine was analyzed for volume, Na, K, urea, creatinine, and osmolality. In addition, three blood samples, taken at baseline, immediately before diuretic administration and at the conclusion of the study, were analyzed for renin and aldosterone as markers of the extracellular fluid volume status. From the above measurements, the GFR (as clearance of
endogenous creatinine), the absolute and fractional excretion rates of Na and K, and the free water clearance were determined. For each study, the degree of activation of TGF was estimated from the increase over baseline in GFR that occurred following administration of bumetanide. This "/i-GFR" result was compared for each subject with the analogous data obtained in the postnephrectomy phase of the study. The measured changes in endogenous creatinine clearance were regarded as a valid measure of changes in GFR during the study as each subject served as his or her own control, and renal function was not severely compromised in any case.? Statistical comparisons were made by paired Student's t test analysis. The protocol was assessed and approved by the Concord Repatriation General Hospital Ethics Committee, and informed consent was granted by each study participant.
RESULTS
Prior to donor nephrectomy, the mean baseline GFR was 115 ± 7 mL/min, which increased by 9.5% ± 2.5% to a mean of 126 ± 9 mL/min following the removal ofTGF by bumetanide. Four to six weeks following donor nephrectomy, the mean baseline GFR was 83 ± 8 mL/min, that is, 72% of the two-kidney GFR, confirming significant hyperfiltration in the remaining kidney. After administration of bumetanide, the mean GFR increased by 15.6% ± 3.1% to 96 ± 9 mL/ min (Fig 1), a significantly greater fractional increase than that demonstrated before nephrectomy (P < 0.025). This result indicates an increase in tonic TGF activity in the uninephrectomized state. The time course of the increase in creatinine clearance following removal of TGF is shown in Fig 2 in both the preoperative and postnephrectomy state. In both instances, the values are normalized for the mean creatinine clearance preadministration of diuretic. This demonstrates a Percent increase in GFR 35
30 25
%
20 15 10
Pre
Post
Nepllrectom~
Nephrectolll~
Fig 1. Percent increase in GFR following administration of bumetanide prenephrectomy and postnephrectomy in paired studies.
350
POLLOCK AND FIELD
13
12 12
11
%
110
10 10
*p< 0.05,
pre \'s post nephrectom~'
80
Post·diuretic
Buseline
Fig 2. Time course of the normalized creatinine clearance prenephrectomy (.) and postnephrectomy (0).
stable baseline GFR, which transiently decreases in both studies following the administration of bumetanide, presumably due to a lag phase in the replacement of salt and water. Subsequently, during the phase of quantitative salt and water replacement, an increased percentage elevation in the normalized GFR over baseline is observed in the postnephrectomy compared with the prenephrectomy state, which reaches statistical significance in the second and third postdiuretic clearance periods. The baseline urinary Na excretion was similar in the preoperative and postnephrectomy phases of study (0.11 ± 0.04 v 0.10 ± 0.02 mmol/min), and although the natriuresis in response to the administration of bumetanide was higher in the
two-kidney state, this difference failed to reach statistical significance (1.17 ± 0.08 v 0.91 ± 0.10 mmol/L; P = 0.07). The baseline fractional Na excretion showed considerable interindividual variation, but was similar before and following nephrectomy (0.72% ± 0.2% v 0.96% ± 0.4%; P = NS). Following the administration of bumetanide, the fractional excretion of Na increased significantly above baseline to a similar extent prenephrectomy and postnephrectomy (7.4% ± 1.5% v 7.9% ± 1.6%). Administration of bumetanide in the prenephrectomy study resulted in an increase in urine flow rate from 7.9 ± 1.9 to 18.4 ± 1.8 mL/min, whereas in the postnephrectomy study this effect was somewhat blunted, being 5.4 ± 1.2 to 14.0 ± 1.0 mL/min. Biochemical parameters, summarized in Table 1, demonstrate no change in serum electrolytes, fractional Na excretion, or free water clearance following nephrectomy, and a similar pattern of change in response to bumetanide. Specifically, serum potassium and urea decreased following bumetanide administration and, although free water clearance tended to increase, this was not significant. Following nephrectomy both urea and creatinine increased significantly, reflecting the 28% decrease in GFR. The hematocrit was lower following uninephrectomy, consistent with a degree of unreplaced blood loss at the time of the operation.
Table 1. Biochemical Parameters, Before and After Bumetanide Administration, Prenephrectomy Versus Postnephrectomy Prenephrectomy Baseline Creatinine clearance (mLjmin) Serum Na (mmoljL) Serum K (mmoIfL) Urea (mmoIfL) (mgjdL) Creatinine (/lmoljL) (mgjdL) Hematocrit (%) Osmolality (mOsmjkg) C H20 (mLjmin) FEN. (%)
115 138.3 4.03 4.4 (26.3 65 (0.75 45 280 4.5 0.72
± ± ± ± ± ± ± ± ± ± ±
7 0.9 0.17 0.3 1.7) 3 0.03) 1 4 1.8 0.2
* P < 0.05 v baseline prenephrectomy values. t P < 0.05 v baseline postnephrectomy value.
Postnephrectomy Bumetanide
Baseline
126 ± 9* 136.9 ± 1.0 3.64 ± 0.12* 4.0 ± 0.3* (23.9 ± 1.7) 62 ± 4 (0.70 ± 0.04) 44 ± 1 277 ± 5 7.8±1.4* 7.4 ± 1.5*
83 ± 8* 136.5 ± 1 4.3 ± 0.09 6.02 ± 0.53* (36.0 ± 3.2) 99 ± 7* (1.12 ± 0.08) 38.5 ± 2* 283 ± 4.8 2.5 ± 1.3* 0.96 ± 0.43
Bumetanide
96 ± 134.1 ± 3.94 ± 5.8 ± (34.7 ± 98 ± (1.11 ± 38.5 ± 274 ± 4.4 ± 7.9 ±
9*t 1.1* 0.14 0.59* 3.5) 6* 0.71) 2* 3* 0.3*t 1.6*t
351
TGF ACTIVATION AFTER NEPHRECTOMY IN MAN
No significant change was observed in body weight over the duration of the study, either before or after nephrectomy. The pattern of plasma renin and aldosterone response during the study is as shown in Table 2. No significant change occurred in either parameter, although both showed a final tendency to decrease, and no difference was observed between the two-kidney and the uninephrectomized state. It is likely that no significant alteration in volume status occurred during the studies in view of the similarities in body weight and the measured renin and aldosterone. DISCUSSION
The current study provides clinical evidence that unilateral nephrectomy in humans is followed by activation of TGF in the remaining kidney. This result, obtained by pharmacological blockade of the TGF system in normal subjects, is similar to that previously demonstrated by us in rats using direct micropuncture techniques. 4,5 It is proposed that this is an appropriate response, and serves to protect the remaining nephrons from injury, which may result from, or be exacerbated by, hyperfiltration or increases in glomerular capillary pressure in the solitary kidney. This protective response in limiting an otherwise uncontrolled elevation in GFR may explain in part the well-documented observation that, in the absence of other injurious factors, deterioration in kidney function is rarely a clinical problem following nephrectomy. 8· 10 Since reduction in renal mass is associated with increased GFR in the remaining nephrons, it might be expected that hemodynamic injury to the glomeruli may occur in the long term, as has been proposed in relation to progressive renal disease in generaL 11 It has usually been considered that the reason that this does not occur in humans is because uninephrectomy in itself may not reduce the renal mass to the critical level to trigger the mechanisms for the structural and functional abnormalities described in rat models with hyperfiltration. 10 We would propose that activation of TGF in the remaining nephrons plays a part in limiting the extent to which hyperfiltration develops in the solitary kidney. Indeed, in our own animal models of renal hypertrophy produced by both diabetes and ablation of renal tissue, TGF is found to be activated, resulting in a lower GFR
Table 2. Plasma Renin Activity and Aldosterone Prestudy, Preadministration of Bumetanide, and Poststudy in Both Prenephrectomy and Postnephrectomy States
Renin (ng/mL/h) Prenephrectomy Postnephrectomy Aldosterone (pg/mL) Prenephrectomy Postnephrectomy
Baseline
Diuretic Administration
Postdiuretic
0.19 ± 0.06 0.34 ± 0.10
0.15 ± 0.04 0.16 ± 0.04
0.09 ± 0.03 0.10 ± 0.05
259 ± 99 297 ± 69
372 ± 169 216 ± 70
154 ± 31 216 ± 88
than would otherwise be present. 4,5 When a more severe insult occurs, that is, either more severe diabetes or a greater extent of renal ablation, feedback activation is unable to prevent a more substantial increase in single-nephron GFR. This may explain the higher incidence of long-term renal damage in limited reports of patients with a solitary functioning kidney who have a further partial nephrectomy (25% to 75% loss of remaining renal mass). In these cases, hyperfiltrationinduced renal injury is evident in greater than 50% of cases. 12 It is acknowledged that recent reports indicate that hyperfiltration in itself may not be a sufficient insult to produce glomerular injury, 13 Nonetheless, in patients with any degree of preexisting renal impairment or with systemic hypertension, a reduction in renal mass and resultant hyperfiltration may result in overt expression of hemodynamically mediated renal damage. It is unlikely that the increase in TGF activity present in the baseline state postnephrectomy would occur as a function of the decrease in hematocrit, as this has not been shown to influence TGF. Thus, the present clinical study supports the view that an adaptive mechanism appropriately arising in a hyperfiltering kidney is activation of TGF, which serves to limit the increase in GFR and reduce the risk of structural damage. Findings consistent with this conclusion are available in previous studies assessing the renal response to a protein load in patients with unilateral nephrectomy. Tapson et al,14 in investigating the "functional renal reserve" in patients some 10 years following unilateral nephrectomy, demonstrated an increase of 22% in creatinine clearance following ingestion of a protein meaL An even more impressive increase in GFR of
POLLOCK AND FIELD
352
61.8% under similar conditions was observed by Manno et al,15 16 months following donor nephrectomy. Although not recognized as such by the investigators, it is likely that these effects were mediated in part by increases in TGF activity following nephrectomy, since it has been shown that high protein feeding suppresses TGF. 16 Thus, the marked increase in GFR following a protein meal in each of these studies suggests a high baseline activation of TGF in the subjects with reduced renal mass. The increase in TGF activity following nephrectomy may not be sustained at the same level over prolonged periods of time. While our patients were studied 4 to 6 weeks following nephrectomy, during which time the majority of compensatory hypertrophy occurs,9,1O,17 it may be that as time goes on TGF is less effective in controlling the hyperfiltration response. Indirect evidence for this comes from Regazzoni et al,18 who assessed the increase in GFR following protein loading in two groups of patients either less than 10 years, or more than 20 years following nephrectomy, and compared them with control subjects. Patients who had a nephrectomy less than 10 years previously had an increase in GFR of29%, whereas those in whom nephrectomy was greater than 20 years ago had no change or a decrease in GFR. This suggests that the ability to decrease GFR by TGF during hyperfiltration is lost after a period of time following nephrectomy, and indeed the time span identified for
loss ofTGF activity to become manifest is similar to that required to produce hyperfiltration injury in humans. Our findings of activated TGF following nephrectomy may have implications for the management of patients subjected to a reduction of renal mass in a variety of clinical circumstances. If the enhanced TGF is viewed as playing a protective role, as proposed above, maneuvers that blunt this response might best be avoided, particularly in those patients with more extensive renal ablation or comorbid illness such as systemic hypertension. The most important recommended measures would thus be limitation of dietary protein intake, for the reasons discussed, and avoidance of loop-acting diuretic drugs, which are known to interfere with the mediation of the feedback signal, 6 a principle used in the design of this study. In summary, the results of this study suggest that a compensatory increase in solitary kidney GFR following uninephrectomy in normal humans is associated with an appropriately activated TGF system that serves to limit the resultant level of hyperfiltration. We speculate that maneuvers acting to blunt this response might be deleterious in clinical conditions of reduced renal mass. ACKNOWLEDGMENT We are grateful to Drs P. Collett, G. Elder, J. Mahony, L. P. Roy and D. Tiller who allowed us to enroll their patients in this study.
REFERENCES 1. Hostetter TH, Olsen JL, Rennke HG, et al: Hyperfiltration in remnant nephrons: A potentially adverse response to renal ablation. Am J Physiol 241 :F85-F93, 1981 2. Fotino S: The solitary kidney: A model of chronic hyperfiltration in humans. Am J Kidney Dis 13:88-98, 1989 3. Schnermann J, Briggs 1: Function of the tubuloglomerular apparatus: Local control of glomerular hemodynamics, in Seldin DW, Giebisch G (eds): The Kidney: Physiology and Pathophysiology. New York, NY, Raven, 1992, pp 1249-1290 4. Pollock CA, Lawrence lR, Field MJ: Tubuloglomerular feedback in renal hypertrophy. Kidney Int 39:S32:SI06-109, 1991 5. Pollock CA, Bostrom TE, Dyne M, et al: Tubular sodium handling and tubuloglomerular feedback in compensatory renal hypertrophy. Pflugers Arch 420:159-166, 1992 6. Lorenz lN, Weihprecht H, Schnermann 1, et al: Renin release from isolated juxtaglomerular apparatus depends on macula densa chloride transport. Am 1 Physiol 260:F486F493,1991 7. Schuster VL, Seldin DW: Renal clearance, in Seldin DW,
Giebisch G (eds): The Kidney: Physiology and Pathophysiology. New York, NY, Raven, 1992, pp 943-978 8. MacKay EM, MacKay LL, Addis T: The degree of compensatory renal hypertrophy following unilateral nephrectomy I: The influence of age. 1 Exp Med 56:255-265, 1932 9. Kaufman 1M, DiMeola Hl, Siegel Nl, et al: Compensatory adaptation of structure and function following progressive renal ablation. Kidney Int 6: 10-17, 1974 10. Vincenti F, Amend WJC, Kaysen G, et al: Long term renal function in renal donors. Transplantation 36:626-629, 1983 II. Brenner BM, Meyer TW, Hostetter TH: Dietary protein intake and the progressive nature of kidney disease: The role of haemodynamically mediated glomerular injury in the pathogenesis of progressive glomerular sclerosis in aging, renal ablation and intrinsic renal disease. N Engl 1 Med 307:652659, 1982 12. Novick AC, Guz B, Steinmuller DR, et al: Long-term follow up after partial removal of a solitary kidney. N Engl J Med 325:1058-1062, 1991 13. Bidani AK, Mitchell KD, Schwartz MM, et al: Absence
TGF ACTIVATION AFTER NEPHRECTOMY IN MAN
of glomerular injury or nephron loss in a normotensive rat remnant kidney model. Kidney Int 38:28-38, 1990 14. Tapson JS, Mansy H, Marshall SM, et al: Renal functional reserve in kidney donors. Q J Med 60:725-732, 1986 15. Manno C, D'Elia F, Mingarelli M, et al: Effects of an acute protein load on urinary albumin excretion in kidney donors. Clin Nephrol 35:59-65, 1991 16. Seney FD, Persson AEG, Wright FS: Modification of
353 tubulogiomerular feedback signal by dietary protein. Am J Physiol 252:F83-F90, 1987 17. Katz A: Renal function immediately after unilateral nephrectomy: Relation to the mechanism of compensatory kidney growth. Yale J Bioi Med 43: 164-172, 1970 18. Regazzoni BM, Guignard JP, Arbus GS, et al: Delayed increase in filtration reserve after unilateral nephrectomy. Kidney Int 39: 1332, 1991
O
N THEORETICAL GROUNDS, there is reason to anticipate that the hyperfiltration state which prevails in the nephrons of the remaining kidney following nephrectomy may contribute to glomerular sclerosis and obliteration in the long term, with progressive proteinuria, hypertension, and a decrease in glomerular filtration rate (GFR).l However, this prediction of hemodynamically mediated injury, based largely on experimental work in rats, has received little clinical verification in studies reported to date, although the limited precision of physiological investigations makes definitive conclusions on this matter difficult. 2 In particular, little has been done to define the role of intrinsic renal mechanisms involved in GFR regulation in allowing the remaining kidney to adapt to its increased workload without undergoing the pathological consequences that might be expected to From the Department of Medicine, University of Sydney, Concord Hospital, New South Wales, Australia. Received February 12, 1992; accepted in revisedform June 16, 1992. Supported by a scholarship (to CA.P,) from the National Health and Medical Research Council of Australia, and by the Department of Veterans Affairs, Commonwealth of Australia. Address reprint requests to Carol Pollock, MB, BS, PhD, FRACP, Senior Lecturer in Medicine, Department of Medicine, Wallace Freeborn Building, Royal North Shore Hospital, NSW, 2065 Australia. © 1992 by the National Kidney Foundation, Inc. 0272-6386/92/2004-0005$3.00/0 348
result from an excessively high filtration rate in the remaining nephrons. Tubuloglomerular feedback (TGF) is a system that normally operates to control the GFR of individual nephrons in relation to the salt and water load delivered to the distal tubule,3 and thereby aids in the maintenance of extracellular fluid homeostasis. We have recently observed that TGF activity is increased in animal models of compensatory hypertrophy,4,5 and have proposed this to be a protective mechanism limiting hyperfiltration to the remaining nephron units. From this it might be surmised that manipulations which impair the TGF response may theoretically contribute to long-term deterioration of renal function, particularly in the presence of other factors that may themselves result in renal impairment. This study was therefore designed to test the hypothesis that enhanced TGF activity develops following a reduction in renal mass in man. Specifically, we wished to measure the GFR (as reflected in creatinine clearance) of normal human subjects before and 4 to 6 weeks after removal of one kidney, and to define and compare the contribution of intrarenal TGF to the control of GFR in the same subjects before and after nephrectomy. METHODS
Assessment ofTGF Activity Clearly, in the intact human subject, an indirect means for assessing the state of activation of TGF was required. The
American Journal of Kidney Diseases, Vol XX, No 4 (October), 1992: pp 348-353
349
TGF ACTIVATION AFTER NEPHRECTOMY IN MAN
technique we devised involved studying the effect of a single dose of the loop-acting diuretic bumetanide on the GFR in the donors before and after nephrectomy. This agent is suitable to define the extent of activation of TGF in human subjects, since it is known to bind specifically to the NajKjCI triple cotransporter in the thick ascending limb and macula densa cells within the nephron, and at the latter site acts to interrupt the afferent limb of the feedback response. 6 Thus, as long as Na and volume homeostasis are maintained during the study, the extent to which the GFR increases after acute administration of bumetan ide may be taken as a measure of the degree of preexisting feedback activation, which may be compared in the subjects in their preoperative and postnephrectomy state.
Experimental Protocol Four women and two men, aged 32 to 62 years, were included in the study. Each had been evaluated before this study by their attending physician as being suitable for donor nephrectomy for the purposes of transplantation. Thus, no significant other illnesses were present. Exclusion criteria from the study included a refusal to be studied on two occasions, or a known or suspected hypersensitivity to sulfonamide chemicals. One woman, aged 56 years, took a daily conjugated estrogen compound; otherwise, no regular medications were ingested by the patients before inclusion in the study. In one patient, a 55-year-old woman, an affective disorder became manifest subsequent to the kidney donation, requiring instigation of therapy with lithium carbonate. It was elected to continue with the second assessment in this patient as any tubular effects induced by the lithium are accounted for in the study design. Patients reported for assessment between 8 and lOAM after consuming a light breakfast, which was similar in each study for each patient. Baseline blood and urine samples were collected, after which a moderate water diuresis was initiated and maintained by the ingestion of 10 mLjkg water, followed by further volumes of5 mLjkg half hourly throughout the study. Blood and urine specimens were collected at 30-minute intervals. After a I-hour equilibration period, three consecutive 30-minute blood and urine specimens were obtained, following which the subjects were given a I-mg oral dose of burnetanide, and oral hydration continued. Six further half-hourly urine collections were made by spontaneous voiding, with blood samples obtained on each occasion. Immediate measurements of urine volume and Na concentration were made, which allowed for replacement of extra Na and volume losses induced by the diuretic agent above the baseline excretion levels. Such losses were replaced as extra volumes of normal saline in the 30-minute period subsequent to that in which the loss was measured. Therefore, this protocol maintained an approximately stable extracellular volume state in the prediuretic and postdiuretic periods. The subject was weighed before and at the conclusion of the study. For each blood sample, Na, K, urea, creatinine, hematocrit, and osmolality were determined, and urine was analyzed for volume, Na, K, urea, creatinine, and osmolality. In addition, three blood samples, taken at baseline, immediately before diuretic administration and at the conclusion of the study, were analyzed for renin and aldosterone as markers of the extracellular fluid volume status. From the above measurements, the GFR (as clearance of
endogenous creatinine), the absolute and fractional excretion rates of Na and K, and the free water clearance were determined. For each study, the degree of activation of TGF was estimated from the increase over baseline in GFR that occurred following administration of bumetanide. This "/i-GFR" result was compared for each subject with the analogous data obtained in the postnephrectomy phase of the study. The measured changes in endogenous creatinine clearance were regarded as a valid measure of changes in GFR during the study as each subject served as his or her own control, and renal function was not severely compromised in any case.? Statistical comparisons were made by paired Student's t test analysis. The protocol was assessed and approved by the Concord Repatriation General Hospital Ethics Committee, and informed consent was granted by each study participant.
RESULTS
Prior to donor nephrectomy, the mean baseline GFR was 115 ± 7 mL/min, which increased by 9.5% ± 2.5% to a mean of 126 ± 9 mL/min following the removal ofTGF by bumetanide. Four to six weeks following donor nephrectomy, the mean baseline GFR was 83 ± 8 mL/min, that is, 72% of the two-kidney GFR, confirming significant hyperfiltration in the remaining kidney. After administration of bumetanide, the mean GFR increased by 15.6% ± 3.1% to 96 ± 9 mL/ min (Fig 1), a significantly greater fractional increase than that demonstrated before nephrectomy (P < 0.025). This result indicates an increase in tonic TGF activity in the uninephrectomized state. The time course of the increase in creatinine clearance following removal of TGF is shown in Fig 2 in both the preoperative and postnephrectomy state. In both instances, the values are normalized for the mean creatinine clearance preadministration of diuretic. This demonstrates a Percent increase in GFR 35
30 25
%
20 15 10
Pre
Post
Nepllrectom~
Nephrectolll~
Fig 1. Percent increase in GFR following administration of bumetanide prenephrectomy and postnephrectomy in paired studies.
350
POLLOCK AND FIELD
13
12 12
11
%
110
10 10
*p< 0.05,
pre \'s post nephrectom~'
80
Post·diuretic
Buseline
Fig 2. Time course of the normalized creatinine clearance prenephrectomy (.) and postnephrectomy (0).
stable baseline GFR, which transiently decreases in both studies following the administration of bumetanide, presumably due to a lag phase in the replacement of salt and water. Subsequently, during the phase of quantitative salt and water replacement, an increased percentage elevation in the normalized GFR over baseline is observed in the postnephrectomy compared with the prenephrectomy state, which reaches statistical significance in the second and third postdiuretic clearance periods. The baseline urinary Na excretion was similar in the preoperative and postnephrectomy phases of study (0.11 ± 0.04 v 0.10 ± 0.02 mmol/min), and although the natriuresis in response to the administration of bumetanide was higher in the
two-kidney state, this difference failed to reach statistical significance (1.17 ± 0.08 v 0.91 ± 0.10 mmol/L; P = 0.07). The baseline fractional Na excretion showed considerable interindividual variation, but was similar before and following nephrectomy (0.72% ± 0.2% v 0.96% ± 0.4%; P = NS). Following the administration of bumetanide, the fractional excretion of Na increased significantly above baseline to a similar extent prenephrectomy and postnephrectomy (7.4% ± 1.5% v 7.9% ± 1.6%). Administration of bumetanide in the prenephrectomy study resulted in an increase in urine flow rate from 7.9 ± 1.9 to 18.4 ± 1.8 mL/min, whereas in the postnephrectomy study this effect was somewhat blunted, being 5.4 ± 1.2 to 14.0 ± 1.0 mL/min. Biochemical parameters, summarized in Table 1, demonstrate no change in serum electrolytes, fractional Na excretion, or free water clearance following nephrectomy, and a similar pattern of change in response to bumetanide. Specifically, serum potassium and urea decreased following bumetanide administration and, although free water clearance tended to increase, this was not significant. Following nephrectomy both urea and creatinine increased significantly, reflecting the 28% decrease in GFR. The hematocrit was lower following uninephrectomy, consistent with a degree of unreplaced blood loss at the time of the operation.
Table 1. Biochemical Parameters, Before and After Bumetanide Administration, Prenephrectomy Versus Postnephrectomy Prenephrectomy Baseline Creatinine clearance (mLjmin) Serum Na (mmoljL) Serum K (mmoIfL) Urea (mmoIfL) (mgjdL) Creatinine (/lmoljL) (mgjdL) Hematocrit (%) Osmolality (mOsmjkg) C H20 (mLjmin) FEN. (%)
115 138.3 4.03 4.4 (26.3 65 (0.75 45 280 4.5 0.72
± ± ± ± ± ± ± ± ± ± ±
7 0.9 0.17 0.3 1.7) 3 0.03) 1 4 1.8 0.2
* P < 0.05 v baseline prenephrectomy values. t P < 0.05 v baseline postnephrectomy value.
Postnephrectomy Bumetanide
Baseline
126 ± 9* 136.9 ± 1.0 3.64 ± 0.12* 4.0 ± 0.3* (23.9 ± 1.7) 62 ± 4 (0.70 ± 0.04) 44 ± 1 277 ± 5 7.8±1.4* 7.4 ± 1.5*
83 ± 8* 136.5 ± 1 4.3 ± 0.09 6.02 ± 0.53* (36.0 ± 3.2) 99 ± 7* (1.12 ± 0.08) 38.5 ± 2* 283 ± 4.8 2.5 ± 1.3* 0.96 ± 0.43
Bumetanide
96 ± 134.1 ± 3.94 ± 5.8 ± (34.7 ± 98 ± (1.11 ± 38.5 ± 274 ± 4.4 ± 7.9 ±
9*t 1.1* 0.14 0.59* 3.5) 6* 0.71) 2* 3* 0.3*t 1.6*t
351
TGF ACTIVATION AFTER NEPHRECTOMY IN MAN
No significant change was observed in body weight over the duration of the study, either before or after nephrectomy. The pattern of plasma renin and aldosterone response during the study is as shown in Table 2. No significant change occurred in either parameter, although both showed a final tendency to decrease, and no difference was observed between the two-kidney and the uninephrectomized state. It is likely that no significant alteration in volume status occurred during the studies in view of the similarities in body weight and the measured renin and aldosterone. DISCUSSION
The current study provides clinical evidence that unilateral nephrectomy in humans is followed by activation of TGF in the remaining kidney. This result, obtained by pharmacological blockade of the TGF system in normal subjects, is similar to that previously demonstrated by us in rats using direct micropuncture techniques. 4,5 It is proposed that this is an appropriate response, and serves to protect the remaining nephrons from injury, which may result from, or be exacerbated by, hyperfiltration or increases in glomerular capillary pressure in the solitary kidney. This protective response in limiting an otherwise uncontrolled elevation in GFR may explain in part the well-documented observation that, in the absence of other injurious factors, deterioration in kidney function is rarely a clinical problem following nephrectomy. 8· 10 Since reduction in renal mass is associated with increased GFR in the remaining nephrons, it might be expected that hemodynamic injury to the glomeruli may occur in the long term, as has been proposed in relation to progressive renal disease in generaL 11 It has usually been considered that the reason that this does not occur in humans is because uninephrectomy in itself may not reduce the renal mass to the critical level to trigger the mechanisms for the structural and functional abnormalities described in rat models with hyperfiltration. 10 We would propose that activation of TGF in the remaining nephrons plays a part in limiting the extent to which hyperfiltration develops in the solitary kidney. Indeed, in our own animal models of renal hypertrophy produced by both diabetes and ablation of renal tissue, TGF is found to be activated, resulting in a lower GFR
Table 2. Plasma Renin Activity and Aldosterone Prestudy, Preadministration of Bumetanide, and Poststudy in Both Prenephrectomy and Postnephrectomy States
Renin (ng/mL/h) Prenephrectomy Postnephrectomy Aldosterone (pg/mL) Prenephrectomy Postnephrectomy
Baseline
Diuretic Administration
Postdiuretic
0.19 ± 0.06 0.34 ± 0.10
0.15 ± 0.04 0.16 ± 0.04
0.09 ± 0.03 0.10 ± 0.05
259 ± 99 297 ± 69
372 ± 169 216 ± 70
154 ± 31 216 ± 88
than would otherwise be present. 4,5 When a more severe insult occurs, that is, either more severe diabetes or a greater extent of renal ablation, feedback activation is unable to prevent a more substantial increase in single-nephron GFR. This may explain the higher incidence of long-term renal damage in limited reports of patients with a solitary functioning kidney who have a further partial nephrectomy (25% to 75% loss of remaining renal mass). In these cases, hyperfiltrationinduced renal injury is evident in greater than 50% of cases. 12 It is acknowledged that recent reports indicate that hyperfiltration in itself may not be a sufficient insult to produce glomerular injury, 13 Nonetheless, in patients with any degree of preexisting renal impairment or with systemic hypertension, a reduction in renal mass and resultant hyperfiltration may result in overt expression of hemodynamically mediated renal damage. It is unlikely that the increase in TGF activity present in the baseline state postnephrectomy would occur as a function of the decrease in hematocrit, as this has not been shown to influence TGF. Thus, the present clinical study supports the view that an adaptive mechanism appropriately arising in a hyperfiltering kidney is activation of TGF, which serves to limit the increase in GFR and reduce the risk of structural damage. Findings consistent with this conclusion are available in previous studies assessing the renal response to a protein load in patients with unilateral nephrectomy. Tapson et al,14 in investigating the "functional renal reserve" in patients some 10 years following unilateral nephrectomy, demonstrated an increase of 22% in creatinine clearance following ingestion of a protein meaL An even more impressive increase in GFR of
POLLOCK AND FIELD
352
61.8% under similar conditions was observed by Manno et al,15 16 months following donor nephrectomy. Although not recognized as such by the investigators, it is likely that these effects were mediated in part by increases in TGF activity following nephrectomy, since it has been shown that high protein feeding suppresses TGF. 16 Thus, the marked increase in GFR following a protein meal in each of these studies suggests a high baseline activation of TGF in the subjects with reduced renal mass. The increase in TGF activity following nephrectomy may not be sustained at the same level over prolonged periods of time. While our patients were studied 4 to 6 weeks following nephrectomy, during which time the majority of compensatory hypertrophy occurs,9,1O,17 it may be that as time goes on TGF is less effective in controlling the hyperfiltration response. Indirect evidence for this comes from Regazzoni et al,18 who assessed the increase in GFR following protein loading in two groups of patients either less than 10 years, or more than 20 years following nephrectomy, and compared them with control subjects. Patients who had a nephrectomy less than 10 years previously had an increase in GFR of29%, whereas those in whom nephrectomy was greater than 20 years ago had no change or a decrease in GFR. This suggests that the ability to decrease GFR by TGF during hyperfiltration is lost after a period of time following nephrectomy, and indeed the time span identified for
loss ofTGF activity to become manifest is similar to that required to produce hyperfiltration injury in humans. Our findings of activated TGF following nephrectomy may have implications for the management of patients subjected to a reduction of renal mass in a variety of clinical circumstances. If the enhanced TGF is viewed as playing a protective role, as proposed above, maneuvers that blunt this response might best be avoided, particularly in those patients with more extensive renal ablation or comorbid illness such as systemic hypertension. The most important recommended measures would thus be limitation of dietary protein intake, for the reasons discussed, and avoidance of loop-acting diuretic drugs, which are known to interfere with the mediation of the feedback signal, 6 a principle used in the design of this study. In summary, the results of this study suggest that a compensatory increase in solitary kidney GFR following uninephrectomy in normal humans is associated with an appropriately activated TGF system that serves to limit the resultant level of hyperfiltration. We speculate that maneuvers acting to blunt this response might be deleterious in clinical conditions of reduced renal mass. ACKNOWLEDGMENT We are grateful to Drs P. Collett, G. Elder, J. Mahony, L. P. Roy and D. Tiller who allowed us to enroll their patients in this study.
REFERENCES 1. Hostetter TH, Olsen JL, Rennke HG, et al: Hyperfiltration in remnant nephrons: A potentially adverse response to renal ablation. Am J Physiol 241 :F85-F93, 1981 2. Fotino S: The solitary kidney: A model of chronic hyperfiltration in humans. Am J Kidney Dis 13:88-98, 1989 3. Schnermann J, Briggs 1: Function of the tubuloglomerular apparatus: Local control of glomerular hemodynamics, in Seldin DW, Giebisch G (eds): The Kidney: Physiology and Pathophysiology. New York, NY, Raven, 1992, pp 1249-1290 4. Pollock CA, Lawrence lR, Field MJ: Tubuloglomerular feedback in renal hypertrophy. Kidney Int 39:S32:SI06-109, 1991 5. Pollock CA, Bostrom TE, Dyne M, et al: Tubular sodium handling and tubuloglomerular feedback in compensatory renal hypertrophy. Pflugers Arch 420:159-166, 1992 6. Lorenz lN, Weihprecht H, Schnermann 1, et al: Renin release from isolated juxtaglomerular apparatus depends on macula densa chloride transport. Am 1 Physiol 260:F486F493,1991 7. Schuster VL, Seldin DW: Renal clearance, in Seldin DW,
Giebisch G (eds): The Kidney: Physiology and Pathophysiology. New York, NY, Raven, 1992, pp 943-978 8. MacKay EM, MacKay LL, Addis T: The degree of compensatory renal hypertrophy following unilateral nephrectomy I: The influence of age. 1 Exp Med 56:255-265, 1932 9. Kaufman 1M, DiMeola Hl, Siegel Nl, et al: Compensatory adaptation of structure and function following progressive renal ablation. Kidney Int 6: 10-17, 1974 10. Vincenti F, Amend WJC, Kaysen G, et al: Long term renal function in renal donors. Transplantation 36:626-629, 1983 II. Brenner BM, Meyer TW, Hostetter TH: Dietary protein intake and the progressive nature of kidney disease: The role of haemodynamically mediated glomerular injury in the pathogenesis of progressive glomerular sclerosis in aging, renal ablation and intrinsic renal disease. N Engl 1 Med 307:652659, 1982 12. Novick AC, Guz B, Steinmuller DR, et al: Long-term follow up after partial removal of a solitary kidney. N Engl J Med 325:1058-1062, 1991 13. Bidani AK, Mitchell KD, Schwartz MM, et al: Absence
TGF ACTIVATION AFTER NEPHRECTOMY IN MAN
of glomerular injury or nephron loss in a normotensive rat remnant kidney model. Kidney Int 38:28-38, 1990 14. Tapson JS, Mansy H, Marshall SM, et al: Renal functional reserve in kidney donors. Q J Med 60:725-732, 1986 15. Manno C, D'Elia F, Mingarelli M, et al: Effects of an acute protein load on urinary albumin excretion in kidney donors. Clin Nephrol 35:59-65, 1991 16. Seney FD, Persson AEG, Wright FS: Modification of
353 tubulogiomerular feedback signal by dietary protein. Am J Physiol 252:F83-F90, 1987 17. Katz A: Renal function immediately after unilateral nephrectomy: Relation to the mechanism of compensatory kidney growth. Yale J Bioi Med 43: 164-172, 1970 18. Regazzoni BM, Guignard JP, Arbus GS, et al: Delayed increase in filtration reserve after unilateral nephrectomy. Kidney Int 39: 1332, 1991
