Homeostasis of Solute and Water by the Transplanted Kidney

This chapter analyzes the capacity of the transplanted kidney to transport solute and water. The intrinsic capacity of the transplanted kidney to regulate glomerular filtration rate (GFR), to exhibit T-G feedback are described along with alterations that flow from retrieval ischemia and surgical engraftment. Water homeostasis with respect to osmotic diuresis, tubular transport and tubular responses to hormone are described. Pathologic alterations of tubular transport are described including renal tubular acidosis potassium handling mineral metabolite handling, handling of magnesium and urate. Renal transport altered by standard immunosuppressive drug utilization is then described with particular reference to cyclosporin A, tacrolimus, and sirolimus.

Keywords

T-G feedback, ischemia, RTA, cyclosporin A, tacrolimus, electrolyte handling

Introduction

The transplanted kidney is a healthy organ responding to a multiplicity of complex and variant clinical experiences and insults. There is no single model to adequately explain the manner in which solutes are handled by the transplanted kidney or the physiologic derangements that might be encountered. There is an ordered sequence of transplant events that can be isolated and provide the impetus for unique alterations in function that can be delineated. The hemodynamic state of the donor prior to harvesting can adversely affect early function of the graft. Perioperative circumstances, including the effects of warm and cold ischemia on the donor kidney, produce functional abnormalities similar to those intentionally produced by investigators examining ischemic acute renal failure. Intraoperative factors including hypotension, blood loss, and technical damage establish conditions capable of deranging the function of the transplant. Additionally, the transplanted kidney, both early and several months after engraftment, is a denervated organ with all the consequences to tubular function attendant upon that state. The use of newer immunosuppressive agents, which directly alter physiologic function of the kidney, also affects the performance of the new graft.

The placement of a functioning allograft in a setting in which ischemic effects have been minimized and engraftment has been uncomplicated may result in the passage of salt and water during the early post-transplant period. The nature and amount of salt and water are a function of volume and osmolar loads imposed by the azotemic state, which presents excess solute loads and stimuli for the elaboration of saluretic hormones. The effect of other comorbid disorders associated with uremia, such as secondary hyperparathyroidism, which may persist post-transplant, can alter the handling of such solutes as calcium, phosphorus, hydrogen ion, and bicarbonate even when excretory function is excellent.

The maintenance of a functioning allograft depends on forestalling immunologically mediated assault. For all but identical twins, transplant recipients are maintained on immunosuppressive medications, some of which may have an impact on salt and water regulation. For example, many recipients receive corticosteroids early in the transplant period regardless of the additional immunosuppressive medicines during the maintenance regimen. These drugs have profound effects on sodium, chloride, potassium, and water regulation. Other immunosuppressive agents including calcineurin inhibitors and rapamycin may alter renal transplant tubular function. Since the early 1980s, calcineurin inhibitors have been the mainstay of maintenance immunosuppression in the allograft recipient. Both cyclosporine and tacrolimus have the ability to profoundly alter renal blood flow and thus renal function through reduced glomerular filtration with resultant effects on the handling of sodium, potassium hydrogen and uric acid. Rapamycin, one of the newer agents to be added to the armamentarium may also lead to tubular dysfunction and altered solute transport. On this background, immunologic assault in the form of acute and chronic rejection can profoundly alter the handling of salt and water by the transplanted kidney, possibly related to effector cells or their released cytokines.

This review, updated from its previous version, first explores the intrinsic capacity of the transplanted kidney to modulate solute and water, and to maintain the internal milieu in homeostatic balance. By using the classic observations of Bricker et al. and more recent studies of tubular function that minimize warm ischemia and avoid the confounding variables of immunosuppressive drugs and rejection, the handling of solutes and water by the denervated transplanted organ will be characterized.

The remainder of this review deals with the various derangements of solute and water handling observed clinically after successful engraftment. These separate syndromes will be explained employing clinical data and observations in the human. When experimental data from transplant or related fields of research derived from animal models are applicable, they will be discussed and placed in proper context in an attempt to explain the etiology of or the physiologic basis that underlies the clinical syndromes.

The effect of immunosuppressive therapies on renal function and metabolic derangements has been extensively studied. This revised chapter makes a special effort to explicate the impact of these potent agents, particularly that of calcineurin inhibitors and rapamycin. Those experiments that reveal the multiplicity of effects of these agents in general are reviewed in an attempt to suggest a unifying theory as to their physiologic consequences.

Intrinsic Capacity of the Denervated Transplanted Isograft

To observe the intrinsic capacity of the transplanted kidney to maintain solute and water homeostasis, an experiment must be designed to minimize the effects of ischemia related to harvest, the solute load that the azotemic patient presents to the new kidney, the use of immunosuppressive medications, and the effects of rejection. These criteria were best met by the classic renal autotransplantation studies reported by Bricker et al. in the dog. These researchers left one kidney intact, excised the contralateral kidney, and retransplanted it into a discreet hemibladder. In this two-kidney autotransplantation model in nonuremic animals, renal hemodynamics, maximal concentration and dilution capacity, response to vasopressin, sodium, and potassium excretion, response to modulation of plasma volume by hypertonic and isotonic infusion, and the response to mercurial diuretics were studied. When renal hemodynamics had returned to normal after surgery in the transplanted kidney, the measured tubular functions were virtually equal in the just transplanted kidney to those observed in the contralateral, undisturbed kidney. Sodium excretion rates factored for respective solute loads were also equivalent to the contralateral, undisturbed kidney as was the ability to maximally concentrate or dilute the urine after several days. This study illustrates the ability of the denervated autotransplanted kidney to maintain normal solute and water balance.

In healthy individuals, head-out water immersion increases sodium excretion. Interestingly, in renal denervated dogs, the natriuretic response to head-out water immersion is completely abolished, suggesting a major role for the renal nervous system for this response. Yet, despite renal denervation, it has been shown that patients with well-functioning renal transplants of less than two months duration are capable of a normal natriuretic response to head-out water immersion. This observation suggests that renal nerve activity is not essential for a normal natriuretic response to head-out water immersion in renal transplant recipients. When six renal transplant recipients were compared to, six normal individuals subjected to head-out water immersion, it was demonstrated that the denervated, transplanted kidneys of the six recipients were able to maintain a sodium excretory response to head-out water immersion identical to that obtained in the normal controls. These studies, taken together with the earlier reports, illustrate that the denervated autotransplanted kidney is able to maintain normal solute and water balance.

More precise experimental observations about renal tubular function in the transplanted kidney were made in the studies of Muller-Suur et al. and Norlen et al. These investigators used a rat model in which recipient animals underwent bilateral nephrectomy just prior to isotransplantation. Using standard micropuncture techniques to assess tubular function, Muller-Suur et al. were able to show that five minutes after reestablishment of circulation to the donor kidney that had suffered the least amount of cold ischemia, there was no statistical difference in the single-nephron glomerular filtration rate (GFR) of the transplant compared to the untransplanted normal kidney (13.4 nl/min per 100 g of body weight vs 14.1). After an initial polyuric phase related to solute load, a phenomenon discussed later, the passage of sodium and potassium into the urine was equivalent in the transplant to that of the normal, untransplanted kidney. Norlen et al. measured proximal tubule reabsorptive capacity using micropuncture techniques in a similar experimental rat model. They demonstrated that there was normal proximal tubular fluid reabsorption (TF/P inulin ratio of 2.1 [± 0.4]) as well as normal whole-kidney function and normal single-nephron GFR. These studies show that bulk solute handling by the denervated, isotransplanted kidney is normal on a whole-kidney clearance basis. These nonrejecting grafts can maintain a relatively undisturbed total GFR, single-nephron GFR, and proximal tubular absorptive capacity as determined by micropuncture.

The functional integrity of the more distal nephron segments as well as the intrinsic capacity of the transplanted kidney to exhibit tubuloglomerular (T-G) feedback was investigated further by Muller-Suur et al. in the rat isotransplant model. Two experimental paradigms were employed. In the first, the isograft was punctured in the presence of the normal contralateral kidney; in the second, the puncture occurred in the solitary isograft after contralateral nephrectomy. In contrast to the observations by Norlen et al. the investigators found that the whole-kidney GFR of the transplanted kidney fell approximately 40% when the contralateral kidney was left in place. Yet when the remnant kidney was removed, the GFR in the isotransplant rose toward but not entirely to the two-kidney control value. These animals were then studied 15 hours after isotransplantation at a time when single-nephron GFR had risen essentially to normal in order to measure the effects of T-G feedback. There was normal urine flow rate, normal sodium and potassium excretion, a normal TF/P inulin, and a slightly reduced GFR. When unilateral nephrectomy was performed, urine flow and sodium/potassium excretion rates increased in both the transplanted kidney and in the control. Interestingly, T-G feedback was found to be intact, permitting appropriate modulation of GFR in response to solute load by the transplanted kidney when the contralateral, normal kidney was in place. An increase in osmotic and solute load that would occur after the contralateral kidney was removed would be expected to have produced a substantial reduction in GFR if no other compensatory mechanisms were at play. In the actual experimental situation, the single remaining transplanted kidney demonstrated a decreased sensitivity of the T-G feedback mechanism, measured as a shift in the T-G set-point ( Fig. 94.1 ). This attenuation of the feedback sensitivity set-point allowed the GFR to rise toward normal rather than fall in the transplanted single kidney to maintain normal clearance despite the increased solute load reaching the macula densa. Muller-Surr et al. advanced the hypothesis that increased interstitial pressure and volumes to which the transplanted kidney is exposed after contralateral nephrectomy explains resetting of T-G feedback. In summary, the isotransplanted kidney, when studied with a functioning renal remnant, retains its capacity to modulate GFR to solute load, but resets the T-G feedback set-point to maintain GFR in the face of increased bulk blood flow when placed in an environment devoid of additional functioning nephron mass represented by a normal remnant kidney.

Kidney transplantation between identical twins provides an interesting clinical counterpart to the rat and dog isograft studies. Bricker et al. analyzed the intrinsic capacity of the human kidney allograft to maintain solute and water homeostasis in the original set of identical twins transplanted at the then Peter Bent Brigham Hospital (now Brigham & Women’s Hospital) in Boston in 1954, an advance in clinical medicine recognized for the Nobel Prize of 1990. Recipients of an identical twin kidney allograft require no immunosuppressive therapy and rarely exhibit rejection. When compared to the remaining kidney function of the identical donor sibling, the transplanted organ was able to maintain an equivalent filtration fraction and showed a normal ability to concentrate and dilute urine as well as to acidify and to alkalinize the urine when challenged. Further, the allograft could increase and decrease the rate of sodium excretion in the face of load, to respond to various volume- and drug-related stimuli, and to retain the capacity to alter filtration fraction and renal blood flow acutely while maintaining extracellular fluid volume and composition. One can conclude from isograft studies in animals and from the observations in identical twin human allografts that the transplanted denervated kidney retains the capacity to modulate solute and maintain volume homeostasis.

Ischemia Attendant to Harvest, Preservation, and Engraftment

Experimental models and clinical observation demonstrate that the transplanted kidney has the intrinsic capacity to maintain solute and water homeostasis. Unfortunately, the allograft in clinical settings is exposed to injurious environmental pressures to which it must respond by modifications of physiologic function. This section of the chapter will attempt to isolate several of the insults to which the functioning allograft is exposed and reveal the means by which the transplant has successfully responded to these challenges or delineate the derangements that flow from unsuccessful responses.

Although every attempt is made to minimize the ischemic effect of kidney transplantation by paying scrupulous attention to donor management, to conditions of harvest, organ preservation, and to conditions of engraftment, some degree of ischemic injury is a uniform phenomenon. Even in the early studies of Bricker et al. of autotransplantation in the dog, an initial drop in GFR and renal plasma flow that normalized several days after surgery was observed. This initial drop in GFR was felt to be due to tissue ischemia induced by the operative procedure and did not have any lasting effects on kidney function as revealed by their measurements.

Muller-Suur et al. also noted important consequences of early ischemia in their Munich-Wistar rat isograft model. They observed that even in the group of animals that received the least amount of protocol-determined cold ischemia prior to implantation, a group able to maintain normal single-nephron GFR, there still were increases in urine flow rate with diminished urinary concentration (approximately 100 mOsm/kg) during the first three hours postoperatively, thought a consequence of surgical ischemic injury. During this initial polyuric phase, there was an additional increase in sodium and potassium excretion paralleling the enhanced urine flow rate. An initial polyuria and the formation of hypotonic urine, even when GFR is normal, were felt to be a universal consequence of the unavoidable ischemia of the transplantation event.

These findings have been confirmed in reperfusion studies in renal allografts in humans. Alejandro et al., noted that in the ensuing hours of post-transplant, fractional excretion of rates of sodium were elevated to >20% with an impairment of urinary concentrating ability even in those with normofiltration. Although significant tubular dysfunction occurred, by morphometric analysis, only ~2% of proximal tubular cells exhibited necrosis and obstruction of the tubular lumen. Unlike studies in the native kidney and supporting the work of others, this points away from an important contribution of tubular obstruction to postischemic injury in the transplanted allograft.

In turn, it has been suggested that afferent arteriolar constriction with an attendant fall in the glomerular perfusion pressure is the primary mechanism by which glomerular filtration is affected in the freshly transplanted kidney. One contributing mechanism may be that the high fraction of filtered sodium reaching the macula densa mediates afferent vasoconstriction via tubulo-glomerular feedback. In patients with persistent hypofiltration post-transplantation, loss of polarity of proximal tubular cells with redistribution of cytoskeletal proteins including Na ⁺ /K ⁺ -ATPase, fodrin and ankyrin from the basolateral membrane to the cytoplasm has been noted with subsequent impairment of proximal tubular sodium reabsorption.

In addition to tubuloglomerular feedback, vascular mediators are felt to contribute to a state of afferent arteriolar vasoconstriction. With sustained injury and delayed function of the transplant, both elevation in plasma renin activity and endothelin-1 levels have been observed. In addition, there appears to be resistance of afferent arterioles to the vasodilator actions of endogenous atrial natruiretic peptide, which is present in excess. Thus, postischemic allograft injury appears to be accompanied by an imbalance that favors constrictor hormones over those that dilate the afferent arteriole with a resultant depression in GFR by lowering of glomerular perfusion pressure.

Urinary exosomes released into the urine by fusion of the outer membrane of multivesicular bodies with the apical plasma membrane of renal epithelial cells are shown to include membrane and cytosolic proteins, which have the characteristics of all renal tubule epithelial cells, podocytes, and transitional epithelial from the urinary collecting system. Importantly, several of these proteins found in urinary exosomes have been implicated in various kidney diseases and suggest that the examination of urinary endosomes may allow for a non-invasive measurement of site-specific biomarkers.

Several studies have demonstrated changes in excretion of urinary exosomal Na ⁺ /H ⁺ exchanger isoform 3 (NHE3) and fetuin-A after renal ischemia/reperfusion. NHE3, a sodium transporter protein in the proximal tubule, was present in the urine within 24 hours after renal allograft transplantation with levels returning to undetectable 48 hours after transplantation relating to a marked improvement in renal function. Conversely, NHE3 levels were not increased in the urine in the setting of acute allograft rejection. This data suggests that NHE3 may be a temporal marker of renal ischemia/reperfusion injury. Urinary exosomal fetuin-A levels have been shown to be increased after ischemia/reperfusion in allograft recipients but also in donor patients suggesting this may be a response to a change in renal hemodynamics and not specific to renal ischemia/perfusion injury.

The excretion of urinary exosomal aquaporin-1 has been studied both in animal models and in human allograft recipients. Aquaporin-1 is a water channel protein abundantly expressed in renal epithelial cells of the proximal tubules and the descending limb. It is known that renal ischemia/reperfusion in the native kidney leading to acute kidney injury is associated with decreased renal aquaporin-1 expression with subsequent data showing a decreased abundance of urinary exosomal aquaporin-1 in Sprague-Dawley rats subjected to unilateral renal ischemia/reperfusion as well as a renal transplant recipient at two and six days after renal transplantation despite a dramatic decrease in the plasma creatinine concentration following the operation. There was no decrease in urinary exosomal aquaporin-1 excretion in both animal models of nephrotic syndrome and patients with proteinuria. These data suggest that urinary exosomal aquaporin-1 may be a useful urinary biomarker to predict posttransplant acute kidney injury related to ischemia/reperfusion injury. Follow-up studies in larger sets of patients will be needed to confirm these findings and validate the utility of the role of urinary exosomal proteins in the detection of renal/ischemia reperfusion injury.

Norlen et al. used microsphere injection techniques to evaluate alterations in intrarenal blood flow during varying lengths of warm and cold ischemia of the renal transplant. Analysis of cross sections of rat kidney showed that microspheres injected during the reperfusion phase after two hours of cold ischemia showed no areas of preferential blood flow. Longer degrees of ischemia (12–16 hours) resulted in poor filling of the medulla. Except for the polyuric phase, which always seems to accompany such ischemia, there was no functional or structural abnormality after two hours of cold ischemia. On the other hand, after 12–16 hours of cold ischemia, there were dilated and collapsed tubules interspersed with morphologically “normal” tubules. The dilated tubules exhibited no glomerular filtration and were thought a consequence of obstruction by swollen epithelial cells in the corticomedullary region. This study provides anatomic correlates to the ischemic engraftment and preservation procedures.

The different studies in animals taken together show that short-term ischemia may derange volume homeostasis and concentration ability. Longer periods of ischemia may result in delayed reperfusion of the corticomedullary region, possibly due to cellular swelling in that area. The effect of the more prolonged ischemia to the transplant is a fall in GFR and an early loss of concentrating ability of the tubule. These alterations are potentially reversible spontaneously and may lead to no long-term sequelae. Clinically, the absence of urine formation after surgery suggests longer ischemic injury, whereas some element of the universally encountered polyuria is a consequence of shorter periods of ischemia.

Most data, however, would suggest that early ischemia/reperfusion injury of short duration does not lead to untoward long-term consequence to the allograft. Prolonged injury on the other hand, may contribute to late renal allograft deterioration and failure. To evaluate the association between initial ischemia/reperfusion injury occurring secondary to organ retrieval, storage, and transplantation and late renal allograft deterioration and failure, Azuma et al. studied the patterns of proteinuria, cellular infiltration, cytokine expression, and glomerular sclerosis over time in 344 Lewis and Fischer rats after 45 minutes of warm ischemia of a singe kidney. Intracellular adhesion molecule I, endothelin, and major histocompatibility complex (MHC) class II expression were found to be upregulated within two to five days after the injury, which enhances the antigen load presented to the recipient and the appropriate signal pathways to enhance antigen recognition, culminating in a drive for additional acute and later chronic rejection. In the animals, proteinuria developed after 8 weeks, and glomerulosclerosis, arterial obliteration, and interstitial fibrosis occurred after 16 weeks. These data suggest that early ischemia and reperfusion, if severe enough, may not only contribute to early dysfunction but also to late renal deterioration and chronic rejection. Further studies are needed for better understanding of the factors mediating these abnormalities.