| 2008 JASN IMPACT FACTOR 7.505 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pathophysiology of the Renal Biopsy |

* Department of Pathology, Beth Israel Deaconess Medical Center, and Harvard Medical School, Boston, Massachusetts; and
Department of Pathology and Renal Division, Department of Medicine, Beth Israel Deaconess Medical Center, and Harvard Medical School, Boston, Massachusetts
Correspondence: Dr. Isaac E. Stillman, Department of Pathology, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, MA 02215. Phone: 617-667-4344; Fax: 617-667-7120; E-mail: istillma{at}bidmc.harvard.edu
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
| CASE REPORT |
|---|
|
|
|---|
|
| RENAL INJURY OF ATN |
|---|
|
|
|---|
A more recent biopsy study confirmed the limited nature of the findings and identified two tubular lesions significantly less severe in recovering versus sustained ATN: single cell necrosis and brush border loss.3 The presence of nucleated cells within the vasa recta (possibly representing hematopoiesis or homing mesenchymal cells) has been regarded as a characteristic feature.4 Ultrastructural studies confirmed only limited tubular injury, finding little necrosis and observing discrete cell loss mostly affecting the proximal tubule (pars recta), collecting duct and the medullary thick ascending limb (mTAL).5 Biopsies of ATN in the transplant setting have shown similar findings; that is, very limited tubular necrosis, which often stands in sharp contrast to the marked and prolonged acute renal failure.6 Another example of structural and functional dyssynchrony is seen in some cases (Type I) of acute humoral rejection. Given the minimal histologic alterations, it is not surprising that they are considered "ATN like."7
In summary, the histopathologic findings of ATN are inconstant. When present, they are essentially limited to the tubulo-interstitium, and often subtle and mild. Indeed, recent proposals to update the nomenclature reflect this. The term acute kidney injury (AKI) has been suggested in place of acute renal failure, reflecting the spectrum of declining GFR short of total "failure." Similarly, given the variable degree of necrosis in ATN, the replacement term acute tubular "injury" may indeed be more appropriate.
Failure of Animal Models
Although human ATN has a multifactorial basis, for the purposes of research and conceptualization, etiological agents have been grouped into two major categories: ischemia and nephrotoxins.8 Animal models have been developed using both ischemia (primarily warm ischemia with reperfusion) and drugs. However, the insults used have been, for the most part, too extreme, as evidenced by the extensive necrosis typically produced. Those models therefore have at best very limited clinical, pathophysiologic, and morphologic relevance to the human situation.8
Indeed, many of the nephrotoxins which are directly tubulotoxic in animal studies (such as cisplatin, gentamicin, and cephaloridine) produce minimal histologic changes in the human.9 Furthermore, the distinction between ischemia and nephrotoxins is simplistic. Some nephrotoxins (cyclosporine and amphotericin) might be more accurately considered hypoxia-inducing agents, as they diminish renal perfusion, causing injury to zones (medullary ray and inner stripe) known to have limited oxygen availability.10 Furthermore, the hypoxia-mediated renal injury induced by some nephrotoxins (amphotericin, heme pigments, and possibly radiocontrast agents) is likely also associated with direct tubulotoxicity.
Most cases of ATN occur in patients with predisposing comorbidities, in sharp contrast to the healthy animals used in most models of ATN. Consequently, intense insults, such as prolonged ischemia or mega-doses of nephrotoxins, are required to achieve reproducible experimental renal failure, which in turn is associated with tubular pathology that is far more extensive than is seen in humans. Unfortunately, such animal studies have become the reference point for analysis and interpretation of human ATN. For example, the model of renal artery obstruction with subsequent reflow (warm ischemia reperfusion) produces extensive necrosis of proximal tubules, lesions far beyond what is typical for humans.11 Indeed, multiple clinical trials, based on successful interventions in rats, have failed in humans.9 Gentamicin is an example in which suprapharmacologic doses are necessary to cause tubular necrosis and renal failure in healthy animals.10 Its distinctive lesion (intracellular laminar myelin type whorls) shows no correlation with renal toxicity in humans. However, when combined with additional insults (such as ischemia, volume depletion, obstruction, or organ necrosis), animals do develop renal failure with clinically relevant doses.10 The combination of otherwise nontoxic doses of cisplatin and lipopolysaccharides results in severe renal failure and mortality in rats.12 Myoglobin per se is relatively nonrenotoxic unless combined with other insults (volume depletion or prostaglandin synthesis inhibition). Under those circumstances, renal retention of the tubulotoxic agent may increase, and casts may form, leading to intrarenal obstruction, another possible mechanism for ATN.13 Casts were inconspicuous in our biopsy, and the CPK was not markedly elevated, suggesting that this mechanism was not relevant in our case.
In summary, with the possible exception of calcineurin inhibitors, there is little correlation between the histology observed in many animal studies of ATN and the human situation.10
A New Approach
It seems clear that an appropriate model for ATN would be one of acute renal failure accompanied by limited changes by light microscopy, recapitulating the physiologic and pathologic dissociation. Indeed, recent studies in the isolated perfused rat kidney and in in vivo models of prostaglandin inhibitors and contrast nephropapthy seem to fit that goal, as in some protocols there is documented tissue hypoxia and adaptive responses in the setting of relatively intact tubular morphology. Upregulation of hypoxia inducible factors (HIF) with their multiple cell survival target genes have been localized to hypoxic cells. Experiments in isolated perfused kidneys14 and in vivo15 illustrate that failure to generate HIF and HIF-mediated responses is associated with lethal tubular injury. Rosenberger et al. have used high amplification HIF-1
immunohistochemistry to study human allograft ATN. Elevated HIF-1
expression was seen in immediate graft function, as opposed to the low levels noted in primary nonfunction.16 Although counterintuitive, these findings likely reflect higher renal oxygen tensions that are a function of diminished tubular oxygen demand secondary to reduced GFR. Taken together, these studies document acute renal failure with only limited morphologic alterations and suggest novel therapeutic approaches using drugs that upregulate HIF and target genes, such as erythropoietin.17 These strategies have been shown to both promote cell survival in isolated perfused kidneys18 and ameliorate renal failure after warm-ischemia reflow.19
Dynamic Changes in Intrarenal Microcirculation
Changes in renal microcirculation and oxygenation are likely to occur following use of prostaglandin inhibitors, radiocontrast administration, or strenuous exercise, such as marathon running. Compensatory mechanisms in healthy individuals maintain renal function. With increasing intensity of renal insults, and importantly, as they are multiplied, tubular injury develops, as has been documented by urinary detection of biomarkers such as cytokines (IL-6, IL-8, IL-18),20,21 NGAL,22 NAG, and KIM-1.23 Interestingly, IL-6 RNA was upregulated in renal allograft donor biopsies after cold ischemia and reperfusion and was correlated with longer cold ischemia times.24
Numerous studies over the past two decades have focused on the limited oxygen availability (pO2) of the medulla, in both animals and humans.25 This "hypoxic condition" is a function of both the renal microanatomy that enables urinary concentration and the energy demands of tubular reabsorption. Multiple mechanisms, such as prostaglandins, nitric oxide, adenosine enable the preservation of pO2 by balancing regional oxygen supply with transport activity. When the insults are limited, these adaptive responses preserve tubular function and integrity, as in prerenal failure. When they are overwhelmed, as in ATN, tubular dysfunction, as reflected by diminished urine concentrating capacity and sodium reabsorption, ensues.
Furthermore, activation of tubuloglomerular feedback may alter glomerular hemodynamics and reduce filtration rate to an extent out of proportion to tubulointerstitial injury, hence reducing tubular transport and improving medullary oxygenation, a process Thurau and Boylan cleverly called acute renal success.26 Recently developed noninvasive techniques may help document the evolution of these changes over time. BOLD (blood oxygen level detection) MRI studies based on the magnetic resonance of deoxyhemoglobin have demonstrated limited oxygen availability in the human medulla as well as the expected increase in pO2 when transport activity is inhibited by loop diuretics.27 BOLD MRI confirms diminished medullary pO2 in animal models of contrast nephropathy.28 Interestingly, renal sodium MRI can detect early loss of tubular function (diminished sodium concentration gradient) in this model when there is still very limited tubular injury.29
The recent introduction of extended sepsis models more analogous to the human situation (with the use of fluid resuscitation, antibiotics, and inotropes), and the finding of primary toll-like receptors for Gram-negative toxins, such as TLR4 and their effector proteins, such as MyD88, have also led to a deeper understanding of acute renal failure in sepsis.30 The minimal morphologic lesions present underscore their clinical relevance. Indeed, studies of autopsies in patients with sepsis have found apoptotic changes primarily limited to the gastrointestinal tract and the spleen.30 There seems to be agreement, certainly in animals and likely in humans, that both systemic and renal blood flow dynamics are severely altered in sepsis.30
Vasomotor Hypoxic Nephropathy
As we have seen, the pathophysiologic basis for acute renal failure likely relates to alterations in renal blood flow and oxygenation. In our case, acute renal failure was secondary to the action of cocaine, known to induce marked vasoconstriction.31 Blood flow, sufficient to prevent necrosis, but not enough to maintain GFR, is the likely explanation for the parenchymal preservation. This is likely accomplished by "corticomedullary blood redistribution," and tubuloglomerular feedback (diminishing filtration), leading to acute renal success, and preserving medullary pO2. Comparison of our case with another, showing the unusual finding of extensive tubular necrosis is useful in exploring their differing pathophysiologies.
Case 2 was a 71-yr-old man with a hospital course complicated by cardiopulmonary arrest with postresuscitation hemodynamic instability, respiratory compromise, and oliguric renal failure, requiring ventilation and dialysis until his death 13 d later. Postmortem examination of the kidneys revealed marked necrosis of proximal tubules with regenerative changes. Distal tubules appeared spared (Figure 1, B and C). This case parallels the warm ischemia reflow animal model (cardiac arrest followed by reperfusion) and indeed showed widespread tubular necrosis and regeneration at autopsy, findings almost never seen in clinical ATN.
Inasmuch as tubular necrosis is not a necessary prerequisite for ATN, what these two cases share is an important element of hemodynamic dysregulation. This spectrum of changes is detailed in Table 1. How then can we explain the difference in morphologic patterns? Parenchymal preservation results when intrarenal blood flow dysregulation is associated with sufficient pO2. Renal failure results when that flow does not maintain glomerular filtration and tubular function. However, in almost all cases, efferent arteriolar perfusion is sufficient to maintain tubular integrity, as demonstrated in our biopsy, and the reduction in GFR may serve to preserve tubular integrity by decreasing tubular transport work and oxygen consumption. In the contrasting autopsy case, there likely was a period of no renal perfusion, a scenario that is not typical of ATN. This extreme insult, as used in many animal models, led to a similar pattern of widespread tubular necrosis.
|
ATN may therefore be understood as a (mal)adaptive response of the kidney-trading away GFR for the preservation of medullary pO2 and tubular integrity. If so, the effect of therapeutic agents in ATN may be more of a function of their ability to restore normal intrarenal hemodynamics and tubular integrity. Indeed, restoration of GFR alone, resulting in increased tubular oxygen demand, may be counterproductive unless accompanied by reconstitution of the peritubular microcirculation/oxygenation.
Important questions remain as to how this aberration of flow develops and is maintained. Vascular tone is a function of endothelial and smooth muscle cell interaction. Endothelial cells release many vasoactive factors, some potent vasodilators (such as nitric oxide, prostaglandin E2, and prostacyclin) and others vasoconstrictors (such as endothelin, prostanoids, and components of the renin-angiotensin system). Some agents, such as endothelin, angiotensin-II, and adenosine, have diverse effects on cortical vasoconstriction and medullary microcirculation vasodilation, depending on the type and density of their various receptors. Reactive oxygen species signaling may be relevant to the control of vascular dynamics. Elaborated by the mTAL and governing HIF-1 activation, reactive oxygen species are crucially poised to regulate or dysregulate vascular function and integrity.32 Cytokines have also been suggested as early markers for ATN and are known to have a profound effect on vascular tone, both inducing the expression of cytokine receptors and other mediators of vasoconstriction as well the expression of enzymes such as inducible NO synthase and cyclooxygenase-2.33
In summary, the physiologic derangements of ATN are far more striking than the morphologic changes that accompany them. Noninvasive techniques and recently developed experimental models have begun to unravel this paradox.
| DISCLOSURES |
|---|
|
|
|---|
| Acknowledgments |
|---|
| Footnotes |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. Hammerschmidt, I. Loeffler, and G. Wolf Morg1 heterozygous mice are protected from acute renal ischemia-reperfusion injury Am J Physiol Renal Physiol, November 1, 2009; 297(5): F1273 - F1287. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Verghese, S. D. Ricardo, R. Weidenfeld, J. Zhuang, P. A. Hill, R. G. Langham, and J. A. Deane Renal Primary Cilia Lengthen after Acute Tubular Necrosis J. Am. Soc. Nephrol., October 1, 2009; 20(10): 2147 - 2153. [Abstract] [Full Text] [PDF] |
||||
![]() |
R John and A M Herzenberg Renal toxicity of therapeutic drugs J. Clin. Pathol., June 1, 2009; 62(6): 505 - 515. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Hall, R. J. Unwin, N. Parker, and M. R. Duchen Multiphoton Imaging Reveals Differences in Mitochondrial Function between Nephron Segments J. Am. Soc. Nephrol., June 1, 2009; 20(6): 1293 - 1302. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |