Shockwave Lithotripsy Program Project

The Pathology of SWL Injury

Click here to see numbered references cited below:

SWL has been widely viewed as one of the most effective means of removing kidney stones since it was introduced in the early 1980's. In contrast, views on the safety of SWL have changed as clinical experience with it has increased. One of the first clinical reports on SWL expressed the view that shock waves "do not cause damage in passing through body tissue," (1) but less than a year later another reported that renal damage oc-curred in 63-85% of all SWL patients and that 30% of SWL patients experienced immediate post-SWL reduc-tions of effective renal plasma flow (2). Since then, there have been many reports of SWL-induced renal injury and impaired function in human patients and experimental animals (3,4). The more current view on SWL-in-duced renal injury is summed up in the lead sentence of a 1992 Journal of Urology editorial that referred to SWL as "a form of renal trauma," and went on to assert that "some degree of renal injury occurs with virtually every SWL treatment " (5).

Today, no one who utilizes SWL seriously disputes its ability to injure renal tissue. What is disputed, however, are the clinical consequences of that injury. The studies proposed in Project 1 aim at helping to resolve this dispute.

Characteristic of Morphological Lesion: SWL-induced renal injury consists of tubular, vascular and interstitial damage largely localized to the plane of the pressure wave (6, Progress Report). Human biopsy data show interstitial edema and hemorrhage extending from the capsule to the corticomedullary junction, with dis-ruption of renal corpuscles and degeneration of tubules in those areas.(6,7) There is clear evidence of endothelial dam-age, thrombus formation and initiation of inflammatory processes.

The most common gross evidence of damage to renal and surrounding tissue immediately after SWL in-clude 1) hematuria, (2-4,8,9) 2) kidney enlargement, (10-13) and 3) renal and perirenal hemorrhage( 4,10-12,14-32).

The most extensive morphological analysis of the effects of shock waves on renal tissue have come from studies in experimental animals. Histological examination of the kidney within the first 48 hours after SWL re-veals renal vascular and parenchymal injury at and near F2. The vascular injury is characterized by extensive endothelial cell damage in veins, small arteries and peritubular capillaries (7,12,14,19,23,25,27,29). In addi-tion, ruptured walls of veins and arteries appear within F2, resulting in extensive leakage of blood cells and fluid into the extracellular space. Hemorrhage associated with the renal capsule can result in formation of subcapsu-lar hematomas (10-12,14-31). The perirenal microvessels may also be ruptured. Thrombus formation in small veins and arteries is common at F2 (16,19,20,23,30,33). Intrarenal hemorrhage can be so extensive as to obscure parenchymal architecture and cause underestimation of the extent of tubular injury. Nevertheless, extensive tubular injury occurs with SWL, ranging from complete disruption of the normal segmentation of the nephron to focal tubular necrosis (10,12,17,18,23,25,28-30). Increased excretion of renal tubular enzymes provides further evidence of tubular damage (29,33-37).

Characteristics of Functional Lesion: Vascular injury constitutes the primary basis of the renal lesion caused by shock waves (3), and the potential consequences of this injury to renal function are enormous. Destruction or damage of the endothelium of the ma-jor renal blood vessels could impair mechanisms for the regulation of renal vascular tone and mediation of such processes as inflammatory cell trafficking and transmigration, healing, intravascular coagulation and transcapil-lary permeability. Shock waves could affect renal hemodynamics by interfering with the endothelial production of nitric oxide (38) or by stimulating the release of vasoactive substances such as endothelin-1 from endothelial cells (39,40), thromboxane and other prostaglandin's from platelets (41), angiotensin II produced from rennin released from renal tissue (42,43), norepinephrine from renal sympathetic nerves, etc.

The potential adverse consequences of renal vascular injury from SWL attracted the attention of several clinical investigators early in the history of SWL. They examined the effects of shock waves on renal hemodynamics (e.g., glomerular filtration rate [GFR] and renal plasma flow [RPF]) in patients. These studies made valuable contributions to the SWL literature, but they are difficult to assess in any systematic fashion because they involved different populations of patients, different lithotripters and different protocols. As a result, some studies reported no change in one or the other hemodynamic parameter (44,45), whereas others detected reductions in RPF (2,45,46) or GFR (47). There is no a priori reason to fault these findings; more systematic examination of the hemodynamic effects of SWL suggests that such things as patient demographics and timing of post-SWL measurements can affect studies of SWL-induced effects on renal hemodynamics.

Ours was the first group to systematically characterize the acute renal hemodynamic and tubular re-sponses to SWL in an animal model (48) and correlate them with SWL-induced injury to the renal parenchyma. One of our first findings was that a clinical dose of shock waves applied to one kidney reduced GFR in the shocked kidney, and reduced RPF in both kidneys, during the first 4 hours after SWL (both variables returned to pre-SWL values within 24 hours). (49, Progress Report) The reduction of blood flow in the shocked kidneys usu-ally exceeds that in the contralateral unshocked kidneys by 40-50%, and the bilateral reduction of blood flow has been repeatedly observed in our laboratory (see Progress Report). Since the reduction occurs in the contralateral kidneys with no detectable morphologic evidence of injury, we have proposed that it may be mediated by a circu-lating vasoactive agent(s) of humoral or neuronal origin. Elucidation of the mechanism(s) that causes SWL-in-duced reductions of renal blood flow is a principal aim of this renewal application.

Work from two laboratories supports and expands our hypothesis that renal vasoconstriction character-izes the response to SWL. One group measured renal blood velocity with color doppler sonography in human patients before, immediately after, and 1 week after a clinical dose of shock waves (50). They detected evidence of a tran-sient decrease in renal blood flow (elevated renal resistive index [RI]; i.e., blood velocity blood pressure) imme-diately after SWL. The RI returned to pre-SWL levels within 1 week. Another group measured the effect of SWL on RI (also with color doppler sonography) and arterial pressure in a large patient population stratified ac-cording to age. A clinical dose of shock waves (between 2000 and 3500 shocks) reduced renal blood flow (ele-vated RI) to the greatest degree in the oldest group of patients (>64 years). The 39% of these in whom RI re-mained high de-veloped hypertension (51).

Two aspects of these studies (50,51) are of particular interest and importance to the studies proposed in this application. The first is that the increases in renal vascular resistance occurred immediately after SWL, as have we (49,52,53, Progress Report). The second is that a large segment of the elderly human population may be at risk for development of hypertension after SWL (51).

Risk factors: The evaluation of potential risk factors for SWL (e.g., solitary kidneys, kidney size, renal infection, shock wave energy, etc.) comprised a substantial part of the experimental plan in our currently-funded PPG. The re-sults of these ongoing studies have focused our attention on the size of the lesion (as a fraction of total renal mass) as the major determinant of functional impairment, and forms the basis for the experi-ments to be performed in pursuit of Aim 1. Although much is currently known about the nature of the lesion caused by SWL and at least the acute phase of the functional impairment , a definitive corre-lation between the two remains speculative.

Risk factors may intensify the harm produced by SWL without a corresponding increase in benefit; but, by under-standing the risk, it might be possible to minimize it. The literature supports this notion in at least two ways. First, we and others have shown that calcium-channel antagonists prevent or diminish the hemody-namic (49) or enzymuric (54,55) signs of SWL-induced renal impairment. To the extent that such interventions might diminish the renal injury caused by SWL, the prospect such injury holds for subsequent adverse conse-quences may be di-minished or avoided. Second, it is now quite clear that a major acute effect of SWL is in-trarenal and subcapsu-lar hemorrhage. It is not clear, however, to what extent the size of the resulting lesion and subsequent inflam-mation leading to scar formation reflects direct injury caused by the shock waves. Neither is it clear to what ex-tent ischemia and other metabolic events caused by the accumulation of static pools of blood within the intersti-tium contribute to the size of the lesion and extent of the hemodynamic impairment. It therefore seems reason-able to propose (Aim 1) that a relationship may exist between the intrarenal hemorrhage and the size of the re-sulting lesion. The relationship will be examined with the aid of experimental interventions that either increase or decrease intrarenal hemorrhage after SWL (see D. Experimental Design and Methods of Procedure).

One aspect of the renal response to SWL clearly involves reduced renal blood flow. While we and others gen-erally find this reduction to be transient, this cannot be taken to mean that blood flow is not impaired in localized ar-eas within the kidney, such as within the SWL-induced lesion. Indeed, our preliminary measurements of cortical intrarenal blood flow after SWL and those of Kataoka et al. (50) who used color doppler sonography, indicate that blood flow may be diminished in discrete regions within the kidney (our data place the site of the reduced flow at or near the SWL-induced lesion). Moreover, sustained, generalized reduc-tions of renal blood flow may persist indefinitely in older patients after SWL (51). Since urinary endothelin-1 ex-cretion in-creases after SWL in children (56) and pigs , and may play a role in renal vasocon-striction or increases in systemic arterial pressure (40), further study of its participation in the renal response to SWL seems urgent. All of this evidence provides an irrefutable rationale for gaining an understanding of what causes SWL-induced reductions of renal blood flow. Aim 3 of the proposal targets four of the most likely media-tors of renal vasoconstriction in association with SWL-induced injury (see D. Experimental Design and Methods of Pro-cedure).

Outcome of Renal Fibrogenesis: An important finding of our recent studies on the pig is the identification of numerous inflammatory cells, primar-ily platelets and leukocytes lining the injured vessel walls and infiltrating surrounding damaged tubules and in-terstitial cells immediately after SWL that we have termed "lithotripsy nephritis" which progresses to a prominent region of scar at the site of the acute injury 3 months post-SWL. These data suggest that the injury induced by SWL triggers an acute inflammation that induces events that direct the repair process to fibrogenesis resulting in a scar. Other investigators (34) have noted platelets adhering to exposed suben-dothelial connective tissue of a damaged artery as well as large aggregates of platelets, fibrin and leukocytes (neutrophil granulocytes) around damaged tubules and interstitial tissues of the dog kidney 5 hours post-SWL. Banner et al. (57) have described progressive deposits of complement 3 and traces of immunoglobulin G (IgG) in glomeruli of treated pig kidneys regardless of the lithotripter used. A larger number of studies have examined the site of shock wave treatment from two weeks to 3-6 months post-SWL to determine the outcome of the re-parative process (4,14-18,20,23,24,25,26). While a few studies reported no chronic changes at F2 (24), those investigations that treated the kidney with a level of shock wave en-ergy suf-ficient to clinically comminute a renal stone found striking evidence of focal interstitial fibrosis, focal glomeru-lonephritis to chronic tubulointerstitial disease within one to two months of treatment (see our review; Evan and McAteer, 58). In fact, as a result of the data collected by Karalezli et al, (17) in the rabbit shock wave lithotripsy is listed in one of the leading nephrol-ogy texts as a condition known to result in chronic interstitial nephritis (59). There are only a handful of reports detailing the histological changes that occur in the human kidney after SWL but two of those studies have documented chronic changes in human kidneys treated with SWL. LeChevallier et al. (60) imaged the treated kidney by single photon emission computer tomography (SPECT) before and 30 days after treatment, They noted regions of scar in 7 of 12 SWL patient studied and concluded that SWL cannot be considered a nontrau-matic procedure. Umekawa et al. (61) described evidence of anti-glomerular basement membrane (GBM) anti-body-mediated glomerulonephritis in a SWL treated human kidney that progressed to acute renal failure in 90 days post treatment. While the animal and human studies cited above all strongly sug-gest that events initiated in the acute lesion lead to scar formation, there seems to be several different outcomes , i.e., glomerulosclerosis and/or tubulointersitial nephritis. To date all SWL studies (including ours) that have reported chronic changes in the kidney are descriptive and have not attempted to determine the mechanism for this process. Once we un-derstand the mechanisms for the fibrogenesis induced by SWL, it would be our next goal to be able to intervene in this process so as to control the amount of scarring and thereby, possible prevent the long-term loss of renal function or the development of hypertension. Such ap-proaches are well on their way for other forms of renal in-jury (62,63).

The renal literature clearly shows the scarring occurs with any number of different primary insults, whether meta-bolic, immune, degenerative or traumatic (64). This process, whether of glomerular or of tubuloin-terstitial origin, can proceed to progressive renal damage (65) and particularly, in the case of tubulointerstitial sclerosis correlates closely with a fall in renal function detected as a rise in serum creatinine levels (66). Bohle et al. (67) have ex-tended this observation by showing a strong correlation between a rise in serum creatinine lev-els and a reduction in number and area of postglomerular capillaries. This condition would also result in a rise in meas-ured in-trarenal resistive index which correlates well with the on set of hypertension (51). The main fea-ture of glomeru-lar sclerosis is the accumulation of extracellular matrix in the glomerulus and eventual collapse of capil-lary walls resulting in a loss of glomerular function. Tubulointerstitial scarring is characterized by tubular atro-phy and an accumulation of interstitial matrix, which may be produced by infiltrating fibroblasts, renal interstitial fi-broblasts or tubular cells. As the interstitial space continues to enlarge due to the increase in matrix material, more glomeruli or tubules/microvessels are entrapped causing further tissue injury resulting in the release of more cytokines that stimulate more scar formation and an additional loss of functional tissue. Thus, a vicious cycle may be triggered that eventually ends in extensive kidney damage. This is a very different process than what happens in normal wound healing where there is controlled degradation of any extra matrix material after induction and matrix deposition.

Border and Ruoslahti (68) have described inflammation, immune response and tissue repair as normal pro-tec-tive mechanisms for the kidney as it functions in a hostile and dangerous environment. However, these responses may end up harming the organ it is attempting to protect by inducing an aggressive fibrogenic response that ends up causing further tissue damage. Stated another way, the intensity of an inflammatory response is tightly regulated and may resolve completely, or persist and eventually destroy the kidney by progressive scarring. This process has been seen in a variety of glomerular diseases (69,70), diabetic nephropathy (71), and chronic pyelonephritis (72). In SWL no one has studied the fibrogenic process so we do not know if there is a relationship between level of injury and degree of fibrogenesis. It may be that at low "dose" levels of SWL little to no fibrogenesis results which would explain some of the literature. However, at more clinical "dose" levels, the true renal fibrogenic process may be started which is characterized by continued synthesis of extracellular matrix.

To understand how to study this process of fibrogenesis in the SWL treated kidney, one needs to identify the stages (phases) of renal fibrogenesis (73). The phases are 1) induction, 2) inflammatory matrix synthesis and 3) postinflammatory matrix synthesis. In the first stage inflammatory cells (platelets, PMNs, monocytes/macrophages and lymphocytes) infiltrate the site of injury at certain times after injury. These cell induce a release of profibrogenic cytokines. Thus Aim 2 will study the sequencing and kinetics of infiltrating cells. In the next stage there is a continued release of profibrogenic cytokine resulting in increased matrix synthesis and deposition but also an inhibition of matrix degradation. In the last phase there is cessation of the inflammatory stimulus but continued matrix synthesis and deposition.. The cytokines involved in the accumulation of inflammatory cells and matrix synthesis include interleukins (IL-1, 2, 6, & 8), interferon (IFN-a , IFN-b, IFN-g) PDGF-A and PDGF-B, tumor necrosis factor (TNF-a and b); FGF, EGF, IGF-1, TGF-a and b (74). Following renal injury a variety of other genes are inducted for proteins like transcription factors (eg, c-fos), chemoattractants (MCP-1 & MIP-1), protease's (eg, type I and IV collagenase), vasoactive peptides (endothelin, rennin, NOS) and many other enzymes and proteins of unknown function. Thus Aim 2 will determine which factors are induced by SWL and their influence on renal scarring post-SWL.

All or part of these factors may take part of fibrogenesis in SWL. There are too many factors that could be studied so we selected the most likely candidates to be involved. These factors have been divided into two groups, those factors known to be involved in renal fibrogenesis and those physiological factors that could alter renal hemodynamics. The fibrogenic factors include: 1) chemoattactants, 2) cytoki-nes/protease's TGF-b1, IL-1b, TNF-a, FGF-2, and 3) mammalian matrix metalloproteinases (MMPs). The gene expression of other proteins that will be determined are: rennin, endothelin, and iNOS. Gene expression will be studied by reverse-transcription PCR but we will also use the powerful technique of mRNA differential display (RRD) to screen the simultaneous expression of multiple genes or the expression of novel genes (75,76,77). The application of RDD is rather new to renal disease. The physiological factors that will be measured include: ATP levels, nitric oxide (iNOS, cNOS), lipid preoxidation (conjugated dienes, malondialdehyde), protein oxidation (protein sulfhydryl, protein carbonyls) and DNA damage. These factors will be studied in the renal slices by Project 2 (see Project 2 Background for justifications).

Inflammatory cells and damaged tubular cells express a variety of cytokines, including transforming growth fac-tor beta (TGF-b1), which is probably the most important cytokine for fibrogenesis (78). It induces the produc-tion of fibronectin, collagen, and proteoglycans, and inhibits extracellular matrix (ECM) degradation. TGF-b1 ex-pres-sion is elevated in several models of tubulointerstitial fibrosis, including obstructive nephropathy (73). Tis-sue necrosis factor a (TNF-a) may also induce increased matrix synthesis and deposition. The expression of this cy-tokine is reported to increase in several models of renal disease, although its exact role in interstitial inflam-ma-tion and fibrosis remains to be determined (79). In addition to increased matrix production by existing cells, pro-liferation of matrix-producing cells may also contribute to fibrogenesis (65). Stimulated macrophages secrete large amounts of platelet-derived growth factor (PDGF), which is a potent mitogen for fibroblasts but has minimal effects on ECM production (73). IL-1 has a number of biologic functions, including induction of proliferation and production of collagen in normal fibroblasts. Other mitogens for fibroblasts include epidermal growth factor (EGF), insulin-like growth factor 1 (IGF-1), transforming growth factor alpha (TGF-a), and interleukine1 (IL-1). Both EGF and IGF-1 are detected early in models of interstitial fibrosis. A variety of proteinases are released from infiltrating leukocyes as a result of tissue injury. The most important proteinases in renal injury are believed to be the mammalian matrix metalloproteinases (MMPs) (80,81). MMPs are a group of structurally- related enzymes that are involved in ECM turnover. These enzymes have the combined ability to degrade all the components of the ECM; have an absolute requirement for Zn2+; are secreted in an inactive form and undergo proteolytic cleavage to the active form; and are inhibited by specific tissue inhibitors of metalloproteinases (TIMPs). On the basis of molecular, biochemical, and immunologic data, these enzymes can be divided into three subgroups: the collagenases, the gelatinases, and the stromelysins. The collagenases cleave collagens I, II and III at a single location; the gelatinases degrade type IV collagen (hence the older name type IV collagenases) and fibronectin; and the stromelysins have a broad specificity, degrading proteoglycans and other ECM components. The enzymes are stored in latent forms in separate leukocyte granules and are secreted in response to stimulation by various cytokines (growth factors, chemoattractants). In addition to their presence in leukocytes, MMPs, and in particular gelatinase A (72 kD), are present in glomerular mesengial cells, raising the possibility that this enzyme may contribute directly to glomerular basement damage in glomerular injury (82). Nitric oxide (NO), a gas with vasodilatory properties, plays an important role in vascular regulation, immune responses, and tissue damage and repair (83,84). NO is an extremely labile molecule with a half-life of a few seconds, and it is rapidly oxidized to the stable end-products, nitrite and nitrate. NO is synthesized from L-arginine by a constitutive NO synthetase (cNOS) and an inducible enzyme (iNOS). The former enzyme produces NO in extremely small amounts for short periods, whereas the latter produces larger amounts of the gas for longer periods. Unlike cNOS, NO production by iNOS is regulated by cyokines, with TNF-a, IL-1, and INF-g associated with increased NO production. On the other hand, TGF-b and PDGF are potent inhibitors of iNOS. Constitutively-produced NO participates in many aspects of renal function, including glomerular hemodynamics, but the prolonged production of NO may be associated with glomerular injury.

The Pathology of SWL Treatment