|Home | About | Journals | Submit | Contact Us | Français|
To determine if the acute renal oxidative stress and inflammation after extracorporeal shock wave lithotripsy (ESWL), thought to be mediated by ischaemia, is most severe in the portion of the kidney within the focal zone of the lithotripter, and if these effects result primarily from ischaemic injury.
Pigs (7–8-weeks old) received either 2000 shock waves at 24 kV to the lower-pole calyx of one kidney or unilateral renal ischaemia for 1 h. A third group (sham) received no treatment. Timed urine and blood samples were taken for analysis of lipid peroxidation and the inflammatory cytokines, tumour necrosis factor-α (NF-α) and interleukin-6 (IL-6). At 4 h after treatment, kidneys were removed and samples of cortex and medulla were frozen for analysis of cytokines and heme oxygenase-1 (HO-1).
ESWL did not affect urinary excretion of malondialdehyde, but did elicit an eight-fold induction of HO-1 in the portion of the renal medulla within the focal zone of the lithotripter (F2), while remaining unchanged elsewhere in the treated kidney. There was no induction of HO-1 in renal tissue after ischaemia-reperfusion. Urinary excretion of TNF-α increased from the lithotripsy-treated kidney by 1 h after treatment, but was unaffected by ischaemia-reperfusion. As with the HO-1 response after lithotripsy, IL-6 increased only in the renal medulla at F2. By contrast, ischaemia-reperfusion increased IL-6 in all samples from the treated kidney.
These findings show that the acute oxidative stress and inflammatory responses to ESWL are localized to the renal medulla at F2. Furthermore, the differing patterns of markers of injury for ESWL and ischaemia-reperfusion suggest that ischaemia is not the principal cause of the injury response after ESWL.
ESWL continues to be the treatment of choice for uncomplicated kidney stones of < 2 cm in diameter, 25 years after its introduction into the USA. ESWL is noninvasive, well tolerated by patients, and has a low morbidity and high success rate . However, it has been known since 1985 that a clinical dose of SWs induces acute renal injury that extends from the papilla to the outer cortex, with a change in renal function in most, if not all, patients . This focal and predictable injury has two components; a traumatic vascular injury thought to be induced by the physical forces of the shock wave, and an ischaemic/hypoxic response linked to the severely damaged renal vessels. In addition, an inflammatory response, termed ‘lithotripsy nephritis’, quickly ensues at the sites of endothelial injury .
Renal vasoconstriction occurs in both kidneys after ESWL and hypoxic tissue can be seen around sites of intraparenchymal bleeding. These findings suggest that oxidative stress mediated by ischaemia-reperfusion (IR) might contribute to renal injury subsequent to ESWL . Support for this suggestion comes from experiments in which inhibitors of oxidative stress, such as allopurinol, reduced the level of enzymatic markers of tubular damage [4,5] or reduced the severity of the tissue injury after ESWL [6,7]. ESWL induces oxidative stress in the renal cortex , but whether a similar response occurs in the renal medulla, a region most susceptible to the injurious actions of shock waves, is not known. In addition, no direct comparison of ESWL- and IR-induced renal injuries has been made to further define the role of ischaemia in ESWL-induced renal injury.
For this study we hypothesized that while the entire ESWL-treated kidney might undergo oxidative stress, there should be localized areas of greater oxidative injury in the cortex and medulla within the focal zone of the lithotripter (F2), the region which receives the highest energy of the SWs. Furthermore, we expected the inflammatory response and oxidative stress after ESWL to occur in the regions of greatest injury . Finally, if IR mediates most, if not all, of the secondary injury after ESWL, then patterns of oxidative stress and inflammation should be similar with either treatment.
Our objectives were to assess systemic and renal oxidative stress and inflammation after ESWL and IR. Oxidative stress was evaluated by measuring levels of the lipid peroxidation product, malondialdehyde (MDA), and the stress-response protein, heme oxygenase-1 (HO-1) . We also measured levels of the pro-inflammatory cytokines TNF-α and interleukin-6 (IL-6), in combination with HO-1, as additional markers of the inflammatory response.
The experimental protocol used in this study was carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and was approved by the Institutional Animal Care and Use Committee of the Indiana University School of Medicine and Methodist Hospital. Female farm pigs, 7–8 weeks old (10–15 kg, Hardin Farms, Danville, IN, USA) were randomly assigned to one of three treatment groups, undergoing ESWL, IR, or a control (sham) group.
At the beginning of the experiment the pigs were rendered unconscious with an i.m. injection of ketamine (15–20 mg/kg) and xylazine (2 mg/kg), followed by intubation and anaesthesia with isoflurane (1–3%) throughout the experiment. Catheters were placed into an ear vein for the i.v. infusion of fluids, a femoral artery to monitor blood pressure and sample arterial blood, both renal veins to sample renal venous blood, and both ureters to collect urine. Sterile isotonic saline was infused at a rate of 1% body weight/h to maintain hydration and urine flow. Para-aminohippuric acid was infused i.v. at 70 mL/h.
For all treatment groups, after the initial surgical preparation and placement of catheters, the pigs were allowed a 1-h stabilization period before treatment. During this period urine was collected for 25-min. Samples of arterial and venous blood were drawn during each urine collection. On completing the pretreatment blood and urine collections, pigs in the ESWL and control groups were disconnected from the anaesthesia machine and transported unconscious to the lithotripsy suite (transport time ≈ 5 min), where anaesthesia was resumed. The pigs were placed supine into the gantry of the lithotripter (HM-3, Dornier, Wesseling, Germany) and lowered into the water bath (39 °C). The lower pole calyx of the left kidney was targeted to receive SWs by injecting a small amount of contrast medium (Hypaque 60%; Nycomed, Princeton, NJ, USA) through the ureteric catheter into the collecting system, and using the positioning fluoroscope to locate F2. For the ESWL group, 2000 SWs were delivered to F2 at a rate of 2 Hz at an output energy of 24 kV. The sham group was positioned in the same way, but did not receive ESWL. After ESWL the pigs were returned to the surgical suite (again disconnected from the anaesthesia machine for ≈ 5 min).
For the IR group, after placing the catheters, the left renal artery and vein were accessed by a 7.5 cm long flank incision along the border of the psoas muscle and working through the muscle layers. A string was looped around both vessels, the incision was hydrated with sterile saline and covered with wetted gauze. After completing the pretreatment blood and urine collections, a vessel clamp was placed on the renal artery and vein, with confirmation of cessation of blood circulation by renal blanching. After 1 h of ischaemia the clamps were removed to re-establish blood flow.
For all treatment groups, timed blood and urine samples were collected in a similar manner as the pretreatment collections. Plasma and urine samples were frozen on dry ice and stored at − 80 °C until analysis. Butylated hydroxytoluene in ethanol (final concentration 100 μM) was added to plasma and urine samples randomly designated for MDA analysis. At 4 h after treatment the kidneys were perfused with cold isotonic saline and removed. Tissue sections were removed from both kidneys from the lower and upper poles. The sections were then divided into cortex and medulla before snap-freezing in liquid nitrogen. Portions of each tissue sample were designated for assay of cytokines by ELISA or for Western blot analysis of HO-1.
In preparation for ELISA of cytokines, frozen renal tissue was weighed, then homogenized in two volumes of ice-cold buffer containing 10 mM HEPES, 10 mM KCl, 0.1 mM EGTA, 1 mM dithiothreitol, and 0.25 mM phenyl methylsulphonyl fluoride. The homogenate was centrifuged twice at 3000 g for 15 min at 4 °C. Protein was assayed using the Coomassie Plus assay (Pierce, Rockford, IL, USA) and aliquots of the final supernatant were stored at − 80 °C.
Renal microsomes were prepared for Western blot analysis. Briefly, frozen kidney tissue was weighed, then homogenized in three volumes of ice-cold 20 mM potassium phosphate buffer (pH 7.4) containing 135 mM KCL, 0.1 mM EDTA, Complete Protease Inhibitor Cocktail (Roche, Indianapolis, IN, USA), 1 mM sodium orthovanadate, and 0.1 mM phenyl methylsulphonyl fluoride. The homogenates were centrifuged at 10 000 g for 20 min at 4 °C. The supernatant was centrifuged at 100 000 g for 1 h at 4 °C. The microsomal pellet was resuspended in 20 mM potassium phosphate buffer (pH 7.4) containing 1 mM KCL, 10 mM EDTA and protease inhibitors. After assay of protein, aliquots were stored at −80 °C.
MDA was assayed in plasma and urine samples according to the method of Agarwal and Chase . Briefly, samples were reacted in a solution of 189 μM butylated hydroxytoluene, 293 mM phosphoric acid, and 7 mM 2-thiobarbituric acid at 100 °C for 1 h. The MDA-2-thiobarbituric acid complex was extracted into n-butanol and analysed by liquid chromatography with fluorescence detection, using an excitation wavelength of 515 nm and emission wavelength of 553 nm. Concentrations were determined against a standard curve generated with 1,1,3,3-tetraethoxypropanel the recovery rate was 92%.
TNF-α was quantified in plasma, urine and renal homogenates using a swine ELISA TNF-α kit (BioSource/Invitrogen, Carlsbad, CA, USA). For renal homogenates, 50 μg tissue protein was added to wells for assay; the recovery rate of TNF-α was 80%.
IL-6 was measured in plasma, urine and renal homogenates with a Quantikine Porcine IL-6 ELISA kit (R&D Systems, Minneapolis, MN, USA); the mean recovery rate of IL-6 was 104%.
For renal microsomes, 100 μg of protein prepared in sample buffer was separated on a 10% polyacrylamide gel (Invitrogen), and the proteins electrophoretically transferred to a PVDF membrane (Invitrogen). After blocking for 1 h in blocking buffer containing 5% milk, the membrane was incubated with rabbit anti-HO-1 antibody (1:1000; Stressgen SPA-895: Assay Designs, Ann Arbor, MI, USA) for 1 h at room temperature. Following washes, the membrane was incubated with horseradish peroxidase-conjugated donkey antirabbit IgG (1:10 000; Jackson Immunoresearch, West Grove, PA, USA) for 1 h. Chemiluminescence was detected with ECL reagent (GE/Amersham, Piscataway, NJ, USA). After probing for HO-1, membranes were stripped, blocked and re-probed for β-actin using a 1:20 000 dilution of mouse monoclonal anti-β-actin-peroxidase conjugate antibody (Sigma, St. Louis, MO, USA). Band intensities were quantified by densitometry. In control experiments, the identity of putative pig renal HO-1 (32 kDa) in Western blots was verified by cyanogen bromide digest.
Data were analysed by repeated-measure ANOVA using a mixed-effect model. For values measured in plasma and urine samples, the mixed-effect model included treatment group, time, and interactions between treatment group and time as fixed effects. For values measured in tissue, the mixed-effect model included treatment group, tissue type, and interactions between treatment group and tissue type as fixed effects. Potential correlation of values from the same pig was modelled using a compound symmetry variance-covariance structure. All analyses were carried out comparing group means, except for the urinary TNF-α data, which were very skewed. In this case, group median values were compared using a nonparametric mixed-effect model. Post-hoc comparisons were made using the Bonferroni adjustment to control the overall type I error to P = 0.05.
All treatment groups had similar baseline mean (SD) arterial pressure, of 64.4 (7.6) mmHg, and effective renal plasma flow, of 66.6 (23.3) mL/min. MDA was undetectable in the plasma of sham and ESWL-treated pigs at all time points examined (data not shown). While MDA was present in urine, there was no difference in urinary MDA concentration (not shown) or excretion rates (Fig. 1) between sham and ESWL-treated pigs for both untreated and treated kidneys.
HO-1 protein levels were determined in selected renal tissue samples of sham, ESWL and IR-treated pigs. The HO-1 levels were variable among pigs, particularly those in the ESWL group, which might reflect the variable effect we previously reported for ESWL on lesion size and renal function . Initial control experiments showed no significant difference in HO-1 levels of the cortex or medulla of the sham group and the untreated kidney of the ESWL group (Fig. 2). For HO-1 comparisons, this allowed the contralateral kidney to be used as an internal control within the same pig. Lithotripsy did not induce HO-1 in the renal cortex of either pole of the treated kidney (Fig. 3,A). By contrast, there was an eight-fold induction of HO-1 in the medulla of the lower pole (within F2) of ESWL-treated pigs (Fig. 3,B). There was no significant change in HO-1 in the medulla of the opposite pole of the treated kidney (Fig. 3,B).
To further evaluate the injury response, we compared HO-1 levels in cortex and medulla of the lower pole of the treated kidney of the sham, ESWL (within F2) and IR-treated groups (Fig. 4,A,B). There was no significant induction of HO-1 in the renal cortex after ESWL or IR compared to sham (Fig. 4A). However, HO-1 was induced by nine times more in the renal medulla (within F2) of the ESWL-treated than in the sham group (Fig. 4B). IR caused a smaller three-fold up-regulation of HO-1, but this was not statistically significant.
All treatment groups showed highly variable plasma TNF-α levels (Fig. 5,A). Some pigs maintained stable plasma TNF-α levels while others had extremely high baseline TNF-α levels which gradually decreased throughout the experiment.
Urinary excretion of TNF-α from the untreated kidney of ESWL- and IR-treated pigs was similar to that in the sham group. By contrast, TNF-α excretion from the ESWL-treated kidney was significantly greater than in the sham group by 1 h after treatment (Fig. 5,B) and remained elevated thereafter, albeit at lower levels. For IR-treated pigs there was a trend for increased excretion of TNF-α from the treated (ischaemic) kidney at 1 h of reperfusion, but it was not statistically significant (P = 0.096).
Tissue levels of TNF-α in the renal cortex and medulla were similar in all treatment groups at 4 h after treatment (Fig. 5,C). Interestingly, the TNF-α level was significantly lower in the medulla than in the cortex (P = 0.05).
IL-6 was not detectable in most plasma and urine samples. There was no difference in IL-6 levels of plasma and urine among the treatment groups at any time point (not shown). Sham pigs had undetectable or low levels of IL-6 in the renal cortex and medulla (Fig. 6). ESWL-treated pigs showed a localized increase in IL-6 only in the medulla within F2, while all other renal tissue had IL-6 levels that were similar to those in the sham group. By contrast, IL-6 was significantly higher in the cortex and medulla of both poles of the IR-treated (ischaemic) kidney, while remaining unchanged in the untreated kidney.
This study establishes HO-1 as a marker of oxidative stress and inflammation for the renal injury induced by ESWL. ESWL caused a rapid and substantial (eight-fold) induction of HO-1 in the renal medulla at F2. IR produced a pattern of oxidative stress and inflammation different from that caused by ESWL, suggesting thereby that IR is not the primary mediator of the response to renal injury caused by ESWL.
The stress-response protein, HO-1, catalyses the breakdown of heme to biliverdin, iron and carbon monoxide (CO) . Biliverdin is subsequently metabolized to bilirubin by biliverdin reductase. Iron is pro-oxidant and pro-inflammatory, while HO-1, CO and bilirubin normally have cytoprotective, anti-apoptotic, vasodilatory and anti-inflammatory properties. HO-1 induction is protective in many injuries, but under certain conditions HO-1 and its products can be injurious to tissue. In particular, it was shown in cultured fibroblasts that cellular protection might be associated with low levels of induction (less than five-fold), whereas cellular damage might be more likely with higher levels of induction (≥ 15-fold) [10,12,13]. The cellular injury correlated with elevated intracellular reactive iron . High levels of the other products of HO-1, CO and bilirubin, might also be toxic . Whether the eight-fold induction of HO-1 observed in the present study provides protection or exacerbates injury is not known; additional experiments in our laboratory will investigate this question.
The cell type expressing up-regulated HO-1 was not determined. Other studies of renal injury have shown HO-1 to be induced in renal tubules, vascular endothelium, interstitial cells and infiltrating macrophages . Future studies will address this issue in our porcine model of renal injury caused by ESWL.
Lithotripsy causes renal vasoconstriction, blood vessel breakage, intraparenchymal bleeding and reduced renal blood flow , all of which could contribute to tissue hypoxia and promote oxidative stress. Support for this idea comes from studies in which oxidative stress was ameliorated by calcium-channel blockers, allopurinol, vitamins, or other agents with antioxidant properties [8,14–17].
Markers of lipid peroxidation indicative of oxidative stress increase in the urine of small animals after ESWL. By contrast, an earlier study of ESWL-induced oxidative stress in pigs found no change in similar markers in the urine, in agreement with our MDA data .
Delvecchio et al.  analysed markers of oxidative stress within the renal cortex during ESWL and found the highest level of oxidative stress at F2. Despite the recognized susceptibility of the renal medulla to hypoxia  and to SW injury , no previous study has examined ESWL-induced oxidative stress in renal medulla. By systematically analysing renal cortex and medulla sections after ESWL we could more precisely locate oxidative injury than was possible in previous studies. Differences between those studies and ours with respect to tissues sampled, sampling technique, and markers of oxidative stress measured (e.g. lipid peroxidation vs HO-1) make direct comparison difficult. However, our results are consistent with our previous finding that the renal medulla and papilla are more susceptible to SW injury than is cortical tissue . The impact of this localized injury response on renal function remains to be determined.
ESWL-induced inflammation is characterized by the appearance of inflammatory cells within the renal parenchyma within 30 min after treatment, and by such long-term effects as permanent renal scarring and, less commonly, proliferative glomerulopathy [2,21]. In a clinical study to assess acute ESWL-induced inflammation, Dundar et al.  detected no change in plasma and urine levels of the inflammatory cytokines TNF-α and IL-6 in patients 2 h after ESWL. Our data are consistent with those findings, differing only in that by sampling urine directly from each kidney rather than from the urinary bladder, which contains pooled urine, we detected an increase in TNF-α excretion. The necessity of urine collection from each kidney suggests that TNF-α might not be a useful clinical marker of ESWL-induced injury.
In the present study, urinary TNF-α excretion peaked 1 h after ESWL and declined toward basal levels by 4 h after ESWL, while tissue TNF-α was similar to levels in the sham group at 4 h after ESWL. Tissue TNF-α was not measured at 1 h after treatment, so we do not know if tissue TNF-α was elevated at this time, to correlate with the elevated urine TNF-α.
The finding that the present injury markers increased only in the treated kidneys, or in urine from those kidneys, but not in plasma, suggests that the inflammation and oxidative stress observed after ESWL are localized rather than systemic responses. There are some caveats to this conclusion: First, others reported changes in lipid peroxidation in plasma after ESWL, while we could not detect MDA in plasma. This disparity suggests either that our assay lacks sensitivity or that differences in assay methods produce different results. Second, the highly variable plasma TNF-α levels might have masked any potential changes caused by treatment; seven of 38 pigs had baseline TNF-α levels of > 1000 pg/mL. Finally, our inability to detect IL-6 in plasma and urine was not a weakness of the assay, as the manufacturer reported that 11 of 12 porcine plasma samples measured at or below the lowest standard.
With these considerations, the localized oxidative stress and inflammatory responses we found were largely confined to the medulla at F2 in the treated kidneys. By contrast, IR elevated the levels of markers in both poles of the treated kidney, but in a different pattern than for ESWL. This suggests that ischaemia is not the only cause of the injury response to ESWL.
SWs damage renal tissue and impair renal function; two phases in this process have been identified . The first involves direct SW-induced cellular damage and destruction. The second most likely results from local haemorrhage and ischaemia. Direct tissue injury from SWs follows a predictable pattern. The earliest and primary injury from ESWL is vascular, but renal tubules are also damaged. Blood from ruptured vessels subsequently pools in the renal parenchyma causing local ischaemia and an environment for oxidative stress and inflammation to occur.
Two possible mechanisms for the up-regulation of HO-1 after ESWL can be postulated. First, ischaemia associated with tissue injury after ESWL could result both from ESWL-induced vasoconstriction and damage to blood vessels leading to reduced or cessation of blood flow in the damaged tissue. Overall ischaemia to the ESWL-treated kidney should produce oxidative stress throughout the organ. Whether varying levels of vasoconstriction occur in different regions of the kidney during ESWL is unknown. Oxidative stress from ischaemia could cause induction of HO-1 through free radical formation or by destabilization of heme from intracellular heme proteins . Alternatively, or in addition, heme could be released by haemolysis of blood cells by SWs, particularly at sites of blood pooling. Heme is a potent inducer of HO-1 . Likewise, free iron released from heme degradation up-regulates HO-1 . In this study, we investigated the contribution of ischaemia to HO-1 induction. Our results suggest that ischaemia plays a smaller role in acute oxidative stress and inflammation after ESWL. The role of heme and free iron in this acute injury response remains to be determined.
Lithotripsy shows a distinct pattern of response compared with other renal injury models. Endotoxin and IR cause increased plasma and urine TNF-α and IL-6. Various agents, including cisplatin, puromycin aminonucleoside, heme, endotoxin, mercuric chloride, and human nephropathies induce renal tubular HO-1 [24–30]. In most cases the injury response affects the renal cortex or the corticomedullary junction. The response to ESWL might reflect its unique mode of injury in which SWs cause physical tissue damage principally to the renal medulla.
ESWL is considered to be a safe procedure, but recent findings suggest that ESWL might lead to long-term adverse effects, including new-onset hypertension, the induction of diabetes mellitus, and the exacerbation of stone disease . The new perspective on the acute effects of ESWL on the kidney presented herein might lead to methods to protect the kidney from acute injury and thereby prevent some of the potential long-term complications of ESWL.
The authors express appreciation to Drs Rajiv Agarwal and Joe Bidwell for support on technical aspects of this study. This project was funded in part by PHS grants P01-DK43881 and R01-DK67133.