Physiopathology and etiology of stone formation in the kidney and the urinary tract

Kidney stone development is thought to require the formation of crystals in the tubular fluid, followed by crystal retention and accumulation in the kidney [11]. Three pathways of stone formation and growth are currently being investigated (Fig. 2). The first hypothesis, termed the free particle model, states that crystal nuclei form by homogeneous nucleation in the lumen of the nephron under conditions of a phase change (increasing supersaturation) in the dissolved salts present in the ultrafiltrate [12]. Subsequently, these nuclei would grow in size and eventually lodge (be retained) in the lumen of the distal nephron, causing obstruction of that tubular segment. Obviously, free particle formation could occur in the renal collection system at the level of the minor calyx. The second hypothesis, termed the fixed particle model, also requires crystal nuclei to form in the lumen of the nephron, and then adhere to the apical surface of the tubular epithelium [13, 14]. While a number of mechanisms have been proposed to model this crystal–cell attachment step, the most commonly cited model requires renal cell injury, probably as a result of high tubular oxalate levels [15]. Once the crystal–cell attachment step has occurred, the crystal nuclei would be fixed in position and exposed to the potentially supersaturated ultrafiltrate that would facilitate further growth of these crystals. Both these theories could result in the plugging of the nephron and lead to intratubular calcification, termed tubular nephrocalcinosis [11]. Randall, in his historic paper in 1940, described intraluminal calcification (papillary lesion type II) or nephrocalcinosis in only 23 cases and compared it to the more common finding (204 cases) of interstitial plaque (papillary lesion type I), placing the type II lesion in a category different from that of type I [16]. We have shown that, in patients who form brushite stones [17], or who form apatite [hydroxyapatite, Ca₅(PO₄)₃(OH)] stones because of distal renal tubular acidosis (dRTA) [18], or patients with calcium oxalate stone due to obesity bypass procedures [19], or patients with cystinuria [20], their inner medullary collecting ducts (IMCDs) are plugged (tubular nephrocalcinosis) with cystine, which leads to total destruction of the lining cells and focal sites of interstitial fibrosis. However, it is unclear how the free or fixed particle hypotheses could lead to clinical stone formation (i.e. a stone in the renal pelvis large enough to obstruct the ureter), except that either the free or fixed particle process, or both, are occurring in the lumen of the ducts of Bellini and/or in the renal pelvis.

A third pathway suggests that crystals in the urine can become attached to a site of exposed crystalline deposits of interstitial calcium phosphate (termed Randall’s plaque), following the loss of the normal urothelial covering of a renal papilla (Fig. 2). An anchored nidus of urinary crystals could form as an overgrowth on the interstitial plaque, permitting a fixed stone to form and potentially grow over many years. This theory is explicit in stating that sites of interstitial plaque are the initiating lesions. Randall examined 1,154 pairs of cadaveric kidneys by carefully opening the renal pelvis of each kidney and examining each accessible papilla with a hand lens [16]. This examination revealed an “innocent” appearing lesion that was a cream-colored area near the papillary tip and appeared to be subsurface or subepithelial in location. This lesion was seen in 20.5% of all the kidneys he studied. By light microscopy, the lesion was found to be a plaque of calcium salts deposited in the interstitial tissue and definitely not intraluminal in location. Evidence of inflammation was rarely found, and deposits were localized initially to interstitial collagen material as well as tubular basement membranes. Subsequently, the deposits appeared to fill much of the intertubular space. Chemical analysis of the regions of plaque showed calcium, nitrogen, carbon dioxide and phosphorus. His next major observation was the finding of a small stone (2 mm) that projected into the lumen of the renal pelvis. It was firmly attached to an area of visible calcium plaque and found to be composed of calcium phosphate. Such stones were found attached to papillae in 65 kidneys. Randall was able to study several stones still attached to the papillary wall and concluded from these studies that (1) attached stones were growing from and were supported by interstitial calcium plaque; (2) the sites of interstitial plaque could lose their urothelial cell covering, allowing the plaque to be exposed to the calyceal urine, and (3) some detached stones had a concave surface with patches of phosphate material (possibly representing the attachment site to the interstitial plaque). The observations Randall made from 1,154 pairs of post-mortem kidneys were earth shattering in their importance and originality. However, what Randall was not able to determine because of the techniques available to him in the 1930s was the composition of the mineral in the plaque, the type of crystalline material that formed right at the site of overgrowth, and the transition of mineral type (i.e. calcium oxalate, uric acid, calcium phosphate) from the site of attachment into the body of the stone. In addition, Randall lacked extensive clinical data on each of the kidneys he examined and, therefore, suggested that all stones formed according to his hypothesis. Consequently, his idea was easily challenged.

Our group set out to test the hypothesis that regions of Randall’s plaque develop in unique anatomical sites of the kidney and that their formation is conditioned by specific stone-forming pathophysiologies. To test this hypothesis, we employed state-of-the-art digital endoscopic equipment during percutaneous nephrolithotomy (PNL) for stone removal. We chose to study idiopathic calcium oxalate stone formers (ICSFs) as our primary group of stone formers because they represent approximately 75% of all stone formers. This protocol allowed us to obtain digital images of all papillae for plaque surface determinations and permitted us to obtain papillary biopsies of regions with and without plaque. During intraoperative endoscopy, we identified irregularly shaped regions of whitish material generally located on the papillary tip and usually surrounding the openings of the ducts of Bellini (Fig. 3a), the same plaque material described by Randall. The plaque is located deep in the urothelium (suburothelial) and has a smooth outer surface, except where it is pitted due to the removal of an attached stone. Light microscopic examination of Yasue-stained papillary biopsies revealed minute deposits of mineral, first laid down in the basement membrane beneath cells of a specific portion of the urinary tubule, the thin limbs of Henle, as they pass near the tip of renal papilla—close to where urine drips into the minor calyx (Fig. 3b). Initially, these mineral deposits are seen as electron-dense spherical objects of approximately 50 nm in diameter to multilayered spheres with alternating light (mineral layer) and dark (matrix layer) rings or bands (Fig. 3d, insert). The individual deposits appear then to coalesce on the collagen bundles in the interstitial space and become embedded in an electron-dense matrix material, generating islands of plaque. These islands extend down to the basal side of the urothelium and encase the ducts of Bellini (Fig. 3c). In order to determine the mineral composition of the interstitial plaque, we developed a protocol to analyze these small sites of crystalline material using Fourier transform infrared microspectroscopy (μ-FTIR). The mineral in the individual plaque spherules was always determined to be biological apatite (calcium phosphate). In addition, we have identified osteopontin as one of the proteins forming the crystal–matrix interface [21] of these spherules (Fig. 3e), while heavy chain 3 (H3) of the inter-alpha-trypsin molecule [22] was located in the matrix layer (Fig. 3f). No evidence of cell injury, inflammation, interstitial fibrosis or intratubular crystalline material was detected in any of the ICSFs’ papillary biopsies [23].

Randall was convinced that interstitial plaque was a perquisite for stone formation and growth [16]. Unfortunately, he was not able to obtain a detailed clinical history on each of the autopsy kidneys he had studied, so his theory that stones grow attached to sites of interstitial plaque was stated in very general terms. Generality is crucial to our understanding of stone pathogenesis, because if stones indeed grow on plaque as a rule, then how they do so, and how plaque itself forms and enlarges, become primary clinical research objectives. The problem is especially complicated because, in other groups of calcium oxalate stone formers, apart from ICSFs, growth of stone on plaque is not a prominent mechanism. Therefore, we conducted a series of studies at the time of PNL where every attached stone in a group of ICSFs could be accounted for as attached to plaque or not. Figure 4 shows such a stone fixed to a papilla at a site of plaque. Matlaga and colleagues performed a retrospective study of ICSF patients at the time of their PNL procedures for stone removal and found that 11 of 23 patients had attached stones, with 91% of these stones at sites of plaque [24]. Miller and colleagues [25] have recently repeated this study, using a prospective design. They studied nine ICSF patients who had a total of 115 stones, of which 81 were attached to a site of Randall’s plaque. Based on the observed data, the final point estimate for the number of stones attached to plaque was 0.754 (95% confidence interval 0.575, 0.933), P = 0.013, which suggests that, in ICSF patients, stones develop attached to plaque.

As stated earlier in this paper, those attached stones that have been plucked from papillae, generally have been found to have an irregular concave surface that fits against a papilla and a smooth convex surface that faces the urinary space [16]. Often, the concave surface has been found to have a stock or umbilicus, thought to be made up of plaque and in continuity with the interstitial plaque before the stone had been detached from the papilla. Cifuentes-Delatte et al. [26] and Pieras et al. [27] have also observed a “footprint” of a papilla as a depressed zone (“umbilication”) on the concave surface of the stone, which corresponded to the shape of the tip of a papilla, and this concave surface contained mineral detached from interstitial plaque. Daudon and colleagues recently reported on an analysis of 5,401 stones with umbilications [28]. They found the main mineral component of this plaque material to be carbapatite (91%), with lesser amounts of amorphous carbonated calcium phosphate, sodium hydrogen urate or uric acid. These observations again support the ideas laid out by Randall [16]. Thus, calcium oxalate stones in ICSF patients appear to grow like stalactites adherent to the tips of kidney papillae and, typically, are made of layers of calcium oxalate surrounding a core of apatite.