A majority of patients form stones containing calcium oxalate (CaOx) and calcium phosphate (CaP). Of these most form predominantly CaOx stones in the absence of systemic disease, so called ‘idiopathic CaOx stone formers’ (ICSF). These stones form outside the nephrons, on the papillary surfaces, attached to interstitial apatite (Randall’s) plaque (RP). The initial overgrowth is CaP in the form of apatite over which CaOx gradually layers to form the bulk of the stone. The main driving force for overgrowth is familial (idiopathic) hypercalciuria (IH) in which increased vitamin D activity increases intestinal calcium absorption, and bone turnover raising the amount of calcium available for urine excretion. Postprandial fall of overall tubule calcium reabsorption (rCa) permits hypercalciuria without increase of serum calcium or calcium filtered load. Proximal tubule (PT) rCa falls (rCa pt ) which increases delivery of calcium to the thick ascending limbs (TAL); this may increase reabsorption by medullary TAL (mTAL) without water increasing interstitial [Ca] and therefore [Ca] in the descending vas recta. RP forms in the basement membranes of the thin limbs in the papillum, which are bathed by blood from descending vas recta so the reduced rCa pt of IH may drive RP formation just as IH itself increases urine CaP and CaOx super-saturations (SS) that drive overgrowth of CaP and CaOx over RP. The smaller group of patients whose stones are predominantly CaP produce apatite plugs within their papillary collecting ducts of Bellini (BD). Plugs injure epithelial cells and create interstitial inflammation and scarring. Stones do not grow on RP but rather off the ends of plugs. The systemic diseases that cause calcium stones, primary hyperparathyroidism, bowel diseases, renal tubular acidosis, and primary hyperoxaluria all produce apatite BD deposits, with varying amounts of RP, the tissues in each case reflecting the unique physiological abnormalities produced by the systemic disease. Organic stones—uric acid and cysteine, and the rarer drug stones—differ remarkably from calcium stones. Uric acid stones occur when low urine pH (<5.5) increased the concentration of undissociated uric acid above its solubility. Cystine stones and the drug stones are simply conditions in which organic molecules are excreted at rates that create concentrations above solubility. Being so crucial for clinical management, stone analysis is paramount in the evaluation of stone formers. Given calcium stones, 24 hour urine and morning blood samples will usually identify the cause, and long term treatment can be guided by the numerous prospective double blind randomized trials that have been accomplished to date. Uric acid stones will cease given a urine pH above 6. Cystine stones can be managed with fluids, higher pH, and disulfide binding drugs, though success can be difficult to achieve.
kidney stones, calcium oxalate, apatite, uric acid, supersaturation, cystine, hypercalciuria, primary hyperparathyroidism, enteric hyperoxaluria, primary hyperoxaluria, renal tubular acidosis, Randall’s Plaque
Stone forming patients are labeled by the stones they form. The largest group form calcium stones (calcium oxalate (CaOx), calcium phosphate (CaP), or a mixture of both); others form stones containing primarily uric acid (UA), struvite (infection stones), or cystine. The pathogenesis of each stone type differs; clinical evaluation, therefore, begins with analysis of all available stones. A large fraction of the common calcium stone formers have no systemic disease but do excrete more calcium than the average person, an inherited trait (idiopathic hypercalciuria, IH) that probably reflects the simple fact that risk of calcium stones increases with rate of urine calcium excretion. These “idiopathic” calcium stone formers (ICSF) produce stones mainly of CaOx crystals, and are by far the most common kind of patient a physician will encounter in practice. All ICSF studied thus far exhibit the same renal histopathology and surgical anatomy, and appear to form their stones via a single highly defined pathway, and therefore represent what amounts to a well-defined specific disease. By contrast, idiopathic calcium stone formers whose stones contain over 50% CaP (IPSF) have renal histopathology, surgical anatomy, and mechanisms of stone production that differ radically from those of ICSF, though they share IH as a common physiological trait. Patients with calcium stones due to systemic diseases such as primary hyperparathyroidism, distal renal tubular acidosis (RTA), ileostomy, small bowel resection, and obesity bypass procedures exhibit renal histopathologies, mechanisms of stone production, and treatments that are specific to the underlying disease. Despite their heterogeneity, all stone diseases share formation of unwanted solid phase crystals that produce similar clinical syndromes. As well, crystals nucleate and grow according to natural laws one can use in clinical practice.
Stones, Clinical Presentation, and Natural History
Crystals form in urine of virtually all healthy people, but among those destined for stone disease they are coarser and larger, and cause a syndrome of pain and hematuria which often eludes diagnosis for some time, the crystals being dispersed or not looked for. As perhaps a second stage, though the continuity may be more apparent than real, crystals become so large and dense they form a gravel or sand-like material whose passage is gritty and evident. Such attacks of gravel with or without bleeding are usually recognized by patients and diagnosed as a kind of stone problem. Certainly the conventional passage of a formed stone is a step beyond gravel, if only because the formed stone may lodge along the urinary tract and demand a procedure to remove it. Pain is usual though not invariant, as is bleeding. Finally, stones and plaques of crystal may form along the calyces and papillary tips, forming the radiographic pattern of nephrocalcinosis that may or may not be associated with stone passage, gravel, or crystalluria with hematuria. Thus is stone disease divided into at least four great divisions as seen by patients and doctors, though all arise in similar ways.
Nature of Stones and Crystals
Formed stones are a complex of proteins interlaid with crystals, like a kind of disordered mineralized tissue. The matrix proteins include Tamm-Horsfall protein (THP), osteopontin, albumin, prothrombin fragments, and other urine proteins . The crystals are most often CaOx, monohydrate (COM) or dihydrate (COD) ( Table 67.1 ), often admixed with CaP species such as apatite (APA) and brushite (BR), less often admixed with UA. In a smaller number of patients (12% of 2011 patients seen in the University of Chicago Kidney Stone Clinic with stone analyses), stones are predominantly or completely composed of CaP ; CaP stones are more common in women. In another 7% of our patients, stones were predominantly UA. Uncommon stones (each seen in fewer than 2% of patients) include those of cystine, struvite, ammonium acid urate, and drugs ( Table 67.1 ). Because each kind of stone crystal evidences a particular disorder of urine supersaturation, stone analysis is the bedrock of clinical practice. Repeated stones all deserve analysis, as urine chemistry changes with time, and a change in composition may signal a need for change of treatment.
|Crystal Type||US Population a||US Veterans b|
|Calcium oxalate monohydrate||43||63|
|Calcium oxalate dihydrate||61||42|
|Ammonium acid urate||0.4||1|
b Mandel NS, Mandel GS. Urinary tract stone disease in the United States veteran population, II: geographical analysis of variations in composition. J Urol 1989;142:1516–1521; Mandel NS, Mandel GS. Urinary tract stone disease in the United States veteran population, I: geographical frequency of occurrence. J Urol. 1989;142:1513–1515.
The pain from passage of a stone begins as a discomfort, often not called a pain, and progresses over 30 minutes or so to a plateau of extreme severity, which remains constant thereafter unless medications are given or the stone moves. The pain of a stone in the renal pelvis and upper ureter is over the lateral and anterior abdomen on one side, in a broad band from the ribs downward, and is not well localized. As the stone moves down, the pain moves in parallel, toward the bladder, and a downward moving pain is sure evidence of stone movement. At the uretero-vesical junction, stones often lodge and produce urinary frequency and urgency, dysuria, and hematuria, the complex easily misdiagnosed as urinary infection if symptoms begin there. This latter is common with small stones that easily pass the upper ureter but stick at the bladder junction, and give their first symptoms there. When the stone passes into the bladder, obstruction is relieved and pain disappears on the instant, with a speed unlike any other known pain.
The character of colic is not describable. All attest to severity surpassing any other pain that afflicts humankind, but poets and writers alike fall silent when asked to describe it. What little we have been offered suggests a deep boring and hot sensation, but no more detail than that. Colic is diagnosed by its curious timing, progress, peculiar intensity, and magical disappearance. On the left side, colic can resemble diverticulitis, except the latter causes constipation, whereas colic causes diarrhea and vomiting. On the right it mimics biliary colic, but is too lateral, too severe, too detached from eating, and unnatural in its downward progress. We have constructed long lists of differential diagnosis, but find the problem pointless, because no one mistakes renal colic, for long, as anything other than what it is.
Radiography of Stones
CaOx and CaP stones tend to be small (1–10 mm), bright, and circumscribed. UA stones are famously radiolucent on routine radiography, but can be easily visualized by computed tomography (CT) scan done without contrast infusion. Cystine stones are only faintly visible on routine x-rays, so that we often underestimate size, but CT is ideal for them. A struvite stone, being an amalgam of magnesium ammonium and calcium phosphate crystals and much protein that is made by waves of bacterial overgrowth, naturally appears like a gnarled root . Ultrasound detects stones, but is less sensitive for small stones and does not show anatomy well compared with CT. Because non-contrast CT gives an excellent view of the renal pelvis and of whether there is obstruction, it is our preferred mode of visualization. Protocols using lower doses of radiation are under investigation to evaluate their diagnostic accuracy, and may become preferred in order to limit radiation exposure.
Stones pass in inverse proportion to their size, although time to passage may require four to six weeks. Below 5 mm, passage is common, while ureteral stones greater than 7 mm in diameter pass spontaneously less than 50% of the time. Above 10 mm, passage is unlikely. Stones in the distal ureter are more likely to pass than those located more proximally. Large stones in the renal pelvis grow at leisure and can gradually fill up the collecting system. Such behavior is more usual for struvite, UA and cystine stones, but CaOx stones do attain such stature occasionally. A stone that fills at least two calyces is called a stag horn stone, although this old and picturesque name is better reserved for stones that fill all of the calyces, and have the shape of a stag’s horns.
Given that crystals are coarse and often symptom producing, one might think that urinalysis is valuable in clinical practice of stone prevention, but our experience is the opposite. It is true that one can teach urine crystal morphology in a stone clinic, and the appearances of crystals under the microscope are beguiling, especially under polarized light, but the impact on diagnosis is modest. In addition, a single sample of urine may not represent the daily average. However, at least one study shows that persistent crystalluria in a first morning urine while on treatment is correlated with relapse.
COM crystals are small dumbbell shapes that can superficially resemble red blood cells. The more dramatic COD is bipyramidal, as shown in all usual atlases. UA forms a reddish dust, because it absorbs uricine, a bilirubin metabolite and is an indication of low urine pH. Calcium phosphates are often small, and called “amorphous,” though they are in fact crystalline if seen under higher magnification. Calcium phosphates form whenever urine pH is much above 6.5, and may be irrelevant to stone formation or not depending on how well the spot urine pH represents that of the average 24-hour urine. Crystals occur in casts, and such casts have no established significance. Cystine crystals are found in urine of most cystinuric patients, and have little significance as the cool temperature of the room as compared with the body can allow crystals that are not present in vivo . Of course, they may be the first clue as to the correct diagnosis. Perhaps the most valuable finding is a urine pH above eight, which suggests infection with urea hydrolyzing bacteria, and requires follow up with urine culture. Overall, we hesitate to seem skeptical of urinalysis, given its preeminence in nephrology, but in the stone clinic it is more of aesthetic visual than clinical interest.
In the United States, 11% of men and 5.6% of women will report having formed a stone by their seventh decade; the risk is about three times higher in whites compared with African Americans, but has increased in both groups as well as both sexes over the past three decades. Although this may reflect, in part, improved radiologic detection of asymptomatic stones, a true increase in stone formation seems likely based on hospitalization rates. Obesity is associated with increased risk for stone formation, which rises with body mass index and waist circumference, especially in women.
Stone formation is associated with an increased risk of chronic kidney disease, although the mechanism is not known. There also appears to be an increased risk of hypertension among stone formers, and an attendant increase in risk of cardiovascular disease.
There is a significant genetic component to stone formation. In studies comparing stone rates in male twins, the concordance rate for stones was 32% in monozygotic compared with 17% in dizygotic twins. Similarly, in a longitudinal cohort of adult men, those with a family history of stones were 2.5 times more likely to form a first stone over eight years of follow-up than those without. A number of rare monogenic diseases are known which predispose to stone formation. However, most of the genetic predisposition appears to be polygenic, and is strongly influenced by environmental triggers, as seen in diseases such as hypertension and diabetes.
Following a first stone you may confidently predict that 50% of patients will form at least one more stone after eight years and 75% will recur by 15 years. In those with more than one stone, recurrence occurs sooner; approximately 45% of such ICSF will have a new stone within three years. Once recurrent, stones maintain a one- to two-year interval between themselves in the average patient, although the spread is very wide. We have tried in vain to properly characterize high-velocity stone formers. When we have analyzed the stone history of patients with many stones we find the rate of stones per year surprisingly constant and that stone number increases pari passu with how long people wait from their first stone to obtaining preventative treatment. On the other hand, we find many patients in whom a single stone episode will yield a collection of dozens or perhaps hundreds of small stones passed within hours or days, whereas for most people an episode is no more than a single stone. We have never understood how batches of stones are made and passed within a day’s span.
Primary Process of Crystallization
Solid phase begins as the association of ions or molecules in solution reduce the energy of bonds with water molecules so much that they leave the solution as so-called solid-phase nuclei. For calcium and oxalate, or calcium and phosphate, the association involves formation of dissolved salts; for uric acid and cystine, the molecules associate with themselves, mainly by hydrophobic forces. In either case, for practical purposes, one finds that nuclei form as some critical concentration is reached.
The Notion of Supersaturation
An ideal formal statement of the critical concentrations for nucleation is to compare salt or molecular concentrations of a solution to those at the solubility, any excess being called the supersaturation (SS), and implying the solution is carrying more than it can hold. This naive idea gives the right sense of SS, in begging the question of what is meant by “more than it can hold.” If crystals of a salt are incubated for some time, usually 48 hours, with a solution that is initially free of the salt, the concentrations of salt in solution will increase until the crystals no longer dissolve. This concentration is called the solubility and is empirically determined, being otherwise unknowable. A solution that contains the solubility concentration is called saturated. Values for CaOx, calcium monohydrogen phosphate (BR) (a commonly measured initial phase of calcium phosphate), and undissociated UA are (in μM/l) 6.2, 0.39, and 520, respectively.
If one slowly adds calcium and oxalate to a saturated solution it will remain clear. The salt concentration will rise above solubility, and the solution is called supersaturated. The level of SS may be expressed as the ratio of the concentration of the dissolved salt to its solubility. The solution is metastable in that the extra salt, CaOx, for example, will crystallize out and bring the solution to solubility given any of a number of perturbations. The most obvious is to add seed crystals of, for example, CaOx, which will grow until equilibrium is restored. Seeds of CaP, or UA, which are “heterogeneous,” will also promote formation of CaOx nuclei on their surfaces and deplete the system back to solubility. The same is true for cell debris and aggregated proteins. If one continues to add calcium and oxalate, but not a solid seed, SS will rise to a critical point called the upper limit of metastability (ULM), which is an empirical limit at which CaOx nuclei will form spontaneously, and bring the solution back down to solubility.
Actual values for SS and the ULM can be determined in human urine ( Fig. 67.1 ). For CaP ( Fig. 67.1 , open circles) the SS values of normal urine range from 0.1 to 5, versus 2 to 20 for CaOx ( Fig. 67.1 , stars). The ULM can be determined by adding sodium oxalate to aliquots of the urine, to achieve a CaOx ULM, or calcium chloride, to obtain the CaP ULM. Among healthy individuals, the ULM is strongly dependent upon, but considerably above, the SS, as shown by the position of all points above the line of identity. ULM values range from two- to 11-fold above SS for CaP and 11- to 50-fold for CaOx. How ULM varies with SS is not understood, but that it does so reflects a highly protective set of mechanisms that appear to be aimed at defense against renal crystallization.
Relationship of Supersaturation to Stone Composition
As measured clinically SS represents two or three days of urine measurement out of a lifetime, and one wonders to what extent such a slim sampling can be useful in predicting formation of solid phases over the months to years that stones begin and grow in the renal collecting system. In fact, SS appears to be a very robust measure of long-term crystallization forces. Among men and women whose stones were predominantly CaOx ( Fig. 67.2 ), COM SS was above corresponding same-sex normal values, whereas CaP and UA SS values were not. Among patients whose stones were predominantly CaP ( Fig. 67.2 ), CaP SS was above same-sex normal values, whereas COM SS was not. Naturally, because higher urine pH favors CaP SS and reduces UA SS, UA SS is lower than normal in this group. Among patients with UA stones, UA SS is above normal and COM SS is not. Overall, even when measured only a few times, urine SS reflects at least the main components of stones, and therefore one must conclude that renal physiology acts to maintain SS at a reasonably stable level in most stone formers, and that a few urine samples give a strong insight into long-term, stone-forming tendencies of a patient.
The normal values we used to compare with patient data show a marked sex difference. Whereas healthy women have urine COM SS below that of stone-forming women ( Fig. 67.2 ), COM SS in urine of men is nearly at the levels of patients and is much higher than among healthy women. This occurs mainly because, being larger, men have higher traffics of calcium and oxalate into urine than women, but share with them virtually the same daily urine volume. The finding accords well with the epidemiological observation that male stone formers outnumber female stone formers, and further support the value of even sparse SS measurements to characterize the stone forming impulse of an individual.
Given the general rule that stone composition follows SS, one may ask about the exceptions. When individual urine samples fail to show SS corresponding to stone composition, or show SS not reflected in stone composition, can one use such a discrepancy clinically? We found two types of exception. Sometimes SS elevation is present but the predicted mineral phase is absent from all analyzed stones (Type 1). This occurs mainly when urine volumes are very low and multiple SS values are high. The phenomenon suggests some important renal mechanisms that reduce crystallization. Type 2 discrepancies are solid phase in stones without a relevant SS in the urine. Here, we have found very high urine volume is the rule, and assume that matters have changed with the patients after passing a stone, so that fluid intake increased and a past SS was abolished. These exceptions are a kind of sporadic accent on a general backdrop of correspondence, and we propose that a break in correspondence has clinical use in pointing to increased urine volume, especially, which may not hold long term.
Control of Supersaturation
Given that SS is the driving force for crystallization, and that direct experiments document a strong link between SS measurements made clinically and stone mineral composition, as well as gender differences in stone forming potential, the regulation of SS and the factors that influence it are clearly at the center of stone pathophysiology. One might at this point illustrate SS values from healthy men and women, and how they vary with factors such as urine volume, pH and mineral traffic, but our interest is what happens among patients with stones. The following figures are therefore made from our measurements of three 24-hour urine samples obtained prior to any treatment among 4500 patients with documented stone formation. As best is possible, they convey some sense of what one encounters in such a population.
Effects of Water Reabsorption
For CaOx, reduced urine volume raises SS markedly ( Fig. 67.3 ), although even at the highest urine flows SS remains above 1, indicating near universal SS in human urine samples over 24 hours. Put another way, crystals of CaOx are unlikely to dissolve in human urine once formed, and anything that provokes nucleation of CaOx is apt to cause a stone to grow given enough time and provided other defenses do not prevent such from happening. For CaP ( Fig. 67.4 ), volume also matters, but the tilt of the graph and spread of the points make clear how much less dramatic the effect. Also note that levels of SS for CaP in urine are much lower than for CaOx, and half of all samples are below one, indicating that CaP phases need not persist even if initially formed. For UA volume effects are also quite marked (not shown). As in the case of CaP, over half of urines are under-saturated. Overall, urine volume of two liters is a useful point to observe. Above this point, further reduction of SS is possible, but the number of urines thins, because one reaches the limits of plausible behavior. Below one liter, high SS is the rule and increase becomes very important. This supports the clinical maxim of two l/day as a goal for urine volume, while less than one l/day is a serious problem. Of interest, the normal person has a urine volume of 1.4±0.05 liters as an average, yet does not make stones. Urine volume obviously varies over the course of the day, and is lowest overnight, which is the time of day associated with the highest levels of SS. This should be kept in mind when advising patients to increase fluid intake.
Effects of Relevant Urinary Solute Traffic
For CaOx, urine calcium excretion is a main controller of SS ( Fig. 67.3 ); therefore, it should be unsurprising that hypercalciuric states are the most prevalent clinical causes of stone disease. Rather than using hard cut-points for diagnosing hypercalciuria, urine calcium should be thought of as a graded risk factor more like blood pressure, and like blood pressure in relation to stroke, higher is worse than lower. As well, and perhaps like the stroke analogy, the spread of SS about any urine calcium point shows how many other factors must be interacting. One is volume, and the rest are not defined on this presentation. Urine oxalate excretion ( Fig. 67.3 ) has none of the effect of calcium, the graph being almost spherical. This fact is incontrovertible, yet in vitro , and for theoretical reasons, urine oxalate concentration ought to exert a controlling influence equal to that of calcium . The reason is that the range of oxalate excretion is modest, and the effects offset by other factors such as volume and calcium. For CaP matters are similar but with less sharpness ( Fig. 67.4 ). Urine calcium excretion has a striking effect, as for CaOx; volume affects SS modestly, and urine phosphorus excretion virtually not at all. For UA, volume influences SS rather more than for CaP and a bit less dramatically than for CaOx, and UA excretion is utterly without effect.
Effects of Urine pH on Supersaturation
What really matters in the case of CaP and UA SS is urine pH ( Fig. 67.5 ); in a sense the link between SS and mineral content of these two solid phases reflects stability of 24-hour average urine pH. Above pH 6, virtually no urine is supersaturated with UA, and below 5.3 virtually all are supersaturated. For CaP, the effect is less exacting, perhaps because calcium concentration plays an important role, but SS does not occur below pH 5.3. For CaOx, pH has little effect on SS. That it has any reflects increasing calcium complexation with phosphorus and citrate at higher pH levels, so urine calcium ion falls. As noted before, virtually all urines are supersaturated with CaOx.
Regulation of Urine pH in Relation to Supersaturation
We have no desire to duplicate here detailed materials elsewhere in this volume concerning acid excretion and control of renal acid base balance. However, we have something to add of potential value, offering a perspective usually not considered. Constraints of solubility must have affected renal adaptation to water conservation and mechanisms for maintenance of blood pH, because the solubility laws are prebiological. More compelling, a urine pH near 6 would seem almost a necessary choice for the normal state, because the human design includes excretion of both considerable amounts of calcium, phosphorus, and UA. When plots of CaP and UA SS as a function of urine pH are combined as in Fig. 67.5 , a pH about 6 is clearly near a double SS minimum. Among our samples, the mean and median pH were 6.005, a value that represents the actual compromise reached between net acid excretion, ammonium ion excretion, and reabsorption of filtered organic anions such as citrate. This pH effectively removes UA SS as a characteristic of the 24-hour urine.
Daily acid production, from formation of sulfuric acid during protein metabolism, is balanced by proton titration of urine phosphorus, and ammonium excretion. The amounts of urine phosphorus are set mainly by diet intake, although acid loading leads to bone mineral resorption with extra phosphorus availability. Titration of urine phosphorus to the dihydrogen form removes phosphorus from interaction with calcium, which is how low pH lowers CaP SS. Falling pH titrates urate to the dihydrogen form (pK 5.35), which has a limited solubility of 98 mg/l in urine. To keep urine pH near 6, as clearly occurs, ammonia formation in response to acid load must be nicely balanced with phosphorus excretion and proton secretion, otherwise urine pH would run too high or too low. We presume evolution has favored this compromise pH, and suggest the subject may be worth serious experimental attention.
Organic anion reabsorption is increased by acid loading and depressed by alkali loading so that response to alkaline ash diet by the kidney need not result in as great an increase of urine pH as would be required by excretion of bicarbonate as the sole response. Citrate, the most well known of the urine anions, binds urine calcium in a soluble salt, so that increase of pH that does occur is offset by reduced calcium ion activity. The effects of other urine anions on calcium complexation and urine free calcium ion activity are not known. Acid loading lowers urine anions and therefore favors calcium ion activity increase, but abolishes calcium phosphate SS through titration of monohydrogen phosphate to the dihydrogen form. The total response of urine anions to acid-base balance change strongly suggests an evolved defense against CaP crystallization.
Urine SS is influenced by the free-ion activities of the relevant ligands that make up the stone-forming material. As such, it is reduced by high ionic strength that reduces activity coefficients, and therefore very low urine sodium and potassium concentrations raise SS, all other factors being held constant. Oxalate forms salts with magnesium, sodium, and other urine ions, and the ion activities of each of its ligands affect CaOx SS. Likewise, calcium forms phosphate, sulfate, oxalate, citrate, and other salts, so its free-ion activity also varies with multiple ligand concentrations. Even so, the effects of raw calcium excretion are dramatic, and clinically evident, whereas effects of magnesium and sodium are usually not so because the range of values encountered in clinical practice is too narrow to permit important effects. This is a critical point. Clinical reality concerns what actually happens. Some uncommon patients, for example, because of bowel disease, have almost no urine sodium, or magnesium; among their ranks are a few in whom sodium administration or magnesium loading will lower SS and improve stone prevention. We have emphasized volume, pH, and excretion of calcium because they act generally, and with considerable effect, in most patients.
Relationships between Supersaturation and Upper Limit of Metastability
As first described, the ULM would seem a property of solution, as is SS, and between them no obvious connections come to immediate mind. However, we have already shown how among healthy people ULM and SS seem linked ( Fig. 67.1 ), and this also occurs in patients with stones but to a lesser extent than in non-stone formers. Among healthy women and men, ULM for CaP and for CaOx rises with SS, as seen in Fig 67.1 . Among patients with stones the distance above the line of identity, which essentially gauges the defense against crystal formation, is reduced compared with that of healthy individuals. That the ULM to SS distance is smaller for CaP than CaOx, in healthy individuals and patients, suggests that CaP is more likely to form as an initial phase, making a starter template on which CaOx can overgrow.
Inhibitors of Crystallization and Cell Crystal Interactions
Urine certainly slows formation and growth of CaOx and CaP crystals, and we presume that at least late nephron tubule fluid must do the same. We do not know which molecules are primarily responsible for this phenomenon or if defects of inhibition cause stone disease. Perhaps the best evidence for a pathogenetic role of reduced inhibition is the low ULM to SS distance already alluded to in stone formers versus healthy individuals, and the molecules responsible for that effect are unknown. Presently we do not use measurements of individual inhibitors or of urine inhibition in clinical practice. If a commercially practical ULM measurement could be produced, this type of assay might be found useful in predicting the course of treated stone formers, perhaps, but that is mere speculation. At the moment, and when we first wrote this chapter, the whole matter of inhibitors is of mainly theoretical interest.
Calcium Oxalate Molar Ratio
We have found that the calcium/oxalate molar ratio correlated strongly with the ULM for CaOx. However calcium and oxalate molarity both correlate strongly with SS and therefore with ULM so we are concerned that inferences from regression analysis may be misleading.
Because it binds calcium in a soluble complex, citrate is clearly a protection against stone formation, and low citrate a risk factor for stones. At urine concentrations of mM levels, citrate can reduce crystal aggregation. In vitro , citrate inhibits CaOx growth by 33% at 0.5 mM, and inhibits CaP crystallization at concentrations between 0.5 and 2 mM. Adding 2 to 3 mM of citrate to urine decreases the number and size of crystal particles that form when calcium and oxalate are added to increase SS. In urine, citrate increases the ULM for CaP. These data support the role of citrate as a treatment for calcium stones in patients with hypocitraturia.
Pyrophosphate and Phytate
A long and complex history of unfulfilled promise shadows these small and potent molecules. At 16 µM, pyrophosphate inhibits growth of CaOx crystals in vitro , and urine levels of it average 20 to 40 µM. Urine levels of pyrophosphate can vary over a 10-fold range, as well, so clinical events could arise from its deficiency. On the other hand, dialysis of urine does not much reduce growth inhibition effects, suggesting extreme molecular redundancy, and efforts to raise urine pyrophosphate by giving oral orthophosphate supplements do not reduce stone formation significantly. Another small molecule of related character, phytate, inhibits brushite and apatite crystallization; but whereas pyrophosphate inhibited apatite most effectively, phytate inhibited brushite crystallization most effectively. Perhaps modern methods for quantifying these molecules could be important in understanding why some patients with CaP stones produce apatite, others brushite crystals.
Macromolecules and Small Molecules Combined
In addition to the small molecules described above, urine contains a variety of macromolecules, including proteins and glycosaminoglycans, which are active against crystals, and could, in the aggregate, explain the extraordinary ability of urine to prevent crystal growth, aggregation, and nucleation. Presently we recognize at least 10 molecules that have anti-crystal properties, but the links between them and clinical stone disease are tenuous and suggestive: THP, chondroitin sulfate, citrate, calgranulin, osteopontin, bikunin (inter-α-trypsin inhibitor), prothrombin F1 fragment, heparan sulfate, the complement inhibitor CD59, and inorganic pyrophosphate. These molecules can inhibit to various degrees the nucleation, growth and aggregation of CaOx and CaP crystals, and their ability to bind to cells, at concentrations found in urine. A few studies suggest impaired inhibition may play a role in certain stone formers.
Such a system implies extreme redundancy, perhaps an intentional result of evolutionary biology, perhaps fortunate chance. We omit discussion here of the specialized molecular structures involved in crystal inhibition, but recent work using cystine crystals suggests that understanding the molecular basis of crystal inhibition may eventually lead to the ability to design effective inhibitors for some types of stones.
Disorders of Calcium Stone Formation
Idiopathic Calcium Stones: Metabolic Abnormalities
Most calcium stone formers have no systemic disease, and most diagnostic evaluation and treatment concerns urine abnormalities that can increase SS, the established driving force for crystal nucleation and growth: high calcium and oxalate excretions, low citrate excretion, high urine pH and low volume. Calcium excretion of patients ( Fig. 67.6 , upper right panel) frequently exceeds the upper 95th percentiles for normal women (indented) and normal men. Deviations are less marked but still obvious for oxalate and citrate excretions (lower panels). Urine volume of patients overlaps with normals (upper middle panel). Urine pH of CaP stone formers exceeds normals, that of UA stone formers (discussed in a later section) is below normal.
Idiopathic Hypercalciuria (IH)
The common definition of IH is urine calcium excretion (mg/day) above 250 in women and 300 in men, values at about the 80th percentiles for healthy individuals ( Fig. 67.6 ); serum calcium is not elevated, and known causes of hypercalciuria such as hyperthyroidism, Cushing’s syndrome, excess vitamin D intake, sarcoidosis, glucocorticoid use or other systemic diseases are absent. IH is found in 30–60% of ICSF.
Like hypertension, IH is familial, being found in about 50% of first-degree relatives of probands for both sexes. Although IH is clearly genetic, studies of the monogenic hypercalciuric states, in which nephrocalcinosis and/or stones may be found, have not cast much light on pathogenesis of IH, although they offer intriguing clues. Dent’s disease appears to drive hypercalciuria via increased calcitriol levels, as in IH. Bartter syndromes involve defects of thick ascending limb calcium transport. Autosomal dominant hypocalcemic hypercalciuria arises from a gain of function mutation in the cell surface calcium receptor (CaSR), and stones and hypercalciuria are abetted mainly by vitamin D and calcium repletion to control hypocalcemia. Familial hypomagnesemia is caused by defects of paracellin-1. However, studies of patients with IH have not found a significant contribution to pathogenesis by mutations in these genes, nor by mutations in the vitamin D receptor (VDR), or the distal tubule apical calcium channel TRPV5. Patients with IH represent one end of the natural distribution of calcium excretion, and hypercalciuria results from the interplay of genetic endowment with environmental triggers.
The extra urine calcium in IH comes from diet, bone or both. In five studies ( Fig. 67.7 , left panel), one of them containing three separate comparisons calcium absorption by IH patients exceeded normal. High levels of serum calcitriol are common in IH ( Fig. 67.7 , right panel), and this hormone could raise calcium absorption rates. However, many patients with IH have high gut calcium absorption rates with normal calcitriol levels. In this, they resemble an inbred rat colony whose members absorb calcium at high rates because of an over expression of the VDR by intestine. By inference, some human IH could arise in such a way, and one study presents evidence supporting this mechanism.
Another feature of IH compatible with calcitriol stimulation of target tissues is a tendency for bone mineral to be lost during low-calcium diet challenge. In both normal subjects and patients with IH, urine calcium rises as gut calcium absorption (calculated as [diet calcium-fecal calcium]) increases, but patients with IH excrete more calcium than do normal subjects at any level of net calcium absorption, and their urine calcium often exceeds the amount of dietary calcium absorbed. This latter is of great clinical importance, since it means that bone mineral is being lost in the urine, and implies that renal calcium conservation must be impaired. Administration of oral calcitriol to normal subjects can replicate features of IH, particularly the tendency to go into negative calcium balance on a low calcium diet.
Numerous cross sectional studies have found that bone mineral density (BMD) is often decreased in IH and BMD correlates inversely with the level of urine calcium. In a prospective study, 46 subjects with hypercalciuria had BMD re-measured after three years of follow-up, and urine calcium excretion was predictive of the change in femoral z score—the higher the urine calcium excretion, the greater the drop in femoral neck mineral density. Risk of fracture is increased in IH as well, as documented by a 19-year follow-up of a cohort in Rochester, Minnesota, which found a four-fold increase in vertebral fractures among stone formers. Abnormal bone histomorphometry has been found in eight studies of IH to date, with generally increased resorption, low bone formation, and a mineralization defect. The mechanisms are still unclear, but a recent report found increased bone expression of receptor activator of nuclear factor kappaB ligand and osteoprotegerin in patients with IH versus control subjects.
How the extra calcium gets into the urine is now reasonably well known. Potential models were increased filtered load from post-prandial increase of serum calcium, reduced renal tubule calcium reabsorption, or both; in fact, reduced tubule reabsorption mediates hypercalciuria. Fasting and with meals, fractional excretion of calcium by IH exceeds controls, but ultra-filterable calcium concentrations and calcium filtered loads do not differ ( Fig. 67.8 ). Using endogenous lithium clearance as a marker, proximal tubule (PT) reabsorption falls with meals more in subjects with IH than in normal subjects, and the delivery of calcium out of PT is crudely proportional to the magnitude of urine calcium excretion ( Fig. 67.9 ); the increased sodium (Na) delivered out of PT is quantitatively reabsorbed distally, so that no difference was seen in Na excretion on the controlled diet. Whether other tubule sites also participate in the decreased calcium reabsorption seen in IH is still unknown. This work sheds light on the mechanisms for two successful treatments for recurrent calcium stones; both low sodium diet and thiazide, which reduce urine calcium in IH, could work, in part, by raising PT reabsorption.
Although IH involves accelerated calcium transport in gut, kidney and bone, diet plays a role in promoting calcium loss in these subjects. For example, feeding 100 gm. of glucose reduces renal calcium reabsorption to a much greater extent in IH than in normal subjects, contributing to the negative calcium balance in these patients. High salt intake is also associated with increased calcium in the urine. However, higher dietary calcium intake was associated with decreased risk of stone formation in epidemiologic studies. Because of that, and the risk of bone mineral loss on low calcium intakes, low calcium diet is no longer used for treatment of kidney stones.
Having said what one can about pathogenesis and how much is left to discover, therapy is rather well developed and well tested. Most current treatments to prevent stones have as their goal lowering urine SS. A general consensus exists that, in the absence of systemic disease, a first calcium stone can be treated conservatively with increased fluid. A randomized trial in such patients showed a significant decrease in recurrent stones in those patients who increased urine output to over two liters daily. Water is the optimal beverage, but a variety of fluids can be used, if those that are high in sugar or sodium are avoided.
In patients who relapse, additional therapy should be added to fluid intake. Diet modification is one strategy. A randomized trial tested the hypothesis that a diet with a normal calcium intake (30 mmol/day) but low in sodium (50 mmol) and animal protein (52 gm/day) would prevent stones better than a low calcium-low oxalate diet in hypercalciuric men. After 5 years of follow up, significantly fewer men on the normal calcium diet had recurrent stones compared to those on low calcium diet (20 vs. 38%, p=0.04). A second trial of low protein diet for stone prevention was negative, but sodium was not restricted in that study. A diet study has not been done in women, or replicated in other sites. In the United States, the DASH-sodium diet, originally created to lower blood pressure, has many of the features of the study diet used to prevent stones, after elimination of nuts, and was associated with fewer stone episodes in epidemiologic studies, but no trial has validated it as a method of stone prevention.
As a kind of byproduct of their original design, thiazide-type diuretics reduce urine calcium loss and stone recurrence. The mechanism of the reduced calcium loss is probably increased PT calcium reabsorption, with shunting of re-absorbed calcium into bone, as opposed to urine and the effect lasts as long as the drug is taken. Three prospective controlled trials of thiazide have been carried out for three years each ( Fig. 67.10 ) and yielded the same result: placebo-treated patients (open symbols) relapsed at a higher rate than corresponding thiazide-treated patients (closed symbols), the effect reaching statistical significance by the end of year three. Being protective of bone in a disease that can deplete bone of its mineral, and protective against stones in recurrent stone formers, thiazide is the primary treatment for IH. Our preference is for chlorthalidone, because it is long acting and used once daily, although hydrochlorothiazide and indapamide have also been used successfully. Very low doses of 12.5 mg can be effective, and we often begin there; 25 mg is effective for most people. Kaliuresis with potassium depletion is staved off with reduced sodium (100 mEq daily is a good goal) and increased potassium intake, or with amiloride 5 to 10 mg every morning or a potassium supplement.
We are often asked what to do when someone who is in need of stone prevention from IH cannot take thiazide. Use of the diet detailed above, with careful follow-up to ensure compliance, can be helpful. Amiloride alone will not work. Since oral glucose raises urine calcium in IH, avoidance of sugar loads is prudent. Often thiazide side effects arise from too high a dose. Sometimes “drug allergies” are a myth or may be overcome by change to another molecular variant.
Low Urine Citrate
Stone formers, especially females, excrete less citrate than healthy people of the same sex ( Fig. 67.6 ). Citrate inhibits nucleation and growth of calcium crystals, and binds calcium in a soluble complex thereby reducing SS values. The administration of alkali may also have beneficial effects on the skeleton, and can lower urine calcium, although long term data on these effects in stone formers is absent.
The cause of hypocitraturia is usually not apparent. Urine citrate excretion is positively correlated with net GI alkali absorption, which in most cases is provided by fruits and vegetables, and inadequate intake of these foods results in hypocitraturia in some idiopathic stone formers. Filtered citrate is mainly reabsorbed in PT, and citrate in urine is that which escapes reabsorption. Reabsorption is decreased by alkali and increased by acid loads and potassium depletion. Administration of potassium citrate results in metabolism of most of the absorbed citrate to bicarbonate, which allows more filtered citrate to escape reabsorption in PT; urine pH also rises. If only citric acid is administered, a transient increase in urine citrate will occur as citrate that escaped metabolism in the liver is filtered, but metabolized citrate will titrate the ingested proton and no net gain of alkali will occur. Therefore, potassium citrate or bicarbonate will raise urine citrate, but citric acid will not. This may account for the inconsistent results with administration of lemon juice as a substitute for potassium citrate.
What is the best course of action for physicians? We propose measuring citrate as part of every 24-hour urine analysis for stone diagnosis, and raising it when it is low, using a potassium citrate salt. Two trials document efficacy of supplemental potassium citrate in reducing stone recurrence in hypocitraturic calcium stone formers ( Fig. 67.11 ). Dosing should be about 2/3 of urine ammonia excretion. Care should be taken to monitor urine pH, and avoid raising it significantly above 6.2, as CaP SS may increase, and lead to worsened, not improved, stone formation. When thiazide is used for hypercalciuria, and the inevitable fall occurs in blood potassium, urine citrate will fall, and a small but useful trial has shown that potassium citrate snuffs out the otherwise stubborn recurrences of stones, presumed to form because of urine citrate deficiency. Given a reasonable action, no physician needs to despair at the incomplete understanding evident in this tiny area of science. Nor, given better science, is a better treatment likely to be found.
High Urine Oxalate
Urine oxalate may arise from endogenous production in the liver, as an end-product of glyoxalate, amino acid and ascorbic acid metabolism, and from GI absorption of preformed oxalate. Dietary oxalate is absorbed mainly in the colon, and normally about 10% of ingested oxalate is absorbed. Oxalate from both sources has only one fate—renal excretion. Under ordinary circumstances, approximately half of urine oxalate is derived from the diet, and the other half is endogenously produced. Recent experiments using mice in whom the oxalate transporter SLC26A6 was knocked out have led to the understanding that oxalate is also secreted into the gut, and increased urine oxalate may also result from a drop in gut secretion; this has not yet been documented in humans.
Higher urine oxalate values are often encountered in stone formers versus normal subjects ( Fig. 67.6 ) and should be given proper attention. Mild hyperoxaluria (values from 40 to 60 mg/d) may result from a diet high in oxalate or more commonly from a diet low in calcium, as well as from protein loads. Ascorbic acid intake may also contribute significantly to urine oxalate. Oxalate degrading bacteria are normally present in the gut, and absence of such flora may also lead to increased gut oxalate absorption.
Recent work suggests that renal oxalate handling may differ between stone formers and normal subjects. In calcium stone formers on a low oxalate diet, urine oxalate was higher than among a group of normal subjects on the same diet. Plasma oxalate levels did not differ between the two groups, but while oxalate reabsorption was seen in the normal subjects, oxalate secretion was frequently seen in the stone patients, at similar levels of plasma oxalate. This suggests that control of oxalate transport in the kidney may provide stability of plasma oxalate levels.
Very high levels of urine oxalate, of 80 mg/d or more, occur because of small bowel malabsorption or primary oxalate over-production (see below). These are very serious conditions that can damage kidneys and demand a high level of clinical attention.
This strangely combative area of nephrolithiasis hardly deserves the attention it has had if one maintains a purely clinical perspective. The idea that changes in urine oxalate influence CaOx SS more than corresponding changes in urine calcium has been shown to be incorrect. As we have shown, the range of urine oxalate excretions is much smaller than that for urine calcium or volume so oxalate increase as a cause of stones is modest in practice.
Patients with dietary hyperoxaluria should restrict dietary oxalate sources, avoid ascorbic acid supplementation, and ensure a normal calcium intake (800–1200 mg/day) from food. A recent three month pilot study showed that in a group of stone patients with mild hyperoxaluria, urine oxalate fell significantly more with a low sodium, low animal protein, normal calcium (1200 mg) diet compared with a low oxalate diet. This is consistent with findings that stone patients do not appear to have higher oxalate intakes than normal subjects.
High Urine Uric Acid
One of us was not altogether wrong in 1973 describing a tendency toward high urine uric acid excretion (750 and 800 mg/d, women and men, respectively) among patients who formed calcium stones. High levels of uric acid in urine salt out CaOx salts and reduce their solubility. This could foster CaOx overgrowths on Randall’s plaque. Allopurinol, in one prospective controlled trial, reduced CaOx stone recurrence, adding a critical support to a supposed pathogenetic mechanism.
The source of the extra urine uric acid is usually diet that contains above a pound of beef, chicken and fish—combined—daily. Treatment could be as simple as reducing the total to two-thirds pound, not what one would imagine as a low-protein diet. No diet trial has been undertaken. Allopurinol is effective by trial, but can cause toxic epidermolysis. If stones are persistent and surgically active, diet is of no avail, and high urine uric acid is the main problem, then 100 to 300 mg of the drug daily is reasonable.
Idiopathic Calcium Stones: Pathology
Idiopathic Calcium Oxalate Stone Formation: Role of Interstitial Plaque
These most common stone formers have so distinctive a pattern of clinical presentation, altered physiology, renal histopathology and mechanisms for stone formation they deserve a special place and an extensive presentation. No other kind of patient shares their distinctive features. Like all idiopathic calcium stone formers they have no systemic disease, and display a high prevalence of IH. Their stones are predominantly CaOx in composition and form as overgrowths on the renal papillae.
During percutaneous nephrolithotomy (PNL) or ureteroscopy (URS) stones are found firmly attached to the papillary tips over regions of interstitial apatite plaque, which appear visually as white sub-urothelial deposits ( Fig. 67.12 , panels A and B). Plaque forms in the basement membranes of thin limbs of Henle’s loop ( Fig. 67.12 , panel C) as micro-particles within which alternating lamina of crystal and an organic matrix form a tree-ring pattern ( Fig. 67.12 panels D and E). Particles fuse together in the papillary interstitium to form a syncytium ( Fig. 67.12 , panel D) in which islands of crystals float in an organic sea. The material accumulates beneath the basement membranes of inner medullary collecting ducts (IMCD) ( Fig. 67.12 , panel F) and Bellini ducts (BD), and under the urothelium, where it appears as white plaque. The plaque crystal is biological apatite, and the organic layers contain osteopontin and the third heavy chain of the inter alpha trypsin inhibitor, both known inhibitors of CaOx and CaP nucleation and growth. In 30 consecutive ICSF, minute inspection of serial sections from papillary biopsies revealed no crystals in any epithelial compartment, only interstitial plaque.