UPDATE ON DIETARY RECOMMENDATIONS AND MEDICAL TREATMENT OF RENAL STONE DISEASE

The role of calcium, oxalate, sodium, potassium and protein intake on stone formation have been intensively investigated in the last years, because of their known effect upon urinary excretion as either promoters or inhibitors of stone formation. The association between dietary factors and stone formation was first evidenced by the "stone boom" , i.e., the dramatic increase in stone disease incidence in Western industrialized nations, after World War II compared to the period during the war when malnutrition was the rule. The "stone clinic effect" , a phenomenon described by the Mayo Clinic years ago to explain the reduction of stone recurrence in 2/3 of the patients after basic dietary advice, further reinforced the importance of such association [1].

Calcium - In the past, calcium restriction became a very popular recommendation, based on the high incidence of hypercalciuria, around 50% [2], in calcium stone forming patients, its impact on calcium oxalate and phosphate saturation [3], and also because of the contribution of calcium intake and intestinal calcium hyperabsorption to hypercalciuria. An acute oral calcium load test, described in 1975 by Pak et al [4], should clearly distinguish between absorptive and renal hypercalciuria. In a previous evaluation by our group [5], a 24 hr urinary calcium excretion, under conditions of a mean usual calcium intake of 540 mg/day, was determined in CSF patients who previously presented an absorptive or renal response to this test. We observed that the majority of them, 63 and 78% of each group, presented normocalciuria rather than hypercalciuria. Since this apparently normal calcium excretion might have resulted from a combination of high calcium absorption and low calcium intake, those patients where then challenged to a higher calcium intake of 1500 mg/day given as supplement for one week. Regardless of whether there was a former absorptive or renal-like response to the acute load, the higher calcium intake disclosed the presence of subpopulations sensitive to calcium intake in previously normocalciuric patients [5].

Conversely, most of the hypercalciuric patients, when challenged to a higher calcium intake did not present a further increase in their urinary calcium, showing that under conditions of low calcium intake, as it is the case for the Brazilian population [6], patients were already excreting calcium in excess of their intake, hence being considered as dietary calcium-independent. In addition, as the morning urinary fasting calcium/creatinine (Ca/Cr) ratio seemed to be the single parameter which would distinguish between renal and absorptive hypercalciuric patients, with a cutoff value of 0.11, we repeated this determination in 31 patients [5], and found that 87% of them changed their results from values higher than 0.11 to lower values. Taken together, these data suggest that Absorptive and Renal Hypercalciuria should be considered the same rather than two distinct entities, a hypothesis already raised by Coe et al [3,7], representing two extremes of a continuum behaviour resulting from an abnormal regulation of 1,25 vitamin D.

In a large prospective epidemiological study conducted by Curhan et al [8], healthy men with different levels of calcium intake were followed-up for 8 years, and surprisingly it was observed that the lower was the calcium intake, the higher the incidence of stone formation. A hypothesis to explain such unexpected effect was a secondary increase in urinary oxalate due to a decrease binding of oxalate to calcium in the gastrointestinal tract. Nevertheless, Bushinsky et al [9], in an experimental model of genetic hypercalciuric rats, fed increasing amounts of calcium in the diet, observed a proportional increase in urinary calcium compared to the respective controls, not accompanied by a parallel decrease in urinary oxalate. In CSF patients, the effects of modifying calcium intake from 500 to 1500 mg/day upon oxalate excretion have been evaluated by our group [10]. A significant decrease of urinary oxalate has been observed in calcium dietary-independent hypercalciuric but not in dietary-dependent hypercalciuric patients.

Additional studies are still needed to further clarify whether the postulated inverse relationship between colonic oxalate absorption and colonic load of calcium, behaves differently when hypercalciuria is present. Focusing on the bone issue, we and other investigators have addressed the loss of bone mass in hypercalciuric patients [11-17], suggesting the contribution of high animal protein and sodium intake [12-16], and stressing the role of a low calcium diet in such loss [11-18]. In addition, calcium excretion is not solely affected by the intake of calcium, but other nutrients like animal protein, sodium, oxalate and potassium intake might influence calcium excretion as well [12,19,20,21].

In summary, there are many reasons why calcium restriction should be avoided in hypercalciuric patients, as listed below :

• Lack of a clear distinction between absorptive and renal hypercalciuria
• No prospective studies supported the belief that calcium restriction led to a reduction of stone recurrence
• It induces secondary hyperoxaluria
• It predisposes to bone loss due to a negative calcium balance
• In addition, chronic calcium restriction might upregulate vitamin D receptors allowing 1,25 vitamin D to stimulate both intestinal calcium absorption and bone resorption
• Other nutrients like protein, sodium, oxalate and potassium could affect calcium excretion as well

Additional advantages of a higher calcium intake could be speculated [22] : the substitution of meat protein by dairy product-derived protein may provide a higher intake of phosphate which co-precipitates with calcium in the intestinal lumen, so that urinary phosphate does not increase; since calcium and magnesium compete for a common reabsorptive mechanism in the loop of Henle, increases in urinary calcium excretion are expected to induce an increase in urinary magnesium, a known inhibitor of crystal aggregation. Nevertheless, one has to consider that the benefits of a high calcium supply do not apply to calcium supplements, which usually are not taken with meals, hence losing their oxalate chelating properties [23].

Oxalate - Aside from primary and enteric hyperoxaluria, most cases found in CSF patients are represented by "mild hyperoxaluria", defined by levels of urinary oxalate from 40 to 100 mg/day, with a reported frequency of 12 to 63% [24]. Marangella et al [25] have suggested that "mild hyperoxaluria" might be secondary to calcium hyperabsorption. The rational basis for oxalate restriction relies on the fact that calcium oxalate is the main component of most renal stones, and that there is a lower urinary oxalate content than calcium (Ca/Ox ratio is 5:1). This means that small changes in oxalate concentration have much larger effects on CaOx crystallyzation than large changes in calcium concentration. A recent experimental study by Bushinsky et al [26] has shown that increases of dietary oxalate up to 2% in hypercalciuric rats produced an elevation in oxalate excretion and a fall in urinary calcium excretion, probably due to oxalate binding intestinal calcium. In this model, since the higher induced oxaluria was offset by a lower calciuria, the net effect was a decrease of CaOx saturation ratio. These results have raised the question about the necessity, if any, of limiting dietary oxalate for stone formers. In humans, only 10 to 15% of urinary oxalate is derived from the diet [27].

Additionally, the ability of oxalate-rich foods to augment oxalate excretion depends not only on the oxalate content but also on its bioavailability, solubility and salt form. Only spinach is considered to be a high risk food item, for its high amount of bioavailable oxalate content [28]. Peanuts, instant tea, almonds, chocolate and pecans are considered as moderate risk food items [28]. Finally, the effect of dietary oxalate on urine oxalate critically depends upon calcium intake since decreasing calcium load in the intestinal lumen will increase the concentration of free oxalate anions available for absorption, as mentioned above. In healthy subjects, Hess et al [22] have recently shown that the hyperoxaluria caused by a 20-fold oxalate load can be totally prevented by a very high calcium intake of about 4 g/day. Accordingly, we have investigated if this also holds true for smaller amounts of both nutrients in CSF patients (yet unpublished data).

A slight but significant increase of about 20% on oxalate excretion after a 2-fold increase in oxalate intake was observed. However, such an increase was no longer observed when an amount of calcium (430 mg) had been concomitantly ingested. Marshall et al [29] have studied the effects of either oxalate or calcium restriction alone, as well as double restriction in stone forming patients and controls. In patients, oxalate restriction almost did not alter calcium excretion and produced only a very mild decrease in urinary oxalate. CaOx activity was not altered that much. On the other hand, a severe calcium restriction (down to 250 mg/d) caused an important elevation of urinary oxalate only when the supply of dietary oxalate was normal. The combined restriction of calcium and oxalate was the only way to prevent such oxalate elevation, leading to an effective decrease of CaOx product activity far below the formation product [29].

Bataille et al [30] evaluated the probability of stone formation after a combined restriction of calcium and oxalate, and observed that the combined restriction was not able to decrease the probability of stone formation in dietary-independent hypercalciuria patients, inasmuch as a concomitant increase in oxalate excretion was still evidenced in these patients. In summary, the idea that calcium and oxalate must be maintained in balance during meals is unquestionable, but more long-term controlled studies are still needed to answer to the question as to whether double or no restriction should be recommended. In addition, oxalate excretion also depends on oxalate degradation by anaerobic bacteria in the gastrointestinal tract . The absence of this bacteria from the gut increases the risk for hyperoxaluria [ 31].

Protein - The nutrient that clearly has universal effects on most of the urinary parameters involved in stone formation is protein. High protein intake of animal origin contributes to hyperuricosuria due to the purine overload, to hyperoxaluria due to the the higher oxalate synthesis and to hypocitraturia due to the higher citrate tubular reabsorption [32,33]. Additionally, protein-induced hypercalciuria may be caused by higher bone resorption and lower tubular calcium reabsorption to buffer the acid load, and also by the elevated calcium filtered load and by the presence of nonreabsorbable calcium sulfate in the tubular lumen [32]. An acute moderate protein restriction is able to reduce urinary oxalate, phosphate, hydroxyproline, calcium, and uric acid and to increase citrate excretion, as previously reported [34].

Potassium - An epidemiological study has reported that the lower the potassium intake, below 74 mEq/d, the higher the relative risk for stone formation [8]. Such an effect can be ascribed to an increase in urinary calcium and a decrease in urinary citrate induced by a low potassium intake [21]. In a previous series studied by us [35], a low-normal potassium intake and a higher NaCl intake were observed in stone formers when compared to healthy subjects. The overall effect was a significantly higher urine Na/K ratio [35], increasing the risk for stone formation, as previously suggested by Cirillo et al [36].

Sodium -The effect of sodium chloride (NaCl) intake on increasing calcium excretion is well established. Every 100 mmol increase in dietary sodium results in a 25 mg rise in urinary calcium [37]. The adverse effects of a high NaCl intake and the resultant higher calcium excretion have been extensively reported by many investigators [20,38,39]. In a previous evaluation by our group [40], a multiple regression analysis has suggested that a high NaCl intake (³ 16 g/day), was the single variable that was predictive of risk of low bone mineral density in 85 CSF patients (odds ratio: 3.8) after adjustments for age, weight, body mass index, duration of stone disease, calcium and protein intakes and urinary calcium citrate and uric acid. Finally, a high NaCl intake is expected to lower citrate excretion as well [41].

Fluid Intake - A high fluid intake is a very important goal to reduce urine supersaturation. A very well conducted 5-year randomized prospective study [42] involving first stone episode patients has shown lower rates of recurrence (12%) in those with a higher intake of water compared to those without (27%). It should be emphasized that patients received no drug therapy nor were submitted to any dietary change so that the unique efficacy of increasing urinary volume could be validated [42]. To what extent the hardness and mineral composition of water affect stone risk remains controversial [43-45]. As the calcium content of drinking water increases, calcium excretion increases but oxalate excretion falls [44,46]. Water with a large amount of bicarbonate may increase citrate excretion [44] and magnesium content may favorably alter citrate and magnesium excretion [47]. Based on these findings, there is still no definite evidence that hard water, rich in calcium and magnesium, is more lithogenic than soft water.

A very recent epidemiological study based on food-frequency questionnaires has examined the effects of particular beverages on risk of symptomatic kidney stones in women [48]. Consumption of tea, caffeinated and decaffeinated coffee was associated with a reduction of risk of 8 to 10% , while wine decreased the risk by 59%. Conversely, grapefruit juice ingestion was associated with a 44% increased risk for stone formation. The authors speculated that the protective effects of coffee, tea and wine were caused by urinary dilution, determined by the ability of caffeine and alcohol to inhibit antidiuretic hormone. Therefore, the decreased risk for decaffeinated coffee might have been conferred by another mechanism. The adverse effects of grapefruit juice remained unexplained, since other citrus juices, like orange and lemon, may prevent [49,50] and not stimulate stone formation due to their high citrate content. In summary, these results must still be interpreted with caution until adequate long-term randomized trials of dietary interventions are performed.

Vitamin C - The effect of large doses of vitamin C in increasing urinary oxalate excretion is controversial [51,52] and is eventually accounted for by the conversion of vitamin C to oxalate during the analytical procedure [52]. In a large epidemiological study, the intake of vitamin C was not associated with risk of kidney stones in women [53].

Therefore, the influence of diet on renal stone disease seems to be much more complex than foreseen because of multiple interactions between all of these nutrients with the distinct urinary parameters. Dietary recommendation in renal stone disease can be summarized as shown in the table below:

Dietary recommendations in renal stone disease

• Tailored diet to the specific metabolic disturbance and also to individual dietary habit, to ensure compliance.
• Calcium restriction should be avoided
• Calcium and oxalate intakes must be in balance
• Adequacy of animal protein and NaCl intake (moderate restriction)
• Potassium intake should be encouraged
• A high fluid intake to produce at least 2 liters of urine per day.

MEDICAL TREATMENT

Considering the high rate of recurrence of renal stone disease, the importance of medical preventive programs is incontrovertible. Nevertheless, there is a need for sufficient randomized double-blind placebo-controlled trials to validate the efficacy of such programs. The paucity of drug trials can be ascribed to several factors [54]: low patient compliance because of the absence of symptoms between stone episodes; stone disease is heterogeneous and the course is unpredictable, requiring long treatment periods and a follow-up of at least 5 years to show any beneficial effect; there must be a group of patients of a reasonable size and a control group with a similar risk profile to be compared to; the effect of a single drug has to be compared with placebo without associated dietary intervention which in turn might act as a potential confounder. Despite the potential merit of conservative treatment involving just diet and fluid intake modification, the so-called stone clinic effect [1], only fluid intake has been validated by a prospective study [42]. For ethical purposes, placebo groups of most trials generally receive diet and/or fluid recommendations.

Thiazide - Thiazide lowers urine calcium resulting in a fall in calcium oxalate and calcium phosphate supersaturation. Two double blind, randomized, prospective and placebo-controlled trials, one involving 25 patients given hydrochlortothiazide, 25 mg/day [55], and other involving 42 patients given chlorthalidone, 25 or 50 mg/day [56] have shown a significantly lower rate of recurrence after 3 years (up to 25%) compared to placebo (up to 55%). Interestingly, these studies were performed in patients not categorized according to urinary derangement, and response to therapy was independent of baseline urinary biochemistry. We and others [57,58] have addressed the additional benefits of thiazides on bone mass in small series. On the other hand, adverse effects, often dosage related such as sexual impotence, potassium wasting, raised serum cholesterol and glucose tolerance, are reported in almost 23% of cases [56] and thus intolerance should be kept in mind.

Allopurinol - Allopurinol blocks uric acid production, reducing heterogeneous nucleation of calcium oxalate by both uric acid and monosodium urate. In addition, the adsorption of normally occurring macromolecular inhibitors of calcium oxalate crystallization by uric acid or monosodium urate could be possibly averted when using this drug. In the sole double-blind, placebo-controlled study involving 29 subjects receiving Allopurinol 300 mg daily for 3 years, 51% had fewer recurrences than those treated with placebo [59]. Allopurinol poses a lower incidence of side effects, but the drug is effective in reducing stone recurrence only in calcium oxalate stone patients in whom hyperuricosuria is the only metabolic abnormality [60].

Potassium citrate - Potassium citrate reduces urinary saturation of calcium salts by complexing calcium and reducing ionic calcium concentration. Due to its alkalinizing effect, it also increases the dissociation of uric acid, lowers the amount of poorly soluble undissociated uric acid, reducing the propensity to form uric acid stones. The induced decline of urinary calcium during the early period of treatment [61] represents a promising additional advantage of the drug. Potassium citrate is preferable to sodium citrate in the prevention of urolithiasis [61], and the former has been shown to decrease the stone formation rate in a randomized placebo-controlled study involving 18 hypocitraturic patients who received 45 mEq/day of citrate for 3 years [62]. However, adverse effects of gastrointestinal origin including epigastric pain, abdominal distention or diarrhea are common. Promising results with the use of newer citrate salts such as potassium-magnesium, not yet approved by the Food and Drug Administration, have also been shown in patients with idiopathic calcium oxalate nephrolithiasis [63] irrespective of baseline urinary biochemistry.

Other drugs: Potassium-acid phosphate [64] and magnesium hydroxide [56] were shown to have little or no effect on prevention of stone formation. A neutral potassium phosphate preparation was shown to be better than placebo in reducing calcium excretion and raising urinary inhibitors of stone formation, hence inhibiting CaOx crystal agglomeration and spontaneous nucleation on brushite [65].

The latest evolution in the approach to calcium oxalate stone prevention is rejection of the use of a urinary metabolic profile to guide prophylaxis both because this can be time-consuming and expensive and also for the benefits of nonselective therapy, nicely reviewed elsewhere [66]. The use of drugs in patients not categorized according to different urinary derangements and the protective effects not influenced by the baseline urinary chemistry in many of the above mentioned trials further emphasizes the usefulness of such approach. Besides, in a given subject, the stone formation may not be due to a single abnormality. Overall, potassium citrate represents the most suitable drug for unselective treatment because of its indications for hypocitraturic, hypercalciuric, hyperuricosuric and renal tubular acidosis patients. On the other hand, identification of abnormal risk factors for urinary stones is still important to rule out secondary causes of nephrolithiasis, such as cystinuria, hyperoxaluria, renal tubular acidosis and infection stones. Among all of these examples, cystinuria, the most rare, represents the single entity for which an actual specific therapy with tiopronin would be warranted, despite the need for alkalinizing therapy with potassium citrate as well. The dissolution of pure uric acid stones by potassium citrate also takes place during treatment, as suggested in the present case. The single contraindication for potassium citrate would be urinary tract infection because of the alkalinizing properties of the compound.

In conclusion, well-designed large prospective epidemiological studies performed on healthy subjects have contradicted some long-held beliefs, but changes in urinary factors have not been evaluated. Adequate long-term randomized trials of dietary interventions with stone recurrence and long-term measures of urinary composition as end-points, although difficult, should be performed. Another criticism of most drug trials concerns the timing for comparison, i.e., a given patient presented a number of stones all life long, and then the efficacy in reducing the number of stones refers to a period of a few years. The real effect of various foods on lithogenicity still warrants further studies, given the complex interactions of different nutrients. The efficacy of selective versus nonselective treatment should be further compared.