H.G. Tiselius, Department of Urology, Huddinge University Hospital and Centre for Surgical Sciences, Karolinska Institutet, Stockholm, Sweden. e-mail: firstname.lastname@example.org
Recurrent stone formation in the urinary tract is a common and important problem that must be considered in daily urological practice. With a prevalence of> 10% and an expected recurrence rate of ≈ 50%, stone disease has an important effect on the healthcare system. It is generally agreed that patients with uric acid/urate, cystine or infection stones always should be treated pharmacologically. For calcium stone formers the treatment should be chosen according to the severity of the disease. Recurrence in patients with calcium-stone disease can be prevented with general or specific dietary and drinking advice, and with pharmacological therapy. For idiopathic calcium stone formers the most convincing therapeutic effects have been reported with thiazide and alkaline citrate.
The formation of stones in the urinary tract affects 5–10% of the population in Europe and North America [1–3]. An even higher frequency has been reported from other parts of world and there are only a few geographical areas in which stone disease is rare, e.g. in Greenland and in the coastal areas of Japan . However, in regions with both black and white populations stone disease is much more frequent in the latter. In most populations the occurrence in men is two to three times higher than in women.
The annual incidence of stone formation in the industrialised world is generally considered to be 1500–2000 cases per million . The disease most commonly presents with pain, usually an episode of renal stone colic, but other types of pain, haematuria and infection might lead to the diagnosis. The stone needs to be actively removed in ≈ 25% of those affected, and hence such procedures are required in ≈ 500 patients per million [3,5].
The composition of stones in German , French , American  and Swedish  populations is shown in Fig. 1. Although the chemical composition of stones varies widely a common denominator is the high risk of recurrent stone formation when the first stone is removed, although there is considerable variation among individuals .
Over the past two or three decades there has been a dramatic development in the techniques for stone removal. Although the vast majority of stones pass spontaneously, open surgery for stone removal was previously a very common urological procedure. Currently almost all stones can be removed by non- or slightly invasive methods. Despite these achievements the problem of recurrent stone formation remains and despite considerable progress in this field, efforts to stop stone formation have so far been insufficient.
There are several factors that explain a less favourable course of disease after stone removal or spontaneous passage. For a large proportion of patients there is an incomplete understanding of the mechanism of stone formation, and in those with a plausible theoretical cause the appropriate clinical methods to diagnose the pathology, or the specific tools required to correct the abnormality or abnormalities, are not available. Almost all of the methods for preventing recurrence aim to reduce the level of supersaturation and to affect the crystallization-modifying properties. Although the methods available are sometimes quite effective they often fail because of poor patient compliance . This arises because patients have no symptoms between episodes of renal stone colic, there are no simple therapeutic measures, and sometimes the treating urologist, nephrologist or physician has the wrong attitude. Moreover, preventing recurrence must continue for a long time, occasionally for the rest of the patient's life. In some patients the therapeutic programme is less effective than expected because of insufficient understanding of the mechanism of stone formation.
The present author is well aware of the shortcomings in preventing recurrence, but a serious attempt to identify specific risk factors frequently discloses one or several abnormalities that contribute to or explain the pathological crystallization and stone formation. That many patients are currently left with residual fragments in the kidney (that might serve as a nidus for new stone formation) calls for serious attention to procedures for preventing recurrence.
This review summarizes some aspects of the modern principles of medical treatment in patients with stone formation in the urinary tract. To understand the theoretical basis of preventing recurrence, the causes of stone formation and the identification of risk factors will be assessed. However, this information is only fragmentary and for a detailed discussion of these issues other texts should be consulted [9,12–14].
FACTORS OF IMPORTANCE FOR STONE FORMATION IN THE URINARY TRACT
As is evident from Fig. 1, there is a predominance of calcium stones in patients with urinary tract concretions. These stones are composed of mixtures of calcium oxalate (CaOx) and calcium phosphate (CaP), only CaOx or rarely only CaP. Reports indicate a higher rate of recurrence in stones with a greater fraction of CaP . A detailed knowledge of how calcium stones develop in the urinary tract is still lacking and several theoretical models have been proposed. Figure 2 summarizes a tentative series of events that possibly precedes stone formation and this model might help to understand the rationale for various therapeutic regimens. Such a background is important to clarify how the treatment might interfere with intranephronic crystallization.
Experimental studies indicate that the nucleation of crystals might be promoted in nephronic urine by cell membrane fragments, which are released from destroyed proximal tubular cells. This cell degradation appears to be initiated by lipid peroxidation, caused by the production of free radicals [14,16]. Although these cell membrane fragments can bring about the nucleation of both CaOx and CaP, sufficient levels of CaOx supersaturation are usually not present until urine reaches the collecting ducts. In contrast, supersaturation with CaP might occur already in the loop of Henle [17,18] or in the distal part of the distal tubule .
As long as the precipitated crystals remain small they can probably be expelled from the nephron with no effect. Urinary macromolecules, e.g. glycoproteins, proteins and glycosaminoglycans, adherent or bound to the crystals, counteract both crystal aggregation and crystal growth [12,20], and macromolecules are also important for the adherence of crystals to the tubular cells and for crystal–cell interaction [21,22].
A deficient activity or concentration of macromolecules that inhibit crystallization, or of small molecules with these properties, results in large crystal masses. It is reasonable to assume that whereas small crystals can flow freely through the nephron, large crystals will adhere to the tubular walls. Crystals that move slowly through the system also have the time to grow large.
Several inhibitory factors modify the development of both CaP and CaOx crystals. A large precipitate of CaP that is formed above the collecting ducts and that slowly moves through the nephron will start to dissolve when it encounters a low pH in the collecting duct urine. The slowly moving and dissolving CaP crystal mass will accordingly be surrounded by a high concentration of calcium ions, which in turn create a high supersaturation with CaOx and subsequently CaOx nucleation. Further deposition of CaOx and aggregation of CaOx crystals is modified by small and large molecular inhibitors.
The interaction between crystals and cells is certainly very important. Small crystals that are internalized can be dissolved by lysosomal enzymes within the cell or by macrophages at the basolateral membrane. Large crystals might destroy the cell and with an excessive crystal volume, dissolution will be incomplete. Interstitial transport of internalized large crystals to the papilla, and trapped intratubular crystals, might thus provide a crystal surface for subsequent stone formation on the papillary tip.
Pure CaOx stones might be the result of primary nucleation of CaOx in collecting duct urine or of a heterogeneous nucleation of CaOx induced by CaP, but followed by complete dissolution of the CaP crystal phase. Mixed CaOx/CaP stones occur when CaP crystals remain in the stone. With the possible exception of brushite stones, pure CaP stones form only when the urinary pH is constantly alkaline. The latter finding supports the idea that a low collecting duct urinary pH in most stone formers is necessary for CaOx stone formation.
URIC ACID AND URATE STONES
Crystals of uric acid are precipitated when supersaturation with uric acid is high. The important factors are a high excretion of urate, a small urine volume and a low urinary pH. Uric acid can precipitate at normal urate concentrations if the pH is sufficiently low; the formation product of uric acid is 5.0 × 10−9 (mol/L)2 .
Prerequisites for ammonium urate formation are present in urine with a high concentration of urate and infection with urease-producing micro-organisms. The formation product of ammonium urate is 7.2 × 10−4 (mol/L)2. Similarly, a high urate concentration together with a high excretion of sodium cause precipitation of sodium urate; the formation product of sodium urate is 1.3 × 10−3 (mol/L)2 . Formulae on how to calculate the ion-activity products of the three urate salts were published elsewhere .
Cystine precipitates only in patients with homozygous cystinuria, and the risk of cystine stone formation occurs at concentrations of cystine> 2.5–3 mmol/L or at an ion-activity product of> 1.3 × 10−20 (mol/L)3 . The ion-activity product of cystine can be obtained from information on the concentration of cystine and the urinary pH .
Infection with urease-producing micro-organisms results in increased urinary concentrations of ammonium and carbonate ions, and gives an alkaline pH. The prerequisites are thereby fulfilled for the precipitation of both magnesium ammonium phosphate and carbonate apatite. The formation product of magnesium ammonium phosphate is 2.5 × 10−13 (mol/L)3 .
IDENTIFICATION OF RISK FACTORS
Although some general therapeutic recommendations might be useful for all kinds of stones, it is essential to assess as far as possible the nature of the disease in each patient. Only then is it possible to design a programme to prevent recurrence that is as selective and effective as possible [5,24]. Although some authors claim that nonselective treatment might be equally efficient as a specific treatment , it is theoretically more attractive to address one or several identified abnormalities that are considered important for pathological crystallization.
Whenever the stone has been analysed the information is invaluable; it is essential to decide at an early stage whether the patient has formed an infection stone (magnesium ammonium phosphate + carbonate apatite), a uric acid/urate stone, or a cystine stone. For these three groups of stone formers the mechanism of stone formation is reasonably well understood and the principles of preventing recurrence are fairly straightforward (Table 1) . In most other cases analysing the stone composition will show that the constituents are CaOx, CaP or both. Stone analysis can occasionally show unusual stone constituents in patients with certain diseases and under specific forms of pharmacological treatment. When the stone has not been analysed there are clues that can be used for a reasonably correct conclusion on stone composition:
Table 1. Recommended pharmacological treatments in patients with stone disease and various urinary abnormalities associated with increased supersaturation with the stone-forming salt. All patients are also given basic treatment with a high fluid intake and reasonable dietary recommendations.
Type of urine abnormality
High excretion of calcium
Thiazide + Mg (in both cases with K supplements)
High excretion of oxalate (mild hyperoxaluria)
Dietary restriction of oxalate
High excretion of oxalate (enteric hyperoxaluria)
Dietary restriction of oxalate
Reduce fat in diet
High excretion of oxalate (primary hyperoxaluria)
Pyridoxine can be tried
Low excretion of citrate
Potassium magnesium citrate
Low excretion of Mg
Thiazide + Mg (oxide or hydroxide)
High urinary pH (RTA)
Low urinary pH
High excretion of urate
Stone of brushite
Thiazide + Mg
Stone of uric acid
Allopurinol (if high urate in serum or urine)
Potassium citrate or
Sodium potassium citrate
Stone of cystine
Infection stone (intermittently)
Uric acid stones:
• Radiolucent on a plain abdominal film, radio-opaque on CT.
• Urinary pH is usually low.
• A high serum or plasma urate supports the diagnosis.
• Stone fragments rapidly dissolve in tromethamine solution.
• The urinary sediment becomes orange or red-brown.
• Positive Brand's test (nitroprusside reaction).
• Less radio-opaque than calcium stones.
• Typical hexagonal cystine crystals in the urinary sediment.
• Sometimes a multilayered appearance on the plain film.
• Large staghorn morphology (complete or partial).
• History of or present infection with urease-producing micro-organisms.
• Urinary pH is high (usually above 7.0).
• Typical struvite crystals (coffin-lid shape) on microscopy.
• Stone fragments dissolve in renacidin.
If none of these criteria is applicable the radio-opaque stone should be regarded as a CaOx/CaP stone. For these patients the further search for risk factors depends on their stone history and the present situation.
It is generally considered that all patients with infection, uric acid/urate and cystine stone always should be given pharmacological treatment to prevent recurrence. For calcium stone formation the course of the disease is so variable that the institution of specific treatments should be reserved for those patients in whom stone formation is most likely to continue.
For this it is valuable to know that the anticipated recurrence rate is much lower among first-time stone-formers than among those who have formed more than one stone . Moreover, residual fragments in about half of the patients might grow and cause new stones. Based on these statements, calcium stone-formers can be subgrouped in categories according to the principles in Table 2 .
Table 2. Categories of calcium stone-formers based on the history of stone formation and the presence or absence of residual stone material
Start of stone disease when aged < 25 years; stone composed of brushite; strong family history of stone formation; anatomical abnormalities, medical treatment or diseases known to be associated with an increased risk of stone formation.
First-time stone, with no residual stone or fragments
First-time stone, with residual stone or fragments
Mild recurrent stone disease (≥ 2 stones formed) with no residual stones or fragments
Mild recurrent stone disease (≥ 2 stones) with residual stones or fragments
Severe recurrent stone disease or specific risk factors*with or with no residual stones or fragments
Categories Rm and Rs cannot be clearly distinguished but are based on the number of stones that the patient has formed and on the frequency of stone formation or other problems that the stone disease has caused. Calcium stone formers from all categories should have the plasma or serum concentrations of calcium, albumin, phosphate, creatinine, urate and potassium measured. An albumin-corrected calcium level of> 2.60 indicates hypercalcaemia and hyperparathyroidism needs to be excluded by analysis of parathyroid hormone, and ideally also ionized calcium.
A complete urinary analysis is advised for the patients in categories Sres, Rmres and Rs, whereby the combined result of urinary risk factors and the individual history of stone formation provide the basis for therapeutic decisions. Urine can be collected either over 24-h or any other period of the day, and it is generally recommended that at least two samples are collected [28–30].
The present author collects urine samples during one 16 h day period and one 8 h night period. The first is collected in a bottle containing hydrochloric acid and the second with sodium azide . The 16-h sample is analysed for calcium, oxalate, citrate, magnesium, phosphate, creatinine, urea and volume. The 8 h-sample is used for pH and information on the night urine volume. In the latter collection it is also possible to add an analysis of urate, and to the 16 h sample an analysis of sodium.
The levels of each urinary variable do not primarily give useful information and the most important first step is to decide if the patient produces urine that is or might become critically supersaturated with CaOx or CaP. Approximate estimates of the ion-activity products (AP) of both can be obtained with the AP(CaOx) and AP(CaP) indexes with or without standardization. The principles on how to calculate these indices are described in detail elsewhere [9,24].
The formation products of CaOx and CaP are difficult to establish because both CaOx and CaP nucleate heterogeneously. Moreover, the urine samples only give the average APs during the collection periods. No information is obtained on the peaks of supersaturation that certainly occur and during which peaks the risk of nucleation is most pronounced. Because of this the risk-levels of the AP(CaOx) and AP(CaP) indexes have been arbitrarily set to 1.5 and 50, respectively.
Once a risk level of supersaturation has been identified, the individual urine variables can serve as a guide to which type of treatment best serves the purpose of reducing the crystallization.
SOME CONSIDERATIONS RELATED TO PREVENTING RECURRENCE IN CALCIUM STONE DISEASE
It is very important that the patient is motivated to comply with measures that in most cases will require a life-long interference with dietary and drinking habits and life-style, or adherence to a pharmacological treatment. Any patient who starts a therapeutic programme that is more extensive than just general advice should be informed of the need to continue the treatment despite having no symptoms.
There are three levels of treatment that should be considered [3,5]; the first constitutes general advice about dietary and drinking habits. Such advice is beneficial to all patients with calcium stone disease. The second constitutes specific dietary recommendations and/or drinking advice that should be based on the findings in the risk analysis. The aim of these measures is to lower the APs below the risk level without adding pharmacological treatment. The third level always includes pharmacological treatment and is entered when the measures at level two have failed or when it is obvious that other reasons for pharmacological treatment are necessary.
Pharmacological treatment is chosen when the dietary restrictions are assumed to be unsuccessful, but this does not mean that dietary advice should be omitted once pharmacological treatment has been started. Recurrent stone formers will usually benefit from a combination of pharmacological, dietary and drinking therapy. To avoid starting a life-long pharmacological treatment, if this is not absolutely necessary, it is usually feasible to start at level two whenever possible.
But when is pharmacological treatment a reasonable first-choice? There is general agreement that pharmacological treatment should be given to patients with cystine, uric acid/urate and infection stone disease. Moreover, this is also a reasonable step for patients with brushite stones because of the very high recurrence rate with this CaP crystal phase . Uric acid stones or fragments can be successfully dissolved with oral pharmacological treatment [32,33] (Table 1).
For other patients there are several clinical observations and theoretical reasons that favour powerful measures to prevent recurrence. High levels of supersaturation with CaOx and/or CaP in categories Sres and Rmres might be associated with an increased risk of new stone formation [27,34]. Extensive stone formation in category Rs also favours a pharmacological approach. The risk of recurrence is higher in patients with a large proportion of CaP in the stones  and a higher risk has been recorded for stones containing CaOx dihydrate than for those containing CaOx monohydrate .
These notes can be used as a rough guide when the treatment programme is designed, but it is always necessary to base treatment on the specific requirements of each patient.
THEORETICAL POSSIBILITIES FOR PREVENTING RECURRENCE IN PATIENTS WITH CALCIUM STONE DISEASE
In view of the various steps considered important or possibly contributing to the formation of a calcium stone, there are several therapeutic options. Clinical experience is almost exclusively restricted to measures aimed at reduced levels of supersaturation with CaOx and CaP, and the increased crystallization-inhibiting properties after giving alkaline citrate.
There is a striking lack of randomized placebo-controlled or randomized controlled studies, mostly because of the extreme difficulty in pursuing such studies for a sufficiently long period. In the average population of recurrent stone formers the annual frequency of stone formation is < 0.15–0.20 [35,36], thus no definite conclusions can be drawn from treatment periods of < 5–7 years! To maintain patients on placebo treatment for so long is not easy, particularly as compliance even among patients given active medication is surprisingly low [37,38].
Thus a rational programme for preventing recurrence cannot be based entirely on treatment programmes supported by the strict principles of evidence-based medicine. Information must be combined from the few controlled studies completed, from meta-analyses, and from reports on other studies, together with theoretical reasoning.
Several methods have been used in an attempt to affect urine composition and to reduce the risk of abnormal crystallization. Correcting excessive dietary habits together with drinking advice is the mainstay of stone prevention. The most common pharmacological agents used to affect calcium stone formation are thiazides, orthophosphate, cellulose phosphate, magnesium, alkaline phosphate and allopurinol . The influence on saturation levels and crystallization properties caused by changes in urine composition are summarized in Table 3.
Table 3. Theoretical and actual possibilities favourably affecting urinary composition and calcium stone formation
CLINICAL METHODS FOR PREVENTING RECURRENCE IN PATIENTS WITH CALCIUM STONES
Several dietary manipulations have been used to eliminate abnormalities in urinary composition or to increase the solubility of CaOx and CaP. The effects that various dietary constituents have on the composition of urine is summarized in Table 4. Although consumption of animal protein explains several abnormalities in urinary composition, there are no studies in which a diet restricted in protein has produced convincing effects on the rate of stone formation. Nevertheless, reducing an excessive intake of animal protein is recommended, and when such an approach was combined with a high fluid intake, there was a good clinical effect . Further details of the dietary effects on urinary composition are available in other texts [33,40].
Table 4. Effects of dietary constituents on urine composition
Effects on urinary composition
+ve, increased; –ve, decreased.
Oxalate +ve (wheat bran)
(but not with doses of < 2–4 g/day)
Oxalate –ve (?)
(1,25 vitamin D –ve)
Importantly, advice to restrict calcium intake should not be given to calcium stone formers unless it has been shown that they have an excessive intake of calcium. The daily intake of calcium should be 20–25 mmol (800–1000 mg) to maintain a positive calcium balance. A reduced intake of calcium leads to an increased intestinal absorption of oxalate . Supplements of calcium should always be taken together with meals.
The useful effect of urinary dilution has always been considered a mainstay in treating patients with stone formation. The value of the reduced supersaturation that is thereby achieved supervenes the negative effect caused by the associated dilution of urinary inhibitors [42,43]. Borghi et al.  in a randomized study showed the clinical value of a generous drinking regimen.
Administration of thiazides has a pronounced and well documented effect in reducing urinary calcium. There are two placebo-controlled studies with thiazides [45,46], several others with control groups, and follow-up studies of patients with no comparable controls [47–55]. In one of the randomized studies thiazide was given together with magnesium and this regimen tended to be more effective than when thiazides were given alone . It has also been claimed that thiazides can reduce urinary oxalate, but there are few convincing data on such an effect. Long-term treatment with thiazides and insufficient substitution with potassium might lead to hypokalaemia and a concomitant hypocitraturia. An extensive long-term use of thiazides is to some extent limited because of side-effects such as hypotension, weakness and impotence.
Urinary calcium is also reduced after giving orthophosphate [56–58], which increases pyrophosphate excretion, a compound with inhibitory activity on the growth of CaOx and CaP. A significant clinical effect requires large daily doses of phosphate, given at least three times a day. Such a regimen is accepted only by a few patients and side-effects such as diarrhoea and intestinal gas formation makes this form of therapy less attractive. Low-dose orthophosphate was less successful in preventing stone formation .
Binding intestinal calcium to cellulose phosphate has been used to decrease calcium absorption and calcium excretion. Although the urinary excretion can thereby be decreased the oxalate excretion increases as a result of an increased oxalate absorption [60,61]. The use of cellulose phosphate therefore cannot be recommended .
Alkaline citrate has been advised as the method of choice to increase the excretion of urinary citrate. The excretion of citrate increases as a result of the alkali load, which reduces the tubular reabsorption of citrate in the nephron. The simultaneous alkalinization and a high citrate excretion causes several effects that favourably counteract crystallization of both CaOx and CaP . In addition to the direct inhibitory effects of citrate, an increased urinary pH results in dissociation of citrate and of the inhibitory macromolecules, which increases the crystallization inhibiting power of urine.
Administration of alkali also causes reduced calcium excretion. Both sodium potassium citrate and potassium citrate alter urinary composition favourably [63–69], but the admixture of sodium might result in an increased calcium excretion or a less pronounce reduction in urinary calcium [37,70]. Clinical studies subsequently indicated that potassium citrate is superior to sodium potassium citrate in counteracting stone formation . In controlled studies with potassium citrate there was a good effect on stone prevention but the follow-up was < 3 years .
Magnesium is an inhibitor of CaP crystal growth and modifies the interconversion among different CaP crystal phases. The complexing between magnesium and oxalate decreases the AP of CaOx. Despite initially promising effects of magnesium treatment, there are no long-term studies that support an effect in preventing recurrence for either MgO or Mg(OH)2 [11,72]. This lack of effect can probably be explained by an increased excretion of calcium parallel with that of magnesium, and that a low magnesium excretion is an uncommon finding in calcium stone formers. When magnesium was given together with thiazides in the long-term randomized placebo-controlled study cited above , adding magnesium appeared to augment the effect of the thiazide. Long-term thiazide treatment can cause hypomagnesaemia and hypomagnesuria, and supplements of magnesium might thus be of value. Promising results in preventing recurrence were reported in patients treated with potassium magnesium citrate .
Reports of an apparent relationship between hyperuricosuria and calcium stone formation led to the use of allopurinol as a means by which CaOx stone formation could be reduced . The results obtained with allopurinol are conflicting and it is possible that an effect should be expected only when the urinary concentration of urate is high . The occurrence of hyperuricosuria varies among geographical areas. One possibility that should be considered in patients treated with allopurinol is the effect that it might have as an antioxidant in the nephron, thus counteracting the negative effect of free radicals. However, there are only experimental studies in support of such an effect.
The observation that stone disease does not occur among Greenland Inuits or in other populations with a high intake of fish oil has stimulated the therapeutic use of eicosapentaenoic acid as a measure to prevent stones; Buck et al.  reported an impressive clinical effect. The mechanism by which eicosapentaenoic acid might work is assumed to be the reduced excretion of calcium, probably through a direct effect on membrane efflux in tubular cells.
METHODS TO REDUCE HYPEROXALURIA
The excessive intestinal absorption of oxalate in patients with enteric hyperoxaluria can be treated by a restricted intake of oxalate-rich products with or without a diet poor in fat , administration of Oxabsorb®  (a marine hydro-colloid that forms complexes with oxalate), and calcium supplements . For the average stone former with normal or mild hyperoxaluria a small reduction can be anticipated after the dietary restriction of oxalate-rich food. In patients with no intestinal colonization by Oxalobacter formigenes less oxalate excretion was recorded after substituting this bacterial strain . However, this method has not been extensively used in patients with calcium stone disease.
Some patients with primary hyperoxaluria respond to large doses of pyridoxine . The dilution of urine is usually insufficient in these patients. Stone formation in patients with primary hyperoxaluria is extremely difficult to control and these few patients should always be referred to a physician with special knowledge of treating this condition.
GENERAL CLINICAL RECOMMENDATIONS ON PHARMACOLOGICAL TREATMENT
It is difficult to identify the most efficient, simple and rational method to prevent calcium stone recurrence, because the various treatments have usually been applied in selective or general ways. However, there is some evidence that selective therapy is better than general , and theoretically it is more attractive to direct treatment towards that or those abnormalities considered important for each patient.
An extensive review of publications and a consensus discussion held by the Advisory Board of European Urolithiasis Research and a Working Party of Lithiasis resulted in therapeutic recommendations that have been published elsewhere . The conclusion was that thiazides and alkaline citrate, together with recommendations of a high fluid intake, were the most convincing therapeutic alternatives in preventing recurrent calcium stone formation. These results were very much in agreement with the results of a meta-analysis presented by Pearle et al. . Figure 3 summarizes the proportion of patients with no recurrent stone formation during a long-term follow-up of treatment with thiazides and alkaline citrate, compared with the recurrence rate expected in untreated recurrent calcium stone formers.
In the absence of uncritical evidence-based studies, the recommended treatment programme presented in Table 1 is based on what is presently known about how different therapeutic measures affect the physicochemical and crystallization properties and the rate of stone formation.
The expert secretarial assistance by Ms Marie Karlsson is highly appreciated.