In 1972 Greene et al. described an adult patient with hypouricemia that remitted discreetly after pyrazinamide administration and the hypouricemia was associated to hyperuricosuria and hypercalciuria.1 Two years later, Sperling et al. published data on a family of Jewish origin with the described characteristics that also had associated reduction in bone mass. The authors thought they were dealing with a new disease which they named Hypouricemia syndrome with hypercalciuria and reduced bone density (OMIM 242050).2 Subsequently, some patients with the presence of hypouricemia and hypercalciuria were described.3–6 Since 1993 we are not aware of any new published cases of this association, coinciding, approximately, with the description of the transporters involved in the proximal tubular handling of urate.

In 1994, our group presented data on several children with hypouricemia and hypercalciuria, something unusual given the rarity of the association.7 In the present work we have analyzed 8 cases that were diagnosed with hypouricemia and hypercalciuria in infancy and their evolution have been followed until adolescence. At the same time, we have tried to establish whether it is a distinct entity or a variant of idiopathic hypercalciuria (IH).

The initial and final values for age, uric acid levels, calcium/creatinine ratio and z-BMD in patients in the hypouricemia and hypercalciuria group are shown in Table 2. Likewise, this table reflects the initial values of FEUr and urate reabsorbed per 100ml of glomerular filtrate. Peak urinary osmolality was normal in all cases (Table 3). Calcitriol levels (n=7; range: 19–45pg/ml), as well as intact PTH (n=7; range: 6.1–27.9pg/ml) were normal. Five patients had undergone pyrazinamide challenge with an adequate response, such that FEUr dropped to values between 0.11 and 1.84ml/100ml GFR. The z-BMD values were less than −1 in 4/8 cases. In period 2, the calcium/creatinine ratio value remained elevated in only 3 cases and all patients had urate levels above 2mg/dl (Table 2). The z-BMD value had improved in 5 cases and decreased in the remaining 3. Three patients showed ultrasound stones and 2 microcalcifications.

In the group with hypouricemia and hypercalciuria in relation to the group with IH and normouricemia, there were no differences observed in period 1 in terms of age, body mass index, parameters evaluating renal water management, creatinine levels, calcemia and phosphatemia levels and calcium/creatinine ratio, and statistically significant differences were only observed in the parameters studying urinary excretion of urate (Table 3).

Table 4 shows the bone remodeling and BMD parameters in period 1. No differences were observed between the two groups in the deoxypyridinoline/creatinine ratio and the z-BMD value, although there were differences in osteocalcin levels.

In period 2, blood uric acid levels were still significantly lower than in the HI group, although no differences were observed in calciuria or z-BMD values. Likewise, no differences were found in the frequency of children with improved calcium/creatinine ratio or in those with improved z-BMD values. The parameters evaluating urinary excretion of urate were not available during this period (Table 5).

With a physiological blood pH, about 98% of uric acid in blood is in the form of monosodium urate. In urine, when the pH is below 5.5, uric acid is in an undissociated and highly insoluble form and can crystallize. Renal tubular hypouricemia tubulopathy (type 1: OMIM 220150 and type 2: OMIM 612076) was described in 1950 by Praetorius and Kirk.13 These authors observed that, in their patient, hyperuricosuria was very excessive, and urate clearance was higher than creatinine clearance, suggesting the existence of tubular secretion of the ion. Physiological studies performed in the following years led to the assumption that the proximal tubular handling of urate had 3 components, namely, presecretory reabsorption, tubular secretion and a postsecretory reabsorption. These mechanisms have not been confirmed; more recently the complexity and diversity of the transporters involved in proximal tubular urate reabsorption have become known using molecular biology techniques.

Specifically, renal tubular hypouricemia type 1 is caused by mutations in the SLC22A12 gene encoding the luminal urate-anion exchanger urate-anion transporter 1 (URAT1).14 These patients are characterized by showing reduced urate levels (between 0.5 and 1.5mg/dl), elevated FEUr (between 20 and 90/100ml GFR) and an attenuated uricosuria in response to probenecid and pyrazinamide.15 Type 2 renal tubular hypouricemia is caused by mutations in the SLC2A9 gene encoding the basolateral glucose transporter (GLUT9).16 Patients with this variant show even lower urate levels (less than 0.5mg/dl) and FEUr values above 150/100ml GFR.17 The response to pyrazinamide is preserved.18

In the literature related to this subject, there are cases described of renal tubular hypouricemia in which hypercalciuria was present.1–6 The most relevant data of these patients have been collected in Table 6. Some patients with this association have not been included in the table because they were affected by other diseases such as cerebral palsy,19 diabetes mellitus,20 myeloma5 and nephrotic syndrome.21 The urate levels of the patients reviewed ranged from 0.6 to 2mg/dl and the FEUr was always higher than 20/100ml GFR. The cases published by Greene et al.,1 Sperling et al.,2 Frank et al.3 and Smetana and Bar-Khayim4 appear compatible with functional alterations in the URAT exchanger given the attenuated response to pyrazinamide15 (Table 6). In addition, the 3 patients presented in 2 of the above publications3,4 also showed an attenuated response to probenecid (data not shown in Table 5). In the cases of renal tubular hypouricemia published by Gaspar et al.5 and Uribarri and Oh6, the response to pyrazinamide was adequate. This is the expected response in patients with functional alterations in the GLUT9 transporter,18 but in this circumstance urate levels are very low, a fact not observed in the patients mentioned above, so perhaps they were heterozygous carriers of a mutation with loss of function in GLUT9. In any case, from what we have reviewed it seems that hypercalciuria is not specific to either of the 2 types of renal tubular hypouricemia. So far, it has not been described that an alteration in proximal tubular handling of urate is accompanied by a reduction in renal tubular reabsorption of calcium at that level. Moreover, if this would occur, calcium could be recovered in the more distal segments of the nephron, especially in the distal tubule.

As illustrated in Table 6, it can be observed that in the 70s and 80s of the last century, there few cases published with the association of hypouricemia with hypercalciuria. The interest of our group in the subject is based on the fact that in the 1990s we had diagnosed in our hospital several patients with this rare association.7 Many of them had ancestors from the neighboring island of La Gomera. As a consequence, in a study carried out in children living in several towns on that island, we observed that they were carriers of a high frequency of hypercalciuria and very low frequency of hypouricemia,22 which does not invalidate the hypothesis that some of them could have had transient reduced blood levels of urate in the first years of life.

In the present work we have studied 8 patients diagnosed in our hospital in childhood of the association of hypouricemia with hypercalciuria. The intention is to out if they are patients with renal tubular hypouricemia with hypercalciuria or patients with IH and hypouricemia.

In recent years it has been described that IH is not a disease and, therefore, is not a tubulopathy. Hypercalciuria is most likely related to an increase in the number of vitamin D receptors, with the resulting increase in intestinal calcium absorption of calcium23 and increased bone resorption24 resulting in a discrete reduction in BMD. A z-BMD value of less than −1 is observed in 30%–40% of children25 and in 60% of adult patients with IH.26 In this sense, it is possible that some of the cases of so-called Juvenile Idiopathic Osteoporosis are actually extreme forms of IH.27,28 Recently, we have described that many patients with IH, especially women, improve their z-BMD value upon reaching adulthood.29

In our study, in the hypouricemia and hypercalciuria group as compared with the hypercalciuria group with normal urate levels, there were no differences observed in the parameters that evaluate renal management of water, calcemia and phosphatemia levels, the calcium/creatinine ratio and the z-BMD value (Tables 3–5). Statistically significant differences were only observed in parameters of urinary excretion of urate (Table 3), osteocalcin levels (Table 4) and uricemia at the end of follow-up (Table 5). It is noteworthy that many patients in both groups had normalized the calcium/creatinine ratio and improved the z-BMD value in period 2 (Table 5). In this phase, as indicated, the parameters that study urinary urate elimination were not available. This data coincides with that was previously reported by our group about the fact that some patients diagnosed with IH in childhood normalize calciuria, in addition to bone mass, in adolescence or in early adulthood.29

A relationship between urate and calcitriol levels has been described in the literature on the subject. In 1993, Vanholder et al. observed in patients with mild to moderate chronic renal failure that, after short-term administration of allopurinol, plasma urate levels were reduced and calcitriol levels increased.30 In experimental animals, Chen et al. demonstrated that hyperuricemia suppresses renal 1-α hydroxylase leading to a reduction in calcitriol concentration and higher PTH levels.31 Consequently, in patients with hypouricemia, it is possible that the absence of the inhibitory effect of intracellular uric acid on 1-alpha-hydroxylase favors an increase in calcitriol levels and the consequent occurrence of hypercalciuria.6

In summary, in some patients with renal tubular hypouricemia (FEUr greater than 20%) (Table 6) there is hypercalciuria that could be mediated by the aforementioned absence of the inhibitory effect of intracellular uric acid on 1-alpha-hydroxylase. On the other hand, some patients with IH, as observed in our series, have discrete urinary losses of urate (FEUr less than 20%), presumably at the level of the proximal tubule level, although uricemia and calciuria values tend to normalize in many cases with the increase in age and there are no differences in the z-BMD value in relation to patients with IH without hypouricemia.

Thus, our patients with hypercalciuria and hypouricemia would be affected by a variant of IH in which the proximal tubular reabsorption of urate would be discretely reduced for unknown reasons and in a way that could improve.

It should be pointed out that nine patients with hypouricemia and hypercalciuria were not included in the study because basic values necessary for diagnosis were not available. This absence of these data has a limiting effect on our results.

The authors declare that they have no conflicts of interest.