NDT Advance Access originally published online on April 4, 2006
Nephrology Dialysis Transplantation 2006 21(7):1870-1875; doi:10.1093/ndt/gfl067
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author [2006]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Original Articles: Clinical Nephrology
Factors increasing the risk for stone formation in adult patients with cystic fibrosis
Department of Nephrology and Renal Stone Centre, Pellegrini Hospital, Napoli, 1 Cystic Fibrosis Centre, Campania Regional Adult Unit, University Federico II, Naples and 2 Department of Nephrology and Renal Stone Centre, Mauriziano Hospital, Torino, Italy
Correspondence and offprint requests to: Dr Maurizio Terribile, MD, Centro Calcolosi Renale, Dipartimento di Nefrologia, Ospedale dei Pellegrini, Via Portamedina 41, 80134 Napoli, Italia. Email: maurizio.terribile{at}fastwebnet.it
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsDiscussionReferences |
|---|
Background. Patients with cystic fibrosis (CF) are at high risk of nephrolithiasis (NL), but controversy still exists in terms of causes, including low urine output, hypercalciuria, hyperoxaluria, hyperuricosuria and hypocitraturia. Moreover, heterozygotes (H-CF), which may exhibit altered renal concentrating and diluting ability, have never studied so far. We, therefore, evaluated the metabolic and physicochemical data of adult CF and H-CF patients, comparing them to controls (C).
Methods. Twenty-nine CF patients (16 females, aged 28.4±7.1 years), 20 H-CF (12 females, aged 58.6±6.3 years) and 30 C (19 females, aged 39.1±11.5 years) underwent kidney ultrasound and metabolic evaluation to assess stone risk profile.
Results. There was a 21% prevalence of NL in CF vs 15% in H-CF. The CF group had elevated uric acid, but no other serological differences compared with the H-CF and C group. Conversely, the citrate and oxalate content in the urine differed significantly, being lower and higher, respectively. These changes held after correction for urine creatinine. Consequently, urine specimens were more supersaturated with calcium oxalate, despite exhibiting no differences for other relevant constituents. Uric acid increased only after normalization for the body weight and urine creatinine. Lower urine volume and more acidic pH produced mild supersaturation with uric acid in samples from CF, while urine from both H-CF and C remained undersaturated. H-CF had only minor increases in both urine oxalate and calcium oxalate supersaturation.
Conclusions. This study confirms a high prevalence of kidney stones among CF patients associated with supersaturated urine. Their longer survival justifies diets and/or medications aimed at reducing the risk of forming stones.
Keywords: calcium oxalate; cystic fibrosis; hyperoxaluria; hyperuricosuria; hypocitraturia; nephrolithiasis
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Cystic fibrosis (CF) is the most common autosomal recessive disease with fatal outcome in Caucasians, with a prevalence of one in 2500 live births. Advances in the treatment and management of respiratory and pancreatic disorders have dramatically increased the life expectancy of patients with CF. In fact, the average life span has gradually increased from 2 to >30 years. Along with this increased life expectancy, a variety of nutritional, endocrinal and bone diseases are becoming more frequent, including malnutrition, poor growth, pubertal development, CF-related diabetes and osteoporosis. Some recently published studies have also shown an increased prevalence of nephrolithiasis (NL) and nephrocalcinosis (NC), but the reasons are still controversial [1,2]. Several factors promoting kidney stones have been reported to occur in patients with CF, including urinary abnormalities such as low volume [3], hypercalciuria, hypocitraturia [3], hyperoxaluria [4,5], and hyperuricosuria [6,7]. Despite a wide agreement on the association between kidney stones and CF, some doubts still exist about the mechanisms for this association, because of the differences in epidemiology, study design and the number of patients enrolled.
Heterozygous subjects (H-CF) for the cystic fibrosis transmembrane receptor (CFTR) have been reported to have some abnormalities, not only at the respiratory level [8], but also in the kidney, including impaired renal concentrating and diluting ability [9]. No literature data exist concerning the lithogenic risk in this subset.
The purpose of the present study was to assess the risk of NL in adult patients with CF and to compare them with H-CF subjects and healthy controls (C). This was accomplished by carrying out a thorough metabolic evaluation, including the calculation of the state of saturation of urine with respect to stone-forming salts.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Twenty-nine CF patients, 20 H-CF parents of CF patients plus 30 healthy volunteers (C) were enrolled (Table 1). The C subjects had been previously submitted to an ultrasound of the kidney and urinary tract, which excluded the presence of NL. At the time they were studied, all CF patients were receiving daily antibiotics, vitamins, mucolithics and high-energy supplements. They had various degrees of disease severity and most had pseudomonas pulmonary colonization. Since the majority of them had pancreatic insufficiency, they were routinely given pancreatic enzymes. While none of these patients had signs of overt and severe malabsorption, some degree of steatorrhoea was reported by 22/29 (76%) of them. Chronic respiratory insufficiency occurred in 9/29 (69%) of the patients. These patients were on follow-up in a committed regional CF care centre—adult unit, that adheres to standard guidelines established by the CF Foundation. Genotype analysis was performed on these patients who were classified according to the type of mutation. We excluded patients taking drugs or having intercurrent disorders that might influence the excretion of lithogenic or inhibitory substances, or both.
|
Renal function was normal in all the subjects under study (Table 2). All were studied as outpatients according to a standard stone risk profile for complete metabolic evaluation including serum electrolytes and 24 h urinalysis. Arterial blood gas analysis was only performed in CF patients. Urine was analysed for creatinine, sodium, potassium, calcium, magnesium, ammonia, citrate, oxalate, chloride, phosphorus, inorganic sulphate, urate and pH. These data were entered into a computer-based program to calculate urine saturation with respect to calcium oxalate monohydrate (ßCaOx), calcium phosphate dihydrate (brushite, ßbsh) and uric acid (ßUA) [10]. All the subjects were studied while consuming their self-selected home diets. Detailed dietary inquiries were not performed routinely. Therefore, an estimate of the main nutrients was derived from urine parameters. Protein, salt and potassium intakes were deduced by excretions of urea, sodium and potassium, respectively. Net acid excretion, which denotes dietary ash acid, was estimated by the sum of urinary ammonium and titratable acid, and assuming as negligible bicarbonate excretion. Because all the patients were under antibiotic therapy, bacteriological analysis of stool, including the search for Oxalobacter formigenes or other oxalate degrading bacteria, was not carried out.
|
Plain-film X-rays (CF patients) and ultrasounds of abdomen (all the subjects under study) were carried out to assess the presence of stones.
We obtained an informed consent from each patient. All the data were expressed as means±SD. Means were compared by using the non-parametric Wilcoxon or Mann-Whitney tests, for paired or unpaired data, respectively. The chi-square test was used to assess differences in incidence of chemical abnormalities. A probability index of P<0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
CF patients were younger and had lower weight and BMI compared with the other two groups. The male to female ratios did not differ among groups (chi-square = 0.41, P = 0.815). Neverthless, to account for possible sex-dependent misinterpretation of the data, we performed preliminary comparisons between males and females for relevant parameters. The only minor differences were that urine samples from males were slightly more acidic (5.9±0.5 vs 6.2±0.7, P = 0.051) and ßUA higher (0.66±0.74 vs 1.04±0.89, P = 0.044). The overall prevalence of NL and/or NC was 6/29 patients (21%) in the CF group vs 3/20 (15%) in the H-CF group (Table 1). Table 2 shows serum chemistries: the three groups did not differ significantly as to the various measured parameters, save the uric acid concentration which was higher, though within normal ranges, in CF patients as compared with C subjects. Arterial blood gas analyses in the CF group exhibited a state of moderate hypercapnia (pCO2 45.5±6.41 mmHg), associated with slight increases in bicarbonate concentrations (27.5±5.6 mEq/l) and, consequently, normal pHs (7.41±0.20). These data were consistent with a state of moderate respiratory acidosis compensated by metabolic alkalosis.
Urinary parameters are listed in Table 3. The three groups did not differ for most constituents, whereas significant differences were seen between CF patients and both H-CF and C subjects for some of the others. Because of the relevant differences in body size between CF and both H-CF and C groups, we also expressed data as either creatinine or body weight ratios (Table 4). Calcium excretion was lower in CF patients, but this difference disappeared upon correction for body weight (uCa 2.64±1.29, 2.64±1.07, 2.63±1.16 mg/kg/day, in CF, H-CF and C, respectively) or urine creatinine (Table 4). Citrate excretion was markedly decreased in CF patients compared with the other two groups and differences were even greater after correction for the urine creatinine. Conversely, oxalate excretion was higher in CF, and persisted when factored for urine creatinine. Uric acid excretion, which was comparable when expressed as mg/24 h, became slightly higher in CF, when factored for urine creatinine but this was significant. The daily excretion of urea, a reliable index of total protein intake, was similar in the three groups. However, intake became higher in CF patients when related to body weight. A similar trend emerged for net acid excretion which was markedly higher in CF patients than in both H-CF and C groups. The aforementioned differences among groups substantially held when sex-matched comparisons were carried out.
|
|
Table 5 reports the prevalence of metabolic abnormalities that are currently seen as risk factors for urinary calculi. These have been defined according to the wide agreement reported in the literature [4,6]. Hyperoxaluria, hyperuricosuria and hypocitraturia were more common among CF patients, occurring in about 50% of them. On the other hand, hypercalciuria and low urine pH were comparable with the other groups.
|
The features of urine physicochemistry are shown in Table 6. The CF patients had lower 24 h urine output. The state of saturation, with respect to all of the examined specimens was increased, with ßCaOx>ßbsh>ßUA. Figures 1–3
depict ß values in individual subjects. It is shown that ßCaOx was higher than five (which is a critical level of supersaturation), [10] in 21/29 (72%) of CF, 11/20 of H-CF (55%) and 7/30 (23%) of C [chi-square, P<0.001 (CF vs C) and P<0.05 (H-CF vs C)]. The ßbsh exceeded two in 11/29 (40%) of CF, 14/20 (70%) of H-CF and 5/30 (16.7%) of C (P<0.001, H-CF vs C). Urine was supersaturated with uric acid in 12/29 (41%) of CF patients, 4/20 (20%) of H-CF and 7/30 (23.3%) of C (P = 0.07, CF vs C).
z
|
|
|
|
Patients with CF were subdivided according to presence or absence of stone disease. No difference was found between groups with respect to calcium, oxalate, citrate and uric acid excretion. Similarly, no differences emerged as to the state of saturation with the evaluated solid phases.
All CF patients had been subjected to mutational analysis. Distribution of mutations, homozygous rate and incidence of stone disease, are listed in Table 7. The number of patients enrolled did not allow for establishing a phenotype/genotype correlation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
As compared with age-matched individuals, patients with CF have exhibited a higher prevalence of stone disease, with stones often composed of calcium oxalate [1,2,4]. NL has been ascribed to persistence of derangements of metabolic risk factors in these patients, involving uric acid, oxalate, calcium and citrate urinary excretions. Previous studies have dealt with the occurrence and the causes of these abnormalities, supporting some of our findings, conflicting with others.
Generally speaking, difficulties in the evaluation of urine chemistry may arise due to the fact that it is not easy to establish reference ranges that can be applied to CF patients. In fact, they are younger and often underweight compared with healthy individuals, and are given high-energy and high-protein diets to prevent malnutrition. All of these problems may be the confounding factors in the interpretation of the results. In fact, conflicting literature data are reported for urate [4–7] and calcium [4–6] metabolism and excretion in CF patients. Concerning urate, controversy has arisen from dose, drug formulation and responsiveness to pancreatic extracts [11]. In regard to calcium, residual steatorrhoea due to poor compliance or unresponsiveness to pancreatic enzymes [12], vitamin D deficiency [13], altered renal handling [14], and different dietary supplies have been involved.
Contrary to the aforementioned reports, more consensus seems to exist concerning oxalate and citrate excretion. In fact, hyperoxaluria and hypocitraturia were found in a significant proportion of CF patients by many authors [3–6]. However, most of the reported results are weakened by either the lack of control subjects, or the small number and the young age of patients enrolled.
In the present study, we enrolled an acceptable number of adult CF patients, which not only were compared with a similar number of healthy controls, but also, in the case of heterozygous subjects, a population who had never been previously evaluated. The H-CF subjects, who are in approximately one out of 25 cases Caucasians [8], are in heterozygosity for CFTR mutations, and aberrant CFTR expression might explain subtle abnormalities in the renal concentrating and diluting ability seen in these subjects [9].
We chose patients’ parents rather than brothers, not only because this allowed for avoiding new costly genetic screening, but also for a higher motivation of the former. Consequently, while the male to female ratios among groups were acceptably comparable, age and body size differed significantly. This was taken into account in the interpretation of the results, since the relevant urine constituents were corrected for either creatinine or body weight.
The results of chemistry and physicochemistry indicate a substantial increase in the risk for calcium oxalate stone formation. In fact, ßCaOx was 2-fold higher in CF patients as compared with C subjects, and this derived by the presence of clear-cut abnormalities involving principally oxalate and citrate excretion, associated with lower daily urinary output. The increase in saturation degrees with calcium oxalate encompassed part of the overall risk of stone formation, because the decrease in urine citrate concentration can be predicted to result in a decrease of the inhibitor activity.
As mentioned above, urine volume was significantly lower in CF patients, and this contributed to the higher saturation degrees. This meagre urine output may be explained by the increased water losses through the skin and lungs by evaporation. Excessive loss of sodium-rich sweat is a hallmark of the disease. Because of greater sweat sodium concentration, there is no increase in serum osmolality and, consequently, thirst is stimulated less, and less oral fluid is taken. The ultimate result is that fluid intake and urine output, do not take place with sensible and insensible water losses.
However, the differences in ßCaOx between the CF and C groups held after adjusting for urine volume (mean ßCaOx = 6.06, after setting urine volume at the same levels of C). Conversely, ßbsh and ßUA restored to levels comparable with the C subjects (ßbsh = 1.21 and ßUA = 0.52). In other words, the higher risk of stone formation could be offset by simply increasing urine output in the case of brushite and uric acid, but not calcium oxalate. Hyperoxaluria was confirmed as a major determinant of the stone forming risk in CF: oxalate excretion was significantly increased in these patients and hyperoxaluria occurred in about 50% of them. Hyperoxaluria is often intestinal in origin and associated to malabsorptive states with steatorrhoea. This was the case in a recent report by Hoppe et al. [15], whose patients were studied by means of an absorption test using [13C2] oxalate. In that study, however, they failed to find a significant correlation with urinary oxalate excretion. In addition to steatorrhoea, which was partly offset in our patients by the use of pancreatic enzymes, other mechanisms may explain intestinal hyperabsorption of oxalate.
Low dietary calcium is the most frequent cause of intestinal hyperabsorption of dietary oxalate [16]. Despite being normal when factored for body weight, net calcium excretion was lower in our CF patients than C, and consequently, a dietary contribution to hyperoxaluria cannot be ruled out. Hyperoxaluria could theoretically be endogenous in origin, as suggested by increases in glycolate excretion seen in a few CF patients [6]. Severe pyridoxine deficiency may decrease alanine-glyoxylate aminotransferase (AGT) activity, thereby increasing newly oxalate generation [17]. However, severe vitamin B6 deficiency was unlikely in our CF patients given multivitamins and with no evidence of overt malnutrition (BMI ranging between 18.3 and 26.4).
Finally, the regular antibiotic treatment against pulmonary infections may have led to reduction or disappearance of oxalate degrading bacteria in the gut, especially the obligate anaerobe bacterium Oxalobacter formigenes, which is able to degrade oxalate to formiate [18]. Stool analysis was not performed in these patients. However, the fact that all were under long-term antibiotic therapy, made it highly possible for the absence of bacterial degradation of oxalate in their intestine. It is therefore plausible to conclude that hyperoxaluria of CF patients is intestinal in origin and multifactorial.
The occurrence of hyperoxaluria demands medical intervention, aimed at counteracting the underlying causes. Dose of and compliance to pancreatic extracts should be carefully optimized if steatorrhoea persists. Reduction of oxalate-rich food, though difficult to apply, and supplemental calcium, 0.5–1.0 g of calcium carbonate during meals, are able to reduce intestinal absorption and urinary excretion of oxalate [16]. Oral administration of oxalate-degrading bacteria, including Oxalobacter formigenes or probiotics [19], may be potentially useful, provided patients are not on antibiotic therapy.
The second major feature observed herein was hypocitraturia, which occurred in more than half of the patients, with citrate levels nearly half as high as those in controls. Low citrate excretion was previously reported, and recently confirmed, in CF patients and related to both hypokalaemia and acidosis [3,6,15]. In general, both chronic metabolic acidosis and potassium deficiency decrease citrate excretion, by increasing its tubular uptake and intracellular metabolism in renal cells [20]. Neither metabolic acidosis, deduced from arterial blood gas analysis, nor overt potassium deficiency, deduced from levels in serum and urine, were observed in our CF patients. However, citrate excretion was directly related to both arterial blood pH (R = 0.53, P = 0.005) and urinary potassium (R = 0.55, P = 0.002). An important concurrent cause of hypocitraturia was likely represented by elevated fixed acid production and excretion, as reflected by the increase in net and body weight adjusted acid excretion. These findings are a consequence of high (animal) protein intake that is deduced by the data of urea and phosphate excretions [20]. In conclusion, similar to hyperoxaluria, hypocitraturia resulted from different causes, eventually leading to subtle derangements of intracellular proton and potassium content. Potassium supplementation, as potassium citrate at dosages of about 1 mEq/kg b.w./day, should be recommended in CF at the risk of stone disease. Mild hyperuricaemia, hyperuricosuria, and supersaturation with uric acid were also seen in CF patients. These features are likely consequent to elevated urine intake, in keeping with the aforementioned increase in protein intake. A lower urine output and a tendency to more acidic pH in CF patients caused a high incidence of supersaturation with uric acid among CF patients. However, the contribution of hyperuricosuria and uric acid supersaturation to the risk of calcium stone formation is still an object of debate.
Finally, some of the findings in H-CF are worth mentioning. They had a moderate increase in the propensity for calcium oxalate stone crystallization in urine, depending substantially on increased oxalate excretion. Our data do not allow for drawing conclusions or hypothesizing on causes of this mild hyperoxaluria. From the elevated potassium excretion, it is tempting to speculate that their diets were more rich in oxalate containing foodstuffs. The matter of the lithogenic risk in heterozygous CF deserves further studies.
In summary, CF patients are confirmed to be at a higher risk for calcium stone formation when compared with healthy controls. The risk appears to be engendered by the combination of hyperoxaluria, hypocitraturia and low urine output. This emphasizes the need to include measures aimed at counteracting these abnormalities in the medical management of CF patients, as outlined above.
Conflict of interest statement. None declared.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
- Matthews LA, Doershuk CF, Stern RC, Resnick MI. Urolithiasis and cystic fibrosis. J Urol 1996; 155:1563–1564[CrossRef][Medline]
- Gutknecht DR: Kidney stones and cystic fibrosis. Am J Med 2001; 111: 83[CrossRef][Web of Science][Medline]
- Von Der Heiden R, Balestra AP, Bianchetti MG et al. Which factors account for renal stone formation in cystic fibrosis? Clin Nephrol 2003; 59: 160–163[Medline]
- Perez-Brayfield MR, Caplan D, Gatti JM et al. Metabolic risk factors for stone formation in patients with cystic fibrosis. J Urol 2002; 167: 480–484[CrossRef][Web of Science][Medline]
- Turner MA, Goldwater D, David TJ. Oxalate and calcium excretion in cystic fibrosis. Arch Dis Child 2000: 83: 244–247
[Abstract/Free Full Text] - Hoppe B, Hesse A, Bromme S et al. Urinary excretion substances in patients with cystic fibrosis: risk of urolithiasis? Pediatr Nephrol 1998; 12: 275–279[Medline]
- Stapleton FB, Kennedy J, Nousia-Arvanitakis S, Linshaw MA. Hyperuricosuria due to high-dose pancreatic extract therapy in cystic fibrosis. N Engl J Med 1976; 295: 246–248[Abstract]
- Romeo G, Devoto M, Galietta LJ. Why is the cystic fibrosis gene so frequent? Hum Genet 1989; 84: 1–5[CrossRef][Web of Science][Medline]
- Stephens SE, Rigden SP. Cystic fibrosis and renal disease. Paediatr Respir Rev 2002; 3: 135–138[CrossRef][Medline]
- Marangella M, Petrarulo M, Daniele PG, Sammartano S. LithoRisk: a software for calculating and visualising nephrolithiasis risk profiles. G Ital Nefrol 2002; 19: 693–698[Medline]
- Kraisinger M, Hochaus G, Stecenko A, Bowser E, Hendeles L. Clinical pharmacology of pancreatic enzymes in patients with cystic fibrosis and in vitro performance of microencapsulated formulations. J Clin Pharmacol 1994; 34: 158–166[Abstract]
- Proesmans M, De Boeck K. Omeprazole, a proton pump inhibitor, improves residual steatorrhoea in cystic fibrosis patients treated with high dose pancreatic enzymes. Eur J Pediatr 2003; 162: 760–763[CrossRef][Medline]
- Lark RK, Lester GE, Ontjes DA et al. Diminished and erratic absorption of ergocalciferol in adult cystic fibrosis patients. Am J Clin Nutr 2001; 73: 602–606
[Abstract/Free Full Text] - Bohles BS, Gebhardt B, Beeg T, Sewell AC, Solem E, Posselt G. Antibiotic treatment-induced tubular dysfunction as a risk factor for renal stone formation in cystic fibrosis. J Pediatr 2002; 140: 103–109[Medline]
- Hoppe B, von Unruth GE, Blank G et al. Absorptive hyperoxaluria leads to an increased risk for urolithiasis or nephocalcinosis in cystic fibrosis. Am J K Dis 2005; 46: 440–445[CrossRef]
- Holmes RP, Goodman HO, Assimos DG. Contribution of dietary oxalate to urinary oxalate excretion. Kidney Int 2001; 59: 270–276[CrossRef][Web of Science][Medline]
- Marangella M. The primary hyperoxalurias. Contrib Nephrol 2001; 136: 11–32
- Sidhu H, Hoppe B, Hesse A et al. Absence of Oxalobacter formigenes in cystic fibrosis patients: a risk factor for hyperoxaluria. Lancet 1998; 352: 1026–1029[CrossRef][Medline]
- Lieske JC, Goldfarb DS, De Simone C, Regnier C. Use of a probiotic to decrease enteric hyperoxaluria. Kidney Int 2005; 68: 1244–1249[CrossRef][Medline]
- Melnik JZ, Presig PA, Moe OW, Srere P, Alpern RJ. Renal cortical mitochondrial aconitase is regulated in hypo- and hypercitraturia. Kidney Int 1998; 54: 160–165[CrossRef][Web of Science][Medline]
Accepted in revised form: 6. 2.06
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||