Effect of conservative treatment on the renal outcome of children with primary hyperoxaluria type 1

http://www.kidney-international.org & 2009 International Society of Nephrology Effect of conservative treatment on the renal outcome of children with primary hyperoxaluria type 1 Sonia Fargue 1,Jérôme Harambat 2, Marie-France Gagnadoux 3, Michel Tsimaratos 4, Franc oise Janssen 5, Brigitte Llanas 2, Jean-Pierre Berthélémé 6, Bernard Boudailliez 7,Gérard Champion 8, Claude Guyot 9, Marie-Alice Macher 10, Hubert Nivet 11, Bruno Ranchin 1,Rémi Salomon 3, Sophie Taque 12, Marie-Odile Rolland 13 and Pierre Cochat 1 1 Service de Pédiatrie and Centre de Référence des Maladies Rénales Rares, Hospices Civils de Lyon and Université Lyon 1, Lyon, France; 2 Service de Pédiatrie, Hôpital Pellegrin, Bordeaux, France; 3 Service de Néphrologie Pédiatrique, Hôpital Necker Enfants-Malades, Paris, France; 4 Service de Pédiatrie, Hôpital La Timone, Marseille, France; 5 Service de Néphrologie Pédiatrique, Hôpital Universitaire des Enfants Reine Fabiola, Brussels, Belgium; 6 Centre Perharidy, Roscoff, France; 7 Service de Pédiatrie, Centre Hospitalier Universitaire Nord, Amiens, France; 8 Service de Pédiatrie, Centre Hospitalier Universitaire, Angers, France; 9 Clinique Pédiatrique, Centre Hospitalier Universitaire, Nantes, France; 10 Service de Néphrologie Pédiatrique, Hôpital Robert Debré, Paris, France; 11 Service de Pédiatrie, Hôpital Gatien-de- Clocheville, Université Franc ois Rabelais, Tours, France; 12 Service de Médecine de l Enfant et de l Adolescent, Hôpital Sud, Rennes, France and 13 Laboratoire de Biochimie Pédiatrique, Centre de Biologie Est, Lyon, France Primary hyperoxaluria type 1 results from alanine:glyoxylate aminotransferase deficiency. Due to genotype/phenotype heterogeneity in this autosomal recessive disorder,the renal outcome is difficult to predict in these patients and the longterm impact of conservative management in children is unknown. We report here a multicenter retrospective study on the renal outcome in 27 affected children whose biological diagnosis was based on either decreased enzyme activity or identification of mutations in the patient or his siblings. The median age at first symptoms was 2.4 years while that at initiation of conservative treatment was 4.1 years; 6 children were diagnosed upon family screening. The median follow-up was 8.7 years. At diagnosis, 15 patients had an estimated glomerular filtration rate (egfr) below 90, and 7 children already had stage 2 3 chronic kidney disease. The median baseline egfr was 74, which rose to 114 with management in the 22 patients who did not require renal replacement therapy. Overall, 20 patients had a stable egfr, however, 7 exhibited a decline in egfr of over 20 during the study period. In a Cox regression model, the only variable significantly associated with deterioration of renal function was therapeutic delay with a relative risk of 1.7 per year. Our study strongly suggests that early and aggressive conservative management may preserve renal function of compliant children with this disorder, thereby avoiding dialysis and postponing transplantation. Kidney International (2009) 76, 767 773; doi:10.1038/ki.2009.237; published online 1 July 2009 KEYWORDS: orphan disease; primary hyperoxaluria type 1; renal insufficiency; survival Correspondence: Pierre Cochat, Service de Pédiatrie, Centre de Référence des Maladies Rénales Rares, Hôpital Femme Mère Enfant, 59 boulevard Pinel, Bron cedex 69677, France. E-mail: pierre.cochat@chu-lyon.f Received 6 March 2009; revised 11 May 2009; accepted 13 May 2009; published online 1 July 2009 Type 1 primary hyperoxaluria (PH1) is an autosomal recessive disorder caused by a deficit in the liver-specific, peroxisomal, alanine:glyoxylate aminotransferase (AGT). AGT deficiency leads to overproduction of oxalate and deposition of insoluble calcium oxalate crystals in tissues. 1,2 More than 100 mutations have been identified on AGXT, the gene encoding AGT, 3 few being frequent. To this genetic heterogeneity is added a marked phenotypic heterogeneity, and genotype/phenotype correlations have yet to be clearly demonstrated. 4 Pyridoxine is the AGT co-factor and B6 sensitivity has been shown in 10 30% of patients, especially in those with the G170R mutation. 5,6 Calcium oxalate deposition in tissues eventually leads to renal failure and oxalosis, a multiorgan involvement. End-stage renal disease (ESRD) is reached between 25 and 33 years of age in 50% of cases, 7 9 whereas first symptoms appear at a median age of 5 6 years. Conservative measures are based on the reduction of calcium oxalate precipitation by high fluid intake (2 3 l/m 2 /day), urine alkalinization, and crystallization inhibitors such as citrate, bicarbonate, and magnesium. 10 12 Treatment with pyridoxine can potentially reduce oxalate production in B6-sensitive patients. 13,14 Conventional dialysis does not permit sufficient oxalate removal as production rates exceed clearance capacity. 15 17 Liver transplantation is currently the only acceptable option and is mostly performed as combined or sequential liver kidney transplantation when ESRD has been, or is soon to be, reached. 18,19 Although the individual course of the disease is unpredictable, conservative treatment started at the time of diagnosis while renal function was still good, might be able to delay further renal deterioration. This is of particular interest in children with early diagnosis, as in family screening. Kidney International (2009) 76, 767 773 767

S Fargue et al.: Effect of conservative treatment on the renal outcome of children with PH1 The purpose of this study is to describe the renal outcome of patients on conservative treatment in a pediatric population. RESULTS Population characteristics The diagnosis database included 40 childreno15 years of age; 13 patients were excluded because of ESRD at presentation (n ¼ 10, with six patients with infantile PH1), pre-emptive liver transplantation (n ¼ 1), follow-upo1 year (n ¼ 1), and unavailable data (n ¼ 1). Thus, 27 children aged less than 15 years at treatment initiation were included for analysis. Among the 27 patients (sex ratio 0.48), 11 were western Europeans and 14 were Mediterraneans. Consanguinity was reported in 13. A positive family history of PH1 or lithiasis was found in 12. The diagnosis was made on family screening in six children. Clinical presentation The median age at first symptom was 29 months (0.5 94.0), and was before 5 years of age in 22/27 patients (Table 1). Median age at diagnosis was 49 months (0.9 148), with a median therapeutic delay of 7.5 months (0 108). Three children were diagnosed before any renal symptom had occurred, as family screening, and four children were diagnosed within 2 months of first symptoms. Fifteen children had renal function impairment at diagnosis, and one child presented with extrarenal (bone) involvement. Diagnosis of PH1 Diagnosis of PH1 was based on enzymatic deficiency on hepatic biopsy in 16/27 patients with a median corrected enzymatic activity of 0.5 mmol pyruvate/sec per kg protein (0 18) (normal values: 490 750). Mutation screening was carried out for 25/27 patients and at least one AGXT mutation identified for all patients. The c.731t4c (I244T) mutation was found in 7/25 patients of Mediterranean descent, 6 homozygous and 1 heterozygous, representing 26% of alleles searched. The c.508g4a (G170R) mutation was found in 4/25, as a heterozygous mutation (8% of alleles). Other mutations found included 33 34 insc (12% of patients; 8% of alleles), c.568g4a (G190R, 12 and 8%, respectively), c.584t4g (M195R, 8% for both), c.847-1g4c (16 and 8%, respectively), c.847-3c4g (8 and 6%, respectively), and private mutations for the rest. Only one mutation could be identified for one patient. Treatment All patients received fluid intake orally (20/27) or enterally, with nasogastric tube or gastrostomy (7/27) (Table 2). Fluid intake ranged between 2 and 3 l/m 2 per day for 8 children and was over 3 l/m 2 per day for 13 children. Most children also received urine alkalinization, though two did not (one in each group). Alkalinization consisted of sodium or potassium citrate in 14, sodium bicarbonate in 7, and both citrate and bicarbonate in 4. Magnesium was added for 13 children. Sensitivity to pyridoxine was assessed in some patients but formal testing could not be recorded in all. Seventeen children were still treated with B6, 10 having been declared B6 resistant after a median time of treatment of 6.6 years (range 3.5 19). Mean B6 dose ranged between 10 and 20 mg/ kg/day and was over 20 mg/kg/day in two patients of group 2. Renal outcome Median follow-up was 8.7 years (1.8 18.8), with a median age at last follow-up of 12.1 years (4.6 21.2) (Table 3). One patient died during follow-up from a domestic accident. Twenty-five children presented at least one nephrolithiasis during follow-up, with a mean number of renal stone passage of 2.1±0.5 (range 0 7) that required a mean number of shock wave lithotripsy or open surgery of 1.5±0.4 (range 0 8). Nephrocalcinosis was found in 19 patients. Radiological lesions were stable over time in 13 patients, diminished or disappeared in 8, grew or appeared in 3, with data unavailable in 3. Only one patient remained symptom-free during follow-up. Median serum creatinine at the start of the treatment was 52 mmol/l (29 115) and 66 mmol/l (33 246) at the last follow-up. Median glomerular filtration rate (GFR) was 74 ml/min per 1.73 m 2 (29 141) at the start of the treatment and 105 ml/min per 1.73 m 2 at the last follow-up (7 195); 85 (29 141) and 114 (40 195), respectively, when patients with renal replacement therapy (RRT) later on are excluded. Median DGFR was þ 11 ml/min per 1.73 m 2 ( 86 to þ 124), þ 12 ( 40, þ 124) RRT excluded (Figure 1), with a median slope of þ 1.2 ml/min per 1.73 m 2 per year. Of the 15 patients who presented with a decreased GFR at diagnosis (GFRo90 ml/min per 1.73 m 2 ), 7 already had chronic kidney disease (CKD), as opposed to transiently decreased initial GFR: 6 with stage 2 and 1 with stage 3. Three additional patients reached CKD stage 2 4 later (Figures 1 and 2). Survival-free from CKD stage 3 was 92% at 3 years of followup, 78% at 5 years of follow-up (Figure 3a), with an overall prevalence of CKD stage 2 4 of 37% (10/27 children). The median time interval between start of the treatment and CKD stage 3 was 4.1 years (2 9.9). CKD stage 2 occurred at a median chronological age of 13.1 years (9 15.2). Four patients reached ESRD and were treated by hemodialysis at a median age of 17.8 years (15.3 21.3), with a GFR between 7 and 19 ml/min per 1.73 m 2. Another patient was transplanted pre-emptively at 15.1 years with a GFR of 19 ml/min per 1.73 m 2 (Figure 2). Among the five patients who required RRT, four were transplanted at a median age of 18.3 years (15.1 22.1). One patient received an isolated kidney graft and presented with recurrence 4 months post-transplantation, leading to combined liver kidney transplantation with good results. Combined liver kidney transplantation was performed in the other three patients, with kidney graft failure in one, requiring a second kidney transplant 2 months later. The median time interval between the start of the treatment and RRT was 9.6 years (7.1 11), RRT occurring at a median age of 16.2 (14.3 21.2) years. Survival-free from RRT (Figure 3b) 768 Kidney International (2009) 76, 767 773

S Fargue et al.: Effect of conservative treatment on the renal outcome of children with PH1 Table 1 Baseline characteristics of 27 PH1 patients Variable Total (n=27) Group 1 (n=20) Group 2 (n=7) P-value Male, % 48 60 14.3 0.08 Ethnic origin, n (%) 0.81 Western European 11 (41) 7 (35) 4 (57) Mediterranean 14 (52) 11 (55) 3 (43) Other 2 (7) 2 (10) 0 Consanguinity, n (%) 13 (48) 10 (50) 3 (43) 1.00 Family history, n (%) 12 (44) 9 (45) 3 (43) 1.00 Presentation Age at first symptoms, years (range) 2.4 (0.04 7.8) 1.8 (0.04 7.8) 4.0 (1.5 5.6) 0.21 GFRo90 ml/min per 1.73 m 2, n (%) 15 (55) 10 (50) 5 (71) 0.39 CKD at diagnosis, n (%) 7 (26) 2 (10) 5 (71) 0.03 Nephrolithiasis, n (%) 25 (92) 18 (90) 7 (100) 1.00 Nephrocalcinosis, n (%) 19 (70) 12 (60) 7 (100) 0.07 Extrarenal involvement, n (%) 1 (4) 0 1 (14) 1.00 Diagnosis Age at diagnosis, years (range) 4.1 (0.1 12.3) 2.8 (0.1 9.5) 8.0 (2.3 12.3) 0.02 Therapeutic delay, years (range) 0.6 (0.2 9.0) 0.1 (0.2 8.4) 2.4 (0.1 9.0) 0.02 Family screening, n (%) 6 (22) 5 (25) 1 (14) 1.00 Mutation, n (total n=25) (total n=19) (total n=6) 1.00 I244T 7 (28) 6 (32) 1 (17) 0.63 G170R 4 (16) 3 (16) 1 (17) 1.00 Other 14 (56) 10 (52) 4 (66) AGT enzymatic activity, n=16 (mmol pyruvate/sec per kg protein) 0.5 (0 18) 0.1 (0 1.6) 0 (0 0.3) 0.18 AGT, alanine:glyoxylate aminotransferase (normal range: 490 750 mmol pyruvate/sec per kg protein); CKD, chronic kidney disease; GFR, glomerular filtration rate; PH1, primary hyperoxaluria type 1. Results are given as median and range when not otherwise specified. Table 2 Details of conservative treatment Variable Total (n=27) Group 1 (n=20) Group 2 (n=7) P-value Vitamin B6 treatment, n (%) 17 (63) 12 (60) 5 (71) 0.67 Vitamin B6 dose, n (%) o10 mg/kg/day 2 (7) 1 (5) 1 (14) 10 20 mg/kg/day 10 (37) 9 (45) 1 (14) 420 mg/kg/day 2 (7) 0 2 (29) Unavailable 3 (11) 2 (10) 1 (14) Type of hyperhydration, n (%) 1.00 Oral 20 (74) 15 (75) 5 (71) Enteral 7 (26) 5 (25) 2 (29) Fluid intake, n (%) 0.63 2 3 l/m 2 /day 8 (30) 5 (25) 3 (43) 43 l/m 2 /day 13 (48) 10 (50) 3 (43) Unavailable 6 (22) 5 (25) 1 (14) Alkalinization, n (%) 0.30 None 2 (7) 1 (5) 1 (14) Citrate (Na or K) 14 (58) 12 (60) 2 (29) Bicarbonate (Na) 7 (26) 5 (25) 2 (29) Citrate+Na bicarbonate 4 (15) 2 (10) 2 (29) Magnesium, n (%) 13 (48) 11 (55) 2 (29) 0.38 after treatment initiation was 100% at 1 and 7 years and 80% at 10 years (N ¼ 8 children at risk at 10 years). The overall prevalence rate of RRT was 19% (5/27). At 15 years of age, 100% of the children were free from RRT and 73% at 18 years. Determinants in the evolution of renal function According to renal outcome, patients were divided into two groups: group 1 (favorable outcome) included 20 (74%) patients and group 2 (unfavorable outcome that is, decline in GFR420 ml/min per 1.73 m 2 ) included 7 (26%) patients. Kidney International (2009) 76, 767 773 769

S Fargue et al.: Effect of conservative treatment on the renal outcome of children with PH1 Table 3 Outcome of 27 PH1 patients on conservative treatment Variable Total (n=27) Group 1 (n=20) Group 2 (n=7) P-value Median follow-up, years (range) 8.7 (1.8 18.8) 6.7 (1.8 18.8) 9.6 (7.1 11.0) 0.14 Death, n (%) 1 (4) 1 (5) 0 CKD stage 2 4, n (%) (ESRD excluded) 6 (22) 3 (15) 3 (43) ESRD, n (%) 4 (15) 0 4 (57) Age at RRT, years (range) 16.2 (15.1 21.2) Transplantation, n (%) 4 (14.8) 0 4 (57) Age at transplantation, years (range) 18.3 (15.1 22.1) Renal stone passage, mean±s.e. (range) 2.1±0.5 (0 7) 1.7±0.5 (0 7) 3.3±1.2 (0 7) 0.17 ESWL or open surgery, mean±s.e. (range) 1.5±0.4 (0 8) 1.2±0.4 (0 6) 2.6±1.1 (0 8) 0.07 Radiological evolution, n (%) 0.78 Stabilization or improvement 21 (78) 16 (80) 5 (72) Degradation 3 (11) 2 (10) 1 (14) Unavailable 3 (11) 2 (10) 1 (14) GFR evolution Initial GFR (ml/min per 1.73 m 2 ) 74 (29 141) 85 (29 141) 69 (48 100) 0.33 Initial GFR (RRT excluded) 85 (29 141) 85 (29 141) 87 (74 100) Final GFR (ml/min per 1.73 m 2 ) 105 (7 195) 118.5 (51 195) 16 (7 57) Final GFR (RRT excluded) 114 (40 195) 118.5 (29 195) 48.5 (40, 57) DGFR (ml/min per 1.73 m 2 ) 11 ( 86, +124) 17.5 ( 16, +124) 43 ( 86, 35) DGFR (RRT excluded) 12 ( 40, +124) 17.5 ( 16, +124) 40 ( 43, 35) CKD, chronic kidney disease; ESRD, end-stage renal disease; ESWL, extracorporal shock wave lithotripsy; GFR, glomerular filtration rate; PH1, primary hyperoxaluria type 1; RRT, renal replacement therapy. Results are given as median except when otherwise specified. GFR (ml/mn per 1.73m 2 ) 250 200 150 100 50 0 0 24 48 72 96 120 144 168 192 216 240 Time (months) Figure 1 Evolution of glomerular filtration rate in 27 children under conservative treatment from first to last follow-up. Children belonging to group 1 are represented with solid lines, group 2 children, with a decline in renal function by more than 20 ml/min per 1.73 m 2, are represented with dotted lines. Children who were under 1 year of age at diagnosis are represented with filled triangles, children between 1 and 2 years of age with open triangles, and others with a rectangle. GFR, glomerular filtration rate, as estimated by the Schwartz formula. There was no statistical difference between the two groups in terms of sex ratio, ethnic background, consanguinity, family history, diagnosis circumstances, mutations (frequencies of I244T and G170R alleles), enzymatic activity, age at onset, number of renal stone migration, nephrocalcinosis, radiological evolution, initial GFR, treatment modalities (B6 dose, hyperhydration, alkalinization, and use of crystallization inhibitors), and duration of follow-up (Tables 1 3). In univariate analysis (Table 2), patients in group 2 were statistically older at diagnosis (94±18 vs 45±8 months in group 1; P ¼ 0.025), had a longer therapeutic delay between first symptoms and treatment initiation (51±19 vs 13±6 months in group 1; P ¼ 0.01), and were more likely to have CKD stage 2 4 at diagnosis (29 vs 15% in group 1; P ¼ 0.03). In a Cox regression model, the only variable significantly associated with an unfavorable renal outcome was therapeutic delay (P ¼ 0.02; relative risk 1.7 (1.1 2.7) per year delay). CKD stage 2 4 at initial presentation (P ¼ 0.12; relative risk 3.7 (0.7 20.0)) was not significant. DISCUSSION Few studies have addressed the renal outcome of patients with PH under conservative treatment. Our study is the first to evaluate the outcome of renal function exclusively in formally proven PH1 patients, with inclusion of all French children. This is relevant as patients with PH2, and secondary hyperoxaluria, can also reach ESRD, but the time at which this occurs may differ from that of PH1. 9,20,21 In this study, 7/27 children with PH1 deteriorated their renal function after a median follow-up of 7.7 years, 5 of whom required RRT after a median time of 9.2 years. The last GFR was below 60 ml/min per 1.73 m 2 in eight children, five of whom already had impaired GFR at diagnosis. The choice of a cutoff GFR of 60 ml/min per 1.73 m 2 for discrimination was dictated by the notion that as renal function declines, so does the excretion of oxalate, leading to further accumulation. The threshold GFR below which degradation is unavoidable was estimated to be between 40 and 50 ml/min per 1.73 m 2. Such a notion is linked to therapeutic strategies. We, therefore, chose a GFR of 60 (CKD stage 2 or more) to identify patients at low risk of further renal degradation. Among the 15 patients with baseline GFR below 90, 8 improved their GFR, suggesting that a low GFR at diagnosis does not necessarily represent a risk for developing ESRD. This may be related to the presence of initial renal 770 Kidney International (2009) 76, 767 773

S Fargue et al.: Effect of conservative treatment on the renal outcome of children with PH1 GFR (ml/min per 1.73m 2 ) 120 100 80 Non-compliance 60 40 20 7 4 Urolithiasis Pregnancy 6 APN 3 2 1 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Years since start of treatment Figure 2 Decline of renal function in group 2 children during follow-up and clinical events. Group 2, n ¼ 7, defined by decline in GFR of more than 20 ml/min per 1.73 m 2. Patients 1, 2, 3, and 6 were treated by hemodialysis once end-stage renal disease was reached, patient 4 was pre-emptively transplanted (liver kidney transplantation), and patients 5 and 7 are still on conservative treatment. APN, acute pyelonephritis. Cumulative incidence free from CKD 1.0 0.8 0.6 0.4 0.2 0.0 0 1 2 3 4 5 6 7 8 9 10 1112 Years since start of treatment Cumulative incidence free from RRT obstruction and reversible renal impairment in some and to natural maturation of renal function in children below 2 years of age at diagnosis (5/7 with initial GFRo90). Milliner et al. 10 reported 3/9 PH1 patients with ESRD after 9 years of follow-up and a decline in GFR of 1.4 ml/min per 1.73 m 2 per year and later reported 4/12 patients with ESRD after 10 years of mean follow-up. 20 Lieske et al. 9 reported 15/74 PH patients (PH1 and -2 and unclassified hyperoxaluria) with ESRD after 9 years of median follow-up. For van Woerden et al., 22 8/30 patients, adults, and children developed CKD during a median follow-up of 7.7 years, 5 of whom reaching ESRD. Among the 15 screened for diagnosis, 4 developed CKD stage 2 or more. In our study, the median time interval between initiation of conservative measures and stage 3 CKD was 4.1 years and chronological age 13.1 years, whereas it was at a chronological age of 30 years (22.2 37.7) in the study by van Woerden et al. 22 No publication has focused on renal survival in children, discounting those adults with late onset of PH. After 10 years of follow-up, the cumulative incidence of CKD stage 2 or more was 54% in this study. For Milliner et al., 10 cumulative incidences of ESRD at 5, 10, and 15 years of treatment were 4, 11, and 25%, respectively, vs 0 and 20% RRT at 5 and 10 years, respectively, in our study. However, Milliner s study included all types of PH, thus it probably underestimates ESRD in PH1 patients. For Kopp et al., 23 1.0 0.8 0.6 0.4 0.2 0.0 0 1 2 3 4 5 6 7 8 9 10 1112 Years since start of treatment Figure 3 Renal survival in 27 PH1 children. (a) Survival free of chronic kidney disease (CKD, glomerular filtration rateo60 ml/min per 1.73 m 2 ). (b) Survival free of renal replacement therapy (RRT). PH1, primary hyperoxaluria type 1. ESRD was reached in 20% of PH patients by 15 years of age. Median time without ESRD in PH1 was 33 years for Lieske et al., 9 with an actuarial renal survival for all PH of 97, 91, and 64% at 10, 20, and 30 years, respectively. In this study, survival-free from RRT was 100 and 73% at 15 and 18 years of age, respectively, with six patients having reached adulthood (18 years). The outcome in our study is similar to that of Lieske 9 and Kopp. 23 The patients in this study benefited from an early treatment, thus improving outcome. Improvement of renal function has been shown even when renal insufficiency was already present at the time of diagnosis. 22,23 The age at onset of symptoms in our study was lower than that in other studies: median of 2.4 vs 5.5 years for Lieske et al. 9 and 6 years for van Woerden et al., 22 as these studies included adult and pediatric patients, as well as PH1, -2, and atypical PH for Lieske et al. The younger patients in our study belonged to group 1 as they were diagnosed by family screening; however, all patients presented with symptoms before 8 years of age, and only 4/27 after 5 years of age. The diagnosis was also made earlier, with a median age at diagnosis of 4 years, compared with 7 years for van Woerden et al. 22 and 15 years for Lieske et al. 9 The therapeutic delay in our study was therefore shorter (0.6 years), though over 8 years in four, three of whom degraded their renal function later on. The delay to diagnosis is often long in PH, with a reported mean time of 5 years for Lieske et al. 9 and a median time of 8 years for Kopp et al. 10,23 In our study, only the therapeutic delay was associated with a degradation of renal function. This factor had not been previously shown to be associated with an unfavorable outcome, 22 and suggests the importance of irreversible renal damage, even after optimal treatment has been started. This stresses the crucial role of early diagnosis and management of PH. Conversely, the presence of nephrocalcinosis and pyridoxine unresponsiveness was not found to be associated with deterioration of renal function. This might be explained by the small number of patients with nephrocalcinosis. No conclusion can be drawn about B6 response, as formal testing was not available for all patients. The end point of this study was the outcome of PH1 children, a population more homogenous than a mixed adult and children population. Six children had a diagnosis made as part of a family screening but only three of them were symptom-free at that time as nephrocalcinosis was already present in four. Precocious silent renal lesions have been reported by others. 9,22,23 Of the three patients asymptomatic at diagnosis, only one remained symptom-free as the others formed renal stone during follow-up that disappeared after readjustment of treatment to body weight. This points out the continuous benefits of treatment, whatever the stage of the disease may be. There was no advantage to enteral as opposed to oral hydration, as the critical factor is the volume and not the way through which the fluid is taken. Of the four children who reached ESRD, one had a stable renal function with GFR around 50 ml/min per 1.73 m 2 for 7 years and suddenly degraded her renal function within 1 month until Kidney International (2009) 76, 767 773 771

S Fargue et al.: Effect of conservative treatment on the renal outcome of children with PH1 she reached ESRD (Figure 2). This occurred after a complete lapse in compliance, which is known to be a major determinant. Lapses in compliance led to the apparition of renal stones in two patients in a matter of months after the end of this study. The transition to adult care is a sensitive one and adolescents are here also particularly at risk. The frequency of the G170R mutation was only 16% here, although it is reported to be the most frequent mutation amounting to a third of all mutations. 24,25 However, the I244T represented 26%, which reflects on the ethnicity of our population, as it is predominant in Mediterranean countries. 26,27 No phenotype genotype correlation could be established in our study. This study was conducted retrospectively in a limited number of patients, which was due to the low prevalence of the disease, especially in the pediatric age group. The management of patients was rather homogenous and followed international standards. Despite the crucial role of compliance, it was not taken into account as it had not been evaluated in all patients. The GFR was estimated using the Schwartz formula as only a subset of patients were evaluated with inulin clearance. However, our conclusions are based on GFR changes and not on GFR absolute values. In summary, an early and aggressive treatment may preserve renal function in compliant children with PH1. A favorable course results from a delicate balance between the natural history of the disease and its management. Conservative treatment must be adapted to body weight to fit child s growth. The presence of silent symptoms in patients who are screened for diagnosis is a reminder of the necessity of an early diagnosis. A longer follow-up of such patients would enable us to determine if the stability of GFR that we observed can delay the onset of ESRD. New treatments are now being developed, which may help stabilize or slow down the decline in renal function in addition to standard conservative treatment and further delay ERSD. MATERIALS AND METHODS Study design We conducted a multicenter retrospective cohort study. All children aged less than 15 years followed in a French-speaking center for at least 1 year, with an enzymatically or genetically proven PH1 between 1986 and 2006, were identified from the database of the only reference center for enzymatic or genetic diagnosis of PH1 in France (Laboratoire de Biologie Est, Hospices Civils de Lyon). We then contacted all the centers where these children were followed to collect clinical and biological follow-up data. Population and data collection PH1 was defined on the basis of the following criteria: AGT activity below 50% on liver biopsy, or mutations in AGXT, or urinary biochemistry in favor of PH1 in siblings of enzymatically or genetically proven PH1 patients. Enzymatic studies and mutation analysis were performed in the same laboratory. The age at diagnosis was considered to be that of initiation of conservative treatment even if enzymatic or genetic confirmation was performed later. Clinical and laboratory information was collected on chart review. Baseline characteristics such as gender, ethnic origin, family history of nephrolithiasis or PH1, and consanguinity were recorded, as well as clinical presentation, renal events, and their timing (stone passage, nephrocalcinosis, dialysis, and transplantation). The treatment delay was considered to be the time elapsed between the first symptom and initiation of conservative treatment. Laboratory data included plasma creatinine concentration, height and body weight, AGXT mutations, and AGT enzymatic activity. Renal insufficiency at diagnosis was defined as a GFRo90 ml/min per 1.73 m 2 and was classified as transient or chronic if renal function could not be restored 3 months after the initiation of treatment; CKD was staged according to the Kidney Disease Outcome Quality Initiative and Kidney Disease Improving Global Outcome stages, 28,29 and was defined as a GFRo60 ml/min per 1.73 m 2 (stage 3). Conservative treatment Conservative treatment consisted of all children in hyperhydration (X2 3 l/m 2 /day), administered orally or enterally (nasogastric or gastric tube feeding), and urine alkalinization by sodium or potassium citrate or sodium bicarbonate. Pyridoxine was administered in B6-sensitive patients, but sensitivity was not universally investigated in the absence of available guidelines. Details on medical treatment were recorded (type of fluid intake, volume, type and dose of urine alkalinization, and pyridoxine dose). Statistical analysis The main outcome was GFR estimated from Schwartz 30 formula for children under 18 years of age and with the Modification of Diet in Renal Disease formula for adults. 31 For end point analysis, patients with ESRD and pre-emptive liver kidney transplantation were pooled together as RRT may occur at a greater GFR in PH patients. Patients were divided into two groups according to their renal outcome: group 1 included children with stable renal function, whereas group 2 had a decline in GFR over 20 ml/min per 1.73 m 2 during the study period. Results were expressed as median and range or mean±s.d. for continuous variables, and as percentage for categorical variables. w 2 or Student s t-test was used for comparison of normally distributed variables, and Wilcoxon test or Fisher s exact test was used for non-normally distributed variables. The probability of remaining free from CKD or RRT was estimated using a time to failure Kaplan Meier method. A Cox proportional hazards model was performed to determine factors associated with a decline in GFR420 ml/min per 1.73 m 2 during follow-up. All statistical analyses were carried out with the use of SAS software (version 9.1; SAS Institute, Cary, NC, USA). DISCLOSURE All the authors declared no competing interests. REFERENCES 1. Danpure CJ, Jennings PR, Watts RW. Enzymological diagnosis of primary hyperoxaluria type 1 by measurement of hepatic alanine: glyoxylate aminotransferase activity. Lancet 1987; 1: 289 291. 2. Danpure CJ, Rumsby G. Enzymological and molecular genetics of primary hyperoxaluria type 1. Consequence for clinical management. In: Khan SR (ed). Calcium Oxalate in Biological Systems. CRC Press: Boca Raton, 1995, pp 189 205. 3. Purdue PE, Lumb MJ, Fox M et al. Characterization and chromosomal mapping of a genomic clone encoding human alanine:glyoxylate aminotransferase. Genomics 1991; 10: 34 42. 772 Kidney International (2009) 76, 767 773

S Fargue et al.: Effect of conservative treatment on the renal outcome of children with PH1 4. Amoroso A, Pirulli D, Florian F et al. AGXT gene mutations and their influence on clinical heterogeneity of type 1 primary hyperoxaluria. JAm Soc Nephrol 2001; 12: 2072 2079. 5. van Woerden CS, Groothoff JW, Wijburg FA et al. Clinical implications of mutation analysis in primary hyperoxaluria type 1. Kidney Int 2004; 66: 746 752. 6. Monico CG, Olson JB, Milliner DS. Implications of genotype and enzyme phenotype in pyridoxine response of patients with type I primary hyperoxaluria. Am J Nephrol 2005; 25: 183 188. 7. Cochat P, Liutkus A, Fargue S et al. Primary hyperoxaluria type 1: still challenging!. Pediatr Nephrol 2006; 21: 1075 1081. 8. Cochat P, Deloraine A, Rotily M et al. Epidemiology of primary hyperoxaluria type 1. Société de Néphrologie and the Société de Néphrologie Pédiatrique. Nephrol Dial Transplant 1995; 10(Suppl 8): 3 7. 9. Lieske JC, Monico CG, Holmes WS et al. International registry for primary hyperoxaluria. Am J Nephrol 2005; 25: 290 296. 10. Milliner DS, Eickholt JT, Bergstralh EJ et al. Results of long-term treatment with orthophosphate and pyridoxine in patients with primary hyperoxaluria. N Engl J Med 1994; 331: 1553 1558. 11. Broyer M, Jouvet P, Niaudet P et al. Management of oxalosis. Kidney Int Suppl 1996; 53: S93 S98. 12. Leumann E, Hoppe B, Neuhaus T. Management of primary hyperoxaluria: efficacy of oral citrate administration. Pediatr Nephrol 1993; 7: 207 211. 13. Watts RW, Veall N, Purkiss P et al. The effect of pyridoxine on oxalate dynamics in three cases of primary hyperoxaluria (with glycollic aciduria). Clin Sci (Lond) 1985; 69: 87 90. 14. Hoppe B, Langman CB. A United States survey on diagnosis, treatment, and outcome of primary hyperoxaluria. Pediatr Nephrol 2003; 18: 986 991. 15. Marangella M, Petrarulo M, Cosseddu D et al. Oxalate balance studies in patients on hemodialysis for type I primary hyperoxaluria. Am J Kidney Dis 1992; 19: 546 553. 16. Hoppe B, Graf D, Offner G et al. Oxalate elimination via hemodialysis or peritoneal dialysis in children with chronic renal failure. Pediatr Nephrol 1996; 10: 488 492. 17. Hoppe B, Kemper MJ, Bokenkamp A et al. Plasma calcium oxalate supersaturation in children with primary hyperoxaluria and end-stage renal failure. Kidney Int 1999; 56: 268 274. 18. Watts RW, Calne RY, Williams R et al. Primary hyperoxaluria (type I): attempted treatment by combined hepatic and renal transplantation. QJMed1985; 57: 697 703. 19. Cochat P, Schärer K. Should liver transplantation be performed before advanced renal insufficiency in primary hyperoxaluria type 1? Pediatr Nephrol 1993; 7: 212 218; discussion 218 219. 20. Milliner DS, Wilson DM, Smith LH. Phenotypic expression of primary hyperoxaluria: comparative features of types I and II. Kidney Int 2001; 59: 31 36. 21. Johnson SA, Rumsby G, Cregeen D et al. Primary hyperoxaluria type 2 in children. Pediatr Nephrol 2002; 17: 597 601. 22. van Woerden CS, Groothoff JW, Wanders RJ et al. Primary hyperoxaluria type 1 in The Netherlands: prevalence and outcome. Nephrol Dial Transplant 2003; 18: 273 279. 23. Kopp N, Leumann E. Changing pattern of primary hyperoxaluria in Switzerland. Nephrol Dial Transplant 1995; 10: 2224 2227. 24. Coulter-Mackie MB. Preliminary evidence for ethnic differences in primary hyperoxaluria type 1 genotype. Am J Nephrol 2005; 25: 264 268. 25. Rumsby G, Williams E, Coulter-Mackie M. Evaluation of mutation screening as a first line test for the diagnosis of the primary hyperoxalurias. Kidney Int 2004; 66: 959 963. 26. Basmaison O, Rolland MO, Cochat P et al. Identification of 5 novel mutations in the AGXT gene. Hum Mutat 2000; 15: 577. 27. Santana A, Salido E, Torres A et al. Primary hyperoxaluria type 1 in the Canary Islands: a conformational disease due to I244T mutation in the P11L-containing alanine:glyoxylate aminotransferase. Proc Natl Acad Sci USA 2003; 100: 7277 7282. 28. Hogg RJ, Furth S, Lemley KV et al. National Kidney Foundation s Kidney Disease Outcomes Quality Initiative clinical practice guidelines for chronic kidney disease in children and adolescents: evaluation, classification, and stratification. Pediatrics 2003; 111: 1416 1421. 29. Levey AS, Eckardt KU, Tsukamoto Y et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2005; 67: 2089 2100. 30. Schwartz GJ, Haycock GB, Edelmann Jr CM et al. A simple estimate of glomerular filtration rate in children derived from body length and plasma creatinine. Pediatrics 1976; 58: 259 263. 31. Levey AS, Bosch JP, Lewis JB et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999; 130: 461 470. Kidney International (2009) 76, 767 773 773