The spectrum of chronic renal failure among Jordanian children

BACKGROUND: The causes of chronic renal failure (CRF) vary from one country to another. In this study we reviewed our experience with the different types of renal disorders leading to CRF in Jordanian children. METHODS: We investigated CRF in 202 Jordanian children (113 males and 89 females) who presented to the Jordan University Hospital, Amman, in the period from July 1988 to April 2001. The mean age at onset of CRF was 7.5 +/- 3.9 years. Patients were followed for 0.6-12.6 years (mean 6.3 years). RESULTS: The causes of CRF included urological abnormalities and malformations (42.1%), hereditary renal disorders (29.7%), glomerulonephritis (GN)(14.4%), renal hypo- or dysplasia (5%), hemolytic uremic syndrome (HUS) (4.5%), and idiopathic (4.5%). Forty-nine patients required renal replacement therapy, most of them with peritoneal dialysis. Nine patients have undergone renal transplantation. We estimated the prevalence of CRF in children in Jordan to be 51 per million population, and the incidence as 10.7 new cases per million-child population per year. CONCLUSION: The high rate of hereditary disorders in our series is attributed to the high prevalence of parental consanguinity in our community. There was a striking number of patients with non-neurogenic neurogenic bladder in our study. The relative incidence of GN leading to CRF in Jordan is lower than in Europe and North America. The relative incidence of the other causes of CRF in our series is similar to many other countries. The incidence and prevalence of CRF in children in Jordan is high compared to other countries.     

Children with chronic renal failure in Sweden 1978-1985

A survey of chronic renal failure (CRF) in Swedish children was carried out for the period 1978-1985, using age-related cut-off levels for creatinine concentrations corresponding approximately to a glomerular filtration rate of 30 ml/min per 1.73 m2. The mean annual incidence of CRF was 6.9 and of terminal renal failure (TRF) 4.4/million children. The prevalence increased during the study period, for preterminal renal failure from 14.1 (1978) to 26.1 (1985) and for TRF from 12.4 to 16/million children. The main groups of primary renal disease were malformations (42%), hereditary disorders (27%), and glomerular diseases (14%), while pyelonephritis with vesico-ureteral reflux only made up 5%.     

Children with chronic renal failure in Sweden 1978–1985

A survey of chronic renal failure (CRF) in Swedish children was carried out for the period 1978–1985, using age-related cut-off levels for creatinine concentrations corresponding approximately to a glomerular filtration rate of 30 ml/min per 1.73 m². The mean annual incidence of CRF was 6.9 and of terminal renal failure (TRF) 4.4/million children. The prevalence increased during the study period, for preterminal renal failure from 14.1 (1978) to 26.1 (1985) and for TRF from 12.4 to 16/million children. The main groups of primary renal disease were malformations (42%), hereditary disorders (27%), and glomerular diseases (14%), while pyelonephritis with vesico-ureteral reflux only made up 5%.     

Chronic renal failure in children: a report from Port Harcourt,Nigeria (1985–2000)

A 15-year review of children who presented with chronic renal failure (CRF) to the University of Port Harcourt Teaching Hospital, Rivers State of Nigeria, was carried out. Forty-five children (28 boys, 17 girls) with CRF, defined as a glomerular filtration rate below 30 ml/min per 1.73 m(2) body surface area or a rise in serum creatinine above 120 micro mol/l for at least 6 months, often accompanied by other biochemical abnormalities, were identified. The median annual incidence of CRF was 3.0 per million children. The prevalence of CRF increased from 12.5 in the 1985-1990 periods to 15 per million children after 1995. Acquired disorder was the major cause of CRF. Glomerulopathies were the cause in 53.3% of patients, mainly chronic glomerulonephritis (56.5%) and nephrotic syndrome (30.4%). Hepatitis B surface antigen was positive in 2 patients. Congenital disorders accounted for 28.9% of all cases of CRF, which is lower than data from other countries. Posterior urethral valve was the only congenital disorder causing CRF in the study. No child with hereditary renal disorder as a cause of CRF was identified. Children with congenital disorders were diagnosed at an earlier age. The mortality rate was high (46.7%), as most patients were managed conservatively, since there were no permanent facilities for chronic dialysis or renal transplantation in Nigeria. The study shows that CRF is common in Nigerian children, and there is an urgent need for the establishment of facilities for renal replacement therapy.     

Chronic renal failure in children: an epidemiological survey in Lorraine (France) 1975–1990

A comprehensive investigation in Lorraine from 1975 to 1990 identified 127 children (73 boys, 54 girls) under 16 years with chronic renal failure (CRF). From 1975–1980 to 1985–1990 the mean annual incidence of pre-terminal CRF decreased from 12.7 to 7.5 per million children under 16 years of age. The incidence of end-stage renal disease (ESRD) in children increased from 5.6 to 7.5 per million with a peak of 9.1. The prevalence of preterminal CRF was variable (29.4–54) and the prevalence of ESRD increased from 15.5 to 37.0 per million children. Acquired nephropathies were observed in 30.7% and congenital nephropathies in 68.5%. Although patients with acquired nephropathies had only slightly higher serum creatinine levels, they progressed more rapidly to ESRD than those with congenital disease: mean 1.8 years versus 3.85 years after diagnosis of pre-terminal of CRF (P<0.02). Ten years after onset of pre-terminal CRF, 94% with acquired and 69% of those with congenital nephropathies had started renal replacement therapy (P<0.001). It is unclear whether the decrease in preterminal CRF reflects a reduced number of children with kidney disease reaching CRF or is the result of a real delay in the progression due to better therapeutic management.     

Renal failure in Yemen

Renal failure remains a serious cause of mortality in Yemen. Our region has 1.25 million population and our hospital is the central hospital, which has a nephrology department and performs dialysis for the region. Between January 1998 and December 2002, we admitted 547 patients; including children, with acute renal failure (ARF) and chronic renal failure (CRF). CRF was observed in 400 patients, an incidence of 64 per million per year and a prevalence of 320 per million. ARF occurred in 147 persons with an incidence of 23.5 per million per year and a prevalence of 117.5 patients per million. Of all patients, 72% were adults (age range, 20-60 years) with a male preponderance. As a tropical country, malaria (27.9%), diarrhea (13.6%), and other infectious diseases were the main causes. Next most common were obstructive diseases causing CRF and ARF (26.8% and 12.9%, respectively), mainly urolithiasis, Schistosomiasis, and prostatic enlargement. However the cause of CRF in 57.5% of patients was unknown as most persons presented late with end-stage disease (64.7%), requiring immediate intervention. Other causes, such as hepatorenal syndrome, snake bite, diabetes mellitus, and hypertension, showed low occurrence rates. Patients presented to the hospital mostly in severe uremia and without a clear history of prior medications. The major findings were vomiting, acidosis, and hypertension with serum creatinine values ranging between 2.8-45 mg/dL (mean value, 13.4 mg/dL). Anemia was observed in 80.4% of CRF versus 62.6% of ARF patients. Hypertension prevalence was 65.5% among CRF patients, of whom 25% were in hypertensive crisis, whereas among ARF the prevalence was only 26.5%.     

Children with chronic renal failure in the Federal Republic of Germany: I. Epidemiology, modes of treatment, survival. Arbeits- gemeinschaft für P?diatrische Nephrologie

By a nation-wide retrospective survey in the Federal Republic of Germany, epidemiological data were obtained on children with chronic renal failure (CRF) up to the age of 16.0 years. During a 4-year period (1972-1975), an incidence of 6 new cases per year of CRF occurred when referring to a 1 million population of the same age. The incidence of preterminal CRF (serum creatinine greater than 2 mg/dl to end-stage) was 4.4 and its prevalence 6.4 per million per year. The incidence of terminal CRF, analyzed for a 6-year period from 1972 to 1977 increased only slightly with time (from 4.4 to 5.4 per million of the same age per year). The proportion of children with terminal CRF admitted yearly for renal replacement therapy increased during the observation time from 27% to 79% up to the age of 10 years and from 80% to 96% between 10 and 16 years of age. The number of patients alive with terminal CRF rose significantly from 11.9 in 1972 to 22.0 per 1 million of the same age in 1977. At the end of 1977, 46% of all pediatric patients on renal replacement therapy had a functioning graft, compared to 38% in 1972. The increasing number of renal transplantations was accompanied by shortening of the waiting period from first dialysis to grafting. The patient survival on dialysis and after transplantation rose significantly during two subsequent 4-year periods.(ABSTRACT TRUNCATED AT 250 WORDS)     

Chronic renal failure in Iranian children

We investigated chronic renal failure (CRF) in 166 Iranian children (95 boys and 71 girls) from July 1991 to June 1999. The mean age at onset of CRF was 7.9±4.5 years. The most common cause of CRF was congenital urological malformations (78 cases). The second most common cause of CRF was hereditary nephropathy (21%). Glomerular diseases accounted for only 10% of children who later went on to develop renal failure. High rates of cystinosis and primary hyperoxaluria were seen, and these elevated rates could be due to a high prevalence of parental consanguinity. Eighty-six patients required renal replacement therapy, of whom the majority underwent hemodialysis. The prevalence of primary reflux as a cause of CRF was high compared with reports from western countries. Earlier diagnosis and management of urinary tract infections in this group could reduce the prevalence of reflux as a cause of CRF in this population. Received: 15 May 2000 / Revised: 2 October 2000 / Accepted: 5 October 2000     

Epidemiology of chronic renal failure in children: a report from Sweden 1986 – 1994

In a national survey, chronic renal failure (CRF) in Swedish children was studied during the period 1986 – 1994; 118 children (72 boys, 46 girls) with CRF, defined as a glomerular filtration rate below 30 ml/min per 1.73 m² body surface area, were identified. The median annual incidence of CRF was 7.7 and that of terminal renal failure (TRF) 6.4 per million children. The prevalence of preterminal renal failure decreased from 29 to 21 per million children over the study period, while the prevalence of TRF increased from 17.8 in 1986 to 38 per million children in 1994. The increase in TRF prevalence was due to a lower incidence of deaths due to uremia and a slightly increased incidence of TRF compared with an earlier study period, 1978 – 1985. The results point to a more active treatment of uremia in Sweden now than during the period 1978–1985. The congenital causes of CRF (renal malformations, obstructive conditions, and hereditary disorders) accounted for 67.5% of all cases, which is high compared with data from other countries. No child with non-obstructive pyelonephritis as a cause of CRF was identified. Age at detection of CRF and time from detection of CRF to TRF were studied. As a high proportion of children, 42%, reached 16 years of age without entering TRF, the value of presenting time from CRF to TRF for the remaining individuals is questionable. There were only minor differences in primary renal disease, age at presentation, and time from CRF to TRF when the study results were compared with those from 1978 – 1985. Received October 1, 1996; received in revised form and accepted January 21, 1997     

Children with chronic renal failure in the Federal Republic of Germany: II. Primary renal diseases, age and intervals from early renal failure to renal death. Arbeitsgemeinschaft für P?diatrische Nephrologie

In a retrospective survey, 623 children with chronic renal failure (CRF) comprising a 7-year period were registered in the Federal Republic of Germany. The primary renal disease could be classified in 91% of the patients. Pyelonephritis was the most frequent diagnosis (31%), followed by glomerulopathies (20%), renal hypoplasia or dysplasia (14%), cystic kidney disease including nephronophthisis (12%), other hereditary nephropathies (7%), and vascular nephropathies (4%). At the time of first presentation, 23% of the children with preterminal chronic renal failure were younger than 5 years, 34% 5 to 10 years and 43% 10 to 16 years old. At the time of renal death, the serum creatinine level was below 10 mg/dl in 84% of the children below 5 years, compared to 5% in the patients older than 10 years. The mean interval from the first presentation of CRF to the terminal stage was 6 months in vascular nephropathies, 19 months in cystic renal disease, 26 months in glomerular disorders, 32 months in pyelonephritis, and 36 months in hereditary nephropathies and in renal hypoplasia or dysplasia. The range of these intervals is so large, even when diagnostic subgroups are considered, that a reliable prediction of the individual course from the underlying kidney disease is not possible.     

Etiology and outcome of chronic renal failure in hospitalized children in Ho Chi Minh City, Vietnam

The aim of this study was to investigate the etiology and treatment modalities and to determine mortality risks in hospitalized children with chronic renal failure (CRF) in Ho Chi Minh City, Vietnam. We reviewed the records of 310 children with CRF hospitalized in Ho Chi Minh City from January 2001 to December 2005. The average annual number cases was 4.8 per million child population native to Ho Chi Minh City. Median age was 14 years; 85% of patients were in end-stage renal failure. Associated illnesses were anemia (96%), hypertension (74%), and cardiopulmonary diseases (39%). Causes of CRF included glomerulonephritis (30%) and congenital/hereditary anomalies (20%), but in 50% of children, the etiology was unavailable. Seventy-three percent of cases with end-stage renal failure did not benefit from renal replacement therapy. During hospitalization, 47 patients (15%) died. Mortality risks were higher in young children (1–4 years), in boys, and in patients with acquired pathologies. Severe metabolic acidosis was the main predictive factor of mortality by multivariate regression analysis. Our data shows a poor outcome due to late referral and limited facilities for renal replacement therapy in children with CRF hospitalized in Ho Chi Minh City.     

The epidemiology of chronic renal failure and provision of renal services in Albania

Tirana, the only dialysis facility in Albania (pop 4 million),has a stock of 12 patients and three haemodialysis machines.To determine the need for renal services in Albania we studiedthe incidence and outcome of patients with chronic renal failure(CRF) referred to the renal service in Tirana (pop 300000) over1 year. Case-notes of all patients with a serum creatinine concentration300 µmol/l during the study period (1992) were examinedand outcome at 2 years recorded for each patient. In all, 84 patients (mean age 41.6±17.5 years, 56% male)were referred to nephrologists of whom 35 (42%) came from Tirana,giving an annual incidence of 116 per million pop. 77% wereunder 40 and had no co-morbid illness. Glomerulonephritis, themost common renal diagnosis, affected 26% patients. 22% patients(mean age 38±18.1) died within 2 years and only 5% receiveddialysis. The mean age of those who received dialysis was 29±8.3compared with those who were not dialysed (42±18.0).The 59 patients (24 from Tirana i.e. 80 per million) who werealive with advanced CRF (creatinine > 500) had a mean creatinineof 623±93 µmol/l and would be candidates for dialysis.Patients with progressive renal failure in Albania are regularlyfollowed and treated with antihy-pertensives and dietary modification.The need for RRT, however, is not being met even for young patientswith no co-morbidity.     

Long-term outcome of vesicoureteral reflux associated chronic renal failure in children. Data from the ItalKid Project

PURPOSE: The nephropathy associated with vesicoureteral reflux (VUR) is one of the leading causes of chronic renal failure (CRF) in children. We describe the clinical course of the disease based on information available in the ItalKid Project database, and analyze the predictive value of baseline renal function, age at VUR diagnosis and urinary protein excretion in relation to the risk of progressive renal failure. MATERIALS AND METHODS: As of December 31, 2001 the registry included a total of 343 patients (261 males) with a diagnosis of primary VUR, which was the leading single cause of CRF, accounting for 25.4% of all patients with CRF. RESULTS: The estimated risk of end stage renal disease (ESRD) by age 20 years was 56%. The patients with a creatinine clearance (Ccr) of less than 40 ml per minute at baseline had an estimated 4-fold greater risk of ESRD developing in comparison with those whose Ccr was 40 to 75 ml per minute. No significant difference in probability of disease progression to ESRD was found between subjects diagnosed with VUR at age 6 months or less and those diagnosed later (older than 6 months). Furthermore, children with normal urinary protein excretion (a urinary protein [uPr]/urinary creatinine [uCr] ratio of less than 0.2 in 36 patients) and low grade proteinuria (uPr/uCr 0.2 to 0.8 in 34 patients) at baseline showed a significantly slower decrease in mean Ccr than those with moderate proteinuria (uPr/uCr greater than 0.8 in 34 patients). Hypertension and/or antihypertensive treatment (including antiprogressive drugs) were reported in 29.1% of patients. CONCLUSIONS: The results of the present study define the long-term risk of ESRD in a large population of children with CRF and VUR, and provide some critical information for identifying the prognosis.     

Etiology and outcome of chronic renal failure in Indian children

A prospective analysis of all new pediatric cases of chronic renal failure (CRF) was performed at our hospital over a 1-year period. The diagnosis of CRF was based on serum creatinine >2 mg/dl with supportive clinical, laboratory, and radiological findings. There were a total of 48 patients with CRF with a median age of 13 years (range 10 days to 16 years). The causes of CRF included glomerulonephritis (37.5%), obstruction and interstitial (52%), hereditary (6.3%), and undetermined (4.2%). Patients were symptomatic for a mean of 33.2 months (range 10 days to 11 years) at presentation. Eight patients (16.7%) had acute reversible deterioration of renal function at presentation. This was due to accelerated hypertension in 2, infection in 3, volume depletion in 2, and nonsteroidal antiinflammatory drugs in 1 patient. At presentation, 22 (46%) children had mild to moderate renal failure and 26 (54%) had end-stage renal disease. Twenty-one children (43.7%) had associated illness at presentation. Mean follow-up was 22.9 weeks (range 2–126 weeks). At the end of the study period, 10 (21%) patients were on conservative treatment, 7 (14.6%) on maintenance dialysis, 8 (16.7%) patients had functioning allografts, 4 (8.3%) patients had died, and 19 (39.6%) opted against further therapy. We conclude that CRF in Indian children carries a poor prognosis due to late referral and the limited availability and high cost of renal replacement therapy. Received: 31 July 1998 / Revised: 7 December 1998 / Accepted: 13 December 1998     

Etiology of chronic renal failure in Turkish children

The etiology of chronic renal failure (CRF) was studied in 459 Turkish children (205 girls, 254 boys) for the period January 1979-December 1993. Their mean age at onset of CRF was 9.5±4.2 years (range 1–16 years); CRF was defined as a glomerular filtration rate (GFR) below 50 ml/min per 1.73 m² for at least 6 months. When a GFR determination was not available, the serum creatinine concentration was used: greater than 1 mg/dl for children aged 1–3 years, greater than 1.5 mg/dl for those 3–10 years and greater than 2 mg/dl for those 10–16 years. Primary renal disorders were as follows: reflux nephropathy 32.4% glomerular diseases 22.2%, hereditary renal disorders 11.4%, amyloidosis 10.6%, urinary stones 8% and other renal disorders 15.4%. Twenty-three cases of reflux nephropathy (15.4%) were associated with neural tube defects (NTD) and 20 (13.4%) were caused by infravesical obstruction. CRF caused vesicoureteral reflux associated with NTD and amyloidosis are more frequent in our series compared with west European and Nordic countries.     

End-stage renal disease of the Tunisian child: epidemiology,etiologies, and outcome

From December 1989 to December 1993, 90 children under 15 years were admitted to our department for end-stage renal disease; 9 children were less than 5 years and 28 were aged between 5 and 10 years. The sex ratio (M/F) of the children was 2.1. The estimated incidence of pediatric end-stage renal disease in Tunisia is 7 new cases per year and per million child population under 15 years. The chief etiologies of end-stage renal disease are glomerulonephritis (19%), hereditary nephropathies (29%), and malformative uropathies (13%); 26% of the end-stage renal diseases are of unknown etiologies. Our findings, compared with the European data, show a particularly high frequency of primary hyperoxaluria (13.5%) and an unusual proportion of male subjects. Received April 13, 1995; received in revised form and accepted January 4, 1996     

Early prognostic factors of infants with chronic renal failure caused by renal dysplasia

Renal dysplasia (RD) is a common cause of chronic renal failure (CRF) in children. The evolution towards end-stage renal failure is unpredictable due to the paucity of early prognostic factors. In order to identify early prognostic clinical criteria, we have retrospectively analyzed renal function and growth in 11 infants with RD and CRF from birth up to 4 years of age. Children with obstructive RD were not included. Glomerular filtration rate (GFR) was estimated from Schwartz formula. In infants with a GFR below 15 ml/min per 1.73 m² at 6 months of age (group A, n=5), kidney function did not further improve; 4 reached end-stage renal failure between 8 months and 6 years of age. In contrast, infants with a GFR above 15 ml/min per 1.73 m² at 6 months of age (group B, n=6) experienced a significant improvement in renal function during follow-up, and none required renal replacement therapy. During the first 3 months of life all infants with RD and CRF developed severe growth retardation. Between 6 months and 4 years of age, children from group B grew significantly better than those from group A. In conclusion, our experience suggests that GFR, estimated from Schwartz formula at 6 months of age, is a useful prognostic factor in infants with RD and CRF. Infants with a GFR below 15 ml/min per 1.73 m² are at risk of severe growth delay and the need for early renal replacement therapy, whereas those with a GFR above 15 ml/min per 1.73 m² have a relatively favorable long-term prognosis. Received: 4 October 1999 / Revised: 26 October 2000 / Accepted: 26 October 2000     

Chronic renal failure in pediatrics 1996

 The Nephrology Branch of the Chilean Pediatric Society has greatly influenced the development of government health plans regarding the management and care of patients under 18 years with chronic renal failure (CRF). In order to assess the status of children with CRF in Chile up to 1996, a questionnaire was sent to all pediatric nephrologists in charge of those children. The total sample was of 227 patients under 18 years, giving a national prevalence of 42.5 and an incidence of 5.7 per million inhabitants; of these patients, 50.7% were male, 58.6% over 10 years and 15% younger than 5 years. The most frequent etiologies of CRF were: obstructive uropathy, 18.1%; hypo/dysplasia, 16.7%; reflux nephropathy, 16.7%; and glomerulopathies, 16.3%. Although 48% of patients were on conservative medical treatment, 42.2% of these were in end-stage renal disease, 22.9% were on dialysis, and 29.1% had undergone renal transplantation. Of the dialysis group, 75% were on peritoneal dialysis. Of the transplanted children, 78.8% had normal renal function, but 16.7% returned to dialysis. Three-year graft survival and patient survival were 68% and 94%, respectively. Received: 19 January 1998 / Revised: 6 July 1998 / Accepted: 16 July 1998     

Homocysteine,folate, vitamin B<Subscript>12</Subscript> levels,and<Emphasis Type="Italic"> C677T</Emphasis> MTHFR mutation in children with renal failure

Hyperhomocysteinemia is well documented in chronic renal failure (CRF) and premature and progressive occlusive vascular disease is common in CRF. The combined effects of renal failure, folate and vitamin B(12) levels, and a common mutation (C677T) in the methylenetetrahydrofolate reductase (MTHFR) gene that leads to total plasma homocysteine (tHcy) elevation in CRF children were investigated. Forty-two children (15 females) with CRF, mean age 10.3+/-4.7 years, were included. The mean glomerular filtration rate (GFR) was 37.3+/-16.9 ml/min per 1.73 m(2). The control group comprised 33 children (18 females) with a mean age of 8.6+/-3.4 years. There were 40% of CRF patients with hyperhomocysteinemia. Folate and vitamin B(12) deficiencies were identified in 14% (n=6) and 5% (n=2), respectively, of all patients. On univariate analysis, the tHcy serum concentration was negatively correlated with the plasma folate concentration (P<0.05) in controls, and with GFR (P<0.05) in patients. On multiple regression analysis for the predictors of tHcy serum concentrations, folic and vitamin B(12 )were significant in controls, whereas only GFR was significant in CRF children. In our patients no effect of the MTHFR polymorphism on tHcy levels was seen This result, in addition to the limited number of patients, may partially be explained by the low prevalence of folate deficiency in our patients.     

Primary focal segmental glomerular sclerosis in children: clinical course and prognosis

To review the clinical course and identify prognostic factors, we retrospectively analyzed 92 children with steroid-resistant primary focal segmental glomerulosclerosis (FSGS). The mean age of onset was 80.4+/-42.4 months. The mean follow-up duration was 98.2+/-63.3 months. Eighty-five patients presented with nephrotic syndrome and seven presented with asymptomatic proteinuria. Thirty-three patients were initial responders to steroid treatment (late non-responders) and 59 were initial nonresponders. At last follow-up, 36 patients (39.1%) were in complete remission, and 29 (31.5%) progressed to chronic renal failure (CRF). Renal survival rates at 5, 10, and 15 years were 84, 64, and 53%, respectively. By morphological classification, there were tip variants (6.1%), collapsing variants (10.6%), cellular variants (1.5%), perihilar variants (9.1%), and NOS (not otherwise specified, 72.7%). Among the variants, there were no significant differences in age of onset, degree of proteinuria, response to treatment, or progression to CRF. Poor prognostic factors for CRF included: asymptomatic proteinuria at presentation, initial renal insufficiency, higher segmental sclerosis (%), severe tubulointerstitial change, initial nonresponse, and absence of remission. In the multivariate analysis, an increase in the initial serum creatinine and resistance to treatment were independent risk factors for CRF. A more prolonged use of corticosteroid therapy and early introduction of cyclosporin A (CsA) may improve the prognosis for primary FSGS in patients with initial steroid nonresponsiveness.