Patterns and time of initiation of dialysis in US children.

The purpose of this communication is to study the clinical patterns and level of residual renal function at the initiation of dialysis in children in the United States of America (US). Data were reviewed for 7,039 children under the age of 20 years and 647,600 adults extracted from the patient incidence report, obtained from the United States Renal Data Systems (USRDS), who were initiated on dialysis between January 1995 and September 2002. Based on pre-defined exclusion criteria, only 4,808 of the 7,093 (67.8%) pediatric entries were included in the analysis. About 6.9% of the entries were not used because of missing data only, 23.3% because of out of range data only, and 2.0% because of both missing and out of range data, a total of 32.2% exclusions. For adults, 570,808 (88.1%) had acceptable data. The percentage of the 4,808 children who were initiated on dialysis with an estimated GFR greater than 10 mL min(-1) per 1.73 m(2) (early start) was 49.6%. Using logistic regression, the factors affecting the probability of an early start were sex, race, type of insurance, region of the country, age at initiation of dialysis, and the year dialysis was initiated. The highest chance of starting dialysis early (0.77) was for a white male, aged 15-19 years, with insurance and residing in the Northwest part of the US The percentage of 4,808 children who initiated dialysis with an estimated GFR less than 5 mL min(-1) per 1.73 m(2) (late start) was 7.3%. The factors affecting the probability of a late start were sex, race, type of insurance, and the year dialysis was initiated. The greatest chance for a late start of dialysis was for black female patients without insurance (0.21). Mean estimated GFR at the start of dialysis was higher for children than for adults (10.7+/-4.6 vs 8.2+/-4.1 mL min(-1) per 1.73 m(2), respectively, P<0.0001). It was concluded that patterns of management of children with ESRD are not uniform among US children.