The management of the preterm infant often requires rapid assessment of glomerular filtration rate (GFR). We sought to develop a screening test using GFR = kL/Pcr, where GFR is expressed as ml/min/1.73 m2, L is body length in centimeters, Pcr is plasma creatinine concentration, and k is a constant that depends on muscle mass. The value for k in 118 appropriate for gestational age preterm infants (0.34 +/- 0.01 SE) was significantly less than that of full-term infants (0.43 +/- 0.02, P less than 0.001). There was no difference between 12- to 24-hour single-injection inulin clearance and either 0.33 L/Pcr or creatinine clearance in preterm infants. We compared the body habitus of preterm and full-term infants using the assessment of muscle mass from urinary creatinine excretion (UcrV) and from upper arm muscle area (AMA) and volume (AMV), and that of fatness from the sum of five skinfold thickness measurements. During the first year of life, premature infants were found to have a lower percentage of muscle mass than term infants did. On the other hand, they took on a relatively greater amount of subcutaneous fat. There was a very good correlation between AMA or AMV and urinary creatinine excretion (r = 0.91 and 0.94, respectively) in 68 infants with heterogeneous body composition during the first year, indicating the validity of the urinary creatinine measurement. Absolute GFR (ml/min) was also well estimated from AMA or AMV factored by Pcr. We conclude that GFR can be well estimated from 0.33 L/Pcr in preterm infants. The lower value for k reflects the smaller percentage of muscle mass in preterm versus term infants. As a screening test, 1.5 X k or 0.05 L/Pcr predicted low values of GFR with an efficiency of 73%, specificity of 67%, and sensitivity of 88%.