We reviewed 75 pin retrievals after in situ pinning of 35 children with slipped capital femoral epiphysis (SCFE). Four types of pins were used. Ten pins in 8 hips could not be removed because of pin breakage or stripping. Our failure rate was significantly related to pin type and size (p less than 0.039). As compared with the expected values using the chi-square method of analysis, the noncannulated large steel pins and cannulated steel pin groups had fewer failures (p less than 0.001) as compared with the cannulated titanium and noncannulated small steel pin groups. We now avoid using cannulated titanium or noncannulated small pins in treatment of SCFE.