Abstract
The catheter semiquantitative culture roll tip technique has been validated as a discriminator between non-catheter-related bacteremias and catheter-related bacteremias (CRBs) caused by Staphylococcus species. However, this technique has not been specifically validated when used for the evaluation of catheters infected with organisms other than staphylococci. We reviewed catheters that had been submitted for semiquantitative roll tip culture as well as hospital records to determine clinical correlates of infection. Local infection and CRB were defined by standard criteria. Catheter-related sepsis (CRS) was defined as fever, leukocytosis, or hypotension which resolved with catheter removal, without another source of infection. For 195 catheters from 93 patients, gram-negative rods and enterococci were present on 36, fungi were on 25, Corynebacterium species were on 5, Bacillus species were on 3, Staphylococcus species were on 79, and 41 demonstrated no growth. Of 21 episodes of CRB or CRS due to nonstaphylococcal organisms, only 1 (questionable) episode was due to a catheter with < 15 CFU (P < 0.05). Eleven of these 21 episodes of CRB or CRS were due to gram-negative rods and enterococci, of which only the questionable episode was due to a catheter with < 15 CFU. Nine of these 21 episodes of CRB or CRS were due to fungi, none of which were associated with a catheter with < 15 CFU. The data for Staphylococcus species recapitulated published data (none of 21 CRB or CRS episodes were associated with catheters with < 15 CFU) and validated this retrospective technique. The data presented in this study validate the use of the semiquantitative culture technique for the evaluation of catheter-related infections caused by organisms other than staphylococci.