Editor—Neonates who have clinical signs of a bleeding diathesis warrant a coagulation screen to detect an underlying coagulation disorder. If the specimen collected for a clotting profile clots then the test cannot be done and a repeat specimen is required. Often samples that are clotted are not repeated due to the commonly held misconception that for a specimen to clot it must have normal coagulation factor concentrations.

We aimed to test the hypothesis that if a coagulation profile specimen clots then the patient’s clotting profile must be normal. A retrospective analysis was performed, examining all neonatal clotting profile specimens that had clotted, from August 1 1996 to December 26 1998 at the Royal Women’s Hospital, Melbourne, Australia. Clotted specimens were identified and data were collected from the clotting profile of specimens repeated later the same day. The clotting profile included the prothrombin time (PT) (seconds), activated partial thromboplastin time (APTT) (seconds), fibrinogen concentration (g/l) and D-dimer (mg/l). The reference ranges for each parameter were corrected for postnatal and gestational age.

Eighteen clotted specimens were taken from 11 neonates. Only nine were repeated on the same day, two of which clotted again and blood was not re-collected. Of the remaining seven specimens, five had abnormal results. Of those five abnormal results, all had increased PT, APTT, and D-dimer, and two had an accompanying low fibrinogen concentration. In summary, half of the clotted specimens were not repeated, and 71% of repeated specimens were abnormal.

Only 20–30% of most coagulation factors are required for clot formation in vivo,1 and this may also be true for blood collected for analysis. These data support the principle that it is wise not to assume that clotted specimens come from patients with normal clotting profiles. We conclude that if a clotting profile is clinically indicated and the specimen clots, this does not imply normal coagulation, and another specimen should be collected for analysis.