intraventricular haemorrhage . Fibrinolysis in cerebrospinal fluid after

Concentrations of cross linked fibrin degradation products were measured in the cerebrospinal fluid from five 'normal' preterm infants (median 102 ng/ml), four preterm infants with intraventricular haemorrhage (median 315 ng/ ml), and five infants with progressive posthaemorrhagic ventricular dilatation (median 1000 ng/ml). Serial samples of cerebrospinal fluid from one infant showed a peak concentration two weeks after the haemorrhage. Posthaemorrhagic ventricular dilatation (PHVD) occurs after intraventricular haemorrhage (IVH) in preterm infants and is assumed to be caused by small particles of blood clot obstructing the flow of cerebrospinal fluid through the ventricular system, basal cisterns, and arachnoid villi on the surface of the cerebral hemispheres. Little is known about the natural mechanisms by which the body clears clots from the cerebrospinal fluid. Masuda et al concluded that fibrinolytic activity started to increase three to five days after experimental intracerebral haematoma and increased for seven to 10 days, decreasing after 21 to 28 days.' We have carried out a study to determine whether there is evidence of natural fibrinolysis in the cerebrospinal fluid after IVH. used in the multicentre trial of early tapping2 and in a textbook of neonatal neurology.3 Five preterm infants required repeated tapping to control cerebrospinal fluid pressure and excessive head growth. After the first therapeutic lumbar or ventricular tap, fluid was kept frozen for analysis of cross linked fibrin degradation products. In one infant, who had eight lumbar punctures (either for diagnostic or therapeutic purposes) a sample of fluid from each lumbar puncture was kept for analysis of concentrations of cross linked fibrin degradation products. MEASUREMENT OF FIBRINOLYTIC ACTIVITY Assays ofcross linkedfibrin degradation products This assay was performed directly on cerebrospinal fluid using the procedure described by Gaffney et al,4 with the following modification. The polyvinyl plates were coated with a catcher monoclonal antibody, mab NIBn-123 (10 pLg/ ml), which has a similar specificity to the monoclonal antibody used in the original assay; a detector (tag) polyclonal antibody to human fibrinogen (Dakopatts), which was labelled with horseradish peroxidase, was also used. Sensitivity of the assay was in the range 3-5000 ng/ ml. Results The table shows that after IVH the concentraDepartment of Paediatrics and Neonatal Medicine, Hammersmith Hospital, London and Department of Paediatrics, Aker University Hospital, 0514 Oslo 5, Norway A Whitelaw Haematology Department, National Institute for Biological Standards and Control, South Mimms, Hertfordshire L Creiehton Patients and methods COLLECTION OF SAMPLES OF CEREBROSPINAL FLUID Normal preterm infants Because of clinical instability, five preterm infants underwent lumbar puncture (one on two occasions) to exclude infection; they were subsequently found to be free from both infection and intraventricular haemorrhage. For ethical reasons completely healthy infants could not be subjected to lumbar puncture purely for research purposes. Intraventricular haemorrhage Four preterm infants with IVH (without PHVD) visible on cranial ultrasound scan underwent lumbar puncture to exclude infection. P Gaffney Intraventricular haemorrhage progressing to Correspondence to: PHVD Dr Whitelaw. The definition of PHVD was 'ventricular width Accepted 28 February 1991 expanding after IVH to 4 mm over the 97th (Ar(/li Di (lW/il W99166:88-8019) centile' which was the same definition as that Median concentrations of cross linked fibrin degradation products in cerebrospinal fluid in normal infants and those with IVH and PHVD No of Cross linked fibnn infants degradation products (nglml)