[Serious delayed intracranial complications after mild traumatic brain injury in oral anticoagulant use]

BACKGROUND

Patients with mild traumatic brain injury (TBI) who use anticoagulants prior to injury have an increased risk of intracranial complications. Sometimes these complications are delayed, even if the initial CT scan of the head is normal.

CASE DESCRIPTION

An 84-year-old woman who was using acenocoumarol presented elsewhere with mild TBI. She had no focal neurological deficit. The initial CT scan revealed no abnormalities and the patient was discharged home. That evening she had diffuse headache. The next day she was found with a reduced level of consciousness and was brought to our hospital. Her INR was 9.0 and a new CT scan showed an acute, left-sided subdural haematoma with a large mass effect.

CONCLUSION

Serious delayed intracranial complications in patients with mild TBI who use anticoagulants are rare. In these patients INR measurement and a CT scan of the head are always indicated. Admission for observation may be considered. On discharge it is necessary to give clear instructions about warning symptoms.

Safety and efficacy of frameless and frame-based intracranial biopsy techniques

BACKGROUND

Frameless stereotaxy or neuronavigation has evolved into a feasible technology to acquire intracranial biopsies with good accuracy and little mortality. However, few studies have evaluated the diagnostic yield, morbidity, and mortality of this technique as compared to the established standard of frame-based stereotactic brain biopsy. We report our experience of a large number of procedures performed with one or other technique.

PATIENTS AND METHODS

We retrospectively assessed 465 consecutive biopsies done over a ten-year time span; Data from 391 biopsies (227 frame-based and 164 frameless) were available for analysis. Patient demographics, peri-operative characteristics, and histological diagnosis were reviewed and then information was analysed to identify factors associated with the biopsy not yielding a diagnosis and of it being followed by death.

RESULTS

On average, nine tissue samples were taken with either stereotaxy technique. Overall, the biopsy led to a diagnosis on 89.4% of occasions. No differences were found between the two biopsy procedures. In a multiple regression analysis, it was found that left-sided lesions were less likely to result in a non-diagnostic tissue sample (p = 0.023), and cerebellar lesions showed a high risk of negative histology (p = 0.006). Postoperative complications were seen after 12.1% of biopsies, including 15 symptomatic haemorrhages (3.8%). There was not a difference between the rates of complication after either a frame-based or a frameless biopsy. Overall, peri-operative complications (p = 0.030) and deep-seated lesions (p = 0.060) increased the risk of biopsy-related death. Symptomatic haemorrhages resulting in death (1.5% of all biopsies) were more frequently seen after biopsy of a fronto-temporally located lesion (p = 0.007) and in patients with a histologically confirmed lymphoma (p = 0.039).

CONCLUSIONS

The diagnostic yield, complication rates, and biopsy-related mortality did not differ between a frameless biopsy technique and the established frame-based technique. The site of the lesion and the occurrence of a peri-operative complication were associated with the likelihood of failure to achieve a diagnosis and with death after biopsy. We believe that using intraoperative frozen section or cytologic smear histology is essential during a stereotactic biopsy in order to increase the diagnostic yield and to limit the number of biopsy specimens that need to be taken.

Brain parenchyma/pO2 catheter interface: a histopathological study in the rat

Local cerebral oxygenation can be monitored continuously using an intraparenchymal Clark-type pO2 sensitive catheter. Measured values of brain tissue pO2 (PbrO2) not only depend on the clinically interesting balance between oxygen offer and demand, but also on catheter properties and characteristics of the probe tissue interface. Microdamage surrounding pO2-sensitive needles, inserted into various tissues, has been reported; we evaluated histologic changes at the probe tissue interface after insertion of pO2 probes, suitable for clinical use, in the rat brain. The effect of insertion of the probe itself (mechanical damage), the application of micropotential during the measurements, and the effect of time was evaluated using digital image analysis of H&E-stained histological slices. Surrounding the probe tract, a zone of edema with an average radius of 126.8 microm was seen; microhemorrhages with an average surface area of 56.2 x 10(3) microm2 were observed in nearly all cases. The area of edema and the presence of microhemorrhages were not influenced by performed measurements or by time. Intraventricular blood was observed in 10 of 19 rats studied. Measured low PbrO2 values were related to the presence of a microhemorrhage in either probe tract or ventricles. Tissue damage due to the measurements is negligible, and the amount of edema itself does not influence the accuracy or response time of the pO2 probe. Low PbrO2 readings, however, could be caused by local microhemorrhages, undetectable on CT or MRI.