Improvement of the prognosis for children suffering from hydrocephalus requires prompt diagnosis and reliable indication of surgical treatment. Today, intrauterine hydrocephalus is detectable within the first three months of pregnancy; in infancy, before the cranial sutures have fused, pathological growth of the head is the principal sign confirming together with anatomical examinations (ultrasound, CT scan) the indication of operative treatment. In later childhood, surgical treatment is only definitely indicated by symptoms and morphological examination of clearly active hypertensive hydrocephalus. Intermittently normotensive hydrocephalus (not "normal-pressure-hydrocephalus"!) showing symptoms adapted to childhood, however, often requires exact examination of intracranial pressure dynamics, including quantitative volume provocation test. "Step-by-step-procedure" is advisable (Table III).

According to the hypothesis of Jannetta, an arterial compression of the left root entry zone (REZ) of cranial nerves IX and X by looping arteries could play an important role in the pathogenesis of essential hypertension. In an initial anatomical study, the positions of the left vagus and glossopharyngeal nerves in the skull were radiographically determined in 10 cadavers. By using a pattern of REZ topography developed from this information, the angiographic findings in 107 hypertensive and 100 normotensive patients were then compared retrospectively. In 80% of the angiograms of the hypertensive patients that could be evaluated, an artery crossed the left REZ of cranial nerves IX and X. Most frequently, this was the posterior inferior cerebellar artery (35.3% of cases), followed by the vertebral artery (29.4% of cases) and the anterior inferior artery (19.1% of cases). In 9 cases (13%), both the posterior inferior cerebellar artery and the vertebral artery appeared in the REZ. Frequently, a larger diameter of the left vertebral artery was found. The angiograms of normotensive patients that could be evaluated revealed an artery in the REZ in only 34.5% of cases. Our results support the hypothesis that essential hypertension may be associated with neurovascular compression of the left REZ of cranial nerves IX and X.

In 16 patients with possible disturbed CSF circulation longterm recording of intracranial pressure (ICP) was performed. In 9 of the 16 patients the disorder became evident after the application of an exogenous volume-pressure testing procedure. This procedure enabled a classification into a group I--disturbed CSF circulation (8 patients)--and another group II--normal CSF circulation (8 patients). During a routine Xenon-CT-CBF study all patients of both groups were given to 1 g acetazolamide (DIAMOX) intravenously. Before, during and after the administration of DIAMOX the epidural ICP was continuously measured. An increase in ICP was monitored in all patients. In group I the average initial ICP was 13.6 +/- 7.6 mmHg. The maximum ICP was reached within a time interval of 13.1 +/- 4.5 min after DIAMOX administration. At this time the mean ICP was 36.4 +/- 19.2 mmHg (p less than 0.01). The average initial ICP in group II was 6.3 +/- 4.2 mmHg. The maximum ICP was reached within a time interval of 13.6 +/- 1.1 min. At this time the mean ICP was 11.9 +/- 4.8 mmHg (p less than 0.01). In respect to the maximum ICP both groups were significantly different (p less than 0.01). Despite a considerably rising ICP up to values of about 50 to 70 mmHg in several patients of group I only 1 patient complained of being sick. Presumingly, an ICP elevation caused by vasodilatation would be better tolerated than ICP elevations due to other causes.

Intraoperative sector scanning enables routine imaging of the entire anatomy of intracranial and spinal spaces. Almost all of the pathological processes are exactly localised. It is an essential preliminary condition that fully sterilizable, handy ultrasound probes of 6 to 10 MHz including an accurate instrument- and puncture guide are available. Initial orientation projections similar to CT- or MRI scans are chosen using defined anatomical "landmarks". Cavities filled with aqueous fluids like ventricles, arachnoidal cysts, cystic low-grade gliomas are shown as regions of low echo intensity. Other pathological processes are usually imaged by their hyperecho-characteristics: brain oedema has an increased reflexion compared to normal brain tissue; all brain tumors are diffusely echogenic, the image allows preliminary grading e.g. of gliomas. Vascular processes like aneurysms and angiomas are well defined by wall- and perfusion characteristics. Ultrasound guiding enables even the most minutely detailed microsurgical approach; cysts and disorders of CSF circulation can be treated by simple ultrasound-monitored catheterization. Even stereotactic biopsy can often be replaced. Complications, such as hemorrhages, are recognised immediately.

Clinical studies by Jannetta and others implicated that arterial compression of the root entry zone (REZ) of cranial nerves IX and X at the left ventrolateral medulla may represent an etiological factor for arterial hypertension. Positive therapeutic outcomes with reduction of hypertension in 42 of Jannetta's patients by microsurgical decompression initiated further studies. Experience of our group points in the same direction. Four patients treated by microvascular decompression showed lasting reduction of severe hypertension postoperatively. In our previous comparing postmortem explorations and angiographic studies essential hypertensive patients displayed signs of left sided neurovascular compression in opposition to normotone controls or renal hypertensive patients. By using MR-imaging we are currently developing a method of detecting neurovascular compression syndromes in hypertensive patients suitable for surgical management.