Summary A 79-year-old male patient presented with a subacute cauda syndrome caused by an intradural metastasis of the lumbosacral caudate fibres from an adenocarcinoma of the prostate, which had been treated 5 years earlier with external beam radiation therapy.Diagnosis could not be established by repeated magnetic resonance images (MRIs) during a 2-year period of increasingly severe radicular pain. Eventually, a small tumour mass could be visualized on the fourth MRI. Repeated normal serum prostate-specific antigen (PSA) did not hint at a prostate cancer metastasis (range 2.4-5.1 ng ml-1); however, PSA in the cerebrospinal fluid was found to be elevated (29.1 ng ml-'). Empirical radiation therapy of the caudate region did not improve radicular pain. Therefore, an exploratory surgical procedure was conducted, which confirmed the suspicion of an intradural prostate cancer metastasis. In conclusion, PSA in the cerebrospinal fluid provides a useful diagnostic tool for detecting intradural prostate cancer metastasis.Keywords: intradural spinal metastasis; caudate fibres; prostate-specific antigen; prostate cancer; alpha-i-anti-chymotrypsin Clinically occult carcinoma of the prostate has been found at autopsy series in around 70% of men over the age of 80 (Grant et al, 1994;Oschmann et al, 1994). In manifest disease with metastatic spread, skeletal, pulmonary as well as hepatic metastases predominate (Elkin and Mueller, 1954), but intradurally located metastases of the spinal axis are uncommon. Since 1950, 59 patients with histologically confirmed intradural spinal metastasis have been found in scattered reports (Perrin et al, 1982;Chow and McCutcheon, 1996), containing only one case of prostate cancer. It is, however, generally accepted that the blood-brain barrier has to be disrupted in intradural carcinomatosis (Siegal et al, 1987), and therefore that haematogenous dissemination from the primary tumour as well as spinal or epidural metastases by embolization may be the most important and common mechanism for spread of tumour cells into the spinal subrachnoid space (Perrin et al, 1982;Chow and McCutcheon, 1996). CASE REPORTA 79-year old man was evaluated for a 3-year history of increasingly severe radicular pain projecting into the left lumbosacral dermatomes. His medical history included a prostate cancer (T3, NO, MO) diagnosed 5 years earlier by a needle biopsy (moderately differentiated adenocarcinoma of the prostate) and treated with fractionated external beam radiation therapy (45 MeV, 6000 cGy cumulative). Two years later, a radicular pain syndrome developed insidiously, projecting into the left-sided lumbosacral dermatomes [prostate-specific antigen (PSA), prostate-specific acid phosphatase (PAP), cerebro-spinal fluid (CSF) cytology, CSF infection] were repeated every 6 months over the next 2 years with no pathological findings. Radiological investigations using abdominal ultrasound, excretory urography, skeletal scintigraphy and abdominal computerized tomography (CT) could not detect metastatic disease; ev...

Purpose– To investigate the rate of false negative initial cerebral angiography in spontaneous SAH and to ascertain why aneurysms remain undetected. Furthermore to validate CCT in predicting the presence and site of an angiographically missed aneurysm. Methods– Forty‐two patients with spontaneous SAH were investigated, in whom initial cerebral angiography did not reveal any bleeding cause. Repeat‐angiography was performed in all patients 5 to 55 days (mean 15 days) after the bleeding event. All patients underwent CCT scans within 48h after the ictus. Results– In 8 of 42 patients (19%) repeat‐angiography revealed an aneurysm missed on initial angiography. The aneurysms were located on the AcomA (n = 2), the MCA (n = 2), the ACA (n= 1), the PICA (n = 2) and the junction of ICA and PcomA (n = 1). Presumable reasons for missing an aneurysm were spasms detected in four of eight cases on initial angiography and thrombosis of the aneurysm found in two cases at surgery. In two cases, multiple additional views just revealed the aneurysm appearing different in size and shape on repeat‐angiography. CCT blood distribution pattern in four cases indicated presence and site of an aneurysm, while blood distribution was non‐specific in the other four cases. Conclusion– Repeat‐angiography plays an important role in defining the site of an initially occult aneurysm and should be performed in all cases of unexplained SAH. It is of particular importance if vasospasm has compromised the initial angiogram or if one part of the vascular tree is not optimally seen.

No abstract

The perimesencephalic pattern is frequently found in patients with nonaneurysmal subarachnoidal haemorrhage, the prognosis of these patients is excellent. Rarely is the perimesencephalic haemorrhage caused by a ruptured aneurysm. It needs thorough angiographic evaluation.