Spontaneous subarachnoid hemorrhage and negative cerebral panangiography

J Neurosurg 64:537-542, 1986 Spontaneous subarachnoid hemorrhage and negative cerebral panangiography Review of 140 cases MARK S. M. ALEXANDER, M.B.,B.S., P. S. DIAS, M.B., F.R.C.S., F.R.C.S.E.(SN), AND DAVID UTFLEY, M.B., F.R.C.S. Department of Radiodiagnosis, Royal Postgraduate Medical School Hammersmith Hospital and Department of Neurosurgery, Atkinson Morley's Hospital Wimbledon, London, England, and Department of Neurosurger.v, King Faisal Hospital Taif Saudi Arabia ~" One hundred and forty consecutive patients who sustained proven spontaneous subarachnoid hemorrhage (SAH) with negative cerebral panangiography were studied retrospectively. Attention was directed to the presence, amount, and distribution of subarachnoid blood on computerized tomography (CT) scans. It was determined that the finding of blood on CT had a significant association with clinical grade, loss of consciousness, ventricular ratio, fixed ischemic deficit, and total of all complications, but not with epilepsy, hypertension (previously known or detected on admission), treated hydrocephalus, rebleeding, angiographic spasm, and eventual outcome (which was generally excellent on follow-up examination). The distribution of blood, predominantly around the basal cisterns, suggests leakage from ventriculostriate and thalamoperforating vessels as the cause of SAH, and closer study of these vessels is suggested. KEY WORDS 9 subarachnoid hemorrhage 9 computerized tomography 9 prognosis 9 panangiography T HERE are many studies of subarachnoid hemorrhage (SAH) from aneurysms or artefiovenous malformations (AVM's) 3'4' 14,20~22,23,25,28.-9,32 or from other angiographically proven 5,6.36 or identified 34 causes; however, few studies have been published of patients with proven spontaneous SAH who have negative cerebral panangiography. 2'13"24'35 These patients represent a significant subgroup among all cases of SAH (22% of the total cases in the first Cooperative Study, 24 13% in this study). In these cases no feasible mechanism for the SAH has been proposed, and management and advice regarding the future has largely been empirical. Computerized tomography (CT) findings in: these patients have received scant attention, with discussion tending to focus only on the differential findings between cases with and without aneurysms. 32 In previous studies, the relative insensitivity of angiography has been cited to explain the lack of aneurysms or AVM's. 35 However, as the prognosis in this group of patients is so good, 13"24 it appears that a fundamentally different process may be operating. Because of the generally good prognosis, there have been few adequate postmortem studies in these patients and this is partly why the mechanism of the bleeding has remained obscure. In this study of SAH patients with negative panangiography, we have concentrated on the CT findings both in relation to complications and outcome and to the actual distribution of subarachnoid blood on the CT scan. Clinical Material and Methods Of 1041 consecutive patients admitted with proven SAH between January, 1978, and December, 1983, 992 underwent cerebral angiography. Of these patients, 140 had spontaneous SAH, as confirmed by lumbar puncture and CT scanning (95% of cases) or by CT scanning alone (5%), with no evidence of AVM's or aneurysms on high-quality angiography of all intracranial vessels; these 140 patients form the basis for this study. In the past, failure to demonstrate AVM's has been ascribed to the inadequacies of angiography, but with the current advanced techniques this can no longer be held true. J. Neurosurg. / Volume 64/April, 1986 537

M. S. M. Alexander, P. S. Dias and D. Uttley FIG. 1. Computerized tomography (CT) scans with a CT score of 9+ and a ventricular ratio of 1:6.75. There is subarachnoid blood in the anterior interhemispheric fissure, chiasmatic cistern, left Sylvian fissure, interpeduncular cistern, cisterna ambiens, and prepontine cistern. Lateral ventricular blood can also be seen. The importance of complete high-quality cerebral panangiography is paramount, and for this reason we excluded one patient who had incomplete views on fourvessel angiograms. This patient subsequently suffered a fatal rebleed, shown at postmortem examination to be due to a posterior inferior cerebellar artery aneurysm. In 25 cases, angiography was performed twice (13 cases with positive CT findings), in five it was carried out three times (two with positive CT scans), and in one four times in order to ensure full visualization of the vascular tree or to assess vessels after the resolution of spasm. FIG. 2. Upper: Computerized tomography (CT) scans with a CT score of 9+ and a ventricular ratio of 1:7.25. There is blood in the cisterna ambiens, anterior interhemispheric cistern, chiasmatic cistern, interpeduncular cistern, and Sylvian fissure. Lower: CT scans with a score of 9 and a ventricular ratio of 1:6.5. There is blood in the chiasmatic cistern, anterior interhemispheric cistern, Sylvian fissure, interpeduncular cistern, cisterna ambiens, and quadrigeminal cistern. FIG. 3. Computerized tomography scans with a ventricular ratio of 1:3.37 (marked dilatation). 538 J. Neurosurg. / Volume 64/April 1986

SAH and negative panangiography On admission, patients were graded clinically according to the Hunt and Hess system, 17 and any change in grade was noted. Blood pressure on admission, history of hypertension or migraine, loss of consciousness, and presence of epilepsy were recorded. All CT scanning was performed on an EMI 1010 head scanner with a 160 x 160 matrix, using 8-mm thick slices at a 15" angle to the orbitomeatal line, and six images were reproduced in 11 x 8-cm transparencies. All patients underwent CT scanning on admission and were then divided into two groups: one group had blood visible in either the subarachnoid space and/or the ventricles, and the other had no blood demonstrated. Because subarachnoid and ventricular blood resolves over time, 1'~8"2~ there were some patients with negative CT findings who had delayed scanning within 6 to 10 days of SAH (15%) or more than 10 days after the ictus (20%). The amount of subarachnoid blood was quantified on the basis of the "CT score," as described by Gurusinghe and Richardson. 12 In their scoring system, the greatest transverse horizontal diameter (1 mm being equivalent to 3.3 mm in actual terms) of subarachnoid clot on the transparencies is added to the number of vertical slices on which subarachnoid blood is visible. The site of subarachnoid blood in the cisterns and the presence of blood in the ventricles were also recorded (Figs. 1 and 2). Ventricular size was graded according to the method of Vassilouthis and Richardson 33 as a ratio of the width of the lateral ventricles at the level of the foramina of Monro compared to the transverse diameter of the skull at the same level (Fig. 3), as follows: Grade 0 (no dilatation, ratio > 1:6.5), Grade 1 (minimal dilatation, ratio 1:6.4 to 1:5.1), Grade 2 (moderate dilatation, ratio 1:5 to 1:4), and Grade 3 (marked dilatation, ratio > 1:4). The association between CT scan findings and clinical grade, complications (epilepsy, treated hydrocephalus, loss of consciousness, rebleeding, angiographic spasm, and fixed ischemic deficits (Fig. 4)), ventricular ratio, and outcome was assessed using the chi-square test with appropriate adjustment for sample size. The CT score was correlated with complications and known hypertension using the Kendall coefficient of rank correlation. All patients were treated symptomatically and were subjected to bed rest for periods ranging from 3 to 6 weeks. Antifibrinolytic agents were not used. Followup review was possible in 90% of cases, the mean follow-up period being 33 months (range 1 month to 6 years). Outcome was graded as excellent (returned to full activity), good (capable of normal active life with minor disability), fair (returned to active independent life but with disability precluding previous occupation), poor (disability causing dependence), and dead. Results Among the 140 patients, there was an equal incidence of males and females. The mean age in the series was 46.9 years, falling in an approximately Gaussian age distribution (Fig. 5). The attributable and relative risk was calculated for the various complications with regard to the presence or absence of blood on CT scans, and the results are given in Table 1. Parameters that were TABLE i Risk of complications and presence of hypertension related to blood on CT scans* Attributable Risk Relative Complications (/1000 cases) Risk loss of consciousness 127 2.58 epilepsy 38 1.68 fixed ischemic deficits 82 8.20 angiographic vasospasm 54 3.70 treated hydrocephalus 3 1.09 known hypertension 71 1.87 hypertension on admission 129 1.56 * CT = computerized tomography. FIG. 4. Computerized tomography scan showing a left frontoparietal low-density area that was accompanied by a fixed ischemic deficit clinically and an occluded parietal branch of the left middle cerebral artery angiographically. FIG. 5. Sex and age distribution of the 140 patients in this series. TOT = all patients who underwent computerized tomography (CT) scanning; CT+ = patients with positive findings; CT- = patients with negative CT findings. J. Neurosurg. / Volume 64/April 1986 539

M. S. M. Alexander, P. S. Dias and D. Uttley Ii p <0.001 ~..,~.- p < 0.{~.-~ ~,- p < O-GS-~.- 30-8 c~ )D ILOOD zo- ----- 10- O- L O.C. "IT TIT I'V-'V" 1 2 3 FIXED ISCHAEMIC ALL 9 l WORST CLINICAL GRADE mm ~ VENTRIC RATIO IJ DEFICIT COMPLIC FIG. 6. Graph showing the parameters that were significantly related to the presence of blood on computerized tomography (CT) scans. Clinical grading was according to the system of Hunt and Hess./7 LOC = low of consciousness. significantly related to the presence of blood on CT were: clinical grade, loss of consciousness, ventricular ratio, fixed ischemic deficits, and the total of all complications (Fig. 6). On the other hand, no significant relationship was found between blood on CT and epilepsy, hypertension (previously known, or detected on admission), treated hydrocephalus, rebleeding, angiographic spasm, and eventual outcome (excluding death) (Fig. 7). A reanalysis that excluded the negative CT scans obtained either 6 to l0 days or more than l0 days after the ictus did not alter the parameters sufficiently to achieve significance. The CT score 12 for the amount of subarachnoid blood was correlated with the number of patients with complications (fixed ischemic deficit, loss of consciousness, epilepsy, hydrocephalus requiring treatment, and rebleeding). The score was standardized for the number of patients in each CT score group using the Kendall coefficient of rank correlation to give a value of z = 0.60 and a significance level of p < 0.03. A good positive correlation was found between the amount of subarachnoid blood and the occurrence of complications. Using the same method, CT score was correlated with known hypertension and showed no significant relationship (~ = 0.15, p not significant). The distribution of blood on the CT scans is given in Fig. 8. There was a high incidence of blood around the basal cisterns, and in this group five patients, all of whom had blood in the interpeduncular cisterns, also demonstrated angiographic spasm. 7-9"11 Eight patients had blood in the ventricles only. Six of the eight had normal ventricular ratios, and one showed mild and one moderate ventricular dilatation. None required treatment. Half of this group 4 had hypertension undergoing treatment and one, not previously known to be hypertensive, was hypertensive on admission. The suggestion is that ventricular blood alone may be a variant of hypertensive hemorrhage. Eventual outcome was excellent in six and good in two. Only one patient died as a consequence of rebleeding (he bled on four occasions). No site could be found clinically, and at postmortem examination it was believed to be at the basilar artery origin. Finally, angiography showed dysplastic or hypoplas- 100 ~ CT BLOOD NO CT BLOOD z 50 EPILEPSY KNOWN HYPERTENSION ON TREATED REBUEED ANGLO. SPASM EXCELLENT GO00 FAIR HYPERTENSION ADMISSION HYDROCEPHALUS II OUTCOME FIG. 7. Graph showing the parameters that were not significantly related to the presence of blood on computerized tomography (CT) scans. For explanation of outcome see text. DEATH Ir 540 J. Neurosurg. / Volume 64/April, 1986

SAH and negative panangiography -20" 30-9 Ventdcular blood..=. a~ I0- -'-[ S ubarachnoid blo~l VENTRICLES ALL BASAL IPI CA/ LOWER LOWER ONLY CISTERNS PPCIQC AI HF $F FIG. 8. Graph showing the distribution of blood on computerized tomography scans. IP = interpeduncular cistern; CA = cisterna ambiens; PPC = prepontine cistern; QC = quadrigeminal cistern; AIHF = anterior interhemispheric fissure; SF = Sylvian fissure. tic vessels in seven of the CT-positive and seven of the CT-negative patients, and atheroma in six of the CTpositive and six of the CT-negative patients. Discussion In this study we have shown a significant association between the presence of blood on CT and loss of consciousness, clinical grade, ventricular ratio, fixed ischemic deficits, and all complications. We have also demonstrated a significant correlation between the CT score and the complication rate. We could find no significant relationship between blood on CT and epilepsy, hypertension (previously known, or detected on admission), rebleeding, angiographic spasm, and eventual outcome. In view of the lack of difference in the eventual outcome between patients with and without blood on CT, we would suggest that, assuming the absence of evidence to the contrary, no major structural difference exists between the two groups and that the amount of blood in the basal subarachnoid cisterns plays an important role in the development of complications. In the follow-up evaluation of these patients, complete amnesia for the period shortly after the SAH (lasting from days to weeks) was noted, together with headaches and persistent anxiety. Subtle deficits of brain function, irritability, distractability, and perplexability were not uncommon and may suggest ischemia of the limbic lobe structures. 21 This, together with the high incidence of blood in the region of the basal cisterns, suggests leakage from the ventriculostriate and thalamoperforating vessels, some of which supply the limbic system structures. It would seem reasonable to suppose that spasm in these vessels (undetectable on angiography), precipitated by and proportional to the amount of blood in the subarachnoid cisterns, could in turn be responsible for subtle changes in brain function. A case has been reported ~5 in which the SAH occurred through a lamina muscularis deficiency in a small pontine artery. Because of the good prognosis of these patients, there are very few postmortem studies, 1~176 but on the basis of our findings we advise that closer attention be paid to the ventriculostriate and thalamoperforating vessels, both at autopsy in the patients who die of unrelated causes and at angiography, perhaps by using the newer technique of digital subtraction vascular imaging. Recent studies have clearly shown an increased risk of ischemic complications with the use of antifibrinolyric agents. Since the risk of rebleeding is extremely small, we endorse the view that antifibrinolytic therapy is not indicated in this group of patients. 2'19"26"27"3~ Conclusions The good prognosis reported previously in patients with SAH and negative cerebral panangiography has been confirmed in this report. Subtle deficits in these patients may be easily overlooked in a routine neurological examination, but detailed neuropsychological testing may bring to light the nature and extent of any such minor brain dysfunction. A further study is currently in progress to evaluate the neuropsychological status of our patients, and this will be reported in due course. The incidence of complications is related to the presence and amount of subarachnoid blood on CT scans, and the characteristic distribution of blood around the basal cisterns should encourage closer study of the fine ventriculostriate and thalamoperforating vessels as potential sites of SAH. Acknowledgments The authors wish to record their grateful thanks to their neuroradiological and neurosurgical colleagues for their kind cooperation, and to Miss S. Rostron for her help in the preparation of the manuscript. References 1. Adams HP Jr, Kassell NF, Torner JC, et al: CT and clinical correlations in recent aneurysmal subarachnoid hemorrhage: a preliminary report of the Cooperative Aneurysm Study. Neurology 33:98 t-988, 1983 2. B6guelin C, Seiler R: Subarachnoid hemorrhage with normal cerebral panangiography. Neurosurgery 13: 409-411, 1983 3. Bell BA, Kendall BE, Symon L: Computed tomography in aneurysmal subarachnoid hemorrhage. J Neural Neurosurg Psychiatry 43:522-524, 1980 4. Bj6rkesten G, Halonen V: Incidence of intracranial vascular lesions in patients with subarachnoid hemorrhage investigated by four-vessel angiography. J Neurosnrg 23: 29-32, 1965 5. Bryer A, Miller JL: Conn's syndrome presenting as a subarachnoid haemorrhage. A case report. S Aft Med J 62:249-250, 1982 6. Castillo R, Watts C, Pulliam M: Sudden hemorrhage in an acoustic neuroma. Case report. J Neurosurg 56: 417-419, 1982 J. Neurosurg. / Volume 64/April, 1986 541

M. S. M. Alexander, P. S. Dias and D. Uttley 7. Chui M, Battista A, Kricheff II: Vasospasm of the vertebrobasilar system in cases of ruptured intracranial aneurysm. Neurosurgery 12:542-548, 1983 8. du Boulay G: Distribution of spasm in the intracranial arteries after subarachnoid haemorrhage. Acta Radiol (Diagu) 1:257-266, 1963 9. Ecker A, Riemenschneider PA: Angiographic evidence of spasm in cerebral vascular disorders. Neurology 3: 495-502, 1953 10. Eldevik OP, Kristiansen K, TorvikA: Subarachnoid hemorrhage and cerebrovascular spasm. Morphological study of intracranial arteries based on animal experiments and human autopsies. J Neurosurg 55:861-876, 1981 11. Fein JM, Flor WJ, Cohan SL, et al: Sequential changes of vascular ultrastructure in experimental cerebral vasospasm. Myonecrosis of subarachnoid arteries. J Neurosurg 41:49-58, 1974 12. Gurusinghe NT, Richardson AE: The value of computerized tomography in aneurysmal subarachnoid hemorrhage. The concept of the CT score. J Neurosurg 60: 763-770, 1984 13. Hayward RD: Subarachnoid haemorrhage of unknown aetiology. A clinical and radiological study of 51 cases. J Neurol Neurosurg Psychiatry 40:926-931, 1977 14. Hayward RD, O'Reilly GVA: Intracerebral haemorrhage. Accuracy of computerised transverse axial scanning in predicting the underlying aetiology. Lancet 1:1-4, 1976 15. Hochberg FH, Fisher CM, Roberson GH: Subarachnoid hemorrhage caused by rupture of a small superficial artery. Neurology 24:319-321, 1974 16. Hughes JT, Schianchi PM: Cerebral artery spasm. A histological study at necropsy of the blood vessels in cases of subarachnoid hemorrhage. J Neurosurg 48:515-525, 1978 17. Hunt WE, Hess RM: Surgical risk as related to time of intervention in the repair of intracranial aneurysms. J Neurosurg 28:14-20, 1968 18. Imanaga H, Yasamoto M, Jimbo M, et al: Computed tomography in the diagnosis of hemorrhage secondary to intracranial aneurysm. No Shinkei Geka 8:623-631, 1980 19. K~tgstr6m E, Palma L: Influence of antifibrinolytic treatment on the morbidity in patients with subarachnoid hemorrhage. Acta Neurol Scand 48:257-258, 1972 20. Kendall BE, Lee BCP, Clavaria E: Computerised tomography and angiography in subarachnoid haemorrhage. Br J Radiol 49:483-501, 1976 21. Lezak MD: Subtle sequelae of brain damage. Perplexity, distractibility, and fatigue. Am J Phys Med 57:9-15, 1978 22. Liliequist 8, Lindqvist M, Valdimarsson E: Computed tomography and subarachnoid hemorrhage. Neuroradiology 14:21-26, 1977 23. Locksley HB: Report on the Cooperative Study of Intracranial Aneurysms and Subarachnoid Hemorrhage. Section V, Part I. Natural history of subarachnoid hemorrhage, intracranial aneurysms and arteriovenous malformations based on 6368 cases in the Cooperative Study. J Neurosurg 25:219-239, 1966 24. Nishioka H, Torner JC, Graf CJ, et al: Cooperative Study of Intracranial Aneurysms and Subarachnoid Hemorrhage: a long-term prognostic study. III. Subarachnoid hemorrhage of undetermined etiology. Arch Neuroi 41: 1147-1151, 1984 25. Pakarinen S: Incidence, aetiology, and prognosis of primary subarachnoid haemorrhage. A study based on 589 cases diagnosed in a defined urban population during a defined period. Acta Neurol Scand 43 (Suppl 29):1-28, 1967 26. Rydin E, Lundberg PO: Tranexamic acid and intracranial thrombosis. Lancet 2:49, 1976 (Letter) 27. Schisano G: The use of antifibrinolytic drugs in aneurysreal subarachnoid hemorrhage. Surg Neurol 10:217-222, 1978 28. Schneider GH, Ladurner G, Sager WD: [Computed tomography in primary subarachnoid hemorrhage.] Routgenblatter 36:256-259, 1983 (Ger) 29. Scotti G, Ethier R, Melanqon D, et al: Computed tomography in the evaluation of intracranial aneurysms and subarachnoid hemorrhage. Radiology 123:85-90, 1977 30. Smith RR, Clower BR, Peeler DF Jr, et al: The angiopathy of subarachnoid hemorrhage: angiographic and morphologic correlates. Stroke 14:240-245, 1983 31. Sonntag VKH, Stein BM: Arteriopathic complications during treatment of subarachnoid hemorrhage with epsilon-aminocaproic acid. J Neurosurg 40:480-485, 1974 32. van Gijn J, van Dongen KJ: Computerized tomography in subarachnoid hemorrhage: difference between patients with and without an aneurysm on angiography. Neurology 30:538-539, 1980 33. Vassilouthis J, Richardson AE: Ventricular dilatation and communicating hydrocephalus following spontaneous subarachnoid hemorrhage. J Neurosurg 51:341-351, 1979 34. Vincent FM, Zimmerman JE, Auer TC, et al: Subarachnoid hemorrhage -- the initial manifestation of bacterial endocarditis. Report of a case with negative arteriography and computed tomography. Neurosurgery 7:488-490, 1980 35. West HH, Mani RL, Eisenberg RL, et al: Normal cerebral arteriography in patients with spontaneous subarachnoid hemorrhage. Neurology 27:592-594, 1977 36. Yonemitsu T, Niizuma H, Kodama N, et al: Acoustic neurinoma presenting as subarachnoid hemorrhage. Surg Neurol 20:125-130, 1983 Manuscript received July 19, 1985. Address reprint requests to: David Uttley, M.B., F.R.C.S., Consultant Neurosurgeon, Atkinson Morley's Hospital, Wimbledon, London SW20 ONE, England. 542 J. Neurosurg. / Volume 64~April, 1986