(0025-7974 /86/6502-0082$02.00/0 Copyright © 1986 by The Williams & Wilkins Co. Vol. 65, No. 2 Printed in U.S.A. Septic Thrombosis of the Dural Venous Sinuses FREDERICK S. SOUTHWICK, M.D.', E. P. RICHARDSON JR., M.D. AND MORTON N. SWARTZ, M.D. Introduction In the preantibiotic era, uncontrolled infections of the facial area frequently led to septic thrombosis of one or more of the intracranial venous sinuses. The consequences of serious impairment of venous drainage of the central nervous system were often disastrous, resulting in extensive neurological se- quelae or death. Older medical literature (before 1945) frequently reported this type of paramenin- geal infection (3, 19, 22, 25, 27, 46, 71, 77, 121-123, 127), often describing marked neurological impair- ment developing with the onset of these illnesses. Since the introduction of antibiotics and intrave- nous fluid administration such complications have become rarer, and the neurological manifestations have often been more subtle. In order to define more clearly in the current era the symptoms, signs, and laboratory findings associated with these dis- orders, we have retrospectively reviewed all diag- nosed cases of septic dural venous sinus thrombosis at the Massachusetts General Hospital from 1948 to 1984. Anatomic Considerations Before considering the various types of major dural venous sinus thrombosis, it would be helpful to recall the anatomy of the venous drainage system of the brain (Figs. 1 and 2). The superior sagittal sinus is the largest of the intracerebral venous sinuses. It receives blood in its anterior segments from the nasal septum via the ethmoidal veins. This sinus is triangular on cross-section, increasing progressively in caliber in the anteroposterior di- rection. It receives blood from two groups of veins as it progresses toward the occiput: the frontal, From the Department of Medicine, Harvard Medical School, Medical Services (Infectious Disease Unit), Massachusetts Gen- eral Hospital, Boston, Massachusetts. 1 Present address: Infectious Disease Section, Johnson Pavil- ion University of Pennsylvania School of Medicine, Philadel- phia, PA 19104. Address reprint request to: Frederick S. Southwick, 567 John- son Pavilion, University of Pennsylvania School of Medicine, Philadelphia, PA 19104. 82 parietal and occipital superior cerebral veins and the diploic veins, which communicate with the men- ingeal veins. Also, arachnoid granulations, respon- sible for cerebrospinal fluid uptake, project into this large venous channel. This sinus anastomoses with a number of other sinuses including the infe- rior sagittal, straight, and lateral (transverse) si- nuses (43, 62). The lateral sinuses begin at the internal occipital protuberance. In 50 percent of individuals the right lateral sinus is larger than the left and is a direct continuation of the superior sagittal sinus. In this circumstance the left lateral sinus receives blood from the straight sinus. In other individuals a left dominant drainage is present; that is, the superior sagittal sinus drains into the left and the straight sinus into the right lateral sinus. In others the four vessels combine to form a true confluens, and blood from both the superior sagittal and straight sinuses enters both lateral sinuses. Occasionally, only one lateral sinus is present and is responsible for drain- age of both of these venous sinuses (135). Once the lateral sinus leaves the tentorium, it becomes the sigmoid sinus which curves inferiorly and medially toward the jugular vein. In its early course the lateral sinus passes beneath the mastoid air cells contained in the parietal and temporal bones. The lateral sinuses receive small veins from both the middle ear and mastoid air cells (43, 62). At the base of the skull, inferior to the superior sagittal sinus are the cavernous sinuses. These si- nuses are irregular in shape and contain multiple trabeculae. Each cavernous sinus is situated lateral and superior to the sphenoid air sinus, lying on either side of the sella turcica. The two cavernous sinuses are connected by the intercavernous sinuses which pass anteriorly and posteriorly to the sella turcica and pituitary gland. Several cranial nerves are attached by fibrous sheaths along the lateral walls of the cavernous sinuses (Fig. 3). The oculo- motor nerve (III) is most superior followed by the trochlear nerve (IV), and the ophthalmic and max- illary branches of the trigeminal (V) nerve. The abducens nerve (VI) is located more medially, closer to the sphenoid sinus. The internal carotid artery SEPTIC DURAL VENOUS SINUS THROMBOSIS 83 SUP SAGITTAL, = i ~ GE, y i! AS ff GREAT ven OF GALEN FALX CEREBRI STRAIGHT SINUS ~-: CAVERNOUS SINUS . SPHENOID SINUS} SUP PETROSAL SINUS y INF. PETROSAL SINUS TENTORIUM CEREBELL! ANTERIOR and POSTERIOR INTERCAVERNOUS SINUS BULB OF EYE SUP PETROSAL SINUS = MASTOID AIR CELLS R. SIGMOID SINUS “ss INF. SAGITTAL SINUS \ H R. TRANSVERSE SINUS ‘ SUP. SAGITTAL SINUS Fic. 1. Lateral cross-section of the skull demonstrating the major dural ve- nous sinuses. Note the close proximity of the cavernous sinus to the sphenoid air sinus. The anterior segment of the superior sagittal sinus is near the frontal air sinus. Fic. 2. Horizontal cross-section of the skull demonstrating the locations of the major dural venous sinuses, as well as the superior ophthalmic vein. Note the close proximity of the mastoid air cells to the sigmoid sinus (right side). with its surrounding sympathetic plexus is also superior petrosal sinuses which drain into the sig- more medially located. Blood bathes all these struc- —_ moid sinuses (9, 43, 49, 62, 71). tures. The cavernous sinuses receive blood from the The cerebral veins and venous sinuses have no facial veins via the superior and inferior ophthalmic _valves; therefore, blood within them can flow in veins. In turn, they drain into both inferior petrosal _ either direction depending on pressure gradients. sinuses, which enter the jugular veins, and into the _In addition to the larger vessels just described, there 84 SOUTHWICK, RICHARDSON, AND SWARTZ INTERNAL CAROTID A. i HYPOPHYSIS H ‘ tf SPHENOID BRANCH é SINUS We NIN “ Fic. 3. A-P vertical cross-section through the right cavernous sinus. Multiple cranial nerves as well as the internal carotid artery are shown passing through this venous sinus. Just medial to the cavernous sinus is the hypophysis. are many smaller venous channels which, with time, may enlarge in response to occlusion of the larger venous sinuses or veins, and serve as important routes for collateral circulation (43, 62). Methods Criteria for inclusion in the various categories of major dural venous sinus thrombosis were established: a. Evidence of an infectious process abutting on or within the sinus. b. Evidence of thrombotic occlusion of any one of the three major dural venous sinuses. 1. Cavernous-sinus thrombosis: autopsy evidence of pre- mortem thrombosis or angiographic findings consistent with occlusion of this sinus. In the absence of either of these criteria, cases were included based on clinical criteria if proptosis, che- mosis, and ptosis as well as III, IV, and VI cranial nerve paralysis began in one eye and subsequently developed in the other eye. 2. Lateral-sinus thrombosis: neurological manifestations compatible with this diagnosis, and surgical or autopsy evidence of premortem thrombosis or angiographic findings consistent with occlusion of this sinus. 3. Superior sagittal-sinus thrombosis: neurological manifes- tations compatible with this diagnosis, and either autopsy evi- dence of premortem thrombosis or angiographic findings consis- tent with occlusion of this sinus. Identical criteria were used in the review of previous case reports. Septic Cavernous-Sinus Thrombosis Results and discussion Review of case reports from 1940 to 1984 (1, 2, 8, 10, 14, 15, 18, 20, 21, 26, 28, 30, 32, 33, 35, 36, 38— 40, 42, 44, 47, 49, 50, 58, 61, 64, 67, 69, 74-76, 79, 81, 83, 87, 89, 90, 92, 94, 96, 98, 99, 102-105, 108, 109, 115, 117, 128-134, 137-139, 141) as well as our own clinical experience emphasize a number of important changes in septic cavernous-sinus thrombosis since the development of antibiotics. Primary site of infection (Tables 1 and 2): In the preantibiotic era infections of the medial third of the face, including the nose, orbits, tonsils, and soft palate, most commonly resulted in septic throm- bosis of the cavernous sinus. Bacteria readily en- tered the facial veins or pterygoid plexus, reaching the cavernous sinus via the superior or inferior ophthalmic veins. In addition, infections of the ear not infrequently spread by emissary veins from the mastoid air cells to the sigmoid sinuses and later reached the cavernous sinuses by way of the inferior petrosal sinuses (19, 25, 27, 46, 71, 127). Previous antibiotic-era cases have also emphasized the as- sociation of facial infections with this complication. Approximately 60% of patients with septic cavern- ous-sinus thrombosis from 1940-1960 had a preced- ing facial infection (Table 2). In the early antibiotic era, 1940-1945, otitis media was the second most common site of infection predisposing to this com- plication. In the more recent antibiotic era, 1961 to 1984, the sphenoid and ethmoid sinuses have be- come more frequent primary sites of infection pre- disposing to septic cavernous sinus thrombosis. Bacteria from these air sinuses can spread to the cavernous sinus by small emissary veins or by direct contiguity. Our own patients emphasize this asso- ciation; in all eight of our cases the primary site leading to septic cavernous-sinus infection was the sphenoid and/or ethmoid sinuses. As opposed to facial infections that are easily recognized and have been aggressively treated with antibiotics at our hospital, sphenoid sinusitis has often been misdi- agnosed (70). In all eight of our cases this diagnosis was not entertained on initial evaluation, leading to delay in antibiotic therapy and to spread of infection. Early recognition of the clinical symp- toms and signs of sphenoid air-sinus infection and initiation of early treatment of this disease is crit- ical for preventing the severe morbidity and high mortality of cavernous-sinus thrombosis (70). Symptoms and history of predisposing illness (Ta- ble 3): Although headache was rarely mentioned in early descriptions of cavernous-sinus thrombosis, this was the most prominent early symptom in the majority of our patients as well as in approximately half of previous antibiotic-era cases. This complaint provoked seven of our eight patients to seek medical attention before the onset of orbital swelling. Pain was generally restricted to areas innervated by the ophthalmic and/or maxillary branches of the fifth nerve, the frontal and retroorbital areas being the SEPTIC DURAL VENOUS SINUS THROMBOSIS tured £4 TA Jueqsts1Eg IA 4 Pra queqsisieg seared ~forey Yel ‘ede Y PUNE Asyed @Atou Al Wegsts1eg Aep 431 pod Ap WL ped Aup WeT Ped uorssrut “pe soye aq 08 Pe am07ngQ Auroyapt -ouayds pus prow -q30 ‘joorueydure adoro “uypoyeu “Ay Jooruayduresopyo wolssardm09 -Op BBs0} 10L10}80d sjooruayqduresoyyo “urpAmI0IY AIA A’. quewjzBe1], a ‘TA BisaqyseredAy ‘AjU0 Y sorsqed oazau [A ‘AI ‘Hely eatjoves Ajystd -8nys pidnd y ‘ATU Y -nd y eanoeeruou "7 < YW sisourayo Zuruooryeq,, pue sisojdoid ‘sarsjed aazau *A tA ‘TA ‘AI ‘TIT Te2078[8q Seto <— ‘Buynig Os1p pra puny shypexaqeyiq sis0ydord sTexeyeniq A poe", vysayyeqodAy ‘Keyed 1A, ‘ALT eaeepg ‘AreyIe7] ‘dnd 3 pereyp earyover -uou ‘yy stsojdoid ‘sets “Ted oazou "A ‘TA ‘AI ‘III Tezozep1q teuio <— ABreqie] you ys teurepelided "J :rpuny faarjowar Alqersnys ‘dnd aarou '] < Y‘IA ‘AI ‘Ill Jeraqe]Iq SBUIOD <— Aare] 18 poprs-] ssrsarodraroy ‘] ‘udts Zru1ey + ‘eurepe ~{ltded 743s ‘Tpung ‘sie Tysurqeg “y ‘yoou Jy9s ‘uorsedu0o gnouea sIpuny “Yy < J ‘ets -ojdord ezeyeyiq ‘ ‘sareped eazau TA “T‘A ‘AI ‘IIT Tereyeiq Seu09 <— ABrey Ie] sSurpuy [eoSojomnen, 1y8u = YY] = 7 “ssoUT[! 30 180K , snus plousydg proueyds Seseosqe 43001, snus plousydg snus plouaydg snuys prournyg Areyixeu “pio -ueyds plomyyg sesnuls pio wads ‘proud snuts prloueydg woroezut Jo oye [eNTU (8261) _ A/AK LE 8 (0861) - W/4 1 ‘L (pooyq) “da (¥861) wnusaqo0gosny W/24 18 ‘9 (8261) _ W/24 68 ‘g (snurs prouryya) (8961) de sndoztyay a/1k 99 + (ureaq ‘enute Areqirxwur) $900003de138 (W961) H dnoig A/K GT € (pootq) (6r61) snaind ‘§ A/IK 61 % (piousyds ‘ASO ‘poora) snaunp (S961) snao000j6ydoIg G/1K ZL, ‘tT weruesiQ xeg/e8y ese (ses80 HDA) soquiosy) snuys-snousleawo oFdos Jo sornjea] [WOU *L ATAVL 86 SOUTHWICK, RICHARDSON, AND SWARTZ TABLE 2. Initiating infections in cavernous-sinus thrombosis Literature : . Current series . 1946- 1961- 1970- Total Primary site of (8 cases) 1940-1945 1960 1970 1984 (96 cases) infection N (27 cases) No. (% 0. No. (%) (24 cases) (23 cases) —«_ (14 cases) 0. (%) 0. No. (%) No. (%) No. (%) Facial 0 16 (60) 16 (67) 8 (35) 7 (50) 47 (49) Sinusitis (ethmoid or 8 4 (15) 7 (29) 6 (26) 2(14) 27 (28) sphenoid) Dental infection lt 2(7) 1(4) 2(9) 3 (22) 9 (9) Otitis media 0 5 (18) 0 (0) 14) 1(7) 7(7) Other 0 0 (0) 0 (0) 3 (13) 0 (0) 3 (3) Undetermined 0 0 (0) 0 (0) 3 (13) 1(7) 4(4) * Dates of case rts, repo: + Case 7 had both a dental abscess and sphenoid sinus infection. most frequent locations of pain. Retroorbital pain was associated with tearing of the eye in two of our patients, suggesting the diagnosis of ocular mi- graine. Pain was usually sharp and steadily increas- ing in severity. In many instances this symptom probably reflected sphenoid-sinus infection (70). Headache generally preceded fever and perior- bital edema by several days. In two of our patients with more indolent presentations, headache per- sisted for 6 weeks and 6 months before eye signs developed (Cases 7 and 8). Periorbital edema often initially affected only one eye; within 24 to 48 hours the other eye also became involved as the infection spread through the intercavernous sinuses. In four of our patients (Cases 1-4) as well as in other fatal cases, eye complaints were quickly followed by changes in mental status: drowsiness, confusion, inappropriate speech and behavior. Stratification of antibiotic-era cases by year demonstrated no significant change in the frequency of specific symptoms over the 44 years reviewed in this study. A history of predisposing illness was infrequent in the preantibiotic era. With the introduction of antibiotics the association of cavernous-sinus infec- tion with chronic illness has become more frequent (28%), diabetes mellitus and chronic sinusitis being most common. Physical findings (Tables 1 and 3): The classic findings, including ptosis, proptosis, chemosis, and ocular-muscle palsies, were noted in the majority of modern cases, including ours. Exophthalmos and chemosis, manifestations of edema of the perior- bital tissues, were undoubtedly the result of oph- thalmic vein occlusion. Chemosis was a particularly prominent finding in our Case 6. This abnormality developed before other findings, suggesting an ini- tial diagnosis of allergic blepharitis. Generally, che- mosis and proptosis were rapidly followed by the development of impaired eye movements. Function of all of the nerves to the eye muscles (III, IV, VI) was impaired, probably because these nerves are exposed to inflammatory exudate as they pass through the cavernous sinuses. Less frequently, dilated or sluggishly reactive pupils have been de- scribed as one of the manifestations of third-nerve involvement. In 5 of 88 previous cases, isolated sixth-nerve dysfunction has been reported in asso- ciation with thrombosis of this sinus. In two of our cases (Cases 7 and 8), sixth nerve impairment pre- ceded other ocular pareses. This greater suscepti- bility to injury may have an anatomic basis. The third and fourth cranial nerves pass along the lat- eral aspect of the cavernous sinuses and are sur- rounded by a thick fibrous sheath, which may pro- tect them from damage by inflammatory exudate. The sixth nerve, on the other hand, is usually located medially, near the carotid artery, and may be surrounded by blood (49). The ophthalmic and maxillary branches of the fifth cranial nerve also pass along the lateral aspect of the cavernous sinus. Therefore, one would expect fifth-nerve sensory deficits to be as common as eye- muscle dysfunction; however, previous reports have only rarely noted such abnormalities (16/87, 18%). When carefully sought, hyper- or hypoesthesia of the dermatomes served by these branches was found to be present in all but one of our cases. Dysfunction of the ophthalmic and maxillary branches of the trigeminal nerve (including a de- pressed corneal response) should be specifically looked for, since a functional deficit of this nerve in the appropriate setting often indicates inflam- mation of the cavernous sinus. Other frequently observed abnormal physical findings include decreased visual acuity, papill- edema or dilated tortuous retinal veins, nuchal ri- gidity (generally associated with inflammatory cells in the cerebrospinal fluid [CSF]), and lethargy or coma. Deficits in visual acuity (21/95, 22%) fre- quently progressed to blindness (15/95, 16%). The causes of this complication include corneal ulcera- tion from failure of lid closure (36), occlusion of the central retinal artery by pressure at the orbital apex SEPTIC DURAL VENOUS SINUS THROMBOSIS 87 TABLE 3. History and findings in cavernous-sinus thrombosis Current Literature rotq) series (88 (96 ) (8 cases) cases)* %t No. No.t History Periorbital swelling 6 54/74 13 He he 7 36/74 52 Drowsiness 3 20/74 28 Diplopia 1 9/74 12 Eye tearing; photo- 2 2/74 5 phobit Ptosis 1 3/74 5 Predisposing illness 2 21/74 28 Physical findings Fever 7 78/82 94 Ptosis, proptosis, 6 83/86 95 and chemosis III, IV, VI nerve pal- 8 74/85 88 sies Abnormal fundi 5 53/81 65 (papilledema or venous engorge- ment) Lethargy — coma 5 47/87 55 Dilated or sluggishly 3 26/83 32 reactive pupils Decreased visual 1 20/87 22 acuity Fifth nerve dysfunc- 7 16/87 24 tion Abnormal ENT 5 33/87 40 exam Nuchal rigidity 4 34/87 40 Seizures 1 8/88 9 Hemiparesis 3 6/88 9 Laboratory findings Peripheral WBC el- 5 56/59 91 evated CSF pressure ele- 3 11/19 52 vated CSF analysis: 3 10/34 31 meningitis for- mula 5 18/34 55 formula Abnormal sinus X- 6/7 6/33 30 : abnor- 4/4 9/11 87 mal carotid artery Orbital venogram 2/2 Yi 100 abnormal CT scan abnormal 2/4 2/6 40 Brain scan (radionu- 0/1 0/10 0 clide) abnormal CXR§: rounded 1 1/88 8 opacifications * Reported cases in English literature (1940-1984) in which history, physical, laboratory findings could be analyzed. ith abnormality/total no. of cases evaluable. +t Calculated by adding numerators from current and litera- ture cases and dividing by the sum of the denominators for each category. § Chest radiograph. (99), emboli (77), or intracavernous internal carotid arteritis (75). Finally, high intraocular pressure may compromise the posterior ciliary circulation to the optic nerve and produce ischemic optic neurop- athy (36). Abnormalities in the ears, nose, or throat were also frequent (38/95, 40%). Findings have included purulent nasal or posterior pharyngeal exudate, inflammation of the nasal turbinates, and tenderness of the frontal or maxillary sinuses. One prominent finding emphasized in older case reports, palpable engorged frontal veins, was observed in only one antibiotic-era case (83). As found in our analysis of symptoms, stratification of cases by year demonstrated no significant change in the fre- quency of specific physical findings over the 44 years reviewed in our study. In the preantibiotic era, Eagleton (27) and Grove (46) described a slowly progressive form of cavern- ous-sinus thrombosis associated with otitis media, and with spread of infection and thrombosis via the superior petrosal sinus to the posterior section of the cavernous sinus; modern reports, however, have rarely described such subacute or chronic cases (15, 117). Two of our cases (Cases 7 and 8) had pro- longed symptomatic periods and very slow progres- sion of their illness. In both cases the first objective neurological deficit was unilateral sixth-nerve palsy, followed by third-nerve palsy in Case 7 and fourth-nerve dysfunction in Case 8. Unlike most of our acute cases, both of these patients survived. A number of other diseases can mimic cavernous- sinus thrombosis clinically. The differential diag- nosis of this disorder includes other causes of peri- orbital edema: orbital cellulitis, intraorbital abscess (99), allergic blepharitis (Case 6), intracavernous carotid artery aneurysm or arteriovenous fistula (91), idiopathic granulomatous inflamation of the superior orbital fissures and cavernous sinus (To- losa-Hunt syndrome) (55, 56), polyarteritis nodosa associated with cerebral venous sinus thrombosis (Cogan syndrome) (40), nasopharyngeal tumor, meningioma, and trauma. The most common proc- ess mimicking cavernous-sinus thrombophlebitis is orbital cellulitis. Several clinical findings help dif- ferentiate these two disorders. Generally, patients with cavernous sinus thrombophlebitis appear more toxic, and are more likely to have pupillary abnor- malities, visual loss, papilledema or dilated retinal veins, deficits of the fifth cranial nerve, bilateral ocular involvement, and inflammatory cells in the CSF (99). Laboratory findings (Tables 1, 3, and 4): Inflam- matory cells were found in the spinal fluids of 83% of patients in previous case reports and in the CSF of all of our cases (Table 3). The composition of the CSF was consistent with a diagnosis of either 88 SOUTHWICK, RICHARDSON, AND SWARTZ TABLE 4. Bacteriology of septic cavernous-sinus thrombosis Current series . . Literature Total Bacterial species (culianes cbisined (69 cases) (64 canes) No. No. No. (%) Staphylococcus 2 42 44 (69) aureus Streptococcus 1 10 11 (17) species Pneumococcus 0 3 3 (5) Gram-negative 0 3 3 (5) Bacteroides spe- ti) 1 1(2) cies Fusobacterium 1 0 1 (2) species Other 1 0 1(2) purulent meningitis (a predominance of polymor- phonuclear leukocytes, low glucose, and elevated protein) or a parameningeal infection (moderate numbers of polymorphonuclear leukocytes often mixed with mononuclear cells, normal glucose, and slightly elevated protein). As in the preantibiotic era, Staphylococcus aureus has remained the major pathogen associated with this disease. Less frequently, streptococci and pneu- mococci have been the infecting organisms. Gram- negative organisms have rarely been implicated in septic cavernous sinus thrombosis. Except for the finding of anaerobes in two recent cases (39, Case 6) there has been no significant change in the bacteriology of this disease over the 44 years of our study. Radiologic findings (Table 3): Sinus films are indicated, judging by our experience, in all patients with severe headache of unclear etiology, especially if pain is localized to the frontal or periorbital areas. These studies should include tomograms of the sphenoid sinus; although routine sinus views dem- onstrated changes consistent with sinusitis in three of our cases (Cases 2, 4, and 5), in four others routine views failed to demonstrate marked sphen- oid sinus abnormalities later found at autopsy (Case 3) or noted by tomography (Cases 6-8). Radiologic changes in positive studies have included opacifi- cation of the sphenoid sinus and sclerosis of its bony margins. Computed axial tomography has been reported in only a few cases of cavernous-sinus thrombosis (36, 39, 92, 101, 109, 141). Although useful in ac- curately evaluating the sphenoid sinus, this test, even with injection of contrast medium, has not proven sensitive in demonstrating occlusion of this venous sinus. In only one instance were irregular filling defects demonstrated in the cavernous sinus after injection of contrast medium (109). However, the recent introduction of techniques employing infusion of larger amounts of contrast and rapid sequential views of the cavernous sinus appears to offer promise for greater sensitivity. In only one of our cases (Case 8) was computed tomographic (CT) scan of the cavernous sinuses helpful, demonstrat- ing increased density in the area of the thrombosed cavernous sinus on non-contrast study. If therapy with anticoagulants is being considered, CT scan should be performed to rule out associated cortical venous infarcts with attendant cerebral hemor- rhage. Finally, CT scan of the orbit may demon- strate a dilated superior ophthalmic vein suggesting thrombotic occlusion of the cavernous sinus (Case 6, Fig. 4). Views of this area have also proved useful for ruling out an orbital apex abscess which may mimic cavernous-sinus thrombosis or, on occasion, may also lead to this complication. Carotid angiography has proved to be another useful radiologic study. In active cavernous sinus infection the intracavernous segment of this artery was frequently abnormal. In 11 of 15 studies the lumen was narrowed or completely obstructed (18, 21, 36, 38, 75, 117, Cases 4, 5, 7) (Fig. 5), probably secondary to inflammatory spasm and thrombosis. Radiologic changes consistent with mycotic aneu- rysm were less common (2 of 15 studies) (109) (Case 8). Although technically difficult, orbital venog- raphy has remained the definitive method of dem- onstrating occlusion of the cavernous sinus (12, 72). Fic. 4. CT scan, horizontal cross section, demonstrating marked dilation of the right superior ophthalmic vein (arrow). Case 6. SEPTIC DURAL VENOUS SINUS THROMBOSIS 89 Fic. 5. Arteriogram demonstrating narrowing of the intracav- ernous segment of right internal carotid artery (arrow). Case 7. This procedure, however, has rarely been used to diagnose the septic form of this disease. In each of three instances, in which the procedure has been used, it clearly demonstrated nonfilling of the in- fected cavernous sinus (117, Cases 7 and 8) (Fig. 6). Finally chest x-ray has been the least helpful radiologic study, only rarely demonstrating findings consistent with pulmonary infarction (41, 74, 118, 113, Case 3). Pathology (Table 5): Pathologic findings at au- topsy or surgery have been reported in 20 previous antibiotic-era cases (2, 8, 10, 14, 18, 23, 33, 50, 92, 103, 104, 123, 128, 134, 139) as well as in three of our cases. In contrast to the preantibiotic era, when bilateral involvement was the rule in fatal cases, unilateral disease proved to be as frequent (12 au- topsies) as bilateral disease (11 cases). Our three cases all had bilateral involvement. Extension of thrombosis to other venous sinuses, including the inferior petrosal, superior petrosal, inferior sagittal, sigmoid, and lateral, was observed in seven cases (14, 92, 103, 123, 128, Case 1). In three instances infection spread to the pituitary gland, resulting in necrosis (128, 139, Case 1). In a fourth case aseptic infarction of the anterior pituitary was noted (92). Evidence of leptomeningitis was found in nearly half of autopsies (11 cases), as were brain abscesses or subdural empyema (9 cases), usually localized to Fic. 6. A subtraction view of a frontal vein orbital venogram from Case 7. Dye flowed from the frontal veins (top arrows) to periorbital venous plexus (middle arrows). On the left side dye then filled the superior ophthalmic vein (bottom arrow) and then drained into the cavernous sinus. On the right side dye failed to flow into the superior ophthalmic vein or cavernous sinus, indicating probable thrombosis. the frontal-parietal or temporal regions. Cortical vein thrombosis was found in three cases (18, 123, Case 2). In one instance this was associated wi extensive nonhemorrhagic infarction (123), and in a second case with hemorrhagic infarction and a large, fatal intracerebral hemorrhage (Case 2). The latter patient had been receiving anticoagulants. Septic infarcts of other organs were uncommon (four cases) (2, 8, 14, 134). Descriptions of damage to the intracavernous structures have rarely been presented. In two instances partial breakdown of the internal carotid artery wall was observed (128, Case 1). In Case 1 both the media and adventitia of the intracavernous carotid artery were infiltrated with polymorphonuclear leukocytes and necrosis had occurred; the vessel walt was almost completely destroyed and rupture appeared imminent. In this same case the cranial nerves within the cavernous sinus were also found to be infiltrated with granu- locytes. Finally, in Case 3 hemorrhages were ob- served within cranial nerves III, IV, and VI. Treatment and outcome (Tables 6 and 7): Based on our review of the literature as well as our own experience, high-dose intravenous antibiotic ther- apy directed against Staphylococcus aureus, other SOUTHWICK, RICHARDSON, AND SWARTZ TABLE 5. Pathologic findings in septic cavernous-sinus thrombosis : . segs Other areas of Intracavernous Occlusion of other Case Paranasal sinus disease Meningitis brain sinuses 1. Large sphenoid pyocele Mild basilar Abscess and in- Carotid artery: ne- Superior petrosal si- meningitis farction of an- crosis of media nuses only terior pitui- and adventitia; tary; small ab- PMN invasion scess, temporal of all cranial lobe nerves 2. Purulent material in Subdural ab- Large cerebral Cranial nerves III, Thrombosis of cortical post. ethmoid and. scess, R hemorrhage in IV, and VI nor- veins in R posterior sphenoid sinuses temporal R post. tem- mal temporal region and region, poral—ant. oc- of orbital veins bi- with local- cipital regions laterally ized men- (secondary to ingitis venous infarc- tion) 3. Purulent material L. Severe basi- R orbital abscess Hemorrhagic in- None frontal, ethmoid and lar menin- farction of cra- sphenoid sinuses gitis nial nerves III, IV, VI PMN = polymorphonuclear leukocytes. TABLE 6. Treatment and outcome in septic cavernous-sinus thrombosis Literature Current Total Treatment (Bcases) 1940-1945* 1946-1960 1961-1970 1971-1984 (96 cases) No. (27 cases) (24 cases) (23 cases) (14 cases) No. (%) No. (%) No. (%) No. (%) No. (%) Antibiotics 8 27 (100) 24 (100) 23 (100) 14 (100) 96 (100) Heparint 1 6 (22) 12 (50) 6 (26) 5 (36) 30 (31) Corticosteroids 2 0 (0) 0 (0) 2 (9) 4 (29) 8 (8) Surgery (abscess or air sinus 3 3 (11) 2 (8) 2 (9) 2 (14) 12 (12) drainage; craniotomy) Oute (8 cases) (27 cases) (24 cases) (23 cases) (18 cases)$ (95 cases)$ ome No. No. (%) No. (%) No. (%) No. (%) No. (%) Full recovery 0 13 (48) 11 (46) 11 (48) 2 (15) 37 (39) Chronic sequelae 4 7 (26) 7 (29) 5 (22) 6 (46) 29 (30) Oculomotor weakness 2 4 (15) 6 (25) 2(9) 2 (15) 16 (17) Blindness 1 4 (15) 5 (21) 3 (18) 3 (28) 16 (17) Pituitary insufficiency 0 0 (0) 1(4) 1(4) 0 (0) 2 (2) Hemiparesis 1 0 (0) 0 (0) 1 (4) 1(8) 3 (3) Death 4 7 (26) 6 (25) 7 (30) 5 (38) 29 (30) * Dates of case reports by year. t One patient received streptokinase. $ One case outcome not described (82). gram-positive organisms, and anaerobes should be initiated as soon as the diagnosis of cavernous- sinus thrombosis is suspected. The combination of nafcillin and cefotaxime (or chloramphenicol) is a reasonable initial regimen. In addition to antibiotics, heparin administration may be considered, in particular for patients with early unilateral manifestations of cavernous-sinus thrombosis. Heparin was administered in 32% of 86 reported cases (Tables 6 and 7). As shown in Table 7, mortality was lower among those patients who received heparin treatment, 14% vs. 36% (p < 0.05, chi square corrected for small sample size). Review of case protocols revealed no significant difference in the mean ages of the heparinized and nonheparinized patients (29.3 vs. 25.9 years, re- spectively), acuteness of their disease (duration of symptoms before hospitalization, 6.6 vs. 5.1 days), mean body temperature (39.2 vs. 39.3°C), percent- age of individuals with mental status changes (41% vs. 48%) or hemiparesis (4% vs. 9%). Heparin use varied somewhat during the four time periods; it was used most frequently between 1946 and 1960 (Table 6). Any conclusions concerning the efficacy of heparin must be considered tentative since this analysis was retrospective and involved a collection of individual case reports. Authors may have been less likely to report failure of treatment with hep- arin, and a randomized prospective trial comparing heparin to supportive care has not been performed. SEPTIC DURAL VENOUS SINUS THROMBOSIS 91 TABLE 7. Outcome in septic cavernous-sinus thrombosis* Patients Patients treated not treated with with heparin heparin No. % No. % Total 28 100 58 100 Full recovery 13 46 23 40 Blind 7 25 1 12 Oculomotor paralysis 3 uu ll 19 Other sequelae 4 14 2 3 Died 4t 14 21 36 * Reported cases in English literature (1940-1984) totaled 86. ‘Two cases were excluded. One case outcome was not described (82), and a second patient received streptokinase (81). + Mortality was significantly lower for heparinized patients (p <0.05 by chi square test corrected for small samples). Only one of our patients received heparin (Case 2). However, after two days this medication was discontinued and dicumarol therapy begun. Pro- thrombin time was allowed to increase to nearly three times normal. The patient suddenly devel- oped left-side headaches and vomiting, followed by papilledema, coma and death secondary to a large intracerebral hemorrhage and associated brain- stem herniation. Cerebral hemorrhage in associa- tion with anticoagulation has been reported i in only one other case of septic cavernous-sinus thrombosis (96). While receiving heparin this patient suffered from a self-limited subarachnoid hemorrhage as- sociated with coma and spastic paralysis of his right arm. Anticoagulation was rapidly reversed with protamine, and he subsequently made a full recov- ery. Anticoagulation with vitamin K antagonists has been reported in only four other instances (20, 94, 98, 118). All of these patients recovered without serious neurological sequelae, and none developed clinical signs or symptoms suggesting cerebral hem- orrhage. Early administration of heparin may serve to prevent spread of thrombosis to the other cavernous sinus as well as to the inferior and superior petrosal sinuses. Anticoagulation should be considered only if there is no evidence of cortical venous infarction clinically or by CT scan. If administered, heparin should be given by constant intravenous infusion and activated partial thromboplastin times moni- tored closely to maintain values of no greater than twice normal. The hazards of anticoagulation in this setting are clear—bleeding from hemorrhagic sites of cortical venous infarction or from sites of necro- sis of the intracavernous walls of the carotid artery. In rare instances, administration of corticoste- roids may prove helpful, provided appropriate bac- tericidal antibiotics have been employed. In one of our patients (Case 8) as well as in one previous patient (115), their use may have partially reversed cranial nerve dysfunctions caused by inflammation. Corticosteroids may occasionally serve to decrease persistent orbital congestion (36). Also, in cases where septic thrombosis has also spread to the Pituitary gland, their use (in replacement doses) will prevent Addisonian crisis (63). In addition to medical therapy, if sphenoid sinus infection has been documented by sinus tomograms or if orbital abscess is noted on CT scan, surgical drainage of these purulent collections should be performed promptly. Surgical i was per- formed in 12% of cases. Debridement of the infected sphenoid sinuses was followed by rapid improve- ment in all three of our surgically treated cases (Cases 5, 6, 8). Cavernous-sinus exploration and drainage is a difficult surgical procedure, which has generally not been recommended. Performance of this procedure has been reported only once in the antibiotic era (32). This patient did well postoper- atively; however, one eye had to be enucleated at the time of surgery. The outcome following treatment of septic cav- ernous-sinus thrombosis has improved since the introduction of antibiotics in clinical medicine. Mortality has dropped from between 88 and 100% to 30%. Thirty-nine percent of antibiotic-era cases recovered without serious sequelae, while the re- maining patients suffered from chronic residual neurological impairments including ocular palsies, blindness, hemiparesis, and in two instances pitui- tary insufficiency (58, 131) (Table 6). The poor outcome in our patients, half of whom died and the other half of whom had sequelae, probably reflects the serious nature of cavernous-sinus thrombosis associated with sphenoid sinus infection. A similar high mortality and morbidity has been noted in previous cases with sphenoid sinus infection (14, 58, 103, 138). Despite the development of more effective anti- biotics and better supportive care, there has been no significant improvement in outcome of cases treated after 1960 (Table 6) as compared to patients treated in the earlier antibiotic era (1940-1960). The continued poor outcome in patients with this disease emphasizes the importance of earlier rec- ognition and treatment of primary infections asso- ciated with potential spread to the cavernous sinus. Patients should be dissuaded from squeezing facial pimples, and physicians should avoid surgical drainage of facial infections until appropriate sys- temic antibiotics have been administered (103). As demonstrated by our cases, early recognition of sphenoid-sinus infection is also critical for prevent- ing this disease. Once infection has fully developed in the cavernous sinus, injury to intracavernous cranial nerves, and spread to the meninges and 92 SOUTHWICK, RICHARDSON, AND SWARTZ even to the pituitary gland can occur despite appro- priate antimicrobial therapy. Septic Lateral-Sinus Thrombosis Results and discussion Including our 4 cases, 64 cases of septic lateral sinus thrombosis have been reported from 1940 to 1984 (5, 18, 16, 24, 34, 37, 45, 48, 53, 54, 57, 59, 60, 65, 66, 73, 79, 80, 84-86, 88, 93, 100, 107, 110, 113, 126, 136, 140). Primary site of infection (Tables 8 and 9): As observed in the preantibiotic era, the pathogenesis of this disease has been almost exclusively the spread of infection to the lateral and sigmoid si- nuses from the mastoid air cells either via the emissary veins or by direct invasion. Septic lateral- sinus thrombosis with one exception (5) has fol- lowed either acute (49%) or chronic (51%) otitis media. All four of our patients had a history of chronic ear infections for 2 to 31 years (mean, 18.5) before developing this complication. Symptoms and history of underlying diseases (Ta- ble 9): Severe generalized or occipitofrontal head- ache, earache, nausea, vomiting, diplopia, loss of visual acuity, and hemiparesis were the major symptoms described in preantibiotic-era cases (22, 121). Headache (81%) and earache (54%) have re- mained the most frequent symptoms in the anti- biotic era. Ear pain generally preceded headache by several weeks, and the headache has generally been described as severe, persistent, and fronto-tem- poral-occipital in location, with pain usually con- fined to the same side as the ear infection. Head pain is probably a manifestation of developing epi- dural abscess, irritation of the fifth cranial nerve, or septic thrombosis of the lateral sinus. Nausea and vomiting (48%) have also remained common complaints. In one of our patients (Case 1) nausea and vomiting were the most prominent symptoms, initially suggesting to the admitting physician the diagnosis of viral gastroenteritis. Diplopia, photo- phobia, neck stiffness, and cough associated with bloody sputum production have been less com- monly described. Vertigo was a common early com- plaint in three of our cases (Cases 1, 2, and 4) and may have resulted from extension of inflammation to the inner ear. As compared to cavernous-sinus thrombosis, which has an acute onset, septic lateral sinus thrombosis is a subacute illness, and symp- toms generally persisted for several weeks before hospitalization was considered. A history of predisposing illness other than chronic otitis media has been rare, only three pa- tients having had another chronic illness (5, 16, Case 3). This has generally been a disease of young, otherwise healthy individuals (mean age, 22 years). Physical findings (Tables 8 and 9): Evidence for an ear infection and persistent fever despite the administration of appropriate antibiotics should al- ways raise the possibility of lateral-sinus thrombo- sis. Abnormalities on otologic examination were almost always noted (98%) in antibiotic-era cases, particularly posterior auricular swelling and tender- ness (42%), caused by mastoid emissary vein ob- struction (sometimes called Gresinger sign), and purulent drainage from a ruptured tympanic mem- brane (42%). Less commonly, only a dull erythem- atous tympanic membrane was noted (21%). Fever has been reported in the majority of cases (79%); however, absence of fever does not exclude the diagnosis. Prominent bilateral papilledema occa- sionally associated with retinal hemorrhage, a man- ifestation of elevated CSF pressure, has been found in over half of cases (53%). Focal neurological signs were usually not present, with the exception of unilateral sixth-nerve palsy (37%). Dysfunction of this nerve was probably the result of local compres- sion by swelling of the inferior petrosal sinus in the closed space of Dorello’s canal (59). The symptom complex of sixth-nerve palsy, fifth-nerve irritation resulting in temporoparietal and retroorbital pain, and otitis media has been named the Gradenigo syndrome. This rare complex has been noted in several recent cases (59) as well as in one of our patients (Case 4). When present, this symptom complex has provided strong evidence for the pres- ence of lateral-sinus thrombosis or inflammation along the petrous ridge of the temporal bone. Less common findings have included mild nuchal rigidity (31%), depression of mental status (14%), loss of visual acuity (8%), and hemiparesis (2%). Depres- sion of mental status was a prominent finding in three of our four cases (Table 8). Another promi- nent finding in three of our patients (Cases 1, 3, and 4) was lateral-gaze nystagmus. In the latter two cases nystagmus preceded lateral-gaze paralysis by several days. As in our analysis of septic cavernous- sinus thrombosis, stratification of septic lateral- sinus thrombosis cases by year demonstrated no significant change in the frequency of specific symptoms or physical findings over the 44 years reviewed in our study. Laboratory findings (Tables 8, 9 and 10): Useful tests for evaluating patients with suspected lateral- sinus thrombosis include lumbar puncture, which has often revealed an elevation of CSF pressure (77% of patients who had lumbar punctures; mean + S.D. = 453 + 133 mm H,0) (Tables 8 and 9). Elevation of CSF pressure has been thought to be due to interference of CSF resorption by the arach- 93 SEPTIC DURAL VENOUS SINUS THROMBOSIS “sUBU = Y “Yel = T ‘89;400yn9] ZeafonuoYdioursJod = NWd ‘amssard Suuado = -q’Q ‘wajoud [e109 = “_\L “UNod [189 poojq HY = OGM, “avout! Jo avak () IP uotyei0jdxa snurs /3m gp 2800 Terey8y Y ‘AuI04 “DID ‘p/Sua -deployseul [eorper TST ‘dL wuteperided ¥W SutpAmoqdarys + (NWd %8) pus styIpioy Jooruaqdurezopyo 096 O@M -seur queMIeY + Unforuad -a‘] ‘009 ‘dO snquory} [p/3ar jo [eaoutar WU gg esoon|s uorez0jdxe snuts ‘1P/3 OTT [e1948] J ‘Aur0y09 TL (NWd -ployseur [worpe ‘T %0T) sppay ‘joorueydurezoyyo 088 OFM Tensia poqouysu0l + uporued “A'] ‘009 ‘Z'0 snurs [ero -JB] "TT 1240 aug Jo 1p/3ur yeaourar ‘Aa10}99 yg asoons -pioyseur peorped "] ‘1p/San og sjoorueydorezo[yo “AL *(8hjod Asanooel [1M + unformed “A°] %O0l) FF OU aSeuresp sseosqu Teanpida ‘Auroz9 -proyseum [eorped "T ‘joorueydurezopyo + ujorureyued 4190901 [[0,] + UNTPIAeqzwd “AT = au1093nGQ queulyeel], seniatelay wurepel[ided [era4] -1q ‘sersjed aazeu 439 [eseqeriq ‘eze8 [e1038] +] Wo snueysAu “ery yoau yynne ‘Ased Tere; y Seisdoue -Tmay snowmAuomoy Y ‘sureperided yezayetiq peyseur ‘Asped earau 3g 7] ‘e788 [es948] J uo snadezsAu ‘odzey}e7] yoo J7H8 ‘peyuets08iq re urepuv} Apeoys -Un ‘suigzeuI 9sIp erin] ‘e783 [81998] ‘] Uo snuifeysAu ‘Asmoiq sBurpuyy feorZojomneny simp ‘ployseur “zee aTpprar y snp ‘pioysear ‘1e8 aTpprm ‘T vinp ‘plojseut ‘ree a]pprar ‘| aouds feinp -1de ‘pioyseut ‘x80 [prot T uoRoeyut jo sais (proyseut) ‘ds snajoig: (proyeun) snaino “S (proy -seul) ‘de ‘dasgg 214 (e889 HOW) s1soquiory) snuys-[e1978] atydes Jo sernyeoy [BOTY “§ ATAV.L 94 SOUTHWICK, RICHARDSON, AND SWARTZ TABLE 9. History and findings in septic lateral-sinus thrombosis Current series Literature No. with No. with Total finding/total __finding/total History Onsett 1644 days 14+ 10 dayst Chronic otitis media 4/4 29/61 51 Headache 4/4 35/44 81 Earache 3/4 23/44 54 Vomiting 3/4 20/44 48 Vertigo 3/4 3/44 12 Physical findings Fever 4/4 18/24 9 Abnormal ear find- 4/4 48/49 98 ings Papilledema 1/4 26/47 53 VI nerve palsy 2/4 17/47 37 Mild nuchal rigidity 2/4 14/47 31 Nystagmus 3/4 0/47 6 Laboratory findings Leukocytosis 3/4 11/14 18 Increased CSF pres- 2/3 22/28 17 sure Parameningeal CSF 3/3 1/29 31 formula Abnormal mastoid 4/4 20/22 92 (x-ray) Abnormal CT scan _ 3/7 42 Abnormal angio- — 27/27 100 gram:venous phase Abnormal static 6/9 67 brain scan Abnormal dynamic _— 2/2 100 brain scan Chest film: circular 0/4 4/61 6 opacification *Calculated by adding the numerators from current and literature cases and dividing by the sum of the denominators for each category. + Mean time from development of first symptom to hospital- ization. + Mean + S.D. TABLE 10. Bacteriology: Septic lateral-sinus thrombosis Current series* Literature Total Organism (4 cases) (14 cases) (18 cases) No. No. No. (%) Proteus species 1 6 7 (39) S. aureus 1 4 5 (28) E. coli 0 2 2 (11) Bacteroides fragilis 2 0 2(11) Anaerobic streptococci 2 0 2(11) Other 2 3 5 (28) * Multiple organisms isolated frum several cultures (included acombination of Bacteroides fragilis and anaerobic streptococci). noid villi (see Pathology section). In the majority of patients the cellular and biochemical composi- tion of the CSF has been normal. In approximately one-third of patients, the CSF contained moderate numbers of mononuclear cells often mixed with polymorphonuclear leukocytes (parameningeal CSF formula). This abnormal finding was associ- ated with an epidural or brain abscess in 5 of 10 instances (37, 80, 126, Cases 2 and 4). The Queck- enstedt or Tobey-Ayer maneuver has proven unre- liable in diagnosing lateral-sinus thrombosis (78). In only 28% of antibiotic-era cases has the bac- terial pathogen been adequately demonstrated, pos- itive cultures being derived primarily from intra- operative samples (13 specimens). Blood cultures were positive in four instances and CSF in one. The principal organisms associated with septic lateral- sinus thrombosis have reflected the bacteriology of chronic otitis media: Proteus species, Staphylococ- cus aureus, Escherichia coli, and more recently an- aerobes (105, Cases 1 and 2). Bacteroides fragilis was isolated from intraoperative specimens in two of our patients. Radiologic findings (Table 9): Since the primary site of infection leading to septic lateral-sinus thrombosis is almost always the mastoid air cells, mastoid radiographs should always be performed when this diagnosis is suspected. Findings consis- tent with mastoid infection have almost always been demonstrated (92%). Abnormalities have in- cluded increased density with loss of mastoid air- cell trabeculae (50%), bony sclerosis of the mastoid region (31%), and lytic lesions of the temporal or parietal bones (27%). CT scan with contrast has proved helpful in demonstrating concomitant cerebritis in one case (57), a filling defect in the area of the thrombosed sigmoid sinus in another (17), and ring enhance- ment around a thrombosed sigmoid sinus in a third (17). This study should also prove helpful in dem- onstrating associated brain abscess or hemorrhagic infarctions secondary to cortical vein thrombosis. Angiography has remained the most definitive method for demonstrating lateral sinus occlusion. Venous phase views have demonstrated obstruction of one or both lateral sinuses in all 27 previously reported cases in which this study was performed. Carotid angiography was performed in three of our cases (Cases 1-3). However, since the diagnosis was thought to be brain abscess in each instance, venous phase studies were not obtained; the correct diag- nosis was finally made at surgery. Although experience with dynamic radionuclide scan has been limited, having been performed in only two previous cases of septic lateral sinus thrombosis (16), this test may prove helpful in demonstrating superior sagittal and lateral sinus flow or lack of flow. Other less useful tests include static radionuclide brain scan and chest roentgen- ogram (29). Pathology: Operative findings have been de- scribed in 31 antibiotic-era cases (including our 4 cases). Granulation tissue or a frank purulent ab- SEPTIC DURAL VENOUS SINUS THROMBOSIS 95 scess was noted to be overlying the lateral sinus in 68% of cases (18 of 27 previous cases as well as our Cases 1, 2, and 4). Cholesteatomas were found to have eroded through the temporal bone and to be overlying the lateral sinus in eight previous cases (30%). Cholesteatomas were also found in our Cases 1 and 2; however, neither had eroded through the temporal bone. In all cases in which the sinus was opened thrombus was found, often mixed with pu- rulent exudate. Needle aspiration in three of our patients (Cases 2-4) documented lack of blood flow in the lateral sinus. Postmortem examinations have been reported in only four modern cases (5, 48, 59). In addition to thrombosis of one (48, 59) or both lateral sinuses (5), thrombus was also found to have extended into the internal jugular vein in all four patients. In one case, clot also had extended into the posterior section of the superior sagittal sinus (48). Bilateral pulmonary abscesses secondary to septic pulmonary emboli were noted in one instance (59) and a cerebellar abscess in another (48). The side of lateral sinus thrombosis was de- scribed in 39 antibiotic-era cases including our pa- tients: The right lateral sinus was occluded in 51%, the left in 44%; and in 5% bilateral thrombosis was demonstrated. Elevated CSF pressure and/or papilledema were noted more frequently after right- sided thrombosis (75% of patients with occlusion of this side) than after left-sided occlusion (29%). Only one of our cases (Case 4) had thrombosis of the right side. This event was associated with both papilledema and elevated CSF pressure. The other three suffered from thrombosis of the left sinus: Case 1 developed mild papilledema, and Case 3 had marked papilledema and elevated CSF pressure; Case 2 had normal CSF pressures. The greater predisposition to develop CSF pressure elevation following right lateral-sinus occlusion, the sinus which most frequently drains the superior sagittal sinus, supports the suggestion that hydrocephalus of the otitic type is caused by interference with superior sagittal-sinus blood flow and concomitant reduction in CSF resorption by the arachnoid villi. Treatment, complications and outcome (Tables 8 and 11): Treatment with intravenously adminis- tered antibiotics in high dosage should be initiated on admission. Antibiotics should be directed against the most likely pathogens in chronic otitis media and mastoiditis (Proteus species, S. aureus, anaerobes, and E. coli). In most cases a combination of nafcillin and cefotaxime (or chloramphenicol) is appropriate initial therapy pending results of cul- tures. Although Pseudomonas is frequently cul- tured from the external ear canal, this organism has not been isolated from intraoperative speci- mens in antibiotic-era cases. If this organism is found to be the primary pathogen, a combination of carbenicillin (or ticarcillin) with an aminoglyco- side should be administered. Despite the administration of antibiotics, surgery has frequently been required before patients im- proved clinically (73% underwent mastoidectomy). If the patient remains febrile and toxic after 12 to 24 hours of appropriate antibiotic treatment, radi- cal mastoidectomy should be performed and the area overlying the lateral sinus explored. In pre- vious cases infected granulation tissue or purulent collections requiring debridement and were frequently found. The lateral sinus may be needled to document thrombosis. Judging from past experience actual removal of thrombus (performed in 12 antibiotic-era cases) does not improve clinical outcome and, as in our Case 4, may lead to persist- ent venous hemorrhage. Lateral sinus recanaliza- tion, development of collateral venous channels and normalization of CSF pressure have usually fol- lowed removal of the inflammatory focus. Mannitol, glycerol, and/or steroids may be em- ployed to reduce cerebral edema, but these agents have been employed in only four previous cases (57, 82, 88, 113). Serial lumbar punctures every 48 hours may also be considered to reduce elevated CSF pressure (used in four previous patients) (45, 54, Cases 3 and 4). If papilledema and increased CSF pressure persist and visual acuity deteriorates de- spite these measures, subtemporal decompression may be performed (employed in eight previous cases) or venous grafts, bypassing the areas of per- sistent thrombosis, may be considered (performed successfully in two previous cases) (53, 113). In the preantibiotic era ligation of the jugular vein was commonly performed to prevent septic pulmonary emboli. With the advent of antibiotics this compli- cation has become extremely rare, making this pro- cedure unnecessary. As shown in Table 11, jugular vein ligation was not performed in cases treated after 1975. Anticoagulation is not recommended. Cortical veins overlying the mastoid infection may become occluded, resulting in small venous hem- orrhagic infarcts. Such infarcts make the risk of intracerebral hemorrhage following anticoagulation prohibitively high. In the preantibiotic era, complications were fre- quently associated with septic lateral-sinus throm- bosis (78% of cases) (78). Since the development of antibiotics they have been much less frequent, oc- curring in 20% of cases. Complications have in- cluded meningitis (1 case) (84), cerebellar abscess (4 cases) (48, 60, 80, 84), septic pulmonary emboli (4 cases) (25, 59, 65, 79), cavernous-sinus throm- bosis (1 case) (110), cerebral infarct secondary to cortical vein thrombosis (1 case) (66), septic ar- thritis of the hip (1 case) (126), and uncontrollable 96 SOUTHWICK, RICHARDSON, AND SWARTZ TABLE 11. Treatment and outcome in septic lateral-sinus thrombosis Literature Current series Total (4 cases) 1940-1955 1956-1965 1966-1975 1976-1984 (64 cases) No. (8 cases) (16 cases) (26 cases) (10 cases) No. (%) No. No. No. No. Treatment Antibiotics 4 8 16 26 10 64 (100) Mastoidectomy 4 8 10 17 8 47 (73) Exploration of lateral 4 6 9 3 5 27 (42) sinus Ligation of internal jug- 0 1 2 6 0 9 (14) vein Heparin 0 0 2 0 0 2 (3) Outcome Full recovery 2 6 10 20 8 46* (75) Chronic sequelaet 2 0 1 3 2 8 (13) Death 0 2 2 3 0 7 (12) * Outcome not described in 3 cases (60). t Include chronic otitie hydrocephalus, seventh and eighth cranial nerve defects, hemiparesis, ataxia. sepsis (1 case) (110). These complications were evenly distributed in the first three decades of our study. None of these complications developed in cases reported after 1975 or in our cases (Table 11). Mortality has also been reduced since the intro- duction of antibiotics; the case-fatality rate for cases from 1940 to 1984 was 12% as compared to 20 to 30% in the preantibiotic era (78). No deaths have been associated with this disease since 1975 (Table 11). Although full recovery has been the rule (75%), 13% of patients survived with chronic se- quelae that have included otitic hydrocephalus (3 cases) (88, 110, Case 4), chronic facial paralysis (2 cases) (60, 110), decreased visual acuity (3 cases) (88, 113, Case 3), eighth-nerve dysfunction (1 case) (Case 4) and residual hemiparesis (1 case) (66). Septic Thrombosis of the Superior Sagittal Sinus Results and discussion The superior sagittal sinus, the largest of the intracerebral venous sinuses, is less commonly in- volved in septic thrombosis than the cavernous or lateral sinuses. Including our seven patients, only 23 cases have been reported since the development of antibiotics (4, 16, 31, 62, 68, 97, 101, 112, 114, 119, 120). Primary sites of infection (Tables 12 and 13): In the preantibiotic era septic thrombosis of the su- perior sagittal sinus was almost exclusively a com- plication of bacterial meningitis (3, 125). Meningi- tis has remained the major infection predisposing to this disease (48% of cases) (62, 97, 112, 114, 120, Cases 1-6), infection probably spreading from the meninges to this sinus via the diploic veins. Air sinus infection has been the second most frequent infection associated with superior sagittal sinus thrombosis (17%) (16, 101, 120, Case 7). Infections in the ethmoid and maxillary air sinuses may spread to this dural sinus via ethmoidal veins (16). Frontal sinus infection with associated epidural abscess has resulted in cortical vein thrombosis which may be followed by superior-sagittal-sinus thrombosis (120, Case 7). Extension of septic thrombosis from the lateral dural sinus into the superior sagittal sinus has also been reported (9%) (97, 120). Pul- monary infection (97, 120), tonsillitis (68), tooth infection (119), and pelvic infection (4) have also preceded septic thrombosis of the superior sagittal sinus. Symptoms and history of predisposing illness (Ta- ble 14): The pace of this illness has generally been acute, in our experience, the mean time from the development of the first symptom until hospital admission being only 2 days. Other antibiotic case reports have described a more variable onset of 1 to 30 days. The initial symptom in most cases has been severe headache (45%), either generalized, occipital, or, in patients with bacterial sinusitis, localized to the region overlying the frontal, max- illary, or ethmoid sinuses. Headache has often been associated with nausea and vomiting (35%); these symptoms are generally followed within several days by confusion (52%), which in most cases rap- idly progresses to coma. The onset of confusion has frequently been followed by focal or grand mal seizures (57%), which have usually been refractory to anticonvulsant medications. In 35% of antibiotic-era cases a history of a predisposing illness has been reported, including sickle cell anemia (97), osteopetrosis (114), Down syndrome (Case 2), breast cancer (112), and malig- nant lymphoma (Case 3). Two women have devel- oped septic thrombosis of this sinus in association 97 SEPTIC DURAL VENOUS SINUS THROMBOSIS RP seNzZ198 10}0Ul poze /3u gy ‘ons -Jouad [esaae ‘esouquan wor 1p/8a1 gt Tetoes of suas ‘sts0ydord -B10qT|Go svodipe “juemepLg TL NWA | Seuapajided poyzeai ~9p ‘sermojosnitts [wyUONy ‘joo %09) 8 SerTnafeoe ‘eso] £10 pooiq ‘souds ~qwoydurezoyyo ‘urporeydeo “aT = OAM ‘OVE ‘dO -Wrem MoreTUsoUCD Joog = eampide ‘nuts peyUONy samnzies 109001 pezt saduru Amoyeproyseu ‘urporued *A'] PA “Ter0uad <— popis ‘| ‘wag -9ur ‘poo]g ‘ree aTppry] pPp/Aau 9% ev0ony8 ‘wioyduspenb ‘1p/Sur ogg Piovy ‘omayare ‘pid “aL ‘00¢'08 -hd poy poxy ‘eyueur saBuyuaw ‘poojq ouogeyjourexep ‘uTotued “AT = JEM 02S “dO §=—=-@AOU aUTND08I4x0 OU SeMIOD “soenuts sre ‘re0 Opp, Ader} 0} £10;081391 ‘somNz198 Ip/ta [8903 "] pue yf ‘amaund GT e800ny3 FequIn] 1848 UONwyETIP 1p/am 212 Tdnd y 3usrewexy ‘seared ‘dL (NWd ~furey piooey "J ‘suas rye vam ‘AT % 001) 008"ET “trzprug ‘Sruzeyy + ‘wai0o sodura Suyorued Tesoqyerquy pu -a°] OM ‘06S ‘dO <— ured dasp 03 esuodsaz f -@U ‘poolq ‘180 oIppIW, [oxpeam Ppa -[08 ‘ouTjoxows7e3 ‘uloruE ues 9% ap0on|s snuys proweyds “oorueyduresopyo ‘uypioexo “ay Ip/Su ge dL Yeu Ys ‘euoo — umuTaq, ‘sadurueur ‘poojq ‘Suny uyoraed “A"] _ wut0g, seSuruem ‘poojg ‘Suny p/n Aderayy 03 £10,083 9% asoon|? ad ‘samzias poziesued ‘Ip/am gaz, — [R00] yf Seoyduspenb ‘DL (NW soquozodAy ‘otxepjore euojostuperd|Aqyour %09) OSTT ‘eka y uoHonppe posers ‘uypoqued ‘uodureuwy “AT ‘ORM 96 “dO 9p Seul0o daop «- ABreqa] seBuyuour poorg emyound WorzazUT queurzeery, equiv] sSuypuy [eoojomen, joaug “8 A1q8L, O68 ‘suONBIAaIqqe 104 “aeourTt Jo Wak ( ), (snure [oy ~wouy ‘poo]q) -dazys oiqoueeuy (489 pooiq) eng0000}der78 ansjouray gf wstuBzIC (s2889 HDW) ssoquiory) snurs-[ey {Fes zojzedns opjdes Jo soANyWo] VOUT “ZI ATAVL (ret) W/E 9¢ (g961) d/o g (L961) W/4 gg (2961) WHA ug (oz6t) W/6 1g (8961) WAK ¢9 +(SL6T) W/I" > x0g/o8y 98 SOUTHWICK, RICHARDSON, AND SWARTZ TABLE 13. Initiating infections in septic superior sagittal-sinus thro i Current series Literature Total {7 cases) (16 cases) (23 cases) No. No. No. (%) Meningitis 6 5 11 (48) Sinusitis 1 3 4 (17) Mastoiditis 0 2 2 (9) Tonsillitis 0 1 1(4) Pulmonary 0 2 2(9) infection Other YJ 3 3 (13) with pregnancy, one postpartum (4) and a second while 38 weeks pregnant (62). Physical findings (Tables 12 and 14): High fever (70%) and alterations of mental status (65%) have been the most frequent findings associated with septic thrombosis of the superior sagittal sinus in the antibiotic era. The majority of patients were unresponsive to verbal or painful stimuli on arrival at the hospital or within several days of admission. Nuchal rigidity and Kernig and Brudzinki signs have been observed in many of the cases associated with meningitis (26%). Hemiparesis was a frequent finding (43%), the consequence of cortical vein thrombosis and associated venous cerebral infarcts. Signs of brainstem compression, including lack of eye movements even on caloric stimulation, dilated unreactive pupils and generalized flaccid motor pa- ralysis have been noted in three cases (13%) (62, Cases 1 and 5). Papilledema was observed in 17% of cases (31, 68, 120, Case 7); however, patients often did not survive long enough to develop this finding. Patients in whom frontal, ethmoid, or maxillary sinus infections were the predisposing illnesses tended to develop a subacute onset of symptoms. Headache often persisted for several weeks before focal motor deficits or seizures occurred. Since only the anterior segment of the superior sagittal sinus usually thrombosed following these infections, coma, severe cerebral edema, and brain stem her- niation developed less frequently. Laboratory findings (Tables 12, 14, 15): Cerebro- spinal pressure was often elevated (57%), occlusion of the superior sagittal sinus impairing CSF resorp- tion by the arachnoid villi (Tables 12 and 14). In cases with concomitant meningitis, CSF analysis has generally revealed a high granulocyte count, increased protein level, and hypoglycorrachia (62, 112, Cases 1, 3-5), and the causative bacterium isolated on CSF culture (62, Cases 1, 3-6). CSF findings in cases associated with infections at other sites have included a lower CSF granulocyte count, a slightly elevated CSF protein, normal glucose TABLE 14. History and findings in septic superior sagittal-sinus thrombosis Current series Literature No. with finding No. with finding Total Total No. Total No. History Onsett 2.1 + 1.5 dayst 11.6 + 10.7 days Severe headache 4/4 5/16 45 Confusion 6/7 6/16 52 Vomiting 2/4 5/16 35 Seizures 4/7 9/16 57 Predisposing circumstances 2/7 6/16 35 Physical findings High fever UT 9/16 70 Abnormal mental status U7 8/16 65 Motor deficits 4/7 10/16 61 Nuchal rigidity 2/7 4/16 26 Papilledema 7 3/16 17 Laboratory findings Leukocytosis 5/7 2/2 18 Increased CSF pressure 4/4 4/10 57 CSF formula: meningeal 3/4 2/10 36 parameningeal 1/4 5/10 43 CSF analysis: RBC’s 0/4 4/10 29 Abnormal EEG 3/3 3/5 vy Abnormal skull films 2/3 1/2 60 Abnormal CT scan Wl 4/4 100 Abnormal angiogram: 1/72 5/6 15 venous phase * Calculated by adding the numerators from current and literature cases and dividing by the sum of the denominators for each category. + Mean time from development of first symptom to hospitalization. + Means + S.D. SEPTIC DURAL VENOUS SINUS THROMBOSIS 99 TABLE 15. Bacteriology in septic superior sagittal- thrombosis sinus Current series Literature Total Organism (7 cases) (8 cases) (15 cases) No. No. No. (%) S. pneumoniae 3 3 6 (40) S. aureus 0 1 1(7) B-hemolytic 1 1 2 (13) streptococci Anaerobic strep- 1 0 1(7) tococci Klebsiella sp. 1 1 2 (13) Pseudomonas sp. 0 1 1(7) Trichinella sp. 0 1 1m” level, and a negative CSF culture (4, 68, 119, 120, Case 7). In the preantibiotic era this disease was generally associated with meningitis due to Streptococcus pneumoniae or Hemophilus influenzae (3, 125) (Ta- bles 12 and 15). Since the introduction of antibiot- ics a variety of pathogens has been associated with septic thrombosis of this dural sinus. S. pneumoniae has remained the most common pathogen (40% of cases with positive cultures) (62, 120, Cases 4-6), but other gram-positive and gram-negative orga- nisms have also been isolated. In one case throm- bosis of the superior sagittal sinus was associated with severe infestation by Trichinella spiralis, doc- umented by muscle biopsy (31). Dehydration and a hypercoagulable state in addition to possible central nervous system trichinosis may have predisposed this patient to develop dural-sinus thrombosis. Radiologic findings (Table 14): Although the symptoms, signs and CSF findings described may raise the possibility of superior sagittal-sinus thrombosis, they are not specific. Similar findings can be associated with lateral-sinus thrombosis, brain, epidural or subdural abscess; or bacterial meningitis complicated by septic arteritis, extensive cortical vein thrombosis, or severe cerebritis. Sev- eral noninvasive diagnostic techniques can be help- ful in making a specific premortem diagnosis. Com- puterized tomography has proved one of the most helpful studies, demonstrating abnormalities in all five cases in which it was performed (31, 101, 112, Case 7). This study has also proved very helpful in cases of aseptic superior sagittal-sinus thrombosis (6, 11, 17, 52, 95, 116, 142). The most specific findings have been demonstrated only after injec- tion of contrast enhancing material and include: 1. The “empty delta sign,” a triangular area of decreased density surrounded by a ring of increased density. This represents clot within the sinus with contrast material in the surrounding collateral veins and sinus wall. In all reports of septic and aseptic superior sagittal-sinus thrombosis to date, this sign has proved specific for the disorder (17, 101). It has been observed in 50% of the 18 septic and aseptic cases studied. 2. Gyral enhancement following contrast infu- sion: This finding is not specific and has also been reported following septic thrombosis of the domi- nant lateral sinus. This finding probably represents increased venous collateral circulation and has been noted in 4 of the 18 septic and aseptic cases studied (22%). 3. Focal cerebral edema following contrast injec- tion (observed in 3 of 18 cases): This is a less specific finding, which also may be noted in asso- ciation with meningiomas, gliomas, and other in- tracerebral neoplasms. Abnormalities on unenhanced CT scan tend to be less specific or seen late in the course of this disease. Parenchymal hemorrhages secondary to cortical venous thrombosis and hemorrhagic cere- bral infarcts have been noted in approximately one- quarter of septic and nonseptic cases of sagittal sinus thrombosis (9 of 37). Small ventricles are also seen in approximately one-quarter of patients. However, this finding is also seen in pseudotumor cerebri, lateral-sinus thrombosis, and diffuse cere- bral inflammation. The “cord sign” represents a high-density clot in a cortical vein and is a specific sign of thrombosis; however, it is a rare finding, observed in 2 of 37 patients. Venous phase carotid angiography continues to be the most definitive method for demonstrating superior sagittal-sinus thrombosis (4, 31, 68, 101). Angiography may be difficult to perform in the acutely ill, confused patient. If cerebral edema is severe, venous phase filling may be impaired with- out dural-sinus thrombosis. This study has been particularly useful in cases of partial occlusion of the superior sagittal sinus in which CT scan and dynamic brain scan may prove nondiagnostic (4, Case 7) (Fig. 7). Although experience with dynamic brain scans has been limited, this study may prove useful. Non- filling of the superior sagittal sinus should raise the possibility of thrombosis of this sinus (7, 16). At present data on specificity and sensitivity are not available. Other diagnostic studies of limited value have included skull films and static brain scan. Electroencephalograms were performed in eight cases, demonstrating focal electrical abnormalities in five patients (4, 68, 119, Cases 1 and 7), diffuse slow-wave activity in one case (112), and absence of electrical activity in another (Case 5). Pathology (Table 16); Autopsies in the preanti- biotic era generally revealed thrombosis of the en- tire superior sagittal sinus as well as thrombosis of superior cerebral cortical veins. Subdural and intra- cerebral hemorrhages were common. These abnor- 100 SOUTHWICK, RICHARDSON, AND SWARTZ Fic. 7. Lateral view of the venous phase of the carotid arteriogram performed in Case 7. Dye failed to flow into the anterior third of the superior sagittal sinus. Note also the absence of opacified cortical veins in the region of the frontal cortex. The frontal air sinus can be seen in the lower left. malities were associated with the findings of purulent bacterial meningitis (3, 125). Postmortem findings in 12 previous modern cases have been described. Gross and histopathological evidence for bacterial meningitis was found in one-third of au- topsies (62, 68, 99, 120). The entire superior sagittal sinus was thrombosed in 92% of cases. In one case only the middle-third of this sinus was occluded with thrombus (62). In 41% additional dural sinuses were also occluded. The lateral sinus was occluded in two cases (119, 120), and in one of these throm- bus extended into the internal jugular vein (5). All dural venous sinuses were found to be occluded in two cases (97, 120). In a third case only the cavern- ous sinus and great cerebral vein (of Galen) re- mained patent (68). Cortical vein thrombosis was found in half the cases and was associated with hemorrhagic cerebral infarction in each instance (4, 68, 97, 120). Infarction resulted in cerebral edema in one case (119). In addition, pulmonary emboli were observed in 33% of patients (97, 119, 120). Iliac vein thrombosis was present in two pa- tients (119, 120). Postmortem examinations were performed in all of our fatal cases. The pathologic findings are sum- marized in Table 17. In all five cases there was evidence of acute meningeal inflammation near areas of superior sagittal-sinus thrombosis. Case 2 had only a small collection of purulent exudate over the parietal lobe adjacent to the superior sagittal sinus. In the other four cases thick purulent exudate was more widely distributed in the subarachnoid space. Other findings included cerebral edema, brain herniation, cortical cerebritis, cortical vein thrombosis, and hemorrhagic infarction of cerebral cortex. Frequently other dural sinuses were also thrombosed. The sequence of thrombosis could be judged by the histologic age of thrombi in several instances. In Cases 1 and 2 thrombosis began in the superior sagittal sinus, later extending to the torcula and right lateral sinus and, in Case 2, also to the sigmoid sinus. In Case 5, in which thrombosis developed less than 24 hours before death, thrombi of similar age were observed in the left lateral and straight sinuses and great vein of Galen, as well as in the superior sagittal sinus. Patient 6 survived her acute illness, but suffered severe mental retardation, and was deaf and blind. Neurological examination six months after her ill- ness revealed an inability to follow light or to react to noise, persistent flexion of both arms at the elbows and of the right leg. Muscle tone was poor. Deep tendon reflexes were asymmetric (1+ in the right leg and 3+ in the left leg). This child required SEPTIC DURAL VENOUS SINUS THROMBOSIS 101 TABLE 16. Pathologic findings in acute septic thrombosis of the superior sagittal sinus (MGH cases) Duration Extent of Evidence for . . and extent Case meningeal Cerebral edems/ cortical Areas of infarction Co rtical vein of superior Other dural involvement cerebritis i hrombosi 1. Regional meningi- Moderate/No Minimal Hemorrhagic infarcts, Present in areas 3-6days; Torcula; R tis sparing the superior parietal re- of infarction posterior _ lateral base of the brain gion, bilateral (L > % sinus 2. Regional meningi- Moderate/No None Hemorrhagic infarc- Present in areas 3-6days; R lateral tis overlying the tion, R superior pari- _ of infarction posterior sigmoid R parasagittal and R inferior % sinuses gyrus and the R occipital regions occipital inferior 3. Diffuse meningitis, _ND/ND Insular region: None seen Areas with venous <24-48 hr; None most extensive cellular in- vasculitis with- _ focal exudate at base filtration of out occlusion of brain, ven- cortex with triculitis necrosis 4. Diffuse meningitis, Mark Recent necro- Focal infarction, R Thrombosis of § <24 hr, fo- None most extensive lateral sis of outer frontal lobe large vein over- _cal mid- in frontal re- temporal cortical lay- lying R superior _ portion gions (R > L) and cere- ers beneath temporal gyrus ({puru- bellar her- exudate lent niation throm- bus) 5. Diffuse meningitis § Marked/bi- PMN infiltra- Diffuse hypoxic en- Bilateral cortical <24hra, _L lateral lateral tion of the cephalopathy vein thrombosis _ entire sinus, temporal superficial sinus vein of lobe her- cortex Galen niation ND = not described. TABLE 17. Treatment and outcome in septic superior sagittal-sinus thrombosis Current series Literature Total (7 cases) (16 cases) (23 cases) No. No. No. (%) Treatment Antibiotics 7 13 20 (87) Corticosteroids 5 0 4 (17) or urea Surgery 1 4 5 (22) Outcome Full recovery 1 2 3 (18) Chronic sequelae 1 1 2 (9) Died 5 13 18 (78) chronic institutionalization and died of aspiration pneumonia at 7 years of age. Postmortem exami- nation illustrated the permanent pathologic seque- lae of this devastating complication of meningitis. Her superior sagittal sinus, torcula and both lateral sinuses were filled with very firm, grayish-white, fleshy connection tissue. The venous sinuses were only irregularly recanalized, with small lumina vis- ible on histologic section. The right cerebral hemi- sphere contained almost no cortex; the leptomen- inges essentially covered the ventricular wall. The left cerebral hemisphere also appeared shriveled without a distinguishable cortical ribbon (Fig. 8). Microscopic sections revealed a thinned and gliotic cortex with loss of normal neuronal elements. The temporal lobes were the only areas relatively pre- served. There was marked ventricular enlargement, probably secondary to marked cortical atrophy. Treatment and outcome (Tables 12 and 17): In the preantibiotic-era mortality in patients who de- veloped septic thrombosis of the superior sagittal sinus was 100% (3, 125). There has been little improvement in outcome since the development of antibiotics, mortality now being 78%. Physicians can do little to alter the usually devastating conse- quences of this disease. Treatment should include intravenous administration of appropriate antibiot- ics. If a pathogen has not been initially identified, nafcillin and cefotaxime will treat the majority of pathogens responsible for infection of the superior sagittal sinus: S. pneumoniae, other aerobic and anaerobic streptococci, S. aureus, H. influenzae and Klebsiella sp. If Pseudomonas is known to be the primary pathogen, a combination of carbenicillin (or ticarcillin) with an aminoglycoside should be administered. Intravenous mannitol or corticosteroids may re- duce cerebral edema and decrease the risk of brain herniation. However, all four patients who received 102 SOUTHWICK, RICHARDSON, AND SWARTZ Fic. 8. A-P cross-sections of the brain from Case 6. Top left is the most anterior slice. Note the markedly dilated ventricles and the diffuse severe cortical atrophy. these medications died soon after this treatment was started (Cases 1, 3-5), moderate to severe cer- ebral edema being noted in three (Cases 1, 4, 5) on postmortem examination. Mannitol should be ad- ministered with caution, since severe dehydration may predispose to further dural sinus thrombosis. Anticoagulants are contraindicated in this condi- tion, since venous hemorrhagic infarcts have been commonly found to be associated with superior sagittal-sinus thrombosis. Infected air sinuses should be surgically drained to prevent continued spread of infection via the ethmoidal and cortical veins (120, Case 7). As com- pared to septic cavernous and lateral-sinus throm- bosis where surgical drainage has often proved life saving, in superior sagittal disease craniotomy has frequently been followed by rapid clinical deterio- ration (4, 62, 68, 120). This procedure should be avoided unless a large abscess has been demon- strated and the patient’s condition is rapidly worsening. Acute complete occlusion of the superior sagittal sinus has usually proved rapidly fatal, due to asso- ciated cortical vein thrombosis, venous infarction, cerebral edema, and loss of the capacity to effect CSF resorption. When this dural sinus has become occluded during acute meningitis, transtentorial or cerebral herniation may follow. Patients who sur- vive acute complete thrombosis of the superior sag- ittal sinus, as exemplified by our Case 6, will prob- ably be left with multiple severe neurological se- quelae. Patients with partial subacute thrombosis, especially of the anterior segment of the superior sagittal sinus secondary to air sinus infection, may recover with minimal sequelae depending on the extent of venous infarction (101, Case 7). With time, an extensive venous collateral circulation may develop and elevated CSF pressure may resolve. Conclusions With the development of antibiotics, septic thrombosis of the dural venous sinuses has become a rare disease. Over the past 36 years, only 19 cases have been diagnosed at the Massachusetts General Hospital. With two exceptions, the possibility of septic dural-sinus thrombosis was not even consid- ered on initial presentation. Physicians must keep in mind that infections of the facial area, air si- nuses, middle ear, mastoid, or meninges can be complicated by this life-threatening disease. Eight cases of septic cavernous-sinus thrombosis are described and 88 cases reported in the antibiotic era are reviewed: 1. Untreated sphenoid-sinus infection is empha- sized as a major risk factor for septic cavernous- sinus thrombosis. 2. Retroorbital, frontotemporal, or occipital headaches often precede periorbital edema and fe- ver by several days. 3. Chemosis, proptosis, and palsies of all three oculomotor nerves remain common findings in this disease. Other important findings, often not men- tioned in previous reports, include isolated sixth- nerve palsy and hypo- or hyperesthesia of the oph- thalmic and maxillary branches of the fifth nerve. 4. In addition to the classical acute form of cav- ernous sinus thrombophlebitis, slowly progressive cavernous-sinus thrombosis can occur. 5. Useful laboratory studies for diagnosis include: (a) lumbar puncture (which generally reveals poly- morphonuclear pleocytosis), (b) sinus films with overpenetrated lateral views or tomograms (which may demonstrate marginal sclerosis and opacifica- tion of the sphenoid air sinus), (c) CT scan of the brain, sinuses and orbits with and without contrast (which may rarely demonstrate increased density in the area of one or both cavernous sinuses, a dilated superior ophthalmic vein, an orbital abscess, and/or opacification of the sphenoid and ethmoid air sinuses), (d) carotid arteriography (which fre- quently demonstrates narrowing of the intracavern- SEPTIC DURAL VENOUS SINUS THROMBOSIS 103 ous segment of the carotid artery), and (e) orbital venography (the most definitive method of dem- onstrating cavernous-sinus occlusion). 6. Treatment should include intravenous anti- biotics directed against Staphylococcus aureus, other gram-positive organisms, and anaerobes. An- ticoagulation should be considered, if at all, only if there is no evidence of cortical venous infarction. Surgical drainage of air sinus infection should be performed on an emergent basis. 7. Morbidity and mortality remain high. Fifty percent of the patients in our series died, and 29% of other cases reported in the antibiotic era. Survi- vors often had persistent neurologic sequelae. 8. Early recognition and treatment of primary infections, in particular sphenoid air-sinus infec- tion, remain the best means of preventing cavern- ous-sinus thrombosis. Four cases of septic lateral-sinus thrombophle- bitis are described, and 60 previously reported from the antibiotic era are reviewed: 1. This disease is almost exclusively a complica- tion of otitis media and/or mastoid infection. 2. Persistent severe unilateral frontooccipital headache followed by nausea and vomiting, and less commonly by diplopia and vertigo, are the primary symptoms associated with this disease. Earache and ear drainage often precede these symptoms by sev- eral weeks, 3. Evidence for infection is almost always present. on otologic examination. When septic lateral sinus thrombosis develops, fever persists despite appro- priate antibiotic therapy. Papilledema is observed in over half of cases. Focal neurologic signs are uncommon but may include ipsilateral sixth-nerve paralysis. 4. Useful laboratory studies for diagnosis include: (a) lumbar puncture (which often reveals elevated CSF pressure and a few white blood cells, usually lymphocytes), (b) mastoid radiographs (which dem- onstrate abnormalities consistent with infection in nearly all cases), (c) dynamic brain scan (which may demonstrate lack of flow in the lateral sinus), (d) CT scan with contrast (which may demonstrate associated cerebritis, brain abscess, hemorrhagic infarction secondary to cortical vein thrombosis, and/or a filling defect in the area of the sigmoid and lateral sinuses), and (e) carotid angiography with venous phase studies (which remains the most definitive method, aside from surgical exploration, for demonstrating lateral sinus occlusion). 5. Treatment should include intravenous anti- biotics directed against the major pathogens asso- ciated with this disease: Proteus species, E. coli, S. aureus, and anaerobes. Radical mastoidectomy and exploration of the area overlying the lateral sinus are necessary to effect a cure, since purulent ma- terial or infected granulations are always found adjacent to the lateral sinus. 6. Outcome is generally favorable in this disorder; fatalities did not occur in our series and a mortality tate of 12% was observed in previously reported cases in the antibiotic era. Chronic sequelae (otitic hydrocephalus, seventh- or eighth-nerve dysfunc- tion, and decreased visual acuity) were observed in 13% of cases. Seven cases of septic superior sagittal-sinus thrombosis are reported and 16 previously reported cases in the antibiotic era are reviewed: 1. This disease is primarily a complication of meningitis. Infections of the frontal and maxillary sinuses, as well as pulmonary and pharyngeal in- fections, can also Predispose to thrombosis of the superior sagittal sinus. 2. Presenting symptoms include severe headache, confusion, vomiting, and focal or grand mal sei- zures. 3. The most frequent physical findings include high fever, depressed mental status and hemi- or quadriparesis. Papilledema is a less common find- ing. 4. Useful laboratory studies for diagnosis include: (a) CT scan with and without contrast (which may enable specific diagnosis if an empty delta sign is demonstrated), (b) lumbar puncture (CSF pressure is almost always elevated and the CSF formula is almost always consistent with frank meningitis or a parameningeal process), (c) electroencephalogram (which frequently demonstrates focal electrical ab- normalities or diffuse slow-wave activity), (d) sinus films (which in cases without evidence for bacterial meningitis may demonstrate opacification of the frontal, ethmoid, or maxillary sinuses), (e) dynamic brain scan (which may demonstrate lack of blood flow in this dural sinus), and (f) carotid angiography with careful venous phase studies (which remains the most definitive method of diagnosing, in partic- ular, partial superior sagittal-sinus thrombosis). 5. Treatment includes intravenous antibiotics di- rected against the pathogens most commonly as- sociated with this disease: Streptococcus pneumo- niae, S. aureus, or other streptococci, and Klebsiella species. Mannitol and corticosteroids may be help- ful in controlling the often severe cerebral edema. Infected air sinuses may require surgical drainage. Craniotomy should be avoided in most cases, since this procedure is frequently followed by rapid clin- ical deterioration. 6. Despite appropriate treatment, acute complete occlusion of the superior sagittal sinus is usually rapidly fatal (overall mortality, 78%). Patients with subacute thrombosis of the anterior segment of this 104 SOUTHWICK, RICHARDSON, AND SWARTZ dural sinus in association with air-sinus infection may recover with minimal sequelae. Summary From 1940 to 1984, 19 cases of septic dural-sinus thrombosis have been diagnosed at the Massachu- setts General Hospital, and some 136 cases have been reported from other institutions. Septic thrombosis most frequently involves the cavernous sinuses (96 cases). Facial or sphenoid air sinus infection often precede cavernous-sinus disease. In addition to the classical signs of proptosis, chemo- sis, and oculomotor paralysis, isolated sixth-nerve palsy and hypo- or hyperesthesia of the fifth nerve may be found. The major pathogens associated with cavernous-sinus infection include Staphylococcus aureus, other gram-positive organisms, and anaer- obes. Septic lateral-sinus thrombosis (64 cases) is almost exclusively a complication of otitis media and/or mastoid infection. Organisms causing this infection include Proteus species, Escherichia coli, S. aureus, and anaerobes. Septic thrombosis of the superior sagittal sinus (23 cases) most frequently accompanies bacterial meningitis or air sinus infec- tion. Causative organisms include Streptococcus pneumoniae, S. aureus, other streptococci, and Klebsiella species. Because septic dural-sinus thrombosis is rare, this disease is frequently mis- diagnosed. Evaluation should include lumbar punc- ture, air sinus films, and computed tomographic scan with contrast. Other helpful diagnostic tests may include carotid angiography, and dynamic brain scan. Orbital venography is the most defini- tive study in cases of chronic cavernous-sinus thrombosis. Therapy should include intravenous antibiotics and early surgical drainage of purulent exudate in the air sinuses or mastoid regions. Ret- rospective analysis suggests that treatment with heparin may reduce mortality in carefully selected cases of septic cavernous-sinus thrombosis. Anti- coagulation is not recommended in other forms of septic dural-sinus thrombosis. Mortality in the an- tibiotic-era remains high, particularly in patients with septic thrombosis of the cavernous (30%) and superior sagittal (78%) sinuses. References 1. Aberg S. Nasal furuncle with thrombosis of the cavernous sinus and meningitis. Treatment with penicillin, sulfadiazine and heparin. Acta Otolaryngol Stockh 34: 192-99, 1946. 2. Abraham EP, Chain E, Fletcher CM, Gardner AD, Heatley NG, Jennings MA, Florey HW. Further observations on penicillin. Lan- cet 2: 186-87, 1941. 3. Adams R, Kubik CS, Bonner FJ. The clinical and pathological aspects of influenzal meningitis. Arch Pediatr 65: 354-76, 1948. 4. Askenasy HM, Kosary IZ, Braham J. Thrombosis of the longitudinal sinus. Neurology 12: 288-92, 1962. 5. Ata M. Cerebral infarction due to intracranial sinus thromboeis. J Clin Pathol 18: 636-40, 1965. 6. Banna M, Groves JT. Deep vascular jon in dural venous thrombosis on computed tomography. J Comput Assist Tomogr 3: 539-41, 1979. 7. Barnes B, Winestock DP. Dynamic radionuclide scanning in the diagnosis of thrombosis of the superior sagittal sinus. Neurology 27: 656-61, 1977. 8. Bassey OO, Elebute EA. Septic thrombosis of the cavernous sinus. West Afr Med J 17: 39-41, 1968. 9. Bedford MA. The “cavernous sinus”. Br J Ophthalmol 50: 41-46, 1966. 10. Bell RW. Orbital cellulitis and cavernous sinus thrombosis caused by rhabdomyosarcoma of the middle ear. Ann Ophthalmol 4: 1090- 92, 1972. 11. Brant-Zawadzki M, Chang GY, McCarty GE. Computed tomography in dural sinus thrombosis. Arch Neurol 39: 446-47, 1982. 12. Brismar J. Orbital phlebography. Acta Radiol 16: 1-16, 1975. 13. Bronson SR, Dunbar HS. Thrombosis of the dural venous sinuses as a cause of “Pseudotumor cerebri.” Ann Surg 134: 376-85, 1951. 14. Brown P. Septic cavernous sinus thrombosis. Bull Johns Hopkins Hosp 109: 68-75, 1961. 15. Bucky TL, Lahey WJ, Kunkel P. Bilateral cavernous sinus throm- bophlebitis. Conn State Med J 12: 996-99, 1948. 16, Buonanno FS, Moody DM, Ball MR, Cowan RJ, Laster W, Ball JD. Radionuclide sinography: Diagnosis of lateral sinus thrombosis by dynamic and static brain imaging. Radiology 130: 207-13, 1979. 17. Buonanno FS, Moody DM, Ball MR, Laster DW. Computed cranial tomographic findings in cerebral sinovenous occlusion. J Comput Assist Tomogr 2: 281-90, 1978. 18. Casaubon J-N, Dion MA, Larbrissea UA. Septic cavernous sinus thrombosis after rhinoplasty. Plast Reconstruct Surg 59: 119-23, 1977. 19. Cavenagh JB. Cavernous sinus thrombosis. Br Med J 1: 1195-99, 1936. 20. Childress RC, Bitzer W. Cavernous sinus thrombosis. J Fl Med Assoc 51: 94-95, 1964. 21. Clune JP. Septic thrombosis within the cavernous chamber. Am J Ophthalmol 56: 33-39, 1963. 22, Courville CB, Nielsen JM. Fatal complications of otitis media. Arch Otolaryngol 19: 451-59, 1934. 23. D’Arbela PG. Cavernous sinus thrombosis. East Afr Med J 41: 551- 59, 1964. 24. Davidson AS. Otogenic pulmonary infection. J Laryngol 74: 877, 1960. 25. Dixon OJ. The pathologic examination in cavernous sinus throm- bosis. JAMA 87: 1088-92, 1926. 26. Doorly ARC. Thrombophlebitis of the cavernous sinus: recovery. Br Med J 1: 42, 1943. 27. Eagleton W. Cavernous sinus thrombophlebitis and allied septic traumatic lesions of the basal venous sinuses. New York: Macmillan Company, p. 1, 1926. 28. Elfman LK. Thrombosis of the cavernous sinus. Arch Otolaryngol 51: 188-965, 1950. 29. Emery JL. Pulmonary embolism in children. Arch Dis Child 37: 591-95, 1962. 30. Evans HW. Cavernous sinus thrombosis. J Lancet 85: 109-11, 1965. 31. Evans RW, Patten BM. Trichinosis associated with superior sagittal sinus thrombosis. Ann Neurol 11: 216-17, 1982. 32. Fairclough WA. Drainage in infected cavernous sinus thrombosis. Austr NZ J Surg 16: 194-96, 1947. 33. Feinfeld DA, Al-Acjlar G, Lipner HI, Chirayil SJ, Hakim J, Avram MM. Syndrome of inappropriate secretion of antidiuretic hormone. JAMA 240; 856-57, 1978. 34. Foley J. Benign forms of intracranial hypertension. Toxic and otitic hydrocephalus. Brain 78: 1-41, 1955. 35. Fox SL, West GB. Thrombosis of the cavernous sinus. JAMA 134: 1452-56, 1947. 36. Friberg TR, Sogg RL. Ischemic optic neuropathy in cavernous sinus thrombosis. Arch Ophthalmol 96: 453-56, 1978. 37. Gagnon NB, Sierra-Dupont S, Huot LA, Larochelle D. Thrombosis of the lateral sinus. J Otolaryngol 6: 257-61, 1977. 38. Gallagher JP. Septic thrombosis of the cavernous sinus. Med Ann Distr Columbia 29: 278-83, 1960. 39. Gialldrenzi AF, Weiss WW, Furman DJ, Greenwald AM. Septic cavernous sinus thrombosis in a diabetic after dental extraction. J Oral Surg 32: 924-30, 1974. 41. 48. 49. 50. 51. 52. 53. 5A. 55. 56. 28 & F B 8 B SEPTIC DURAL VENOUS SINUS THROMBOSIS . Gilbert WS, Talbot FJ. Cogan’s syndrome. Arch Opthalmol 82: 633- 36, 1969. . Goldenburg M. Sulfanilamide in ophthalmology. Am J Ophthalmol 21: 54-60, 1938. . Goodhill V. Cavernous sinus thrombosis. JAMA 125: 28-31, 1944. . Gray H. Anatomy of the Human Body. Philadelphia: Lea and Febiger, p. 726, 1954. . Greenish BVI. Cavernous sinus thrombosis. Br Med J 1: 876-77, 1945. . Greer M. Benign intracranial hypertension. Neurology 12: 472-76, 1962. ;. Grove WE. Septic and aseptic types of thrombosis of the cavernous sinus. Arch Otolaryngol 24: 29-50, 1936. Haas L. Acute osteomyelitis of the maxilla in infancy with cavernous sinus thrombosis. Br Med J 2: 245-46, 1955. Harpman JA. On the management of otorhinogenic intracranial infections. J Laryngol Otol 69: 180-94, 1955. Harris FS, Rhoton AL. Anatomy of the cavernous sinus. J Neurosurg 46; 169-80, 1976. Harvey JE. Streptokinase therapy and cavernous sinus thrombosis. Br Med J 4: 46, 1974. Henner R, Ridall EG. Thrombosis of cavernous sinus treated with penicillin and heparin. Arch Otolaryngol 41: 295-97, 1945. Hickey WF, Garnick MB, Henderson IC, Dawson DM. Primary cerebral venous thrombosis in patients with cancer—A rarely diag- nosed paraneoplastic syndrome. Am J Med 73: 740-50, 1982. Hitchcock ER, Cowie RA. Sino-jugular venous graft in otitic hydro- cephalus. Acta Neurochirurgica 59: 187-93, 1981. . Horowitz S. Otogenic intracranial hypertension. J Laryngol Otol 63: 363-81, 1949. Hunt WE. Tolosa-Hunt syndrome: one cause of painful ophthal- moplegia. J Neurosurg 44: 544-49, 1976. }. Hunt WE, Meagher JN, LeFever HE, Zeman W. Painful ophthal- moplegia. Neurology 11: 56-62, 1961. . Ingstrup HM, Jorgensen PS. Tentorial changes in sigmoid sinus thrombosis. J Comp Assist Tomogr 5: 760-62, 1981. Ivey KJ, Smith H. Hypopituitarism associated with cavernous sinus thrombosis. J Neurol Neurosurg Psychiat 31: 187-89, 1968. . Jahrsdoerfer RA, Fitz-Huge GS. Lateral sinus thrombosis. South Med J 61: 1271-75, 1968. . Jensen AM. Sinus thrombosis and otogenic sepsis. Acta Otolaryngol 55: 237-43, 1962. . Johnston DF. Cavernous sinus thrombosis treated with penicillin. Lancet 1: 9-10, 1945. Kalbag RM, Woolf AL. Cerebral venous thrombosis: with special reference to primary aseptic thrombosis.New York: Oxford Univer- sity Press, p. 23, 1967. Karlin FJ, Robinson WA. Septic cavernous sinus thrombosis. Ann Emerg Med 13: 449-55, 1984. Khare BB. Pyrrolidi: Itetracycline in cavernous sinus throm- bosis. Br J Ophthalmot 51: 712-13, 1967. Kimmick H, Myers D. Lateral sinus thrombosis. Arch Otolaryngol 68: 156-59, 1958. Kinal ME, Jaeger RM. Thrombophlebitis of dural venous sinuses following otitis media. J Neurosurg 17: 81-89, 1960. Kinnaird PM, Acker JD, Snead OC. Cavernous sinus thrombosis with vascular steal syndrome. J Pediatr 94: 410-13, 1979. Krayenbuhl H. Cerebral venous thrombosis. Clin Neurosurg 14: 1- 24, 1967. Lawton C, Hobin M. Cavernous sinus thrombosis. Canad Nurse 52: 120-21, 1956. 70. Lew PD, Southwick FS, Montgomery WW, Weber AL, Baker AS. a. 74. 1. Sphenoid sinusitis: A review of 30 cases. N Engl J Med 309: 1149— 54, 1983, Lillie HL. Prognosis of septic thrombophlebitis of the cavernous sinus. J Int Coll Surg 15: 754-59, 1951. . Lloyd GAS. The localization of lesions in the orbital apex and cavernous sinus by frontal venography. Br J Radiol 45: 405-14, 1972. }. Lund WS. A review of 50 cases of intracranial complications from otogenic infection between 1961 and 1971. Clin Otolaryngol 3: 495- 501, 1978. Malik SRK, Gupta AK, Singh G, Choudhry S. Pyrrolidinomethyl in cavernous sinus thrombosis. Br J Ophthalmol 54: 113-16, 1970. Mathew NT, Abraham J, Toari GM, lyer GV. Internal carotid artery occlusion in cavernous sinus thrombosis. Arch Neurol 24: 11-16, 1971. 16. 105 McAllen PM, Shaw RE. Cavernous sinus thrombophlebitis. Br J Surg 40: 49-52, 1952. TT. Mehra KS, Somani PN. Multiple emboli in central retinal artery BR F B oe 7. 8 8 91. 101. 102. 103. 104. 105. 106. 107. 108, 109. following cavernous sinus thrombosis. J All-India Ophthalmo! Soc 15: 71-72, 1967. Meltzer PE. Treatment of thrombosis of the lateral sinus. Arch Otolaryngol 22: 131-42, 1935. Merei L. Necrosis of the wall of the sigmoid sinus and the jugular vein. Lung abscess. Recovery. Acta Oto-Laryngol 38: 78-81, 1950. Miglete AW, Harrington JW. Complications of chronic mastoiditis. RI Med J 53: 152-59, 1970. . Miklos A. The cure of cavernous sinus thrombosis phiebitis. Br J Ophthalmol 34: 235, 1950. Morantz RA, Lansky L, Batnitzky S. Non-operative management of peeudotumor cerebri caused by lateral sinus thrombosis. J Kansas Med Soc 465-66, 1980. Morrison LF, Schindler M. Cavernous sinus thrombosis. Arch Oto- laryngol 31: 948-54, 1940. Morse HR. Intracranial complications of chronic mastoiditis. Arch Otolaryngol 63: 142-45, 1956. . Nail BM. Otitic hydrocephalus. South Med J 5: 1168-69, 1966. Neffeon AH. Occult thrombosis of sigmoid and lateral sinuses. Arch ‘Otolaryngol 41: 77-78, 1945. Nicholson WM, Anderson WB. Penicillin in the treatment of cav- ernous sinus thrombophlebitis. JAMA 126: 12-15, 1944. O'Connor AEF, Moffat DA. intracranial hypertension: Otitic J Laryngol Otol 92: 767-75, 1978. hydrocephalus. . Oliver KS, Diab AE, Abu-Jaudeh CN. Thrombophlebitis of the cavernous sinus originating from acute dental infection. Arch Oto- laryngol 48: 36-40, 1948. . Pace E. Thrombosis of the cavernous sinus. Arch Otolaryngol 33: 216-230, 1941. Palmer BW. Unilateral exophthalmos. Arch Otolaryngol 62: 415-24, 1965. . Palmersheim LA, Hamilton MK. Fatal cavernous sinus thrombosis following 3rd molar removal. J Oral Maxillofac Surg 40: 371-76, 1982. . Pang LQ. Intracranial complications of otitis media in this antibiotic era. Hawaii Med J 5: 426-30, 1967. Pascarelli E, Lemlich A. Diplopia and photophobia as premonitory symptoms in cavernous sinus thrombosis. Ann Otol 73: 210-17, 1964. Patronas NJ, Duda EE, Mirfakhraee M, Wollmann RL. Superior sagital sinus thromboais by computed tomography. Surg Neurology Pirkey WP. Thrombosis of the cavernous sinus. Arch Otolaryngol 51: 917-24, 1950. . Poltera AA, Jones AW. Intracranial venous thrombosis in Uganda. East Afr Med J 50: 634-43, 1973. Pratt LW. Cavernous sinus thrombosis. J Maine Med Assoc 50: 817-22, 1969. Price CD, Hameroff SB, Richards RD. Cavernous sinus thrombosis and orbital cellulitis. South Med J 64: 1243-47, 1971. 16: 1966. Rao KCVG, Knipp HC, Wagner EJ. Computer tomographic findings in cerebral sinus and venous thrombosis. Radiology 140: 391-98, 1961. Reid JL, McGuckin F. Cavernous sinus thrombophlebitis. A report on six consecutive recoveries. J Laryngol Otol 61: 273-81, 1926. Rubin HW, Cutler A. Cavernous sinus thrombosis: A report of three cases with autopsy findings. Ann Otol 49: 736-43, 1940. Russel A, Fearing SJ. Cavernous sinus thrombosis in a diabetic. Oral Surg 8: 372-77, 1955." Schall LA. Treatment of septic thrombophlebitis of the cavernous sinus. JAMA 117: 581-84, 1941. Scotti LN, Goldman RL, Hardman DR, Heinz ER. Venous throm- bosis in infants and children. Radiology 112: 393-99, 1974. Sears TP, Wileon FL. Recovery from cavernous sinus thrombosis and staphylococcal pneumonia by combined use of penicillin sulfa- diazine and antistaphylococcus serum. Rocky Mt Med 42: 838-41, 1945. Segall HD, Ahmadi J, McComb JG, Zee C-S, Becker TS, Han JS. Computed tomographic observations pertinent to intracranial ve- 106 110. 111. 112. 113. 114. 115. 116. M7. 118. 119. 120. 121. 122, 123. 124. 125. SOUTHWICK, RICHARDSON, AND SWARTZ nous thrombotic and occlusive disease in childhood. Radiology 143: 441-49, 1982. Seid AB, Sellars AL. The management of otogenic lateral sinus disease at Groote Shuur Hospital. Laryngoscope 83: 397-403, 1973. Shaw RE. Cavernous sinus thrombophlebitis: A review. Br J Surg 1952: 40-48, 1952. Sigsbee B, Deck MDF, Poener JB. Nonmetastatic superior sagittal sinus thromboeis complicating sytemic cancer. Neate oe ise. 46, 1979. Sindou M, Mercier P, Boker J, Brunon J. Bilateral thrombosis of the transverse sinuses: revascularization with venous microsurgical bypass. Surg Neurol 13: 215-20, 1979. Sjolin S. Studies on is. Acta Paediat 48: 532, 1959. Solomon OD, Moees L, Volk M. Steroid therapy in cavernous sinus thrombosis. Am J Ophthalmol 54: 1122-24, 1962. Steinherz PG, Miller LP, Ghavimi F, Allen JC, Miller DR. Dural sinus thrombosis in children with acute lymphoblastic leukemia. JAMA 246: 2837-39, 1981. Stevens J, Robinson K. Chronic cavernous sinus thrombosis: dis- cussion and report of case. J Oral Surg 35: 136-39, 1977. Stool JA, Lomas RD. Thrombophlebitis of cavernous sinus with recovery. Texas State J Med 44: 372-73, 1948. Strauss SI, Stern NS, Mendelow H, Spatz SS. Septic superior sagittal sinus thrombosis after oral surgery. J Oral Surg 31: 560-65, 1973. Stuart EA, O’Brien FH, McNally WJ. Cerebral venous thrombosis. Ann Otol Rhinol Laryngol 60: 406-38, 1951. Symonds CP. Otitic Hydrocephalus. Brain 54: 55-71, 1931. Symonds CP. Otitic hydrocephalus. Neurology 6: 681-85, 1956. Taylor PJ. Cavernous sinus thrombosis. Br J Ophthalmol 41: 228- 37, 1957. Tempae V, Dorun G. Cavernous sinus thrombosis. Arch Otolaryngol 69: 220-23, 1959. Toomey JA, Hutt HB. Thrombosis of the dural sinuses. Am J Dis Child 77: 285-302, 1949. 126. Venezio FR, Naidich TP, Shulman ST. Complications of mastoiditis with special emphasis on venous sinus thrombosis. J Pediatr 101: 509-13, 1982. Walsh FB. Ocular signs of thrombosis of the intracranial venous sinuses. Arch Ophthalmol 17: 46-65, 1937. 128. Weisman AD. Cavernous sinus thrombophlebitis. N Engl J Med 231: 118-22,1944. Welty RF. Bacterial thrombophlebitis of a cavernous sinus with recovery. Arch Otolaryngol 43: 70-72, 1946. Wiesenfeld IH, Phillips E. Thrombophlebitis of a cavernous sinus following extraction of teeth. Arch Otolaryngol 40: 497-500, 1944. Williams E. Hypopituitarism following sinusitis and cavernous sinus thrombosis. Proc R Soc Med 49: 827-28, 1956. Wolf JW. Thrombosis of the cavernous sinus with hemolytic strep- tococcal bacteremia. Arch Otolaryngol 40: 33-37, 1944. 133. Wolfe CT, Wolfe WC. Thrombosis of cavernous sinus with recovery. Arch Otolaryngol 33: 81-85, 1941. Wolfe WC, Gain JF. Thrombosis of the cavernous sinuses. Arch Otolaryngol 40: 79-84, 1944. 135. Woodhal} B. Variations of the cranial venous sinuses in the region of the torcular herophili. Arch Surg 33: 297-314, 1936. 136. Wright JLW, Grimaldi PMGB. Otogenic intracranial complications. J Laryngol Otol 87: 1085-95, 1973. Yarington CT. Septic thrombosis of the cavernous sinus. JAMA 173: 506-08, 1960. 138. Yarington CT. The prognosis and treatment of cavernous sinus thrombosis. Ann Rhinol Laryngol Otol 70: 263-67, 1961. Yarington CT. Thrombosis of the cavernous sinus. Otorhinolaryng Surg 40: 66-71, 1963. Yu WK, Shimo G. Otitic hydrocephalus. Can J Otolaryngol 4: 712- 19, 1975. Zahller M, Spector RH, Skoglund R, Digby D, Nyhan WL. Cavern- ous sinus thrombosis. West J Med 133: 44-48, 1980. Zilkha A, Daiz AS. Computed tomography in the diagnosis of supe- rior sagittal sinus thrombosis. J Comput Assist Tomogr 4: 124-26, 1980. 127. 129. 130. 131. 132. 134. 137. 139, 140. 141, 142.