Journal of Neuro-Ophthalmology, March 1998, Volume 18, Issue 1

Annual Review of Systemic Diseases

Systemic illnesses, such as vasculitides and systemic lupus erythematosus, may have ocular symptoms, or they may present with ocular signs and undiagnosed or occult systemic signs. The recently described ophthalmic and neurologic presentations and new diagnostic modalities for Wegener's granulomatosis, giant cell arteritis, and systemic lupus erythematosus are reviewed. Hypercoagulable states and their role in visual diseases are assessed, and strategies for evaluating patients with hypercoagulable states are described.

Journal of Neuro- Ophthalmology 18( 1): 67- 79, 1998. © 1998 Lippincotl- Raven Publishers, Philadelphia Annual Review of Systemic Diseases: 1995- 1996 Parti *| Larry P. Frohman, M. D., and * Paul Lama, M. D. Systemic illnesses, such as vasculitides and systemic lupus erythematosus, may have ocular symptoms, or they may present with ocular signs and undiagnosed or occult systemic signs. The recently described ophthalmic and neurologic pre­sentations and new diagnostic modalities for Wegener's granu­lomatosis, giant cell arteritis, and systemic lupus erythematosus are reviewed. Hypercoagulable states and their role in visual diseases are assessed, and strategies for evaluating patients with hypercoagulable states are described. This article updates the review that appeared in this journal in two parts ( 1,2). The article reviews recent reports in the literature regarding systemic illnesses that may cause neuro- ophthalmic signs. Whereas the empha­sis is specifically on the diagnosis, etiopathogenesis, and therapy of the neurologic or ophthalmic manifestations of these illnesses, articles reporting significant progress in the understanding of the systemic illness have been reviewed. This part of this recurrent series concentrates on the vasculitides that the neuro- ophthalmologist would most likely encounter, namely Wegener's granulomato­sis ( WG) and the giant cell vasculitides. Recent information on systemic lupus erythematosus ( SLE) is reviewed, as is the role of antiphospholipid antibodies ( APA) in producing neurologic and ophthal­mic disease. Finally, other specific causes of thrombosis responsible for neurologic and ophthalmic symptoms are discussed. In a future issue, the next part of this periodic review will cover recent advances in knowledge of in­fectious and related illnesses ( including AIDS) and sar­coidosis. From the * Department of Ophthalmology and the fDepartment of Neurosciences of UMDNJ- New Jersey Medical School. Supported in part by an Unrestricted Grant from Research to Prevent Blindness, Inc. Address correspondence and reprint requests to Dr. Larry P. Frohman, Department of Ophthalmology, UMDNJ- New Jersey Medi­cal School, University of Medicine and Dentistry of New Jersey, 90 Bergen St., Newark, NJ, 07103 U. S. A. VASCULITIS Wegener's Granulomatosis Wegener's granulomatosis is a necrotizing vasculitis with a predilection for the upper and lower respiratory tracts, with glomerulonephritis a common feature. In 1990, the American College of Rheumatology estab­lished criteria for the classification of WG. These are as follows: ( 1) oral or nasal inflammation manifested by oral ulcers or purulent or bloody nasal discharge; ( 2) abnormal chest roentgenogram with nodules, fixed infil­trated, or cavities; ( 3) abnormal urine sediment with erythrocyte casts or microhematuria ( more than five erythrocytes per high- power field); and ( 4) granuloma­tous inflammation within the wall of an artery or in the perivascular or extravascular area of and artery or arte­riole. The nervous system is ultimately involved in 22% of cases, and the visual system in 60% of cases. Ocular signs may be the presenting signs of Wegener's in up to 17% of cases. The most common neurologic manifestations are pe­ripheral neuropathy ( mononeuritis multiplex) or cranial neuropathy. Cranial nerves II, III, IV, and VI are the most commonly affected. Hearing loss is a common find­ing from mechanisms such as cochlear nerve involve­ment or chronic otitis from Eustachian tube blockage. Neurologic foci may occur as a direct extension from the paranasal sinuses or orbit, from cerebral or meningeal granulomatous inflammation, or from central nervous system ( CNS) vasculitis. Episcleritis, scleritis, marginal corneal ulceration, and uveitis are typical ocular signs. Granulomatous involve­ment of the paranasal sinuses, lacrimal glands, and orbit are frequently seen. Choriocapillaritis, retinal vasculitis, or optic neuropathy may lead to visual loss ( 3). New or infrequently seen neurologic and ophthalmic signs of WG have recently been reported. Tullo and co­workers describe four patients, all with active WG, who developed florid xanthelasma as a sign of WG. Lesions were asymmetric, were worse on the side with worse orbital inflammation, and resolved or improved with therapy. The researchers point out that a markedly yel­low eyelid, especially in the presence of orbital inflam­mation, suggests WG ( 4). 67 68 L. P. FROHMAN AND P. LAMA Wegener's granulomatosis may present in a manner mimicking other neuro- ophthalmic syndromes. Gobel and others report on a 58- year- old woman who presented with a painless breast mass. This led to evaluation of a year of orbital inflammation ( previously diagnosed as orbital pseudotumor) and 6 months of trigeminal neural­gia, as well as a pulmonary nodule. The specimens from both the exenteration of the blind eye and the breast mass were consistent with WG ( 5). Agostini and associates report on a patient in whom ocular WG mimicked malignant melanoma. This patient was believed to have a choroidal melanoma, and had a history of angina pectoris and no respiratory disease. Scintillography results for malignant melanoma were falsely positive. A positive finding on antineutrophilic cytoplasmic antibody ( ANCA) assay led to open- lung biopsy, which led to the diagnosis of WG. The ocular symptoms resolved with treatment using corticosteroids and cyclophosphamide ( 6). Other manifestations seen in the head may be present­ing signs of WG. Ah- See and colleagues report the first case of WG that presented as isolated and widespread involvement of the major salivary glands ( 7). WG lim­ited to the salivary gland tends to be more limited and to carry a better prognosis. This patient did well with im­munosuppressive therapy. Facial nerve palsy is seen in the course of WG but rarely is the presenting sign. Typically, such cases have had a protracted diagnostic evaluation until the underly­ing WG has been established as the etiology of the facial palsy. Hern and others report on a patient in whom the discovery of an elevated erythrocyte sedimentation rate led to obtaining an abnormal ANCA assay result. Nasal mucosal mucosa biopsy established the diagnosis ( 8). Neuro- ophthalmic signs may be present before overt systemic disease is present, or before serologic evidence of the underlying process is available. Newman and col­leagues report on four patients with the limited form of WG who presented with neuro- ophthalmic findings ( 9). All four had motility disturbances at presentation: one with oculomotor and trochlear paresis, one with oculo­motor paresis, one with trochlear and abducens paresis, and one with a horizontal gaze deviation. Other neuro­logic findings in the series included headache, seizures, confusion, focal weakness, dysphagia, and dysphonia. Other cranial neuropathies seen in this group included facial, acoustic, glossopharyngeal, vagus, accessory, and hypoglossal paresis. Magnetic resonance ( MR) imaging abnormalities in this series included meningeal enhance­ment and thickening, gyral enhancement, and T2 white matter abnormalities. The initial ANCA assay results were normal in all four; one developed a positive reac­tion later. All had necrotizing granulomatous inflamma­tion. Thus, in this series, meningocerebral inflammation preceded either systemic symptoms or laboratory abnor­malities. Wegener's granulomatosis may have hypothalamic dysfunction caused by meningeal and intracerebral in­flammation. Czarnecki and Spickler report a case where WG caused an elevated prolactin level and diabetes in­sipidus. The initial Tl- weighted MR imaging scan dem­onstrated a isointense heterogeneous mass in the sella; contrast revealed a thickened infundibulum, with both the infundibulum and hypothalamus enhancing. The im­aging findings nearly resolved with steroid therapy ( 10). Foster and others report a case of WG where the initial signs were an optic neuropathy and contralateral oph­thalmoplegia with supraorbital anesthesia. This 26- year-old man had a history of 11 years of nasal crusting, nonpurulent blood- tinged discharge, and occasional epi-staxis after a polypectomy and ethmoidectomy. Despite a negative finding on ANCA assay, deep nasal biopsies revealed small vessel vasculitis with necrosis consistent with WG, and therapy with cyclophosphamide and pred­nisone led to recovery of vision and ocular rotations ( 11). Wegener's granulomatosis may mimic or overlap other vasculitides affecting the eye. Palmowski and co­workers report on a 46- year- old patient who presented with a picture consistent with Cogan's syndrome with interstitial keratitis and vestibular- auditory dysfunction. This patient also had necrosis of the nasal septum, and the ultimate diagnosis was believed to be WG ( 12). Wegener's granulomatosis may be quiescent for long periods of time and then relapse, including in the ocular system. Piercey and Montanaro report on a patient with biopsy- proven WG ( upper and lower respiratory tract) in remission for 18 years who developed a solitary nidus of orbital recurrence. This was successfully treated with cy­clophosphamide ( 13). Advances have been made in the laboratory, radio­logic, and pathologic armamentarium used to diagnose WG. Many of the advancements in the serologic testing have been in the use and interpretation of the ANCA assay, which has become the primary diagnostic screen­ing test for WG. Edgar summarizes the utility of this indirect immuno-fluorescent test. There are two major staining patterns: cytoplasmic, or classic, ANCA ( c- ANCA), whose target antigen is proteinase 3; and perinuclear ANCA, whose target antigen is myeloperoxidase. Other antigens also can give positive reactions. Although c- ANCA is largely associated with WG, whereas perinuclear ANCA is as­sociated mainly with microscopic polyangiitis in the kid­ney, other diseases may cause a positive finding on ANCA assay. These include autoimmune hepatitis, Henoch- Schoenlein purpura, ulcerative colitis, and ma­lignancies ( 14). Rao and colleagues prospectively studied 346 con­secutive patients believed to have vasculitis based on the c- ANCA assay results. After excluding those with known WG and those who had been on noncorticosteroid immunosuppressive therapy before performing the c- ANCA assay, 212 met the study criteria. Of these, 14 ( 7%) had positive serologic findings for ANCA. Twenty-five patients met the 1990 American College of Rheu­matology criteria for the diagnosis of WG. The overall sensitivity rate for WG was only 28%, yet the specificity rate was high ( 96%). Among the subset of 25 who met J Neuro- Ophthalmol, Vol. IS, No. I, 1998 ANNUAL REVIEW OF SYSTEMIC DISEASES 69 American College of Rheumatology criteria, six were biopsy- positive, and five of these six ( 83%) had positive c- ANCA assay results. Thus, a negative finding on c- ANCA assay does not rule out clinical WG, yet a posi­tive finding is highly specific for WG ( 15). Rao and others also reviewed all relevant articles ex­amining the utility of the c- ANCA assay as a diagnostic test for WG. The reported sensitivity rate of c- ANCA assay ranged from 34% to 92%, and the specificity rate from 88% to 100%. If all data were pooled, the overall sensitivity rate was 66%, and the specificity rate, 98%. If only active disease was examined, the pooled sensitivity rate was 91%, and selectivity rate, 99%. If the pooled data for patients with clinically inactive disease were examined, the sensitivity rate was 63%, and the speci­ficity rate was 99.5%. Rao and others point out that data from the c- ANCA assay must be interpreted in the con­text of disease activity ( 16). The clinical utility of this specificity in limited ocular disease is illustrated by the case reported by Soukiasian and colleagues ( 17). Their patient presented with a con­junctival nodule and scleritis; biopsy of the nodule sug­gested Wegener's. When the ANCA assay finding was positive, the patient was successfully treated with trime­thoprim- sulfamethoxazole ( TS; see later), and the ANCA assay findings normalized. Similarly, Yamashita and coworkers report on a pa­tient who had bilateral exophthalmos for 9 months, be­lieved to be caused by idiopathic orbital pseudotumor. When systemic signs were noted ( microhematuria, puru­lent nasal discharge), an ANCA assay was performed and found to be elevated. The patient was specifically treated with cyclophosphamide and corticosteroids, and the inflammation resolved ( 18). Yamashita and others thus recommend that all patients with orbital pseudotu­mor undergo an ANCA assay. Recent reports have dealt with the utility of ANCA as a marker for disease response to therapy. Power and colleagues report a series of eight patients with scleritis alone or scleritis and peripheral ulcerative keratitis. None had the diagnosis of WG before their evaluation. All eight remitted in response to oral cyclophosphamide therapy, with or without adjunctive corticosteroids. Five patients relapsed with treatment withdrawal; in four of these five ( 80%), the ANCA assay result had not nor­malized during therapy. The ANCA assay showed a sig­nificant rise prognosticating clinical relapse in only one of these five. All three patients who remained in remis­sion continued to have a normal result on ANCA assay. Thus, in this series, if the ANCA level did not normalize with therapy, the patient with ocular manifestations of WG was more likely to relapse ( 19). Gobel and associates studied another assay's utility in diagnosing WG: the antiendothelial cell antibody ( AECA) assay. Elevated AECA assay levels were found in all patients with active WG but not in all of those who had positive c- ANCA. Elevated AECA was seen in some WG patients with inactive disease, but no patient with active WG had a normal AECA level. In general, pa­tients about to enter remission would show a decrease in AECA, and patients entering a relapse would demon­strate an elevation in AECA. Thus, AECA assay may be another useful assay in diagnosing WG and may be use­ful in predicting disease activity and response to therapy ( 20). Provenzale and Allen retrospectively reviewed com­puted tomography ( CT) and MR imaging scans of 15 patients with known WG and found abnormalities found in 7. These were dural thickening with enhancement ( n = 3), infarcts ( n = 2), hyperintense signal on T2 ( n = 2), and abnormalities in the brain stem ( n = 2). Dural thickening with enhancement ( n = 3) is thought to rep­resent granulomatous involvement of the dura. No case demonstrated extension into the CNS from contiguous orbital or sinus involvement. No patients showed evi­dence of CNS vasculitis ( 21). Muhle and others reviewed the imaging findings of a subpopulation of WG patients ( 22). Of 121 patients with biopsy- proven WG, 12 had clinical orbital involvement. Of these, 10 underwent MR imaging scanning; 2 of these 10 also underwent a CT scan. They report that the MR imaging findings of orbital granuloma are low signal in Tl and T2 weighted spin echo sequences, with an inho-mogeneous enhancement characteristic. Nuclear medicine studies also have advanced diagnos­tic imaging capabilities. Roccatello and others, using in­dium scans, studied 16 patients with rapidly progressive glomerulonephritis with arteritis and extensive crescent formation ( 23). Eight of these 16 were known to have WG. They found that significant accumulation of the tracer in the paranasal sinuses suggested that the cause of the glomerulonephritis was WG. This accumulation of indium was observed to reverse with therapy in some cases. They conclude that indium scan may prove to be a useful diagnostic tool in WG. In contrast, in a limited experience, ocular ultrasound was not useful in identifying WG. Janknecht and col­leagues used ultrasound in a patient with a sclerochoroi-dal granuloma and could not differentiate the lesion from uveal melanoma ( 24). Histologic evaluation demon­strated that WG was the underlying process. As with other vasculitides, the cornerstone of diagno­sis of WG remains histopathologic examination. Matsub-ara and coworkers report on the utility of early nasal biopsy in the diagnosis of WG ( 25). They evaluated 11 patients with WG who had undergone nasal biopsy. The pathologic findings in this series were as follows: micro­abscess in vascular walls in 82%, leukocytoclastic cap-illaritis in 73%, fibrinoid necrosis in 45%, leukocytoclas­tic endovasculitis in 27%, palisading granuloma in 100%, microabscess in 91%, and diffuse granulomatous tissue in 82%. Eight of these 11 patients ( 73%) went into remission; the remaining 3 ( 18%) died of their disease. Leukocytic vasculitis was more common in the fatal cases and pre­dicted dissemination and a poor prognosis, whereas the specific form of leukocytoclastic endovasculitis was seen only in the fatal cases. The authors suggest that the J Neuw- Ophlhalmol, Vol. 18. No. I, 1998 70 L. P. FROHMAN AND P. LAMA pathologic subtype may assist the treating clinician in selecting how aggressive a regimen will be employed. The availability of such seminal information may prove to be useful for the clinician as new therapies for WG become available. The traditional therapy of cyclo­phosphamide and glucocorticosteroid is effective but is relatively toxic, and alternate therapies for those with less aggressive illness have been sought. Several articles report on some of these alternative strategies. Le Thi Huong and others investigated the use of an alternate dosing strategy of cyclophosphamide ( 26). Fourteen consecutive patients with active WG received prednisone and monthly pulse cyclophosphamide. One patient died of sepsis related to leukopenia. Long- term remission was induced in 42%, and responders were more likely to have less extensive disease than were nonresponders. Side effects were common, although, aside from the one death, were mild. Reinhold- Keller and colleagues looked at the results of therapy of initial- phase WG with TS ( 1920 mg/ day) in 19 patients. Eleven of these 19 ( 58%) went into complete or partial remission with this therapy. The mean length of this remission was 43 months. Five ( 26%) of the remain­ing eight who did not respond had local extension of their disease, whereas three ( 16%) developed generalized WG ( 27). This group also looked at the utility of this therapeutic regimen in 24 patients with generalized WG who had already received immunosuppressive therapy. Of the patients with generalized WG who received TS after standard therapy, 42% relapsed with a mean time to relapse of 13 months. In the group that received standard therapy and did not get TS, 29% relapsed after a mean duration of 22.5 months. Thus, in this study, TS was not a useful adjunct therapy for advanced disease and may be more useful in initial phases of WG. Stegeman and others, in noting that respiratory infec­tions have been reported to trigger relapses in WG, con­ducted a prospective study of TS at a dose of 1920 mg/ day for 24 months to study if this would diminish the frequency of relapse in patients in whom remission had been induced. Eight of 41 ( 20%) patients had to stop TS because of side effects. Eighty- two percent of those re­ceiving TS were in remission at 24 months, compared with 60% of the placebo group. There were no differ­ences in ANCA titers between the two groups. They concluded that TS prevents relapses in WG ( 28). Another alternate therapy for non- life- threatening WG has been reported by Sneller and coworkers. Weekly low- dose methotrexate and prednisone induced remis­sion in 71% of their patients, with a median time to remission of 4.2 months and a median time to relapse of 29 months. Among the patients who relapsed, six of the eight ( 75%) who were retreated with the regimen had remission reinduced. The authors concluded that weekly low- dose methotrexate with prednisone is a reasonable alternate therapy in selected cases of non- life-threatening WG ( 29). Person reports on the therapy of a case of WG with dapsone ( 30). This 73- year- old man had chronic nasal obstruction and several cutaneous nodules that patho­logically showed granuloma lobular panniculitis. A sub­sequent nasal biopsy showed granulomatous vasculitis consistent with WG. Therapy with TS did not halt dis­ease progression. Dapsone ( which acts by inhibiting neu­trophil myeloperoxidase and neutrophil chemotaxis), at a dose of 50 mg twice daily, controlled his clinical illness and normalized his sedimentation rate, and a mainte­nance dose of 25 mg daily maintained this remission. Alternate therapies for severe disease unresponsive to cyclophosphamide and steroids have been reported. Georganas and associates treated a 32- year- old man, who had had a splenectomy, with severe WG that was refrac­tory to this traditional therapy ( 31). He was initially sta­bilized with plasmapheresis and gamma globulin, and remission was successfully induced with cyclosporine. In Europe, there is an experience with therapy of des­perate cases with rabbit anti- thymocyte globulin. Hagen and others treated five patients, who were unresponsive to or could not tolerate alkylating agents, with anti-thymocyte globulin. Four of five ( 80%) of patients re­sponded, with side effects including labial herpes sim­plex and serum sickness ( 32). Murray reports on a 53- year- old man with WG ( and a pituitary adenoma) who developed an ocular complica­tion, probably secondary to his therapy for WG. While on immunosuppression consisting of cyclophosphamide, 100 mg daily, and 10 mg of prednisolone twice daily, bilateral panuveitis and acute retinal necrosis developed. Fortunately, institution of acyclovir and ganciclovir pre­vented progression of retinal necrosis ( 33). Morgenstern and Pardo report a neurologic complica­tion of therapy for WG ( 34). Their patient developed biopsy- proven progressive multifocal leukoencephalopa-thy after therapy with cyclophosphamide and prednisone. Withdrawal of immunosuppression led to neurologic re­covery. Temporal Arteritis Temporal arteritis ( TA), a giant cell arteritis, is the most common vasculitis likely to be encountered clini­cally by the ophthalmologist and neurologist. TA largely affects large- and medium- sized arteries, which have a well- defined elastic lamina. Temporal arteritis, is some­what of a misnomer, since it is not limited to involve­ment of cranial arteries. Any arterial bed may be af­fected, although extracranial branches of the carotid ar­teries are the most commonly affected vessels. Ischemia from TA is rarely observed below the neck, with pref­erential involvement of the ophthalmic, posterior ciliary, superficial temporal, facial, occipital, and maxillary ar­teries. What these arteries have in common is a high elastic component in their media and adventitia ( 3). Temporal arteritis is a common vasculitis and has well- reported clinical manifestations. Nonetheless, a few unusual clinical features have been described. Sonnen-blick and colleagues report formed visual hallucinations ( the Charles Bonnet syndrome) as the presenting sign of TA in an 87- year- old woman ( 35). The hallucinations were followed by headache and monocular visual loss. J Neuro- Ophlhalmol, Vol. 18, No. I, 1998 ANNUAL REVIEW OF SYSTEMIC DISEASES 71 After biopsy confirmed TA, the institution of corticoste­roid therapy eliminated the hallucinations. Seven months later, the patient experienced a recrudescence of the vi­sual hallucinations, which responded to an increase in the corticosteroid dose. Fineman and others report on a 77- year- old patient with TA whose presentation was a non-embolic branch retinal artery occlusion. Three weeks later, the patient developed ipsilateral ischemic optic neuropathy ( 36). Nassani and coworkers report an un­usual presentation of TA, with bilateral orbital inflam­mation seen clinically and demonstrated on both CT and MR imaging scans ( 37). A temporal artery biopsy showed that TA was the source of the inflammation. Ruiz- Masera and colleagues describe a case of facial artery involvement in TA where the presenting sign was a submandibular mass ( 38). There are cases of TA where it is difficult to differ­entiate between the polymyalgic onset of rheumatoid ar­thritis in the elderly and TA. Kassimos and others found that assaying for serum cytidine deaminase may help to differentiate these two clinically similar disorders ( 39). They found that 36 of 40 ( 90%) patients with TA or polymyalgia rheumatica ( PMR) demonstrated normal levels of cytidine deaminase ( mean 8.64 U/ mL), whereas in the 20 patients with polymyalgic rheumatoid arthritis the mean was 21.33 U/ mL. This was statistically signifi­cant ( p < 0.0001), and they concluded that the level of cytidine deaminase may therefore be a useful marker to differentiate polymyalgic rheumatoid arthritis from TA. Several immunologic markers are being evaluated as to whether they are useful in understanding the patho­genesis of and in guiding rational therapy of TA. Liozon and associates prospectively studied 86 biopsy- proven TA patients and 50 controls to assess if anticardiolipin antibodies ( ACA) were markers for disease activity ( 40). Before therapy, 50% of the TA cases and 8% of the controls were positive for ACA ( 36% had ACA- IgG, 17% had ACA- IgM). In this study, there was no asso­ciation between the presence or level of ACA and isch­emic ocular manifestations of TA. Corticosteroid therapy normalized ACA levels, and relapses were accompanied by an increase in ACA. Their conclusion was that the presence of ACA may be a marker for TA disease ac­tivity. Chakravarty and associates evaluated ACA levels in 99 patients with PMR or TA. They found that among patients with a presentation of pure PMR, the finding of an elevated ACA at time of presentation carried a 4.82 relative risk for the development of TA. They also found that 35 ( 60%) of the patients presenting with TA with a high level of ACA at presentation developed severe vas­cular complications ( defined as stroke or anterior isch­emic optic neuropathy). No patient in their study who did not present as pure TA developed these ischemic vascu­lar complications. They reasoned that an elevated ACA level in a patient presenting with TA may prognosticate for a more severe course ( 41). Salvani and others performed a prospective study of 38 patients with TA or PMR. They studied levels of soluble interleukin- 2 ( sIL- 2R) and CD8- positive cells as­sayed before the institution of therapy, 6 months into corticosteroid therapy, and at the patient's last visit. As opposed to controls, patients with active disease had higher sIL- 2R and lower levels of CD8 cells. With therapy, erythrocyte sedimentation rate, C- reactive pro­tein, and sIL- 2R declined significantly. Despite 6 months of therapy, CD8 levels remained significantly lower than in controls, and sIL- 2R was significantly higher than in controls. They found that if the percentage of CD8 cells after 6 months of therapy was more than one standard deviation below that found in controls, then the patients required significantly longer corticosteroid therapy, and concluded that reduced levels of CD8 cells at 6 months of therapy may identify a group with more severe disease ( 42). Emilie and associates studied the production of serum interleukin- 6 ( IL- 6) levels in patients with TA ( 43). They found that compared with controls, serum IL- 6 levels are increased in patients with TA and fall with corticosteroid therapy. IL- 6- producing cells were demonstrated in all layers of arteries effected, but were particularly rich in areas of the media that were in contact with the internal elastic lamina. In the media, they found that most of the IL- 6- producing cells were macrophages, whereas in the intima, fibroblasts also were found to produce IL- 6. Roche and coworkers studied both IL- 6 and tumor necrosis factor- alpha levels in patients with TA and PMR. They found that IL- 6 was increased before therapy in both PMR and TA patients, whereas tumor necrosis factor- alpha was not. Corticosteroids rapidly reduced cir­culating IL- 6 levels but did not diminish the increased local production of IL- 6. They also found that changes in circulating IL- 6 levels varied in parallel with alterations in the clinical status, and that withdrawal of corticoste­roids led to increased IL- 6 levels. They believed that this suggested that IL- 6 contributes to disease manifestations, and that monitoring IL- 6 may play a role in guiding the dose of corticosteroids in these patients ( 44). Weyand and Gorozny conclude that TA is likely a disease where the vessels carry an as- of- yet unidentified antigenic target. Selected helper T cells migrate to le­sions where cytokines are produced and macrophages are activated. They believe that a genetic risk factor includes the expression of a region of the HLA- DR molecule ( 45). Kraft and coworkers report the promise of color Dopp- Ier ultrasound as a possible new tool for identifying pa­tients with TA ( 46). They studied 10 patients with TA and 8 with PMR, as well as 23 controls, with high-resolution ultrasound. The size of the temporal artery lumen was assessed, as was blood flow velocity. They found that there was a typical hypoechoic halo around the perfused lumen of stenotic or thrombosed superficial temporal arteries. This halo was not seen in controls or in the arteries of PMR patients. Furthermore, they said that the halo disappeared with 10 to 14 days of corticosteroid therapy. Dautzenberg has described a case of dementia where SPECT scanning improved with prednisone therapy ( 47). / Neuro- Ophthalmol, Vol. IS. No. I, 1998 72 L. P. FROHMAN AND P. LAMA The patient was an 81- year- old woman who underwent SPECT scanning to evaluate an acute dementia evolving over a few weeks. This was believed clinically to be vascular, but the initial SPECT scanning suggested Alz­heimer's dementia as the underlying etiology, with poor perfusion in both parietal regions. Since this patient was previously known to have had TA, she was treated with prednisone; the dementia partially reversed, and the right parietal perfusion was seen to improve on SPECT scan 1 year into her therapy. The limitations of the current therapy of TA are well represented by the case reported by Rauser and Ris-mondo, a 74- year- old woman with TA who presented with 2 weeks of diplopia and 4 weeks of temporal head­ache and myalgia, with recent weight loss. The Wester-gren erythrocyte sedimentation rate was 115 mm/ h, and 80 mg of prednisone daily was begun. Results from a temporal artery biopsy were positive. The diplopia and constitutional symptoms resolved within 1 week of therapy, and 2 weeks into therapy the sedimentation rate was 31 mm/ h. Despite this clinical response, 1 week later, the patient had ischemic optic neuropathy left eye ( OS) while still on 80 mg daily of prednisone. At this point she was free of constitutional symptoms and her sedimentation rate was 15 mm/ h ( 48). The two large concerns with the standard protracted corticosteroid therapy for TA are as follows: 1. How should patients be treated who break through or are unresponsive to corticosteroid therapy? and 2. Is there a way to avoid corticosteroid- induced com­plications of therapy? Currently, there is no proven effective alternate regi­men to corticosteroid therapy. Thus, the thrust of most of the investigations of the therapy of TA center around whether there are safe alternate therapies employing ad­junct agents to long- term corticosteroids, or if there are regimens employing other steroid- sparing agents that al­low for a shorter duration of or lesser total dose of cor­ticosteroid therapy. Hernandez- Garcia and others treated 11 patients with newly diagnosed TA with an initial high dose of pred­nisone and rapidly tapered the dose. The patients also received weekly oral methotrexate for 2 years. They found that it took a mean of 14 weeks to get down to a dose of 10 mg of prednisone daily, and 30 weeks until withdrawal of the corticosteroid was possible. While on this regimen, two patients relapsed, and five developed side effects of corticosteroids, whereas no complications of methotrexate were seen. They believed that their small series showed that methotrexate might be a safe and useful therapeutic modality in TA ( 49). van der Veen and others performed a randomized, double- blinded study of 40 patients with PMR and 6 with clinical and biopsy- proven TA ( 50). All patients were begun on 20 nig/ day of prednisone and then were ran­domized to receive either 7.5 mg/ week of methotrexate or placebo. The prednisone was tapered as clinical signs abated, and the erythrocyte sedimentation rate normal­ized. In this study, they found no difference between the two groups in time to remission or relapses, or in the cumulative dose of prednisone required, and concluded that no steroid- sparing effect of weekly methotrexate was demonstrated. In a small pilot study, Kupersmith and colleagues studied whether the addition of methotrexate allowed more rapid tapering of corticosteroid therapy in TA pa­tients ( 51). Their preliminary data show no difference between the ability to taper steroids in patients treated with corticosteroids and placebo versus that in patients treated with corticosteroid and methotrexate. They did, however, in this small series, show an ability to safely taper the patient's prednisone dose more rapidly than has been typically recommended. Thus, it is too early to define whether methotrexate will play a significant role in the therapy of TA. Takayasu's Arteritis Takayasu's arteritis may resemble TA histologically, but it affects a different patient population, namely young adults, usually 15 to 20 years of age, with a marked female preponderance ( 9: 1). The target tissues typically are large arteries such as the aortic arch and its tributaries. Active arteritis is flowed by a vasoocclusive phase. The typical signs of this vasculopathy include absence of or decreased arterial pulses and bruits of the aorta or its major branches. Symptoms from carotid oc­clusive disease are more typical than those caused by vertebrobasilar involvement ( 3). Kumar and others report a 65- year- old man with Taka­yasu's arteritis who had painless progressive visual loss OD over 1 year. He also had transient bilateral visual obscurations lasting up to a few minutes. The patient was light perception OD with a dense cataract preventing inspection of the fundus. The acuity in the left eye was 6/ 24 OS. Dilated episcleral and conjunctival vessels were seen OU, as was bilateral iris neovascularization. The left fundus demonstrated mild temporal pallor with a few anastomotic loops on the disc surface. " Boxcarring" was seen in large arteries, with venous dilation and tor-tuoisty. The fluorescein angiogram showed a prolonged arm to retina time, delayed choroidal filling, and patchy hypofluorescence of the disc. There was early cystoid macular edema, but no disc neovascularization. The au­thors commented that it was unusual to have such severe anterior segment ischemia without more advanced fun­dus findings ( 52). Similar to the study performed by Kraft and others in TA ( 46), Raninen and colleagues looked at the utility of B mode ultrasonography as an aid in the diagnosis of Takayasu's arteritis ( 53). They measured total wall, in-timal plus medial, and adventitial thickness of the com­mon carotid, subclavian, and common femoral arteries, as well as the abdominal aorta. They found that in their 16 patients, compared with their control group, all ves­sels other than the common femoral showed significant increases in total wall and intimal plus medial thickness. ./ Nairo- Ophllialmal, Vol. 18, No. 1. 1998 ANNUAL REVIEW OF SYSTEMIC DISEASES 73 SYSTEMIC LUPUS ERYTHEMATOSUS Reviewing all progress in the understanding of and therapy of SLE is beyond the scope of this review. Here we concentrate on a few specific presentations and points regarding neurologic and ophthalmic involvement in SLE. Ahmadieh and others recently reported in this journal on an 11- year- old girl diagnosed with systemic lupus erythematosus ( SLE) who, 3 months after diagnosis, de­veloped bilateral optic neuropathy with retrobulbar pain on ocular rotation. The initial visual acuity was light perception OU. Therapy with oral prednisolone ( 4 mg/ kg/ day) led only to recovery of counting fingers OU. Whether this patient exhibited APA is not reported in the article. The authors point out that optic neuritis in SLE in children has had a poor outcome in the few cases re­ported ( 54). The presence or absence of neurologic disease at the time of diagnosis of SLE and during its course are mark­ers for disease survival. Golstein and others studied 92 consecutively diagnosed patients with SLE ( 24 of whom had neurologic or ophthalmic disease) to see if the pres­ence of APA correlated with neurologic and ophthalmic involvement. The most common neurologic findings were stroke or transischemic attack ( n = 9), psychotic delirium ( n = 9), and seizures ( n = 6). The five oph­thalmic manifestation seen in the series were central reti­nal artery occlusion ( n = 4) and central retinal vein occlusion ( n = 1). Thirty- two percent of patients without neurologic or ophthalmic illness were positive for APA, whereas 15/ 24 ( 63%) of those with such disease were APA- positive ( p < 0.01). In 13/ 15 cases ( 87%), the detection of APA antedated the neurologic/ ophthalmic sign. There was no correlation between having a positive APA finding and the specific type of neurologic/ ophthalmic manifestation. Golstein and coworkers conclude that a positive APA finding is a risk factor for neurologic and ophthalmic disease in SLE. They do not recommend any preventa­tive therapy for the SLE patient found to harbor APA who has not yet had a neurologic/ ophthalmic complica­tion ( 55). Seaman and colleagues looked at the significance of APA in SLE in children. Twenty- nine patients with SLE diagnosed in childhood were examined, with the pres­ence of APA defined as demonstrating either a false-positive VDRL, the presence of lupus anticoagulant, or ACA- IgG or ACA- IgM. They found that 65% of their patients demonstrated at least one of these three abnor­malities, with a false- positive VDRL in 11/ 28 ( 39%), lupus anticoagulant in 16/ 26 ( 42%), and anticardiolipin antibodies in 18/ 27 ( 67%). In this series, thrombotic events were correlated with the presence of anticardio­lipin antibodies ( 56). Denburg and others recently reported on the relation of neurologic involvement in SLE and the development of guidelines for therapy ( 57). Denburg and colleagues believe that prevailing evidence supports that this is a disease process mediated by autoantibodies evident in the serum and the cerebrospinal fluid. These autoanti­bodies may effect CNS involvement by various mecha­nisms. These include antibody- dependent cytotoxicity, which may cause cell death or demyelination, or by in­terference with cell- cell communication. They point out that there is a body of evidence linking some of these potential mechanisms to specific neurologic presenta­tions in SLE. Thus, the presence of APA in SLE patients is linked to focal neurologic deficits, whereas producing antineuronal antibodies tends to be associated with dif­fuse neurologic or neuropsychiatric involvement, and having antibodies directed against ribosomal P proteins is associated with psychosis and depression. Their con­clusion is that the treatment of neurologic involvement of SLE should be tailored to its presumed etiopathogenesis in the individual patient. Thus, neurologic lupus that is thrombotic is treated using known therapeutic regimens applicable in stroke, such as warfarin, whereas neuro­logic lupus that is believed to be mediated through au­toantibodies is treated with therapeutic agents used in autoimmune disease models such as myasthenia, includ­ing immunosuppressants. HYPERCOAGULABLE STATES Although local anatomic risk factors have been well described in the pathogenesis of nonarteritic ischemic optic neuropathy, it is not always clear when an inves­tigation for other potentially important etiologic factors should be undertaken. Hypercoagulability and prethrom-botic conditions have long been associated with the de­velopment of spontaneous venous thrombosis, but the association of these conditions with arterial thrombosis has been recognized only recently ( 58- 61). Acheson and Sanders ( 62) describe coagulation abnormalities in seven patients with nonarteritic ischemic optic neuropathy over a 2- year period. Four patients had deficiencies in the anticoagulant proteins C and S as well as antithrombin III, two had antiphospholipid antibody syndromes, and one patient had reduced levels of physiologic tissue plas­minogen activator. Six of these seven patients had bilat­eral ischemic optic neuropathy, and four had experienced recurrent episodes in the same eye. Additionally, two of the seven patients had presented before the age of 30 years. Hypercoagulability may be divided into two catego­ries based on whether identifiable alterations exist in the hemostatic pathway. Patients who have identifiable ab­normalities then can be divided into subgroups, depend­ing on the site of alteration ( i. e., blood [ cellular compo­nents] serum [ anticoagulant and fibrinolytic enzymes, presence of lupus anticoagulants or antiphospholipid an­tibodies], or the vascular endothelium). These can be further subdivided, depending on whether the abnormali­ties are congenital or acquired. Until 1994, the most common identifiable disorders of hemostasis in patients with spontaneous venous and arterial thrombotic events were deficiencies in antithrombin III, protein C, and pro­tein S ( 62,63). Congenital protein C deficiency accounts for up to 10% of all patients with deep venous thrombo- J Neiiro- Ophthalmol, Vol. IS. No. I, 1998 74 L. P. FROHMAN AND P. LAMA sis or pulmonary embolism, and antithrombin III defi­ciency accounts for 3% to 8% of these patients. Protein 5 deficiency may account for 5% to 10% of unexplained venous thromboses in patients younger than 45 years of age ( 64). Interestingly, in a study of 40 cases of cerebral venous thrombosis in adults from Saudi Arabia, the most common etiology was Behcet's disease, which occurred in 25% of patients. Protein S and antithrombin III defi­ciency ( 12.5%) and APA ( 10%) were other common causes considered as sources for the thromboses ( 65). Unfortunately, most investigations for hypercoagulable states were unable to define an etiology in approximately 70% of cases. Certainly, labeling most as idiopathic is not useful from a therapeutic or a genetic pathophysi­ologic standpoint. However, in 1993, Dahlbeck and as­sociates ( 66) made the seminal discovery of resistance to activated protein C, which now is believed to account for most cases of familial thrombophilia. Such resistance to protein C was found to be caused by an R506Q mutation in the gene coding for factor V, which was termed factor V Leiden ( 67). As a result of this mutation, factor V cannot be neutralized by protein C on activation by thrombin- thrombomodulin complex, leading to excess factor V activity and a prethrombotic state ( Fig. 1). The discovery of activated protein C resistance not only accounts for a substantial percentage of hereditary thrombotic disorders, but also may provide an explana­tion for the etiology of hypercoagulability in other con­ditions in which the mechanisms of thrombosis have been previously unknown. For example, patients who are homozygous for homocystinuria have only a one in three probability of developing a thrombotic event. This implies that there may be additional factors that are necessary for the development of thrombosis. Man-del and others demonstrated that coexistence of the factor V Leiden mutation in 7 of 11 patients with ho­mocystinuria was associated with thrombotic events in 6 of these 7 patients ( 68). The only patient with both homocystinuria and factor V Leiden mutation who did not develop thrombosis had received warfarin therapy since birth. The four homocystinuria patients without the factor V Leiden mutation have yet to develop thrombosis. Despite the unequivocal importance of activated pro­tein C resistance as a cause of hypercoagulability, knowl­edge of this condition has not been well disseminated throughout the medical community as would be ex­pected, although 37 of the 115 articles published since its discovery have been written in journals typically read by primary care physicians. Additionally, only six of the articles were published in journals read by pathologists ( 69). This may account for the lack of awareness of this condition among community pathologists who are poten­tially invaluable in assisting the treating physician in selecting the appropriate laboratory studies for a hyper­coagulability workup. This was well illustrated in a study conducted at the University of Utah that was designed to establish prevalence data regarding ordering of various diagnostic laboratory tests by community and university physicians seeking to identify a specific thrombotic dis­order. Testing for other inherited thrombotic disorders such as antithrombin III, protein C, and protein S defi­ciency occurred at a rate sixfold greater than that for activated protein C resistance. Furthermore, 37% of the specimens collected for evaluation for activated protein C deficiency were not evaluable because of concomitant anticoagulation therapy ( 69). Resistance to activated pro­tein C was identified by measuring the activated partial thromboplastin time ( aPTT) with added activated protein C, and comparing the resultant value to the aPTT without addition of activated protein C. In normal individuals, addition of activated protein C prolongs the aPTT by two to three and a half times. A ratio of less than 2.0 is presumptive evidence for activated protein C resistance ( 70). Since heparin anticoagulation prolongs the aPTT, specimens sent for determination of activated protein C resistance after heparin therapy was initiated could not be assayed. These statistics underscore the lack of aware- PROTEIN C O-ENDOTHELIUM THROMBOMODULIN 1 t THROMBIN 1 THROMBIN ACTIVATED PROTEIN C ( APC) PROTEIN S APC COFACTOR INACTIVE PROTEIN C V FACTORS V & Vni: C i> INACTIVE FACTORS V & VIILC FIG. 1. Role of protein C in modulation of thrombosis. J Neum- Ophthalmol, Vol. 18, No. 1, 1998 ANNUAL REVIEW OF SYSTEMIC DISEASES 75 ness as well as the lack of understanding of the labora­tory assay used to diagnose this disorder. In summary, activated protein C resistance has emerged as the leading cause of inherited thrombophilia, and therefore appropriate laboratory studies to identify this condition should be performed when evaluating pa­tients with recurrent or unusual thromboses or throm­boembolism. Antiphospholipid Antibodies Central nervous system disease in SLE patients is well known to be associated with the presence of APA ( 71). However, the prevalence rate of CNS disease in SLE patients is variable, with reports ranging from 18% to 70% ( 72). This variability likely results from a lack of standardized methods to define subtle clini­cal CNS manifestations. Such wide variability in re­ported prevalence leads to conflicting reports with respect to its association with other features of SLE, namely the prevalence of APA antibodies. If more specific definitions of CNS disease are outlined, then prevalence rates of its associated features can be ob­tained more consistently. Toubi and others identified 96 of 340 unselected patients at a lupus clinic with CNS manifestations, including 55 with transient ischemic attacks or strokes, 24 with epilepsy, and 12 with psychi­atric disorders, and compared them with a control group of 100 patients without CNS or thromboembolic manifestations in terms of APA positivity and presence of other clinical and serologic markers of lupus disease activity ( 73). Patients with migraine or cognitive disor­ders were excluded. Overall, 55% of the patients with CNS manifestations were positive for APA compared with 20% in SLE control patients. Additionally, only 44% of patients with CNS findings had clinical and se­rologic findings consistent with globally increased lupus disease activity, and the remaining 56% were inactive. Serologic positivity for APA was strongly associated with SLE inactivity. Of the 53 patients who underwent MR imaging evaluation, 33 had changes consistent with vasculopathy and 26/ 33 had positive findings on an­tiphospholipid antibody serologic study. Only 8/ 20 pa­tients with normal results on MR imaging scans had positive APA. In summary, the study demonstrates and confirms that a strong association exists between positive APA serologic findings and CNS disease in SLE pa­tients, and that abnormal results on MR imaging scans highly correlate with positive APA antibodies in patients with a restricted definition of CNS lupus. Additionally, CNS disease can occur in the absence of other evidence of lupus disease activity. Focal cerebral infarction is the most common mani­festation of arterial thrombosis in patients with the an­tiphospholipid syndrome ( 74). However, the prognosis, and the clinical, serologic, and other laboratory features require clarification. Levine and colleagues studied 81 patients consecutively with APA, who, over a 7- year period, developed cerebral thrombo- occlusive events ( 75). Patients with the highest titers of anticardiolipin antibody had the shortest times to ischemic occlusive events. The average age was 10 years younger than the average stroke patient, and women were more often af­fected than men. Additionally, over 50% of patients had more than one recurrent occlusive event, usually occur­ring within 1 year of follow- up. The presence of APA in pathologic conditions other than stroke and recurrent fetal loss has led to the emergence of studies of prevalence of these antibodies and their potential role in the pathogenesis of other neurologic and nonneurologic conditions. Although, several reports ( 76- 79) indicate a possible association and perhaps a pathogenic role in patients with migraine, these reports have not been substantiated by investiga­tions on patients with migraine. Silvestrini and others, however, suggest that patients with migraine and APA may represent a specific subset of patients with migraine ( 80). In a study of 16 patients with migrainous stroke without other known causes of infarction, 6 of these pa­tients were positive for the presence of APA antibodies. Additionally, those with positive serologic results had fewer risk factors for stroke than those without positive serologic evidence, thus indicating a possible causal or influential role for APA in the development of migrain­ous stroke. Apropos to our earlier discussion regarding APA in other disease states is a case report of a patient with type I diabetes mellitus who presented with a central retinal artery occlusion and was found to have positive serologic results for anticardioplipin antibodies ( 81) Since patients with type I diabetes generally are believed to have de­veloped diabetes on the basis of an autoimmune process leading to insulitis and subsequent loss of ( 3 cells of the pancreas, it is not surprising that ACA antibodies may coexist, suggesting a more generalized disturbance of the immune system. The appearance of ACA antibodies has been related in two experimental models of diabetes in­sulitis ( 82,83). This implies that development of throm­botic events in patients with type I diabetes may be partly related to the presence of ACA antibodies. Until now we have focused on the complications re­lated to APA through their effects on the coagulation pathway. However, Orefice and others ( 84) report a case of benign intracranial hypertension in a patient with pri­mary antiphospholipid syndrome and no evidence of ce­rebral sinus thrombosis. Benign intracranial hypertension in patients with SLE often results from cerebral sinus thrombosis with impairment of CSF outflow. The only abnormal manifestations in this patient were high titers of APA and lupus anticoagulant. Despite the numerous reports attempting to pathoge-netically link various neurologic disorders with the pres­ence of APA, caution must be exercised before attribut­ing causal roles to these antibodies. Definitive patho­physiologic data have not been attained. Moreover, especially in disorders with other existing immunologic derangements, the presence of these antibodies may merely reflect a biological epiphenomenon and therefore are causally unrelated to the underlying disorder in ques­tion. ./ Neum- Ophlhalmol, Vol. IS, No. I, 1998 76 L. P. FROHMAN AND P. LAMA INTRINSIC PATHWAY HMWK FACTOR XII t> Xlla FACTOR XI V - f> XIa FACTOR IX - VIHa PL, Ca++ FACTOR X PROTHROMBIN V, Ca++, PL -^ THROMBIN EXTRINSIC PATHWAY DAMAGED ENDOTHELIUM t> TISSUE FACTOR, Ca++ FACTOR VII - I> FACTORVIIa FrBRTNOGEN -^- i> FIBRTN MONOMER FEB RTN " POLYMER XHIa CROSS- LINKED FIBRTN FIG. 2. Intrinsic and extrinsic coagulation pathways. SERINE PROTEASE COMPLEX V AT HEPARIN SERINE PROTEASE HEPARIN COFACTOR II •=[ THROMBIN HEPARIN INACTIVE THROMBIN INACTIVE SERINE PROTEASE ( COAGULATION FACTORS) FIG. 3. Role of antithrombin III ( AT) and heparin cofactor II in modulation of thrombosis. J Neuw- Ophthalmol, Vol. IS, No. I, 1998 ANNUAL REVIEW OF SYSTEMIC DISEASES 77 LABORATORY EVALUATION OF HYPERCOAGULABLE PATIENTS Once a prethrombotic state is suspected, systematic evaluation of the hemostatic and thrombotic pathways should be undertaken. The initial workup begins with a thorough history and physical examination followed by standard global tests of hemostasis such as platelet count, bleeding time, and measurements of the intrinsic and extrinsic clotting pathways ( aPTT and prothrombin time; Fig. 2). Abnormalities in these initial laboratory tests should direct the ordering physician toward appropriate confirmatory tests to arrive at a specific diagnosis. How­ever, it is more likely that in most cases of suspected prethrombotic states, the initial screening laboratory evaluation is within normal limits. Based on our previous discussion of congenital hypercoagulable states, labora­tory assay for activated protein C resistance should be the next step. It is important that the blood sample be obtained before initiating anticoagulation. If this test re­sult is negative, then the next step should be evaluation for anticoagulant deficiencies such as antithrombin III, protein C, and protein S. Often, however, since the range of normal is wide for each of these anticoagulants and minor deficiencies may be clinically significant, it is im­perative that functional assays also be performed. Addi­tionally, there may be existing conditions that interfere with determination of these functional assays, leading to false measurements of their activity. For example, hep­arin cofactor II ( HC- II) is a naturally occurring antico­agulant that inhibits thrombin. Because HC- II inhibits one sixth of the amount of thrombin inhibited by anti­thrombin III, patients with minor reductions in functional or antigenic antithrombin III may not be detected, lead­ing to an overestimation of its activity by 5% to 10% ( Fig. 3). With this in mind, Demers and colleagues de­veloped a new assay that uses factor Xa inhibition to determine functional antithrombin III activity ( 85). Fac­tor Xa is not inhibited by HC- II, and thus patients with borderline antithrombin III activity can be appropriately identified. Aside from measurements of anticoagulant deficien­cies, serum markers reflecting increased coagulative ac­tivity may be useful in determining the nature of the hypercoagulable state. During the early stages of blood coagulation, activated protein C may appear in the cir­culation when subcoagulant amounts of thrombin are generated. Activated protein C may complex with al­pha- 1 antitrypsin, which has a longer half- life than acti­vated protein C- antithrombin II complex. Increased amounts of these complexes reflect a stimulated coagu­lation pathway. Since the former complex has a longer half- life, it is more easily detected at an early stage of coagulation and is a sensitive marker of increased activ­ity of the coagulation process, and therefore is a useful marker in prethrombotic states ( 86). Evaluation for the presence of the lupus anticoagulant or other antiphospholipid antibodies begins with mea- PRE- KALLIKREIN FACTOR XUa- HAGEMAN FACTOR KALLIKREIN PLASMINOGEN V r- JLASMINOGEN PPRR On AAC PTTITVVA AT TODRR 1^ ACTIVATOR FACTOR XI ACTOR XIa PLASMIN - PLASMINOGEN V PROTHROMBIN OTHROMBIN FIBRINOGEN ^> FIBRIN MONOMER 6 CLOT LYSIS H> FIBRINOPEPTIDES FIG. 4. Fibrinolytic pathway. J Neuro- Ophthalmol, Vol. IS, No. 1, 1998 78 L. P. FROHMAN AND P. LAMA surement of the aPTT. An elevated PTT is suggestive but not diagnostic of the presence of the lupus anticoagulant. A mixing study ( using the patient's serum mixed with normal control serum) is necessary. If this mixture re­sults in normalization of the PTT, then this eliminates the presence of a lupus anticoagulant as the cause for PTT elevation and is implicit evidence for factor deficiency. Factor XII deficiency ( Hageman factor) may lead to el­evation of the PTT with a thrombotic tendency rather than a hemorrhagic diathesis, as in other factor defi­ciency states, as a result of interruption of the fibrinolytic pathway ( Fig. 4). Factor XII is necessary in the activa­tion of plasminogen to plasmin through its catalytic ef­fect in activating plasminogen preactivator to plasmino­gen and conversion of prekallekrein to kallekrein. Con­versely, a normal PTT value does not rule out the presence of a circulating anticoagulant, since the concen­tration of these antibodies may not be in sufficiently high titers to result in prolongation of the PTT, but yet may be a clinically significant cause of thrombosis. Dilution of phospholipid by using kaolin ( kaolin clotting time) in­creases the sensitivity of the assay. The most sensitive assay for anticardiolipin antibody is Russell's viper venom time. REFERENCES 1. Newman NJ. Neuro- ophthalmology and systemic disease. Part I. An annual review. J Neuroophthalmol 1995; 15: 109- 21. 2. Newman NJ. Neuro- ophthalmology and systemic disease. Part II. An annual review. J Neuroophthalmol 1995; 15: 241- 53. 3. Frohman LP, Bielory L. Ocular and neuro- ophthalmic vasculitis. In: Burde RM and Slamovits TL eds. Advances in Clinical Oph­thalmology, volume 2. St Louis, Mosby 1995: 89- 132. 4. Tullo AB, Durrington P, Graham E, et al. Florid xanthelasmata ( yellow lids) in orbital Wegener's granulomatosis. Br J Ophthal­mol 1995; 79: 453- 56. 5. Gobel U, Kettritz, R, Kettritz U, Thieme U, Schneider W, Luft FC. Wegener's granulomatosis masquerading as breast cancer. Arch Intern Med 1995; 155: 205- 7. 6. Agostini HT, Brautigam P, Loftier KU. Subretinal tumour in a patient with a limited form of Wegener's granulomatosis. Acta Ophthalmol Scand 1995; 73: 460- 3. 7. Ah- See KW, McClaren K, Maran AG. Wegener's granulomatosis presenting as major salivary gland enlargement. J Laryngol Otol 1996; 110: 691- 3. 8. Hern JD, Hollis LJ, Mochloulis G, Montgomery PQ, Tolley NS. Early diagnosis of Wegener's granulomatosis presenting with fa­cial nerve palsy. J Laryngol Otol 1996; 110: 459- 61. 9. Newman NJ, Slamovits TL, Friedland S, Wilson WB. Neuro-ophthalmic manifestations of meningocerebral inflammation from the limited form of Wegener's granulomatosis. Am J Ophthalmol 1995; 120: 613- 21. 10. Czarnecki EJ, Spickler EM. MR demonstration of Wegener granu­lomatosis of the infindibulum, a cause of diabetes insipidus. Am J Neuroradiol 1995; 16( suppl 4): 968- 70. 11. Foster WP, Greene JS, Millman B. Wegener's granulomatosis pre­senting as ophthalmoplegia and optic neuropathy. Otolaryngol Head Neck Surg 1995; 112: 758- 62. 12. Palmowski AM, Hille K, Ruprecht KW. Non- syphilitic interstitial keratitis and inner ear deafness in the initial phase of Wegener's granulomatosis. Klin Monatsbl Augenheilkd 1994: 205: 364- 7. 13. Piercey S, Montanaro A. Recurrent Wegener's granulomatosis: a case report and review. Ann Allergy Asthma Immunol 1996; 76: 317- 20. 14. Edgar JDM. The clinical utility of ANCA positivity. Ann Rheum Dis 1996; 55: 494- 6. 15. Rao JK, Allen NB, Feussner JR, Weinberger M. A prospective study of antineutrophil cytoplasmic antibody ( c- ANCA) and clini­cal criteria in diagnosing Wegener's granulomatosis. Lancet 1995; 346: 926- 31. 16. Rao JK, Weinberger JM, Oddone EZ, Allen NB, Landsman P, Feussner JR. The role of anti- neutrophillic cytoplasmic antibody ( c- ANCA) testing in the diagnosis of Wegener granulomatosis: a literature review and meta- analysis. Ann intern Med 1995; 123: 925- 32. 17. Soukiasian SH, Jakobiec FA Niles JL, Pavan- Langston, D. Trime­thoprim- sulfamethoxazole for scleritis associated with limited Wegener's granulomatosis: use of histopathology and antineurtro-phil cytoplasmic antibody ( ANCA) test. Cornea 1993; 12: 174- 80. 18. Yamashita K, Kobayashi S, Kondo M, Ishikura H, Shibuya Y, Hayasaka S. Elevated anti- neutrophilic cytoplasmic antibody titer in a patient with atypical orbital pseudotumor. Ophthalmologica 1995; 209: 172- 5. 19. Power WJ, Rodriguez A, Neves RA, Lane L, Foster CS. Disease relapse in patients with ocular manifestations of Wegener's granu­lomatosis. Ophthalmology 1995; 102: 154- 60. 20. Gobel U, Eichhorn J, Kettritz R, et al. Disease activity and autoan­tibodies to endothelial cells in patients with Wegener's granulo­matosis. Am J Kidney Dis 1996; 28: 186- 94. 21. Provenzale JM, Allen NB. Wegener granulomatosis: CT and MR findings. Am J Neuroradiol 1996; 17: 785- 92. 22. Muhle C, Nolle B, Brinkmann G, et al. Magnetic resonance to­mography and computerized tomography of Wegener's granulo­matosis of the orbits. Aktuelle Radiol 1994; 4: 229- 34. 23. Roccatello D, Picciotto G, Gigliola G, et al. Indium- 111- labeled granulocyte head accumulation in patients with Wegener's granu­lomatosis. Am J Nephrol 1995; 15: 500- 6. 24. Janknecht P, Mittelviefhaus H, Loffler KU. Sclerochoroidal granu­loma in Wegener's granulomatosis simulating a uveal melanoma. Retina 1995; 15: 150- 3. 25. Matsubara O, YoshimuraN, Doi Y, Tamura A, Mark EJ. Nasal biopsy in the early diagnosis of Wegener's ( pathergic) granuloma­tosis: significance of pallisading granuloma and leukocytoclastic vasculitis. Virchows Arch 1996; 426: 13- 9. 26. Le Thi Huong D, Papo T, Piete JC, et al. Monthly intravenous pulse cyclosphosphamide therapy in Wegener's granulomatosis. Clin Exp Rheumatol 1996; 14: 9- 16. 27. Reinhold- Keller E, De Groot K, Rudert H, Nolle B, Heller M, Gross WL. Response to trimethoprim/ sulfamethoxazole in Wege­ner' s granulomatosis depends on the phase of the disease. Q J Med 1996; 89: 15- 23. 28. Stegeman CA, Cohen Tervaert JW DeJong PE, Kallengerg CG. Trimethoprim/ sulfamethoxazole ( co- trimoxazole) for the preven­tion of relapses of Wegener's granulomatosis. N Engl J Med 1996; 335: 16- 20. 29. Sneller MC, Hoffman GS, Talar- Williams C, Kerr GS, Hallahan CW, Fauci AS. An analysis of 42 Wegener's granulomatosis pa­tients treated with methotrexate and prednisone. Arthritis Rheum 1995; 38: 608- 13. 30. Person JR. Limited Wegener's granulomatosis treated with dap-sone. Int J Dermatol 1995; 34: 870- 1. 31. Georganas C, Iokimidis D, Iatrou C, et al. Relapsing Wegener's granulomatosis: successful treatment with cyclosporin A. Clin Rheumatol 1996; 15: 189- 92. 32. Hagen EC, de Keizer RJ, Andrassy K, et al. Compassionate treat­ment of Wegener's granulomatosis with rabbit anti- thymocyte globulin. Clin Nephrol 1995; 43: 351- 9. 33. Murray PI. Wegener's granulomatosis, pituitary adenoma, and BARN. Int Ophthalmol 1995; 18: 361- 2. 34. Morgenstern LB, Pardo CA. Progressive multifocal leukencepha-lopathy complicating treatment for Wegener's granulomatosis. J Rheumatol 1995; 22: 1593- 5. 35. Sonnenblick M, Nesher R, Rozenman Y, Nesher G. Charles Bon­net syndrome in temporal arteritis, j Rheumatol 1995; 22: 1596- 7. 36. Fineman MS, Savino PJ, Federman JL, Eagle RC Jr. Branch retinal aretry occlusion as the initial sign of giant cell arteritis. Am J Ophthalmol 1996; 122: 428- 30. 37. Nassani S, Cocito L, Arcuri T, Favale E. Orbital pseudtumor as a J Neuro- Ophthalmol, Vol. 18, No. 1, 1998 ANNUAL REVIEW OF SYSTEMIC DISEASES 79 presenting sign of temporal arteritis. Clin Exp Rheumatol 1995; 13: 367- 9. 38. Ruiz- Masera JJ, Alamillos- Granados FJ, Dean- Ferrer A, et al. Sub­mandibular swelling as the first manifestation of giant cell arteritis. J Craniomaxillofac Surg 1995; 23: 119- 21. 39. Kassimos D, Kirwan JR, Kyle V, Hazleman B, Dieppe P. Cytidine deaminase may be a useful maker in differentiating elderly onset rheumatoid arthritis from polymyalgia rheumatica/ giant cell arteri­tis. Clin Exp Rheumatol 1995; 13: 641- 4. 40. Liozon F, Jauberteau- Marchan MO, Boutros- Toni F, et al. Ann Med Interne ( Paris) 1995; 146: 541- 7. 41. Chakravarty K, Pountain G, Merry P, Byron M, Hazleman B, Scott DG. A longitudinal study of anticardiolipin antibody in polymy­algia rheumatic and giant cell arteritis. J Rheumatol 1995; 22: 1694- 7. 42. Salvani C, Boiardi L, Macchioni P, et al. Role of peripheral CD8 lymphocytes and soluble IL2 receptor in predicting the duration of corticosteroid treatment in polymylagia rheumatic and giant cell arteritis. Ann Rheum Dis 1995; 54: 64CM1. 43. Emilie D, Liozon E, Crevon MC, et al. Production of interleukin 6 by granulomas of giant cell arteritis. Hum Immunol 1994; 39: 17- 24. 44. Roche NE, Fulbright JW, Wagner AD, Hunder GG, Grozny JJ, Weyand CM. Correlation of interleukin- 6 production and disease activity in polymyalgia rheumatica and giant cell arteritis. Arthritis Rheum 1996; 39: 1264- 7. 45. Weyand CM, Gorozny JJ. Giant cell arteritis as an antigen driven disease. Rheum Dis Clin North Am 1995; 21: 1027- 39. 46. Kraft HE, Moller DE, Volker L, Schmidt WA. Color Doppler ultrasound of the temporal arteries: a new method for diagnosing temporal arteritis. Klin Monatsbl Augenheilkd 1996; 208: 93- 5. 47. Dautzenberg Paul LJ, Leitjens Jan PAM. Reversible perfusion dis­order on brain SPECT after treatment with prednisone in temporal arteritis. Clin Nucl Med 1995; 20: 463- 4. 48. Rauser M, Rismondo V. Ischemic optic neuropathy during corti­costeroid therapy for giant cell arteritis. Arch Ophthalmol 1995; 113: 707- 8. 49. Hernandez- Garcia C, Soriano C, Morado C, et al. Methotrexate treatment in the management of giant cell arteritis. Scand J Rheum 1994; 23: 295- 8. 50. van der Veen MJ, Dinant HJ, Vam Booma- Frankfort C, Van Al-bada- Kuipers GA, Bijlsma JW. Can methotrexate be used as a steroid sparing agent in the treament of polymy algia rheumatic and giant cell arteritis? Ann Rheum Dis 1996; 55: 218- 23. 51. Kupersmith MJ, Langer R, Paget S, Mitnick H, Speira H. Visual Outcome in Patients with Giant Cell Arteritis After One Year of Therapy. Presented at the 1997 North American Neuro- ophthal-mology Society Annual Meeting, Keystone, Colorado. 52. Kumar G, Kumar A, Menon V. Ophthalmic manifestations of pulseless disease: a case report. Indian J Ophthalmol 1995; 43: 79- 81. 53. Raninen RO, Kupari MM, Pamilo MS, Pajari RI, Poutanen VP, Hekali PE. Arterial wall thickness measurements by B mode ul­trasonography in patients with Takayasu's arteritis. Ann Rheum Dis 1996; 55: 461- 5. 54. Ahmadieh H, Roodpeyma S, Azarmina M, Soheilian M, Sajjadi SH. Bilateral simultaneous optic neuritis in childhood systemic lupus erythematosus: a case report. J Neuroophthalmol 1994; 14: 84- 6. 55. Golstein M, Meyer O, Bourgeois P, et al. Neurological manifes­tations of systemic lupus erythematosus: role of antiphopholipid antibodies. Clin Exp Rheumatol 1993; 11: 373- 9. 56. Seaman DE, Londino AV Jr, Kowh CK, Medsger TA Jr, Manzi S. Antiphospholipid antibodies in pediatric systemic lupus erythema­tosus. Pediatrics 1995; 96: 1040- 5. 57. Denburg JA, Denburg SD, Carbotte RM, Sakic B, Szetchman H. Nervous system lupus: pathogenesis and rationale for therapy. Scand J Rheumatol 1995; 24: 263- 73. 58. Coller BS, Owen J, Jesty J, et al. Deficiency of plasma protein S, protein C, or antithrombin III and arterial thrombosis. Arterioscle­rosis 1987; 7: 456- 62. 59. Meade TW, Cooper J, Miller GJ, et al. Antithrombin III and arte­rial disease. Lancet 1991; 338: 850- 1. 60. Eastern JD, Mills JL, Beckett WC. Hypercoagulable states in arte­rial thromboembolism. Surg Gynecol Obstet 1992; 174: 211- 5. 61. Levy JP, Gonzalez FM, Rush DS, Haynes JL. Hypercoagulable states as an evolving risk for spontaneous venous and arterial thrombosis. J Am Coll Surg 1994; 178: 266- 70. 62. Acheson JF, Sanders MD. Coagulation abnormalities in ischemic optic neuropathy. Eye 1994; 8: 89- 92. 63. Mannucci P, Tripodi A. Laboratory screening of inherited throm­botic syndromes. Thromb Haemost 1987; 57: 247. 64. Bick RL, Pegram M. Syndromes of hypercoagulability and throm­bosis: a review. Semin Thromb Hemost 1994; 20: 109- 32. 65. Daif A, Awada A, Al- Rajeh S, et al. Cerebral venous thrombosis in adults: a study of 40 cases from Saudi Arabia. Stroke 1995; 26: 1193- 5. 66. Dahlbeck B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: prediction of a co-factor to activated protein C. Proc Natl Acad Sci USA 1993; 90: 1004- 8. 67. Bertina RM, Koelman BPC, Koster T. Mutation in blood coagu­lation Factor V associated with resistance to activated protein C. Nature 1994; 369: 64- 7. 68. Mandel H, Brenner B, Berant M, et al. Coexistence of herditary homocystinuria and Factor V Leiden: effect on thrombosis. N Engl J Med 1996; 334: 763- 8. 69. Florell SR, Rodgers GM. Utilization of testing for activated protein C resistance in a reference laboratory. Am J Clin Pathol 1996; 106: 248- 52. 70. Koepke JA. Testing for hereditary hypercoagulability, activated protein C resistance. Am J Clin Pathol 1996; 106: 161- 2. 71. Asherson RA, Khamashta MA, Gil A, et al. Cerebrovascular dis­ease and antiphospholipid antibodies in SLE, lupus- like disease and the primary antiphospholipid syndrome. Am J Med 1989; 86: 391- 9. 72. Sibley JL, Wojciech P, Olszynski WSP, et al. The incidence and prognosis of central nervous system disease in SLE. J Rheumatol 1992; 19: 47- 52. 73. Toubi E, Khamashta MA, Panarra A, et al. Association of an­tiphospholipid antibodies with central nervous system disease in systemic lupus erythematosis. Am J Med 1995; 99: 397- 401. 74. Harris EN, Exner T, Hughes GRV, Asherson RA. Phospholipid binding antibodies. Boca Raton, FL: CRC press, 1991. 75. Levine SR, Brey RL, Sawaya KL, et al. Recurrent stroke and thrombo- occlusive events in the antiphospholipid syndrome. Ann Neurol 1995; 38: 119- 24. 76. Briley DP, Coull BM, Goodnight SH. Neurological disease asso­ciated with antiphospholipid antibodies. Arch Neurol 1989; 25: 221- 7. 77. Levine SR, Joseph R, D'Andrea G, Welch KMA. Migraine and the lupus anticoagulant. Cephalalgia 1987; 7: 93- 9. 78. Hogan MJ, Brunet DG, Ford DM, Lillicrap D. Lupus anticoagu­lant, antiphospholipid antibodies and migraine. Can J Neurolol Sci 1988; 15: 420- 5. 79. Shuaib A, Barklay L, Lee MA, Suchowersky O. Migraine and antiphospholipid antibodies. Headache 1989; 29: 42- 5. 80. Silvestrini M, Matteis M, Troisi E, et al. Migrainous stroke and the antiphospholipid antibodies. Eur Neurol 1994; 34: 316- 9. 81. Manzanares JM, Conget I, Rodriguez- Villar C, et al. Antiphospho­lipid syndrome in a patient with type I diabetes presenting as retinal artery occlusion. Diabetes Care 1996; 19: 92- 3. 82. Anzai K, Nakamura M, Nagafuchi S, et al. Production of anticar­diolipin antibody in AKR/ J mice with streptozotocin- induced in-sulitis and diabetes. Diabetes Res Clin Pract 1993; 20: 29- 37. 83. Anzai K, Nagafuchi S, Okubo K, Ono J. Appearance of autoanti­bodies to the complex of beta 2 glycoprotein I and cardiolipin in NOD mice ( Abstr). Diabetologia 1994; 37( suppl 1): A94. 84. Orefice G, DeJoanna G. Coppola M, et al. Benign intracranial hypertension: a non- thrombotic complication of the primary an­tiphospholipid syndrome. Lupus 1995; 4: 324- 6. 85. Demers C, Henderson P, Blajchman MA, et al. An antithrombin III assay based on factor Xa inhibition provides a more reliable test to identify congenital antithrombin III deficiency than an assay based on thrombin inhibition. Thromb Haemost 1993; 69: 231- 5. 86. Espana F, Gilabert J, Vicente V, et al. Activated protein C: alpha- 1- antitrypsin ( APCalAT) complex as a marker for in vitro diag­nosis of prethrombotic states. Thromb Res 1992; 66: 499- 508. / Neuro- Ophthalmol, Vol. 18. No. I, 1998