Multiple sclerosis associated with uveitis in two large clinic-based series

Methods.

Medical records of consecutive patients seen between 1987 and 1993 in an MS clinic (n = 1,098) and in a uveitis clinic (n = 1,530) were reviewed by a group of neurologists and ophthalmologists to select patients with MS and uveitis. These two clinics are in two different large general hospitals in Paris, and have a wide catchment area, with patients seen in both primary and referral care. We used the databases from both clinics to select patients with uveitis from the MS clinic and to select patients with neurologic symptoms from the uveitis clinic. The charts of selected patients were reviewed carefully by both a neurologist and an ophthalmologist. Inclusion criteria were uveitis and definite MS as defined by strict criteria to exclude other inflammatory diseases.9 Patients were classified according to disease severity (assessed using Kurtzke’s Expanded Disability Status Scale [EDSS]) and clinical course of the neurologic disease. Age at onset of MS and duration of follow-up were recorded for each patient, allowing determination of a Kurtzke index. Best-corrected visual acuity testing, type of uveitis, and presence of pars planitis and periphlebitis (venous sheathing) on dilated funduscopy were recorded. Uveitis was classified anatomically according to the International Uveitis Study Group criteria10 as anterior, posterior, intermediate (pars planitis), or panuveitis, and by temporal profile (acute or chronic). These patients were contacted by phone to schedule a follow-up visit in the MS or uveitis clinic, and to obtain information regarding the severity and temporal profile of MS and uveitis and family histories of MS, uveitis, and autoimmune diseases. When a follow-up visit was not possible, we performed a telephone interview of patients and family members. These were structured interviews designed by both an ophthalmologist and a neurologist.

Results.

We identified 28 patients (20 women and 8 men; mean age, 47 years; range, 28 to 67 years) with definite MS and uveitis from a total of 2,628 patients (1%): 12 from the 1,098 patients followed in the MS clinic (1.1%) and 16 from the 1,530 patients followed in the uveitis clinic (1%). Detailed clinical characteristics of these patients are presented in the table.

Mean age at onset of neurologic symptoms was 34.2 years (range, 15 to 55 years). MS was relapsing-remitting in 19 of 28 patients (67.8%), secondary progressive in 8 of 28 patients (28.6%), and primary progressive in one patient. The mean duration was 12 years (range, 1 to 30 years). The mean EDSS score was 3.4 (range, 0 to 6.5), with 68% of patients having a score of 5 or less, and 36% having a score of 2 or less. The mean Kurtzke index was 0.49 (range, 0 to 1.5).

Mean age at onset of uveitis was 33.7 years (range, 7 to 57 years). Uveitis was remitting in 22 of 28 patients (78.5%) and chronic in 6 of 28 patients (21.5%). It was bilateral in most patients (22 of 28; 78.5%). The range of visual acuity of affected eyes was 20/20 to 20/100 after a mean follow-up of 13.4 years (range, 1 to 38 years), with 29 of 50 affected eyes better than 20/50. We were not able to obtain accurate visual acuity in two patients who were lost to follow-up. The most common types of uveitis were pars planitis (in 10 of 28 patients; 35.7%) and panuveitis (in 11 of 28 patients; 39.3%). Isolated anterior uveitis was observed in 4 of 28 patients (14.3%) and posterior uveitis was observed in 3 of 28 patients (10.7%). Retinal periphlebitis (venous sheathing) was noted in 11 of 28 patients (39.3%).

In 16 of 28 patients (57.1%), neurologic symptoms occurred from 1 to 31 years (mean, 9.1 years) before onset of uveitis, whereas in 11 patients (39.3%), neurologic symptoms occurred from 1 to 23 years (mean, 9.8 years) after the onset of uveitis. In one patient both neurologic and ocular symptoms developed at the same time. There was no correlation between the type of MS and uveitis, presence of periphlebitis, or degree of neurologic or ocular disability.

Detailed family histories were obtained for 19 patients, four of whom had at least one first-degree relative with MS, uveitis, or other autoimmune diseases (see the table). These four patients had pars planitis. Therefore, a family history of MS or autoimmune diseases or both was found in one-third of our patients with pars planitis.

There were no differences between the two groups of patients selected from the MS and uveitis clinics.

Discussion.

The association of MS and uveitis is rare, occurring in only approximately 1% of 1,098 patients from an MS clinic and 1,530 patients from a uveitis clinic. Although clinic-based series are sometimes biased toward more severe cases, there is no evidence that neurologic or visual problems were a significant bias in our two series. The frequency of uveitis among patients with MS is approximately 10 times that of the general population.1 However, its exact frequency in previous series varies widely—from 0.4 to 26.9%.1-7 These differences may be explained in part by the variations in patient populations studied (recruitment in neurology or ophthalmology departments), and by the diverse inclusion criteria and classifications used for both MS and uveitis. Indeed, a majority of studies were performed before the development of MRI and autoantibody studies, and various neurologic inflammatory diseases more commonly associated with uveitis might have been misdiagnosed as MS.1,6,8 To exclude patients with other diseases in our study, all selected patients were examined by both a neurologist and an ophthalmologist considered experts in inflammatory diseases of the brain and the eye. Moreover, we used strict criteria for the diagnosis of MS, and included only patients with definite MS.9 This might explain the relatively low frequency of the association of MS and uveitis found in our study. Furthermore, the delay, often in years (mean, 9 years in our study), between the onset of ocular and neurologic symptoms emphasizes the importance of performing not just one diagnostic search, but sequential ones throughout the patient’s course. Symptoms such as eye pain, redness of the eye, and unilateral visual loss in the absence of a relative afferent pupillary defect should alert the neurologist to the possibility of uveitis, and a prompt slit-lamp examination as well as dilated funduscopy.

It is generally believed that in patients with MS, uveitis has a broad clinical spectrum.1-8 In our study most patients had pars planitis or panuveitis (see the table). Pars planitis is characterized by intravitreal inflammation, exudates, and snowbank formation along the pars plana, and its diagnosis may be difficult.1,7,8 Because this study was retrospective, we cannot be certain that ophthalmoscopy was performed with scleral depression of the peripheral retina and pars plana in all patients. Therefore, it is possible that we may have underestimated the frequency of pars planitis. Indeed, pars planitis seems to be the most common type of uveitis occurring in patients with MS.1,6-8 Conversely, Malinowski et al.7 found that among 54 patients with pars planitis, eight (14.8%) developed MS over a mean follow-up period of 89 months, and four additional patients developed optic neuritis. Kaplan-Meier analysis of these data showed a 16.2 ± 6.2% risk of patients with pars planitis developing MS, and a 20.4 ± 6.7% risk of patients with pars planitis developing either MS or optic neuritis within 5 years of the retinal disease, with the finding of retinal periphlebitis increasing the rate of development of these conditions.7 Periphlebitis (venous sheathing) was found in 39.3% of our patients. These findings have been reported in 6 to 26% of patients with MS1 and, although this was not the case in our patients, they sometimes correlate with the severity of neurologic dysfunction in patients with MS.1,2

The cause of the association of MS and uveitis is unclear because the etiology of both diseases remains unknown. Although it is possible that patients with both ocular and neurologic manifestations represent a disease similar to MS but etiologically different from other forms of MS in which the uveal tract remains uninvolved, the neurologic and radiologic presentations of these patients do not differ from those with MS without uveitis. Malinowski et al.7 suggested that there might be a similar genetic predisposition to the development of MS and uveitis and noted that 11.1% of their 54 patients with pars planitis had a positive family history for MS in a first-degree relative.1 Two of our patients with pars planitis had a family history of MS, suggesting that this particular type of uveitis might be linked specifically to the development of MS.

The association of MS and uveitis is rare. Pars planitis and panuveitis are most commonly encountered. The delay between the onset of neurologic and ocular symptoms emphasizes the importance of a sequential diagnostic search throughout the patient’s course. The cause of this association is unclear, but genetic studies of both diseases might show a common genetic predisposing factor.