Uveitis as a prognostic factor in multiple sclerosis

Introduction

Uveitis has been reported in several multiple sclerosis (MS) cohorts with highly varying incidences often in the form of intermediate uveitis (IU) that afflicts the vitreous and peripheral retina.^{1 –3} Likewise, 1%–14% of uveitis patients followed in ophthalmology outpatient clinics have received the diagnosis of MS.^4,5 Previous studies have failed to identify a major difference in the clinical features of MS patients with uveitis.^3,6 To investigate whether uveitis has a prognostic impact on the clinical course of MS, we compared the clinical and demographic features of MS patients with and without uveitis.

Patients and methods

The case records of 6024 MS patients fulfilling the revised McDonald’s criteria ⁷ and followed between 1993 and 2013 were reviewed and patients with a history of uveitis were included. Uveitis was diagnosed in uveitis clinics and anatomically classified according to the International Uveitis Study Group criteria as anterior (iridocyclitis), posterior (chorioretinitis), intermediate uveitis (pars planitis) or panuveitis. ⁸ Systemic vasculitis, chronic bacterial or viral infections and sarcoidosis were excluded with relevant serological tests, chest X-ray and serum angiotensin-converting enzyme levels. None of the patients had the clinical (e.g. recurring oral aphthae, genital ulcers) and neuroimaging features of Behçet’s disease. Demographic, neuroimaging and clinical features were assessed in each patient.

For statistical analyses, we randomly recruited 100 MS patients without uveitis, using the simple random sampling method. All 6024 MS patients were sorted according to their last names and every 60th name in the patient list was recruited. Since, in our outpatient clinic, all MS patients with visual complaints are examined by an ophthalmologist on a routine basis, the control MS patients were known not to have clinically manifest uveitis. The requirement for informed consent was waived because of the nature of this retrospective chart review study. Gender, age at MS onset, type and duration of MS, presence of cerebrospinal fluid (CSF) oligoclonal bands (OCB), treatment types, Expanded Disability Status Scale (EDSS) and progression index (PI) scores were compared between MS patients with and without uveitis using Fisher’s exact test, Student’s t-test and Mann Whitney U test. Moreover, direct logistic regression was performed to assess the impact of presence of uveitis, gender, age at MS onset and MS duration on the likelihood that patients fulfill the diagnostic criteria of benign MS (i.e. EDSS of less than 4.0 in 15 years of disease duration). P < 0.05 was considered as significant.

Results

Comparison of MS patients with and without uveitis

Duration of MS, OCB positivity and treatment types were identical among MS patients with and without uveitis. Although MS patients with uveitis showed trends toward exhibiting higher female to male ratio and age at MS onset, these differences did not reach statistical significance. Notably, none of the uveitis patients had developed the progressive forms of MS during their follow-up, whereas 36% of MS patients without uveitis had primary (n = 6) or secondary (n = 30) progressive MS (p < 0.001). Moreover, final EDSS and PI scores of MS patients with uveitis were significantly lower than those without uveitis (p = 0.004 and <0.001, respectively) (Table 1). Only two of 41 patients in the uveitis group had a final EDSS score of ≥4.0, as opposed to 36/100 patients in the control group (p < 0.001 by Fisher’s exact test). Also, 13/41 MS patients with uveitis fulfilled the criteria for benign MS as opposed to 15/100 control MS patients (p = 0.035 by Fisher’s exact test). In regression analysis, presence of uveitis and duration of MS were the only two independent variables that made a unique statistically significant contribution to the model. The strongest predictor of benign MS was the presence of uveitis. Patients with uveitis were more than 200 times more likely to have benign MS than controls, controlling for all other factors in the model (Table 2).

OPEN IN VIEWER

Table 1.

Comparison of clinical and demographic features of MS patients with and without uveitis.

	MS patients with uveitis (n = 41)	MS patients without uveitis (n = 100)	p value
Gender (number of women)	35/41 (85%)	72/100 (72%)	0.12 c
Mean age at MS onset (SD, range)	31.1 (10.9, 13–52)	27.9 (9.1, 17–51)	0.09 d
Mean MS duration (years; SD, range)	10.6 (7.1, 2–30)	10.3 (8.2, 4–33)	0.26 d
Oligoclonal band positivity a	33/41 (80%)	76/100 (76%)	0.66 c
RRMS	41/41 (100%)	64/100 (64%)	<0.001 c
SPMS or PPMS	0/41 (0%)	36/100 (36%)
Azathioprine	21/41 (51%)	36/100 (36%)	0.13 c
IFN or GA	27/41 (66%)	59/100 (59%)	0.57 c
NAT	3/41 (7%)	11/100 (11%)	0.76 c
Mitoxantrone	2/41 (5%)	12/100 (12%)	0.35 c
Median final EDSS score (IQR, range)	2.0 (1.7, 0.0–4.5)	3.0 (4.5, 1.0–9.5)	0.004 e
Median progression index (IQR, range) b	0.2 (0.2, 0.0–1.8)	0.4 (0.4, 0.1–2.7)	<0.001 e

MS: multiple sclerosis; EDSS: Expanded Disability Status Scale; SD: standard deviation; RRMS: relapsing–remitting MS; SPMS: secondary progressive MS; PPMS: primary progressive MS; IFN: interferon; GA: glatiramer acetate; NAT: natalizumab; IQR: interquartile range (difference between the upper and lower quartiles). ^aCerebrospinal fluid (CSF) oligoclonal bands (pattern 2 or 3), ^bProgression index: EDSS/disease duration (years). ^cFisher’s exact test, ^dStudent’s t-test, ^eMann-Whitney U test.

OPEN IN VIEWER

Table 2.

Logistic regression analysis performed to determine the contribution of uveitis on benign MS occurrence.

	OR	95% CI	p value
Presence of uveitis	272.6	14.6–5096.8	<0.001
Gender	1.6	0.3–9.3	0.62
Age at MS onset	1.0	0.9–1.1	0.54
Duration of MS	1.2	1.1–1.3	0.01

MS: multiple sclerosis; OR: odds ratio; CI: confidence interval.

No significant difference could be found between clinical and demographic features of MS patients developing uveitis before MS (n = 15) and MS patients developing uveitis after or simultaneously with the onset of MS (n = 26) (data not shown). Likewise, there were no differences between the features of MS patients with IU (n = 25) and other forms of uveitis (n = 16) (data not shown).

Discussion

The findings of our study were comparable with those of previously published MS cohorts with uveitis. In particular, frequency of uveitis in MS, age at uveitis onset, time between MS and uveitis and anatomical distribution of uveitis were all in agreement with most of the previous reports.^{1 –3} As has been repeatedly recognized in previous studies,^1,2,9 IU was the dominant uveitis type in our MS cohort. Also similar to previous reports,^3,6 the type of uveitis (IU vs non-IU) and timing of uveitis with respect to MS onset (before vs after the first MS attack) did not affect neurological disability.

However, our study did show that MS patients with uveitis display significantly lower neurological disability scores. The single study focused on the influence of uveitis on neurological disability has failed to find a similar difference. ³ In our study, we used newer diagnostic criteria for MS and all MS patients with visual problems were routinely examined by an ophthalmologist especially for detection of IU, which may be difficult to diagnose. These differences in patient assessment may have accounted for a recruitment bias, thereby increasing MS patients with uveitis and/or good prognosis. Because of the retrospective nature of our study, patients with subclinical uveitis and no evident visual loss and those who have not reported their visual complaints because of the mildness of their symptoms may have been included in the control group. Prospective studies performing a thorough ophthalmologic examination in all recruited patients might provide a better evaluation of the impact of uveitis on disability.

The association between reduced disability and uveitis in MS patients might be explained by epidemiological factors or genetic factors that influence susceptibility to developing uveitis together with MS. Unraveling the exact immunopathogenic mechanisms underlying this association might aid in the improvisation of novel treatment methods for MS.