Intermediate uveitis: pattern of etiology, complications, treatment and outcome in a tertiary academic center

Background Patients with intermediate uveitis (IU) represent a heterogenous group characterized by a wide spectrum of etiologies and regional differences. Aim of the study was to analyze the characteristics of patients with IU examined in an academic center in Germany. Methods We conducted a retrospective analysis of the clinical records of all patients with intermediate uveitis referred to the Eye Center, University of Freiburg from 2007 to 2014. Diagnosis followed the Standardization in Uveitis Nomenclature (SUN) criteria. Data analysis included: etiology of IU, demographics, complications, treatment and visual acuity. Results We identified 159 patients with intermediate uveitis during that period. Mean age at diagnosis was 35 years. Most are female (64%), and the mean duration of IU was 6.1 years (range 1 month – 35 years). Etiology of IU was idiopathic in 59%. Multiple sclerosis (MS) (20%) and sarcoidosis (10%) were frequent systemic causes of IU. Other etiologies including infectious diseases (tuberculosis, borreliosis) or immune-mediated conditions (eg, after vaccination) were present in 11%. The pattern of complications included macular edema (CME) (36%), cataract (24%), secondary glaucoma (7%), and epiretinal membrane formation (19%). Periphlebitis and optic neuritis were more frequent in conjunction with MS. Treatment comprised local and systemic steroids, immunosuppressive agents, biologics, and surgery. Best corrected visual acuity was better than 20/25 in 60% of the eyes after more than 10 years of follow-up. Conclusions In our German academic center, most IU cases were idiopathic or associated with MS or sarcoidosis. In contrast to other countries, infectious cases were rare. Patients’ overall visual prognosis is favorable even when the duration of IU has been long and and despite numerous complications.

We conducted this study to analyze the demographic and clinical data of patients with IU in our academic center in Central Europe.

Methods
This was a retrospective study including all patients with intermediate uveitis examined at the Eye Center, University of Freiburg between 2007 and 2014. Intermediate uveitis was classified according to recommendations by the SUN working group [3].
Our study received institutional review board approval (EK Freiburg 19/15). Patient consent was not required as this was a retrospective, pseudoanonymous chart review. Patients diagnosed with any disorder other than intermediate uveitis were excluded. All patients were examined in a specialized uveitis center and treated in a multidisciplinary setting. If necessary, the appropriate specialists were consulted to determine any suspected underlying systemic or infectious disease.
In the case of sarcoidosis we collected chest radiographs, computer tomographies, bronchoalveolar lavage results, biopsies and laboratory data, if available. Diagnosis of infectious IU was based on serological testing and systemic manifestations, if applicable.
Data analysis included: etiology of IU, demographics, complications, treatment modalities, visual acuity and final outcome. Continous factors are presented as mean, standard deviation, standard error of the mean and confidence interval. Categorial data are presented as percentages. We used chi-sqare statistics for hypothesis testing. Change in visual acuity is presented as Box-and Whisker Plot.
All calculations were performed with the R-platform using only core functionality [19].
The age at diagnosis varied with the underlying origin of IU. Patients with idiopathic IU were the youngest (mean 32.9 years (SD 17.7; SEM 1.8; CI ± 3.  interferon alpha) (Fig. 3). The main indications for initiating therapy are summarised in Table 2. Some patients got more than one therapy. Usually we started treatment with oral, parabulbar or intravitreal steroids. If there was no stable remission with less than 7.5 mg prednisolon equivalent, an immunosuppressive or biologic agent was added.
A total of 62% of the IU patients developed at least one complication. Cystoid macular edema was the most frequent complication (36.5%). Nearly a quarter suffered from cataract (23.9%), 19% from epiretinal membrane, 5% from retinal detachment, and 7% from glaucoma (Fig. 4). Periphlebitis and optic neuritis were significantly related to MS-associated IU (p < 0.001 Chi Square Test).
The overall prognosis was favorable. As Fig. 5 illustrates, visual acuity was stable over time in most patients. At the end of follow-up, 75% of the eyes had a best corrected visual acuity better than 20/25 ( Table 1). As shown in Fig. 6, the percentage of eyes with visual acuity of 20/25 or better was slightly decreasing with follow-up. After a follow up of at least 10 years more than 60% fulfilled this criterium.

Discussion
Our study demonstrates that IU in Central European patients is mostly non-infectious and idiopathic, requiring therapy in 80% of cases, and that it has an overall favorable prognosis. However, many patients experience at least one of many complications (eg. cataract, glaucoma, CME, epiretinal membrane). Many of these patients fulfilled the criteria for the older term pars planitis, which is restricted by SUN for "that subset of intermediate uveitis associated with snowbank or snowball formation in the absence of an associated infection or systemic disease" [3].
Like in our cohort, most other researchers have noted that IU usually affects young adults. The mean age at diagnosis varies between 22.6 and 33 years of age [14,16,[21][22][23]. In contrast to other studies, we differentiated age by etiology. We observed a marked difference in age at diagnosis depending on the underlying disease. The youngest patients suffered from idiopathic IU, the oldest from infectious IU. In addition, we detected in conjunction with infectious IU a biphasic age distribution, with one peak in children and a second one in the fifth decade.
In Europe, the US and China, IU is usually idiopathic [1,9,11,13,14,16]. In contrast, in other parts of the world such as India, there is a marked proportion of infectious IU rising to 58% [24]. In these countries, tuberculosis is a very common comorbidity; as the cause of IU in Europe and US it is rare [17]. On the other hand, MS is a frequent underlying disease in IU in our patients and in the US [14]. The proportion of MS in IU patients varies from 7 to 30.4% [1,21,[25][26][27][28][29][30][31]. In our cohort, MS was very significantly associated with periphlebitis, a particular indication of IU. Others have observed the same [21,32]. Since IU might be the first manifestation of MS and early treatment seems to improve the overall prognosis, it is important to screen all IU patients for MS [33][34][35]. About 10% of our patients have sarcoidosis, a diagnosis that is also frequent elsewhere in the world [1,9,11,16,31,36]. We have found that soluble interleukin 2 receptor is a useful screening parameter [37].
Immunization as a cause of IU is currently under discussion. As described by several groups, FSME immunization may trigger IU in some patients [38][39][40][41][42]. JIA is associated with anterior uveitis. Our three cases suffered from IU and the pediatricians found no other underlying disease than JIA.
Many IU patients suffer from complications. The development of cataract, glaucoma, CME, epiretinal membrane formation, retinal detachment, periphlebitis or optic neuritis is similar worldwide [4,9,10,14,30,43]. Cataract and glaucoma might be caused by IU itself or by treatment of IU, especially with corticosteroids. There is ample evidence that CME and epiretinal membrane formation correlate with poor visual prognosis [4,9,16].
As in the smaller study by Donaldson et al, nearly 2/3 of our patients required therapy [14]. Main treatment indications in our series were CME or severe vitreous inflammation. Systemic, intraocular and parabulbar corticosteroids are the predominant therapeutic options. Only a quarter of our patients received immunosuppressive agentsmore frequently than in China and the US [8,14,16]. Regarding our cohort's MS patients: about 10% received biologics, mainly interferon.
Despite the many complications, IU's overall prognosis is encouraging. Most patients have retained best corrected visual acuity of 20/40 or better [4,14,16,43]. The decrease of visual acuity during follow-up in our study might be biased by the fact, that patients with no complications and good visual acuity were lost for follow up. In our specialized center, those with complications were followed for a longer time.
Our study is limited by its retrospective character. Nevertheless, we report on a large number of patients and have delivered useful data for daily clinical practice.

Conclusions
In our German academic center, most IU cases were idiopathic or associated with MS or sarcoidosis. In contrast to other countries, infectious cases were rare. Patients' overall visual prognosis is favorable even when the duration of IU has been long and and despite numerous complications.