Customize your JAMA Network experience by selecting one or more topics from the list below.
Han YS, Rivera-Grana E, Salek S, Rosenbaum JT. Distinguishing Uveitis Secondary to Sarcoidosis From Idiopathic Disease: Cardiac Implications. JAMA Ophthalmol. 2018;136(2):109–115. doi:10.1001/jamaophthalmol.2017.5466
What are the implications in distinguishing uveitis secondary to sarcoidosis from idiopathic disease?
In this case series that included 249 consecutive patients with uveitis, the referring diagnosis was idiopathic for 179 (71.9%). Nineteen patients referred for idiopathic uveitis subsequently received a diagnosis of presumed sarcoidosis, usually on the basis of findings on chest computed tomography, and 4 of these 19 were found to have episodes of ventricular tachycardia requiring implantable cardiac defibrillators.
The present case series shows the potential utility of distinguishing sarcoidosis-associated uveitis from idiopathic uveitis.
Idiopathic disease is the most frequent diagnosis in a uveitis clinic. The need to distinguish sarcoidosis from idiopathic uveitis is controversial. However, cardiac involvement in sarcoidosis can be life-threatening.
To report a series of patients with uveitis and cardiac sarcoidosis to illustrate the importance of categorizing the causes of uveitis.
Design, Setting, and Participants
This retrospective observational case series reviewed the medical records of 249 patients with uveitis who were referred to the Casey Eye Institute between July 1, 2008, and February 28, 2017.
Main Outcomes and Measures
We describe patients who initially received a diagnosis of idiopathic uveitis but subsequently received a diagnosis of sarcoidosis. Clinical data, including ophthalmologic findings, were collected. We summarized the number of patients who initially presented with idiopathic uveitis, the number of patients who recived a classification of idiopathic uveitis after evaluation, the number of patients who underwent chest computed tomography or an electrocardiogram, and the number of patients with ocular sarcoidosis.
Of 33 patients with sarcoidosis, 21 (63.6%) were women and the mean (SD) age was 53.5 (13.8) years. Of 249 patients, the referring diagnosis was idiopathic uveitis for 179 (72%). After history, examination, and laboratory testing, 127 (51%) were still considered to have idiopathic disease. Fifty-three of the 179 patients (30%) with idiopathic disease underwent chest computed tomography scanning. A diagnosis of presumed sarcoidosis, usually on the basis of a chest computed tomography scan, was made in 19 patients (36.2%). As 14 patients (5.6%) were previously known to have sarcoidosis, 33 patients (13.3%) were evaluated with definite or presumed ocular sarcoidosis. We obtained electrocardiograms as a screen for cardiac sarcoidosis on 14 (42.4%) of these patients. Nine patients with abnormal electrocardiogram results were referred to cardiologists. Four of the 19 patients (21.1%) who were referred for idiopathic uveitis but subsequently received a diagnosis of presumed sarcoidosis were found to have episodes of ventricular tachycardia that required implantable cardiac defibrillators. Distinguishing ocular sarcoidosis from idiopathic uveitis had potentially life-saving implications for these patients.
Conclusions and Relevance
The present case series shows the potential utility of distinguishing sarcoidosis-associated uveitis from idiopathic uveitis. We suggest that patients older than 40 years with a history of idiopathic uveitis be evaluated with chest computed tomography and an electrocardiogram if sarcoidosis is suggested on ophthalmic examination.
Uveitis is a major cause of visual loss and accounts for 10% of acquired blindness.1,2 Although understanding the pathogenesis for uveitis is evolving, about 24% to 55% of patients with uveitis have been commonly labeled as having idiopathic uveitis or undifferentiated uveitis. The prevalence of idiopathic uveitis is similar in studies from the United States and from other countries.3-15 Therefore, idiopathic uveitis is the most common diagnosis in most US and European clinics for uveitis.3 The frequency of the diagnosis of idiopathic uveitis was recently confirmed in 3 prominent, multicenter double-masked randomized clinical trials.16-18 Some examples of idiopathic uveitis may result from incomplete history taking or from incomplete laboratory testing. Some patients who are categorized as having idiopathic uveitis may actually have sarcoidosis,19 tubulointerstitial nephritis and uveitis,20 or ankylosing spondylitis.21,22
Sarcoidosis is a chronic, granulomatous multisystem disease of unclear cause that affects predominantly the lungs and other organs, including the skin, nervous system, eyes, and heart.23,24 The annual incidence of sarcoidosis varies worldwide.24-26 The annual US age-adjusted incidence, in the number of new cases per 100 000, was highest among African American females (39.1 cases), followed by African American males (29.8 cases), white females (12.1 cases), and white males (9.6 cases).27 While sarcoidosis usually occurs in individuals between the second and fifth decades of life, there is a bimodal distribution and a second peak occurs in women older than 50 years in Europe and Japan.24,28
Ocular sarcoidosis occurs in about 25% to 50% of patients with systemic sarcoidosis.29-31 Ocular sarcoidosis may manifest as anterior uveitis that mainly affects younger patients or as panuveitis that is often seen in the middle and older ages.30 As the disease progresses, ocular sarcoidosis can involve all structures of the eye, such as the uvea, cornea, sclera, optic nerve, orbit, and visual pathway. Many argue that there is no need to go “hunting” for sarcoid because making a diagnosis of ocular sarcoidosis does not change the treatment compared with the treatment for idiopathic uveitis.
Cardiac involvement in sarcoidosis is rare, affecting approximately 6% of all patients.29,32-34 Cardiac involvement, however, can be life-threatening. We reported 4 patients with sarcoid-associated uveitis and ventricular tachycardia presumably secondary to sarcoidosis. In each case, establishing the cause of the uveitis led to a potentially life-saving intervention.
This retrospective review was approved by the institutional review board of Oregon Health & Science University and included patients who were referred with a diagnosis of uveitis to the Casey Eye Institute between July 1, 2008, and February 28, 2017. Informed consent was not obtained because this was a retrospective review of medical records.
All patients were referred with a diagnosis of uveitis. The patients had undergone a variable amount of laboratory investigation before being referred. All patients who were referred to the uveitis clinic underwent an ophthalmic examination and provided a detailed medical history. In obtaining the history, we searched for clues, such as bowel symptoms or chronic low back pain, that might point to a specific cause. For all patients who remain in the idiopathic category after history and examination, we obtained a serologic test for syphilis and chest radiography results. Assuming that both these tests yielded negative or normal results, we recommended a chest computed tomography (CT) for patients 40 years or older. We were hesitant to recommend a chest CT for younger patients because of the amount of radiation. We accepted symmetric hilar or mediastinal adenopathy as diagnostic of presumed sarcoidosis in patients who also had uveitis based on the First International Workshop of Ocular Sarcoidosis Guidelines.30
A diagnosis of presumed ocular sarcoidosis on the basis of a chest CT examination showing symmetric hilar adenopathy and the presence of uveitis was first described by Winterbauer et al.35 In the report by Kaiser et al19 on the use of chest CT in evaluating patients with sarcoidosis, all 14 patients with uveitis and symmetric adenopathy who underwent mediastinal biopsy had noncaseating granulomas that were detected histologically.
SPSS, version 20.0 (IBM) was used for data analysis. The association between multiple peripheral chorioretinal atrophic lesions (MPCALs) and severe cardiac sarcoidosis necessitating implantable cardiac defibrillator (ICD) was analyzed using odds ratios. A P value of less than .05 was considered statistically significant.
Over 8 years and 8 months of medical record review, 249 patients were referred for a diagnosis of uveitis. One hundred seventy-nine of these patients (72%) were considered to have idiopathic uveitis at the time of referral. Fifty-three (30%) underwent chest CT scans, and 17 patients (34%) had scans that were found to have adenopathy consistent with sarcoidosis. One additional patient who previously received a diagnosis of idiopathic uveitis was presumed to have ocular sarcoidosis on the basis of hilar adenopathy that was identified by chest radiography. Another patient was presumed to have sarcoidosis, as this was the best unifying diagnosis for the combination of uveitis, orbital mass, and optic neuropathy. Most patients were women, and the demographics of these patients are presented in Table 1. Patients who did not have a CT scan were younger than age 40 years or had a phenotype of uveitis, such as sudden onset, unilateral, or anterior, that was not likely to be associated with sarcoidosis. In addition, CT scans were recommended for some patients who declined because of concerns about cost, radiation, or inconvenience. Because 14 of the original 249 patients had a prior diagnosis of sarcoidosis, the series consisted of 33 patients with presumed ocular sarcoidosis. After evaluation in the uveitis clinic, the number of patients with idiopathic uveitis went from 179 to 127.
Of the 33 patients with definite or presumed ocular sarcoidosis, 14 (42.4%) underwent electrocardiograms (EKGs). The clinic began to order EKGs routinely for patients with ocular sarcoidosis only in the last 3 to 4 years. Furthermore, many patients with known sarcoidosis before referral had recently received EKG results. Nine EKGs yielded abnormal results such that patients were referred to a cardiologist. Cardiac sarcoidosis with ventricular tachycardia was diagnosed in 4 (44.4%), each of whom had a chest CT scan that showed a symmetric hilar adenopathy and each of whom had initially received a diagnosis of idiopathic disease. Thus 4 of 33 patients (12%) with ocular sarcoidosis and 4 of 19 patients (21%) with presumed ocular sarcoidosis and an initial diagnosis of idiopathic uveitis had cardiac sarcoidosis.
The age of these 4 patients at presentation in the uveitis clinic ranged from 66 to 67 years (mean [SD] 66.8 [0.3] years), all 4 patients were women, and they were all white. Table 2 summarizes clinical information on the participants in this report. Multiple peripheral chorioretinal atrophic lesions were observed in 3 of the 4 patients (75%).
The time between the onset of ocular symptoms and the first visit to the Casey Eye Institute with a presumptive diagnosis of idiopathic uveitis varied, ranging from 9 weeks to more than 10 years. Three patients received a diagnosis of sarcoidosis-associated uveitis within 4 weeks of referral, whereas it took 128 weeks for 1 patient to receive a diagnosis. The median time from the diagnosis of sarcoidosis-associated uveitis to the diagnosis of cardiac sarcoidosis was 50 weeks (interquartile range, 12-118 weeks). Three patients received a diagnosis of cardiac involvement within 48 weeks to 140 weeks. One patient received a simultaneous diagnosis of ocular sarcoidosis and cardiac sarcoidosis (Table 2). The length of time to diagnose sarcoidosis and cardiac sarcoidosis was affected by changes in the diagnostic algorithm used to evaluate patients with uveitis. These changes were a result of publications19,26,30,35-38 and the evolving clinical experience.
All 4 patients had negative findings on angiotensin-converting enzyme tests and chest radiography, but a chest CT revealed sarcoidosis. Magnetic resonance imaging (MRI), positron emission tomography (PET), and abdominal CT results revealed that all 4 patients had pulmonary involvement. Two of 4 patients (50%) had no cardiovascular symptoms before undergoing screening with an EKG or Holter monitor. All patients showed ventricular tachycardia. Two patients experienced premature ventricular contractions. After undergoing medical treatments following the manifestations of cardiac sarcoidosis, all 4 patients underwent ICD implantation. All 4 patients showed stable cardiac function and no active ocular inflammation at the last follow-up after treatment of ocular sarcoidosis and cardiac involvement. Treatment information is provided in Table 2. All patients were treated with oral corticosteroids and a steroid-sparing immunomodulatory drug, either azathioprine (1.5-2.5 mg/kg/day) or methotrexate (15-25 mg/week either orally or sucutaneously) with daily folic acid supplementation.
Of the 4 patients with sarcoid uveitis combined with cardiac sarcoidosis, there were 3 patients (75%) with MPCAL. Of the 29 patients with sarcoidosis-associated uveitis without cardiac sarcoidosis, 12 patients (41.4%) had MPCAL. Multiple peripheral chorioretinal atrophic lesions were not significantly or more frequently observed in patients with severe cardiac sarcoidosis necessitating ICD implantation compared with those without (odds ratio, 4.25; 95% CI, 0.39-45.96; P = .31).
To our knowledge, idiopathic uveitis remains the most common diagnosis in most tertiary referral clinics. The term idiopathic is unsatisfying to both physicians and patients. However, the treatment for most forms of noninfectious uveitis is nonspecific. Therefore, skeptics argue that categorizing uveitis into a specific cause has no therapeutic implication. In this article, we described 4 patients who were referred with a diagnosis of idiopathic uveitis. With limited testing, we were able to establish a diagnosis of sarcoidosis as the likely cause. Further, the diagnosis of sarcoidosis had implications that were potentially life-saving. Although Portland, Oregon, has a small African American population for a city its size, this article validates the use of chest CT scanning in selected patients with uveitis, as noted in a previous study from Cleveland, Ohio,19 a city with a larger African American population.
Cardiac manifestations are not clinically apparent in most patients with sarcoidosis, and sudden cardiac death or tachyarrhythmia may be the initial manifestation of cardiac sarcoidosis (CS).39,40 The prevalence of clinically silent CS is reported in 3.7% to 54.9% of patients with CS.41 A few studies have reported that some patients with clinically silent CS have cardiac events, although there is considerable controversy regarding the prognosis of patients with clinically silent CS.41 Therefore, cardiac monitoring on asymptomatic patients with sarcoidosis may be required. In this study, 2 of 4 patients (50%) were asymptomatic before the diagnosis of CS. Only 1 patient had definitive cardiovascular symptoms. Clinically evident cardiac involvement was reported in 2.4% to 6.5% of patients with systemic sarcoidosis.29,32-34 In autopsy studies, cardiac involvement was reported in 20% to 30% of patients with sarcoidosis.39,42 In Japan, cardiac sarcoidosis is reportedly responsible for up to 85% of sarcoidosis-related deaths.31,43
Our findings validate a report from Japan on cardiac sarcoidosis and uveitis. Umazume and colleagues29 reported that severe cardiac involvement that led to implantation of a pacemaker was found in 7 patients (6.5%) with ocular sarcoidosis between 1997 and 2009. For a definite diagnosis of cardiac sarcoid, the presence of noncaseating granulomas on myocardial tissue with no alternative cause is recommended.38,44 However, a myocardial biopsy is not often performed, and only approximately 25% of patients with cardiac sarcoidosis have a noncaseating granuloma identified on an endomyocardial biopsy.45 Therefore, the diagnosis of most cases of cardiac sarcoidosis has been “probable” using World Association for Sarcoidosis and Other Granulomatous Disorders criteria, which consist of (1) treatment-responsive cardiomyopathy or atrioventricular node block, (2) reduced left ventricular ejection fraction in the absence of other clinical risk factors, (3) unexplained sustained (spontaneous or induced) ventricular tachycardia, (4) Mobitz type 2 or third-degree heart block, (4) delayed enhancement or T2 prolongation on a cardiac MRI, (5) patch uptake on a dedicated cardiac PET, (6) positive gallium uptake study, and (7) a defect on perfusion scintigraphy or single-photon emission CT.38,44 Imaging modalities, such as cardiac MRI and PET, may reveal early granulomatous disease and have been increasingly used for diagnosing preclinical cardiac involvement. All patients in this case series did not undergo a myocardial biopsy, but chest CT, cardiac MRI, and PET results revealed the evidence of sarcoidosis.
If there is a mechanism with which to predict cardiac sarcoidosis, it will be quite worthwhile. Umazume and colleagues29 investigated whether ocular findings of sarcoidosis can predict severe cardiac involvement resulting in pacemaker implantation. They concluded that MPCALs were observed significantly and more frequently in patients (6 of 7 patients) with concurrent ocular sarcoidosis and severe cardiac sarcoidosis, suggesting that severe cardiac involvements may be predicted by specific fundus lesions.29 The reason why MPCALs are significantly observed in patients with severe cardiac sarcoidosis and whether MPCALs would appear before cardiac involvement remain unclear. As Umazume and colleagues29 have noted, microcirculatory disturbances may be involved in forming fundus atrophic lesions and myocardial scars. In this case series, 3 of 4 patients (75%) necessitating ICD implantation showed MPCAL on fundus examination results. However, on statistical analysis MPCALs were not significantly more frequently observed in patients with severe cardiac sarcoidosis who needed ICD implantation compared with those without. This result fails to confirm this conclusion by Umazume and colleagues.29 Considering that few patients underwent ICD implantation in this case series, further research may be required. Patients with ocular sarcoidosis may have not only MPCALs but also extensive chorioretinal atrophy in the posterior pole secondary to longstanding involvement from ocular sarcoidosis.46 In this case series, patient 4 showed peripapillary chorioretinal atrophy, but this was not extensive.
Immunomodulatory therapy, such as prednisone and methotrexate, was recommended for cardiac sarcoidosis based on the presence of ventricular arrhythmias, hypermetabolic activity on a cardiac PET scan, and/or left ventricular dysfunction.47 The combination of medications could reduce toxicity and enhance steroid sparing. The combination of prednisone and methotrexate compared with prednisone alone has been reported as less toxic and more effective.48 Two patients (patient 1 and patient 4) received combination treatment with prednisone and azathioprine, and the other 2 patients were treated with prednisone and methotrexate following a diagnosis of cardiac sarcoidosis.
An ICD has been considered as the main therapy for patients with cardiac sarcoidosis with sustained ventricular tachycardia, prior cardiac arrest, or a left ventricular ejection fraction of 35% or less, despite optimal medical therapy and a period of immunosuppression.38 The implantation of a pacemaker has been considered as the necessary therapy for patients with an advanced atrioventricular block.38 In the present case series, all 4 patients underwent ICD implantation because of ventricular tachycardia, and 1 patient (patient 4) underwent a pacemaker implantation because of a complete atrioventricular block. All patients showed stable cardiac function without cardiovascular complications at the last follow-up.
The major limitation of this study is that the conclusion is based on only 4 patients. However, the conclusion is validated by comparable observations published previously by Umazume and colleagues.29
This case series demonstrates the need to recognize cardiac sarcoidosis in elderly patients with sarcoidosis-associated uveitis who present initially with idiopathic uveitis. Our current practice is to obtain a CT scan for all patients with idiopathic uveitis, an age older than 40 years, and a phenotype of disease that is suggestive of sarcoidosis. Generally, this means bilateral disease with intermediate or posterior involvement. We would also recommend CT scans for a disease that affects only the anterior segment if it had granulomatous features. If the CT scan reveals a bilateral hilar adenopathy, we obtain an EKG. We also question the patient regarding possible symptoms to suggest arrhythmia or heart failure. If the EKG results show any conduction delay or rhythm disturbance, or if the review of systems suggests cardiac disease, we refer the patient to a cardiologist for additional cardiac imaging and a more detailed characterization of the rhythm. Although additional testing, such as Holter monitoring or imaging with cardiac MRI, could be considered, most sarcoidosis experts in the United States follow a similar approach to ours.47
Corresponding Author: James T. Rosenbaum, MD, Casey Eye Institute, Oregon Health & Science University, 3181 SW Sam Jackson Park Rd, Mailcode L467AD, Portland, OR 97239 (email@example.com).
Accepted for Publication: October 17, 2017.
Published Online: January 11, 2018. doi:10.1001/jamaophthalmol.2017.5466
Author Contributions: Dr Rosenbaum had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Salek, Rosenbaum.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: All authors.
Obtained funding: Rosenbaum.
Administrative, technical, or material support: Salek.
Supervision: Salek, Rosenbaum.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Rosenbaum reports receiving consulting fees from Gilead, Abbvie, UCB, Regeneron, and Eyevensys; research support from Alcon Research Institute; speaking fees from Mallincrokdt; and royalties from UpToDate. He owns stock in Novartis. No other disclosures are reported.
Funding/ Support: This work was supported by National Institutes of Health grants EY020249 and EY06572, the William and Mary Bauman Foundation, the Stan and Madelle Rosenfeld Family Trust, and Research to Prevent Blindness.
Role of the Funder/Sponsor: The funding sources played no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: