Aims To determine the clinical features of patients with uveitis with biopsy-proven sarcoidosis, document differences in these features according to ethnicity, age and sex, and assess the diagnostic value of biochemical and imaging examinations.
Methods Retrospective study of 83 biopsy-proven sarcoid uveitis cases seen at two Departments of Internal Medicine and two Departments of Ophthalmology between April 2004 and March 2014.
Results Caucasian patients presented with uveitis at a later age than non-Caucasian (58 years vs 41 years; p=0.001) and had more often a chronic form (78.3% vs 43.8%; p=0.01). Women had higher rates of chronic macular oedema than men (48.3% vs 14.3%; p=0.01). There were no statistically significant differences between patients aged ≤50 years and patients aged >50 years. ACE levels were high (>62 U/L) in 61.7% and lysozyme levels high (>16.7 mg/L) in 83.9% of tested patients. Chest X-rays and CTs were suggestive of sarcoidosis in 62.8% and 91.2% of cases, respectively. Among 21 patients with positive tomography and negative X-rays, 13 were Caucasian women >50 years. Endoscopic ultrasound-guided fine-needle aspiration of intrathoracic nodes contributed to the diagnosis in 7 patients with normal labial salivary gland and transbronchial biopsies. Any of the enzyme tests together with any of the imaging tests identified 100% of the patients.
Conclusions In this largest European series of biopsy-proven sarcoidosis to date, the outstanding diagnostic ability of enzyme test plus imaging test couple suggests that the recourse to invasive procedures should be limited to patients with ocular involvement that would justify systemic treatments.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Sarcoidosis is a systemic inflammatory disorder of unknown aetiology that is characterised, pathologically, by non-caseating epithelioid-cell granulomas that affect primarily the lungs and the lymphatics, and, clinically, by various symptoms and courses.1 However, ocular sarcoidosis may develop in the absence of any apparent systemic involvement and may even be the main or sole site of the disease.2 In the prospective ACCESS study, 11.8% of 727 patients with histologically proven sarcoidosis had ocular problems as presenting symptoms.3
In ocular sarcoidosis, any eye structure may be involved but uveitis remains the most frequent finding; it may affect up to 30% of patients with sarcoidosis during the course of the disease.4 ,5 However, the frequency of ocular involvement varied between reports. In Japan, ocular lesions are the most frequent presentations of sarcoidosis; they were reported in 50–93% of patients with sarcoidosis.5 ,6 In the USA, ocular sarcoidosis was more prevalent in Black than in Caucasian subjects and the former tended to be younger at first presentation to the ophthalmologist and had more frequently anterior granulomatous inflammations.7 In contrast, chronic posterior uveitis was more commonly seen in Caucasian female patients with late disease onset. In Europe, only a few studies—mostly from England—have reported the clinical features of patients presenting with sarcoid uveitis.8 ,9
Patients with only ocular symptoms present a difficult challenge because establishing a definitive diagnosis of sarcoidosis using intraocular biopsy may be associated with significant morbidity. In these patients, the diagnosis is usually established on the basis of granulomatous inflammation, elevated serum ACE, elevated lysozyme, and chest imaging with findings characteristic of hilar lymphadenopathy or pulmonary nodules. The latter three tests have been reported to have various sensitivities and specificities. However, in 29 out of 63 patients with biopsy-proven ocular sarcoidosis (of which 39 were African-American), Birnbaum et al10 showed that each of ACE or lysozyme identified more patients when used in combination with chest imaging. Recently, the minimally invasive endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA, which used a curved linear array ultrasonic bronchoscope and provided high-resolution images of the mediastinum) succeeded in establishing the diagnosis of sarcoidosis making practically unnecessary the use of mediastinoscopy.11
The purposes of the present study were: (1) determining the clinical features of uveitis in a large group of patients with biopsy-proven sarcoidosis; (2) searching for differences in clinical features according to ethnicity, sex and age; and, most importantly, (3) assessing the diagnostic value of various tests.
Patients and methods
Records of all patients were seen at two Departments of Internal Medicine and two Departments of Ophthalmology of Hospices Civils de Lyon between April 2004 and March 2014.
The study identified first the records of patients older than 18 years with a diagnosis of uveitis and a biopsy-proven sarcoidosis.1 The records of patients with other granuloma-forming diseases (such as tuberculosis or fungal infections) were excluded. The demographic characteristics, medical histories and sarcoidosis-related investigations were collected and analysed by two investigators (CF and PS) using a standardised form. All white patients were considered as Caucasian; patients of African (Maghreb included) or Asian origin were considered as non-Caucasian.
These data included: (1) eyelid, conjunctiva, cornea, anterior chamber, iris, lens, vitreous and retina examination findings at presentation; (2) the biopsy site and histology results; (3) the results of available diagnostic tests: ACE and lysozyme levels, chest X-ray examination12 and chest CT.
Because the above-cited tests or examinations were required by distinct physicians (ophthalmologists, pulmonologists or internists) all patients did not have the same series of results. Some patients with suspicious features had conjunctiva, peripheral node or skin biopsies whereas others underwent labial salivary glands biopsies, bronchial lung biopsies, EBUS-TBNA of intrathoracic nodes or mediastinoscopy.
Uveitis was classified according to the Standardization of Uveitis Nomenclature criteria and chronic uveitis was defined as inflammation lasting more than 3 months.13
According to the standards of our laboratory, ACE levels were considered high if >62 U/L and lysozyme levels high if >16.7 mg/L. A chest X-ray was considered positive when the image revealed signs of hilar lymphadenopathy, pulmonary granuloma or a ground-glass parenchymal appearance. A chest CT was considered positive in the presence of lymph nodes in the hilum or mediastinum with short axis diameter >10 mm, perilymphatic pulmonary nodules or other parenchymal lung abnormalities.14
All imaging findings were interpreted by the radiologists who carried out the imaging examinations.
Each observed variable was expressed as number (percentage) if categorical and mean (SD) or median (range) if continuous. Caucasian and non-Caucasian patient characteristics were compared using non-parametric tests (Wilcoxon test for continuous variables and χ2 or Fisher's test for categorical variables).
The analyses were carried out with R software (V.2.15, http://cran.r-project.org/). A p value <0.05 was considered for statistical significance.
The study examined the records of 83 adult patients with biopsy-proven sarcoidosis. The non-Caucasian group included 21 patients (15 Maghrebi, 4 black African and 2 Asian).
The mean follow-up duration was 4.76 years (range: 1–300 months). Thirty-seven patients (44.6%) were aged ≤50 years.
All the patients had biopsies suggestive of sarcoidosis: 24 from bronchi, 16 from mediastinal lymph nodes, 15 from labial minor salivary glands, 13 from skin, 8 from conjunctiva, 3 from peripheral lymph nodes and 1 from each of kidney, liver and stomach. Seven patients with normal bronchi and normal labial minor salivary gland biopsies had EBUS-TBNA of intrathoracic nodes that showed granuloma. In one patient with normal chest X-ray and normal chest CT, a further 18-fluorodeoxyglucose positron-emission tomography (18-FDG PET) identified occult sites and a subsequent subclavicular biopsy revealed a non-caseating granuloma without evidence of infection. This patient case study has been published elsewhere.15
Among the 83 patients, 54 (65.1%) had chronic uveitis and 22 (26.5%) had acute uveitis. Five patients had less than 3 months follow-up and uveitis status was unknown in two patients. Four patients had a family history of sarcoidosis.
Ethnicity-related, sex-related and age-related differences
The prevalence of ophthalmological signs and sarcoidosis features varied according to the group considered (tables 1⇓–3). Caucasian patients presented with uveitis at a significantly later age than non-Caucasian patients. In Caucasian patients, uveitis was more often chronic and complicated by chronic macular oedema (CME) (table 1).
Whereas women had higher rates of CME than men, the other features of uveitis showed no statistically significant differences between men and women (table 2).
Chest X-ray abnormalities were more frequent in patients aged ≤50 years than in older patients but no other statistically significant differences between patients could be attributed to older age (table 3).
Diagnostic values of serum markers and chest imaging
Serum markers (namely, ACE and lysozyme) were measured in 82 patients (81 patients for ACE and 62 for lysozyme). High ACE levels were found in 50 patients (61.7%) and high lysozyme levels in 52 patients (83.9%) (table 4). At least one marker was high in 68 patients (82.9%) and only one in 34 patients. Among the 20 patients who had only ACE measurements, only 7 had results within the normal range, of whom one was taking oral prednisone at the time of testing. Two patients were taking an ACE inhibitor at initial presentation but their lysozyme levels were high. Anyway, there were no differences in the values of serum markers according to ethnicity, sex or age (tables 1⇑–3).
A chest X-ray was performed in 78 patients and a chest CT in 80. X-ray findings consistent with sarcoidosis were found in 62.8% of X-rays and 91.2% of CTs (table 4). Among the patients with negative X-rays, 28 underwent chest CTs of whom 21 were suggestive of sarcoidosis. Of the latter patients, 13 were Caucasian women older than 50 years. The proportions of positive chest X-rays and CTs were identical between Caucasian and non-Caucasian, men and women, and those under and over 50 years (tables 2⇑–4).
Among the patients who underwent ACE tests and chest X-rays, 80.5% had at least one result suggestive of sarcoidosis. Among those who had ACE tests and either chest X-rays or chest CTs, this percentage rose to 96.3% (79 out of 82 patients). Lysozyme levels together with chest X-rays were able to identify 59 patients (73.7%) with biopsy-proven sarcoidosis whereas lysozyme levels together with either chest X-rays or chest CTs were able to identify 80 patients (98.7%). Among the 82 patients who had at least one serum test together with chest X-ray, 78 (95.1%) had at least one test result consistent with a diagnosis of sarcoidosis. The result of one serum test (ACE or lysozyme) together with that of one imaging test (chest X-ray or CT) identified 100% of the patients with biopsy-proven sarcoidosis (table 4).
The present study displays the clinical presentations and the test findings in one of the largest series of patients with biopsy-proven sarcoidosis. Unlike several previous studies on this topic, all the patients had uveitis and a large proportion of them underwent a chest X-ray and a chest CT. Besides, unlike other reports,7 ,10 ,16 ,17 three out of four patients were Caucasian, which helps make comparisons with previous European studies on sarcoidosis manifestations.9 ,18
The above-presented findings contrast with those of another historical large English series of 123 patients with biopsy-proven sarcoidosis and ocular involvement.9 In that series, the most common presentation was acute or chronic iridocyclitis (72%), followed by some form of conjunctival involvement whereas, here, more than 75% of the cases were diagnosed within a year after the first symptoms which, in 71 patients, were those of uveitis. This difference is most probably due to the fact that most of the patients here were sent by the ophthalmologist for assessment of sarcoidosis.
In the present series, Caucasian patients presented with sarcoid uveitis at an older age than non-Caucasian and had significantly higher rates of posterior involvement and CME. A similar finding was reported in a study of systemic sarcoidosis7 and in a series of proven ocular or posterior segment ocular sarcoidosis. These ethnicity-related differences may be due to genetic and environmental differences, which reflects the complex nature of the disease.19
In agreement with another report,20 sex-related differences were also observed: women had a higher rate of posterior segment involvement or CME than men. Rothova et al and Khalatbari et al17 have reported higher rates of CME and worse visual acuities in women than in men.9 Moreover, Caucasian women seem to be a special clinical subgroup of patients with ocular sarcoidosis. As reported here and elsewhere, this subgroup had the highest mean age at initial examination and significantly higher rates of chronic forms and CMEs whatever the age.
Here, ACE and lysozyme tests succeeded in identifying, respectively, 62% and 84% of the patients with biopsy-proven sarcoidosis. These results are in disagreement with those of Birnbaum et al10 who reported that these levels were high in, respectively, only 40% and 42% of 24 patients tested. However, these authors did not specify whether these patients were receiving systemic immunosuppression, which may influence serum ACE and lysozyme levels. The literature reported various sensitivities and specificities of ACE and lysozyme tests in ocular sarcoidosis (58–84% sensitivity and 83–95% specificity for ACE and 60–78% sensitivity and 76–95% specificity for lysozyme), although none of these studies included more than 22 patients with biopsy-proven sarcoidosis.21 ,22 Anyway, because ACE and lysozyme tests were found normal in many patients with biopsy-proven sarcoidosis, a negative result should not exclude a clinical diagnosis of sarcoidosis. In fact, imaging techniques alone identified more patients than single or joint ACE and lysozyme tests.
In parallel, here too, the findings of chest X-rays and chest CTs were considered positive in respectively, 63% and 91% of tested patients. These findings are close to those of Birnbaum et al10 who found images consistent with sarcoidosis in 69% of 25 patients who underwent chest X-rays and 100% of 19 patients who underwent CTs. Obtained in a larger group, the present results confirm with stronger evidence that chest CT should be reserved for patients with negative chest X-rays but still suspected of sarcoidosis. This is especially true in Caucasian women >50 years.14
In the present series, biochemical tests (ACE and lysozyme) and chest imaging (X-rays and CT) were able to identify 82 out of 83 patients with biopsy-proven sarcoidosis. Unfortunately, the study lacked a control group of patients with non-sarcoid uveitis; thus, we cannot comment on the false-positive rates of these diagnostic tests. However, molecular imaging has recently made important progress in diagnosing sarcoidosis. Indeed, 18-FDG PET provided an insight into the metabolism of this disease and proved particularly effective in detecting occult disease and assessing disease activity during treatment.15 Here, 18-FDG PET was useful in diagnosing sarcoidosis in only one patient with normal serum markers and chest imaging.
Recently, several studies have reported that EBUS-TBNA outperforms the diagnostic yield of transbronchial biopsy by means of bronchoscopy.11 EBUS-TBNA is available in our teaching hospital since 3 years and was useful for the diagnosis of sarcoidosis in seven patients with previous normal labial salivary gland and transbronchial biopsies. This supports the role of EBUS-TBNA in establishing a definite diagnosis of sarcoidosis in a large series of sarcoid uveitis cases.
Despite its multiple assets, the present study has several limitations. First, it was carried out within departments belonging to the same institution. However, this institution is a secondary reference centre and most of the patients had a previous short history and almost all had undergone a standard screening protocol including medical history, physical examination, chest X-ray and chest CT. Second, like most retrospective studies, full data could not be obtained for all patients, particularly the full list of ancillary tests and examinations.
In conclusion, to our knowledge, the present study reports on the largest European series of biopsy-proven sarcoidosis. It displays the main clinical features of the disease and the diagnostic values of key examinations. Overall, the study confirms the existence of two groups of patients with sarcoid uveitis: one young and multiethnic group in which ophthalmological findings are polymorphous and another group of elderly Caucasian women with mostly chronic posterior uveitis.23 In patients with uveitis suggestive of sarcoidosis, serum ACE or lysozyme level and chest X-ray would miss many diagnoses, especially among elderly Caucasian women. Serum ACE test plus serum lysozyme and chest imaging (X-ray only if positive, CT if X-ray negative) may identify a high majority of the patients; in the present series, all biopsy-proven cases could be identified.
Contributors CF, LK, LP, YJ, ChB, CaB, PS collected the data. CF, LK, DM-B, LP, YJ, ChB, CaB, PS performed the statistical analysis and interpretation of data. CF, LK, DM-B, LP, JI, YJ, ChB, CaB, PS drafted and revised the manuscript. All authors read and approved the final version.
Competing interests None.
Ethics approval This study was conducted with the approval of the institutional review board of Hospices Civils de Lyon.
Provenance and peer review Not commissioned; externally peer reviewed.