To determine the value of anterior chamber paracentesis with polymerase chain reaction (PCR) in patients with anterior uveitis.
Retrospective observational case series.
setting : Bascom Palmer Eye Institute. patient population : Fifty-three patients with a diagnosis of anterior uveitis who underwent anterior chamber paracentesis with PCR. observation procedures : Anterior chamber paracentesis with PCR of aqueous fluid. main outcome measure : Diagnostic utility and frequency of management change based on anterior chamber paracentesis and PCR.
There were 15 of 53 (28%) acute and 35 of 53 (65%) chronic anterior uveitis patients. PCR positivity of herpes simplex virus, cytomegalovirus, varicella zoster virus, and Epstein-Barr virus in our population were 4 of 53 (8%), 1 of 47 (2%), 1 of 35 (3%), and 1 of 18 (6%). Overall, 7 of 53 patients (13%) had a change in management because of PCR results from anterior chamber paracentesis. Four patients encountered paracentesis complications (4/53, 7.5%), 1 with long-term sequelae.
Anterior chamber paracentesis with PCR had a relatively low diagnostic utility and resulted in few management changes in patients with suspected infectious anterior uveitis.
Anterior, intermediate, and posterior uveitis and panuveitis as a whole contribute to greater than 10% of visual impairment in the western world. Anterior uveitis is the most common anatomic subtype of uveitis, based on the International Uveitis Study Group classification, and comprises up to 60% of uveitis cases.
Anterior uveitis derives from both noninfectious and infectious causes. Distinguishing infectious from noninfectious anterior uveitis is important as monotherapy with corticosteroids can exacerbate an undiagnosed infection. Among the most common noninfectious etiologies are systemic diseases such as HLA B27 spondyloarthropathies, sarcoidosis, and Behçet disease. When considering infectious anterior uveitis, factors such as history, biomicroscopic features, host immune status, and geographic location are helpful in determining the most probable pathogens. In the western world, herpes simplex virus (HSV) and varicella zoster virus (VZV) are recognized as the most common causes of anterior uveitis.
Cytomegalovirus (CMV) anterior uveitis is a newly described cause of infectious herpetic anterior uveitis. Recent studies from Singapore and Thailand have identified CMV DNA in 20% to 24% of anterior chamber taps. The frequency of CMV DNA recovery in the United States, however, has not been reported. Making a correct diagnosis of CMV anterior uveitis is important, as its treatment (oral valganciclovir) is a medication with significant side effects and is therefore not empirically used. Furthermore, lengthy treatment is often required as up to 75% of patients recur upon discontinuation of the drug.
The diagnostic approach to anterior uveitis is primarily guided by history and clinical examination. Laboratory screening, imaging, and invasive sampling techniques are used in diagnostic dilemmas and for confirmation of etiology. Anterior chamber paracentesis is an invasive, but reportedly safe, procedure that samples aqueous humor and is used to aid in the diagnosis of anterior, intermediate, and posterior uveitis. There are few studies on the diagnostic utility of anterior chamber paracentesis with polymerase chain reaction (PCR) in patients with posterior uveitis, but this has yet to be investigated in anterior uveitis. In the 2 studies that looked at anterior chamber paracentesis and its impact on the management of posterior uveitis, treatment was altered in 20% and 24% of the patients. The purpose of our study was to retrospectively review cases of suspected infectious anterior uveitis that underwent anterior chamber paracentesis with PCR to assess the diagnostic utility and frequency of management change in these patients. A secondary goal was to look at the frequency of CMV recovered from aqueous specimens and to compare this to the frequency reported in Singapore and Thailand.
Subjects and Methods
Approval for this study was obtained from the University of Miami Institutional Review Board for the retrospective review of existing patient records. The methods adhered to the tenets of the Declaration of Helsinki and were HIPAA compliant. We performed a retrospective analysis of patients with anterior uveitis who underwent PCR testing of aqueous humor from January 1, 2007 to April 15, 2012. Patients were identified by cross-referencing the PCR testing list maintained by the University of Miami Ocular Microbiology Laboratory (n = 595) with the presence of an International Classification of Disease (ICD-9) code for anterior uveitis (364.00, 364.01, 364.02, 364.03, 364.04, 364.05, 364.10, 364.11, 364.21, 364.22, 364.23, 364.24, or 364.3) (n = 24 000). Using this method, 130 charts were selected for review. Seventy-seven charts were excluded based on an ocular diagnosis other than anterior uveitis (eg, panuveitis, chorioretinitis, anterior and intermediate uveitis, and vitritis), leaving 53 patients who were included in this analysis. A PubMed literature search using combinations of the terms “uveitis,” “anterior uveitis,” “iridocyclitis,” “cytomegalovirus,” “herpes simplex virus,” “polymerase chain reaction,” and “anterior chamber paracentesis” was used to review the literature on anterior chamber paracentesis in anterior uveitis and the frequency of CMV DNA recovery in anterior uveitis.
All charts were reviewed for patient demographics, disease course and treatments prior to paracentesis, ocular inflammation prior to paracentesis, intraocular pressures and best-corrected visual acuity (BCVA) at time of paracentesis, slit-lamp examination at time of paracentesis, PCR results, and postparacentesis management.
Disease course and anterior chamber inflammation was classified in accordance with the Standardization of Uveitis Nomenclature (SUN). The presence of trace/0.5 cell and greater was defined as anterior chamber inflammation. The uveitis course was categorized into acute, recurrent, and chronic as described by the SUN definitions: acute = first episode of anterior chamber inflammation with sudden onset and less than 3 months’ duration; recurrent = greater than 1 episode of anterior chamber inflammation with at least 3 months without treatment between episodes; chronic = persistent anterior chamber inflammation greater than 3 months that recurs upon discontinuation or taper of treatment.
Ocular hypertension was defined as IOP >21 mm Hg in the affected eye. Visual acuity before paracentesis was categorized as losses of none to mild (20/20-20/40), moderate (20/50-20/200), or severe (20/200 and lower). Endotheliitis was defined as the presence of anterior chamber inflammation and keratic precipitates (KP) with overlying stromal or epithelial corneal edema. Fuchs uveitis syndrome (FUS) and Posner-Schlossman syndrome (PSS) were recorded as present if identified as such by the treating physician.
Aqueous fluid extraction was performed at the slit lamp with topical anesthesia after placement of a sterile lid speculum. A 27- or 30-gauge needle was used, with extraction of 0.1 to 0.2 mL of aqueous humor followed by injection of balanced salt solution in some cases. Complications secondary to paracentesis were recorded either immediately after paracentesis or within 1 month.
Real-time PCR was performed by Focus Diagnostics (Cypress, California, USA) within 48 to 72 hours of collection. Samples underwent analysis for combinations of HSV, VZV, CMV, or Epstein-Barr virus (EBV) DNA based on clinical suspicion. Limits of PCR detection were as follows: (1) HSV1 and HSV2: 100 copies/mL; (2) VZV: 500 copies/mL; (3) CMV: 200 copies/mL; (4) EBV: 100 copies/mL.
The main outcome measures included the frequency of viral presence by PCR and the frequency of management change (defined as any change in treatment strategy directly attributable to PCR findings).
Descriptive statistics were applied to continuous data while categorical variables were summarized by frequencies using IBM SPSS statistical software version 19 (SPSS, Inc, Chicago, Illinois, USA). Logistic regression analysis was used to assess which presenting factors were significantly associated with a positive PCR result.
A total of 53 patient charts were reviewed. Demographics of patients are listed in Table 1 . Mean age was 56, with a slight female predominance. Race/ethnic composition of our population was 66% white and 30% black, with 28% of the population self-identifying as Hispanic. The majority of patients were immunocompetent (94%).
|Mean age, y ± SD (range)||56.3 ± 17.2 (1286)|
|Sex, female (%)||27 (51%)|
|Black (%)||16 (30%)|
|Ethnicity, non-Hispanic (%)||38 (72%)|
|Setting of presentation and/or care|
|Referral, outside||24 (45%)|
|Emergency room a||14 (26.5%)|
|Referral, within institution||9 (17%)|
|First visit at clinic, no referral||6 (11.5%)|
|Reason for paracentesis|
|Persistent AC inflammation on CS||17 (32%)|
|Persistent AC inflammation on CS & ACV||17 (32%)|
|Clinically suspected viral etiology||12 (22.5%)|
|IOP >21 mm Hg on maximum medical therapy||4 (7.5%)|
|Suspected viral etiology in FUS||2 (4%)|
|Persistent AC inflammation on CS & MTX||1 (2%)|
|Treatments prior to paracentesis|
|Topical CS||50 (94%)|
|Course of oral ACV||23 (43%)|
|Oral corticosteroid||10 (19%)|
|Topical antibiotic||10 (19%)|
|Topical NSAID||5 (9%)|
|Oral methotrexate||2 (4%)|
|Treatments at time of paracentesis|
|Topical CS||44 (83%)|
Approximately two-thirds of the cohort had chronic anterior uveitis (65% [35/23]), one-third had acute anterior uveitis (28% [15/53]), and the minority had recurrent anterior uveitis (6% [3/53]). Almost all patients received topical corticosteroids and 43% received acyclovir before paracentesis. Indications for paracentesis included: (1) persistent anterior chamber inflammation while on topical corticosteroids (33% [17/53]); (2) persistent anterior chamber inflammation while on topical corticosteroids and acyclovir (33% [17/53]); (3) clinical features and/or presentation (KP, PSS-like, iris atrophy) that led to a suspected viral etiology (22.5% [12/53]); (4) IOP >21 on maximum medical therapy (7.5% [4/53]); (5) FUS with suspected viral etiology (4% [2/53]); and (6) persistent anterior chamber inflammation on topical corticosteroids and a systemic immunosuppressant (2%[1/53]) ( Table 1 ). Eleven of 53 patients (21%) had “suspicion for CMV” specifically documented in their medical records.
Clinical characteristics upon aqueous fluid extraction are described in Table 2 . Thirteen percent of patients (7/53) had visual acuity of less than 20/200. Ocular hypertension was found in 40% of patients (21/53), with a mean IOP of 24 ± 14 mm Hg. Two-thirds of the patients (34/53) had keratic precipitates ( Figure ), most commonly pigmented or fine (35% [12/34], 32% [11/34]) and located inferiorly (77% [26/34]). Clinical presentations of FUS-like and PSS-like anterior uveitis were uncommon (7.5% [4/53] and 6% [3/53], respectively).
|Affected eye, right (%)||22 (41.5%)|
|Affected eye, bilateral (%)||3 (6%)|
|Vision affected eye, 20/20 to 20/40 (%)||23 (43.5%)|
|Vision affected eye, 20/50 to 20/100 (%)||23 (43.5%)|
|Vision affected eye, 20/200 and worse (%)||7 (13%)|
|Mean IOP in affected eye (mm Hg ± SD)||24 ± 14|
|Ocular hypertension||21 (40%)|
|AC inflammation||49 (92%)|
|Keratic precipitates||34 (64%)|
|Corneal edema||12 (23%)|
|Iris atrophy||8 (15%)|
|Posterior synechiae||7 (13%)|
|Cystoid macular edema||5 (9%)|
|Fuchs uveitis syndrome||4 (7.5%)|
|Posner-Schlossman syndrome||3 (6%)|
Regarding PCR testing, HSV and CMV were the 2 most commonly tested pathogens (92% [49/53] and 89% [47/53], respectively). In the 53 samples, the overall PCR positivity was 11% (6/53, 95% CI 4%-23%) ( Table 3 ). HSV was the most common pathogen detected and was found in 4 of 49 patients (8%). CMV, VZV, and EBV were all detected once (2% [1/47], 3% [1/35], and 6% [1/18], respectively). One patient was positive for both HSV and EBV. There were 5 anterior chamber paracentesis complications in 4 of the 53 patients (7.5%; 95% CI 2%-18%). Two patients developed hyphema, 1 an FUS patient. One patient developed a post-tap bleb, which resolved. The last patient experienced 2 complications: a Seidel-positive wound for 2 days and subsequent development of a permanent central scotoma.