Aims To describe ocular syphilis presentations to a tertiary referral eye hospital over a 5-year period and to document HIV coinfection frequency.
Methods A retrospective chart review was conducted of consecutive ocular syphilis presentations to Sydney Eye Hospital from 2007 to 2012. Inclusion criteria were positive syphilis serology, ocular inflammation on clinical examination and appropriate syphilis treatment. Outcome measures were clinical features at presentation and best-corrected visual acuity (BCVA) at interval follow-up.
Results Thirty-seven eyes of 25 patients were included in the series. Patients were predominantly male (92.0%, p<0.05) with mean age 43.7±14.0 years. Eight (32.0%) patients had confirmed HIV coinfection, three newly diagnosed with HIV. Twelve (32.4%) eyes demonstrated anterior segment involvement with anterior uveitis. Twenty-five (67.6%) eyes demonstrated posterior segment involvement, including panuveitis, acute syphilitic posterior placoid chorioretinitis, retinitis, necrotising retinitis, punctate retinitis and optic neuritis. There was a significant improvement in BCVA for involved eyes (p<0.05) at 1 month and 2–3 months follow-up.
Conclusions The clinical findings of 37 eyes with ocular syphilis demonstrated a broad spectrum of clinical manifestations. Rates of HIV coinfection were high, with patients exhibiting both anterior and posterior segment inflammation. Visual outcome improved following syphilis treatment.
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.
Ocular syphilis is caused by infection with the Gram-negative spirochaete treponema pallidum. Its onset is often insidious, with reports of ocular involvement during all stages of syphilis.1 ,2 The time course between initial infection and ocular presentation is variable. The clinical presentation of ocular syphilis is similarly broad, with involvement of any structure of the eye.3 The complication of syphilitic uveitis, if untreated, has the potential to cause significant visual impairment. In addition, there is the need to exclude neurosyphilis in patients with ocular syphilis. The impact and clinical manifestations of ocular syphilis have been reported previously through numerous case reports and small case series.2 ,4–6 There remains a need for ongoing reports of the clinical presentation, ocular signs and outcomes of syphilitic eye disease.
Coinfection with HIV in patients with ocular syphilis is an important consideration. In parallel with the development of highly active antiretroviral therapy (HAART) and an increase in the number of HIV diagnoses, there has been a resurgence in reported syphilis cases throughout the world.7 Data from the National Notifiable Diseases Surveillance System (NNDSS) report incidence rates in 2012 for Australia to be 6.9 cases per 100 000 population for syphilis of <2 years duration.8
Men who have sex with men (MSM) have been identified to be at greater risk of primary syphilis infection.9 In addition, a significant association has been found between recent or past syphilis infection and HIV seroconversion.10 HIV coinfection rates among this population have been reported to reach 27%.11 This remains an important consideration within Sydney, Australia, due to its large MSM population.
The purpose of this study was to describe ocular syphilis presentations to a tertiary referral eye hospital and to document frequency of HIV coinfection.
Materials and methods
A retrospective chart review was conducted of consecutive presentations to Sydney Eye Hospital of ocular syphilis from January 2007 to November 2012. The Sydney Eye Hospital is a uveitis tertiary referral centre located within central Sydney. It is colocated with Sydney Hospital and the Sydney Sexual Health Clinic.
A structured query language test was used within the South Eastern Area Laboratory Services’ database to identify all reactive cases of Treponema pallidum antibodies, Treponema pallidum particle agglutination test and fluorescent treponemal antibody-absorption test at the Sydney Eye Hospital, Sydney Hospital and Sydney Sexual Health Clinic from January 2007 to November 2012.
A database was subsequently created to identify all patients with positive syphilis serology who presented to Sydney Eye Hospital over this period. This database consisted of all patients presenting to Sydney Eye Hospital where a diagnosis of syphilis was considered in the diagnostic workup and relevant serological investigation conducted. Medical records were subsequently reviewed of all patients testing positive for syphilis serology.
Inclusion criteria were positive syphilis serology with evidence of ocular inflammation on clinical examination, in addition to systemic syphilis treatment. Syphilis serology was deemed positive if initial treponemal investigation was reactive in addition to either a positive non-treponemal investigation or positive second treponemal investigation. Patients were excluded if there was insufficient clinical detail on systemic syphilis management.
All patients underwent detailed ophthalmic history and examination at the time of presentation. Investigations were directed by clinical findings. Outcome measures recorded included patient demographics, presentation and examination details. Best-corrected visual acuity (BCVA) was recorded at initial presentation and at varied follow-up intervals.
In Australia, syphilis is a reportable disease with laboratories required to notify the local Public Health Unit of syphilis cases. Therefore, this database of positive syphilis serology accurately reflects the number of patients affected by ocular syphilis and treated at Sydney Eye Hospital over the study period.
Statistical and data analysis
Data extracted from medical record review were entered into Microsoft Excel 2008 (Microsoft, Redmond, USA) for initial analysis and development of descriptive statistics. Further statistical analysis was performed using SPSS V.20 (SPSS, Chicago, USA).
Uveitis diagnosis was classified according to the Standardization of Uveitis Nomenclature working group criteria.12 BCVA was converted to logarithm of the minimum angle of resolution (LogMAR) equivalent for calculation of average visual acuity and statistical analysis. Due to variability of follow-up duration, ‘interval acuities’ were used to assess response to syphilis therapy.13 BCVA was recorded, where available, at 1 month, 2–3 months and 5–7 months from initial presentation.
Syphilis stage was classified based upon clinical history and serology results according to the Centers for Disease Control case definitions 2012.14 Ocular hypertension was defined according to the Ocular Hypertension Treatment Study Group as an intraocular pressure (IOP) >21 mm Hg.15
Categorical variables were compared using χ2 test or Fisher's exact test when expected outcome was less than five. Continuous variables were compared using the Student t test or Wilcoxon signed-rank test where specified. Eye-based analysis was used where the outcome being assessed was LogMAR equivalent as this enabled clearer interpretation of bilateral cases. Statistical significance was set at p<0.05. p Values were adjusted for multiple comparisons with a Bonferroni correction.16
This study received ethics approval from the South Eastern Sydney Local Health District Human Research Ethics Committee and followed the tenets of the Declaration of Helsinki.
Fifty-six syphilis serology positive cases were identified during the study period, of which 37 eyes of 25 patients met inclusion criteria for the series. There were 25 patients with a diagnosis of ocular syphilis managed through the uveitis clinic where a total of 736 patients were seen (25/736=3.4%). Demographic details of eligible patients are described in table 1. Eight (32.0%) patients had confirmed HIV coinfection, including three newly diagnosed with HIV. Patient follow-up duration ranged from 5 days to 30 months with a median duration of 6 months.
Eight patients were excluded from the series due to inadequate syphilis serology, that is, record of only a single positive treponemal or non-treponemal test. Eight patients were excluded due inadequate syphilis treatment details. Four patients had no evidence of active ocular inflammation on initial clinical examination and were therefore excluded. Eleven patients were excluded due to a diagnosis other than syphilis being deemed more likely after clinical examination and investigation.
Features on presentation
The most frequent ocular presenting symptom was perceived change in vision, followed by red eye and pain. Seven (28.0%) patients recalled previous symptoms consistent with primary syphilis infection, including genital and oral ulcers. One patient reported previous symptoms suggestive of secondary syphilis infection with a widespread maculopapular rash.
Clinical examination and consultations
Detailed clinical examination was undertaken on all patients at presentation (see online supplementary table S1). Twelve (32.4%) and 25 (67.6%) eyes demonstrated predominant anterior and posterior segment inflammation, respectively. Frequently encountered signs of anterior segment involvement included keratic precipitates, cells, flare, fibrin and posterior synechiae. Frequently encountered posterior segment signs included vitritis with snowbanking, various manifestations of retinitis, in addition to optic disc swelling. Posterior segment signs correlated with cases of panuveitis, acute syphilitic posterior placoid chorioretinitis, retinitis, necrotising retinitis, punctate retinitis and optic neuritis.
Ten eyes of seven patients were diagnosed with optic neuritis correlating with evidence of optic disc swelling on clinical examination. Four eyes with optic neuritis had features of vitritis, and a relative afferent pupillary defect was identified in three eyes. This subgroup comprised two HIV positive and four HIV negative patients, with two patients confirmed to have neurosyphilis on lumbar puncture. Initial BCVA for this subgroup ranged from 6/6 (cases 22 and 23) to count fingers (case 24, right eye).
IOP was higher for involved eyes (n=37; 14.97±7.47 mm Hg, range 4–46 mm Hg) than uninvolved eyes (n=12; 13.42±3.06 mm Hg, range 9–19 mm Hg); however, this did not reach statistical significance (p=0.45). Four eyes (10.8%) met criteria for ocular hypertension. There was a significant improvement in BCVA at 1 month and 2–3 months follow-up intervals for involved eyes (table 2). There was no significant change in BCVA at follow-up intervals for uninvolved eyes. BCVA in both eyes of case 8 worsened during follow-up intervals. Progress in clinical examination revealed significant bilateral cataract formation, contributing to poor BCVA at follow-up intervals.
A variety of medical specialities were consulted for patient review and opinion on clinical management. Sexual Health Medicine, Infectious Diseases and General Medicine were the most frequently consulted services, involved in 22, 10 and 9 cases, respectively. Other services consulted included Immunology, Microbiology, Otolaryngology and Respiratory Medicine.
A range of auxiliary investigations were undertaken, guided by clinical history and examination findings and directed by the treating clinician. Optical coherence tomography (OCT) was performed on nine eyes of five patients. Bilateral cystoid macular oedema was confirmed on OCT of case 20. Retinal nerve fibre layer thinning was identified on OCT in the left eye of case 23 and both eyes of case 25. B scan was performed on the left eye of case 13 identifying a fibrin mass in the posterior segment that resolved following treatment. Fundus fluorescein angiography (FFA) demonstrated hyperfluorescence with leakage at the optic disc in the left eye of case 21. FFA performed on the left eye of case 22 confirmed an area of retinal vasculature leakage. Electrophysiology studies with visually evoked potentials were used to confirm optic neuritis in cases 22, 23 and 25. MRI of the brain was performed on three patients. MRI brain was suggestive of right optic nerve atrophy in case 24 and bilateral optic neuritis in case 25.
Lumbar puncture to investigate for neurosyphilis was performed in 14 (56.0%) cases. Of these, six (24.0%) cases demonstrated positive syphilis serology on cerebrospinal fluid (CSF) investigation.
A 14-day course of intravenous benzylpenicillin was the most frequently prescribed systemic syphilis therapy, administered to 14 cases (56.0%). Seven cases (28.0%) were prescribed intramuscular benzathine penicillin while four cases (16.0%) received oral doxycycline due to penicillin allergy. Twelve cases (48.0%) were administered prednisone in addition to antibiotic therapy.
In addition to syphilis-specific therapies, topical glucocorticoids, cycloplegics and IOP-lowering medications were also frequently prescribed. Case 20 received bilateral orbital floor triamcinolone injections. Case 14 received intravitreal foscarnet for treatment of the initial diagnosis of acute retinal necrosis. Case 15 required laser retinopexy and vitrectomy of the left eye for management of retinal detachment.
This study described the clinical characteristics and outcomes of a cohort of patients with positive syphilis serology and syphilitic eye disease over a 5-year period. Posterior segment inflammation was more frequent than anterior segment involvement, affecting 25 (67.6%) and 12 (32.4%) eyes, respectively. Patients were typically men, with eight (32.0%) confirmed to be HIV positive. Admission for intravenous antibiotic therapy, further investigation or management of ocular complications was necessary in 16 (64.0%) patients. Patient demographics were comparable with the previous literature.5 The majority of patients in this cohort reported their country of origin as being outside of Australia.
Syphilis staging is divided into primary, secondary, latent, neurosyphilis, congenital or late benign and cardiovascular syphilis.14 While ocular syphilis has been reported across all disease stages, it often occurs during the latent disease stage.4 The most common stage recorded at presentation was latent syphilis of unknown duration, accounting for 44.0% of the cases in this series. Given that the chancre of primary syphilis often goes unnoticed prior to healing, it is not surprising that the duration of syphilis was frequently difficult to determine.
The association between ocular syphilis and HIV is well described. Genital ulcers due to primary syphilis result in disruption of primary immune barrier defences, thereby increasing the risk of HIV transmission. Rates of syphilis and HIV coinfection have been reported to be as high as 70% during disease outbreaks.17 HIV coinfection may result in an acceleration of the clinical manifestations of syphilis in addition to the potential for early neurosyphilis.18 HIV coinfection was confirmed in 32.0% of patients in this series. This is a conservative estimate given HIV status was not available on five patients. The presentation of ocular syphilis heralded diagnosis of HIV in three patients.
A recent syphilis diagnosis has been found to be an independent predictor of HIV seroconversion.10 At a population level, the introduction of HAART is reported to have had a limited impact on ocular syphilis levels.7 This is in contrast to other ocular opportunistic infections such as cytomegalovirus retinitis whose incidence has declined.19 Despite the limited impact of HAART on ocular syphilis levels, it remains a valuable additional treatment option in the acute setting, having been demonstrated to reduce the odds of neurosyphilis by 65%.18
The division between anterior and posterior segment involvement in syphilitic uveitis varies throughout the literature. The review by Amaratunge et al 20 of 143 patients with syphilitic uveitis found the breakdown of anterior, posterior and panuveitis to be 20%, 55% and 25%, respectively. In this study, over two-thirds of patients presented with posterior segment inflammation.
The association between HIV status and ocular segment involvement has been analysed in previous case series with a suggestion that HIV positive patients tend to present with predominantly posterior segment involvement.20 ,21 It is possible that posterior segment involvement in ocular syphilis is not solely associated with positive HIV status, but may also be influenced by the degree of immunosuppression. A review of 101 cases of ocular syphilis in HIV positive patients found posterior uveitis to be significantly more frequent in those with a CD4 count of <200 cells/mm3.22 In this series, posterior uveitis was identified in HIV positive patients with a wide range of CD4 cell counts, 190–800 cells/mm3.
The relationship between HIV infection and increasing severity of ocular syphilis remains an unresolved issue.23 In the review by Tucker et al,22 visual prognosis following syphilis treatment was good despite HIV coinfection. Similarly, in the analysis by Eandi et al5 of HIV positive and negative patients presenting with acute syphilitic posterior placoid chorioretinitis (ASPPC), there were no discernable differences between presentation and outcome with respect to HIV coinfection. In this study, both HIV positive and negative patients generally had improved BCVA following treatment for syphilis.
Ocular syphilis has protean manifestations that can involve any ocular tissue.3 A wide range of ocular signs were encountered in this series, including anterior uveitis, panuveitis, acute syphilitic posterior placoid chorioretinitis, retinitis, necrotising retinitis, punctate retinitis and optic neuritis. Although anterior uveitis was common, affecting 12 (32.4%) eyes, various forms of retinitis were the predominant ocular sign in 13 (35.1%) eyes. In addition, ocular syphilis was often found to present with a ‘hot eye’, with 19 (51.4%) eyes displaying evidence of inflammation in both the anterior and posterior segment.
Optic nerve involvement in ocular syphilis may present as optic atrophy, inflammatory disc oedema or papilloedema or with gumma of the optic disc.3 Optic disc oedema frequency has been reported at 13% in patients presented with posterior segment ocular syphilis.2 FFA on these patients may demonstrate fluorescein staining or leakage at the optic nerve head.24 In this study, 10 (27%) eyes had ocular examination findings consistent with optic neuritis. Other investigations such as OCT, FFA, electrophysiology and MRI brain were useful in confirming diagnosis. Visual prognosis in cases of optic neuritis in this series was generally good with all eyes, demonstrating either improvement in BCVA or stabilisation at their initial BCVA at final recorded interval follow-up.
Syphilitic uveitis has previously been divided into granulomatous and non-granulomatous forms of the disease. In recent years, there have been increasing reports emphasising the prevalence of non-granulomatous syphilitic uveitis.4 ,6 This trend was observed in the study cohort with mutton-fat, or granulomatous, keratic precipitates found in only three eyes.
Appropriate systemic syphilis therapy has been demonstrated to positively impact upon visual outcome and is often taken as an indication of correct diagnosis.25 Current treatment guidelines advise the use of a single dose of intramuscular benzathine penicillin for management of primary, secondary and early latent syphilis.26 Patients with late latent syphilis should receive three doses of intramuscular benzathine penicillin over a 2-week period.26 Those with neurosyphilis require intravenous benzylpenicillin therapy for up to 17 days.26 In this study, BCVA was found to significantly improve for involved eyes after appropriate syphilis treatment, at 1 month and 2–3 months follow-up points. The effectiveness of syphilis therapy on BCVA outcome may be influenced by the site of ocular involvement upon presentation.27
Complications, generally secondary to intraocular inflammation, were observed in a moderate proportion of patients. Nine eyes developed posterior synechiae, three eyes developed cataract and two eyes developed hypopyon. Ocular hypertension was found in four eyes, which responded to topical IOP-lowering medications. Elevated intraocular pressure is common in patients with uveitis and has been reported to affect up to 18% of patients with uveitis and positive syphilis serology.28 Additionally, the syphilitic uveitis population appears to be significantly more affected by this complication than other causes of uveitis.28
There have been several previous case series describing ocular syphilis throughout the world, notably in Singapore,29 China6 and the USA.24 Chao et al24 highlighted the re-emergence of ocular syphilis describing four patients with CSF confirmed neurosyphilis. Placoid lesions consistent with ASPPC were identified in two of these patients, both HIV negative.24 A recent report of 18 eyes from 12 patients with ocular syphilis in a Singaporean population identified an HIV positive coinfection rate of 25%.29 Isolated anterior and panuveitis were present in two-thirds of eyes in the Singaporean series.29 A Chinese study of 35 eyes of 19 patients reported by Yang et al6 noted higher rates of posterior uveitis, with over 85% of eyes displaying evidence of posterior segment involvement. HIV coinfection was identified in four (21%) patients in the Chinese series.6 The frequency of HIV coinfection in this series was 32%, which is higher than that identified in the Singaporean and Chinese series.6 ,29
This study has limitations. As a retrospective study patient follow-up period varied, as did investigations which varied according to each clinical scenario. In addition, given syphilitic uveitis comprises a small percentage of all uveitis presentations, the sample size of this series remains small.
In conclusion, this study describes the clinical characteristics and visual outcomes of patients presenting with ocular syphilis over a 5-year period to the Sydney Eye Hospital. Ocular syphilis continues to be an important cause of uveitis and optic neuritis among the HIV positive and MSM population. Ocular syphilis was found to manifest with a broad, and often complex, variety of ocular signs. BCVA was found to significantly improve at 1 month and 2–3 months follow-up after appropriate systemic syphilis treatment. The possibility of syphilis as a cause for uveitis and optic neuritis should continue to be considered in the diagnostic workup of at-risk patients and in those for whom there is no other clear discernable cause.
Contributors All authors are given credit for (1) substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content and (3) final approval of the version to be published.
Competing interests None.
Ethics approval South Eastern Sydney Local Health District Human Research Ethics Committee
Provenance and peer review Not commissioned; externally peer reviewed.