- Email: [email protected]
Prognostic Factors in Syphilitic Uveitis
Prognostic Factors in Syphilitic Uveitis Florence Hoogewoud, MD,1,* Laure Frumholtz, MD,2,* Paul Loubet, MD,3,4 Caroline Charlier, MD, PhD,5 Philippe Blanche, MD,2,6 David Lebeaux, MD, PhD,5 Nadjet Benhaddou, MD,7 Neila Sedira, MD,8,11 Laetitia Coutte, MD,2,6 Clelia Vanhaecke, MD,9 Odile Launay, MD, PhD,6 Claire Le Jeunne, MD, PhD,2 Emmanuel Héron, MD,8,11 Dominique Monnet, MD, PhD,1 Olivier Lortholary, MD, PhD,5 José-Alain Sahel, MD, PhD,10,11 Nicolas Dupin, MD, PhD,7,9 Antoine Brézin, MD, PhD,1 Marie-Hélène Errera, MD, PhD,10,11 Sawsen Salah, MD,1 Matthieu Groh, MD2,12 Purpose: To identify predictors of treatment success in syphilitic uveitis (SU). Design: Retrospective multicentric analysis of patients treated for SU. Participants: A total of 95 eyes (66 patients, mean [standard deviation] aged 49 [12.5] years, 31 [47%] of whom were human immunodeficiency virus [HIV]þ) were analyzed. Methods: Activity of SU was assessed at 1 week and 1 month after treatment onset, and at last follow-up. Improvement was defined by a 2-step decrease of both anterior chamber and vitreous haze inflammation levels, and by the size reduction in chorioretinal lesions. Main Outcome Measures: Recovery was defined as the resolution of inflammation in all anatomic structures at 1 month. Results: Panuveitis and posterior uveitis were the most frequent findings. Inflammatory parameters were higher in HIVþ patients. Recovery was reported in 65% and 85% of eyes at 1 month and at last follow-up, respectively. In multivariate analysis, after adjusting for initial best-corrected visual acuity and the antimicrobial treatment regimen, clinical improvement at 1 week (corrected risk ratios [cRR], 3.5 [2.3e3.8]; P ¼ 0.001) was predictive of recovery at 1 month, whereas the use of periocular dexamethasone injections (cRR, 0.05 [0.02e0.6]; P ¼ 0.01) and methylprednisolone pulses negatively affected the outcomes of eyes. Conclusions: Early improvement is the strongest predictor of ophthalmological recovery in SU. Ophthalmology 2017;-:1e9 ª 2017 by the American Academy of Ophthalmology Supplemental material is available at www.aaojournal.org.
Although syphilis was highly endemic in the preantibiotic era,1 it nearly disappeared in high-income countries after the discovery of penicillin and the implementation of screening and treatment programs.2 However, similar to other sexually transmitted diseases (STDs), a resurgence of syphilis recently has been reported, particularly among human immunodeficiency virus (HIV)infected patients and in the men having sex with men (MSM) population.3 Between 2001 and 2014, the annual incidence of primary and secondary syphilis in the United States increased from 2.1 to 6.3 cases per 100 000 inhabitants.4 Likewise, an outbreak of ocular syphilis occurred recently in the United States.5 Ocular complications (0.6%e15% of cases)6 are more frequent during the secondary and early latent phases of syphilis, but, similar to neurosyphilis, they have been reported in all stages of the disease.7 Although any segment of the eye can be involved, panuveitis and posterior uveitis are the most frequent ocular findings.6 Besides superficial primary syphilis (e.g., canker of the lid or conjunctiva), it has long been hypothesized that syphilitic uveitis (SU) was closely related to neurosyphilis, and current treatment guidelines recommend treating it as ª 2017 by the American Academy of Ophthalmology Published by Elsevier Inc.
such.8 Thus, treatment with intravenous (IV) penicillin G (6 mUI 3 or 4 times per day for 10e14 days) is demanding and may extend the length of hospital stay. Moreover, cases of treatment failure have been reported.9 Although ceftriaxone,10 oral amoxicillin plus 11,12 azithromycin,13 and doxycycline14 have probenecid, been shown to be effective for early stages of syphilis, there are few data regarding the safety and efficacy of alternative treatments for neurosyphilis and SU.15 This study aims to identify predictors of treatment success of SU and to evaluate the efficacy of alternative therapies (e.g., ceftriaxone and benzathine penicillin G [BPG]).
Methods Patients A retrospective multicenter study was conducted from January 2003 to April 2016 in 2 tertiary ophthalmic centers (Cochin and the Quinze-Vingts National Ophthalmology Hospital). Adult patients treated for SU were identified using the medical information system databases. All diagnoses of SU were reassessed by the investigators (F.H., S.S.). Syphilitic uveitis was defined by the http://dx.doi.org/10.1016/j.ophtha.2017.06.003 ISSN 0161-6420/17
1
Ophthalmology Volume -, Number -, Month 2017 presence of ocular inflammation compatible with the diagnosis of syphilis, the positivity of both serum Treponema pallidum hemagglutination assay (TPHA) and Venereal Disease Research Laboratory (VDRL) tests, and the exclusion of alternate diagnoses. Patients with optic neuritis (without any sign of intraocular inflammation) and congenital syphilis were not included. This study was approved by the local ethics committee.
improvement was defined by a 2-step decrease of the levels of both anterior chamber (AC) cells and vitreous haze inflammation, and by size reduction in chorioretinal lesions. Recovery (main outcome measure) was defined as the resolution of inflammation in all anatomic ocular structures. Ocular complications included posterior synechiae, ocular hypertension, glaucoma, macular edema, serous retinal detachment, vasculitis, and optic atrophy.
Baseline Measurements Data collection was performed using a standardized anonymous form. Baseline demographic, clinical, and paraclinical parameters were retrieved, including age, gender, socioeconomic status, sexual orientation, history of syphilis and other STDs, dermatologic and systemic clinical symptoms due to syphilis, HIV status, TPHA and VDRL levels at baseline and 3 months after onset of treatment, type, dose, and route of administration of anti-Treponema drugs, and use of glucocorticoids (topical or systemic). Ophthalmological findings included type and duration of ocular symptoms at presentation, best-corrected visual acuity (BCVA), slit-lamp examination (including grading of anterior segment and vitreal inflammation according to the Standardization of Uveitis Nomenclature [SUN] Guidelines),16 laser-flare meter (when available), intraocular pressure measurement, and fundus examination. The SUN guidelines were used to classify uveitides subtypes.16 When available, cerebrospinal fluid (CSF) data also were collected (routine biochemical, cytologic analyses, and microbiological testing [i.e., VDRL, Fluorescent Treponemal Assay absorption test, and polymerase chain reaction]).17 Lumbar puncture was considered abnormal in case of pleocytosis (>10 cells/mm3), hyperproteinorrachia (>0.4 g/L), or positive CSF microbiological test results.
Outcomes Data from ophthalmological examinations at 1 week (84 days), 1 month (3012 days), and last follow-up after antimicrobial treatment onset were collected. Last follow-up was defined as month 1 examination for patients whose follow-up ended at 1 month and as the last date of follow-up for those who had additional follow-up visits. As defined by the SUN guidelines,16
Statistical Analyses Characteristics of patients and eyes are reported as numbers and percentages for categoric variables and as mean (standard deviation) or median (interquartile range) for continuous variables. For univariate analysis, the Wilcoxon rank-sum or KruskaleWallis tests, and Fisher exact test were used as appropriate. Missing data for each variable were excluded from the denominator. Patient subsets were differentiated on the basis of their HIV status and treatment regimens (“close to standard of care”: group A, 14 days of IV penicillin G; “pragmatic approach”: group B, 5 days of IV penicillin G followed by ceftriaxone or BPG; “nonvalidated treatment regimens”: group C, ceftriaxone or BPG; and group D, oral doxycycline). High ocular inflammation was defined as a composite variable that was considered positive if anterior chamber cell or vitreous haze grades were >2 (when appropriate, taking into account the anatomic location of ocular inflammation). Full details of the variables entered in model building are available in Supplementary material (available at www.aaojournal.org). Factors associated with recovery at 1 month were identified in a backward stepwise logistic regression model considering “recovery at 1 month” (positive/negative) as the dependent variable. All variables with a P value <0.2 in univariate analysis were tested in the model. To exclude variables with high collinearity from the multivariate analysis, variables significantly associated with recovery at 1 month in univariate analysis were assessed for bivariate correlation using Spearman’s test. The fit of the model was tested using the HosmereLemeshow goodness of fit test. Results are expressed as crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs). Because the prevalence of our outcome (“recovery at 1 month”) was frequent (>10%), the ORs were corrected by the Zhang and Yu method18 to avoid
Figure 1. Flowchart showing the search strategy and inclusion/exclusion criteria for the study population.
2
Hoogewoud et al
Treatment of Syphilitic Uveitis
Table 1. Baseline Ophthalmological Findings According to Human Immunodeficiency Virus Status (n ¼ 95 eyes) Variable BCVA <2/10 2/10e5/10 >5/10 Median IOP, mmHg (IQR) Anatomic location of ocular inflammation Isolated anterior uveitis Isolated intermediate uveitis Isolated posterior uveitis Isolated papillitis Retinitis (or chorioretinitis) with papillitis Retinitis (or chorioretinitis) without papillitis Panuveitis AC cells >0.5 Hypopyon Median flare, photons/ms (IQR) Vitreous haze >0.5 Ocular complications Posterior synechiae Serous retinal detachment Cystoid macular edema Glaucoma Retinal vasculitis (fundoscopy)
HIVD (n [ 46)
HIVL (n [ 49)
Total (n [ 95)
P Value
15 (33) 14 (30) 17 (37) 12 [10e14]
9 (18) 19 (39) 21 (43) 13 [12e15]
24 (25) 33 (35) 38 (40) 13 [11e15]
0.29 0.03
7 1 12 10 2
(15) (2) (26) (83) (17)
7 2 26 10 12
(14) (4) (53) (38) (46)
14 3 38 20 14
(15) (3) (40) (53) (37)
1.0 1.0 0.01 -
0
4 (15)
4 (11)
-
26 (57)
14 (29)
40 (42)
0.007
33 (72) 3 (7) 48 [18e88]
16 (33) 1 (2) 16 [8e33]
49 (52) 4 (4) 33 [11e65]
<103 0.35 0.01
29 (63)
18 (37)
47 (49)
0.02
14 (30) 2 (4) 3 (7) 0 7 (15)
4 (8) 2 (4) 0 0 6 (12)
18 (19) 4 (4) 3 (3) 0 13 (14)
0.008 1.0 0.10 0.77
Data are presented as no. (%) unless otherwise specified. AC ¼ anterior chamber; BCVA ¼ best-corrected visual acuity; HIV ¼ human immunodeficiency virus; IOP ¼ intraocular pressure; IQR ¼ interquartile range.
overestimation of the risk ratio. Corrected risk ratios (cRRs) are thus displayed as final results of the multivariate logistic regression model. P values <0.05 were considered statistically significant. All analyses were conducted using STATA software (V12, 1996-2014 StataCorp LP, College Station, TX).
Results Patients’ Identification Screening from the medical information database identified 98 patients with ocular inflammation and both serum-positive VDRL and TPHA. After exclusion of 32 patients, 66 patients (95 eyes) were included in the study (Fig 1), among whom 10 (15%) had received anti-Treponema drugs (oral azithromycin, n ¼ 6; 1 BPG injection, n ¼ 1; oral sulfadiazine, n ¼ 1; oral doxycycline, n ¼ 1) as probabilistic treatment of chorioretinitis or as treatment of secondary cutaneous syphilis for less than 7 days before baseline ophthalmological evaluation.
Baseline Characteristics According to Human Immunodeficiency Virus Status Baseline demographic and clinical characteristics of the 66 patients (61 male, among whom 33/48 [69%] were MSM) are shown in Table S1 (available at www.aaojournal.org). Thirty-one patients (47%) were seropositive for HIV (median CD4þ T Lymphocytes [CD4] count: 313/mm3; range, 30e1750/mm3), among whom
16 of 30 (53%, unknown antiretroviral (ARV) treatment for 1 patient) received ARV therapy at the time of SU diagnosis, with an undetectable viral load in 10 of 25 patients (40%). Patients positive for HIV were younger (mean age, 44 vs. 53 years, P ¼ 0.002), more frequently MSM (87% vs. 52%, P ¼ 0.01), and more frequently reported other concomitant STDs (24% vs. 3%, P ¼ 0.02). Lumbar puncture was performed in 52 patients (79%), among whom 41 (79%) had abnormal CSF findings, without differences between HIV serostatus (P ¼ 0.73). Abnormal CSF findings did not correlate with a specific subtype of uveitis. Hyperproteinorrachia was the most frequent abnormal feature, reported in 67% of all cases. The Fluorescent Treponemal Assay absorption was the most sensitive microbiological CSF test, with positive results in 17 of 19 cases (89%), whereas VDRL and polymerase chain reaction were positive in 18 of 46 cases (39%) and 1 of 16 cases (6%), respectively. Baseline ophthalmological findings according to HIV status are shown in Table 1. Figure S1 (available at www.aaojournal.org) provides iconographic features of uveitides that were encountered. At disease onset, 29 patients (44%) were referred with bilateral ocular involvement. Among the 95 eyes (right eye, n ¼ 50, left eye, n ¼ 45) involved, 46 were from HIVþ patients and 49 were from HIV patients. Among the 33 eyes (35%) with keratic precipitates, only 2 (6%) were granulomatous. There was no difference in baseline BCVA with regard to HIV status. Panuveitis was the most frequent type of uveitis in HIVþ patients (57% vs. 29%, P ¼ 0.007), whereas posterior uveitis was more frequent in HIV patients (26% vs. 53%, P ¼ 0.01). Inflammatory parameters were significantly higher in HIVþ patients’ eyes, with
3
Ophthalmology Volume -, Number -, Month 2017 Table 2. Baseline Ophthalmological Findings According to Treatment Regimens (n¼71 eyes) Characteristics Visual acuity <2/10 2/10e5/10 >5/10 Median IOP, mmHg (IQR) Anatomic location of ocular inflammation Isolated anterior uveitis Isolated intermediate uveitis Isolated posterior uveitis Isolated papillitis Retinitis (or chorioretinitis) with papillitis Retinitis (or chorioretinitis) without papillitis Panuveitis Crystalline lens Clear Cataract Pseudophakia AC cells [0.5e2] [3e4] Hypopyon Median flare, photons/ms (IQR) Vitreous haze [0.5e2] [3e4] Ocular complications Ocular hypertension Posterior synechiae Serous retinal detachment Cystoid macular edema Glaucoma Retinal vasculitis (fundoscopy) Prior treatment with anti-treponemal drugs Topical glucocorticoids Eye drops Periocular dexamethasone injections Systemic glucocorticoids Methylprednisolone pulses (followed by oral prednisone) Oral prednisone (not preceded by methylprednisolone pulses)
Group A (n [ 40) Group B (n [ 21) Group C (n [ 10) Total (n [ 71) P Value 16 (40) 9 (23) 15 (37) 13 [11e15]
1 (5) 8 (38) 12 (57) 12.5 [10.5e14]
4 (10) 0 23 (58) 11/23 (48) 10/23 (43) 2/23 (9) 13 (33)
1 (10) 6 (60) 3 (30) 11.5 [10e17]
18 (25) 23 (32) 30 (43) 13 [11e15]
5 2 3 2/3 1/3
(23) (10) (14) (67) (33) 0 11 (52)
1 (10) 0 3 (30) 3/3 (100) 0 0 6 (60)
10 2 29 16 11 2 30
21 (100) 0 0
10 (100) 0 0
70 (99) 1 (1) 0
12 (57) 6 (29) 2 (10) 23.5 [14e64.8]
1 (10) 4 (40) 0 162 .2 [7.7e185]
26 (37) 17 (24) 4 (6) 33 [13e64.8]
15 (41) 2 (5)
12 (57) 3 (14)
4 (50) 1 (13)
31 (47) 6 (9)
0.40 0.19 0.003 0.18 1.0 0.02 0.79 0.60 0.31 -
0 8 (20) 4 (10) 0 0 4 (10) 5 (12) 19 (49) 17 (44) 5 (13)
0 3 (14) 0 1 (5) 0 4 (19) 4 (19) 12 (57) 14 (67) 1 (5)
0 2 (20) 0 0 0 2 (20) 1 (10) 6 (60) 5 (50) 1 (10)
0 13 (18) 4 (6) 1 (2) 0 10 (14) 10 (14) 37 (53) 36 (51) 7 (10)
0.90 0.44 0.47 0.50 0.90 0.75 0.21 0.75
3 (8) 10 (25)
1 (5) 2 (10)
1 (10) 3 (30)
5 (7) 15 (21)
1.0 0.26
39 (98) 1 (2) 0 13 (33) 7 (18) 2 (5) 37.9 [13.4e57.4]
(14) (3) (41) (55) (38) (7) (42)
0.01 0.38
Data are presented as no. (%) unless otherwise specified. Treatment regimens: group A, 14 days of IV penicillin G; group B, 5 days of IV penicillin G followed by ceftriaxone or BPG; group C, ceftriaxone or BPG exclusively. AC ¼ anterior chamber; IOP ¼ intraocular pressure; IQR ¼ interquartile range.
>0.5 AC cells in 72% versus 33% (P < 0.001), >0.5 vitreous haze in 63% versus 37% (P ¼ 0.02), median flare being 48 versus 16 photons/ms (P ¼ 0.01), and posterior synechiae in 30% versus 8% (P ¼ 0.008) in HIVþ and HIV patients’ eyes, respectively. Among HIV-infected patients, there was a trend toward higher inflammatory parameters in patients with <200/mm3 CD4þ T lymphocytes (Table S2, available at www.aaojournal.org).
Baseline Characteristics According to Treatment Regimens Of the 66 patients enrolled in the survey, 16 were subsequently excluded from further analyses of ocular outcomes (Fig 1). Patients were treated with 18 different regimens that could be merged in 4 different groups (group A, n ¼ 27; group B, n ¼ 14; group C, n ¼ 8; group D, n ¼ 1) (Table S3, available at www.aaojournal.org), as
4
described previously. Patients’ baseline demographic and ophthalmological characteristics according to treatment regimens (49 patients, 71 eyes) are shown in Table S4 (available at www.aaojournal.org) and Table 2, respectively (data from the unique patient from group D are not shown). Eyes belonging to group A patients had poorer baseline BCVA (<2/10 BCVA in 40%, 5%, and 10% of all 3 groups, respectively, P ¼ 0.01) and more frequent posterior uveitis (32%, 5%, and 10%, P ¼ 0.03) (Table 2). Yet, there was no difference between groups regarding ocular complications and treatment with topical or systemic glucocorticoids.
Ocular Outcomes According to Treatment Groups (n ¼ 71 Eyes) Ocular outcomes of the 71 eyes with >1 month of follow-up after treatment onset are shown in Table 3. Time periods between baseline
Hoogewoud et al
Treatment of Syphilitic Uveitis
Table 3. Ocular Outcomes According to Treatment Regimens (n ¼ 71 eyes) Characteristics
Group A (n [ 40)
Group B (n [ 21)
Group C (n [ 10)
Total (n [ 71)
P Value
1-wk assessment* Median time between baseline and 1-wk examinations, days (IQR) Clinical outcome Worse Stable Improvement Recovery 1-mo assessment Median time between baseline and 1-mo examinations, days (IQR) Recovery Isolated papillitis Posterior uveitis (with the exclusion of isolated papillitis) or panuveitis Isolated anterior and/or intermediate uveitis Last follow-up Median time between baseline examination and last follow-up, days (IQR) Recovery (among patients with persistent inflammation at 1 mo) Overall recovery Isolated papillitis Posterior uveitis or panuveitis (with the exclusion of isolated papillitis) Isolated anterior and/or intermediate uveitis BCVA at last follow-up <2/10 2/10e5/10 >5/10 Median visual acuity improvement at last follow-up, logMAR (IQR)
36 7 [6e8.5]
19 5 [5e7]
6 6 [5e7]
61 7 [5e8]
0.16
(17) (19) (61) (3) 40 28.5 [25e36]
(5) (16) (58) (21) 21 27 [25e40]
0 2 (33) 4 (67) 0 10 26 [22e39]
(11) (20) (61) (8) 71 28 [25e37]
0.28 0.92
22 (55) 6 (60) 14 (54)
18 (86) 1 (50) 10 (83)
6 (60) 0 5 (83)
46 (65) 7 (47) 29 (65)
0.06 0.26 0.17
6 7 22 1
1 3 11 4
7 12 37 5
2 (50)
7 (100)
1 (100)
10 (83)
0.15
40 88.5 [72e126]
21 83 [78e92]
10 97 [91e97]
71 88 [74e102]
0.35
9/11 (82)
2/3 (66)
3/3 (100)
14/17 (82)
1.0
31/40 (78) 7/10 (70) 22/26 (85)
20/21 (95) 1/2 (50) 12/12 (100)
9/10 (90) 3/3 (100) 5/6 (83)
60/71 (85) 11/15 (73) 39/44 (87)
0.20 0.54 0.39
2/4 (50)
7/7 (100)
1/1 (100)
10/12 (83)
0.15
3 (8) 9 (24) 26 (68) 0.3 [0.6 to 0]
0 1 (5) 20 (95) 0.1 [0.5 to 0.05]
0 2 (20) 8 (80) 0.3 [0.7 to 0]
3 (4) 12 (17) 54 (78) 0.3 [0.56 to 0.02]
0.18 0.69
Data are presented as no. (%) unless otherwise specified. BCVA ¼ best-corrected visual acuity; IQR ¼ interquartile range; logMAR ¼ logarithm of the minimum angle of resolution. Treatment regimens: group A, 14 days of IV penicillin G; group B, 5 days of IV penicillin G followed by ceftriaxone or BPG; group C, ceftriaxone or BPG exclusively. *Missing data for 10 eyes.
and subsequent (i.e., 1-week, 1-month, and last follow-up) evaluations were comparable among treatment groups. Of the 61 eyes that were evaluated at 1 week, there was no difference between treatment regimens regarding the rates of early improvement at 1 week. Seven eyes (11%) had worse ophthalmological findings at 1 week, but none had general symptoms suggestive of JarischeHerxheimer’s reaction (otherwise reported in 3 patients). Of note, none of these 7 eyes recovered at the 1-month follow-up visit. Overall, complete recovery was reported in 46 eyes (65%) at 1 month, with a trend toward higher recovery rates in eyes from group B (55% vs. 86% vs. 60%, P ¼ 0.06), without difference between those treated with ceftriaxone or BPG (recovery at 1 month: 15/18 and 3/3 eyes, respectively, P ¼ 1.0). Although not reaching statistical significance (P ¼ 0.07), recovery at 1 month was reduced in eyes with isolated papillitis. Forty-four eyes (62%) had additional follow-up after the 1-month visit. At the last follow-up visit (median, 88 days; interquartile range, 74e102), 14 of 17 eyes (82%) with persistent inflammation at the 1-month visit finally recovered (missing data for 18/35 eyes without recovery at the 1-month visit). At the last
follow-up, overall recovery was reported in 60 eyes (85%), without any difference between treatment groups (P ¼ 0.20) and baseline ophthalmological findings. All 3 patients (5 eyes) with a 4-fold decrease of VDRL titers at 3 months had concomitant ophthalmological recovery. At last follow-up, BCVA was stable or had improved in all eyes, again without difference between treatment groups (P ¼ 0.18). Overall, 54 eyes (78%) reached >5/10 of BCVA at last follow-up visit.
Prognostic Factors of Treatment Success at 1 Month after Treatment Onset Prognostic factors of treatment success are shown in Table 4. In univariate analysis, 2/10 of baseline BCVA (crude OR, 3.4; 95% CI, 1.2e9.7; P ¼ 0.02) and early improvement at 1 week (crude OR, 14; 95% CI, 3.8e51.6; P ¼ 0.001) were significantly associated with recovery at 1 month, but age, sex, HIV status, delay between onset of visual symptoms and baseline ophthalmological evaluation, uveitides subtypes, high ocular inflammation, VDRL titer, abnormal lumbar puncture, type of
5
Ophthalmology Volume -, Number -, Month 2017 Table 4. Predictors of Ophthalmological Recovery 1 Month after Treatment Onset Characteristics Total Age <50 y 50 y Sex Male Female HIV status Negative Positive Duration between onset of ocular symptoms and baseline evaluation (days) 21 >21 Baseline visual acuity <2/10 2/10 Anatomic location of ocular inflammation Anterior and/or intermediate uveitis Isolated papillitis Posterior or panuveitis (with the exclusion of isolated papillitis) High ocular inflammation No Yes VDRL titer (UI/ml) 64 >64 Abnormal LP No Yes Treatment A B and C Oral prednisone (not preceded by methylprednisolone pulses) No Yes Methylprednisolone pulses (followed by oral prednisone) No Yes Glucocorticoid eye drops No Yes Periocular dexamethasone injections No Yes 1-wk evaluation Worse or stable Improvement or recovery
Month 1 Recovery
Crude OR 95% CI
P Value
aOR* 95% CI
cRRy
P Value
46/71 (65) 0.06 17 (52) 29 (76)
1 1.0 [0.9e1.1]
42 (66) 4 (57)
1 0.7 [0.2e3.4]
22 (67) 24 (63)
1 0.8 [0.3e2.3]
24 (75) 22 (56)
1 0.4 [0.2e1.2]
11 (46) 35 (75)
1 3.4 [1.2e9.7]
10 (83) 7 (47) 29 (66)
1 0.2 [0.1e1.1] 0.5 [0.1e2.0]
33 (65) 13 (65)
1 1.0 [0.3e3.0]
23 (58) 23 (74)
1 2.1 [0.8e5.8]
8 (80) 34 (67)
1 0.5 [0.2e2.6]
22 (55) 24 (77)
1 2.8 [1.0e8.0]
0.70
0.81
0.14 0.02
0.73 1 1.4 [0.2e10.3]
1 1.2 [0.3e2.0]
1 4.3 [0.6e30.4]
1 1.5 [0.8e1.8]
0.07 0.26 0.98
0.21
0.49
0.06
0.14
0.3 38 (68) 8 (53)
1 0.5 [0.2e1.7] 0.04
46 (70) 0
1 NA
NA
NA
1 0.02 [0.005e0.3]
1 0.05 [0.02e0.6]
1 29.9 [4.3e201.0]
1 3.5 [2.3e3.8]
0.94 22 (65) 23 (64)
1 1.0 [0.4e2.6]
44 (70) 1 (14)
1 0.07 [0.01e0.6]
5 (26) 35 (83)
1 14.0 [3.8e51.6]
0.02
0.01
0.001
0.001
Data are presented as no. (%) unless otherwise specified. aOR ¼ adjusted odds ratio; CI ¼ confidence interval; cRR ¼ corrected risk ratio; HIV ¼ human immunodeficiency virus; LP ¼ lumbar puncture; NA ¼ not applicable; OR ¼ odds ratio; VDRL ¼ Venereal Disease Research Laboratory. *Backward stepwise analysis adjusted on treatment and baseline visual acuity. y Using the method of Zhang and Yu18 (see “Methods” section).
antibiotic regimen, and use of glucocorticoid eye drops or oral prednisone were not. Conversely, the periocular dexamethasone injections and methylprednisolone pulses were significantly associated with treatment failure. In the final multivariate model, after adjusting for the initial BCVA and the antimicrobial treatment regimen, clinical improvement at 1 week (cRR, 3.5;
6
95% CI, 2.3e3.8; P ¼ 0.001) was the only predictive factor of recovery at 1 month, whereas the use of periocular dexamethasone injections (cRR, 0.05; 95% CI, 0.02e0.6; P ¼ 0.01) and methylprednisolone pulses (no treatment success in patients who received pulses) negatively affected the outcomes of eyes.
Hoogewoud et al
Treatment of Syphilitic Uveitis
Discussion Although the incidence of SU has increased in the last decade,5 the literature dedicated to this topic remains scarce, with 143 patients reported in the meta-analysis published in 20109 and recent case series of 50 patients.19,20 To our knowledge, this work is the largest series of patients with SU reported to date and specifically studies factors associated with ophthalmological recovery. A stringent definition of cases of SU was applied, and expert ophthalmologists in the field of uveitis reviewed all cases. The study population’s baseline characteristics were in line with published data, with posterior uveitis and panuveitis being the most frequent features3,6,19e24 and anterior inflammation being almost exclusively nongranulomatous.20,23e25 Surprisingly, significantly higher inflammatory parameters were reported in eyes from HIVþ patients than in those from HIV patients. One might speculate that HIV infection could be a corollary of highrisk sex practices leading to multiple syphilis recontaminations and increased ocular inflammation. Yet, the level of immunosuppression also seemed to correlate with ocular inflammation, with higher vitreous haze grades and more posterior synechiae reported in eyes from patients with <200/mm3 CD4þ T lymphocytes (Table S2, available at www.aaojournal.org). In line with previous studies, HIV status did not influence visual prognosis.20,22,26 One-month evaluation was arbitrarily chosen as the main outcome measure because SU is an acute condition with usually a rapid improvement after onset of antibiotics. In univariate analysis, baseline BCVA 2/10 and early clinical improvement at 1 week were the only factors significantly associated with recovery at 1 month, whereas the use of periocular dexamethasone injections and methylprednisolone pulses was significantly correlated to treatment failure. In multivariate analysis, after adjusting for initial BCVA (the only baseline ophthalmological characteristic that differed between treatment groups) and treatment regimens (that was forced in the model to answer to the primary assumption of the work), the remaining independent prognostic factor of recovery at 1 month was early improvement at 1 week (cRR, 3.5; 95% CI, 2.3e3.8; P ¼ 0.001). Thus, rather than the antimicrobial drug used, early improvement seems to best predict ophthalmological recovery. These findings suggest that selected patients with early favorable clinical course after onset of penicillin G could be safely switched to an alternative antibiotic regimen. According to our data, the latter could rely on ceftriaxone (IV or intramuscularly 2 g/day) which has a good CSF penetration and is easily completed in ambulatory care. By contrast, because all eyes with clinical worsening at early evaluation failed to recover at 1 month, it remains an open question whether these corresponding patients could benefit from reinforced antimicrobial therapy. Next, although eyes treated with periocular dexamethasone injections or methylprednisolone pulses were severely involved (i.e., with frequent panuveitides or low baseline visual acuity), our findings provide no evidence for the benefit of such treatments. Yet, because all
periocular dexamethasone injections were performed before (n ¼ 2 eyes) or within the first 6 days (n ¼ 5 eyes) after onset of antibiotics, the time frame between antimicrobial and anti-inflammatory treatments also might be a critical issue. Altogether, recovery rates of eyes were high (65% at 1 month and 85% at last follow-up). Thus, in line with previous studies,19,20 our data confirm that SU carries a good prognosis. Yet, isolated papillitides (otherwise described as “uveopapillitides”) might require close attention.22 As reported in brucellosis,27 in human leukocyte antigen B27eassociated uveitis,28 and more recently in SU,20 these tend to recover slowly (53% of recovery at 1 month in the present study). Nevertheless, it remains unclear whether this trend is due to a reduced or slow response to antibiotics or to a shear inflammatory process, and whether such eyes could benefit from treatment intensification. Conversely, further studies are warranted to determine whether 1-week improvement could correlate with ophthalmological findings at last follow-up. Study Limitations We acknowledge some limitations to our study, namely, baseline discrepancies between groups and loss for followup inherent to its retrospective design. Next, both departments of Ophthalmology were tertiary referral centers for the management of uveitides, which might have led to a selection bias. Yet, this work provides new data suggesting that 1-week improvement rather than the antimicrobial regimen best predicts ophthalmological recovery in SU, whereas there is no evidence for the benefit of periocular dexamethasone injections and methylprednisolone pulses. These findings are likely to give tools to physicians for effective and individualized management of patients with SU.
References 1. Krause RM. Syphilis during 1900-1910: similarities to present-day AIDS. Allergy Proc. 1991;12:127-132. 2. Fenton KA, Breban R, Vardavas R, et al. Infectious syphilis in high-income settings in the 21st century. Lancet Infect Dis. 2008;8:244-253. 3. Hughes G, Field N. The epidemiology of sexually transmitted infections in the UK: impact of behavior, services and interventions. Future Microbiol. 2015;10:35-51. 4. Butler NJ, Thorne JE. Current status of HIV infection and ocular disease. Curr Opin Ophthalmol. 2012;23: 517-522. 5. Tuddenham S, Ghanem KG. Ocular syphilis: opportunities to address important unanswered questions. Sex Transm Infect. 2016;92:563-565. 6. Mathew RG, Goh BT, Westcott MC. British Ocular Syphilis Study (BOSS): 2-year national surveillance study of intraocular inflammation secondary to ocular syphilis. Invest Ophthalmol Vis Sci. 2014;55:5394-5400. 7. Davis JL. Ocular syphilis. Curr Opin Ophthalmol. 2014;25: 513-518.
7
Ophthalmology Volume -, Number -, Month 2017 8. Workowski KA, Bolan GA, Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2015. MMWR Recomm Rep. 2015;64:1-137. 9. Amaratunge BC, Camuglia JE, Hall AJ. Syphilitic uveitis: a review of clinical manifestations and treatment outcomes of syphilitic uveitis in human immunodeficiency virus-positive and negative patients. Clin Experiment Ophthalmol. 2010;38: 68-74. 10. Liang Z, Chen Y-P, Yang C-S, et al. Meta-analysis of ceftriaxone compared with penicillin for the treatment of syphilis. Int J Antimicrob Agents. 2016;47:6-11. 11. Rolfs RT, Joesoef MR, Hendershot EF, et al. A randomized trial of enhanced therapy for early syphilis in patients with and without human immunodeficiency virus infection. The Syphilis and HIV Study Group. N Engl J Med. 1997;337:307-314. 12. Tanizaki R, Nishijima T, Aoki T, et al. High-dose oral amoxicillin plus probenecid is highly effective for syphilis in patients with HIV infection. Clin Infect Dis. 2015;61:177-183. 13. Riedner G, Rusizoka M, Todd J, et al. Single-dose azithromycin versus penicillin G benzathine for the treatment of early syphilis. N Engl J Med. 2005;353:1236-1244. 14. Ghanem KG, Erbelding EJ, Cheng WW, Rompalo AM. Doxycycline compared with benzathine penicillin for the treatment of early syphilis. Clin Infect Dis. 2006;42:e45-49. 15. Dowell ME, Ross PG, Musher DM, et al. Response of latent syphilis or neurosyphilis to ceftriaxone therapy in persons infected with human immunodeficiency virus. Am J Med. 1992;93:481-488. 16. Jabs DA, Nussenblatt RB, Rosenbaum JT, Standardization of Uveitis Nomenclature (SUN) Working Group. Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop. Am J Ophthalmol. 2005;140: 509-516. 17. Vanhaecke C, Grange P, Benhaddou N, et al. Clinical and biological characteristics of 40 patients with neurosyphilis and
18. 19. 20. 21. 22. 23. 24. 25. 26.
27. 28.
evaluation of Treponema pallidum nested polymerase chain reaction in cerebrospinal fluid samples. Clin Infect Dis. 2016;63:1180-1186. Zhang J, Yu KF. What’s the relative risk?: A method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998;280:1690-1691. Moradi A, Salek S, Daniel E, et al. Clinical features and incidence rates of ocular complications in patients with ocular syphilis. Am J Ophthalmol. 2015;159:334-343.e1. Fonollosa A, Martinez-Indart L, Artaraz J, et al. Clinical manifestations and outcomes of syphilis-associated uveitis in northern Spain. Ocul Immunol Inflamm. 2016;24:147-152. Ormerod LD, Puklin JE, Sobel JD. Syphilitic posterior uveitis: correlative findings and significance. Clin Infect Dis. 2001;32: 1661-1673. Tsuboi M, Nishijima T, Yashiro S, et al. Prognosis of ocular syphilis in patients infected with HIV in the antiretroviral therapy era. Sex Transm Infect. 2016;92:605-610. Yang P, Zhang N, Li F, et al. Ocular manifestations of syphilitic uveitis in Chinese patients. Retina Phila Pa. 2012;32: 1906-1914. Northey LC, Skalicky SE, Gurbaxani A, McCluskey PJ. Syphilitic uveitis and optic neuritis in Sydney, Australia. Br J Ophthalmol. 2015;99:1215-1219. Anshu A, Cheng CL, Chee S-P. Syphilitic uveitis: an Asian perspective. Br J Ophthalmol. 2008;92:594-597. Eandi CM, Neri P, Adelman RA, et al. Acute syphilitic posterior placoid chorioretinitis: report of a case series and comprehensive review of the literature. Retina Phila Pa. 2012;32:1915-1941. Sungur GK, Hazirolan D, Gurbuz Y, et al. Ocular involvement in brucellosis. Can J Ophtalmol. 2009;44:598-601. Monheit BE, Read RW. Optic disk edema associated with sudden-onset anterior uveitis. Am J Ophthalmol. 2005;140: 733-735.
Footnotes and Financial Disclosures Originally received: March 23, 2017. Final revision: June 2, 2017. Accepted: June 2, 2017. Available online: ---.
12
Present address: Department of Internal Medicine, Hôpital St. Louis, Paris, France.
Manuscript no. 2017-629.
1
Department of Ophthalmology, National Referral Center for Rare Ocular Diseases, Hôpital Cochin, Paris, France.
2
Department of Internal Medicine, National Referral Center for Rare Systemic Autoimmune Diseases, Hôpital Cochin, Paris, France.
3
IAME, UMR 1137, INSERM, Hôpital Bichat, Paris, France.
4
Department of Infectious Diseases, Hôpital Bichat-Claude Bernard, Paris, France.
5
Necker-Pasteur Center for Infectious Diseases and Tropical Medicine, Hôpital Necker Enfants Malades, Paris, France.
6
Department of Infectious Diseases, Hôpital Cochin, Paris, France.
7
National Referral Center for Syphilis, Hôpital Cochin, APHP, Paris, France.
8
Department of Internal Medicine, Centre Hospitalier National des QuinzeVingt, Paris, France.
9
Department of Dermatology, Hôpital Cochin, Paris, France.
10
Department of Ophthalmology IV, Centre Hospitalier National des Quinze-Vingt, Paris, France.
11
DHU ViewMaintain, Sorbonne-Pierre et Marie Curie University, Paris, France.
8
Presented at: the 17èmes Journées Nationales d’Infectiologie, Lille, France, June 7e9, 2016; 73ème Congrès de la Société Nationale Française de Médecine Interne, Lille, France, June 29 to July 1, 2016; Journées Dermatologiques de Paris, December 6e10, 2016, Paris, France; and European Congress of Clinical Microbiology and Infectious Diseases, 27th Annual Congress, April 22e22, 2017, Vienna, Austria. *F.H. and L.F. contributed equally to this work. Financial Disclosure(s): The author(s) have made the following disclosure(s): F.H.: Support Prof Max Cloëtta Foundation, Zurich, Switzerland, which had no role in the design or conduct of this research. P.L.: Consultant Pfizer. C.C.: Grants Agence Nationale de la Recherche and Contrat pour la Recherche Clinique. C.L.J.: Payment for lectures Adalimumab (AbbVie). O.L.: Board Member Astellas, MSD, Pfizer; Consultant Gilead Sciences, Novartis. J.-A.S.: Consultant Sanofi, GenSight Biologics, Pixium Vision; Grants Banque Publique d’Investissement, ERC Synergy “HELMHOLTZ,” Foundation Fighting Blindness, and LabEx LIFESENSES (ANR-10LABX-65); Stock GenSight Biologics, Chronocam, Chronolife, Pixium Vision, Tilak Healthcare, Sparing Vision.
Hoogewoud et al
Treatment of Syphilitic Uveitis
M.G.: Expenses paid e Laboratoire Français du Fractionnement et des Biotechnologies, Octapharma. Author Contributions: Conception and design: Hoogewoud, Loubet, Le Jeunne, Monnet, Lortholary, Dupin, Brezin, Salah, Groh Data collection: Hoogewoud, Frumholtz, Blanche, Lebeaux, Benhaddou, Sedira, Coutte, Vanhaecke, Launay, Heron, Monnet, Lortholary, Sahel, Dupin, Brezin, Errera, Salah, Groh Analysis and interpretation: Hoogewoud, Frumholtz, Loubet, Charlier, Blanche, Lebeaux, Monnet, Lortholary, Brezin, Errera, Salah, Groh Obtained funding: Not applicable Overall responsibility: Hoogewoud, Frumholtz, Loubet, Charlier, Lebeaux, Benhaddou, Sedira, Coutte, Vanhaecke, Launay, Le Jeunne, Heron, Monnet, Lortholary, Sahel, Dupin, Brezin, Errera, Salah, Groh
Abbreviations and Acronyms: ARV ¼ antiretroviral; BCVA ¼ best-corrected visual acuity; BPG ¼ benzathine penicillin G; CI ¼ confidence interval; cRR ¼ corrected risk ratio; CSF ¼ cerebrospinal fluid; HIV ¼ human immunodeficiency virus; IV ¼ intravenous; MSM ¼ men having sex with men; OR ¼ odds ratio; STD ¼ sexually transmitted disease; SU ¼ syphilitic uveitis; SUN ¼ Standardization of Uveitis Nomenclature; TPHA ¼ Treponema pallidum hemagglutination assay; VDRL ¼ Venereal Disease Research Laboratory. Correspondence: Matthieu Groh, MD, Department of Internal Medicine, Hôpital St Louis, 1 Avenue Claude Vellefaux, 75010 Paris, France. E-mail: matthieu.groh@ aphp.fr.
9
Although syphilis was highly endemic in the preantibiotic era,1 it nearly disappeared in high-income countries after the discovery of penicillin and the implementation of screening and treatment programs.2 However, similar to other sexually transmitted diseases (STDs), a resurgence of syphilis recently has been reported, particularly among human immunodeficiency virus (HIV)infected patients and in the men having sex with men (MSM) population.3 Between 2001 and 2014, the annual incidence of primary and secondary syphilis in the United States increased from 2.1 to 6.3 cases per 100 000 inhabitants.4 Likewise, an outbreak of ocular syphilis occurred recently in the United States.5 Ocular complications (0.6%e15% of cases)6 are more frequent during the secondary and early latent phases of syphilis, but, similar to neurosyphilis, they have been reported in all stages of the disease.7 Although any segment of the eye can be involved, panuveitis and posterior uveitis are the most frequent ocular findings.6 Besides superficial primary syphilis (e.g., canker of the lid or conjunctiva), it has long been hypothesized that syphilitic uveitis (SU) was closely related to neurosyphilis, and current treatment guidelines recommend treating it as ª 2017 by the American Academy of Ophthalmology Published by Elsevier Inc.
such.8 Thus, treatment with intravenous (IV) penicillin G (6 mUI 3 or 4 times per day for 10e14 days) is demanding and may extend the length of hospital stay. Moreover, cases of treatment failure have been reported.9 Although ceftriaxone,10 oral amoxicillin plus 11,12 azithromycin,13 and doxycycline14 have probenecid, been shown to be effective for early stages of syphilis, there are few data regarding the safety and efficacy of alternative treatments for neurosyphilis and SU.15 This study aims to identify predictors of treatment success of SU and to evaluate the efficacy of alternative therapies (e.g., ceftriaxone and benzathine penicillin G [BPG]).
Methods Patients A retrospective multicenter study was conducted from January 2003 to April 2016 in 2 tertiary ophthalmic centers (Cochin and the Quinze-Vingts National Ophthalmology Hospital). Adult patients treated for SU were identified using the medical information system databases. All diagnoses of SU were reassessed by the investigators (F.H., S.S.). Syphilitic uveitis was defined by the http://dx.doi.org/10.1016/j.ophtha.2017.06.003 ISSN 0161-6420/17
1
Ophthalmology Volume -, Number -, Month 2017 presence of ocular inflammation compatible with the diagnosis of syphilis, the positivity of both serum Treponema pallidum hemagglutination assay (TPHA) and Venereal Disease Research Laboratory (VDRL) tests, and the exclusion of alternate diagnoses. Patients with optic neuritis (without any sign of intraocular inflammation) and congenital syphilis were not included. This study was approved by the local ethics committee.
improvement was defined by a 2-step decrease of the levels of both anterior chamber (AC) cells and vitreous haze inflammation, and by size reduction in chorioretinal lesions. Recovery (main outcome measure) was defined as the resolution of inflammation in all anatomic ocular structures. Ocular complications included posterior synechiae, ocular hypertension, glaucoma, macular edema, serous retinal detachment, vasculitis, and optic atrophy.
Baseline Measurements Data collection was performed using a standardized anonymous form. Baseline demographic, clinical, and paraclinical parameters were retrieved, including age, gender, socioeconomic status, sexual orientation, history of syphilis and other STDs, dermatologic and systemic clinical symptoms due to syphilis, HIV status, TPHA and VDRL levels at baseline and 3 months after onset of treatment, type, dose, and route of administration of anti-Treponema drugs, and use of glucocorticoids (topical or systemic). Ophthalmological findings included type and duration of ocular symptoms at presentation, best-corrected visual acuity (BCVA), slit-lamp examination (including grading of anterior segment and vitreal inflammation according to the Standardization of Uveitis Nomenclature [SUN] Guidelines),16 laser-flare meter (when available), intraocular pressure measurement, and fundus examination. The SUN guidelines were used to classify uveitides subtypes.16 When available, cerebrospinal fluid (CSF) data also were collected (routine biochemical, cytologic analyses, and microbiological testing [i.e., VDRL, Fluorescent Treponemal Assay absorption test, and polymerase chain reaction]).17 Lumbar puncture was considered abnormal in case of pleocytosis (>10 cells/mm3), hyperproteinorrachia (>0.4 g/L), or positive CSF microbiological test results.
Outcomes Data from ophthalmological examinations at 1 week (84 days), 1 month (3012 days), and last follow-up after antimicrobial treatment onset were collected. Last follow-up was defined as month 1 examination for patients whose follow-up ended at 1 month and as the last date of follow-up for those who had additional follow-up visits. As defined by the SUN guidelines,16
Statistical Analyses Characteristics of patients and eyes are reported as numbers and percentages for categoric variables and as mean (standard deviation) or median (interquartile range) for continuous variables. For univariate analysis, the Wilcoxon rank-sum or KruskaleWallis tests, and Fisher exact test were used as appropriate. Missing data for each variable were excluded from the denominator. Patient subsets were differentiated on the basis of their HIV status and treatment regimens (“close to standard of care”: group A, 14 days of IV penicillin G; “pragmatic approach”: group B, 5 days of IV penicillin G followed by ceftriaxone or BPG; “nonvalidated treatment regimens”: group C, ceftriaxone or BPG; and group D, oral doxycycline). High ocular inflammation was defined as a composite variable that was considered positive if anterior chamber cell or vitreous haze grades were >2 (when appropriate, taking into account the anatomic location of ocular inflammation). Full details of the variables entered in model building are available in Supplementary material (available at www.aaojournal.org). Factors associated with recovery at 1 month were identified in a backward stepwise logistic regression model considering “recovery at 1 month” (positive/negative) as the dependent variable. All variables with a P value <0.2 in univariate analysis were tested in the model. To exclude variables with high collinearity from the multivariate analysis, variables significantly associated with recovery at 1 month in univariate analysis were assessed for bivariate correlation using Spearman’s test. The fit of the model was tested using the HosmereLemeshow goodness of fit test. Results are expressed as crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs). Because the prevalence of our outcome (“recovery at 1 month”) was frequent (>10%), the ORs were corrected by the Zhang and Yu method18 to avoid
Figure 1. Flowchart showing the search strategy and inclusion/exclusion criteria for the study population.
2
Hoogewoud et al
Treatment of Syphilitic Uveitis
Table 1. Baseline Ophthalmological Findings According to Human Immunodeficiency Virus Status (n ¼ 95 eyes) Variable BCVA <2/10 2/10e5/10 >5/10 Median IOP, mmHg (IQR) Anatomic location of ocular inflammation Isolated anterior uveitis Isolated intermediate uveitis Isolated posterior uveitis Isolated papillitis Retinitis (or chorioretinitis) with papillitis Retinitis (or chorioretinitis) without papillitis Panuveitis AC cells >0.5 Hypopyon Median flare, photons/ms (IQR) Vitreous haze >0.5 Ocular complications Posterior synechiae Serous retinal detachment Cystoid macular edema Glaucoma Retinal vasculitis (fundoscopy)
HIVD (n [ 46)
HIVL (n [ 49)
Total (n [ 95)
P Value
15 (33) 14 (30) 17 (37) 12 [10e14]
9 (18) 19 (39) 21 (43) 13 [12e15]
24 (25) 33 (35) 38 (40) 13 [11e15]
0.29 0.03
7 1 12 10 2
(15) (2) (26) (83) (17)
7 2 26 10 12
(14) (4) (53) (38) (46)
14 3 38 20 14
(15) (3) (40) (53) (37)
1.0 1.0 0.01 -
0
4 (15)
4 (11)
-
26 (57)
14 (29)
40 (42)
0.007
33 (72) 3 (7) 48 [18e88]
16 (33) 1 (2) 16 [8e33]
49 (52) 4 (4) 33 [11e65]
<103 0.35 0.01
29 (63)
18 (37)
47 (49)
0.02
14 (30) 2 (4) 3 (7) 0 7 (15)
4 (8) 2 (4) 0 0 6 (12)
18 (19) 4 (4) 3 (3) 0 13 (14)
0.008 1.0 0.10 0.77
Data are presented as no. (%) unless otherwise specified. AC ¼ anterior chamber; BCVA ¼ best-corrected visual acuity; HIV ¼ human immunodeficiency virus; IOP ¼ intraocular pressure; IQR ¼ interquartile range.
overestimation of the risk ratio. Corrected risk ratios (cRRs) are thus displayed as final results of the multivariate logistic regression model. P values <0.05 were considered statistically significant. All analyses were conducted using STATA software (V12, 1996-2014 StataCorp LP, College Station, TX).
Results Patients’ Identification Screening from the medical information database identified 98 patients with ocular inflammation and both serum-positive VDRL and TPHA. After exclusion of 32 patients, 66 patients (95 eyes) were included in the study (Fig 1), among whom 10 (15%) had received anti-Treponema drugs (oral azithromycin, n ¼ 6; 1 BPG injection, n ¼ 1; oral sulfadiazine, n ¼ 1; oral doxycycline, n ¼ 1) as probabilistic treatment of chorioretinitis or as treatment of secondary cutaneous syphilis for less than 7 days before baseline ophthalmological evaluation.
Baseline Characteristics According to Human Immunodeficiency Virus Status Baseline demographic and clinical characteristics of the 66 patients (61 male, among whom 33/48 [69%] were MSM) are shown in Table S1 (available at www.aaojournal.org). Thirty-one patients (47%) were seropositive for HIV (median CD4þ T Lymphocytes [CD4] count: 313/mm3; range, 30e1750/mm3), among whom
16 of 30 (53%, unknown antiretroviral (ARV) treatment for 1 patient) received ARV therapy at the time of SU diagnosis, with an undetectable viral load in 10 of 25 patients (40%). Patients positive for HIV were younger (mean age, 44 vs. 53 years, P ¼ 0.002), more frequently MSM (87% vs. 52%, P ¼ 0.01), and more frequently reported other concomitant STDs (24% vs. 3%, P ¼ 0.02). Lumbar puncture was performed in 52 patients (79%), among whom 41 (79%) had abnormal CSF findings, without differences between HIV serostatus (P ¼ 0.73). Abnormal CSF findings did not correlate with a specific subtype of uveitis. Hyperproteinorrachia was the most frequent abnormal feature, reported in 67% of all cases. The Fluorescent Treponemal Assay absorption was the most sensitive microbiological CSF test, with positive results in 17 of 19 cases (89%), whereas VDRL and polymerase chain reaction were positive in 18 of 46 cases (39%) and 1 of 16 cases (6%), respectively. Baseline ophthalmological findings according to HIV status are shown in Table 1. Figure S1 (available at www.aaojournal.org) provides iconographic features of uveitides that were encountered. At disease onset, 29 patients (44%) were referred with bilateral ocular involvement. Among the 95 eyes (right eye, n ¼ 50, left eye, n ¼ 45) involved, 46 were from HIVþ patients and 49 were from HIV patients. Among the 33 eyes (35%) with keratic precipitates, only 2 (6%) were granulomatous. There was no difference in baseline BCVA with regard to HIV status. Panuveitis was the most frequent type of uveitis in HIVþ patients (57% vs. 29%, P ¼ 0.007), whereas posterior uveitis was more frequent in HIV patients (26% vs. 53%, P ¼ 0.01). Inflammatory parameters were significantly higher in HIVþ patients’ eyes, with
3
Ophthalmology Volume -, Number -, Month 2017 Table 2. Baseline Ophthalmological Findings According to Treatment Regimens (n¼71 eyes) Characteristics Visual acuity <2/10 2/10e5/10 >5/10 Median IOP, mmHg (IQR) Anatomic location of ocular inflammation Isolated anterior uveitis Isolated intermediate uveitis Isolated posterior uveitis Isolated papillitis Retinitis (or chorioretinitis) with papillitis Retinitis (or chorioretinitis) without papillitis Panuveitis Crystalline lens Clear Cataract Pseudophakia AC cells [0.5e2] [3e4] Hypopyon Median flare, photons/ms (IQR) Vitreous haze [0.5e2] [3e4] Ocular complications Ocular hypertension Posterior synechiae Serous retinal detachment Cystoid macular edema Glaucoma Retinal vasculitis (fundoscopy) Prior treatment with anti-treponemal drugs Topical glucocorticoids Eye drops Periocular dexamethasone injections Systemic glucocorticoids Methylprednisolone pulses (followed by oral prednisone) Oral prednisone (not preceded by methylprednisolone pulses)
Group A (n [ 40) Group B (n [ 21) Group C (n [ 10) Total (n [ 71) P Value 16 (40) 9 (23) 15 (37) 13 [11e15]
1 (5) 8 (38) 12 (57) 12.5 [10.5e14]
4 (10) 0 23 (58) 11/23 (48) 10/23 (43) 2/23 (9) 13 (33)
1 (10) 6 (60) 3 (30) 11.5 [10e17]
18 (25) 23 (32) 30 (43) 13 [11e15]
5 2 3 2/3 1/3
(23) (10) (14) (67) (33) 0 11 (52)
1 (10) 0 3 (30) 3/3 (100) 0 0 6 (60)
10 2 29 16 11 2 30
21 (100) 0 0
10 (100) 0 0
70 (99) 1 (1) 0
12 (57) 6 (29) 2 (10) 23.5 [14e64.8]
1 (10) 4 (40) 0 162 .2 [7.7e185]
26 (37) 17 (24) 4 (6) 33 [13e64.8]
15 (41) 2 (5)
12 (57) 3 (14)
4 (50) 1 (13)
31 (47) 6 (9)
0.40 0.19 0.003 0.18 1.0 0.02 0.79 0.60 0.31 -
0 8 (20) 4 (10) 0 0 4 (10) 5 (12) 19 (49) 17 (44) 5 (13)
0 3 (14) 0 1 (5) 0 4 (19) 4 (19) 12 (57) 14 (67) 1 (5)
0 2 (20) 0 0 0 2 (20) 1 (10) 6 (60) 5 (50) 1 (10)
0 13 (18) 4 (6) 1 (2) 0 10 (14) 10 (14) 37 (53) 36 (51) 7 (10)
0.90 0.44 0.47 0.50 0.90 0.75 0.21 0.75
3 (8) 10 (25)
1 (5) 2 (10)
1 (10) 3 (30)
5 (7) 15 (21)
1.0 0.26
39 (98) 1 (2) 0 13 (33) 7 (18) 2 (5) 37.9 [13.4e57.4]
(14) (3) (41) (55) (38) (7) (42)
0.01 0.38
Data are presented as no. (%) unless otherwise specified. Treatment regimens: group A, 14 days of IV penicillin G; group B, 5 days of IV penicillin G followed by ceftriaxone or BPG; group C, ceftriaxone or BPG exclusively. AC ¼ anterior chamber; IOP ¼ intraocular pressure; IQR ¼ interquartile range.
>0.5 AC cells in 72% versus 33% (P < 0.001), >0.5 vitreous haze in 63% versus 37% (P ¼ 0.02), median flare being 48 versus 16 photons/ms (P ¼ 0.01), and posterior synechiae in 30% versus 8% (P ¼ 0.008) in HIVþ and HIV patients’ eyes, respectively. Among HIV-infected patients, there was a trend toward higher inflammatory parameters in patients with <200/mm3 CD4þ T lymphocytes (Table S2, available at www.aaojournal.org).
Baseline Characteristics According to Treatment Regimens Of the 66 patients enrolled in the survey, 16 were subsequently excluded from further analyses of ocular outcomes (Fig 1). Patients were treated with 18 different regimens that could be merged in 4 different groups (group A, n ¼ 27; group B, n ¼ 14; group C, n ¼ 8; group D, n ¼ 1) (Table S3, available at www.aaojournal.org), as
4
described previously. Patients’ baseline demographic and ophthalmological characteristics according to treatment regimens (49 patients, 71 eyes) are shown in Table S4 (available at www.aaojournal.org) and Table 2, respectively (data from the unique patient from group D are not shown). Eyes belonging to group A patients had poorer baseline BCVA (<2/10 BCVA in 40%, 5%, and 10% of all 3 groups, respectively, P ¼ 0.01) and more frequent posterior uveitis (32%, 5%, and 10%, P ¼ 0.03) (Table 2). Yet, there was no difference between groups regarding ocular complications and treatment with topical or systemic glucocorticoids.
Ocular Outcomes According to Treatment Groups (n ¼ 71 Eyes) Ocular outcomes of the 71 eyes with >1 month of follow-up after treatment onset are shown in Table 3. Time periods between baseline
Hoogewoud et al
Treatment of Syphilitic Uveitis
Table 3. Ocular Outcomes According to Treatment Regimens (n ¼ 71 eyes) Characteristics
Group A (n [ 40)
Group B (n [ 21)
Group C (n [ 10)
Total (n [ 71)
P Value
1-wk assessment* Median time between baseline and 1-wk examinations, days (IQR) Clinical outcome Worse Stable Improvement Recovery 1-mo assessment Median time between baseline and 1-mo examinations, days (IQR) Recovery Isolated papillitis Posterior uveitis (with the exclusion of isolated papillitis) or panuveitis Isolated anterior and/or intermediate uveitis Last follow-up Median time between baseline examination and last follow-up, days (IQR) Recovery (among patients with persistent inflammation at 1 mo) Overall recovery Isolated papillitis Posterior uveitis or panuveitis (with the exclusion of isolated papillitis) Isolated anterior and/or intermediate uveitis BCVA at last follow-up <2/10 2/10e5/10 >5/10 Median visual acuity improvement at last follow-up, logMAR (IQR)
36 7 [6e8.5]
19 5 [5e7]
6 6 [5e7]
61 7 [5e8]
0.16
(17) (19) (61) (3) 40 28.5 [25e36]
(5) (16) (58) (21) 21 27 [25e40]
0 2 (33) 4 (67) 0 10 26 [22e39]
(11) (20) (61) (8) 71 28 [25e37]
0.28 0.92
22 (55) 6 (60) 14 (54)
18 (86) 1 (50) 10 (83)
6 (60) 0 5 (83)
46 (65) 7 (47) 29 (65)
0.06 0.26 0.17
6 7 22 1
1 3 11 4
7 12 37 5
2 (50)
7 (100)
1 (100)
10 (83)
0.15
40 88.5 [72e126]
21 83 [78e92]
10 97 [91e97]
71 88 [74e102]
0.35
9/11 (82)
2/3 (66)
3/3 (100)
14/17 (82)
1.0
31/40 (78) 7/10 (70) 22/26 (85)
20/21 (95) 1/2 (50) 12/12 (100)
9/10 (90) 3/3 (100) 5/6 (83)
60/71 (85) 11/15 (73) 39/44 (87)
0.20 0.54 0.39
2/4 (50)
7/7 (100)
1/1 (100)
10/12 (83)
0.15
3 (8) 9 (24) 26 (68) 0.3 [0.6 to 0]
0 1 (5) 20 (95) 0.1 [0.5 to 0.05]
0 2 (20) 8 (80) 0.3 [0.7 to 0]
3 (4) 12 (17) 54 (78) 0.3 [0.56 to 0.02]
0.18 0.69
Data are presented as no. (%) unless otherwise specified. BCVA ¼ best-corrected visual acuity; IQR ¼ interquartile range; logMAR ¼ logarithm of the minimum angle of resolution. Treatment regimens: group A, 14 days of IV penicillin G; group B, 5 days of IV penicillin G followed by ceftriaxone or BPG; group C, ceftriaxone or BPG exclusively. *Missing data for 10 eyes.
and subsequent (i.e., 1-week, 1-month, and last follow-up) evaluations were comparable among treatment groups. Of the 61 eyes that were evaluated at 1 week, there was no difference between treatment regimens regarding the rates of early improvement at 1 week. Seven eyes (11%) had worse ophthalmological findings at 1 week, but none had general symptoms suggestive of JarischeHerxheimer’s reaction (otherwise reported in 3 patients). Of note, none of these 7 eyes recovered at the 1-month follow-up visit. Overall, complete recovery was reported in 46 eyes (65%) at 1 month, with a trend toward higher recovery rates in eyes from group B (55% vs. 86% vs. 60%, P ¼ 0.06), without difference between those treated with ceftriaxone or BPG (recovery at 1 month: 15/18 and 3/3 eyes, respectively, P ¼ 1.0). Although not reaching statistical significance (P ¼ 0.07), recovery at 1 month was reduced in eyes with isolated papillitis. Forty-four eyes (62%) had additional follow-up after the 1-month visit. At the last follow-up visit (median, 88 days; interquartile range, 74e102), 14 of 17 eyes (82%) with persistent inflammation at the 1-month visit finally recovered (missing data for 18/35 eyes without recovery at the 1-month visit). At the last
follow-up, overall recovery was reported in 60 eyes (85%), without any difference between treatment groups (P ¼ 0.20) and baseline ophthalmological findings. All 3 patients (5 eyes) with a 4-fold decrease of VDRL titers at 3 months had concomitant ophthalmological recovery. At last follow-up, BCVA was stable or had improved in all eyes, again without difference between treatment groups (P ¼ 0.18). Overall, 54 eyes (78%) reached >5/10 of BCVA at last follow-up visit.
Prognostic Factors of Treatment Success at 1 Month after Treatment Onset Prognostic factors of treatment success are shown in Table 4. In univariate analysis, 2/10 of baseline BCVA (crude OR, 3.4; 95% CI, 1.2e9.7; P ¼ 0.02) and early improvement at 1 week (crude OR, 14; 95% CI, 3.8e51.6; P ¼ 0.001) were significantly associated with recovery at 1 month, but age, sex, HIV status, delay between onset of visual symptoms and baseline ophthalmological evaluation, uveitides subtypes, high ocular inflammation, VDRL titer, abnormal lumbar puncture, type of
5
Ophthalmology Volume -, Number -, Month 2017 Table 4. Predictors of Ophthalmological Recovery 1 Month after Treatment Onset Characteristics Total Age <50 y 50 y Sex Male Female HIV status Negative Positive Duration between onset of ocular symptoms and baseline evaluation (days) 21 >21 Baseline visual acuity <2/10 2/10 Anatomic location of ocular inflammation Anterior and/or intermediate uveitis Isolated papillitis Posterior or panuveitis (with the exclusion of isolated papillitis) High ocular inflammation No Yes VDRL titer (UI/ml) 64 >64 Abnormal LP No Yes Treatment A B and C Oral prednisone (not preceded by methylprednisolone pulses) No Yes Methylprednisolone pulses (followed by oral prednisone) No Yes Glucocorticoid eye drops No Yes Periocular dexamethasone injections No Yes 1-wk evaluation Worse or stable Improvement or recovery
Month 1 Recovery
Crude OR 95% CI
P Value
aOR* 95% CI
cRRy
P Value
46/71 (65) 0.06 17 (52) 29 (76)
1 1.0 [0.9e1.1]
42 (66) 4 (57)
1 0.7 [0.2e3.4]
22 (67) 24 (63)
1 0.8 [0.3e2.3]
24 (75) 22 (56)
1 0.4 [0.2e1.2]
11 (46) 35 (75)
1 3.4 [1.2e9.7]
10 (83) 7 (47) 29 (66)
1 0.2 [0.1e1.1] 0.5 [0.1e2.0]
33 (65) 13 (65)
1 1.0 [0.3e3.0]
23 (58) 23 (74)
1 2.1 [0.8e5.8]
8 (80) 34 (67)
1 0.5 [0.2e2.6]
22 (55) 24 (77)
1 2.8 [1.0e8.0]
0.70
0.81
0.14 0.02
0.73 1 1.4 [0.2e10.3]
1 1.2 [0.3e2.0]
1 4.3 [0.6e30.4]
1 1.5 [0.8e1.8]
0.07 0.26 0.98
0.21
0.49
0.06
0.14
0.3 38 (68) 8 (53)
1 0.5 [0.2e1.7] 0.04
46 (70) 0
1 NA
NA
NA
1 0.02 [0.005e0.3]
1 0.05 [0.02e0.6]
1 29.9 [4.3e201.0]
1 3.5 [2.3e3.8]
0.94 22 (65) 23 (64)
1 1.0 [0.4e2.6]
44 (70) 1 (14)
1 0.07 [0.01e0.6]
5 (26) 35 (83)
1 14.0 [3.8e51.6]
0.02
0.01
0.001
0.001
Data are presented as no. (%) unless otherwise specified. aOR ¼ adjusted odds ratio; CI ¼ confidence interval; cRR ¼ corrected risk ratio; HIV ¼ human immunodeficiency virus; LP ¼ lumbar puncture; NA ¼ not applicable; OR ¼ odds ratio; VDRL ¼ Venereal Disease Research Laboratory. *Backward stepwise analysis adjusted on treatment and baseline visual acuity. y Using the method of Zhang and Yu18 (see “Methods” section).
antibiotic regimen, and use of glucocorticoid eye drops or oral prednisone were not. Conversely, the periocular dexamethasone injections and methylprednisolone pulses were significantly associated with treatment failure. In the final multivariate model, after adjusting for the initial BCVA and the antimicrobial treatment regimen, clinical improvement at 1 week (cRR, 3.5;
6
95% CI, 2.3e3.8; P ¼ 0.001) was the only predictive factor of recovery at 1 month, whereas the use of periocular dexamethasone injections (cRR, 0.05; 95% CI, 0.02e0.6; P ¼ 0.01) and methylprednisolone pulses (no treatment success in patients who received pulses) negatively affected the outcomes of eyes.
Hoogewoud et al
Treatment of Syphilitic Uveitis
Discussion Although the incidence of SU has increased in the last decade,5 the literature dedicated to this topic remains scarce, with 143 patients reported in the meta-analysis published in 20109 and recent case series of 50 patients.19,20 To our knowledge, this work is the largest series of patients with SU reported to date and specifically studies factors associated with ophthalmological recovery. A stringent definition of cases of SU was applied, and expert ophthalmologists in the field of uveitis reviewed all cases. The study population’s baseline characteristics were in line with published data, with posterior uveitis and panuveitis being the most frequent features3,6,19e24 and anterior inflammation being almost exclusively nongranulomatous.20,23e25 Surprisingly, significantly higher inflammatory parameters were reported in eyes from HIVþ patients than in those from HIV patients. One might speculate that HIV infection could be a corollary of highrisk sex practices leading to multiple syphilis recontaminations and increased ocular inflammation. Yet, the level of immunosuppression also seemed to correlate with ocular inflammation, with higher vitreous haze grades and more posterior synechiae reported in eyes from patients with <200/mm3 CD4þ T lymphocytes (Table S2, available at www.aaojournal.org). In line with previous studies, HIV status did not influence visual prognosis.20,22,26 One-month evaluation was arbitrarily chosen as the main outcome measure because SU is an acute condition with usually a rapid improvement after onset of antibiotics. In univariate analysis, baseline BCVA 2/10 and early clinical improvement at 1 week were the only factors significantly associated with recovery at 1 month, whereas the use of periocular dexamethasone injections and methylprednisolone pulses was significantly correlated to treatment failure. In multivariate analysis, after adjusting for initial BCVA (the only baseline ophthalmological characteristic that differed between treatment groups) and treatment regimens (that was forced in the model to answer to the primary assumption of the work), the remaining independent prognostic factor of recovery at 1 month was early improvement at 1 week (cRR, 3.5; 95% CI, 2.3e3.8; P ¼ 0.001). Thus, rather than the antimicrobial drug used, early improvement seems to best predict ophthalmological recovery. These findings suggest that selected patients with early favorable clinical course after onset of penicillin G could be safely switched to an alternative antibiotic regimen. According to our data, the latter could rely on ceftriaxone (IV or intramuscularly 2 g/day) which has a good CSF penetration and is easily completed in ambulatory care. By contrast, because all eyes with clinical worsening at early evaluation failed to recover at 1 month, it remains an open question whether these corresponding patients could benefit from reinforced antimicrobial therapy. Next, although eyes treated with periocular dexamethasone injections or methylprednisolone pulses were severely involved (i.e., with frequent panuveitides or low baseline visual acuity), our findings provide no evidence for the benefit of such treatments. Yet, because all
periocular dexamethasone injections were performed before (n ¼ 2 eyes) or within the first 6 days (n ¼ 5 eyes) after onset of antibiotics, the time frame between antimicrobial and anti-inflammatory treatments also might be a critical issue. Altogether, recovery rates of eyes were high (65% at 1 month and 85% at last follow-up). Thus, in line with previous studies,19,20 our data confirm that SU carries a good prognosis. Yet, isolated papillitides (otherwise described as “uveopapillitides”) might require close attention.22 As reported in brucellosis,27 in human leukocyte antigen B27eassociated uveitis,28 and more recently in SU,20 these tend to recover slowly (53% of recovery at 1 month in the present study). Nevertheless, it remains unclear whether this trend is due to a reduced or slow response to antibiotics or to a shear inflammatory process, and whether such eyes could benefit from treatment intensification. Conversely, further studies are warranted to determine whether 1-week improvement could correlate with ophthalmological findings at last follow-up. Study Limitations We acknowledge some limitations to our study, namely, baseline discrepancies between groups and loss for followup inherent to its retrospective design. Next, both departments of Ophthalmology were tertiary referral centers for the management of uveitides, which might have led to a selection bias. Yet, this work provides new data suggesting that 1-week improvement rather than the antimicrobial regimen best predicts ophthalmological recovery in SU, whereas there is no evidence for the benefit of periocular dexamethasone injections and methylprednisolone pulses. These findings are likely to give tools to physicians for effective and individualized management of patients with SU.
References 1. Krause RM. Syphilis during 1900-1910: similarities to present-day AIDS. Allergy Proc. 1991;12:127-132. 2. Fenton KA, Breban R, Vardavas R, et al. Infectious syphilis in high-income settings in the 21st century. Lancet Infect Dis. 2008;8:244-253. 3. Hughes G, Field N. The epidemiology of sexually transmitted infections in the UK: impact of behavior, services and interventions. Future Microbiol. 2015;10:35-51. 4. Butler NJ, Thorne JE. Current status of HIV infection and ocular disease. Curr Opin Ophthalmol. 2012;23: 517-522. 5. Tuddenham S, Ghanem KG. Ocular syphilis: opportunities to address important unanswered questions. Sex Transm Infect. 2016;92:563-565. 6. Mathew RG, Goh BT, Westcott MC. British Ocular Syphilis Study (BOSS): 2-year national surveillance study of intraocular inflammation secondary to ocular syphilis. Invest Ophthalmol Vis Sci. 2014;55:5394-5400. 7. Davis JL. Ocular syphilis. Curr Opin Ophthalmol. 2014;25: 513-518.
7
Ophthalmology Volume -, Number -, Month 2017 8. Workowski KA, Bolan GA, Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2015. MMWR Recomm Rep. 2015;64:1-137. 9. Amaratunge BC, Camuglia JE, Hall AJ. Syphilitic uveitis: a review of clinical manifestations and treatment outcomes of syphilitic uveitis in human immunodeficiency virus-positive and negative patients. Clin Experiment Ophthalmol. 2010;38: 68-74. 10. Liang Z, Chen Y-P, Yang C-S, et al. Meta-analysis of ceftriaxone compared with penicillin for the treatment of syphilis. Int J Antimicrob Agents. 2016;47:6-11. 11. Rolfs RT, Joesoef MR, Hendershot EF, et al. A randomized trial of enhanced therapy for early syphilis in patients with and without human immunodeficiency virus infection. The Syphilis and HIV Study Group. N Engl J Med. 1997;337:307-314. 12. Tanizaki R, Nishijima T, Aoki T, et al. High-dose oral amoxicillin plus probenecid is highly effective for syphilis in patients with HIV infection. Clin Infect Dis. 2015;61:177-183. 13. Riedner G, Rusizoka M, Todd J, et al. Single-dose azithromycin versus penicillin G benzathine for the treatment of early syphilis. N Engl J Med. 2005;353:1236-1244. 14. Ghanem KG, Erbelding EJ, Cheng WW, Rompalo AM. Doxycycline compared with benzathine penicillin for the treatment of early syphilis. Clin Infect Dis. 2006;42:e45-49. 15. Dowell ME, Ross PG, Musher DM, et al. Response of latent syphilis or neurosyphilis to ceftriaxone therapy in persons infected with human immunodeficiency virus. Am J Med. 1992;93:481-488. 16. Jabs DA, Nussenblatt RB, Rosenbaum JT, Standardization of Uveitis Nomenclature (SUN) Working Group. Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop. Am J Ophthalmol. 2005;140: 509-516. 17. Vanhaecke C, Grange P, Benhaddou N, et al. Clinical and biological characteristics of 40 patients with neurosyphilis and
18. 19. 20. 21. 22. 23. 24. 25. 26.
27. 28.
evaluation of Treponema pallidum nested polymerase chain reaction in cerebrospinal fluid samples. Clin Infect Dis. 2016;63:1180-1186. Zhang J, Yu KF. What’s the relative risk?: A method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998;280:1690-1691. Moradi A, Salek S, Daniel E, et al. Clinical features and incidence rates of ocular complications in patients with ocular syphilis. Am J Ophthalmol. 2015;159:334-343.e1. Fonollosa A, Martinez-Indart L, Artaraz J, et al. Clinical manifestations and outcomes of syphilis-associated uveitis in northern Spain. Ocul Immunol Inflamm. 2016;24:147-152. Ormerod LD, Puklin JE, Sobel JD. Syphilitic posterior uveitis: correlative findings and significance. Clin Infect Dis. 2001;32: 1661-1673. Tsuboi M, Nishijima T, Yashiro S, et al. Prognosis of ocular syphilis in patients infected with HIV in the antiretroviral therapy era. Sex Transm Infect. 2016;92:605-610. Yang P, Zhang N, Li F, et al. Ocular manifestations of syphilitic uveitis in Chinese patients. Retina Phila Pa. 2012;32: 1906-1914. Northey LC, Skalicky SE, Gurbaxani A, McCluskey PJ. Syphilitic uveitis and optic neuritis in Sydney, Australia. Br J Ophthalmol. 2015;99:1215-1219. Anshu A, Cheng CL, Chee S-P. Syphilitic uveitis: an Asian perspective. Br J Ophthalmol. 2008;92:594-597. Eandi CM, Neri P, Adelman RA, et al. Acute syphilitic posterior placoid chorioretinitis: report of a case series and comprehensive review of the literature. Retina Phila Pa. 2012;32:1915-1941. Sungur GK, Hazirolan D, Gurbuz Y, et al. Ocular involvement in brucellosis. Can J Ophtalmol. 2009;44:598-601. Monheit BE, Read RW. Optic disk edema associated with sudden-onset anterior uveitis. Am J Ophthalmol. 2005;140: 733-735.
Footnotes and Financial Disclosures Originally received: March 23, 2017. Final revision: June 2, 2017. Accepted: June 2, 2017. Available online: ---.
12
Present address: Department of Internal Medicine, Hôpital St. Louis, Paris, France.
Manuscript no. 2017-629.
1
Department of Ophthalmology, National Referral Center for Rare Ocular Diseases, Hôpital Cochin, Paris, France.
2
Department of Internal Medicine, National Referral Center for Rare Systemic Autoimmune Diseases, Hôpital Cochin, Paris, France.
3
IAME, UMR 1137, INSERM, Hôpital Bichat, Paris, France.
4
Department of Infectious Diseases, Hôpital Bichat-Claude Bernard, Paris, France.
5
Necker-Pasteur Center for Infectious Diseases and Tropical Medicine, Hôpital Necker Enfants Malades, Paris, France.
6
Department of Infectious Diseases, Hôpital Cochin, Paris, France.
7
National Referral Center for Syphilis, Hôpital Cochin, APHP, Paris, France.
8
Department of Internal Medicine, Centre Hospitalier National des QuinzeVingt, Paris, France.
9
Department of Dermatology, Hôpital Cochin, Paris, France.
10
Department of Ophthalmology IV, Centre Hospitalier National des Quinze-Vingt, Paris, France.
11
DHU ViewMaintain, Sorbonne-Pierre et Marie Curie University, Paris, France.
8
Presented at: the 17èmes Journées Nationales d’Infectiologie, Lille, France, June 7e9, 2016; 73ème Congrès de la Société Nationale Française de Médecine Interne, Lille, France, June 29 to July 1, 2016; Journées Dermatologiques de Paris, December 6e10, 2016, Paris, France; and European Congress of Clinical Microbiology and Infectious Diseases, 27th Annual Congress, April 22e22, 2017, Vienna, Austria. *F.H. and L.F. contributed equally to this work. Financial Disclosure(s): The author(s) have made the following disclosure(s): F.H.: Support Prof Max Cloëtta Foundation, Zurich, Switzerland, which had no role in the design or conduct of this research. P.L.: Consultant Pfizer. C.C.: Grants Agence Nationale de la Recherche and Contrat pour la Recherche Clinique. C.L.J.: Payment for lectures Adalimumab (AbbVie). O.L.: Board Member Astellas, MSD, Pfizer; Consultant Gilead Sciences, Novartis. J.-A.S.: Consultant Sanofi, GenSight Biologics, Pixium Vision; Grants Banque Publique d’Investissement, ERC Synergy “HELMHOLTZ,” Foundation Fighting Blindness, and LabEx LIFESENSES (ANR-10LABX-65); Stock GenSight Biologics, Chronocam, Chronolife, Pixium Vision, Tilak Healthcare, Sparing Vision.
Hoogewoud et al
Treatment of Syphilitic Uveitis
M.G.: Expenses paid e Laboratoire Français du Fractionnement et des Biotechnologies, Octapharma. Author Contributions: Conception and design: Hoogewoud, Loubet, Le Jeunne, Monnet, Lortholary, Dupin, Brezin, Salah, Groh Data collection: Hoogewoud, Frumholtz, Blanche, Lebeaux, Benhaddou, Sedira, Coutte, Vanhaecke, Launay, Heron, Monnet, Lortholary, Sahel, Dupin, Brezin, Errera, Salah, Groh Analysis and interpretation: Hoogewoud, Frumholtz, Loubet, Charlier, Blanche, Lebeaux, Monnet, Lortholary, Brezin, Errera, Salah, Groh Obtained funding: Not applicable Overall responsibility: Hoogewoud, Frumholtz, Loubet, Charlier, Lebeaux, Benhaddou, Sedira, Coutte, Vanhaecke, Launay, Le Jeunne, Heron, Monnet, Lortholary, Sahel, Dupin, Brezin, Errera, Salah, Groh
Abbreviations and Acronyms: ARV ¼ antiretroviral; BCVA ¼ best-corrected visual acuity; BPG ¼ benzathine penicillin G; CI ¼ confidence interval; cRR ¼ corrected risk ratio; CSF ¼ cerebrospinal fluid; HIV ¼ human immunodeficiency virus; IV ¼ intravenous; MSM ¼ men having sex with men; OR ¼ odds ratio; STD ¼ sexually transmitted disease; SU ¼ syphilitic uveitis; SUN ¼ Standardization of Uveitis Nomenclature; TPHA ¼ Treponema pallidum hemagglutination assay; VDRL ¼ Venereal Disease Research Laboratory. Correspondence: Matthieu Groh, MD, Department of Internal Medicine, Hôpital St Louis, 1 Avenue Claude Vellefaux, 75010 Paris, France. E-mail: matthieu.groh@ aphp.fr.
9