Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 https://doi.org/10.1186/s13643-021-01758-7

SYSTEMATIC REVIEW UPDATE Open Access Safety and efcacy of diferent antibiotic regimens in patients with ocular toxoplasmosis: systematic review and meta‑analysis John E. Feliciano‑Alfonso1, Juliana Muñoz‑Ortiz2,3, María Alejandra Marín‑Noriega2, Andrés Vargas‑Villanueva2, Laura Triviño‑Blanco2, Natalia Carvajal‑Saiz2 and Alejandra de‑la‑Torre2*

Abstract Background: Ocular toxoplasmosis (OT) is the most common cause of posterior uveitis, which leads to visual impair‑ ment in a large proportion of patients. Antibiotics and corticosteroids lower the risk of permanent visual loss by con‑ trolling infection and infammation. However, there remains disagreement regarding optimal antibiotic therapy for OT. Therefore, this systematic review and meta-analysis were performed to determine the efects and safety of existing antibiotic treatment regimens for OT. Methods: MEDLINE, EMBASE, The Cochrane Central Register of Controlled Trials, LILACS, WHO International Clinical Trials Registry Platform portal, ClinicalTrials.gov, and Gray Literature in Europe (“OpenGrey”) were searched for relevant studies; manual searches of reference lists were performed for studies identifed by other methods. All published and unpublished randomized controlled trials that compared antibiotic schemes known to be efective in OT at any dos‑ age, duration, and administration route were included. Studies comparing antibiotics with placebo were excluded. This review followed standard methodological procedures recommended by the Cochrane group. Results: Ten studies were included in the narrative summary, of which four were included for quantitative synthesis (meta-analysis). Interventions were organized into three groups: intravitreal clindamycin versus pyrimethamine sul‑ fadiazine, trimethoprim sulfamethoxazole versus other antibiotics, and other interventions. The frst comparison+ favored intravitreal clindamycin+ (Mean diference (MD) 0.10 logMAR; 95% confdence interval 0.01 to 0.22). However, this fnding lacks clinical relevance. Other outcomes= showed no statistically signifcant= diferences between the treatment groups. In general, the risk of performance bias was high in evaluated studies, and the quality of the evidence found was low to very low. Conclusions: No antibiotic scheme was superior to others, and the selection of a treatment regimen depends on multiple factors; therefore, treatment should be chosen based on safety, sulfa allergies, and availability. Keywords: Toxoplasmosis, ocular, Toxoplasma gondii, Uveitis, Anti-bacterial agents, Therapeutics

Background [1, 2]. OT results from an acquired or congenital infec- Ocular toxoplasmosis (OT) is a primary cause of pos- tion by the obligate intracellular protozoan parasite terior uveitis worldwide, particularly in South America Toxoplasma gondii (Tg), which can infect both humans and animals. An estimated 25%–30% of the human pop- *Correspondence: [email protected] ulation is infected with Tg [3]. Seroprevalence is low 2 NeURos Research Group, Escuela de Medicina Y Ciencias de La Salud, (approximately 10%–30%) in Southeast Asia, Northern Universidad del Rosario, Carrera 24 # 63 C 69, Bogotá, Colombia Europe, and North America. In Central and Southern Full list of author information is available at the end of the article

© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat​ iveco​ mmons.​ org/​ licen​ ses/​ by/4.​ 0/​ . The Creative Commons Public Domain Dedication waiver (http://creat​ iveco​ ​ mmons.org/​ publi​ cdoma​ in/​ zero/1.​ 0/​ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 2 of 14

Europe, the prevalence reported varies between 30 as an adjuvant treatment. Oral prednisolone at 1 mg/kg/ and 50%. Latin America and some tropical countries in day is administered from the third day until 2–6 weeks of Africa have higher seroprevalence (up to 80%) [4]. treatment. Intravitreal therapy also includes clindamycin Tg infection in humans can be congenital or post- combined with dexamethasone [15]. natally acquired through contaminated hands, food, In survey studies, ophthalmologists were asked about water, consumption of undercooked and raw meat, the therapies they used to manage active OT; the results organ transplantation, blood transfusion, or vertical revealed a diverse range of treatment regimens, and many transmission [5–7]. Toxoplasmosis has a variety of clin- ophthalmologists expressed doubtfulness about ques- ical manifestations in humans. Congenital toxoplasmo- tions regarding the disease. Te diversity of responses sis is associated with intrauterine growth restriction, suggests widespread uncertainty regarding appropriate neurological, mental disease, hydrocephalus, encepha- treatment for this disease; moreover, there is a need for litis, retinochoroiditis, hearing and cardiovascular continuing medical education regarding OT [16]. Tere abnormalities, and fetal loss; it may also be asymp- have been some publications concerning this topic [13, tomatic [8, 9]. Acquired infection is asymptomatic in 17, 18]. Notably, a systematic review of the literature more than 80% of immunocompetent patients. In con- was published in 2003 regarding antibiotics for toxoplas- trast, 20% of infected individuals may experience ocular mic retinochoroiditis; however, it compared antibiotic compromise, fever, or cervical lymphadenopathy; these schemes versus placebo and concluded that there was a manifestations are occasionally associated with myal- lack of evidence to support routine antibiotic treatment gia, asthenia, or other nonspecifc clinical signs [10]. [18]. Another systematic review compared antibiotic Te most common manifestation of OT is toxoplasmic treatment versus placebo in patients with OT; it found retinochoroiditis, which typically comprises a unilat- a lack of evidence to support routine antibiotic treat- eral, unifocal, necrotizing retinochoroidal lesion that ment for toxoplasma retinochoroiditis to prevent visual appears as a whitish-yellow region with a blurred mar- impairment and found weak evidence regarding the pre- gin, with or without an accompanying old lesion (often vention of recurrences. Tus, the authors of that study associated with vitritis) [11]. concluded that the risk of recurrence is likely reduced Te treatment goals in OT are reducing the risk of after long‐term treatment with systemic antibiotics [13]. visual impairment by avoiding the multiplication of the Nonetheless, in ophthalmological clinical practice, anti- parasite during the active stage of retinochoroiditis, and biotic treatment must be provided to achieve treatment diminishing the necrotizing infammation and subse- objectives for patients with OT. Our systematic review quent harm to adjacent tissues [12]. In addition, because and meta-analysis will complement a recently published antibiotic treatment is only efective against the tachy- meta-analysis regarding treatment in immunocompetent zoite form of the parasite, it is important to treat each patients [19]. active lesion, which presumably reduces the risk of recur- Tis systematic review and meta-analysis aimed to rence [12, 13]. evaluate the evidence-based information regarding the Antibiotic therapy in acute lesions of OT is highly safety and efcacy of diferent existing antibiotic regi- recommended for the reduction of intraocular infam- mens in patients with OT worldwide. mation and the prevention of retinal and optic disc dam- age. Unfortunately, current treatment schemes do not Methods eradicate Tg tissue cysts; thus, they do not prevent fur- Protocol and registration ther reactivation of toxoplasma retinochoroiditis unless Te study protocol was developed based on the Preferred prophylaxis is started [14]. Reporting Items for Systematic Reviews and Meta-Anal- Various treatment regimens are often used in clini- ysis Protocols (PRISMA-P) guidelines [20], and it was cal practice. Te classic triple therapy is pyrimeth- previously published in 2019 [21]. In the same way, it was amine + sulfadiazine (PYR/SDZ) combined with folinic registered in PROSPERO, with the registration number acid, which is the frst choice for congenital toxoplasmo- CRD42018085468 [22]. sis and immunosuppressed patients; PYR/SDZ + clin- Eligibility criteria damycin + folinic acid is known as quadruple therapy; trimethoprim + sulfamethoxazole (TMP/SMX) can be We included all published and unpublished randomized used alone or combined with clindamycin; for individuals controlled trials (RCTs) comparing therapies that have allergic to sulfa, azithromycin + pyrimethamine + folinic been used in OT. RCTs provide the highest quality of acid or clindamycin + pyrimethamine + folinic acid can evidence according to the grading of recommendation be used; and clindamycin alone can be used for intravit- assessment, development, and evaluation (GRADE) clas- real therapy [14]. In addition, corticosteroids can be used sifcation [23, 24]. Terapies described in the literature for Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 3 of 14

OT include trimethoprim-sulfamethoxazole, pyrimeth- independently compared the full text with the inclusion amine, sulfadoxine, sulfadiazine, clindamycin, tetra- criteria. Finally, disagreements were resolved by consen- cyclines, clarithromycin, azithromycin, atovaquone, sus or by a third reviewer (ADLT). minocycline, spiramycin, rifabutin, trimetrexate, linco- mycin, dapsone, sulfafurazole, ciprofoxacin, doxycycline, Data collection process miokamycin, erythromycin, macrolide, sulfonamide, sul- A data collection form was designed. Then, two famerazine, nifurtimox, methotrexate, alone or in com- review authors (JMO and JFA) independently bination [13, 25–28]. We also included therapies at any extracted relevant details regarding the design and dosage, duration, and administration route, oral or intra- results of each study. Disagreements were resolved vitreal. RCTs comparing antibiotics with placebo were by consensus or by independent evaluation by a third excluded because a systematic review of this comparison review author (ADLT). One review author (JFA) has already been conducted by Pradhan et al. [13]. Par- entered data into Review Manager 5.3 (RevMan 5.3) ticipants included were patients of any age who received software and a third reviewer (ADLT) checked them antibiotic treatment for acute OT worldwide, and those to ensure data quality. with healed scars who received prophylactic antibiotic treatment to prevent recurrent or new lesions, includ- Risk of bias in individual studies ing immunocompetent patients, immunosuppressed We used the Cochrane group “Risk of Bias” tool for patients, pregnant women, and children. RCTs and criteria in the Cochrane Handbook for Sys- tematic Reviews of Interventions to assess these risks Outcomes in the relevant domains of the reported methods and Primary outcome measures were changes in visual results from each included study. Te fve domains acuity at least three months after the start of treat- are selection bias, performance bias, attrition bias, ment and the number of recurrences at the end of detection bias, reporting bias, and other biases. Tese follow-up. Secondary outcome measures included the domains can be classifed as high, unclear, and low risk behavior of ocular inflammation signs according to of bias [29]. the Standardization of Uveitis Nomenclature (SUN), size of lesion at the end of the follow-up, adverse drug reactions, and duration of the active lesion, as stated Dichotomous data in the study protocol [21]. We presented the results as risk ratios (RRs) with 95% confdence intervals (CIs) for dichotomous data. Information sources We used a combination of exploded controlled vocabu- Continuous data lary with thesaurus Science Health Descriptors (DeCS for We used the mean diference for continuous data if its Spanish acronym), Medical Subject Heading (MeSH) outcomes were measured in the same manner among and Embase Subject Headings (Emtree), and free-text multiple trials. We used the standardized mean difer- terms (considering spelling variants, plurals, synonyms, ence to combine trials that measured the same outcome acronyms, and abbreviations) with feld labels, trunca- using diferent methods. tion, proximity operators, and Boolean operators. We conducted our search in the MEDLINE, EMBASE, Management of missing data Cochrane Central Register of Controlled Trials, and LILACS electronic databases, from inception to March We contacted study investigators to obtain miss- 2018 (Annex 1). Besides, the search was updated in ing data. We analyzed available data without making November 2020. For identifcation of additional studies, assumptions or imputing data to adjust for missing the following resources were used: WHO International data. Clinical Trials Registry Platform portal, ClinicalTrials. gov, Gray Literature in Europe (“OpenGrey”), and man- Assessment of heterogeneity ual searches within reference lists of all relevant studies We assessed statistical heterogeneity in each meta- identifed by other methods. analysis using the I­2 statistic. We regarded heteroge- neity as substantial if ­I2 was > 40% and < 70%, which Study selection required using a random-efects model in the analysis. Studies obtained by electronic databases were inde- If there was substantial heterogeneity and I­2 was > 70%, pendently reviewed by two authors (JFA and AVV), meta-analysis was not performed; instead, a narrative evaluating titles and abstracts of all studies. Next, they summary of data was conducted. Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 4 of 14

Assessment of reporting biases assume that studies estimated the same underlying treat- We used funnel plots to assess publication bias when ment efect. Statistical analysis using RevMan 5.3 and at least 10 studies were available for meta-analysis. In GRADE assessment (Annex 2) was performed by one addition, we evaluated possible sources of asymmetry author (JFA). We assessed statistical heterogeneity in in funnel plots, following the method described in the each meta-analysis using the ­I2 statistic. Additional infor- Cochrane Handbook for Systematic Reviews of Interven- mation about the methodology is presented in the pub- tions [29]. lished protocol [21].

Data synthesis Results We used a random-efects model to calculate our meta- Literature search results analysis if eligible studies represented clinically varied In total, 131 studies were identifed in the electronic populations. In addition, we used a fxed-efect meta- databases (44 in Embase, 43 in MEDLINE, 32 in analysis for combining data where it was reasonable to Cochrane Central Register of Controlled Trials, and 12

Fig. 1 Study fow diagram Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 5 of 14

in LILACS); 18 references were identifed in the addi- attributes; the Soheilian 2005 [30] study did not have a tional sources. After duplicates had been removed, 94 high risk of bias in any evaluated attributes. Te bias risk records were screened; 15 were selected for full-text assessments for all included studies are summarized in evaluation. Five studies were excluded for reasons indi- Figs. 2 and 3. cated in Fig. 1 [30–35]. In the case of Sadoughi et al. It was not possible to assess the publication bias since 2006 study [35], the same population was also evalu- the number of studies necessary to evaluate it was not ated by Soheilian 2005; thus, we choose the study with enough. a longer follow-up, which corresponded to Soheilian 2005 [30]. Efects of interventions The remaining 10 studies were included for narra- Intravitreal clindamycin versus PYR/SDZ tive summary of data (Balaskas 2012 [36], Baharivand Te number of studies that provided data for this com- 2013 [37], Bosch-Driessen 2002 [38], Colin 1989 [39], parison was two RCT, and both were included in the Ghavidel 2017 [40], Kartasasmita 2017 [41], Lashay meta-analysis. Te Soheilian 2011 [44] and Baharivand 2017 [42], Ortega 2000 [43], Soheilian 2005 [30], 2013 [37] studies compared head-to-head intravitreal Soheilian 2011 [44]); four provided sufficient informa- clindamycin versus PYR/SDZ. Regarding changes in tion for quantitative synthesis (meta-analysis) (Bahari- visual acuity, the meta-analysis revealed statistically vand 2013 [37], Lashay 2017 [42], Soheilian 2005 [30], signifcant diferences in favor of intravitreal clinda- Soheilian 2011 [44]). mycin (mean diference = 0.11 logMAR; 95% CI = 0.03 In the update 2020, 21 new studies were identified to 0.20) (Fig. 4). However, this result is not considered in the electronic databases (11 in Embase, 7 in MED- clinically relevant, as will be discussed below. Addi- LINE, 3 in Cochrane Central Register of Controlled tionally, in the random-efects model, there were no Trials, and 0 in LILACS). After duplicates had been statistically signifcant diferences (MD 0.10 logMAR; removed, 12 records were screened. None of the stud- 95% CI = -0.01 to 0.22). No signifcant diferences were ies evaluated in the title and abstract phase met the reported in the two studies regarding the number of selection criteria. recurrences and lesion size. In the Soheilian 2011 [44] study, lesion size was reported in pixels with a mean percentage of changes of 1.4% (95% CI = -14.6% to Characteristics of included studies 17.4%, p = 0.86). In contrast, in the Baharivand 2013 Most of the 10 included studies were conducted in Asia, [37] study, retinal lesion size was reported as ≤ 1 disc particularly in Iran (Baharivand 2013 [37], Ghavidel area (DA), < 1 and ≤ 2 DA, and < 2 and ≤ 3 DA. Bahari- 2017 [40], Lashay 2017 [42], Soheilian 2005 [30], Soheil- vand et al. [37] regarded improvement in retinal lesion ian 2011 [44]) and Indonesia (Kartasasmita 2017 [41]). size as a lesion size that decreased in terms of DA clas- Te remaining studies were conducted in Europe (Bal- sifcation (improvement of 65.6% in intravitreal clinda- askas 2012 [36], Bosch-Driessen 2002 [38], Colin 1989 mycin group versus 67.6% in PYR/SDZ group, p = 0.86). [39]) and North America (Ortega 2000 [43]). All 10 Although none of the studies reported ocular infam- studies were published between 1989 and 2017. Eight mation according to the SUN, this outcome was evalu- studies used the classic therapy PYR/SDZ as a com- ated as a reduction of vitreous cells to the level of 0 or parison group. Tree studies included patients under trace after treatment, which was regarded as the reso- 18 years of age (Colin 1989 [39], Ortega 2000 [43], lution of vitreous infammation. Tus, for this outcome, 2 Soheilian 2005 [30]); patient age ranged between 7 and there was no statistical heterogeneity ­(I = 0%), and no 72 years in all included studies. In general, the stud- signifcant diferences were found between the treat- ies had a small sample size, from 19 patients (Balaskas ment groups (RR = 1.04; 95% CI = 0.83 to 1.31) (Fig. 5). 2012 [36]) to 72 patients (Ghavidel 2017 [40]). A sum- In the Baharivand 2013 study, the arm that evaluated mary of all included studies can be found in Table 1, and intravitreal clindamycin presented no adverse drug Annex 3 provides detailed information regarding the 10 reactions, whereas in the PYR/SDZ arm, one case of included studies. adverse drug reactions was reported (hepatotoxicity). In the Soheilian 2011 study, there were two cases of adverse drug reactions in the PYR/SDZ arm (skin rash Risk of bias and thrombocytopenia), which were not included in Overall, nine studies had a high risk of performance bias; the efcacy analysis, while in the intravitreal clindamy- seven studies had low risks of selection bias (random cin arm, there were four cases of adverse drug reactions sequence generation), detection bias, and other bias. Te (three subconjunctival hemorrhages and one transient Ortega 2000 [43] study had a high risk of bias in most raised intraocular pressure). Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 6 of 14 ± 0,29 logMAR 14 /22 (64%) 7 /17 (41%) 10/14 (71%) 5/18 (28%) 5 / 9 (56%) NR 1 (2.9%) 23 (67.6%) 28 (82.4%) 28/34 5 (14.7%) 0.35 9 (100%) NR NR NR NR NR Control ± 0,35 logMAR 8/ 24 (33%) 9/ 22 (41%) 14 /20 (70%) 5 /24 (21%) 5 /15 (33%) NR 0 21 (65.6%) 28 (87.5%) 27/32 4 (12.5%) 0.38 NR NR NR NR NR NR Intervention 0.5 logMAR units 0.5 logMAR = (all) improvement improvement than more (decrease 0,5 optic disk diam ‑ in 3 months) eter from vitreous within vitreous from 4 w) at 3 months) low up) low (major) improvement (grade 0/trace) (grade lines) (minor) Adverse drug reactions drug reactions Adverse Retinal lesion size Retinal lesion size Ocular infammation (No infammatory cells Improvement Improvement in VA (> Number of recurrences ‑ fol (at least one year Changes in VA Adverse drug reactions drug reactions Adverse Retinal lesion size Retinal lesion size Ocular infammation Improvement Improvement in VA (gaining > = 2 Snellen Number of recurrences Changes in VA Adverse drug reactions drug reactions Adverse Changes in VA Number of recurrences Improvement in VA Ocular infammation Retinal lesion Size Outcomes folinic + folinic folinic + folinic folinic + folinic + prednisone + prednisone + prednisone acid oral acid oral acid oral PYR/SDZ route: Administration PYR/SDZ route: Administration PYR/SDZ route: Administration Control Treatment ‑ + pred ‑ + dexa ‑ folinic + folinic + prednisone + Azithro mycin acid oral methasone intravitreal nisone oral PYR route: Administration Clindamycin route: Administration Azithromycin route: Administration Intervention Treatment = 24 = 32 = 10 = 22 = 34 = 9 = 46 = 66 = 19 N Intervention Control N Intervention Control N Intervention Control Participants Netherlands Iran Switzerland Location Bosch-Driessen, 2002 Baharivand, 2013 Baharivand, Balaskas, 2012 1 asas Table year Author, Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 7 of 14 ± 4,6% ± 0.01 logMAR Snellen acuity) 3 (23%) 36.6% 10 (77%) NR 10/13 0.30 52.5% NR NR NR NR NR 20 (55.5%) 638.89 µm 4 (11,1%) NR 0.39 logMAR (20/49 0.39 logMAR NR 1 36% NR NR NR NR Control ± 6,5% ± 0.04 logMAR Snellen acuity) 4 (28.5%) 24.2% 7 (50%) NR 10/14 0.24 57.50% NR NR NR NR NR 4 (12.5%) 354.86 µm 18 (50%) NR 0.35 logMAR (20/44 0.35 logMAR NR 1 21% NR NR NR NR Intervention (mild) (reduction) (vitreous infamma ‑ (vitreous tory cells clearance) (percentage of lesion (percentage reduction in the area w) third (all) (reduction during treatment) (during 24 months after treatment) (all) Adverse drug reactions drug reactions Adverse Retinal lesion size Retinal lesion size Ocular infammation Number of recurrences Improvement in VA Changes in VA Retinal lesion size Retinal lesion size Changes in VA Number of recurrences Improvement in VA Ocular infammation drug reactions Adverse Adverse drug reactions drug reactions Adverse Retinal lesion size Retinal lesion size Number of recurrences Number of recurrences Improvement Improvement in VA Changes in VA Ocular infammation Adverse drug reactions drug reactions Adverse Number of recurrences Changes in VA Ocular infammation Improvement in VA Retinal lesion Size Outcomes ‑ ‑ ‑ folic + folic + Pred ‑ + methyl + predniso + Predniso nisone oral prednisolone acid oral lone oral lone oral Intramuscular TMP/SMX Administration route: route: Administration PYR/SDZ route: Administration PYR/SDZ route: Administration PYR/SDZ route: Administration acid Folinic route: Administration Control Treatment ‑ + pred ‑ + predni + oral ‑ + Clindamy ‑ + methylpredni nisone oral cin solone oral solone oral subconjunctival injections prednisolone Azithromycin route: Administration TMP/SMX Administration route: route: Administration Azithromycin route: Administration Clindamycin route: Administration Intervention Treatment = 14 = 36 = 14 = 13 = 14 = 36 = 15 = 14 tion = 27 Interven ‑ = 28 = 72 = 29 N Control N Intervention Control N Intervention Control N Intervention Control Participants Iran Indonesia Iran Not exactly reported Location (continued) Lashay, 2017 Lashay, Kartasasmita, 2017 Ghavidel, 2017 Ghavidel, Colin, 1989 Colin, 1 Table year Author, Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 8 of 14 + —651,553 ± 0,19 logMAR /30 (2,9%) /36 pixels a a 61% 20 (69%) 3 (10,3%) 0,56 logMAR NR - - - - - 2 89,606 15 (55,5%) 2 0,29 NR Control 1 + —143,997 ± 0,24 logMAR /31 (2,8%) Pixels a 1 59% 17 (56,7%) 3 (10%) 0,52 logMAR NR G1 1 (7.6%) G2 5 (31.2%) G3 4 (36.3%) G1 1 (7.6%) G2 4 (18.1%) G3 0 (0%) NR NR NR NR 4/34 116,994 15 (51,7%) 2 0,44 NR Intervention Patients excluded from the study. See3 details in Annex the study. from excluded Patients a after treatment) infammatory cells (0–trace cells) 6 w after treatment) (increase) ‑ “Complica (called treatment” tions from mild) and apparently (all) (calculated in pixels in pixels (calculated ‑ envi using MATLAB ronment) (trace or no vitreous (trace or no vitreous cells) (eyes) (increase) Adverse drug reactions Adverse Retinal lesion size 6 w (mean reduction Ocular infammation (Reduction of vitreous Number of recurrences Changes in VA Changes in VA Improvement Improvement in VA Number of relapses drug reactions Adverse Changes in VA Improvement in VA Ocular infammation Retinal lesion size Adverse drug reactions drug reactions Adverse Retinal lesion size Retinal lesion size Ocular infammation Number of recurrences Number of recurrences Changes in VA Changes in VA Improvement Improvement in VA Outcomes folinic + folinic Visual Acuity, w: weeks. Acuity, Visual + prednisolone + prednisolone folinic + SDZ folinic acid oral acid oral PYR/SDZ route: Administration - PYR route: Administration Control Treatment Not reported, VA NR Not reported, + Clinda ‑ + dexa ‑ ‑ + pred + oral ‑ pred folinic + folinic acid + folinic acid + prednisone ‑ + PYR pred + PYR Clin ‑ nisolone oral SMX nisone mycin oral SMX damycin nisone oral methasone intravitreal TMP/SMX Administration route: route: Administration G1 TMP/ G2 TMP/SMX route: Administration G3 TMP/ route: Administration Clindamycin route: Administration Intervention Treatment = 13 = 22 = 11 = 30 = 34 = 29 = 34 = 59 = 46 = 68 N Intervention Control N Intervention G1 Intervention G2 Intervention G3 N Intervention Control Participants Iran Mexico Iran Location (continued) Total sample, N Total TMP/SMX trimethoprim/ sulfamethoxazole, SDZ Sulfadiazine, Pyrimethamine, PYR Soheilian, 2005 Ortega, 2000 Soheilian, 2011 Author, year Author, 1 Table Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 9 of 14

Fig. 2 Risk of bias graph: review authors’ judgments regarding each risk of bias item, presented as percentages across all included studies

TMP/SMX versus another antibiotic azithromycin group and 0.39 logMAR in the PYR/SDZ Two RCTs compared TMP/SMX versus another antibi- group, although this diference was not statistically sig- otic, and both were included in the meta-analysis. Te nifcant (p = 0.33). In addition, there was no diference in Soheilian 2005 and Lashay 2017 [42] studies compared the size of the retinal lesion detected. In terms of safety, TMP/SMX versus PYR/SDZ and TMP/SMX versus only four patients in the azithromycin group had gas- azithromycin, respectively. In terms of ocular infam- trointestinal adverse drug reactions. For the PYR/SDZ mation, regarded as the percentage of patients with a group, 20 patients presented adverse drug reactions (16 reduction of vitreous infammatory cells (0 or trace cells) gastrointestinal and 4 dizziness) (p < 0.01 compared with after treatment, there were no signifcant diferences the azithromycin group); however, no severe adverse (RR 1.08; 95% CI 0.59 to 1.98) with moderate het- drug reactions were observed, such as bone marrow = 2 = erogeneity ­(I = 66%) (Fig. 6). Likewise, both treatments suppression. showed a similar number of adverse drug reactions that Te Bosch-Driessen 2002 [38] study compared ranged from mild to signifcant. A summary of the results pyrimethamine + azithromycin (PYR/AZ) versus PYR/ can be found in Table 1. Annex 3 provides detailed SDZ for 4 weeks. Tis study did not report explicit base- information. line data for included patients. Tere were no signifcant diferences in visual acuity improvement at 3 months, Other interventions as well as in the number of recurrences to the frst year Six out of ten studies included data about other ther- of treatment or the disappearance of infammatory cells apeutic schemes, and none were included in the from vitreous within 4 weeks. Te frequency of adverse meta-analysis. drug reactions was higher in the PYR/SDZ group than Te Kartasasmita 2017 [41] study compared TMP/ in the PYR/AZ group (64% versus 33%, respectively; SMX + oral clindamycin versus PYR/SDZ for 3 weeks p = 0.04); more patients discontinued treatment in the and found that the mean percentage of lesion remis- PYR/SDZ group (14%), compared with the PYR/AZ sion from the frst visit to the last visit was 57.5% group (0%) (p = 0.1). in the frst group, while it was 52.5% in the second Te Balaskas 2012 [36] study compared azithromycin group (p = 0.72). Tis result was obtained by compar- with PYR/SDZ in 19 patients without a specifc duration ing the reduction in retinal lesion size between the of treatment, with a follow-up of 3 months. Tis study pretreatment evaluation and the third week of treat- investigated changes in lesion size, time of the sharpening ment evaluation. However, in the frst week of treat- of lesion borders, time to scarring, and time to disease ment, the authors found that the mean percentage of inactivity in days; these results did not demonstrate sta- patients with lesion remission was higher in the TMP/ tistically signifcant diferences. Skin rash was observed SMX + oral clindamycin group than in the PYR/SDZ in one patient in the PYR/SDZ group. Treatment fail- group (71.7% versus 19.8%; p = 0.001). Tere were no ure was documented in one patient in the azithromycin safety considerations in this study. group. Te Ghavidel 2017 [40] study compared azithromy- Te Colin 1989 [39] study compared subconjunctival cin versus PYR/SDZ for 6 weeks. Te authors reported injections of clindamycin (3 weeks of treatment) ver- that visual acuity improved by 0.35 logMAR in the sus PYR/SDZ (6 weeks of treatment) in 29 patients. Te Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 10 of 14

to resolve infammation (p = 0.044) and achieve healing (p = 0.016). Although there were adverse drug reactions, these were not reported explicitly. Table 1 synthesized the results of the included studies in this systematic review.

Discussion Agreements and disagreements with other studies or reviews Our meta-analysis showed that, when comparing the efectiveness of intravitreal clindamycin and oral PYR/ SDZ, there were signifcant diferences in visual acuity in favor of intravitreal clindamycin (Fig. 4). However, there were no clinically signifcant diferences in the use of either treatment because the changes in visual acuity corresponded to 0.1 logMAR (i.e., one line in the Snellen chart). Signifcant changes in visual acuity are defned as worsening or gaining ≥ 15 letters (i.e., three lines in standardized optotypes) [45, 46]. It is important to highlight the fndings of the meta-analy- sis by Pradhan et al. in 2016 [13], where only one study (Felix et al. 2014) [47] evaluated visual acuity outcome measured at least 3 months after the start of treatment, showing similar changes in the TMP/SMX and placebo groups [13]. Regarding lesion size, the measurement methods in the included studies did not allow a meta-analyt- ical comparison of the fndings. Furthermore, in the meta-analysis by Pradhan et al., the outcome of lesion size was considered, but the analyzed studies did not report it [13]. Te methods described in the literature for measuring lesion size include the following: fundo- scopic images, in which the lesion and optic disc areas were measured using Photoshop software and the ratio of the lesion to optic disc area was calculated [48]; a Fig. 3 Risk of bias summary: review authors’ judgments regarding program written in the MATLAB environment made to each risk of bias item for each included study determine the margin of the retinitis and to measure the area of interest in pixels [44]; and the retinal lesion size in terms of ≤ 1 disc area (DA), < 1 and ≤ 2 DA, and < 2 authors reported visual acuity but did not specify the and ≤ 3 DA. “Improvement” was considered a lesion measurement units or whether they represented near or size that decreased in terms of DA classifcation [37]. It far visual acuity. At 14 months after treatment, there were is important to create consensus regarding lesion size 21% recurrences in the clindamycin group and 36% in the measurements and the percentage of lesion reduction in PYR/SDZ group (p-value not reported). In the PYR/SDZ future investigations. group, one patient exhibited Stevens-Johnson syndrome, Even though the SUN working group was frst pub- and another patient exhibited a skin rash. In the clinda- lished in 2005 [49], intraocular infammation measures mycin group, one patient exhibited corneal ulceration were not used in most of the included studies, and this with infammation. outcome could not be objectively measured. Terefore, in Te Ortega 2000 [43] study was an RCT with three further studies, the use of this classifcation is crucial to treatment groups: TMP/SMX + PYR versus TMP/ standardize and compare the intraocular infammation SMX + clindamycin versus TMP/SMX + PYR + clindamy- outcomes. cin for 8 weeks. All groups exhibited an increase in visual Concerning the antibiotic interventions evaluat- acuity (p < 0.05) compared with baseline. Notably, only ing recurrences in OT, it is important to notice that all the TMP/SMX + clindamycin group required less time the regimens of antibiotics were not included in our Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 11 of 14

Fig. 4 Comparison between intravitreal clindamycin and PYR/SDZ, mean changes in visual acuity

systematic review. For example, an observational study treatment, indicating that further rigorous research is performed in patients treated with intensive therapy very likely to have an important impact on the conf- in the active phase, followed by long-term treatment dence of the estimated efect and, in consequence, could with pyrimethamine sulfadoxine (PYR/SDX) (it was not change it. included in our review since it was not a RCT), revealed Zhang et al. [19] published a network meta-analysis a 90.9% probability of 3-year recurrence-free survival about treatment in ocular toxoplasmosis. Te overall after the frst intervention [50]. Future RCTs may include confdence rating of Zhang et al. review is critically low, PYR/SDX scheme, as it is one of the alternative therapies according to the AMSTAR 2 (A Measurement Tool to accessible in diferent countries where PYR/SDZ is not Assess Systematic Reviews 2) [51]. Indeed, more than available. one critical faw was found in the AMSTAR 2 assess- Regarding the number of recurrences, there were no ment in domains 2, 7, 9, 13, and 15. Notably, the net- statistically signifcant diferences between PYR/SDZ work meta-analysis of Zhang et al. did not report the and intravitreal clindamycin in the evaluated studies. risk of bias of the included studies. In contrast, our Nevertheless, the reduction of recurrences in patients study accomplishes all the items required in AMSTAR with OT has been evaluated in long-term therapy stud- 2, obtaining high overall confdence. Furthermore, our ies after the acute stage of infection has been controlled. systematic review and meta-analysis provide an accu- In the meta-analysis by Pradhan et al., the authors found rate and comprehensive summary of the results of the that treatment with antibiotics compared with placebo or available studies that address the question of interest. no treatment probably reduces the risk of recurrent toxo- Nonetheless, we recognize that although the AMSTAR plasma retinochoroiditis (Risk ratio 0.26, 95% CI 0.11 to 2 was not intended to handle the special requirements 2 0.63; 227 participants; 3 studies; I­ = 0%; moderate-qual- of a network meta-analysis [51], a modifed AMSTAR ity evidence). should be developed to assess network meta-analy- Additionally, in the meta-analysis by Pradhan et al., ses [52]. Recently published papers showed that the there was low-quality evidence regarding visual acuity tool called CINeMA (Confdence in Network Meta- changes, intraocular infammation, and adverse drug Analysis) contributes to assessing the credibility of reactions [13]. Tis agrees with the results of our met- the results of a network meta-analysis and should be analysis, where we found that the evidence quality of taken into account when properly evaluating this type the included studies was from low to very low (GRADE of study design [53, 54]. A comparison of AMSTAR 2 methodology results are shown in Annex 2). Tis means results of our work and Zhang et al.’s study is available that there are no well-conducted RCTs regarding OT in Annex 4.

Fig. 5 Comparison between intravitreal clindamycin and PYR/SDZ, resolution of vitreous infammation Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 12 of 14

Fig. 6 Comparison between TMP/SMX and another antibiotic, resolution of vitreous infammation

Regarding the safety of the analyzed drugs, it was strategy; thus, secondary outcomes were considered for found that those studies that reported information on the quantitative analysis [21]. In addition, missing data PYR/SDZ adverse drug reaction had a frequency that and supplementary material were requested from some was in the range of 0.05% (Sohelian 2011 [44]) to 100% authors; however, they did not respond or did not have (Balaskas 2012 [36]), including all degrees of severity. access to those data. In the same way, concerning TMP/SMX, adverse drug reaction was reported from 2.8% (Sohelian 2005 [30]) to Conclusions 23% (Lashay 2017 [42]), and through various degrees of severity. Tus, overall, there were no signifcant difer- In general, the quality of the evidence found was low to ences in adverse drug reactions for the diferent treat- very low. Although visual acuity analysis revealed a sta- ment regimens compared to each other, except for the tistically signifcant diference in the frst comparison of azithromycin-containing arms. Indeed, Ghavidel et al. intravitreal clindamycin and oral PYR/SDZ, the result [40] found a better tolerance in the azithromycin group was not clinically relevant. Regarding the remaining than its comparator, which also occurred in Bosch- outcomes, none of them showed statistically signifcant Driessen 2002 [38]. Terefore, azithromycin appears to diferences (i.e., no treatments were superior to others). have a better safety profle among the diferent options Terefore, the selection of treatment regimen must be analyzed through the studies included in this systematic performed on an individual basis, considering the safety review. of each therapeutic regimen, (azithromycin appears to Finally, according to the results of this systematic have the best safety profle), medical history of sulfa review, some combinations of antibiotic schemes have allergy, and the availability of medications ofered within not been compared in clinical trials but are used in the each nation’s health system. Although there is no con- clinical practice. Terefore, we consider that it is crucial sensus regarding the best treatment regimens in OT, the to know the results of comparisons between combined treatment should include at least two antibiotics along with corticosteroids. clindamycin + TMP / SMX + prednisone versus TMP / SMX + prednisone versus clindamycin + prednisone. Also, quadruple-drug therapies such as TMP / SMX + clin- Abbreviations damycin prednisone versus PYR / SDZ azithro- OT: Ocular toxoplasmosis; PYR/SDZ: Pyrimethamine sulfadiazine; PYR/ + + SDX: Pyrimethamine sulfadoxine; PYR/AZ: Pyrimethamine+ azithromycin; + + mycin + prednisone. We also consider it relevant to TMP/SMX: Trimethoprim sulfamethoxazole; SUN: Standardization of Uveitis compare sulfadiazine and sulfadoxine in well-conducted Nomenclature. + clinical trials, to understand the superiority or noninferior- ity between these sulfa drugs in OT. Supplementary Information The online version contains supplementary material available at https://​doi.​ org/​10.​1186/​s13643-​021-​01758-7. Potential biases in the review process and limitations

We aimed to minimize potential biases by conducting Additional fle 1. a highly sensitive search for trials; we followed rigorous Additional fle 2. methods as recommended by the Cochrane group “Risk Additional fle 3. of Bias” tool for RCTs and other criteria considered in the Cochrane Handbook for Systematic Reviews of Interven- Additional fle 4. tions [29]. Most primary outcomes established in the study pro- Acknowledgements We thank Universidad del Rosario for helping us with the manuscript style tocol were not settled in the articles found by the search correction. Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 13 of 14

We thank Ryan Chastain-Gross, Ph.D., from Edanz Group (https://​en-​ 10. Robert-Gangneux F, Dardé M-L. Epidemiology of and diagnostic strate‑ author-​servi​ces.​edanz​group.​com) for editing a draft of this manuscript and Dr. gies for toxoplasmosis. Clin Microbiol Rev. 2012;25:264–96. Juliana Reyes-Guanes for helping with the manuscript review process. 11. Dodds EM. Ocular toxoplasmosis: Clinical presentation, diagnosis and therapy. Focal Points Clin Modul Ophthalmol. 1999;17:10. Authors’ contributions 12. Casoy J, Nascimento H, Silva LMP, Fernández-Zamora Y, Muccioli C, Dias JEF: Management, research idea, protocol writing, title and abstracts screen‑ JR de O, et al. Efectiveness of Treatments for Ocular Toxoplasmosis. Ocul ing, data extraction, manuscript writing, discussion, and meta-analysis. JMO: Immunol Infamm. 2019;1–7. Protocol writing, data extraction, manuscript writing, and discussion. MAM: 13. Pradhan E, Bhandari S, Gilbert RE, Stanford M. Antibiotics versus no Protocol writing, title and abstracts screening, manuscript writing, and discus‑ treatment for toxoplasma retinochoroiditis. Cochrane Database Syst Rev. sion. AVV: Research idea, protocol registration, title and abstracts screening. 2016;CD002218. LT: Title and abstracts screening, and full-text reading. NC: Title and abstracts 14. Vasconcelos-Santos DV. Ocular Toxoplasmosis. In: Dumitrescu RG, Verma screening, and full-text reading. ADLT: Management, research idea, protocol M, editors. Cancer Epigenetics Precis Med [Internet]. New York, NY: writing, manuscript writing, discussion, and thematic authority. The author(s) Springer New York; 2019 [cited 2019 Apr 7]. p. 79–97. Available from: read and approved the fnal manuscript. http://link.springer.com/https://​doi.​org/​10.​1007/​978-3-​030-​03140-4_5 15. de-la-Torre A, Stanford M, Curi A, Jafe GJ, Gomez-Marin JE. Therapy for Funding Ocular Toxoplasmosis. Ocul Immunol Infamm. 2011;19:314–20. This research did not receive any specifc grant from funding agencies in the 16. Lum F, Jones JL, Holland GN, Liesegang TJ. Survey of ophthalmologists public, commercial, or not-for-proft sectors. about ocular toxoplasmosis. Am J Ophthalmol. 2005;140:724-e1. 17. Cortés JA, Roncancio Á, Uribe LG, Cortés-Luna CF, Montoya JG. Availability of data and materials Approach to ocular toxoplasmosis including pregnant women. Curr The data that support the fndings of this study are available from the cor‑ Opin Infect Dis. 2019;32:426–34. responding author, ADLT, upon reasonable request. 18. Stanford MR, See SE, Jones LV, Gilbert RE. Antibiotics for toxoplasmic retinochoroiditis: an evidence-based systematic review. Ophthalmol‑ ogy. 2003;110:926–32. Declarations 19. Zhang Y, Lin X, Lu F. Current treatment of ocular toxoplasmosis in immunocompetent patients: A network meta-analysis. Acta Trop. Consent for publication 2018;185:52–62. Not applicable. 20. Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred reporting items for systematic review and meta-analysis Competing interests protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1. The authors declare that they have no competing interests. 21. Feliciano-Alfonso JE, Vargas-Villanueva A, Muñoz-Ortiz J, de-la- Torre A. Antibiotic treatment for ocular toxoplasmosis: a system‑ Author details atic review and meta-analysis: study protocol. Syst Rev [Internet]. 1 Departamento de Medicina Interna, Facultad de Medicina, Universidad 2 2019;8:146–146. Available from: https://​www.​ncbi.​nlm.​nih.​gov/​ Nacional de Colombia, Bogotá, Colombia. NeURos Research Group, Escuela pubmed/​31221​217 de Medicina Y Ciencias de La Salud, Universidad del Rosario, Carrera 24 # 63 C 3 22. PROSPERO: International prospective register of systematic reviews 69, Bogotá, Colombia. Escuela Superior de Oftalmología-Instituto Barraquer [Internet]. [cited 2020 Nov 26]. Available from: https://​www.​crd.​york.​ac.​ de América, Bogotá, Colombia. uk/​prosp​ero/​displ​ay_​record.​php?​Recor​dID ​85468 23. Djulbegovic B, Guyatt GH. Progress in evidence= -based medicine: a Received: 6 March 2020 Accepted: 2 July 2021 quarter century on. The Lancet. 2017;390:415–23. 24. GRADE Working Group. Grading quality of evidence and strength of recommendations. BMJ. 2004;328:1490. 25. Wakefeld D, Cunningham ET Jr, Pavesio C, Garweg JG, Zierhut M. Controversies in ocular toxoplasmosis. Ocul Immunol Infamm. References 2011;19:2–9. 1. Wakefeld D, Chang JH. Epidemiology of uveitis. Int Ophthalmol Clin. 26. Engstrom RE Jr, Holland GN, Nussenblatt RB, Jabs DA. Current practices 2005;45:1–13. in the management of ocular toxoplasmosis. Am J Ophthalmol. 2. Huang PK, Jianping C, Vasconcelos-Santos DV, Arruda JSD, Dutta Majum‑ 1991;111:601–10. der P, Anthony E, et al. Ocular toxoplasmosis in tropical areas: analysis and 27. Gilbert RE, Harden M, Stanford M. Antibiotics versus control for toxo‑ outcome of 190 patients from a multicenter collaborative study. Ocul plasma retinochoroiditis. Cochrane Database Syst Rev. 2002; Immunol Infamm. 2018;26:1289–96. 28. Holland GN, Lewis KG. An update on current practices in the manage‑ 3. Montoya J, Liesenfeld O. Toxoplasmosis The lancet. 2004;363:1965–76. ment of ocular toxoplasmosis1, 2. Elsevier; 2002. 4. Pappas G, Roussos N, Falagas ME. Toxoplasmosis snapshots: global status 29. Higgins J. Cochrane handbook for systematic reviews of interventions. of Toxoplasma gondii seroprevalence and implications for pregnancy and Version 5.1. 0 [updated March 2011]. The Cochrane Collaboration. congenital toxoplasmosis. Int J Parasitol. 2009;39:1385–94. Www Cochrane-Handb Org. 2011; 5. Khan K, Khan W. Congenital toxoplasmosis: An overview of the neurologi‑ 30. Soheilian M, Sadoughi M-M, Ghajarnia M, Dehghan MH, Yazdani S, cal and ocular manifestations. Parasitol Int. 2018;67:715–21. Behboudi H, et al. Prospective randomized trial of trimethoprim/ 6. Laurence Baril BC Thierry Ancelle, Véronique Goulet, Philippe Thul‑ sulfamethoxazole versus pyrimethamine and sulfadiazine in the treat‑ liez, Véronique Tirard-Fleury. Risk Factors for Toxoplasma Infection ment of ocular toxoplasmosis. Ophthalmology. 2005;112:1876–82. in Pregnancy: A Case-Control Study in France. Scand J Infect Dis. 31. Abreu M, Martins M, Belfort R. Clindamycin versus spiramycin in ocular 1999;31:305–9. toxoplasmosis: a randomised double masked study. Arq Bras Oftalmol. 7. Fernàndez-Sabé N, Cervera C, Fariñas MC, Bodro M, Muñoz P, Gurguí M, 1988;51:26. et al. Risk Factors, Clinical Features, and Outcomes of Toxoplasmosis in 32. Jeddi A, Azaiez A, Bouguila H, Kaoueche M, Malouche S, Daghfous F, Solid-Organ Transplant Recipients: A Matched Case-Control Study. Clin et al. Value of clindamycin in the treatment of ocular toxoplasmosis. J Infect Dis. 2012;54:355–61. Fr Ophtalmol. 1997;20:418. 8. Wallon M, Garweg JG, Abrahamowicz M, Cornu C, Vinault S, Quantin C, 33. Prasil P, Plisek S. Bostik P [Ocular toxoplasmosis - seeking a strategy for et al. Ophthalmic Outcomes of Congenital Toxoplasmosis Followed Until treatment]. Klin Mikrobiol Infekcni Lek. 2014;20:112–5. Adolescence. Pediatrics. 2014;133:e601–8. 34. Raskin E, Alves M, Eredia G, Raskin A, Ruiz AM. Ocular toxoplasmosis: A 9. Oz HS, Tobin T. Atovaquone ameliorate gastrointestinal Toxoplasmosis comparative study of the treatment with sulfadiazine and pyrimeta‑ complications in a pregnancy model. Med Sci Monit Int Med J Exp Clin mine versus sulphametoxazole-trimethoprim. Rev Bras Oftalmol. Res. 2012;18:BR337–45. 2002;61:335–8. Feliciano‑Alfonso et al. Syst Rev (2021) 10:206 Page 14 of 14

35. Sadoughi M-M, SOHEYLIAN M, Dehghan M-H, Behboudi H, Anisian Aug 19]. Available from: http://​scielo.​isciii.​es/​pdf/​pap/​v18n71/​1139-​7632-​ A, Yazdani S. Short-term results of two treatment regimens in ocular pap-​18-​71-​00267.​pdf toxoplasmosis: Trimethoprim/sulfamethoxazole versus pyrimethamine 47. Felix JPF, Lira RPC, Zacchia RS, Toribio JM, Nascimento MA, Arieta CEL. and sulfadiazine. 2006; Trimethoprim-sulfamethoxazole versus placebo to reduce the risk of 36. Balaskas K, Vaudaux J, Boillat-Blanco N, Guex-Crosier Y. Azithromycin recurrences of Toxoplasma gondii retinochoroiditis: randomized con‑ versus Sulfadiazine and Pyrimethamine for non-vision-threatening trolled clinical trial. Am J Ophthalmol. 2014;157:762–6. toxoplasmic retinochoroiditis: a pilot study. Med Sci Monit Int Med J 48. Vishnevskia-Dai V, Achiron A, Buhbut O, Berar OV, Musika AA, Elyashiv SM, Exp Clin Res. 2012;18:CR296. et al. Chorio-retinal toxoplasmosis: treatment outcomes, lesion evolution 37. Baharivand N, Mahdavifard A, Fouladi RF. Intravitreal clindamycin plus and long-term follow-up in a single tertiary center. Int Ophthalmol. dexamethasone versus classic oral therapy in toxoplasmic retino‑ 2019;1–11. choroiditis: a prospective randomized clinical trial. Int Ophthalmol. 49. Group S of UN (SUN) W. Standardization of uveitis nomenclature for 2013;33:39–46. reporting clinical data. Results of the First International Workshop. Am J 38. Bosch-Driessen LH, Verbraak FD, Suttorp-Schulten MS, van Ruyven RL, Ophthalmol. 2005;140:509–516. Klok AM, Hoyng CB, et al. A prospective, randomized trial of pyrimeth‑ 50. Borkowski PK, Brydak-Godowska J, Basiak W, Switaj K, Zarnowska-Prymek amine and azithromycin vs pyrimethamine and sulfadiazine for the H, Olszynska-Krowicka M, et al. The Impact of Short-Term, Intensive treatment of ocular toxoplasmosis. Am J Ophthalmol. 2002;134:34–40. Antifolate Treatment (with Pyrimethamine and Sulfadoxine) and Antibiot‑ 39. Colin J, Harie J. Presumed toxoplasmic chorioretinitis: comparative study ics Followed by Long-Term, Secondary Antifolate Prophylaxis on the of treatment with pyrimethamine and sulfadiazine or clindamycin. J Fr Rate of Toxoplasmic Retinochoroiditis Recurrence. PLoS Negl Trop Dis. Ophtalmol. 1989;12:161–5. 2016;10:e0004892. 40. Ghavidel LA, Milani AE, Bagheri M. Comparison of azithromycin and 51. Shea BJ, Reeves BC, Wells G, Thuku M, Hamel C, Moran J, et al. AMSTAR 2: pyrimethamine/sulfadiazine treatment in ocular toxoplasmosis in North a critical appraisal tool for systematic reviews that include randomised West of Iran. Crescent J Med Biol Sci. 2017;4:80–4. or non-randomised studies of healthcare interventions, or both. bmj. 41. Kartasasmita A, Muntur WP, Enus S, Iskandar E. Rapid resolution of 2017;358:j4008. toxoplasma chorioretinitis treatment using quadruple therapy. Clin 52. Using the AMSTAR checklist for network meta-analysis: does it ft? | The Ophthalmol Auckl NZ. 2017;11:2133. 24th Cochrane Colloquium [Internet]. [cited 2020 Jan 28]. Available from: 42. Lashay A, Mirshahi A, Parandin N, Riazi Esfahani H, Mazloumi M, Reza http://​2016.​collo​quium.​cochr​ane.​org/​abstr​acts/​using-​amstar-​check​list-​ Lashay M, et al. A prospective randomized trial of azithromycin versus netwo​rk-​meta-​analy​sis-​does-​it-​ft trimethoprim/sulfamethoxazole in treatment of toxoplasmic retinocho‑ 53. Nikolakopoulou A, Higgins JP, Papakonstantinou T, Chaimani A, Del roiditis. J Curr Ophthalmol. 2017;29:120–5. Giovane C, Egger M, et al. CINeMA: An approach for assessing confdence 43. Ortega Larrocea G, Becerril Galindo M, Navarro Montes A, Arellanes GL. in the results of a network meta-analysis. PLoS Med. 2020;17:e1003082. Clindamicina vs pirimetamina en el tratamiento de la toxoplasmosis 54. Papakonstantinou T, Nikolakopoulou A, Higgins JP, Egger M, Salanti G. ocular activa. Rev Mex Oftalmol. 2000;74:5–10. CINeMA: Software for semiautomated assessment of the confdence in 44. Soheilian M, Ramezani A, Azimzadeh A, Sadoughi MM, Dehghan MH, the results of network meta‐analysis. Campbell Syst Rev. 2020;16:e1080. Shahghadami R, et al. Randomized trial of intravitreal clindamycin and dexamethasone versus pyrimethamine, sulfadiazine, and prednisolone in treatment of ocular toxoplasmosis. Ophthalmology. 2011;118:134–41. Publisher’s Note 45. Beck RW, Maguire MG, Bressler NM, Glassman AR, Lindblad AS, Ferris FL. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ Visual acuity as an outcome measure in clinical trials of retinal diseases. lished maps and institutional afliations. Ophthalmology. 2007;114:1804–9. 46. Ana Martínez Rubioj JGA. Valoración de la agudeza visual [Internet]. Rev Pediatr Aten Primaria Vol18 No71 Madr. Julsep 2016. 2016 [cited 2019

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions