Vision-threatening ocular toxoplasmosis involving the fovea diagnosed by intraocular fluid mNGS: A case report Authors: Sijian Xie, Li Wu, Ying Liu, Shangping Jiang, Yijing Yang, Qinghua Peng Published: 2026 ABSTRACT AbstractOcular toxoplasmosis is the most common cause of infectious posterior uveitis worldwide and may lead to irreversible visual impairment when the macula is involved. Definitive diagnosis remains challenging because clinical manifestations overlap with those of other infectious and inflammatory retinal disorders. We report a case of active ocular toxoplasmosis involving the fovea in an immunocompetent 44-year-old man presenting with progressive unilateral visual loss and central scotoma. Ophthalmic examination revealed keratic precipitates, mild vitritis, and a yellow-white inflammatory lesion involving the macular region. Fluorescein angiography demonstrated progressive hyperfluorescent staining, and optical coherence tomography showed a hyperreflective subfoveal lesion. Serological testing was positive for anti-Toxoplasma gondii IgM and IgG antibodies. Intraocular fluid analysis demonstrated positive anti-Toxoplasma IgG, while metagenomic next-generation sequencing (mNGS) directly identified Toxoplasma gondii DNA, providing molecular evidence supporting the diagnosis. The patient denied cat ownership or close feline exposure but reported occasional consumption of raw salmon. Treatment consisted of oral trimethoprim-sulfamethoxazole combined with methylprednisolone and two intravitreal injections of clindamycin plus dexamethasone. Significant regression of the retinal lesion was observed after treatment. This case highlights the diagnostic value of intraocular fluid mNGS in ocular toxoplasmosis and emphasizes the importance of considering ocular toxoplasmosis in patients presenting with posterior uveitis involving the macula, even in the absence of traditional epidemiological risk factors.Keywords: Ocular toxoplasmosis, Toxoplasma gondii, Metagenomic next-generation sequencing, Infectious uveitis, Retinochoroiditis Introduction Ocular toxoplasmosis (OT) is the most common infectious cause of posterior uveitis and a major cause of vision-threatening retinochoroiditis worldwide [1,2]. The disease results from infection with Toxoplasma gondii, an obligate intracellular protozoan parasite capable of establishing chronic tissue infection [1]. Although congenital infection has historically been considered the predominant cause of ocular disease, increasing evidence suggests that acquired infection contributes substantially to the global burden of OT [1]. The differential diagnosis of retinochoroiditis is broad and includes infectious causes such as cytomegalovirus (CMV) retinitis, acute retinal necrosis (ARN) caused by herpes simplex virus (HSV) or varicella-zoster virus (VZV), syphilitic posterior uveitis, tuberculous chorioretinitis, ocular toxocariasis, fungal chorioretinitis, and diffuse unilateral subacute neuroretinitis (DUSN), as well as non-infectious entities including sarcoidosis, Behçet disease, multifocal choroiditis, and primary vitreoretinal lymphoma [[3], [4], [5]]. The overlapping clinical features of these conditions necessitate a systematic diagnostic approach. Clinical diagnosis is often based on characteristic funduscopic findings, including focal necrotizing retinochoroiditis associated with vitreous inflammation [1,2]. However, atypical presentations, unusual lesion locations, and overlapping features with other infectious or inflammatory retinal diseases may complicate diagnosis [2]. Serological testing may support clinical suspicion but cannot independently establish ocular involvement because anti-Toxoplasma antibodies frequently persist following previous systemic infection [6]. Metagenomic next-generation sequencing (mNGS) has recently emerged as a promising diagnostic modality for infectious ocular diseases [[7], [8], [9]]. Unlike targeted molecular assays such as conventional polymerase chain reaction (PCR), mNGS enables unbiased pathogen detection directly from intraocular samples and can simultaneously screen for multiple pathogens in a single assay, which is particularly advantageous when the differential diagnosis is broad [7,8]. The methodology typically employs short-read sequencing platforms (e.g., Illumina and MGI) with library preparation optimized for low-biomass clinical samples, and pathogen identification is performed by aligning sequencing reads against comprehensive reference databases [9]. The diagnostic accuracy of mNGS for intraocular infections has been demonstrated in several studies, with reported diagnostic performance varying widely across studies, with sensitivities ranging from 38.4% to 100% and specificities ranging from 15.8% to 100%, depending on the pathogen burden and specimen quality [7,8]. However, mNGS is associated with higher costs compared with targeted PCR assays, and its cost-effectiveness is most favorable in diagnostically challenging cases where multiple pathogens are under consideration and conventional diagnostic approaches have been inconclusive [9]. We report a case of active fovea-involving ocular toxoplasmosis in an immunocompetent adult in whom intraocular fluid mNGS provided molecular evidence supporting the diagnosis and facilitated timely targeted treatment. Case presentation A 44-year-old Chinese man presented to the Department of Ophthalmology at Ningxiang Hospital of Traditional Chinese Medicine on March 25, 2026, with progressive visual loss and a central dark shadow in the left eye for more than 10 days. His medical history was notable only for myopia (refractive error). He denied previous ocular surgery, ocular trauma, diabetes mellitus, immunodeficiency, malignancy, or autoimmune disease. The patient reported no pet ownership, no close contact with cats or other felines, and no travel history to areas known for high toxoplasmosis prevalence. The patient resides in Hunan Province, China. In recent decades, the positivity rate of human toxoplasmosis infection in China has shown a continuous upward trend, reaching approximately 8.2% [6]. However, he reported consuming raw salmon approximately two to three times annually. It should be noted that fish are not recognized biological intermediate hosts of Toxoplasma gondii, and raw salmon consumption is not an established risk factor for ocular toxoplasmosis. This exposure is mentioned only as a possible but unconfirmed dietary factor [1]. On examination, uncorrected visual acuity was 0.06 (Snellen equivalent: 20/300) in the right eye and 0.02 (Snellen equivalent: 20/1000) in the left eye (the patient was moderately myopic; the right eye was otherwise normal and correctable to 1.0). Intraocular pressure was 21 mmHg in the right eye and 17 mmHg in the left eye. Slit-lamp examination revealed keratic precipitates in the left eye with anterior chamber inflammation graded as 1 + cells and 1 + flare according to the Standardization of Uveitis Nomenclature (SUN) criteria [10]. Fundus examination demonstrated a yellow-white inflammatory lesion involving the macular region, loss of the foveal reflex, and vitreous haze graded as 1 + according to the SUN criteria [10] (Fig. 1). An arrow in Fig. 1 indicates the area of vitreous haze. Initial ophthalmic findings. (A) Slit-lamp photograph showing keratic precipitates in the left eye. (B) Ultra-widefield fundus photograph showing a yellow-white inflammatory lesion centered at the macula. Arrow indicates the area of vitreous haze. Fluorescein angiography revealed early-phase hypofluorescent blockage, which refers to the masking of background choroidal fluorescence by the inflammatory infiltrate, followed by progressive staining and intense hyperfluorescence in the late phase, findings consistent with active inflammatory retinochoroiditis [2] (Fig. 2). The hyperfluorescence in the late phase represents leakage of fluorescein dye from damaged retinal vessels and the choriocapillaris at the site of active inflammation. An arrow in Fig. 2 highlights the main inflammatory lesion. Fluorescein angiography findings. (A) Early-phase hypofluorescent blockage. (B) Progressive staining during the intermediate phase. (C) Intense late-phase hyperfluorescence indicating active retinal inflammation. Arrow indicates the main inflammatory lesion. Optical coherence tomography (OCT) demonstrated a hyperreflective subfoveal lesion involving the central macula with disruption of normal retinal architecture (Fig. 3). Arrows in Fig. 3 indicate the inflammatory lesion and the area of foveal involvement. Optical coherence tomography of the macula. Hyperreflective inflammatory lesion involving the foveal region with disruption of normal retinal architecture. Arrows indicate the inflammatory lesion and foveal involvement. Laboratory investigations showed positive anti-Toxoplasma gondii IgM and IgG antibodies in serum. Quantitative serum anti-Toxoplasma gondii IgG was > 650.0 IU/mL, exceeding the upper limit of detection, and IgM was 24.60 COI (both positive; chemiluminescence and electrochemiluminescence immunoassays), indicating recent or active infection. The remaining TORCH panel provided no serological evidence of active viral retinitis: rubella virus IgM was equivocal (0.98 COI) with positive IgG, consistent with past infection or vaccination, while cytomegalovirus IgM and herpes simplex virus type 2 IgG were negative; the complete quantitative results are provided in Supplementary Table S1. Intraocular fluid analysis demonstrated positive anti-Toxoplasma IgG antibodies (5.63 U/mL; reference, <4 U/mL). The Goldmann–Witmer coefficient (GWC) was not calculated because paired serum and intraocular fluid samples with simultaneously determined antibody titers were not available. To further establish the etiology, intraocular fluid was submitted for mNGS. Sequencing directly identified Toxoplasma gondii DNA sequences, providing molecular evidence supporting the diagnosis of active ocular toxoplasmosis (Table 1). Table 1Metagenomic next-generation sequencing results from intraocular fluid. Sample type: Aqueous humor (left eye). Sequencing platform: MGISEQ-200 (MGI, Shenzhen, China); 50-bp single-end reads. Total sequencing reads: 78,118,076. Overall Q30 ratio: 95.03%.GenusReadsRel. Abund. (%)SpeciesInterpretationToxoplasma335255.40Toxoplasma gondiiHigh-priority pathogenDetection of Toxoplasma gondii DNA sequences (3352 species-specific sequencing reads; relative abundance, 55.40%). The sequence number 3352 represents the number of sequencing reads that were uniquely aligned to the Toxoplasma gondii reference genome at the species level. This is not a sequence deposit reference number. In addition, five low-abundance commensal or environmental organisms were detected (Cutibacterium acnes, 852 reads; Malassezia restricta, 240 reads; Staphylococcus epidermidis, 206 reads; Acinetobacter junii, 96 reads; Moraxella osloensis, 33 reads), which the laboratory interpreted as suspected background microbes rather than pathogens. Metagenomic next-generation sequencing results from intraocular fluid. Sample type: Aqueous humor (left eye). Sequencing platform: MGISEQ-200 (MGI, Shenzhen, China); 50-bp single-end reads. Total sequencing reads: 78,118,076. Overall Q30 ratio: 95.03%. Detection of Toxoplasma gondii DNA sequences (3352 species-specific sequencing reads; relative abundance, 55.40%). The sequence number 3352 represents the number of sequencing reads that were uniquely aligned to the Toxoplasma gondii reference genome at the species level. This is not a sequence deposit reference number. In addition, five low-abundance commensal or environmental organisms were detected (Cutibacterium acnes, 852 reads; Malassezia restricta, 240 reads; Staphylococcus epidermidis, 206 reads; Acinetobacter junii, 96 reads; Moraxella osloensis, 33 reads), which the laboratory interpreted as suspected background microbes rather than pathogens. mNGS methodology Intraocular fluid (aqueous humor, approximately 0.1–0.2 mL) was collected via anterior chamber paracentesis under aseptic conditions before the initiation of any anti-infective therapy; the patient had not received any antimicrobial treatment prior to sample collection. DNA-based mNGS was performed at a commercial clinical laboratory (Beijing GiantMed Medical Diagnostics Laboratory, Beijing, China). In brief, DNA was extracted using the QIAamp DNA Micro Kit and libraries were constructed using the QIAseq Ultralow Input Library Kit (QIAGEN, Hilden, Germany); sequencing was performed on the MGISEQ-200 platform (MGI, Shenzhen, China) with 50-bp single-end reads, yielding a total of 78,118,076 reads with an overall Q30 ratio of 95.03%. Negative (“no-template”) and positive controls were processed in parallel with each batch. The detailed procedures, quality-control metrics, bioinformatics pipeline, and positivity criteria are provided in the Supplementary Methods. Toxoplasma gondii was identified with 3352 species-specific sequencing reads (relative abundance, 55.40%) and was flagged as a high-priority pathogen. In addition, five low-abundance commensal or environmental organisms were detected and were interpreted by the laboratory as suspected background or commensal microbes rather than pathogens (Table 1); no other pathogen of clinical significance was identified, and no drug-resistance genes were detected. Additional laboratory testing demonstrated moderately elevated liver enzyme levels (alanine aminotransferase [ALT]: 141.6 U/L, reference range: 9.0–50.0 U/L; aspartate aminotransferase [AST]: 44.7 U/L, reference range: 15.0–40.0 U/L; gamma-glutamyl transferase [GGT]: 126.0 U/L, reference range: 10.0–60.0 U/L). Human immunodeficiency virus serology and syphilis testing were negative. The tuberculin skin test (PPD) was strongly positive. Antinuclear antibodies and extractable nuclear antigen antibodies were negative; anti-mitochondrial M2 antibody (AMA-M2) was positive but was considered unrelated to the ocular inflammation. Cytokine profiling of the aqueous humor by cytometric bead array demonstrated markedly elevated interleukin-6 (IL-6, 165.1 pg/mL) with interleukin-10 (IL-10) within the reference range (3.7 pg/mL), yielding an IL-10/IL-6 ratio of 0.0224; vascular cell adhesion molecule-1 (VCAM-1), vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and IL-8 were also elevated, supporting an active intraocular inflammatory process with disruption of the blood–ocular barrier (complete panel in Supplementary Table S1). Based on the clinical presentation, multimodal ocular imaging findings, serological results, and molecular evidence provided by mNGS, a diagnosis of active ocular toxoplasmosis involving the fovea was established. The patient received oral trimethoprim-sulfamethoxazole (trimethoprim 160 mg/sulfamethoxazole 800 mg, twice daily, for a total duration of 6 weeks) combined with oral methylprednisolone. Oral methylprednisolone was initiated at a dose of 32 mg once daily, starting 2 days after the initiation of systemic antibiotic therapy to ensure adequate antimicrobial coverage before immunosuppression, and was subsequently tapered over 6 weeks. This timing is consistent with the standard practice of initiating corticosteroid therapy 2–3 days after systemic antibiotic initiation to reduce the risk of parasite proliferation [11,12]. Because of the central location of the lesion and the risk of permanent visual loss, intravitreal clindamycin (1 mg/0.1 mL) combined with dexamethasone (400 μg/0.1 mL) was administered on March 30 and April 2, 2026. Intravitreal clindamycin is an uncommon adjunctive therapy reserved for vision-threatening lesions that encroach upon the fovea, as it provides high local drug concentrations while minimizing systemic adverse effects [11,12]. The known risks of intravitreal injections, including retinal toxicity and endophthalmitis, were carefully weighed against the risk of permanent central visual loss from untreated foveal toxoplasmosis; in this case, the benefits of local therapy were judged to outweigh the procedural risks. Progressive regression of the retinal lesion was observed following treatment. The patient was discharged on April 3, 2026, with continued oral therapy and scheduled outpatient follow-up. The follow-up OCT findings for the patient on April 8 (Day 14 after treatment), April 22 (Day 30 after treatment), and June 17 (3-month follow-up) are presented in Fig. 4. The best-corrected visual acuity (BCVA) of the left eye was 0.04 at discharge (April 3), 0.12 (Snellen equivalent: 20/160) on April 22, and 0.30 (Snellen equivalent: 20/63) at the 3-month follow-up (June 17), compared with the presenting uncorrected visual acuity of 0.02 (Snellen equivalent: 20/1000). Post-treatment follow-up findings. (A) Fundus photograph showing regression of the inflammatory lesion. (B) Optical coherence tomography demonstrating improvement of retinal morphology following treatment. Follow-up time points: Day 14 (April 8, 2026), Day 30 (April 22, 2026), and 3-month follow-up (June 17, 2026). Discussion This case highlights the diagnostic value of intraocular fluid mNGS in vision-threatening ocular toxoplasmosis involving the fovea [[7], [8], [9]]. Ocular toxoplasmosis is usually diagnosed on the basis of characteristic necrotizing retinochoroiditis supported by serological findings [1,2]. However, diagnosis may be challenging when the lesion is located in the posterior pole or directly involves the foveal region. In the present case, the macular location was clinically important because foveal involvement is associated with a high risk of irreversible central visual impairment [13]. Therefore, rapid etiological clarification and prompt treatment were essential. The differential diagnosis of unilateral posterior uveitis with a macular inflammatory lesion includes several infectious and non-infectious entities. Cytomegalovirus retinitis was considered, but it usually occurs in severely immunocompromised patients and is characterized by extensive retinal necrosis with hemorrhage [3]. Acute retinal necrosis, most commonly caused by herpes simplex virus or varicella-zoster virus, typically presents with peripheral retinal necrosis, occlusive vasculitis, and severe intraocular inflammation [4]. Syphilitic posterior uveitis may mimic many forms of intraocular inflammation, but serological testing was negative in this patient [5]. Tuberculous chorioretinitis was also considered. Although the tuberculin skin test was strongly positive, suggesting latent tuberculosis infection (which is not uncommon in Chinese adults), the patient had no systemic symptoms, no radiographic evidence of active pulmonary tuberculosis, and chorioretinal morphology and angiographic findings that were not characteristic of tuberculous chorioretinitis; moreover, the lesion resolved completely with anti-toxoplasma therapy without anti-tuberculous treatment, arguing against this diagnosis [14]. Additional differential diagnoses were considered given the unilateral macular involvement. Ocular toxocariasis typically presents with a peripheral or posterior pole granuloma with vitreous bands and occurs predominantly in children with a history of exposure to dogs or soil contaminated with Toxocara eggs; our patient was a middle-aged adult without such exposure history. Fungal chorioretinitis, most commonly caused by Candida species, is usually associated with risk factors such as immunosuppression, indwelling catheters, or intravenous drug use [15], none of which were present in our patient. Granulomatous uveitis associated with sarcoidosis was considered unlikely because of the absence of bilateral hilar lymphadenopathy, pulmonary infiltrates, or elevated serum angiotensin-converting enzyme levels [16]. Multifocal choroiditis typically presents with multiple lesions in both eyes rather than a solitary macular lesion [17], and primary vitreoretinal lymphoma, which can masquerade as chronic uveitis, was considered less likely given the acute onset, the absence of vitreous clumps, and the favorable response to anti-infective therapy; moreover, the intraocular IL-10/IL-6 ratio (0.0224, well below 1) and the normal IL-10 level (3.7 pg/mL) argued against a lymphomatous masquerade, in which an elevated IL-10/IL-6 ratio is typically observed [18]. In addition, intraocular anti-Toxocara IgG was negative (1.29 U; reference, <1.8 U), providing further evidence against ocular toxocariasis. Non-infectious causes such as Behçet disease and sarcoidosis were unlikely because there were no compatible systemic manifestations. The final diagnosis was supported by the combination of multimodal imaging, positive serum and intraocular anti-Toxoplasma antibodies, and direct detection of Toxoplasma gondii DNA in intraocular fluid by mNGS [1,2]. A key strength of this case is the molecular evidence provided by mNGS. Although positive anti-Toxoplasma IgM and IgG antibodies supported prior or recent exposure, serology alone cannot prove active ocular infection because antibodies may persist for years after systemic infection [6]. In contrast, detection of Toxoplasma gondii DNA directly from intraocular fluid provided strong etiological evidence. Compared with targeted PCR assays, mNGS offers several advantages: (1) it enables unbiased, hypothesis-free pathogen detection without requiring a priori suspicion of a specific pathogen; (2) it can simultaneously detect multiple pathogens, including bacteria, viruses, fungi, and parasites, in a single assay; and (3) it is particularly valuable when the differential diagnosis is broad and conventional diagnostic tests have been inconclusive [7,8]. However, the cost-effectiveness of mNGS must be carefully considered. The cost of mNGS is substantially higher than that of targeted PCR assays (approximately 3–10 times higher, depending on the platform and sequencing depth). Therefore, mNGS is most appropriately used in diagnostically challenging cases where multiple infectious etiologies are under consideration, conventional PCR panels have been negative or are not available, and a definitive microbiological diagnosis is essential to guide targeted therapy [9]. In the present case, the macular location of the lesion, the risk of irreversible visual loss, and the broad differential diagnosis justified the use of mNGS as an adjunctive diagnostic tool. In this case, mNGS increased diagnostic confidence, helped exclude alternative infectious etiologies, and supported timely initiation of targeted anti-toxoplasma therapy. Another notable feature is the absence of classical epidemiological risk factors. The patient denied cat ownership or close feline exposure but reported occasional consumption of raw salmon. This exposure has no established causal association with ocular toxoplasmosis and should be regarded only as a possible but unconfirmed exposure. Fish are not recognized biological intermediate hosts of Toxoplasma gondii, and contamination during food processing or preparation remains a speculative route of foodborne transmission [1]. In a recent comprehensive review, Miyagaki et al. (2024) noted that foodborne transmission of toxoplasmosis is primarily associated with consumption of undercooked meat from infected intermediate hosts (e.g., pork, lamb) or ingestion of oocysts from contaminated water or produce, with no definitive evidence linking fish consumption to toxoplasmosis acquisition [1]. Toxoplasma gondii exhibits three major clonal lineages (Types I, II, and III) that differ in their geographic distribution and clinical phenotype. In North America and Europe, Type II strains predominate and are most commonly associated with human infections, including ocular toxoplasmosis. In South America, atypical and recombinant genotypes are more prevalent and are associated with more severe ocular disease [19]. In China, the predominant genotype is the Chinese 1 (ToxoDB #9) lineage, which is distinct from the classical North American and European clonal types [20]. Although genotyping was not performed in the present case, future studies incorporating mNGS-based genotyping of intraocular Toxoplasma gondii isolates may provide valuable epidemiological and clinical insights. The patient responded favorably to combined systemic and intravitreal therapy [11,12]. Because the lesion involved the fovea, aggressive treatment was clinically justified to reduce the risk of permanent visual loss [13]. Anatomical regression of the lesion and improvement in best-corrected visual acuity of the left eye from 0.04 at discharge to 0.30 at the latest follow-up (presenting uncorrected visual acuity, 0.02) was consistent with a favorable anatomical and functional response following treatment. However, as this is a single case report, conclusions regarding treatment efficacy cannot be drawn. This report has several limitations. Goldmann–Witmer coefficient analysis was not performed because paired serum and intraocular fluid antibody titers were not simultaneously available. The source of infection could not be definitively established. The mNGS analysis was performed without concurrent targeted PCR confirmation, and genotyping of the Toxoplasma gondii isolate was not performed. Nevertheless, direct identification of Toxoplasma gondii DNA by mNGS, together with compatible clinical and imaging findings and a favorable therapeutic response, provided strong support for the diagnosis. Longer follow-up remains necessary because ocular toxoplasmosis may recur even after successful treatment [2,11]. Conclusion Ocular toxoplasmosis should be considered in patients presenting with unilateral posterior uveitis and macular inflammatory lesions, even in the absence of traditional epidemiological risk factors. Intraocular fluid mNGS may provide valuable molecular evidence to support the diagnosis and facilitate timely targeted treatment in diagnostically challenging cases. The cost-effectiveness of mNGS is most favorable when the differential diagnosis is broad and conventional diagnostic approaches have been inconclusive. This case further highlights the potential role of advanced molecular diagnostics in the evaluation of infectious uveitis. CRediT authorship contribution statement Sijian Xie: Writing – original draft. Li Wu: Writing – original draft. Ying Liu: Data curation. Shangping Jiang: Data curation. Yijing Yang: Writing – review & editing. Qinghua Peng: Supervision. Ethics declaration Written informed consent to take part in the study and to publish the article has been obtained from all participants or their legal representatives. The privacy rights of participants have been observed. This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. This study was approved by the Ethics Committee of Ningxiang Traditional Chinese Medicine Hospital. (Approval No. 2026041002) This research follows the CARE guidelines and the CARE checklist. Ethics approval and consent to participate: This study complied with the Declaration of Helsinki and was authorized by the Ethics Committee of Ningxiang Hospital of Traditional Chinese Medicine (No. 2026041002). Written informed consent was obtained from the patient for publication of this case report and accompanying images. Declaration of Generative AI and AI-assisted technologies in the writing process During the preparation of this work the authors used an AI-assisted language editing tool solely for language polishing and clarity improvement. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article. Funding This research was supported by Scientific Research Project of the Education Department of Hunan Province (Project ID: 25B0352) and the Teaching Reform Research Project of Ordinary Higher Education Institutions in Hunan Province (Project ID: HNJG-20230513). Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements We sincerely appreciate the guidance provided by Professor Jing Lu. Footnotes Supplementary data associated with this article can be found in the online version at doi:10.1016/j.idcr.2026.e02741. Supplementary material Supplementary material Data availability The data supporting the findings of this study are available from the corresponding author upon reasonable request. They are not publicly available owing to patient privacy considerations. References Associated Data Supplementary Materials Supplementary material Data Availability Statement The data supporting the findings of this study are available from the corresponding author upon reasonable request. They are not publicly available owing to patient privacy considerations.