Validation of a deep learning system for the detection of diabetic retinopathy in Indigenous Australians.

BACKGROUND/AIMS: Deep learning systems (DLSs) for diabetic retinopathy (DR) detection show promising results but can underperform in racial and ethnic minority groups, therefore external validation within these populations is critical for health equity. This study evaluates the performance of a DLS for DR detection among Indigenous Australians, an understudied ethnic group who suffer disproportionately from DR-related blindness. METHODS: We performed a retrospective external validation study comparing the performance of a DLS against a retinal specialist for the detection of more-than-mild DR (mtmDR), vision-threatening DR (vtDR) and all-cause referable DR. The validation set consisted of 1682 consecutive, single-field, macula-centred retinal photographs from 864 patients with diabetes (mean age 54.9 years, 52.4% women) at an Indigenous primary care service in Perth, Australia. Three-person adjudication by a panel of specialists served as the reference standard. RESULTS: For mtmDR detection, sensitivity of the DLS was superior to the retina specialist (98.0% (95% CI, 96.5 to 99.4) vs 87.1% (95% CI, 83.6 to 90.6), McNemar's test p<0.001) with a small reduction in specificity (95.1% (95% CI, 93.6 to 96.4) vs 97.0% (95% CI, 95.9 to 98.0), p=0.006). For vtDR, the DLS's sensitivity was again superior to the human grader (96.2% (95% CI, 93.4 to 98.6) vs 84.4% (95% CI, 79.7 to 89.2), p<0.001) with a slight drop in specificity (95.8% (95% CI, 94.6 to 96.9) vs 97.8% (95% CI, 96.9 to 98.6), p=0.002). For all-cause referable DR, there was a substantial increase in sensitivity (93.7% (95% CI, 91.8 to 95.5) vs 74.4% (95% CI, 71.1 to 77.5), p<0.001) and a smaller reduction in specificity (91.7% (95% CI, 90.0 to 93.3) vs 96.3% (95% CI, 95.2 to 97.4), p<0.001). CONCLUSION: The DLS showed improved sensitivity and similar specificity compared with a retina specialist for DR detection. This demonstrates its potential to support DR screening among Indigenous Australians, an underserved population with a high burden of diabetic eye disease.

Risk Stratification for Diabetic Retinopathy Screening Order Using Deep Learning: A Multicenter Prospective Study.

Purpose: Real-world evaluation of a deep learning model that prioritizes patients based on risk of progression to moderate or worse (MOD+) diabetic retinopathy (DR). Methods: This nonrandomized, single-arm, prospective, interventional study included patients attending DR screening at four centers across Thailand from September 2019 to January 2020, with mild or no DR. Fundus photographs were input into the model, and patients were scheduled for their subsequent screening from September 2020 to January 2021 in order of predicted risk. Evaluation focused on model sensitivity, defined as correctly ranking patients that developed MOD+ within the first 50% of subsequent screens. Results: We analyzed 1,757 patients, of which 52 (3.0%) developed MOD+. Using the model-proposed order, the model's sensitivity was 90.4%. Both the model-proposed order and mild/no DR plus HbA1c had significantly higher sensitivity than the random order (P < 0.001). Excluding one major (rural) site that had practical implementation challenges, the remaining sites included 567 patients and 15 (2.6%) developed MOD+. Here, the model-proposed order achieved 86.7% versus 73.3% for the ranking that used DR grade and hemoglobin A1c. Conclusions: The model can help prioritize follow-up visits for the largest subgroups of DR patients (those with no or mild DR). Further research is needed to evaluate the impact on clinical management and outcomes. Translational Relevance: Deep learning demonstrated potential for risk stratification in DR screening. However, real-world practicalities must be resolved to fully realize the benefit.

Recent evidence of economic evaluation of artificial intelligence in ophthalmology.

PURPOSE OF REVIEW: Health economic evaluation (HEE) is essential for assessing value of health interventions, including artificial intelligence. Recent approaches, current challenges, and future directions of HEE of artificial intelligence in ophthalmology are reviewed. RECENT FINDINGS: Majority of recent HEEs of artificial intelligence in ophthalmology were for diabetic retinopathy screening. Two models, one conducted in the rural USA (5-year period) and another in China (35-year period), found artificial intelligence to be more cost-effective than without screening for diabetic retinopathy. Two additional models, which compared artificial intelligence with human screeners in Brazil and Thailand for the lifetime of patients, found artificial intelligence to be more expensive from a healthcare system perspective. In the Thailand analysis, however, artificial intelligence was less expensive when opportunity loss from blindness was included. An artificial intelligence model for screening retinopathy of prematurity was cost-effective in the USA. A model for screening age-related macular degeneration in Japan and another for primary angle close in China did not find artificial intelligence to be cost-effective, compared with no screening. The costs of artificial intelligence varied widely in these models. SUMMARY: Like other medical fields, there is limited evidence in assessing the value of artificial intelligence in ophthalmology and more appropriate HEE models are needed.

Cost-Utility Analysis of Deep Learning and Trained Human Graders for Diabetic Retinopathy Screening in a Nationwide Program.

INTRODUCTION: Deep learning (DL) for screening diabetic retinopathy (DR) has the potential to address limited healthcare resources by enabling expanded access to healthcare. However, there is still limited health economic evaluation, particularly in low- and middle-income countries, on this subject to aid decision-making for DL adoption. METHODS: In the context of a middle-income country (MIC), using Thailand as a model, we constructed a decision tree-Markov hybrid model to estimate lifetime costs and outcomes of Thailand's national DR screening program via DL and trained human graders (HG). We calculated the incremental cost-effectiveness ratio (ICER) between the two strategies. Sensitivity analyses were performed to probe the influence of modeling parameters. RESULTS: From a societal perspective, screening with DL was associated with a reduction in costs of ~ US$ 2.70, similar quality-adjusted life-years (QALY) of + 0.0043, and an incremental net monetary benefit of ~ US$ 24.10 in the base case. In sensitivity analysis, DL remained cost-effective even with a price increase from US$ 1.00 to US$ 4.00 per patient at a Thai willingness-to-pay threshold of ~ US$ 4.997 per QALY gained. When further incorporating recent findings suggesting improved compliance to treatment referral with DL, our analysis models effectiveness benefits of ~ US$ 20 to US$ 50 depending on compliance. CONCLUSION: DR screening using DL in an MIC using Thailand as a model may result in societal cost-savings and similar health outcomes compared with HG. This study may provide an economic rationale to expand DL-based DR screening in MICs as an alternative solution for limited availability of skilled human resources for primary screening, particularly in MICs with similar prevalence of diabetes and low compliance to referrals for treatment.

Impact of radical dose escalation by endoluminal brachytherapy and induction chemotherapy in the definitive treatment of locally advanced esophageal cancer - A retrospective study.

Background: The loco-regional recurrence rate remains the main concern in the treatment of esophageal cancer. However, there are controversial data regarding the benefit of dose escalation in the treatment of esophageal cancer. The study examines the implications of dose escalation with endoluminal brachytherapy after induction chemotherapy and definitive chemoradiation in cases of carcinoma esophagus. Material and Methods: Total 31 biopsy-proven patients with inoperable, locally advanced esophageal cancer of stage IIA-IIIB were enrolled from January 2006 till December 2018. All patients underwent two cycles of three weekly induction chemotherapy followed by definitive external beam radiotherapy of 45-50.4 Gray (Gy) at 1.8 Gy per fraction along with chemotherapy, followed by intraluminal brachytherapy boost of two fractions with 5 Gy each. Overall survival (OS) was censored at death or the last follow-up. Results: Of 31 patients, 26 (83.97%) received concurrent chemotherapy and 30 (96.77%) completed radiation therapy. At the end of 3 months, 10 (32.2%), 13 (41.9%), 5 (16.1%), and 3 (9.6%) had complete response, partial response, stable disease, and progression of the disease, respectively. Distal failure was seen in five (16.1%) cases. The median OS was 28 months. OS at 2 years and 5 years was 20 (64.5%) and 9 (28.3%), respectively. At the end of 3 months, 17 (54.8%) of patients had no dysphagia, four (12.9%) of patients had improvement of more than 2 points in dysphagia score, five (16.1%) of patients had no change in the score and five (16.1%) of the patients had worsening of the dysphagia score by 1 point. Median dysphagia-free survival was 10.7 months, eight (25.8%) developed dysphagia after the dysphagia-free interval and two (6.4%) had worsening of dysphagia after treatment. There were no reported grade III or grade IV acute toxicities. Conclusion: The protocol has shown an acceptable survival and dysphagia-free interval. The study suggests intraluminal brachytherapy with induction chemotherapy and definitive chemoradiation is a feasible option in locally advanced esophageal cancer.

Central Pontine Myelinolysis: A Case Report.

Central pontine myelinolysis (CPM) classically occurs due to rapid rise in serum osmolarity. Most cases have been associated with a history of chronic alcohol abuse, malnutrition, diuretic abuse, and hyponatremia. The pathological process of CPM starts in the central pons near median raphe and spreads out "like a brush Fire" into the surrounding basis pontis. Extrapontine sites such as internal capsule, basal ganglia, cerebellum, and cerebrum can also be affected. We report a case of 60-year-old male with history of chronic alcoholism who presented to us with severe neurological deficits 10 days after his episode of severe hyponatremia. How to cite this article: Tiwari R, Kumari A. Central Pontine Myelinolysis: A Case Report. Indian J Crit Care Med 2022;26(9):1049-1051.

The Swallow Tail Sign of Substantia Nigra: A Case-Control Study to Establish Its Role in Diagnosis of Parkinson Disease on 3T MRI.

Background and Objectives The loss of swallow tail sign (STS) has been studied for the diagnosis of Parkinson's disease (PD). The study aims to establish the role of STS on high-resolution 3D susceptibility-weighted images (SWI) on 3T MRI in clinically diagnosed cases of PD and compare with control population. Methods and Materials Forty-five patients with clinically diagnosed PD and Parkinson plus syndrome (PPS) formed the study group and were compared with 45 controls without any neurological disease and normal brain magnetic resonance imaging (MRI). Presence or absence of STS was studied on 1-mm thick axial 3D SWI images in bilateral substantia nigra by two radiologists independently, followed by consensus reading. Bilateral absent, unilateral absent, and faintly present STS were considered as absent STS and predicted PD or PPS, and bilateral presence was considered as a positive STS, and was assessed keeping the clinical diagnosis as the gold standard. Results The sensitivity of the absent STS was 75.55%, specificity 97.77%, positive predictive value 97.14%, negative predictive value 80% and accuracy 86.66%, in the diagnosis of PD or PPS, with odd ratio of 132 (confidence interval 15.97-1098.75). Kappa coefficient was 0.80 ( p < 0.001) for both inter- and intrarater agreement, suggesting high reproducibility for the detection of STS. Conclusions Absence of the STS is a good predictor of degeneration of the nigrosome 1 in the substantia nigra in the PD or PPS patients; hence, it can act as a useful marker of these diseases.

Real-time diabetic retinopathy screening by deep learning in a multisite national screening programme: a prospective interventional cohort study.

BACKGROUND: Diabetic retinopathy is a leading cause of preventable blindness, especially in low-income and middle-income countries (LMICs). Deep-learning systems have the potential to enhance diabetic retinopathy screenings in these settings, yet prospective studies assessing their usability and performance are scarce. METHODS: We did a prospective interventional cohort study to evaluate the real-world performance and feasibility of deploying a deep-learning system into the health-care system of Thailand. Patients with diabetes and listed on the national diabetes registry, aged 18 years or older, able to have their fundus photograph taken for at least one eye, and due for screening as per the Thai Ministry of Public Health guidelines were eligible for inclusion. Eligible patients were screened with the deep-learning system at nine primary care sites under Thailand's national diabetic retinopathy screening programme. Patients with a previous diagnosis of diabetic macular oedema, severe non-proliferative diabetic retinopathy, or proliferative diabetic retinopathy; previous laser treatment of the retina or retinal surgery; other non-diabetic retinopathy eye disease requiring referral to an ophthalmologist; or inability to have fundus photograph taken of both eyes for any reason were excluded. Deep-learning system-based interpretations of patient fundus images and referral recommendations were provided in real time. As a safety mechanism, regional retina specialists over-read each image. Performance of the deep-learning system (accuracy, sensitivity, specificity, positive predictive value [PPV], and negative predictive value [NPV]) were measured against an adjudicated reference standard, provided by fellowship-trained retina specialists. This study is registered with the Thai national clinical trials registry, TCRT20190902002. FINDINGS: Between Dec 12, 2018, and March 29, 2020, 7940 patients were screened for inclusion. 7651 (96·3%) patients were eligible for study analysis, and 2412 (31·5%) patients were referred for diabetic retinopathy, diabetic macular oedema, ungradable images, or low visual acuity. For vision-threatening diabetic retinopathy, the deep-learning system had an accuracy of 94·7% (95% CI 93·0-96·2), sensitivity of 91·4% (87·1-95·0), and specificity of 95·4% (94·1-96·7). The retina specialist over-readers had an accuracy of 93·5 (91·7-95·0; p=0·17), a sensitivity of 84·8% (79·4-90·0; p=0·024), and specificity of 95·5% (94·1-96·7; p=0·98). The PPV for the deep-learning system was 79·2 (95% CI 73·8-84·3) compared with 75·6 (69·8-81·1) for the over-readers. The NPV for the deep-learning system was 95·5 (92·8-97·9) compared with 92·4 (89·3-95·5) for the over-readers. INTERPRETATION: A deep-learning system can deliver real-time diabetic retinopathy detection capability similar to retina specialists in community-based screening settings. Socioenvironmental factors and workflows must be taken into consideration when implementing a deep-learning system within a large-scale screening programme in LMICs. FUNDING: Google and Rajavithi Hospital, Bangkok, Thailand. TRANSLATION: For the Thai translation of the abstract see Supplementary Materials section.

Role of Chest X-Ray in Coronavirus Disease and Correlation of Radiological Features with Clinical Outcomes in Indian Patients.

Recent literature has reported that radiological features of coronavirus disease (COVID-19) patients are influenced by computed tomography. This study aimed to assess the characteristic chest X-ray features of COVID-19 and correlate them with clinical outcomes of patients. This retrospective study included 120 COVID-19 patients. Baseline chest X-rays and serial chest X-rays were reviewed. A severity index in the form of maximum radiological assessment of lung edema (RALE) score was calculated for each lung, and scores of both the lungs were summed to obtain a final score. The mean ± standard deviation (SD) and frequency (%) were determined, and an unpaired t test, Spearman's rank correlation coefficient, and logistic regression analyses were performed for statistical analyses. Among 120 COVID-19 patients, 74 (61.67%) and 46 (38.33%) were males and females, respectively; 64 patients (53.33%) had ground-glass opacities (GGO), 55 (45.83%) had consolidation, and 38 (31.67%) had reticular-nodular opacities, with lower zone distribution (50%) and peripheral distribution (41.67%). Baseline chest X-ray showed a sensitivity of 63.3% in diagnosing typical findings of SARS-CoV-2 pneumonia. The maximum RALE score was 2.13 ± 1.9 in hospitalized patients and 0.57 ± 0.77 in discharged patients (p value <0.0001). Spearman's rank correlation coefficient between maximum RALE score and clinical outcome parameters was as follows: age, 0.721 (p value <0.00001); >10 days of hospital stay, 0.5478 (p value <0.05); ≤10 days of hospital stay, 0.5384 (p value <0.0001); discharged patients, 0.5433 (p value <0.0001); and death, 0.6182 (p value = 0.0568). The logistic regression analysis revealed that maximum RALE scores (0.0932 [0.024-0.367]), (10.730 [2.727-42.206]), (1.258 [0.990-1.598]), and (0.794 [0.625-1.009]) predicted discharge, death, >10 days of hospital stay, and ≤10 days of hospital stay, respectively. The study findings suggested that the RALE score can quantify the extent of COVID-19 and can predict the prognosis of patients.

Whipple's disease: Rare case of malabsorption with hyperpigmentation in a female.

Whipple's disease is a multisystem disorder and responds well to antibiotic therapy if treated timely. It is seen in the fourth to fifth decades of life with a male to female ratio of 10:1. It mainly affects the intestine, the central nervous system, and joints. To the best of our knowledge, we present the first case of duodenal Whipple's disease in an Indian female, who presented with hyperpigmentation and chronic diarrhea with malabsorption. Whipple's disease was diagnosed based on specific upper GI endoscopic and histopathology findings.

Impact of nodal boost irradiation and MR-based brachytherapy on oncologic outcomes in node-positive cervical cancer.

This study aimed to prospectively evaluate the impact of dose-escalated irradiation of nodal metastases on clinical outcomes compared to no boost in patients with node-positive, bulky, locally advanced cervical cancer (LACC) undergoing standard chemoradiation and MRI-based brachytherapy. METHODS: This comparative study included 161 patients with node-positive LACC treated with definitive chemoradiation and MRI-based brachytherapy. The prospective Boost arm accrued 71 patients to receive nodal boost either sequentially or simultaneously to an equivalent dose of 60 Gy. The control arm comprised 90 patients treated before this protocol period with no additional nodal boost. RESULT: Baseline patient and tumor characteristics were similar in both groups. All patients had at least one tumor dimension >5 cm at presentation, and 31% had para-aortic node involvement. With a median follow-up of 36 months (IQR:19-50.5), the overall 3-year Local control rate was 88.8%. The 3-year Regional control (93% vs. 80%, p = 0.035) was statistically better in the Boost arm. No nodal failure was observed in nodes <3 cc and < 2 cm, even in the No-boost arm. There was no significant difference in Disease-free survival (67.6% vs. 58.9%,p = 0.454) and Overall Survival (78.9% vs. 74.4%,p = 0.87) between the two arms. Incidence of acute or late toxicities did not differ significantly with nodal boost or the boost delivery technique. CONCLUSION: The addition of external radiation nodal boost to standard treatment of high-volume cervical cancer has improved pelvic control with an acceptable rate of toxicities. However, high systemic failures continue to pose a challenge in improving survival outcomes.

Magnetic resonance imaging of perianal fistulas in clinically symptomatic patients and the correlation among plain, contrast, and post jelly MRI fistulography

Abstract Objectives To evaluate different types of perianal fistulas and their complications on magnetic resonance imaging (MRI) and to compare plain, contrast, and jelly magnetic resonance fistulography findings. Materials and Methods This prospective study was performed in 30 patients who presented with perianal pus discharge or external fistulous opening. Magnetic resonance imaging of the perianal region before and after giving intravenous contrast and after injecting jelly through a percutaneous opening was performed on a 3T scanner and the results were correlated. Results The mean age of the patients was 40.13±13.88 years (range 19-75 years). The male to female ratio was 14:1. The most common type of fistula was St. James classification type I, which was seen in 13 patients (43%), followed by type IV in 30%, type III in 16%, type II in 6.66%, and type V in 3.33% of the patients. Using agreement analysis, we compared the number of primary and secondary tracts, internal openings, and horseshoe tracts and found a significant agreement between plain and post Jelly MRI fistulography (kappa statistic close to 1).When comparing plain and contrast MRI, there was significant agreement in the primary and secondary tracts, while statistically insignificant results were obtained (p>0.05) for the horseshoe tract and internal openings. Contrast injection was helpful in 7 subjects (23.3%) as peripheral enhancement of abscesses were better delineated. Conclusion Magnetic resonance imaging is the one stop diagnostic modality for perianal fistulas. Acquisition of axial (Ax) T2, axial T2 FS, coronal T2 and coronal T2 FS sequences without administering intravenous contrast or jelly is usually sufficient for the diagnosis of fistulas and their complications.

Generation of a tissue-specific transgenic model for K8 phosphomutants: A tool to investigate the role of K8 phosphorylation during skin carcinogenesis in vivo.

Keratin 8/18, the predominant keratin pair of simple epithelia, is known to be aberrantly expressed in several squamous cell carcinomas (SCCs), where its expression is often correlated with increased invasion, neoplastic progression, and poor prognosis. The majority of keratin 8/18 structural and regulatory functions are governed by posttranslational modifications, particularly phosphorylation. Apart from filament reorganization, cellular processes including cell cycle, cell growth, cellular stress, and apoptosis are known to be orchestrated by K8 phosphorylation at specific residues in the head and tail domains. Even though deregulation of K8 phosphorylation at two significant sites (Serine73 /Serine431 ) has been implicated in neoplastic progression of SCCs by various in vitro studies, including ours, it is reported to be highly context-dependent. Therefore, to delineate the precise role of Kereatin 8 phosphorylation in cancer initiation and progression, we have developed the tissue-specific transgenic mouse model expressing Keratin 8 wild type and phosphodead mutants under Keratin 14 promoter. Subjecting these mice to 7,12-dimethylbenz(a)anthracene/12-O-tetradecanoylphorbol-13-acetate-mediated skin carcinogenesis revealed that Keratin 8 phosphorylation may lead to an early onset of tumors compared to Keratin 8 wild-type expressing mice. Conclusively, the transgenic mouse model developed in the present study ascertained a positive impact of Keratin 8 phosphorylation on the neoplastic transformation of skin-squamous cells.

A comparative study evaluating the dose volume parameters in 3D conformal radiation of left sided whole breast irradiation including regional lymphnodes - a need of resource constrained countries.

BACKGROUND: The purpose of this study was to compare four 3D conformal radiation techniques in treatment of left breast cancer patients. MATERIALS AND METHODS: Radiation was planned for 20 patients to the left breast and regional lymph nodes using four techniques: partially wide tangents, photon-photon mix, photon-electron mix and 30/70 photon-electron mix. All plans were evaluated for internal mammary nodes (IMN) coverage, hotspot and normal tissue constraints. RESULT: The 85% of planning target volume (PTV) coverage was lesser for upper IMN than the lower IMN (below the lower border of the clavicular head) for all four techniques. The lower IMN coverage was better for partially wide tangent (80.46%) and photon-photon mix (88.88%). The lowest value of hotspot was seen in the partially wide tangent technique (112.69% ± 1.92). Hotspot is unacceptably high in both photon-electron mix and 30/70 photon-electron mix (> 120%). Left lung mean dose for all techniques on a pair-wise comparison showed no statistical difference. Left lung V20 values for partially wide tangent was 37.56% ± 8.17 and for photon-photon mix it was 40.49% ± 3.36. The mean heart dose with partially wide tangent was 9.43 ± 3.15 Gy and with photon-photon mix it was 10.10 ± 2.70 Gy. The mean heart dose for photon-electron mix was 7.56 ± 1.95 Gy and for 30/70 photon-electron mix it was 7.98 ± 2.16 Gy. CONCLUSION: No single technique satisfies all the criteria. The decision should be made on a case-by-case basis, considering the anatomy of the patient, availability of electron facilities and setup accuracy and reproducibility.

Longitudinal Screening for Diabetic Retinopathy in a Nationwide Screening Program: Comparing Deep Learning and Human Graders.

OBJECTIVE: To evaluate diabetic retinopathy (DR) screening via deep learning (DL) and trained human graders (HG) in a longitudinal cohort, as case spectrum shifts based on treatment referral and new-onset DR. METHODS: We randomly selected patients with diabetes screened twice, two years apart within a nationwide screening program. The reference standard was established via adjudication by retina specialists. Each patient's color fundus photographs were graded, and a patient was considered as having sight-threatening DR (STDR) if the worse eye had severe nonproliferative DR, proliferative DR, or diabetic macular edema. We compared DR screening via two modalities: DL and HG. For each modality, we simulated treatment referral by excluding patients with detected STDR from the second screening using that modality. RESULTS: There were 5,738 patients (12.3% STDR) in the first screening. DL and HG captured different numbers of STDR cases, and after simulated referral and excluding ungradable cases, 4,148 and 4,263 patients remained in the second screening, respectively. The STDR prevalence at the second screening was 5.1% and 6.8% for DL- and HG-based screening, respectively. Along with the prevalence decrease, the sensitivity for both modalities decreased from the first to the second screening (DL: from 95% to 90%, p = 0.008; HG: from 74% to 57%, p < 0.001). At both the first and second screenings, the rate of false negatives for the DL was a fifth that of HG (0.5-0.6% vs. 2.9-3.2%). CONCLUSION: On 2-year longitudinal follow-up of a DR screening cohort, STDR prevalence decreased for both DL- and HG-based screening. Follow-up screenings in longitudinal DR screening can be more difficult and induce lower sensitivity for both DL and HG, though the false negative rate was substantially lower for DL. Our data may be useful for health-economics analyses of longitudinal screening settings.

Development of a novel outcome prediction score (PEDSS) for pediatric convulsive status epilepticus: A longitudinal observational study.

OBJECTIVE: The objectives of this study were to evaluate ENDIT score and develop a novel outcome prediction score for outcome of pediatric convulsive status epilepticus (CSE) at the hospital and 3 months postdischarge. METHODS: Children and adolescents aged 1 month to 14 years, presenting with CSE to a tertiary care teaching center in North India from January 2017 to March 2019, were screened for enrollment. In-hospital and 3-month postdischarge outcome were defined as poor if Pediatric Cerebral Performance Category Scale (PCPCS) score dropped by ≥2 levels. RESULTS: Overall, 61 patients were enrolled for final analysis after applying exclusion and inclusion criteria. The area under the receiver operating characteristic (ROC) curve for ENDIT score in predicting mortality and differentiating good from poor outcome at the hospital and at 3 months postdischarge was 0.74 (95% confidence interval [CI] = 0.58-0.89), 0.7 (95% CI = 0.57-0.83), and 0.72 (95% CI = 0.6-0.82), respectively. Based on predictors in the present cohort that were significantly different between good and poor outcome cases at the hospital and 3 months postdischarge, a new six-point score named PEDSS (pre-status epilepticus PCPCS, background electroencephalographic abnormalities, drug refractoriness, semiology, and critical sickness) was developed. The area under ROC curves for PEDSS score in predicting mortality and differentiating good from poor outcome at the hospital and at 3 months postdischarge were 0.93 (95% CI = 0.87-0.99), 0.8 (95% CI = 0.7-0.9), and 0.89 (95% CI = 0.8-0.96), respectively. The best cutoff PEDSS scores for predicting mortality and poor outcome at the hospital and at 3 months postdischarge were ≥4, ≥3, and ≥3, respectively. SIGNIFICANCE: The PEDSS score has high predictive accuracy for mortality and differentiating good from poor outcome at the hospital and 3 months postdischarge in pediatric CSE. Future studies should be planned to validate it in various geographical and health care settings and in adults.