Diagnostic challenges in pediatric-onset multiple sclerosis: a case report

Introduction

Background

Pediatric-onset multiple sclerosis (POMS) is a rare disorder with an estimated global annual incidence of 0.87 per 100,000 individuals (1); however, this incidence can vary from 0.05 to 2.85 per 100,000 children (2-4). Most POMS cases are diagnosed during adolescence (2-8), but multiple sclerosis (MS) can also rarely present in children under 10 years of age (2,9).

Rationale and knowledge gap

Timely diagnosis of POMS remains crucial for initiating appropriate disease-modifying treatment (DMT) and reducing long-term disability (10). However, diagnosing acquired demyelinating syndromes (ADSs) in pediatric patients can be difficult, especially after the first demyelinating event. Several criteria have been proposed to differentiate POMS from other ADS, such as acute disseminated encephalomyelitis (ADEM), neuromyelitis optica spectrum disorders (NMOSD) and myelin oligodendrocyte glycoprotein antibody spectrum diseases (MOGAD) (11,12). These include age at onset, presence of specific clinical symptoms (such as encephalopathy), specific antibodies or oligoclonal bands (OCB) and magnetic resonance imaging (MRI) findings (11,13,14). In this clinical case, we describe the challenges encountered in diagnosing POMS in a 7-year-old male patient with numerous atypical features, including gender, age at onset, encephalopathy, uncertain OCB and MRI findings. However, specific biomarkers for the diagnosis of POMS are lacking; therefore, the patient received a final diagnosis only after his third episode of central nervous system (CNS) demyelination.

Objective

To highlight the challenges encountered by physicians when managing the complexities of diagnosing childhood ADS. We present this article in accordance with the CARE reporting checklist (available at https://pm.amegroups.com/article/view/10.21037/pm-25-34/rc).

Case presentation

A 7-year-old boy presented to the University Hospital Emergency Department (ED) with a 5-day history of nausea, vomiting, gait instability, strabismus in the left eye, and head tilting to the left. His medical history included learning difficulties at school, several concussions, and no chronic diseases or medical interventions. The patient had an older brother with mild developmental delay; the family history was otherwise unremarkable. On physical examination, the patient was encephalopathic, exhibited monocular divergent strabismus in the left eye, binocular horizontal and vertical nystagmus, and head tilting to the left. Muscle strength was intact, and his gait was stable; however, due to encephalopathy, specific coordination tests could not be fully assessed. Due to suspicion of ADS, patient was evaluated according to the Expanded Disability Status Scale (EDSS), which measures the current level of disability in patients with MS (15). EDSS ranges from 0 to 10, considering ambulation and neurologic impairment of eight functional systems of CNS. An EDSS score of 0 indicates normal neurological status; meanwhile, a score of 10 indicates death due to MS. The patient received a score of 3.5 and was admitted for further investigations (Figure 1). Cerebrospinal fluid (CSF) analysis revealed normal white blood cell counts, protein levels, and glucose levels. Aquaporin-4 and myelin oligodendrocyte glycoprotein antibodies were negative; however, OCB were positive in the CSF and negative in the serum. MRI of the brain revealed multiple lesions in the cortex, subcortex, periventricular areas of all lobes, corpus callosum, brainstem, medulla, and cerebellar peduncles. Spinal MRI was not performed as the child did not demonstrate extremity weakness, sensory changes, or symptoms of bowel or bladder dysfunction. The patient was diagnosed with ADEM and treated with intravenous methylprednisolone (IVMP) 27 mg/kg/day for 5 days according to guidelines (16-18). He showed significant improvement and was symptom-free within 10 days. A follow-up consultation with MRI after three months was recommended, but was not completed due to noncompliance.

Figure 1 Clinical and investigation findings during the onset of the disease. 1: FLAIR axial images with multiple hyperintense plaques, involving juxta cortical, also periventricular white matter, and corpus callosum. 2: FLAIR sagittal image with perpendicular plaques to the lateral ventricles, making the appearance of Dawson fingers. 3: FLAIR axial image infratentorial plaques in pons, cerebellum, middle cerebellar peduncles. 4: T2 sagittal image, multiple T2 bright plaques. 5: STIR sagittal image, multiple hyperintense plaques. CSF OCB, cerebrospinal fluid oligoclonal bands; EDSS, Expanded Disability Status Scale; FLAIR, fluid-attenuated inversion recovery; IVMP, intravenous methylprednisolone; MRI, magnetic resonance imaging; ND, not done; STIR, short tau inversion recovery.

Five months after the initial episode, the patient returned to the ED with a one-day history of vertigo, diplopia, and complaints of slowed reaction time. Three weeks prior, he had received the combined measles, mumps, and rubella (MMR) vaccine. On physical examination, he exhibited binocular horizontal nystagmus, monocular diverging strabismus in the left eye, mild lower-extremity ataxia, and an EDSS score of 3.5. A repeat MRI (Figure 1) revealed new contrast-enhancing lesions in the subcortical right frontal lobe and periventricular left parietal lobe. An MRI of the spine was also performed, revealing multiple contrast non-enhancing lesions in the cervical spinal cord. Despite these findings, OCB results were negative in both serum and CSF. Given the mild encephalopathy, absence of previously positive OCB, and the patient’s age, he was diagnosed with multiphasic ADEM and treated with IVMP for 5 days, followed by an oral prednisolone taper 1 mg/kg/day (maximal dose 60 mg) over 11 weeks.

Five months later (10 months after the first episode), the patient experienced another relapse with similar clinical features, reaching an EDSS of 3.5. The patient also showed increased impulsivity and emotional lability at home. MRI revealed multiple new contrast-enhancing lesions in the pons, subcortical and periventricular regions of the right frontal lobe, and the thoracic level of the spinal cord (Figure 1). The patient was diagnosed with relapsing-remitting POMS and received acute treatment with IVMP for 5 days followed by oral prednisolone taper over 11 weeks. OCB testing was not repeated since the patient fulfilled the 2017 McDonald criteria for MS (14). Furthermore, the parents expressed reluctance in their son undergoing another lumbar puncture because of concerns of possible complications.

Cognitive testing with the Wechsler Intelligence Scale for Children revealed an overall intelligence quotient (IQ) of 55 (0.1 percentile) at 8 years old, with the lowest scores achieved in processing speed and visual-spatial processing. No previous testing was conducted prior to the first hospitalization, despite suspicions of learning difficulties due to incompliance.

Long-term treatment with subcutaneous glatiramer acetate 20 mg three times per week was initiated; however, due to repeated and severe relapses (with EDSS reaching up to 5.5) over 7 months of therapy with glatiramer acetate, treatment was escalated to intravenous rituximab starting with an induction dose of 375 mg/m². Unfortunately, during the first infusion, the patient experienced an anaphylactic reaction with progressive whole body urticaria, pruritus, mild eyelid edema, and dry cough, which resolved after stopping the infusion. The treatment was then switched to intravenous natalizumab 200 mg (7 mg/kg per dose) once every 4 weeks for 3 months followed by once every 6 weeks. Since starting natalizumab, the patient has been relapse-free, with no new lesions appearing for over two-year period. Despite this, cognitive difficulties result in a baseline EDSS of 3, last IQ overall result at the age of 10 years was 58, previously at the age of 8 years was 55. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Helsinki Declaration and its subsequent amendments. Approval for this case was obtained from the Central Medical Ethics Committee of Latvia Nr. 3/18-03-21. Written informed consent was obtained from the patient’s parents for the publication of this case report and accompanying images. A copy of the written consent is available for review by the editorial office of this journal.

Discussion

Key findings

We present a diagnostically challenging case of a 7-year-old boy with a final diagnosis of POMS, though only after the third demyelinating event. Our patient had multiple atypical features, which hindered diagnosis, including age of onset, gender, suspicions of possible encephalopathy, uncertain MRI and OCB findings. Previous cases of POMS in prepubertal children as young as 2 and 4 years old have been reported (9,19), both of which had negative OCB and were diagnosed as ADEM. Neither patient presented with encephalopathy, one of the main reasons why ADEM was suspected in our patient in the first episodes, as well as OCB and radiological features.

Strengths and limitations

Strengths of our clinical case are highlighting the numerous challenges in navigating the differential diagnosis of ADS in young children. Limitations of our clinical case include inability to generalize our findings as alternative diagnoses, such as ADEM and MOGAD, are known to be more likely in children under the age of ten. In hindsight, spinal MRI in the first episode would have been done. OCB in CSF generally has high sensitivity for MS, depending on the methods used; it can be present in cases of other inflammatory CNS diseases. In addition, it is important to remember that negative OCB does not exclude a POMS diagnosis, especially early in the disease course or in the pediatric population (14).

Comparison with similar research

After the first episode, the patient was diagnosed with ADEM due to polyfocal clinical manifestations and accompanying encephalopathy. ADEM is the most common ADS in children, particularly at this age (4,5,20,21). POMS was also a possible differential diagnosis, supported by positive OCB; however, OCB can be present in 0% to 53% of ADEM cases (2,5,22-24). Furthermore, encephalopathy and polysymptomatic events do not exclude a POMS diagnosis, as encephalopathy may be present in up to 20% of POMS cases (6), and patients may present with polysymptomatic events in up to 67% of cases, especially prepubertal children (2,7,20,22,23), however most common presenting symptoms of POMS include optic neuritis, sensory or motor symptoms and brainstem symptoms (25,26).

Recognizing encephalopathic features in children with cognitive impairment during ADS episodes is difficult. The definition of encephalopathy includes behavioral changes and/or alterations in consciousness, such as irritability (11,27), which can be difficult to assess in children with learning difficulties or psychiatric comorbidities. However, verifying encephalopathy is crucial for adhering to ADEM diagnostic criteria (11). Several studies indicate that children with MS have a wide range of cognitive impairments (7,26), with 30% to 50% presenting at least mild cognitive deficits, which can already be evident in the early stages of the disease (7,9,13,24,28,29). Additionally, it has been reported that lower IQ scores are associated with younger age at MS onset. Our patient had learning difficulties before his first hospitalization, but additional testing, including a cognitive assessment, was never done. It cannot be excluded that cognitive deficit was the presenting symptom of POMS in our patient. However, learning difficulties have also been identified in the patient’s brother, therefore causative or contributing genetic and social factors are more likely in this case.

Regarding imaging, ADEM typically shows large, bilateral, diffuse lesions on MRI, which are poorly marginated and usually lack contrast enhancement; in contrast, MS lesions are well-marginated, with periventricular localization and involvement of the corpus callosum (6,30). The presence of T1 hypointense lesions, T2 periventricular lesions, and brainstem lesions are sensitive and specific predictors of MS in children with acute demyelinating events (9). Our patient initially exhibited multiple diffuse, bilateral, marginated lesions appearing bright on fluid-attenuated inversion recovery (FLAIR) and T2 with no contrast enhancement or restricted DWI localized in the periventricular and juxtacortical white matter including the corpus callosum. Infratentorial hyperintensities and Dawson fingers were also present (Figure 1). No chronic black holes or T1 hypointensities were seen. Overlapping radiologic features in our patient (appearance of diffuse, bilateral, contrast non-enhancing lesions) made the initial differential diagnosis difficult.

During the second episode, the patient experienced another polysymptomatic event with mild encephalopathy. ADEM typically has a monophasic course, but 10–36% of patients experience another demyelinating event (27), including multiphasic ADEM, which occurs in up to 35% of cases (5). The main differential diagnoses were multiphasic ADEM and POMS. Repeat investigations showed the absence of OCB. OCB, if present in ADEM, are usually transient (5), whereas OCB are found in 45% to 90% of MS patients (2,7,20,22), though they are less common in children younger than ten (7,22). Imaging revealed new contrast-enhancing lesions in the subcortical and periventricular areas, and multiple lesions in the cervical spinal cord. Spinal cord involvement can also be seen in ADEM and has been reported in 18% to 80% of ADEM cases (21). Another factor was the recent MMR vaccine. Vaccinations have been reported to trigger ADEM in 4% to 18% of cases (21), though there is no evidence of increased MS relapse risk after vaccinations (31). Based on the clinical findings, including possible mild encephalopathy during a polysymptomatic event, the possible vaccine trigger, and the absence of previously present OCB, the patient was diagnosed with multiphasic ADEM.

Explanations of findings

Subsequently, the patient experienced a non-encephalopathic event with multiple new contrast-enhancing and non-enhancing lesions (Figure 1). It was at this point that the patient was diagnosed with MS, and DMT was initiated. Upon retrospective analysis, we might consider the second OCB result as a possible false negative, though a repeat lumbar puncture was not performed during the third episode or later.

Regarding treatment, in case of ADS acute attack, first-line therapy is IVMP 20–30 mg/kg/day (maximum of 1 g) for 3–5 days; however, approaches to subsequent tapering with oral steroids may vary. Some experts support the use of tapering in patients with ADEM and NMOSD started at 1 mg/kg/day (up to a maximum of 60 mg/day) and tapered over 3–6 weeks in case of ADEM and even up to 2–6 months until long-term immunosuppression is established, if indicated (17). In our case, oral steroid tapering was initiated after the second and third attacks, followed by MS diagnosis and DMT. In case of POMS compared to adult-onset MS, there is a significant disparity in the availability of DMT and DMT for children remains mostly off-label (6,32). DMT choice is based on limited data from available randomized controlled clinical trials (RCTs); nonetheless, there are data available derived from adult RCTs and pediatric observational studies (32). Our patient initially received DMT with glatiramer acetate with insufficient treatment response, followed by rituximab with unfortunate anaphylactic reaction and currently is on natalizumab treatment. Dosing for subcutaneous glatiramer acetate is recommended to be 20 mg daily or 40 mg three times per week in children older than 10 years old (32). In younger children, the dosing regimen is unclear and not well studied. In children under 12 years of age or weighing <40 kg, two regimes of rituximab administration have been recommended: 500 mg separated into two administrations over 2 weeks or four consecutive doses of 375 mg/m² separated from each other by 1 week (6). Natalizumab can either be administered using standard adult regimen of 300 mg intravenously or weight based with 3–5 mg/kg per dose given intravenously every 4 weeks (33); we slightly modified the regimen to 7 mg/kg given every 6 weeks. After initiation of natalizumab, our patient did not experience clinical attacks and no new contrast enhancing and non-enhancing lesions in the brain or spinal cord have been identified. Currently, although our patient’s physical status is stable, cognitive impairment is the primary reason for his current baseline EDSS of 3.

Implications and actions needed

This clinical case illustrates the disease course and diagnostic challenges of early-onset POMS. Diagnostic delay is still a major problem for POMS patients, particularly for those in pre-pubertal age (19,34) due to clinical, radiological and laboratory feature overlap with other ADS as well as the rarity of MS in this age group (1-3). To improve diagnostic accuracy there is a need for increased awareness about MS in children. Although over the years the diagnostic criteria have changed several times and include clinical, radiographic, and laboratory criteria, it still might not be enough to avoid misdiagnosis and diagnostic delay, if atypical features are present. There is still need for further diagnostic accuracy improvements, particularly for patients in pre-pubertal age.

Patient perspective

Since starting natalizumab, the patient has been relapse-free for over two years and with no new lesions appearing on MRI. Nevertheless, increasing cognitive difficulties have left his baseline EDSS at 3. Despite the ongoing condition, the patient expresses satisfaction with the current stability and lack of serious relapses. He finds the present treatment more agreeable compared to the subcutaneous injections (glatiramer acetate) he previously received, which he described as sometimes frightening and painful. His aspiration is to eventually reach a point where he no longer feels the impact of the illness and its associated treatment in his daily life. However, the patient expressed that natalizumab infusions are burdensome to him and his family because they require regular in-hospital visits and as he lives outside of the capital city, travel times are dependent on the national intercity bus schedules.

Conclusions

Early recognition of POMS is crucial, yet remains a significant problem in younger patients. Delayed POMS diagnosis remains a significant issue, and efforts to reduce the time to diagnosis are essential for early DMT initiation in order to improve long-term prognosis and reduce disability.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the CARE reporting checklist. Available at https://pm.amegroups.com/article/view/10.21037/pm-25-34/rc

Peer Review File: Available at https://pm.amegroups.com/article/view/10.21037/pm-25-34/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://pm.amegroups.com/article/view/10.21037/pm-25-34/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Helsinki Declaration and its subsequent amendments. Approval for this case was obtained from the Central Medical Ethics Committee of Latvia Nr. 3/18-03-21. Written informed consent was obtained from the patient’s parents for the publication of this case report and accompanying images. A copy of the written consent is available for review by the editorial office of this journal.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Yan K, Balijepalli C, Desai K, et al. Epidemiology of pediatric multiple sclerosis: A systematic literature review and meta-analysis. Mult Scler Relat Disord 2020;44:102260. [Crossref] [PubMed]
2. Pohl D, Hennemuth I, von Kries R, et al. Paediatric multiple sclerosis and acute disseminated encephalomyelitis in Germany: results of a nationwide survey. Eur J Pediatr 2007;166:405-12. [Crossref] [PubMed]
3. Ketelslegers IA, Catsman-Berrevoets CE, Neuteboom RF, et al. Incidence of acquired demyelinating syndromes of the CNS in Dutch children: a nationwide study. J Neurol 2012;259:1929-35. [Crossref] [PubMed]
4. Jeong A, Oleske DM, Holman J. Epidemiology of Pediatric-Onset Multiple Sclerosis: A Systematic Review of the Literature. J Child Neurol 2019;34:705-12. [Crossref] [PubMed]
5. Massa S, Fracchiolla A, Neglia C, et al. Update on Acute Disseminated Encephalomyelitis in Children and Adolescents. Children (Basel) 2021;8:280. [Crossref] [PubMed]
6. Jakimovski D, Awan S, Eckert SP, et al. Multiple Sclerosis in Children: Differential Diagnosis, Prognosis, and Disease-Modifying Treatment. CNS Drugs 2022;36:45-59. [Crossref] [PubMed]
7. Deiva K. Pediatric onset multiple sclerosis. Rev Neurol (Paris) 2020;176:30-6. [Crossref] [PubMed]
8. Jancic J, Nikolic B, Ivancevic N, et al. Multiple Sclerosis in Pediatrics: Current Concepts and Treatment Options. Neurol Ther 2016;5:131-43. [Crossref] [PubMed]
9. Sivaraman I, Moodley M. Multiple sclerosis in the very young: a case report and review of the literature. Neurodegener Dis Manag 2016;6:31-6. [Crossref] [PubMed]
10. Kopp TI, Blinkenberg M, Petersen T, et al. Long term effect of delayed treatment on disability in patients with paediatric onset multiple sclerosis: A prospective Danish cohort study. Mult Scler Relat Disord 2020;40:101956. [Crossref] [PubMed]
11. Krupp LB, Tardieu M, Amato MP, et al. International Pediatric Multiple Sclerosis Study Group criteria for pediatric multiple sclerosis and immune-mediated central nervous system demyelinating disorders: revisions to the 2007 definitions. Mult Scler 2013;19:1261-7. [Crossref] [PubMed]
12. Chou IJ, Wang HS, Whitehouse WP, et al. Paediatric Multiple Sclerosis: Update on Diagnostic Criteria, Imaging, Histopathology and Treatment Choices. Curr Neurol Neurosci Rep 2016;16:68. [Crossref] [PubMed]
13. Teleanu RI, Niculescu AG, Vladacenco OA, et al. The State of the Art of Pediatric Multiple Sclerosis. Int J Mol Sci 2023;24:8251. [Crossref] [PubMed]
14. Thompson AJ, Banwell BL, Barkhof F, et al. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol 2018;17:162-73. [Crossref] [PubMed]
15. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33:1444-52. [Crossref] [PubMed]
16. Wang CX. Assessment and Management of Acute Disseminated Encephalomyelitis (ADEM) in the Pediatric Patient. Paediatr Drugs 2021;23:213-21. [Crossref] [PubMed]
17. Hofer S, Bauder F, Mori AC, et al. Management of Acute Demyelinating Attacks in the Pediatric Population: A Swiss Consensus Statement. Clin Transl Neurosci 2021;5:17.
18. Waldman AT, Gorman MP, Rensel MR, et al. Management of pediatric central nervous system demyelinating disorders: consensus of United States neurologists. J Child Neurol 2011;26:675-82. [Crossref] [PubMed]
19. Arkar U, Vipotnik Vesnaver T, Osredkar D, et al. Multiple sclerosis in a 4-year-old boy: a case report and literature review. Front Neurol 2024;15:1359938. [Crossref] [PubMed]
20. Bizjak N, Osredkar D, Perković Benedik M, et al. Epidemiological and clinical characteristics of multiple sclerosis in paediatric population in Slovenia: A descriptive nation-wide study. Mult Scler Relat Disord 2017;18:56-9. [Crossref] [PubMed]
21. Lim H, Hwang SK, Lee YJ, et al. The Wide Variety of Acute Disseminated Encephalomyelitis in Children: A Clinical Perspective. Ann Child Neurol 2022. doi: 10.26815/acn.2022.00220.
22. Mikaeloff Y, Caridade G, Assi S, et al. Prognostic factors for early severity in a childhood multiple sclerosis cohort. Pediatrics 2006;118:1133-9. [Crossref] [PubMed]
23. Dale RC, de Sousa C, Chong WK, et al. Acute disseminated encephalomyelitis, multiphasic disseminated encephalomyelitis and multiple sclerosis in children. Brain 2000;123:2407-22. [Crossref] [PubMed]
24. Boesen MS, Blinkenberg M, Koch-Henriksen N, et al. Implications of the International Paediatric Multiple Sclerosis Study Group consensus criteria for paediatric acute disseminated encephalomyelitis: a nationwide validation study. Dev Med Child Neurol 2018;60:1123-31. [Crossref] [PubMed]
25. Aljomah LS, Yeh EA. Pediatric multiple sclerosis: Improving outcome through high-efficacy therapies. Neurotherapeutics 2025;22:e00631. [Crossref] [PubMed]
26. Langille MM, Islam T, Burnett M, et al. Clinical Characteristics of Pediatric-Onset and Adult-Onset Multiple Sclerosis in Hispanic Americans. J Child Neurol 2016;31:1068-73. [Crossref] [PubMed]
27. Paolilo RB, Deiva K, Neuteboom R, et al. Acute Disseminated Encephalomyelitis: Current Perspectives. Children (Basel) 2020;7:210. [Crossref] [PubMed]
28. Johnen A, Elpers C, Riepl E, et al. Early effective treatment may protect from cognitive decline in paediatric multiple sclerosis. Eur J Paediatr Neurol 2019;23:783-91. [Crossref] [PubMed]
29. Simone M, Viterbo RG, Margari L, et al. A Randomized Computer-Assisted Rehabilitation Trial of Attention in Pediatric Multiple Sclerosis: A Post Hoc Analysis. Brain Sci 2021;11:637. [Crossref] [PubMed]
30. Hahn JS, Tenembaum S. 17 - Clinical and biological features of acute disseminated encephalomyelitis. In: Chabas D, Waubant EL, editors. Demyelinating Disorders of the Central Nervous System in Childhood. Cambridge University Press; 2011:183-201.
31. Mailand MT, Frederiksen JL. Vaccines and multiple sclerosis: a systematic review. J Neurol 2017;264:1035-50. [Crossref] [PubMed]
32. Mavridi A, Bompou ME, Redmond A, et al. Current and Emerging Treatment Options in Pediatric Onset Multiple Sclerosis. Sclerosis 2024;2:88-107.
33. Chitnis T. Disease-modifying therapy of pediatric multiple sclerosis. Neurotherapeutics 2013;10:89-96. [Crossref] [PubMed]
34. Brola W, Steinborn B. Pediatric multiple sclerosis - current status of epidemiology, diagnosis and treatment. Neurol Neurochir Pol 2020;54:508-17. [Crossref] [PubMed]