Sage Journals: Discover world-class research

Abstract

Background:

Neuromyelitis optica spectrum disorder (NMOSD) may be triggered by environmental risk factors.

Objectives:

We aimed to explore and integrate the recent research advances in this field. Here we describe relevant studies and summarize current knowledge on non-genetic factors that influence the onset of the disease.

Design:

Systematic review.

Methods:

We performed a systematic review up to May 21, 2024, following preferred reporting items for systematic reviews and meta-analyses guidelines. Two independent reviewers evaluated the quality of the included studies using the Joanna Briggs Institute checklist for risk of bias assessment.

Data sources:

MEDLINE, EMBASE, Scopus, and Web of Science databases.

Results:

A total of 15,869 articles were evaluated. Of those 50 studies met the eligibility criteria. A total of 21,410 NMOSD patients were included in the studies; 17,080 patients were females. Totally, 14 risk factors, including vitamin D deficiency, vaccination, virus infections, lifestyle, and dietary factors, were assessed. A total of 37% of the included articles were conducted in East Asia, mainly focusing on the effects of infection and vitamin D deficiency. These studies suggested vitamin D deficiency as a possible NMOSD risk factor. A total of 25% of the studies included Caucasian populations from Western countries. They showed that smoking decreased the odds of NMOSD, in contrast to observations from Eastern studies. Few cases reported NMOSD onset after COVID-19 vaccination. Antibodies against Epstein-Barr virus, Mycobacterium paratuberculosis, and Helicobacter pylori were observed to be more frequently positive in the serum of NMOSD patients. Lower protein and fat and higher carbohydrate intakes were correlated with NMOSD development.

Conclusion:

Vitamin D deficiency, cigarette smoking, Mycobacterium avium subspecies paratuberculosis infection, and diet were reported as environmental risk factors for NMOSD. The difference in the onset of NMOSD between Asian and Caucasian populations could be affected by smoking and vitamin D deficiency. Knowledge of modifiable risk factors for NMOSD may be beneficial in preventing and improving disease outcomes.

Keywords

environment epidemiology neuromyelitis optica spectrum disorder (NMOSD)risk factors

Introduction

Neuromyelitis optica spectrum disorder (NMOSD) is an inflammatory autoimmune disease of the central nervous system (CNS).¹ Serum immunoglobulin G autoantibodies (NMO-IgG) are found in most NMOSD patients.² The target antigen of NMO-IgG is the astrocyte water channel aquaporin-4 (AQP4) in the CNS.^3,4 AQP4-IgG is a distinct biomarker for NMOSD and mediates pathogenesis by binding to the autoantigen AQP4 on CNS astrocytes, leading to complement-dependent astrocyte injury, inflammation, subsequent demyelination, and neuronal damage.^5,6 NMOSD is clinically characterized by predominant involvement of the optic nerves and the spinal cord.

The prevalence of NMOSD is variable geographically and ethnically.⁷ NMOSD is prevalent in populations that are not Caucasian, especially in Asians,⁸ and epidemiological studies indicate that the prevalence of NMOSD in East Asian countries (Japanese, Chinese, and Koreans) is ~3.5 per 100,000.^9,10 In blacks, the prevalence is higher compared to whites, as some studies report it as ~1.8 per 100,000.⁸ The prevalence among Caucasians is reported to be ~1 per 100,000. Additionally, the age of onset, disease course, and outcome differ between races.¹¹ Non-Caucasians patients experience a younger age of onset, severe motor and visual disabilities and more frequent attacks on the brainstem.¹² Although the exact etiology of NMOSD remains under investigation, emerging evidence suggests that a combination of genetic and environmental factors contributes to various prevalences in different regions. Genetic studies confirm that HLA-DRB1*08:02 and HLA-DRB1*16:02 are risk alleles for NMOSD, while HLA-DRB1*09:01 is associated with a lower risk of NMOSD onset.¹³ The coexistence of NMOSD with other autoimmune diseases is well recognized, and the genetic autoimmune dependency of NMOSD is partly illuminated.¹⁴ Recently, a report that MHC class I-related biological processes and the interferon-gamma-mediated signaling pathway may be involved in the pathogenesis of NMOSD coexisting with systemic lupus erythematosus, Sjögren’s syndrome, and autoimmune thyroid disease.¹⁵

As to the effect of environmental factors, previous literature indicates that infections, smoking, and dietary habits increase the risk of CNS inflammatory disorders, including Multiple Sclerosis.^16–19 However, recent studies have yielded conflicting results regarding the effects of these factors on NMOSD.²⁰ Epstein-Barr virus (EBV) infection is considered as one of the main MS risk factors, but several studies observed no significant increase in humoral response against EBV in NMOSD patients compared to healthy individuals.²¹ It has long been considered that latitude significantly influences the prevalence of MS, whereas no study has confirmed this effect on the onset of NMOSD. Some studies have reported that vitamin D deficiency, habitual risk factors such as smoking, and infections with atypical agents, such as Mycobacterium avium, are possible risk factors for NMOSD.^22–24 Also, diet may influence NMOSD occurrence; a high-carbohydrate and low-dairy diet has been reported to increase the risk of NMOSD.^11,25

Identifying environmental and genetic risk factors is crucial for a deeper understanding of this disease, which in turn aids in establishing prevention strategies, assessing disease risk, and enhancing clinical efficacy. To the best of the authors’ knowledge, no comprehensive systematic review has been published on the environmental risk factors of NMOSD development. This systematic scoping review aims to explore the non-genetic risk factors of NMOSD, focusing on the causes of regional disparity.

Materials and methods

Search strategy

A systematic literature search of MEDLINE, EMBASE, Scopus, and Web of Science was conducted to evaluate non-genetic and environmental risk factors of NMOSD using the keywords neuromyelitis optica (NMO), environmental, risk factor, and each risk factor name, covering from 1999 to May 21, 2024. Furthermore, a manual search of the references and related reviews was performed. Database-appropriate syntax was used to optimize searching. The exact queries used in each database are available in the Supplemental Material.

Data were reviewed from human population studies, including cross-sectional, case-control, and cohort studies written in English. Two of the authors (M.M.L. and E.B.) separately searched, screened the records, and reviewed the titles and abstracts of identified records. A manual hand search of the reference lists of primary identified articles and relevant reviews to discover additional studies and congress abstracts. The two authors (M.M.L. and E.B.) read the full text of potentially eligible records and checked for the following inclusion criteria: (1) human population-based studies and (2) data on the environmental risk factors of NMOSD. Review articles, letters, editorials, and genetic studies were excluded (Table 1). Two experts (M.M.L. and E.B.) independently reviewed the articles. Disagreements were resolved by consensus. The preferred reporting items for systematic reviews and meta-analyses flow chart demonstrated the study selection (Figure 1).^26–28 Initially, 15,871 records from PubMed (1387), Scopus (2795), Embase (9311), and Web of Science (2378) were included. Before the screening, 9522 duplicate records were removed. Further, 6041 records were excluded due to irrelevant titles and abstracts, and from the remaining 307 articles, 261 were not related, 34 were conference abstracts, and 12 were review articles. Fifty articles were included in the current study.

Table 1.

Eligibility considerations in study selection.

Criteria	Inclusion	Exclusion
Population	Human studies of all ages and both genders	Animal studies
Interventions	Covering data on the environmental risk factors of NMOSD
Outcome	Study reporting the primary outcome
Disease definition	NMOSD diagnosis using the Wingerchuk 1999, 2006, or 2015 criteria	Non-confirmed diagnosis, or other demyelinating diseases
Study design	Population-based studies, observational studies including cohort studies, case-control, cross-sectional, and case series	Review articles, case reports, letters, editorials, and genetic studies
Sample size	No restriction
Type of publication	Peer-reviewed journal articles	Books
Publication date	From 1999 to May 21, 2024
Language	Studies published in English	Studies not published in English

NMOSD, neuromyelitis optica spectrum disorder.

Figure 1.

PRISMA chart of the study.

Data extraction and quality assessment

Data extraction and quality assessment were performed by two independent researchers (M.M.L., E.B.). Quality assessment was evaluated based on the Newcastle-Ottawa tool, which was designed by the Cochrane Handbook for the assessment of observational studies.²⁹ Repeated publications of the same data were excluded. The following information was collected: author and year, study type, risk factor(s), demographic data, sample size, and outcome. Measures of effect (e.g., odds ratio (OR), relative risk, confidence interval (CI), and hazard ratio (HR)) from each eligible study were extracted and then summarized in Tables 2 –8.

Table 2.

The effect of patients’ demographics on NMOSD development.

Author, year	Country	Design	Sample size cases	Sample size controls	Age cases^a	Age controls^a	Gender cases (F/M)	Gender controls (F/M)	NMOSD duration^b
Akaishi, 2020	Japan	Cross-sectional	125	–	45.2 ± 16.1	–	113/12	–	N/A
Aubart, 2022	France	Cross-sectional	19	–	8.68	–	12/7	–	NA
Barzegar, 2023	Iran	Case-control	35	37	36.63 ± 9.34	37.57 ± 9.75	27/35	28/37	6 (4.0–8.0)
Bo, 2017	Italy	Case-control	34	38	51.32	54.37	29/5	28/10	N/A
Bo, 2018	Italy	Case-control	34	38	51.32	54.37	29/5	28/10	N/A
Cossu, 2023	Japan	Case-control	20	40	45 ± 16	44 ± 15	15/5	30/10	8 ± 7
Eskandarieh, 2018	Iran	Case-control	100	400	35.83 ± 9.72	33.67 ± 8.37	84/16	340/60	N/A
Etemadifar, 2019	Iran	Case-control	30	30	33.6		25/5	23/7	N/A
Fang, 2023	China	Cross-sectional	81	–	47.75–60	–	–	–	–
Francis, 2022	UK	Cohort	2	–	39		2/0	–	–
Glehn, 2012	Brazil	Cross-sectional	34	–	38 (14–76)	–	26/8	–	N/A
Graves, 2014	USA	Case-control	31	94	N/A	N/A	24/12	112/112	N/A
Kim, 2022	South Korea	Cohort	10	–	36.5 (22–67)	–	6/4	–	–
Koga, 2010	Japan	Case-control	19	33	53 (16–71)	59 (17–77)	17/2	13/20	N/A
Lee, 2023	Canada	Cross-sectional	38	–	39 (20–82)	–	23/15	–	–
Li, 2014	China	Cross-sectional	88	352	34.1 ± 15.9	–	71/17	308/68	N/A
Long, 2013	China	Case-control	52 NMO 24 high risk NMO	27	N/A	N/A	46/6 NMO 16/8 HR NMO	–	NMO: 56.98 ± 62.28 Months/HR NMO: 20.92 ± 37.50
Malli, 2023	India	Case-control	99	243	41.6 ± 12.6	35.2 ± 11.23	90/9	149/94	10.9 ± 7.2
Masuda, 2014	Japan	Case-control	50	52	51 (29–73)	32	48/2	41/11	8 (0–39.9)
Mello, 2020	France	Cross-sectional	36	–	–	–		–	–
Mirmosayyeb, 2021	Iran	Case-control	220	2174	36.65 ± 10.62	35.92 ± 24.44	75.9/24.1	78.1/29.1	N/A
Mori, 2011	Japan	Case-control	8	33	48.5 (26–68)	65.0 (52–87)	8/1	11/9	N/A
Netravathi, 2022	India	Cross-sectional	2	–	49.5		2/0	–	–
Papais-Alvarenga, 2002	Brazil	Cross-sectional	24	N/A	–	20/4	–	7.7 (8.2) years	N/A
Paz, 2021	Brazil	Case-control	26 case	26	<30 years: 6(23.1)/⩾30 years: 20(76.9)	<30 years: 6(23.1)/⩾30 years: 20(76.9)	25/1	25/1	N/A
Rezaeimanesh, 2019	Iran	Case-control	70	164	34.0 ± 12.5	41.0 ± 24.0	60/10	102/62	N/A
Rezaeimanesh, 2019	Iran	Case-control	70	164	35.34 ± 9.87	42.94 ± 15.31	85%	62%	37.56 ± 20.74
Rezaeimanesh, 2021	Iran	Case-control	70	164	35.34 ± 9.87	42.94 ± 15.31	85.70%	62.20%	3.12 ± 1.72
Rezaeimanesh, 2020	Iran	Case-control	70	164	35.34 ± 9.87	42.94 ± 15.31	85%	62%	37.56 ± 20.74
Rotstein, 2023	Canada	Case-control	122	956	52.7 ± 14.6	48.5 ± 11.4	107/15	70/10	NA
Samim, 2022	India	Cross-sectional	5	–	33	–	5/0	–	–
Sellner, 2010	Germany	Case-control	52	56	44.9 ± 15.8	41.3 ± 14.0	43/9	30/26	N/A
Simon, 2015	USA	Case-control	98	488	46 ± 13	48 ± 17	80/18	293/195	3 ± 4
Siuko	Finland	Cohort	184	–	34	–	140/44	–	N/A
Slavin, 2018	Italy	Cross-sectional	34	–	51.32	–	29/5	–	N/A
Wang, 2018	China	Cross-sectional	10	–	58.1 (31–80)	–	5/5		9 (2–37)
Yang, 2021	China	Case-control	36	122	52 (10–86)	30 (20.59)	21/15	84/38	6.5 (0.1–336) months
Yoshimura, 2015	Japan	Case-control	116	367	N/A	N/A	102/14	N/A	AQP4−: 12.01 ± 7.92
Yoshimura, 2015	Japan	Case-control	116	367	–		102/14	N/A	AQP4+: 12.62 ± 10.54, AQP4−: 12.01 ± 7.92
Zatjirua, 2011	South Africa	Case-control	14	N/A	N/A	3/10	3/10	N/A	N/A

Presented as mean ± SD or median (range).

Presented as years.

AQP4, aquaporin-4; HR, hazard ratio; N/A, not assessed; NMO, neuromyelitis optica; OR, odds ratio.

Table 3.

The effect of patients’ lifestyle on NMOSD development.

Author, year	Outcome			p-Value	OR (95% CI)
Author, year	Status	NMOSD	Controls	p-Value	OR (95% CI)
Rotstein, 2023	Ever smoker	33%	46.10%		0.4 (0.3–0.7)
Simon, 2015	Ever smoker	25%	40%	0.007	0.5 (0.3–0.8)
Graves, 2014	Tobacco exposure in early childhood	13%	25.80%	0.25	0.50 (0.16, 1.5)
Eskandarieh, 2018	Current smoker	16%	9.50%	0.8	1.92 (0.93–3.97)
	Smoking duration >10 years	5%	20%	0.1	2.10 (0.85–5.15)
	Passive smoker	54%	40%	0.01	1.79 (1.12–2.85)
	Passive smoking duration >20 years	28%	17%	0.002	2.46 (1.37–4.40)
	Water-pipe smoking, ever	27%	15%	0.003	2.39 (1.33–4.29)
	Opioid, at least 6 months	6%	9%	0.1	2.83 (0.83–9.65)
	Cannabis, at least 6 months	1%	5%	0.7	0.68 (0.06–6.74)
	Stimulants, at least 6 months	1%	4%	0.7	0.67 (0.06–6.97)
	Hallucinogens, at least 6 months	1%	2%	0.7	1.51 (0.12–19.04)
	Whisky, vodka, at least 6 months	19%	10%	0.02	2.11 (1.08–4.11)
	Beer, at least 6 months	14%	11%	0.4	1.37 (0.67–2.77)
	Wine, at least 6 months	20%	40%	0.01	2.32 (1.21–4.43)
Yang, 2021	Active and passive smoking	NA	NA	0.384	1.392 (0.666–20.914)
Rezaeimanesh, 2019	Inverse association between dietary TAC and odds of NMOSD			<0.01
Rezaeimanesh, 2019	Increased odds of NMOSD per every 10 g increase in total sugar intake			<0.001	1.72 (1.43–2.03)
Rezaeimanesh, 2020	Decreased odds of NMOSD in fifth and sixth deciles of low carbohydrate diet				0.22 (0.05–0.87) and 0.24 ( 0.06–0.93) in the fifth and sixth deciles
Rezaeimanesh, 2021	Higher whole grains and legumes consumption in the control group vs the case group			<0.05	<1
Eskandarieh, 2018	Dairy consumption: Low (50–525 g/week)	58%	27%	<0.001	18.09 (6.91–47.37)
	Dairy consumption: Moderate (526–1082 g/week)	37%	32%	0.001	9.89 (3.71–26.36)
	Sea food: Low (0–37.5 g)	67%	23%	<0.001	13.91 (6.13–31.57)
	Sea food: Moderate (37.5–225 g)	25%	34%	<0.001	4.05 (1.71–9.58)
	Egg: Low (0–2/week)	66%	31%	<0.001	9.39 (4.03–21.86)
	Egg: Moderate (3–4/week):	27%	39%	0.01	2.96 (1.21–7.22)
	Red meat: Low (0–180 g/week):	56%	41%	<0.001	3.07 (1.68–5.62)
	Red meat: Moderate (181–360 g/week):	30%	32%	0.015	2.33 (1.17–4.63)
	Chicken: Low (0–225 g/week)	41%	34%	0.003	4.05 (1.59–10.35)
	Chicken: Moderate (226–450 g/week)	52%	44.5%	0.003	4.11 (1.64–10.28)
	Fruits/vegetables: Low (0–740 g/week)	53%	29%	<0.001	6.23 (3.07–12.62)
	Fruits/vegetables: Moderate (741–1320 g/week)	35%	34%	<0.001	3.23 (1.56–6.66)
	Fat: Low (0–57.5 g/week)	47%	30%	<0.001	3.01 (1.64–5.49)
	Fat: Moderate (58–158 g)	32%	33.2%	<0.001	1.93 (1.02–3.68)
	Multivitamin (monthly)	4%	18%	0.002	5.49 (1.90–15.85)
	Calcium (monthly)	5%	12%	0.1	2.55 (0.91–7.10)
	Iron (monthly)	6%	27%	<0.001	7.64 (2.98–16.61)
	Folic acid (monthly)	12%	17%	0.2	1.55 (0.76–3.15)
	B12 (monthly)	5%	13%	0.03	2.78 (1.05–7.36)
	Vitamin C (monthly)	7%	20%	0.002	3.80 (1.63–8.87)
	Vitamin D (monthly)	4%	6.5%	0.2	2.18 (0.67–7.07)
Paz, 2021	Mean dietary inflammatory index	5.15 ± 0.80	2.64 ± 1.77	2.51 (1.73–3.27), the risk of individuals developing the disease increased 30.25 on average (95% CI: 6.70–136.47) with a DII ⩾4.41
	Vitamin D supplementation	Yes: 5 (19.2) no: 21 (80.8)	Yes: 5 (19.2) no: 21(80.8)	1.000
Huang, 2023	MNA	20.4 ± 3.3	23.3 ± 2.5	0.002
Eskandarieh, 2018	Light physical activity weekly in minutes (13–19 years old age)	Low		<0.001	6.32 (3.08–12.95)
		Moderate		<0.001	3.95 (1.89–8.28)
	Heavy physical activity weekly in minutes (13–19 years old age)	Low		<0.001	16.11 (7.03–36.91)
		Moderate		<0.001	2.60 (1.06–6.35)
Paz, 2021	Physical activity	Yes: 13(50.0%) No: 13(50.0%)	Yes: 11 (42.3%) No: 15 (57.7%)	0.781
Paz, 2021	Obesity	Obesity: 7 (29.2%) Non-obesity: 23 (88.5%)	Obesity: 3 (11.5%) Non-obesity: 23 (88.5%)	0.164
	Excess weight:	Excess weight: 13 (50.0%) No excess weight: 13 (50.0%)	Excess weight: 9 (34.6%) No excess weight: 17 (65.4%)	0.262
Rafieea, 2020	Stress number: Zero	28 (18.3%)	72 (18.0%)	1.00 Ref
	One event	33 (21.6%)	103 (25.8%)	0.46	0.80 (0.44–1.45)
	Two events	39 (25.5%)	119 (29.8%)	0.65	1.24 (0.71–2.17)
	Three or more events	53 (34.6%)	106 (26.5%)	0.43	1.24 (0.71–2.17)
	Stress load: 0	28 (18.3%)	71 (17.8%)	0.46	1.00 Ref
	0–75	55 (35.9%)	175 (43.9%)	0.88	0.81 (0.47–1.40)
	76–150	48 (31.4%)	113 (28.3%)	0.37	1.04 (0.59–1.82)
	>150	22 (14.4%)	40 (10.0%)	0.35	1.36 (0.68–2.72)
	Death of spouse	1 (0.7%)	8 (2.00%)	0.28	0.32 (0.04–2.59)
	Death of first-degree relatives	27 (17.6%)	171 (42.9%)	<0.001	0.28 (0.18–0.45)
	Family disruption	36 (23.5%)	31 (7.8%)	<0.001	3.65 (2.16–6.16)
	Jail term	6 (3.9%)	5 (1.3)	0.06	3.21 (0.96–10.69)
	Homeless periods	5 (3.3%)	3 (0.8%)	0.04	4.45 (1.05–18.89)
	Marriage	17 (11.2%)	87 (21.8%)	0.005	0.45 (0.25–0.78)
	Close family members’ serious disease	40 (26.1%)	100 (25.1%)	0.79	1.05 (0.69–1.62)
	Exam for university entrance	29 (19.0)	85 (21.3)	0.53	0.86 (0.53–1.37)
	Being in debt	49 (32.00)	86 (21.6)	0.01	1.71 (1.13–2.59)
	Getting fired from work	6 (3.9)	15 (3.8)	0.92	1.07 (0.66–1.74)
	Joblessness	54 (35.3)	69 (17.3)	<0.001	1.04 (0.39–2.74)
	Divorce	19 (12.4)	14 (3.5)	<0.001	2.60 (1.71–3.97)
	Migration	15 (9.8)	36 (9)	0.77	3.89 (1.90–7.99)
	Close family member suicide	4 (2.6)	8 (2.00)	0.66	1.09 (0.58–2.06)
	Retirement	6 (3.9)	9 (2.3)	0.28	1.31 (0.38–4.42)
	Depression history	55 (35.9)	82 (20.6)	<0.001	1.76 (0.61–5.05)
	Stressful life events	125 (81.7)	330 (82.7)	0.78	2.17 (1.44–3.26)

CI, confidence interval; MNA, mini-nutritional assessment score; N/A, not assessed; NMOSD, neuromyelitis optica spectrum disorder; OR, odds ratio; TAC, total antioxidant capacity.

Table 4.

The effect of infection on NMOSD development.

Author, year	Infection analysis	Outcome	p-Value	OR (95% CI)
EBV
Cossu, 2023	ELISA	No significant increase in antibody response against EBV	>0.05
Masuda, 2014	ELISA	Higher AB titer against EBV in NMO	0.005
Simon, 2015	ELISA	Higher EBV seropositivity in NMO compared to controls	0.04
Bo, 2018	ELISA	No significant humoral response	>0.05
Graves, 2014	ELISA	No significant humoral response	>0.05
Yoshimura, 2015	ELISA	No significant humoral response	>0.05
Koga, 2010	ELISA	No significant humoral response	>0.05
Eskandarieh, 2018	Questionnaire	Positive history in 16% of NMOSD, 1% of controls	<0.001	79.35 (8.83–712.75)
Helicobacter pylori
Long, 2013	Indirect immunofluorescence assay	Higher seropositivity in HR NMO and NMO+ HR NMO	0.0050.028	9.311 (1.033–83.917)6.350 (1.219–33.072)
Malli, 2023	ELISA	Not significant
Yoshimura, 2015	ELISA	Higher anti-HP antibody seropositivity in AQP4+ patients	0.0088
Barzegar, 2023	ELISA	No significant humoral response	>0.05
Herpes simplex virus
Etemadifar, 2019	ELISA	Not significant	>0.05
Sellner, 2010	ELISA	Not significant	>0.05
Graves, 2014	ELISA	No significant humoral response	>0.05
Koga, 2010	ELISA	No significant humoral response	>0.05
Cytomegalovirus
Sellner, 2010	ELISA	Higher seropositivity in AQP4+ compared to MS, but not statistically significant compared to controls	0.003 and 0.05, respectively
Graves, 2014	ELISA	No significant humoral response	>0.05
Koga, 2010	ELISA	No significant humoral response	>0.05
VZV
Koga, 2010	ELISA	No significant humoral response	>0.05
Eskandarieh, 2018	Questionnaire	Positive history in 69% of NMOSD, 70.1% of controls	1.09 (0.66–1.80)
Wang, 2018	GWAS	Three cases with NMOSD among 10 VZV related myelitis patients
Human herpesvirus
Sellner, 2010	ELISA	Not significant	>0.05
Human T-lymphotropic virus
Glehn, 2012	ELISA	No anti-HTLV positive cases	>0.05
Koga, 2010	ELISA	No significant humoral response	>0.05
MUMPS
Mori, 2011	PCR (RNA detection)	2 positive PCR cases in the NMO group, 0 in others	0.038
Koga, 2010	ELISA	No significant humoral response	>0.05
Eskandarieh, 2018	Questionnaire	Positive history in 39% of NMOSD, 28% of controls	1.49 (0.92–2.41)
Enterovirus
Mori, 2011	PCR (RNA detection)	No positive PCR cases in the NMO group and in others	>0.05
MAP
Bo, 2017	ELISA	Elevated levels of antibodies against MAP	<0.0002
Slavin, 2018	ELISA	Seropositivity abundance in NMO	0.0032
Chlamydia pneumoniae
Yoshimura, 2015	ELISA	Higher antibody seropositivity in AQP4+ patients	0.0284
Parvovirus B-19
Koga, 2010	ELISA	No significant humoral response	>0.05
Rubella
Koga, 2010	ELISA	No significant humoral response	>0.05
Eskandarieh, 2018	Questionnaire	Positive history in 18% of NMOSD, 12.8% of controls	0.2	1.46 (0.79–2.68)
Measles
Koga, 2010	ELISA	No significant humoral response	>0.05
Eskandarieh, 2018	Questionnaire	Positive history in 33.0% of NMOSD, 33.8% of controls	0.79 (0.47–1.30), p = 0.4
Campylobacter jejuni
Koga, 2010	ELISA	No significant humoral response	>0.05
Haemophilus influenza
Koga, 2010	ELISA	No significant humoral response	>0.05
Mycoplasma pneumoniae
Koga, 2010	ELISA	No significant humoral response	>0.05
Hepatitis B and C
Koga, 2010	ELISA	No significant humoral response	>0.05
Eskandarieh, 2018	Questionnaire	Positive history in 1% of NMOSD, 2.5% of controls	0.3	0.30 (0.03–2.57)
Treponema pallidum
Koga, 2010	ELISA	No significant humoral response	>0.05
Fang, 2023	CBA	One AQP+ case
Mycobacterium tuberculosis
Li, 2014	N/A	No significant association	>0.05
Zatjirua, 2011	Review of records	Preceding or simultaneous diagnosis of PTB in 79% of NMOSD and 14% of controls	4.6 (1.71–15.49)
Papais-Alvarenga 2002	Prior PTB in 5/24 patients
Covid-19

Aubart, 2022	NA	One case of post-COVID AQP4+
Dengue
Mello, 2020	PCR	One case of NMOSD with amplification for DENV-1 in serum

AB, anti body; AQP4, aquaporin-4; CBA,cell-based assay; CI, confidence interval; EBV, Epstein-Barr virus; GWAS, genom-wide association studies; HP, Helicobacter pylori; HR, hazard ratio; HTLV, human T-lymphotropic virus; MAP, Mycobacterium avium subsp. paratuberculosis; NMO, neuromyelitis optica; NMOSD, neuromyelitis optica spectrum disorder; OR, odds ratio; PCR, polymerase chain reaction; PTB, pulmonary TB; RNA, ribonucleic acid; TAC, total antioxidant capacity; VZV, Varicella zoster virus.

Table 5.

The effect of vaccination on NMOSD development.

Author, year	Vaccine type	Interval from vaccine/infection to NMOSD onset (days)	Vaccine dose
Francis, 2022	ChAdOx1S	13	N/A
Kim, 2022	mRNA-1273 AstraZeneca	mRNA-1273: 11, AstraZeneca: 19	1st dose
Lee, 2023	N/A	N/A	N/A
Samim, 2022	BBV151	17	1st
	BBV152	14	2nd
	ChAdOx1nCoV	36	2nd
	ChAdOx1nCoV	42	1st
	ChAdOx1nCoV	13	2nd
Netravathi, 2022	ChAdOx1nCoV	36	2nd
	ChAdOx1nCoV	42	1st

NMOSD, neuromyelitis optica spectrum disorder; RNA, ribonucleic acid.

Table 6.

The effect of hormonal factors on NMOSD development.

Author, year	Factor	Outcome		p-Value	OR (95% CI)
Author, year	Factor	NMOSD	controls	p-Value	OR (95% CI)
Eskandarieh, 2018	Ever used OCP	72.3%	52.7%	0.008	2.11 (1.21–3.66)
		OCP use duration: <1 year	39% controls	65.2%	0.7
	OCP use duration: 1–3 years	16.7%	15.2%	1.20 (0.51–2.79)
	OCP use duration >4 years	18.3%	19.7%	0.85 (0.36–2.01)
	IUI, IVF	8.6%	7.1%	0.8	1.14 (0.39–3.29)
Graves, 2014	Breastfed	53.3%	73.7%	0.05	0.42 (0.18, 0.99)
	Cesarean delivery	36.7%	22.2%	0.18

CI, confidence interval; IUI, intrauterine insemination; IVF, in vitro fertilization; NMOSD, neuromyelitis optica spectrum disorder; OCP, oral contraceptive pills; OR, odds ratio.

Table 7.

The effect of regional factors on NMOSD development.

Author, year	Outcome	p-Value	OR (95% CI)
Latitude
Tian, 2020	Not significant	0.3
Bukhari, 2017	Not significant	>0.05
Wu, 2022	Not significant	>0.05
Sun exposure
Eskandarieh, 2018	13–19 years old age: Low: 78% NMOSD, 27.9% Controls	<0.001	17.97 (7.11–45.40)
	13–19 years old age: Moderate 16% NMOSD, 41% controls	0.2	1.85 (0.67–5.08)
	>20 years old age: Low 52% NMOSD, 28% Controls	<0.001	10.83 (4.40–25.47)
	>20 years old age: Moderate 37% NMOSD, 36% Controls	<0.001	5.21 (2.27–11.94)

CI, confidence interval; NMOSD, neuromyelitis optica spectrum disorder; OR, odds ratio.

Table 8.

The effect of vitamin D deficiency on NMOSD development.

Author, year	Assay method	Vitamin D cutoff	NMOSD duration ((mean ± SD)or(median (range))	Vitamin D level ((NMOSD/controls)(mean ± SD or deficiency %))	p-Value
Min, 2014	LC-MS/MS	⩽20	5.30 ± 6.10	Deficiency (<50 nmol/L) in 92.1% NMOSD/66.2% controls	<0.001
Shan, 2016	EIA	⩽20	N/A	46.25 ± 22.38/61.01 ± 19.09	<0.01
Gao, 2019	HPLC-MS/MS	N/A	1.50 (0.36–4.21)	10.12 ± 4.19/21.98 ± 5.50	<0.001
Puangricharoen, 2022	CLIA	⩽20	N/A	20.18 ± 5.90/31.16 ± 1.69	<0.001
Kusumadewi, 2018	CLIA	N/A	3 (1–16)	16(6–33.4)/15.7(5.5–34.4)	N/S
Nayanajeehwi, 2023	ELISA	N/A	2.6 ± 0.70	19.16 ± 3.72/16.11 ± 2.52	N/S
Ashtari, 2022	ELISA	<10	N/A	24.88 ± 15.2/21.56 ± 18.7	N/S

N/A, not assessed; NMOSD, neuromyelitis optica spectrum disorder; N/S, Not significant; OR, odds ratio.

Results

As shown in Figure 2, the search resulted in 50 articles after the removal of duplicates and irrelevant studies from the following countries: China: 10, Iran: 10, Japan: 6, Brazil: 3, India: 3, Canada: 2, France: 2, Italy: 2, USA: 2, Australia: 1, Finland: 1, Germany: 1, Indonesia: 1, Korea: 1, South Africa: 1, South Korea: 1, Sri Lanka: 1, Thailand: 1, UK: 1. The remaining records were all written in English, and the method of the studies were case-control or cross-sectional. Except for one study,³⁰ all included studies had moderate to high qualities based on the NOS system. In total, 21,410 NMOSD patients were included in the studies, of which 17,080 were females.

Figure 2.

Distribution of the studies on environmental risk factors of NMOSD.

Eligible articles assessed 14 factors such as vitamin D deficiency, smoking, substance and alcohol abuse, vaccination, infection, diet, hormonal factors, seasonality, month of birth, latitude, sun exposure and physical activity and weight, stressful life events, and socioeconomic status (SES). A total of 37% of the included articles were conducted in East Asia, mainly focusing on the effect of infection and vitamin D deficiency.^22,31–46 Thirty-one percent of the Asian articles included in this review were published from non-east-Asian countries,^9,11,47–59 62% of which were from Iran.^{9,11,47–54,59} Iranian studies mainly focused on different environmental risk factors, such as diet, smoking, hormonal factors, and SES, history of oral contraceptive pills use (OCPs), month of birth, sun exposure and physical activity, and socioeconomic factors were evaluated as risk factors of NMOSD.^{9,11,47–51,59} Twenty-five percent of the studies included in the present review are from countries with mostly Caucasian populations.^{23,24,25,60–68} The main focus of the studies was smoking and infections.^{23–25,61,64–68.}

We summarize the articles in Tables 2 –8.

Demographics

Table 2 summarizes the effect of demographics on NMOSD onset. In the study of Rafiee et al.,⁴⁷ marriage showed a negative impact on the odds of NMOSD development (OR = 0.10; 95% CI: 0.03–0.35). There was no relationship between NMOSD and the SES of patients and parental educational level.⁴⁷ In contrast, mothers’ education was significantly associated with NMOSD risk in the study by Graves et al., more than high school but less than college: OR = 0.46, 95% CI: 0.10–2.05, college and higher OR = 0.16, 95% CI: 0.03–0.79. The authors also evaluated daycare (OR = 0.33, 95% CI: 0.14–0.78) and age at the start of daycare (<1 year old: OR = 0.39, 95% CI: 0.13–1.15 and >1 years old: OR = 0.28, 95% CI: 0.09–0.87).²⁵ Akaishi et al.³² observed a spring-summer predominance in the onset of AQP4+NMOSD. In another study, Siuko et al.⁶³ reported that the clinical onset of NMOSD in April–June was two times more prevalent than October–December (p = 0.03, OR = 2.0 (95% CI: 1.09–3.68)). Mirmosayyeb et al.⁵¹ observed that the birth rate was reduced in September–October in the NMOSD group compared to controls.

Lifestyle

Table 3 shows possible lifestyle-related risk factors for NMOSD.

Smoking, substance abuse, and alcohol

Regarding the impact of smoking, a significant disparity was observed between studies on Asians and Caucasians. In two studies from Canada and the USA, patients with a history of smoking had a significantly lower risk of NMOSD. Rotstein et al.²³ reported a decreased risk of NMOSD in ever smokers in the crude (OR = 0.4, 95% CI: 0.3–0.7) and adjusted (OR: 0.5, 95% CI: 0.4–0.9) analyses. However, no correlation was observed in the analysis limited to Ontario residents (OR = 1.1, 95% CI: 0.5–2.2). In the study of Simon et al.,⁶⁴ patients were less likely to report a history of smoking (OR = 0.5, 95% CI: 0.3–0.8). In contrast, smoking did not affect the risk of NMOSD in Asian studies, and in a study from Iran, passive smoking even increased the susceptibility.¹¹ Yang et al.³³ observed no statistical significance between active and passive smoking and the risk of NMOSD occurrence. Similarly, the study by Graves et al.²⁵ did not support tobacco exposure in early childhood as an environmental risk factor (p = 0.25, OR = 0.50, 95% CI: 0.16–1.5). Eskandarieh et al. found that being a passive smoker, passive smoking for more than 20 years and a positive history of smoking waterpipe showed a significant correlation with NMOSD risk (OR = 1.79, 95% CI: 1.12–2.85, OR = 2.46, 1.37–4.40 and OR = 2.39, 95% CI: 1.33–4.29, respectively). However, current smoking and smoking for more than 10 years did not correlate with the development of NMOSD.¹¹

Substance abuse of opioids, cannabis, stimulants, and hallucinogens did not increase the risk of NMOSD. Alcohol consumption at least once for 6 months led to an OR of 2.32 (95% CI: 1.21–4.43) for wine and 2.11 (95% CI: 1.08–4.11) for whiskey, and beer consumption did not differ between the two groups.¹¹

Diet

Table 3 summarizes the possible association between diet and NMOSD risk. Rezaeimanesh et al.⁴⁸ determined the odds of developing NMOSD with a focus on the impact of dietary total antioxidant capacity (TAC). Compared to the first quartile of TAC, the risk of NMOSD reduced across the quartiles (p-trend <0.01). Adjusted for age and gender, reduced risk of NMOSD was observed in second, third, and fourth quartiles with the following odds: 76% (OR: 0.24; 95% CI: 0.10–0.54) for second, 87% (OR: 0.13; 95% CI: 0.05–0.34) for third, and 97% (OR: 0.03; 95% CI: 0.00–0.12) for the fourth quartile.⁵⁰ In another study, they concluded that lower protein and fat and higher carbohydrate intakes might be associated with an increased risk of developing NMOSD. Adjusting for age, sex, BMI, and energy intake, a higher score of a low carbohydrate diet resulted in a significantly lower OR for NMOSD for the fifth (OR: 0.22; 95% CI: 0.05–0.86) and sixth (OR: 0.24; 95% CI: 0.06–0.93) decile (p-trend = 0.11).⁴⁹ In line with the findings mentioned above, lower sugar intake was associated with decreased odds of NMOSD in another study by Rezaeimanesh et al.⁴⁹ Per every 10 g increase in total sugar intake, the authors reported a 1.72-fold increase in odds of NMOSD (CI: 1.43–2.03; p-for-trend: 0.00). Moreover, various types of dietary sugar showed the following odds: glucose: 0.02 (CI: 0.00–0.08; p-for-trend: 0.00), fructose: 0.02 (CI: 0.00–0.10; p-for-trend: 0.00), galactose: 0.23 (CI: 0.08–0.61; p-for-trend: 0.00), lactose: 0.19 (CI: 0.06–0.58; p-for-trend: 0.00), and sucrose: 0.16 (CI: 0.05–0.51; p-for-trend: 0.00). In another study, Rezaeimanesh et al. reported a 90% (95% CI: 0.02–0.39) and 92% (95% CI: 0.01–0.52) reduced odds of NMOSD in the third and fourth quartiles of whole grains intake. In addition, in the third and fourth quartiles of legume intake, an 81% (95% CI: 0.05–0.71) and 95% (95% CI: 0.01–0.27) reduction in the odds of NMOSD was observed.^48–50

In the study of Eskandarieh et al., the following dietary habits during adolescence showed a significant association with NMOSD risk: low dairy consumption, weekly (OR = 18.09; 95% CI = 6.91–47.37), low seafood intake (OR = 13.91; 95% CI: 6.13–31.57), low red meat consumption (OR = 3.07, 95% CI: 1.68–5.62), low egg consumption (OR = 9.39, 95% CI: 4.03–21.86), low chicken consumption (OR = 4.05, 95% CI: 1.59–10.35), low-fat consumption (OR = 3.01, 95% CI: 1.64–5.49), and low fruit and vegetable consumption (OR = 6.23; 95% CI: 3.07–12.62). Moreover, Eskandarieh et al. reported that monthly consumption of multivitamins (OR = 5.49; 95% CI: 1.90–15.85), iron (OR = 7.64; 95% CI: 2.98–16.61), vitamin B12 (OR = 2.78, 95% CI: 1.05–7.36), and vitamin C (OR = 3.80, 95% CI: 1.63–8.87) decreases the risk of NMOSD. However, calcium, folic acid, and vitamin D consumption did not differ in the chance of NMOSD onset (p > 0.05).¹¹

Paz et al.³⁰ reported a higher risk of developing NMOSD when the dietary inflammatory index was ⩾4.41 (OR = 30.25; 95% CI: 6.70–136.47). In the study of Huang et al.⁶⁹ using the mini-nutritional assessment score, the nutritional score of NMOSD patients was significantly lower than healthy controls.

Other lifestyle factors

Physical activity was assessed in two articles. In the study of Eskandarieh et al., low and moderate amounts of light physical activity at 13–19 years of old age were associated with being assigned to the NMOSD group (OR = 6.32, 95% CI: 3.08–12.95 and OR = 3.95, 95% CI: 1.89–8.28). Patients more frequently reported low and moderate amounts of heavy physical activity at 13–19 years old age, compared to controls (OR = 16.11, 95% CI: 7.03–36.91 and OR = 2.60, 95% CI: 1.06–6.35, respectively).¹¹

In the study of Paz et al.,³⁰ physical activity and weight did not differ between cases and controls. A study by Rafiee et al.⁴⁷ showed the association of NMOSD and first-degree relatives (OR = 5.10; 95% CI (1.78–14.61), homelessness periods (OR = 4.35; 95% CIs (1.18–14.74), joblessness (OR = 4.24; 95% CI (1.91–5.15)), family disruption (OR = 12.68; 95% CI: 3.53–45.46), and divorce (OR = 14.18; 95% CI (1.91–23.15), university entrance exam (OR = 0.31; 95% CI (0.11–0.88)) showed a negative association with NMOSD risk (Table 3).

Infection

The association of a single infectious agent with autoimmune diseases is not conclusively established; patients with autoimmune diseases often experience relapses associated with seasonal infections during winter, suggesting an influence of environmental factors in addition to well-established genetic predispositions such as polymorphism in the HLA system.⁷⁰

Table 4 summarizes the possible association between the history of infection and the risk of NMOSD development.

Epstein–Barr virus

The history of EBV infection was evaluated in eight studies, three of which reported it as a risk factor. According to Masuda et al., serum anti-early antigen (anti-EA IgG) was more frequently positive in NMOSD patients, showing a more active viral replication than the controls. Furthermore, higher antibody titers of anti-EA IgM and anti-EA IgG were detected in NMOSD patients compared to controls (p < 0.001 and p = 0.005, respectively).³⁸ In the study of Simon et al., EBV seropositivity was more common in NMOSD patients than the controls, and they seemed likely to have a history of infectious mononucleosis (OR = 1.4, 95% CI: 0.8–2.7). However, an antibody against Epstein-Barr nuclear antigen (Anti-EBNA1 IgG Ab) was not associated with the risk of NMOSD.⁶⁴ Eskandarieh et al.¹¹ revealed a higher frequency of mononucleosis in NMOSD cases than in healthy controls (OR = 79.35, 95% CI: 8.83–712.75). Other studies did not observe an association.^11,35,40,46

Helicobacter pylori

Four studies compared the rate of H. pylori (HP) seropositivity in NMOSD patients and healthy controls. Seropositivity against HP in NMOSD patients was only observed in East Asian studies. In the study of Long et al.,³⁴ higher seropositivity was observed in high-risk NMOSD patients (OR = 9.311, 95% CI = 1.033–83.917) and NMO + high-risk NMO (OR = 6.350, 95% CI: 1.219–33.072). Yoshimura et al.³⁵ reported a higher anti-HP antibody seropositivity in AQP4+ patients (p = 0.0088). The other two studies did not report any significance.³⁵

Cytomegalovirus

Among three studies that evaluated the seroprevalence rate of Cytomegalovirus IgG in NMOSD patients, a trend was observed in the study of Sellner et al.⁶⁵ in AQP4+ patients (p = 0.05), while other studies did not show any significance.^25,43

Mumps

Mori et al.³⁹ reported that polymerase chain reaction results of mumps virus Ribonucleic acid (RNA) were more frequently positive in cerebrospinal fluid of NMOSD patients compared to controls. The other two studies did not support the correlation (p > 0.05).^11,43

Mycobacterium avium

The correlation between NMOSD onset and M. avium ssp. paratuberculosis (MAP) infection was strongly supported in the study of Bo et al.²⁴ and Slavin et al.⁶⁶ The sera of NMOSD patients showed a higher antibody level against peptides derived from MAP, compared to the healthy control group (p < 0.0002), in the study of Salvin et al.,⁶⁶ the greater humoral response against two virulence factors of MAP, protein tyrosine phosphatase A, and protein kinase G detected in NMOSD patients (p < 0.0001 and p = 0.0032, respectively).

Chlamydia pneumoniae

Yoshimura et al.³⁵ reported a higher Chlamydia pneumoniae antibody seropositivity in AQP4+ patients than controls.

Mycobacterium tuberculosis

Tuberculosis infection was evaluated in three studies, one of which reported it as a risk factor (OR: 4.6 (95% CI: 1.71–15.49)).^42,71,72 Papais-Alvarenga et al.⁷² reported a history of pulmonary TB in 5 out of 24 evaluated NMOSD patients. One study did not report any significance.

Treponema pallidum

Fang et al.⁴¹ reported one AQP4+ case among 81 patients with neurosyphilis. Koga et al.⁴³ did not report any significance.

Covid-19

Aubart et al.⁶⁷ reported a case of AQP4 seropositive among 19 children who had a history of COVID-19 infection and manifested a CNS inflammatory disease.

Dengue

Mello et al.⁶⁸ reported one case of NMOSD with amplification for DENV-1 in serum among 36 patients.

Varicella zoster virus

Wang et al.⁷³ reported 3 cases of NMOSD among 10 varicella zoster virus related myelitis patients.

Other infections

Studies did not evaluate the following microorganisms as possible risk factors of NMOSD development: parvovirus B-19, rubella, measles, Campylobacter jejuni, Haemophilus influenzae, Mycoplasma pneumoniae, hepatitis B and C, and enterovirus.

Human herpesvirus, human T-lymphotropic virus, and Herpes simplex virus.^{11,39,43,54,65,73}

Vaccination

Table 5 summarizes the studies that report NMOSD onset after COVID-19 vaccination.^{41,52,57,58,71} Francis et al.⁶⁰ reported two patients who developed AQP4+ NMOSD 13 days after ChAdOx1S vaccination. Kim et al.⁴⁴ also reported two patients who developed NMOSD after the first dose of the vaccine. The interval from vaccine to clinical onset ranged from 11 days in the mRNA-1273 vaccine to 19 days in AstraZeneca. Lee et al.⁶¹ also reported an AQP4+ patient after COVID-19 vaccination; however, vaccine type and the interval from vaccination to NMOSD onset were not specified in this study. In another study, Samim et al.⁵⁵ observed five cases of post-vaccination NMOSD, which occurred between 13 and 36 days after ChAdOx1nCoV and BBV152 vaccines. Netravathi et al.⁷⁴ reported two female patients who developed NMOSD after the first and second doses of ChAdOx1nCoV.

Hormonal factors

Table 6 demonstrates the association between hormonal factors and the risk of NMOSD. In the study of Eskandarieh et al., the history of Intrauterine Insemination or In Vitro Fertilization did not relate to NMOSD onset. Regarding the use of contraceptives, they observed that NMOSD patients report a more frequent use of OCP (OR = 2.11, 95% CI: 1.21–3.66). However, the duration of OCP use did not alter the risk.¹¹ Graves et al. reported that breastfeeding is associated with lower odds of having NMOSD (OR = 0.42, 95% CI: 0.18–0.99), cesarean delivery was not associated with increased odds of NMO compared with vaginal delivery (p > 0.05).

Regional factors

Table 7 shows the possible impact of regional factors on NMOSD onset.

Latitude

No association was found regarding the effect of latitude on NMOSD onset.^31,46,62

Sun exposure

In the study of Eskandarieh et al., low sun exposure at 13–19 years old was associated with NMOSD onset (OR = 17.97, 95% CI: 7.11–45.40, p < 0.001). Low and moderate sun exposure at >20 years old also increased the risk of onset (OR = 10.83, 95% CI: 4.40–25.47, p < 0.001 and OR = 5.21, 95% CI: 2.27–11.94, p < 0.001, respectively).¹¹

Vitamin D

Table 8 indicates the possible association between vitamin D deficiency and NMOSD development. All seven studies were conducted in Asia.^{22,31,33,53–55} While East Asian studies evaluated vitamin D deficiency as a risk factor,^22,36,37 another study from Thailand also reported significance,⁵⁷ and three other studies did not observe any significance. The significant results are as follows: Min et al.²² reported a significant difference in the frequency of vitamin D deficiency in NMOSD cases versus controls. Puangsricharoen et al.⁵⁷ and Shan et al.³⁷ also observed lower vitamin D levels in NMOSD patients.

Levels of 25(OH)D, 25(OH)D2, and 25(OH)D3 were reported to be lower in the study of Gao et al.³⁶

In two other studies, no significant difference was observed between the vitamin D levels of NMOSD patients and healthy controls.^56,58 Ashtari et al.⁵⁹ observed no difference between vitamin D levels in NMOSD and MS patients.

Discussion

The current systematic review summarizes the environmental factors associated with NMOSD onset. Even though the exact pathogenesis of the disease is yet to be understood, addressing triggers might help reach a prompt and accurate diagnosis. Besides individual factors such as age, gender, education, and ethnicity, studies indicate an environmental involvement in the onset of NMOSD. Environmental factors with seasonal variation influence the development of neurological conditions related to NMOSD. Spring-summer predominance in NMOSD onset was observed in the studies, which might be related to a higher prevalence of viruses like adenoviruses or varicella zoster and lower vitamin D levels during winter; moreover, changes in electrolytes due to dehydration could also affect disease onset.^7,75 Environmental risk factors in this study have mainly focused on vitamin D deficiency, smoking, infection, and diet.

Effect of ethnicity on NMOSD risk

According to the literature, the prevalence of NMO is 0.5–4 per 100,000; however, in certain racial groups it may reach 10 per 100,000.⁷⁶

East Asians

NMOSD prevalence in China, Japan, and Korea has been reported as ~3.5 per 100,000, which is higher compared to other racial groups. A nationwide survey reported a prevalence of 3.42/100,000 in Japan, reaching 4.1/100,000 in other Japanese studies. The prevalence among Chinese is estimated as 3.1/100,000, in Korea, it has been recorded as 3.56 per 100,000.⁸ Genetic studies indicate that Chinese and Japanese share HLA-DPB1*05:01 and HLA-DRB1*16:02, which increases NMOSD risk.⁷⁷ A third (37%) of the included articles was conducted in East Asia, mainly focusing on the effect of infection and vitamin D deficiency. Studies support that low vitamin D is significantly associated with higher odds of MS onset, and vitamin D supplements are proven to lower the relapses and activity measured by MRI in MS cases.⁷⁸ Vitamin D response elements form when the active form of vitamin D binds to intracellular vitamin D receptors, and directly influence the regulation of innate and adaptive immunity. Vitamin D may play its role in the pathogenesis of NMOSD by directly affecting AQP4-IgG as a part of humoral immunity, as it induces B cell apoptosis, which leads to a decrease in the production of antibodies.^79,80 Moreover, vitamin D enhances an anti-inflammatory environment by inducing IL-10 and suppressing IL-6 and TNF production.⁸¹

Vitamin D deficiency is more prevalent in Asia than in other regions.⁸² When vitamin D deficiency is defined as less than 50 nmol/L, the prevalence is estimated as 18%—30% in the USA, 40% in Europe, 34% in Africa, and 54% in Asia. Several reasons are mentioned to be responsible for this issue. Asians tend to wear clothes that cover larger areas of the skin, leading to a reduction in ultraviolet B to penetrate the skin and vitamin D synthesis. Moreover, developing countries in Asia lack access to proper vitamin D supplements, unlike those in America and Europe.⁸² All seven studies that evaluated vitamin D deficiency as a possible risk factor were conducted in Asia. East Asian studies evaluated vitamin D deficiency as a risk factor; another study from Thailand reported significance, and three other studies did not observe any significance.^{22,31,34,53–55} Consequently, countries in East Asia have relatively lower levels of vitamin D. Based on a meta-analysis by Jiang, the mean level of 25(OH)D concentration in Korea and China is 49.3 and 52.6 nmol/L, while in Indonesia and Sri Lanka, the mean level reaches 65.8 and 69.6 nmol/L. Different prevalence of vitamin D deficiency and a small sample size might explain the disparity.⁸²

The other main focus of East Asian studies was infections. Several case reports have reported the onset and worsening of NMOSD after bacterial and viral infection; however, we only included studies with a population of more than three cases.^83,84 Seropositivity against HP in NMOSD patients was only assessed in East Asian studies.^34,35 It is important to note that the highest prevalence of HP among the normal population is found in African countries, followed by Asia, and the lowest in Oceania. Among Asian countries, East Asians such as China and Korea show the highest prevalence, while Indonesia and Sri Lanka have the lowest.⁸⁵

Studies indicate a significant association between MAP and NMOSD onset; however, controversial results have been reported regarding other infections such as EBV and mumps, although they are considered triggers of MS.⁸⁶ Viruses and bacteria can, for example, trigger NMOSD onset through molecular mimicry and increased blood-brain barrier permeability, thus allowing the entry of antibodies into the CNS.

Even though tuberculosis is more common in Asian populations compared to Europe and North America, the strength of the evidence is relatively weak regarding the impact of this infection in provoking disease onset. Out of three studies evaluating TB in NMOSD, two report frequent prior pulmonary tuberculosis in NMOSD patients (79% and 21% positive history). The studies above evaluated small populations and did not report the AQP4-IgG status of the patients, leaving this as an open question. A possible effect of TB might be related to the fact that aquaporin extracellular loops in the Mycobacterium share homologous epitopes with human AQP4, so molecular mimicry may create autoantibodies that lead to astrocyte damage.⁸⁷ Interestingly, a prospective study showed that in patients with steroid-refractory NMOSD, anti-TB treatments (a four-drug regimen consisting of isoniazid, rifampicin, pyrazinamide, and streptomycin and three three-drug regimens consisting of isoniazid, rifampicin, and pyrazinamide) improve neurological deficits.⁸⁸

It is important to note that the abovementioned factors, such as vitamin D deficiency and some infections, are reportedly more frequent in Africans; however, due to a limited number of studies, environmental risk factors of NMOSD have not been properly assessed in the African region.

Caucasians/Whites

The prevalence of NMOSD among Caucasians ranges from 0.55/100,000 in Australia and New Zealand to 1.04/100,000 in Sweden. In Denmark, the prevalence is reported to be 1.09/100,000; however, recent studies indicate that by including the total clinical phenotype of NMOSD (AQP4-negative cases and myelin oligodendrocyte glycoprotein antibody-positive cases), the prevalence reaches 4.4 per 100,000. The prevalence of NMOSD in Caucasians is remarkably less than the rates observed in Asia; however, in Hungarians, it reaches 1.91, which may be due to the admixture of Asian genes.⁸ In northern Europe, the HLA-types HLA-DRB1*03 and HLA-DRB1*01:01 have been observed to increase the susceptibility of NMOSD.⁷⁷ The main focus of the studies in Caucasian populations was smoking and infections.

Regarding the impact of smoking, a significant disparity was observed between studies on Asians and Caucasians. In two studies from Canada and the USA, respectively, patients with a history of smoking had a significantly lower risk of NMOSD.^23,64 In contrast, smoking did not impact the risk of NMOSD in Asian studies, and in a study from Iran, passive smoking even increased the susceptibility.^11,33 It has been established that smoking can affect several both innate and adaptive immune responses, and the effect of smoking in the onset of MS has been previously established.¹⁷ The function of toll-like receptors, dendritic cells, and natural killer cells is altered by smoking, and the number and activity of CD8+ cells increase in smokers.⁸⁶ Smoking also leads to an increase in memory B cells and a decrease in regulatory B cells.⁸⁹ Moreover, studies indicate that smoking can damage astrocytes, which may consequently lead to NMOSD development.⁹⁰

Nevertheless, some studies report the anti-inflammatory effect of nicotine, which may help explain the paradox. Nicotine can decrease the proliferation rate of lymphocytes and leukocyte infiltration into the CNS. Jiang et al. showed that in a model of MS, the infiltration of proinflammatory monocytes and neutrophils is inhibited by nicotine through its impact on nicotinic acetylcholine receptors. Moreover, in nicotine-treated animal models, CCL2 and CXCL2 are reportedly reduced. Such chemokines involve proinflammatory cell chemotaxis and might affect disease onset.⁹¹ More studies are needed further to evaluate the effect of smoking habits on NMOSD occurrence.

Diet

Studies show that high carbohydrate and sugar intake increases the risk of NMOSD. A possible explanation for the effect of sugar on the nervous system might be dysbiosis of gut microbiota and the pro-inflammatory effect. By increasing pro-inflammatory factors like CRP, sugar consumption can release reactive oxygen species.⁹² Sugar increases intestinal permeability and dysbiosis in gut microflora, leading to the stimulation of proinflammatory factors, such as TNF-α and IL-6, that can cause systemic inflammation.⁹³ By the Brain-Gut axis hypothesis, modifying sugar intake can improve the microflora profile and reduce inflammation, which helps decrease the odds of NMOSD.⁹⁴ A high-energy and high-carbohydrate diet causes excess weight, which some studies indicate is a risk factor for neurodegenerative disease.⁹⁵ Adipose tissue is also a significant factor in autoimmune diseases through its impact on the proliferation of TH17 cells.⁹⁶

Socioeconomic status

Regarding SES, the included study did not show any significance, and according to the literature, MS follows the same pattern. In a systematic review by Goulden et al.⁹⁷ in 2015, no association was observed between SES and NMOSD risk. Another cohort in Denmark found the same result.⁹⁸

Hormonal factors

It was observed in the studies that, in contrast to OCP use, breastfeeding prevents NMOSD onset. The same result was reported when assessing MS cases. In a 2016 case-control study, it was observed that a history of OCP use is related to a 50% increase in the risk of MS.⁹⁹ It is believed that breastfeeding alters the risk of MS onset, which might be due to ovarian suppression that leads to a reduction in inflammatory activity.¹⁰⁰ Moreover, being breastfed as an infant also lowers the risk of MS onset and delays it, as it contains IL-10 and transforming growth factor-beta, which have anti-inflammatory properties.¹⁰¹

The development of NMOSD might be due to the same reasons as well.

Limitations

Like any other systematic review, a limitation of this study is publication bias and heterogeneity in studies that make conclusions difficult for the readers; moreover, limited studies address the environmental risk factors of NMOSD, especially in African populations.

Conclusion

The review of 50 environmental risk factors in the current study showed that vitamin D deficiency, cigarette smoking, MAP infection, and diet had a significant association with NMOSD. In East Asians, vitamin D deficiency contributed to increased risk of NMOSD. Controversial results were reported regarding the effect of smoking among Caucasians and other races. Further studies are required to elucidate the findings presented in this review study.

Supplemental Material

sj-docx-1-tan-10.1177_17562864251363293 – Supplemental material for Environmental risk factors of neuromyelitis optica spectrum disorder: a systematic review

Supplemental material, sj-docx-1-tan-10.1177_17562864251363293 for Environmental risk factors of neuromyelitis optica spectrum disorder: a systematic review by Mahsa Mohammadi Lapevandani, Elham Bazmi, Shima Jahani, Nasrin Asgari and Mohammad Ali Sahraian in Therapeutic Advances in Neurological Disorders

Footnotes

Appendix

Acknowledgements

None.

Declarations

ORCID iDs

Mahsa Mohammadi Lapevandani

Elham Bazmi

Shima Jahani

Nasrin Asgari

Mohammad Ali Sahraian

Supplemental material

Supplemental material for this article is available online.

References

Wingerchuk

Banwell

Bennett

, et al. International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology 2015; 85(2): 177–189.

Lennon

Wingerchuk

Kryzer

, et al. A serum autoantibody marker of neuromyelitis optica: distinction from multiple sclerosis. Lancet (London, England) 2004; 364(9451): 2106–2112.

Gupta

Khandelwal

Nelson

, et al. Structural basis of aquaporin-4 autoantibody binding in neuromyelitis optica. Sci Adv 2025; 11(8): eadq7560.

Zhang

Verkhratsky

Shi

FD.

Astrocytes and microgliain multiple sclerosis and neuromyelitis optica. In: Verkhratsky

de Witte

Aronica

, et al. (eds) Handbook of clinical neurology. Elsevier, Vol. 210, 2025, pp. 133–145.

Cacciaguerra

Meani

Mesaros

, et al. Brain and cord imaging features in neuromyelitis optica spectrum disorders. Ann Neurol 2019; 85(3): 371–384.

Patterson

Goglin

SE.

Neuromyelitis optica. Rheum Dis Clin North Am 2017; 43(4): 579–591.

Papp

Magyari

Aktas

, et al. Worldwide incidence and prevalence of neuromyelitis optica: a systematic review. Neurology 2021; 96(2): 59–77.

Hor

Asgari

Nakashima

, et al. Epidemiology of neuromyelitis optica spectrum disorder and its prevalence and incidence worldwide. Front Neurol 2020; 11: 501.

Rezaeimanesh

Sahraian

Moghadasi

, et al. Epidemiology of neuromyelitis optica spectrum disorder in Tehran, Iran: the prevalence, baseline characteristics, and clinical aspects. Neurol Sci 2020; 41(9): 2647–2648.

10.

Watanabe

Isobe

Niino

, et al. Prevalence of, and disability due to, multiple sclerosis and neuromyelitis optica spectrum disorder in Japan by the Fifth Nationwide Survey. Neurology 2024; 103(10): e209992.

11.

Eskandarieh

Nedjat

Abdollahpour

, et al. Environmental risk factors in neuromyelitis optica spectrum disorder: a case-control study. Acta Neurol Belg 2018; 118(2): 277–287.

12.

Kim

Mealy

Levy

, et al. Racial differences in neuromyelitis optica spectrum disorder. Neurology 2018; 91(22): e2089–e2099.

13.

Ghafouri-Fard

Azimi

Taheri

A comprehensive review on the role of genetic factors in neuromyelitis optica spectrum disorder. Front Immunol 2021; 12: 737673.

14.

Shahmohammadi

Doosti

Shahmohammadi

, et al. Autoimmune diseases associated with neuromyelitis optica spectrum disorders: a literature review. Mult Scler Relat Disord 2019; 27: 350–363.

15.

Wang

Shi

Zhao

, et al. The causal relationship between neuromyelitis optica spectrum disorder and other autoimmune diseases. Front Immunol 2022; 13: 959469.

16.

Barbero

Mazzucca C

Scotti

Comi

, et al. The role of diet in multiple sclerosis onset: a prospective study using UK biobank. Nutrients 2024; 16(11): 1746.

17.

Abbasi

Kasbi N

Ghane

Ezabadi S

Kohandel

, et al. Lifetime exposure to smoking and substance abuse may be associated with late-onset multiple sclerosis: a population-based case-control study. BMC Neurol 2024; 24(1): 327.

18.

Mortazavi

Moghadasi

Almasi-Hashiani

, et al. Waterpipe and cigarette smoking and drug and alcohol consumption, and the risk of primary progressive multiple sclerosis: a population-based case-control study. Curr J Neurol 2023; 22(2): 72–81.

19.

Bafrani

Asadigandomani

Kasbi

, et al. The coincidence of multiple sclerosis and primary vasculitis; from the bench of pathology to the bedside of treatment: a systematic review of case reports. Neurol Sci 2025; 46: 351–364.

20.

Gholizadeh

Exuzides

Sinnott

, et al. Assessment of disability and disease burden in neuromyelitis optica spectrum disorders in the CIRCLES cohort. Sci Rep 2024; 14(1): 26150.

21.

Hedström

AK.

Risk factors for multiple sclerosis in the context of Epstein-Barr virus infection. Front Immunol 2023; 14: 1212676.

22.

Min

Waters

Vincent

, et al. Low levels of vitamin D in neuromyelitis optica spectrum disorder: association with disease disability. PLoS One 2014; 9(9): e107274.

23.

Rotstein

Wolfson

Carruthers

, et al. A national case-control study investigating demographic and environmental factors associated with NMOSD. Mult Scler 2023; 29(4–5): 521–529.

24.

Niegowska

Arru

, et al. Mycobacterium avium subspecies paratuberculosis and myelin basic protein specific epitopes are highly recognized by sera from patients with Neuromyelitis optica spectrum disorder. J Neuroimmunol 2018; 318: 97–102.

25.

Graves

Grandhe

Weinfurtner

, et al. Protective environmental factors for neuromyelitis optica. Neurology 2014; 83(21): 1923–1929.

26.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ (Clin Res ed) 2021; 372: n71.

27.

Liberati

Altman

Tetzlaff

, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Internal Med 2009; 151(4): W65–W94.

28.

Hutton

Salanti

Caldwell

, et al. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations. Ann Intern Med 2015; 162(11): 777–784.

29.

C-KL

Mertz

Loeb

Newcastle-Ottawa Scale: comparing reviewers’ to authors’ assessments. BMC Med Res Methodol 2014; 14: 45.

30.

Paz

ÉS

Maciel

PMCT

D’Almeida JAC , et al. Excess weight, central adiposity and pro-inflammatory diet consumption in patients with neuromyelitis optica spectrum disorder. Mult Scler Relat Disord 2021; 54: 103110.

31.

Yang

Gao

, et al. Incidence of neuromyelitis optica spectrum disorders in China: a large cohort study using claim data. BMJ Open 2022; 12(1): e048942.

32.

Akaishi

Tarasawa

Matsumoto

, et al. Associations between neuromyelitis optica spectrum disorder, Sjögren’s syndrome, and conditions with electrolyte disturbances. J Neurol Sci 2023; 452: 120742.

33.

Yang

Yin

Liu

, et al. A preliminary study of association of cigarette smoking with risk of neuromyelitis optica spectrum disorder. Medicine 2021; 100(37): e27234.

34.

Long

Gao

Qiu

, et al. Helicobacter pylori infection in neuromyelitis optica and multiple sclerosis. Neuroimmunomodulation 2013; 20(2): 107–112.

35.

Yoshimura

Isobe

Matsushita

, et al. Distinct genetic and infectious profiles in Japanese neuromyelitis optica patients according to anti-aquaporin 4 antibody status. J Neurol Neurosurg Psychiatry 2013; 84(1): 29–34.

36.

Gao

Yao

Ding

, et al. Low levels of vitamin D and the relationship between vitamin D and Th2 axis-related cytokines in neuromyelitis optica spectrum disorders. J Clin Neurosci 2019; 61: 22–27.

37.

Shan

Tan

Zhang

, et al. Serum 25-hydroxyvitamin D(3) is associated with disease status in patients with neuromyelitis optica spectrum disorders in south China. J Neuroimmunol 2016; 299: 118–123.

38.

Masuda

Mori

Arai

, et al. Epstein-Barr virus persistence and reactivation in neuromyelitis optica. J Neurol Neurosurg Psychiatry 2015; 86(10): 1137–1142.

39.

Mori

Hosoya

Hiwasa

, et al. Detection of mumps virus RNA in cerebrospinal fluid of patients with neuromyelitis optica. Neurol Sci 2011; 32(5): 795–799.

40.

Cossu

Tomizawa

Sechi

, et al. Epstein-Barr virus and human endogenous retrovirus in Japanese patients with autoimmune demyelinating disorders. Int J Mol Sci 2023; 24(24): 17151.

41.

Fang

Liu

, et al. Secondary immunoreaction in patients with neurosyphilis and its relevance to clinical outcomes. Front Neurol 2023; 14: 1201452.

42.

Zhong

Qiu

, et al. Association between neuromyelitis optica and tuberculosis in a Chinese population. BMC Neurol 2014; 14: 33.

43.

Koga

Takahashi

Kawai

, et al. A serological analysis of viral and bacterial infections associated with neuromyelitis optica. J Neurol Sci 2011; 300(1–2): 19–22.

44.

Kim

Park

, et al. Onset of various CNS inflammatory demyelination diseases following COVID-19 vaccinations. Mult Scler Relat Disord 2022; 68: 104141.

45.

Wang

Zhou

ZhangBao

, et al. Causal associations between prodromal infection and neuromyelitis optica spectrum disorder: a Mendelian randomization study. Eur J Neurol 2023; 30(12): 3819–3827.

46.

Tian

D-C

Yuan

, et al. Incidence of neuromyelitis optica spectrum disorder (NMOSD) in China: a national population-based study. Lancet Reg Health West Pac 2020; 2: 100021.

47.

Rafiee

Tarjoman

Moghadasi

, et al. Stressful life events, socioeconomic status, and the risk of neuromyelitis optica spectrum disorder: a population-based case-control study. Mult Scler Relat Disord 2020; 46: 102544.

48.

Rezaeimanesh

Ariyanfar

Sahraian

, et al. Whole grains and legumes consumption in association with neuromyelitis optica spectrum disorder odds. Curr J Neurol 2021; 20(3): 131–138.

49.

Rezaeimanesh

Jahromi

Ghorbani

, et al. Low carbohydrate diet score and odds of neuromyelitis optica spectrum disorder: a case-control study. Int J Vitam Nutr Res 2022; 92(5–6): 321–330.

50.

Rezaeimanesh

Jahromi

Sahraian

, et al. The association between dietary total antioxidant capacity and NMO-IgG seropositivity in patients with neuromyelitis optica spectrum disorder. Clin Neurol Neurosurg 2021; 209: 106903.

51.

Mirmosayyeb

Barzegar

Afshari-Safavi

, et al. Evaluation of month of birth in neuromyelitis optica spectrum disorders (NMSOD) and multiple sclerosis (MS). Mult Scler Int 2021; 2021: 8874999.

52.

Malli

Pandit

D’Cunha

, et al. Helicobacter pylori infection may influence prevalence and disease course in myelin oligodendrocyte glycoprotein antibody associated disorder (MOGAD) similar to MS but not AQP4-IgG associated NMOSD. Front Immunol 2023; 14: 1162248.

53.

Barzegar

Nouri

Mirmosayyeb

, et al. Association between Helicobacter pylori infection and seronegative neuromyelitis optica spectrum disorder. Caspian J Neurol Sci 2023; 9(1): 9–14.

54.

Etemadifar

Izadi

Sabeti

, et al. Anti-HSV-2 antibody in patients with MS and NMO. Mult Scler Relat Disord 2019; 28: 286–289.

55.

Samim

Dhar

Arshad

, et al. Co-VAN study: COVID-19 vaccine associated neurological diseases—an experience from an apex neurosciences centre and review of the literature. J Clin Neurosci 2023; 108: 37–75.

56.

Nayanajeehwi

Wijesundara

Jayasinghe

, et al. Serum redox status, cytokines and vitamin D levels of multiple sclerosis and neuromyelitis optica spectrum disorder in Sri Lanka: a comparative pilot study. Asian J Res Rep Neurol 2023; 6(1): 129–141.

57.

Puangsricharoen

Vanikieti

Jindahra

, et al. Serum vitamin D levels and status in thai optic neuritis subjects: a case-control study. Clin Ophthalmol (Auckland, NZ) 2022; 16: 3381–3389.

58.

Kusumadewi

Imran

Witjaksono

, et al. Low vitamin D-25(OH) level in Indonesian multiple sclerosis and neuromyelitis optic patients. Mult Scler Relat Disord 2018; 25: 329–333.

59.

Ashtari

Madanian

Zarkesh

, et al. Serum levels of interleukin-6 and vitamin D at the onset of multiple sclerosis and neuromyelitis optica: a pilot study. J Res Med Sci 2022; 27: 67.

60.

Francis

Elhadd

Camera

, et al. Acute inflammatory diseases of the central nervous system after SARS-CoV-2 vaccination. Neurol Neuroimmunol Neuroinflamm 2023; 10(1): e200063.

61.

Lee

Muccilli

Schneider

, et al. Acute central nervous system inflammation following COVID-19 vaccination: an observational cohort study. Mult Scler (Houndmills, Basingstoke, England) 2023; 29(4–5): 595–605.

62.

Bukhari

Prain

Waters

, et al. Incidence and prevalence of NMOSD in Australia and New Zealand. J Neurol Neurosurg Psychiatry 2017; 88(8): 632–638.

63.

Siuko

Kivelä

Setälä

, et al. The clinical spectrum and prognosis of idiopathic acute optic neuritis: a longitudinal study in Southern Finland. Mult Scler Relat Disord 2019; 35: 215–220.

64.

Simon

Schmidt

Loud

, et al. Risk factors for multiple sclerosis, neuromyelitis optica and transverse myelitis. Mult Scler (Houndmills, Basingstoke, England) 2015; 21(6): 703–709.

65.

Sellner

Cepok

Kalluri

, et al. Aquaporin 4 antibody positive central nervous system autoimmunity and multiple sclerosis are characterized by a distinct profile of antibodies to herpes viruses. Neurochem Int 2010; 57(6): 662–667.

66.

Slavin

Caggiu

, et al. High levels of antibodies against PtpA and PknG secreted by Mycobacterium avium ssp. paratuberculosis are present in neuromyelitis optica spectrum disorder and multiple sclerosis patients. J Neuroimmunol 2018; 323: 49–52.

67.

Aubart

Roux

Durrleman

, et al. Neuroinflammatory disease following severe acute respiratory syndrome coronavirus 2 infection in children. J Pediatr 2022; 247: 22–28.e2.

68.

Mello

CDS

Cabral-Castro

Silva de Faria

, et al. Dengue and chikungunya infection in neurologic disorders from endemic areas in Brazil. Neurol Clin Pract 2020; 10(6): 497–502.

69.

Huang

Wang

, et al. The nutritional risk in patients with neuromyelitis optica spectrum disorder. Mult Scler Relat Disord 2023; 77: 104900.

70.

Ercolini

Miller

SD.

The role of infections in autoimmune disease. Clin Exp Immunol 2009; 155(1): 1–15.

71.

Zatjirua

Butler

Carr

, et al. Neuromyelitis optica and pulmonary tuberculosis: a case-control study. Int J Tuberc Lung Dis 2011; 15(12): 1675–1680.

72.

Papais-Alvarenga

Miranda-Santos

Puccioni-Sohler

, et al. Optic neuromyelitis syndrome in Brazilian patients. J Neurol Neurosurg Psychiatry 2002; 73(4): 429–435.

73.

Wang

Zhang

, et al. Long-term outcomes of varicella zoster virus infection-related myelitis in 10 immunocompetent patients. J Neuroimmunol 2018; 321: 36–40.

74.

Netravathi

Dhamija

Gupta

, et al. COVID-19 vaccine associated demyelination & its association with MOG antibody. Mult Scler Relat Disord 2022; 60: 103739.

75.

Çag

Haciseyitoglu

Distribution of adenovirus and rotavirus by age and seasons in children diagnosed with viral gastroenteritis. Mediterr J Infect Microb Antimicrob 2021; 10: 10.

76.

Bagherieh

Afshari-Safavi

Vaheb

, et al. Worldwide prevalence of neuromyelitis optica spectrum disorder (NMOSD) and neuromyelitis optica (NMO): a systematic review and meta-analysis. Neurol Sci 2023; 44(6): 1905–1915.

77.

Matsushita

Masaki

Isobe

, et al. Genetic factors for susceptibility to and manifestations of neuromyelitis optica. Ann Clin Transl Neurol 2020; 7(11): 2082–2093.

78.

Ascherio

Munger

White

, et al. Vitamin D as an early predictor of multiple sclerosis activity and progression. JAMA Neurol 2014; 71(3): 306–314.

79.

Allen

Kelly

Basdeo

, et al. A pilot study of the immunological effects of high-dose vitamin D in healthy volunteers. Mult Scler (Houndmills, Basingstoke, England) 2012; 18(12): 1797–1800.

80.

Müller

Odum

Bendtzen

1,25-Dihydroxyvitamin D3 selectively reduces interleukin-2 levels and proliferation of human T cell lines in vitro. Immunol Lett 1993; 35(2): 177–182.

81.

Liu

Tan

Zhou

, et al. Vitamin D status and the risk of neuromyelitis optica spectrum disorders: a systematic review and meta-analysis. J Clin Neurosci 2024; 119: 185–192.

82.

Jiang

, et al. High prevalence of vitamin D deficiency in Asia: a systematic review and meta-analysis. Crit Rev Food Sci Nutr 2023; 63(19): 3602–3611.

83.

Naserb

Moghadasi A

. Environmental and genetic risk factors in the development of neuromyelitis optica. Exp Rev Ophthalmol 2020; 15(1): 1–9.

84.

Naser

Moghadasi A

Rezaeimanesh

. Microorganisms in pathogenesis and management of neuromyelitis optica spectrum disorder. In: Role of microorganisms in pathogenesis and management of autoimmune diseases. Vol II. Kidney, central nervous system, eye, blood, blood vessels and bowel. Singapore: Springer, 2023, pp. 195–210.

85.

Burucoa

Axon

Epidemiology of Helicobacter pylori infection. Helicobacter 2017; 22(Suppl. 1).

86.

Maroufi

Mortazavi

Sahraian

, et al. Environmental risk factors of multiple sclerosis in the Middle East and North Africa region: a systematic review. Curr J Neurol 2021; 20(3): 166–184.

87.

Papazisi

Gorton

Kutish

, et al. The complete genome sequence of the avian pathogen Mycoplasma gallisepticum strain R(low). Microbiology (Reading, England) 2003; 149(Pt 9): 2307–2316.

88.

Feng

Guo

Huang

, et al. Anti-tuberculosis treatment for Devic’s neuromyelitis optica. J Clin Neurosci 2010; 17(11): 1372–1377.

89.

Eskandarieh

Moghadasi

Sahraiain

, et al. Association of cigarette smoking with neuromyelitis optica-immunoglobulin G sero-positivity in neuromyelitis optica spectrum disorder. Iran J Neurol 2019; 18(3): 93–98.

90.

Lee

S-B

Kim

J-H

Cho

M-H

, et al. Impact of commercial cigarette smoke condensate on brain tissue co-cultured with astrocytes and blood-brain barrier endothelial cells. J Toxicol Environ Health A 2017; 80(10–12): 533–541.

91.

Jiang

St-Pierre

Roy

, et al. Infiltration of CCR2+Ly6Chigh proinflammatory monocytes and neutrophils into the central nervous system is modulated by nicotinic acetylcholine receptors in a model of multiple sclerosis. J Immunol (Baltimore, MD: 1950) 2016; 196(5): 2095–2108.

92.

Della

Corte KW

Perrar

Penczynski

, et al. Effect of dietary sugar intake on biomarkers of subclinical inflammation: a systematic review and meta-analysis of intervention studies. Nutrients 2018; 10(5): 606.

93.

Esposito

Bonavita

Sparaco

, et al. The role of diet in multiple sclerosis: a review. Nutr Neurosci 2018; 21(6): 377–390.

94.

Lombardi

De Meirleir

Subramanian

, et al. Nutritional modulation of the intestinal microbiota; future opportunities for the prevention and treatment of neuroimmune and neuroinflammatory disease. J Nutr Biochem 2018; 61: 1–16.

95.

Endo

Yokote

Nakayama

The obesity-related pathology and Th17 cells. Cell Mol Life Sci 2017; 74(7): 1231–1245.

96.

Bodén

Wennberg

Van Guelpen

, et al. Dietary inflammatory index and risk of first myocardial infarction; a prospective population-based study. Nutr J 2017; 16(1): 21.

97.

Goulden

Ibrahim

Wolfson

Is high socioeconomic status a risk factor for multiple sclerosis? A systematic review. Eur J Neurol 2015; 22(6): 899–911.

98.

Nielsen

Jørgensen

Bager

, et al. Socioeconomic factors in childhood and the risk of multiple sclerosis. Am J Epidemiol 2013; 177(11): 1289–1295.

99.

Hellwig

Chen

Stancyzk

, et al. Oral contraceptives and multiple sclerosis/clinically isolated syndrome susceptibility. PLoS One 2016; 11(3): e0149094.

100.

Collorone

Kodali

Toosy

AT.

The protective role of breastfeeding in multiple sclerosis: latest evidence and practical considerations. Front Neurol 2022; 13: 1090133.

101.

Hajibagher

Naeeni R

Soltanzadeh

Nafissi

, et al. The association of breastfeeding and multiple sclerosis: does formula have any other risks? Arch Adv Biosci 2024; 15(1): 1–7.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB