Prevalence and Predictors of Low Folate Levels among Stroke Survivors in a Country without Mandatory Folate Food Fortification: Analysis of a Ghanaian Sample

Abstract

Background:

While additional folic acid (FA) treatment has a neutral effect on lowering overall vascular risk in countries that mandate FA fortification of food, meta-analytic data suggest that folate supplementation reduces stroke risk in certain patient subgroups, and among people living in countries without mandatory folate food fortification. However, the burden of folate deficiency among adults with stroke in the world’s poorest continent is unknown.

Purpose:

To assess the prevalence and predictors of folate deficiency among recent ischemic stroke survivors.

Methods:

We analyzed data among consecutively encountered ischemic stroke patients aged ≥18 years at a tertiary medical center in Kumasi, Ghana between 10/2020 – 08/2021. We identified a modest sample of stroke free adults to serve as a comparator group. Fasting serum folate was measured using a radioimmunoassay and a cut-off of 4ng/mL used to define folate deficiency. Factors associated with serum folate concentration were assessed using a multilinear regression model.

Results:

Comparing stroke cases (n=116) with stroke-free comparators (n=20), mean folate concentration was lower among stroke cases (7 ng/ml vs. 10.2 ng/ml, p=0.004). Frequency of folate deficiency was higher among stroke cases vs. stroke-free controls (31% vs 5%, p=0.02). Male sex (beta coefficient of −2.6 (95% CI: −4.2, −0.9) and LDL (β: −0.76; −1.4, −0.07) were significantly associated with serum folate concentration.

Conclusion:

Almost one in three ischemic stroke survivors have folate deficiency potentially accentuating their risk for further adverse atherosclerotic events in a setting without folate fortification. A clinical trial of folate supplementation among stroke survivors is warranted.

INTRODUCTION

Vitamin supplementation with folic acid (FA) reduces hyperhomocysteinemia (Hcy), a vascular risk factor with a strong, graded, and independent association with stroke and other vascular diseases.¹ Available data suggest that a 25% reduction in total plasma homocysteine level is accompanied by a 19% lower stroke risk.^1,2 Evidence from clinical trials on the effect of folate supplementation on mitigation of stroke risk has however been conflicting. A synthesis of clinical trial data has shown that supplemental FA had a neutral effect on reducing vascular events in countries with a mandate for folate fortification of food, for example the United States.^3,4 Nevertheless, in countries without mandatory FA food fortification, data from 13 RCTs showed that FA therapy reduced future stroke risk by 15% (95% CI, 5%–23%) compared with control populations.⁵ Furthermore, trials using lower dosage of folate such as 0.8mg daily and those with additional Vitamin B₁₂ reported greater benefits in stroke risk reduction from folate therapy.⁶

In view of these conflicting lines of evidence, the panel who wrote the 2021 ASA/AHA guidelines on secondary stroke prevention did not recommend folate supplementation for vascular risk reduction after stroke.⁷ The panel recommended further studies to further understand the factors which modify the relationship between folate supplementation and risk stroke risk reduction such as renal function, genetic variants in folate metabolism and preexisting homocysteine levels. ⁷ Beyond vascular event risk reduction, folate level exposure has been associated with post-stroke cerebral atrophy⁸, with FA supplementation retarding progression of substrates of cognitive impairment such as cerebral white matter lesions.⁹ There is therefore a clear need for further studies to characterize the associations between folate exposure and vascular risk reduction and cognitive performance among stroke survivors at risk of adverse neurovascular outcomes. These studies are especially warranted in low-resource settings where folate fortification is not mandated, and the burden of stroke is escalating.^10–14 The objective of the present study is to assess the frequency of folate deficiency among recent Ghanaian stroke survivors and factors associated with serum levels of folate in this setting.

METHODS

Study design:

This was a cross-sectional study conducted at the Neurology clinic of the Komfo Anokye Teaching Hospital. The study was approved by the Committee on Human Research Publication and Ethics of the Kwame Nkrumah University of Science and Technology.

Study participants:

Eligible participants included were 18 years old with a recent CT-scan confirmed ischemic stroke and any of the following additional conditions: documented diabetes mellitus or previous treatment with oral hypoglycemic or insulin; documented hypertension >140/90mmHg or previous treatment with anti-hypertensive medications. The present data analysis are limited to baseline data of study participants whose serum samples were assessed for folate acid concentration. We also recruited a limited number of age-matched stroke-free adults to serve as a comparator group for serum folate measurements. Control subjects were community dwelling adults who were matched to stroke cases in a ratio of ≈1:5 and were recruited for the purpose of serving as a comparator group for folate assay measurements. Beside age and sex, controls were not matched for any other variable. Age matching was within a range of +/− 5 years.

Data collection:

We administered structured questionnaires to stroke survivors or proxies to collect data on demographic characteristics, history of vascular risk factors and lifestyle factors such use of cigarette and alcohol. A detailed medical history including history of hypertension or diabetes diagnosis and current medications lists taken were obtained. Anthropometric evaluations including measurement of weight, height and waist circumference were performed by Study nurses. Body mass index (BMI) of each participant was then derived by dividing the weight in kilograms by the square of the height in meters.

Fasting blood samples were obtained at venesection between 0800 and 1000 am. Participants were instructed to abstain from ingesting alcohol for 3 days, and from smoking and eating for 12 hours (overnight fast). Blood samples were collected into vacuum tubes, with separation of serum for storage at −80C until analysis. Serum samples were assayed for concentrations of folate, lipid profile parameters namely total cholesterol, low density lipoprotein cholesterol, high-density lipoprotein cholesterol and triglycerides. Serum folate concentrations were measured using a radioimmunoassay (Quantaphase II; Bio-Rad, Hercules, CA) in one batch.

We assessed cognitive performance using the Montreal Cognitive Assessment (MOCA)¹⁵ and the Vascular Neuropsychological Battery (V-NB).¹⁶ While the MOCA is considered a test of general cognitive functioning, the V-NB is comprised of a battery of tests which evaluates the functions of specific cognitive domains and was patterned after the NINDS-CSN Harmonization Standards 60-minute neuropsychological protocol.¹⁵ Specifically, the V-NB assesses 4 key domains namely executive function, memory/learning, language and visuospatial/visuoconstructive skills using validated test items. Vascular Cognitive Impairment was defined as impairment in at least 1 cognitive domain (memory/learning, executive domain, visuospatial/visuoconstructive skills and language) and normal or mild impairment of activities of daily living independent of motor/sensory symptoms according to the American Stroke Association/American Heart Association Vascular Cognitive Impairment Guideline.¹⁷ Post stroke dementia was defined according to the DSM IV criteria as impairment in > 2 cognitive domains of sufficient severity to affect the individuals ability to perform activities of daily living independent of motor or sensory symptoms.¹⁷ Functional impairment was defined as a Barthel Index score of less than 75%.¹⁸

The average thickness of the left and right common carotid arteries (CCA) obtained at 3 angles (anterior, lateral and posterior bilaterally) at the optimum angle of insonation was reported as the overall CIMT for the CCA.¹⁹ Measurements were also taken at the carotid bifurcations and internal carotid arteries (ICA) bilaterally. Replicate image acquisitions were independently performed by two trained sonographers who were blinded to the status of study participants and risk factor status.

Definition of risk factors

We collected basic demographic and lifestyle data including, socioeconomic status, cardiovascular risk profile, cigarette smoking, and alcohol use.

• Hypertension: Blood pressure was recorded at baseline and daily for 7 days. Hypertension was defined as a sustained elevation of blood pressure ≥140/90 mmHg >72 hours after stroke, a premorbid history of hypertension, use of antihypertensive drugs before stroke or >72 hours after stroke onset. Definition of hypertension in controls was self-reported history of hypertension or use of antihypertensive drugs or average of 3 recorded BP at first clinical encounter ≥140/90mmHg.²⁰

• Diabetes mellitus was defined based on history of diabetes mellitus, use of medications for DM, an HBA1c >6.5% or a fasting blood glucose (FBG) levels > 7.0 mmol/l at first encounter in controls or measured after the post-acute phase in cases due to the known acute transient elevation of glucose as a stress response after stroke.²¹

• Alcohol use was categorized into current users (users of any form of alcoholic drinks) or never/former drinker while alcohol intake was categorized as low drinkers (1–2 drinks per day for female and 1–3 drinks per day for male) and high drinker (>2 drinks per day for female and >3 drinks per day for male. 1 drink or 1 unit of alcohol = 8g of alcohol).²⁰

• Smoking status was defined as current smoker (individuals who smoked any tobacco in the past 12 months) or never/former smoker.²⁰

Statistical analysis:

We compared demographic and clinical characteristics of stroke survivors by folate status into low folate versus normal folate concentrations. Low serum folate concentration was defined as serum folate level < 4ng/ml.²² The means or medians of continuous variables were compared using either the Students t-test or Mann Whitney’s U-test. Categorical variables were compared using the Chi-squared test. A multivariate linear regression analysis was fitted to identify factors independently associated with serum folate concentration. A p-value of <0.05 was considered statistically significant. Model diagnosis and fit were assessed using residual plots analysis. Statistical analysis was performed using SAS 9.4.

RESULTS

Comparison of frequency of folate deficiency among stroke survivors versus stroke-free unmatched controls:

We enrolled 116 stroke cases between 8^th February 2019 to 11^th December 2020 and 20 stroke-free controls between 14^th December 2019 to 12^th February 2020. Frequency of folate deficiency among stroke-free controls (n=20) was 5% vs 31% among stroke survivors (n=116), p=0.02; mean serum folate concentration among controls was higher at 10.2 ng/ml vs 6.7 ng/ml among stroke cases, p=0.004. The mean (SD) age of stroke-free comparator group was 55.9 (9.5) years vs 58.6 (12.0) years, p = 0.35. There were 8 (40%) females in the stroke-free group vs 62 (53.4%), in the stroke group, p=0.34.

Comparison of stroke survivors by serum folate status:

The mean ± SD concentration of serum folate level among 116 ischemic stroke cases of 6.7 ± 4.6 ng/ml. Among study participants 36 (31.0%) had low serum concentration of folate. Demographic characteristics such as mean age, gender distribution, educational status and location of domicile of stroke survivors did not differ significantly by folate status (Table 1). The frequency of vascular risk factors such as diabetes, cigarette use, alcohol use and lipid panel concentrations did not differ between the two groups although the median estimated glomerular filtration rate was significantly lower in those with low folate concentrations. Furthermore, those with low folate level had thicker intimal media thickness at the carotid bulb of 1.38 ± 0.81 mm vs 1.11 ± 0.39 for those with normal folate concentration, p=0.02.

Table 1:

Comparison of demographic and clinical characteristics of Ghanaian ischemic stroke survivors by serum folate status

Characteristic	Low folate N=36	Normal folate N=80	P-value
Age, mean ± SD	59.1 ± 12.7	58.3 ± 11.7	0.74
Sex, male (%)	20 (55.6)	34 (42.5)	0.19
Educational level, n (%)			0.87
None	8 (22.2)	15 (18.8)
Primary	14 (38.9)	38 (47.5)
Secondary	8 (22.2)	15 (18.8)
Tertiary	5 (13.9)	12 (15.0)
Type of domicile, n (%)			0.70
Rural	18 (51.4)	38 (47.5)
Urban	17 (49.6)	42 (52.5)
Stroke types, n (%)			0.33
Large artery atherosclerosis	8 (22.2)	21 (26.2)
Small vessel occlusion	5 (13.9)	19 (23.8)
Unknown	23 (63.9)	40 (50.0)
Modified Rankin score, median [IQR]	3 [2 – 4]	3 [2 – 3.5]	0.32
National Institute of Health Stroke Score, mean ± SD	18.1 (3.5)	17.7 (3.5)	0.60
Diabetes mellitus, n (%)	14 (38.9)	38 (47.5)	0.39
Cigarette use, n (%)	0 (0.0)	2 (2.5)	1.00
Alcohol use, n (%)	9 (25.0)	13 (16.3)	0.26
Baseline SBP, mean ± SD	164.0 ± 18	160.2 ± 18.6	0.31
Baseline DBP, mean ± SD	96.8 ± 15.9	92.2 ± 13.8	0.12
BMI, mean ± SD	26.2 ± 3.6	25.4 ± 5.1	0.42
Waist circumference (cm), mean ± SD	93.2 ± 12.4	90.9 ± 10.9	0.32
Total serum cholesterol (mmol/l)	4.52 ± 1.20	4.07 ± 1.42	0.084
HDL serum cholesterol (mmol/l)	1.30 ± 0.99	1.20 ± 0.24	0.545
LDL serum cholesterol (mmol/l)	2.71 ± 1.01	2.36 ± 1.26	0.115
Serum triglycerides (mmol/l)	1.54 ± 0.69	1.27 ± 0.71	0.058
Categorised eGFR			0.183
>89	6 (18%)	26 (34%)
60–89	15 (45%)	33 (43%)
15–59	12 (36%)	18 (23%)
Missing	2	3
eGFR, median [IQR]	69 [53.0–86.5]	81.0 [59.5–89.0]	0.05
Cognitive measures
Montreal Cognitive Assessment score, mean ± SD	14.8 ± 6.6	16.8 ± 6.2	0.12
Cognitive status baseline			0.87
Vascular Dementia	3 (8.6)	6 (7.5)
Vascular cognitive impairment	6 (17.1)	11 (13.8)
No Vascular cognitive impairment	26 (74.3)	63 (78.8)
Intimal media thickness
Subclinical atherosclerotic markers
Common carotid IMT (mean ±SD)	1.03 ± 0.17	1.05 ± 0.21	0.52
Carotid bulb IMT (mean ±SD)	1.38 ± 0.81	1.11 ± 0.39	0.02
Internal carotid IMT, (mean ±SD)	0.92 ± 0.20	0.93 ± 0.22	0.78
Antihypertensive medications, median [IQR]	3 [3 – 3]	3 [2 – 3]	0.05
Number on statin, n (%)	28 (77.8)	66 (82.5)	0.55

IMT = Intimal media thickness

Factors Associated with Serum Folate concentration among Stroke Survivors:

In linear regression modelling we found, sex, LDL-concentration and category of glomerular filtration rate to be associated with serum folate concentration as a continuous outcome variable. In the adjusted multiple linear regression model, male sex was associated with low folate exposure, beta coefficient of −2.6 (95% CI: −4.2, −0.9) and higher serum LDL was significantly associated with lower serum folate exposure, beta coefficient of −0.76 (−1.4, −0.07). While a low estimated glomerular filtration rate of 15–59 ml/min was significantly associated with serum folate concentration with a beta coefficient of −2.3 (95% CI: −4.6, −0.04) in bivariate analysis, this association was lost in our fully adjusted models.

DISCUSSION

Approximately 30% of Ghanaian ischemic stroke survivors in this study had low serum folate concentrations. Among a limited sample of stroke free adults, only 5% had folate deficiency placing stroke survivors at a higher predisposition towards lower folate exposure. Surprisingly, studies on the frequency of folate deficiency among stroke survivors were quite limited in our literature search. A study published in 2007, reported a rather high prevalence of 68% of elderly stroke survivors with folate deficiency (defined as serum folate < 6ng/mL) in Taiwan.⁸ However, among the general adult population, the prevalence of folate deficiency varies between <0.5% in the U.S. after folate fortification²³, <5% in Latin America²⁴ to 32.4% among elderly Nigerians²⁵ and 40% in North China²⁶.

Overall, male sex and higher LDL-cholesterol concentration were associated with lower serum folate exposure among stroke survivors in a linear regression model. A higher proclivity of males towards folate deficiency has been similarly observed among stroke free adults from previous studies conducted in China²⁶, Israel²⁷ and in the US²⁸. It has been suggested that certain behavioral patterns which are commoner among males such as tendency towards cigarette smoking, and alcohol consumption may partially explain these sex differences in folate exposure.²⁶ The inverse association between LDL-cholesterol and serum folate concentration is of considerable pathophysiological interest. Low folate exposure promotes hyperhomocysteinemia which in turns orchestrates elevations in cholesterol synthesis via activation of 3-hydroxy-3-methylglutaryl coenzyme A reductase in vascular endothelial cells.²⁹ An elegant mechanistic study among adults with atherosclerotic vascular risk factors clearly demonstrated that low dose supplementation with folate (0.4 mg daily for 12 weeks) significantly decreased concentrations of total and LDL-cholesterol.³⁰ Of note, a meta-analysis of 10 folate supplementation trials demonstrated significant reduction in progression of carotid intimal media regression, especially among individuals with high atherosclerotic CVD disease risk or chronic kidney disease.³¹ In the present study among stroke survivors, we found that those with folate deficiency had significantly lower estimated glomerular filtration rate and higher carotid bulb intimal media thickness in bivariate analysis (Table 1). We however did not find significant associations between folate levels and cognitive performance.

Implications and Future directions:

To date, there have been few clinical trials dedicated to evaluating the role of folate supplementation for secondary prevention among stroke survivors. The VISP trial was conducted in US and Canadian medical centers (folate fortification programs existent) among ischemic stroke survivors and found no difference in recurrent cerebral infarction, coronary heart disease or death with either a high dose Vitamin B formulation (2.5 mg folic acid, 0.4 mg of cobalamin, and 25mg of pyridoxine) or a low-dose formulation (20 μg of folic acid, 6 μg of cobalamin, and 200 μg of pyridoxine).³² The risk ratio observed in the VISP study was 1.0 (95% CI: 0.8–1.1).³² The VITATOP study conducted in 20 countries of 19 had no folate fortification, compared a high-dose Vitamin B strategy (2mg of folic acid, 0.5 mg of cobalamin and 25 mg of pyridoxine) vs placebo over a median of 3.4 years and found a risk ratio of 0.91 (95% CI: 0.82 to 1.00, p=0.05).³³ A sub-analysis of the VITATOP study found differential benefits of vitamin B supplementation among stroke types and etiologic subtypes with small vasculopathy namely primary intracerebral hemorrhage and small vessel occlusive ischemic strokes.³³ Lessons from these two landmark studies of folate supplementation trials among stroke survivors suggest that future trials should: (i) compare the relative effectiveness by stroke types and etiologic subtypes and (ii) compare a graduated dose strategy to identify differential benefits of low vs high dose folate compared with placebo. Indeed, it has been posited that at high doses folate could be detrimental by promoting methyl group donation will cause methylation of homocysteine to methionine. Furthermore, a methyl donor, high levels of folate may inadvertently promote other methylation reactions such as methylation of arginine to form asymmetric dimethylarginine (ADMA), which potently inhibits the endothelial nitric oxide synthetase.³⁴ Well- designed mechanistic trials would be required to identify stroke populations and settings under which folate supplementation would be most beneficial.^35–39

Limitations:

This is an observational cross-sectional study and therefore causal inferences cannot be drawn between the observed associations reported. Serum folate is more prone to short-term exposure rather than red blood cell folate levels. We did not measure red cell folate concentration in the present study. A detailed dietary history and homocysteine levels would have been helpful but this data was not collected. Stroke-free controls were not matched exactly by age but within an age range of +/− 5years. Stroke free status was assessed using a validated questionnaire for verifying stroke free status.⁴⁰

Conclusion:

1 in 3 Ghanaian ischemic stroke survivors have folate deficiency which could have implications for secondary prevention after stroke in a setting without folate fortification.

Table 2.

Factors Associated with Serum Folate levels Among Ghanaian stroke survivors

Characteristic	N	Unadjusted Beta (95% CI)	p-value	Adjusted Beta (95% CI)	p-value
Sex	115
Female		—
Male		−2.5 (−4.1, −0.84)	0.004	−2.6 (−4.2, −0.92)	0.003
Age at baseline (years)	115	0.05 (−0.02, 0.12)	0.189
Diabetes mellitus	115	1.4 (−0.29, 3.1)	0.106
Total serum cholesterol (mmol/l)	115	−0.56 (−1.2, 0.06)	0.080
HDL serum cholesterol (mmol/l)	115	−0.12 (−1.6, 1.4)	0.871
LDL serum cholesterol (mmol/l)	115	−0.88 (−1.6, −0.18)	0.015	−0.76 (−1.4, −0.07)	0.031
Serum triglycerides (mmol/l)	115	−0.38 (−1.6, 0.83)	0.541
Urine protein (umol/l)	105	−0.07 (−0.15, 0.02)	0.115
NIHSS Score	115	−0.21 (−0.45, 0.04)	0.102
MOCA score at baseline	115	−0.02 (−0.15, 0.11)	0.797
Carotid bulb intimal media thickness	115	−1.8 (−3.8, 0.19)	0.079
Categorised eGFR	110
>89		—		--
60–89		−1.1 (−3.2, 0.92)	0.285	−0.92 (−2.9, 11)	0.40
15–59		−2.3 (−4.6, −0.04)	0.049	−2.2 (−4.4, 0.04)	0.054

Highlights.

• 31% of Ghanaian ischemic stroke survivors have low serum folate levels

• Males and higher LDL-cholesterol concentration were associated with low folate concentrations

• Folate deficiency is highly prevalent among stroke survivors in a setting without folate fortification