Significant outcomes
Folate, but not cobalamin, deficiency was common in patients admitted to an acute psychiatric unit
Patients with mental disorders had higher total homocysteine (tHcy) levels with folate deficiency than controls, something which might indicate a higher prevalence of genetic variants in the folate metabolism in this population
Limitations
No genetic testing for the methylenetetrahydrofolate reductase (MTHFR) polymorphism was done
The control group included unselected patients from primary health care and could potentially include patient with mental disorders
As data were anonymous, it was not possible to control for potential medication interference with folate and cobalamin metabolism
Introduction
Studies indicate that patients with mental disorders have a poor diet and an increased prevalence of vitamin insufficiency compared to the general population (Aucoin et al., Reference Aucoin, Lachance, Cooley and Kidd2020; Ljungberg et al., Reference Ljungberg, Bondza and Lethin2020; Gabriel et al., Reference Gabriel, Oliveira, Bruna De, Berk, Brietzke, Jacka and Lafer2023). Essential micronutrients are vital to all cellular processes in the body, and particularly reduced levels of folate and cobalamin have been associated with neuropsychiatric symptoms and disease, including schizophrenia, bipolar disorder and major depression (Bottiglieri, Reference Bottiglieri1996).
Folate in the form of 5-methyltetrahydrofolate and cobalamin in the form of methylcobalamin are necessary for remethylation of the amino acid homocysteine to methionine, an important methyldonor in metabolism (Selhub, Reference Selhub1999). In adults, total homocysteine (tHcy) is primarily a metabolic marker of folate status, but it is also a marker for cobalamin and to a lesser extent for vitamin B6 status and deficiency of either of these vitamins will increase tHcy concentrations (Schneede et al., Reference Schneede, Refsum and Ueland2000).
The enzyme methylenetetrahydrofolate reductase (MTHFR) transforms 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate, the methyldonor in the remethylation of homocysteine to methionine (Bagley and Selhub, Reference Bagley and Selhub1998). In the Nordic population, 5–8% have a polymorphism in the gene encoding for the 5,10-MTHFR (C677T, Ala --> Val) enzyme, late reductase (MTHFR) (C677T, Ala --> Val) enzyme, converting 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate (Jacques et al., Reference Jacques, Bostom, Williams, Ellison, Eckfeldt, Rosenberg, Selhub and Rozen1996). When serum folate levels are low, the MTHFR polymorphism will impair folate metabolism and the remethylation of homocysteine to methionine and thereby increase plasma tHcy levels (Savage et al., Reference Savage, Lindenbaum, Stabler and Allen1994; Green et al., Reference Green, Allen, Bjorke-Monsen, Brito, Gueant, Miller, Molloy, Nexo, Stabler, Toh, Ueland and Yajnik2017a).
The MTHFR polymorphism has been linked to an increased risk of bipolar disorder, schizophrenia and major depression in the overall population (Zhang et al., Reference Zhang, Yang, Gai, Cheng, Guo, Sun and Hu2022).
As vitamin deficiencies and genetic variants may cause neurologic and psychiatric symptoms (Bottiglieri, Reference Bottiglieri1996; Hutto, Reference Hutto1997), it is relevant to investigate vitamin status in patients with mental disorders. The aim of this study was to compare folate and cobalamin status in patients admitted to an acute psychiatric unit to patients from primary health care, in order to reveal factors which may be important in the follow-up of patents with mental disorders.
Materials and methods
Study population
The study group was patients with mental disorders admitted to the psychiatric acute unit at Haukeland University Hospital from January 1st through December 31st 2022 (n = 981). The routine test panel for these patients included serum cobalamin, plasma tHcy, serum creatinine including glomerular filtration rate (GFR), whereas serum folate was requested if indicated (in 100 of the 981 patients). In February 2024, serum folate was added to the routine test panel and test results from patients admitted to the psychiatric unit during the period from February 15 to October 15 2024 (n = 603) were included in order to find the prevalence of low folate and cobalamin status in patients with mental disorders.
The control group included unselected patients from the primary health care (controls), who had their serum folate, cobalamin and tHcy requested by their general practitioner (n = 32,201). All blood samples were analysed at the Department of Medical Biochemistry and Pharmacology at Haukeland University Hospital, Bergen, Norway, from January 1st through December 31st 2022.
The data on serum folate, serum cobalamin, plasma/serum tHcy, serum creatinine and estimated glomerular filtration rate (eGFR), in addition to gender and age were anonymously extracted from the laboratory data system. If a patient had multiple blood tests, only the first available test results from the defined study period were included. Patients <18 years were excluded in both groups. All controls (n = 32,201) and all admitted patients with mental disorders (n = 1584) had data on plasma/serum tHcy. Missing data in the control group included serum folate (n = 1693), serum cobalamin (n = 1582) and serum creatinine (n = 3884) and in the mental disorders admitted group was serum folate (n = 943), serum cobalamin (n = 3) and serum creatinine (n = 9).
The Regional Committee for Medical and Health Research Ethics, Region Western Norway, ref. number 2023/630455 evaluated and confirmed that the study ensured anonymity of the laboratory data and waived the need for informed consent.
Biomarker analyses
The serum and plasma analyses were analysed in the routine accredited laboratory at Haukeland University Hospital, Bergen, Norway, on Roche Modular E and Cobas e 602 and Roche Modular P and Cobas c 702 (Roche Diagnostics, Basel, Switzerland). Serum folate and serum cobalamin were analysed with a competitive electrochemiluminescence immunoassay. Plasma and serum tHcy samples were measured using an enzymatic assay. Serum creatinine was analysed using an enzymatic reaction cascade with photometric detection. eGFR was automatically calculated with the CKD-EPI formula (Levey and Stevens, Reference Levey and Stevens2010). Analytical coefficient of variation was 5.7% for tHcy, 10% for folate, 7% for cobalamin and 3% for creatinine. The manufacturer states that no assay interference was found at therapeutic concentrations using common drug panels for folate, cobalamin or tHcy assays (Sonntag and Scholer, Reference Sonntag and Scholer2001).
The majority of the patients admitted with mental disorders had their tHcy measured in plasma, only 5% of the samples were analysed in serum. The opposite was true for the controls, as the majority of their tHcy samples were analysed in serum and only 0.5% in plasma.
The limit for quantification of folate was 1.4–45.4 nmol/L and for cobalamin 75–1476 pmol/L. For values below or above these limits, the numeric values for the limits of detection were used.
Decision limits for folate and cobalamin deficiency
Plasma tHcy starts to increase when serum folate falls below ∼ 25–27 nmol/L, indicating suboptimal intracellular folate stores, and increases more sharply below ∼ 10 nmol/L, indicating biochemical deficiency (de Benoist, Reference De Benoist2008; Chen et al., Reference Chen, Rose, Qi, Williams, Yeung, Berry, Hao, Cannon and Crider2019; Bjørke-Monsen and Ueland, Reference Bjørke-Monsen and Ueland2023). In adults with adequate folate status (serum folate >10 nmol/L), plasma tHcy start to increase when serum cobalamin falls below ∼500 pmol/L, indicating suboptimal intracellular cobalamin stores, with a steeper increase when serum cobalamin falls below 250 pmol/L indicating biochemical deficiency (Green et al., Reference Green, Allen, Bjørke-Monsen, Brito, Guéant, Miller, Molloy, Nexo, Stabler, Toh, Ueland and Yajnik2017b; Bjørke-Monsen and Lysne, Reference Bjørke-Monsen and Lysne2023). Studies show that genomic instability in human cells is minimised when serum folate >25 nmol/L, serum cobalamin >300 pmol/L and plasma homocysteine <7.5 μmol/L in adults (Fenech, Reference Fenech2012; Fothergill et al., Reference Fothergill, Crider, Rose, Bose, Guetterman, Johnson, Jabbar, Zhang, Pfeiffer, Qi, Williams, Kuriyan, Bonam and Finkelstein2023), indicating that these figures represent a vitamin replete condition.
Based on changes in the metabolic marker tHcy, serum folate <10 nmol/L and serum cobalamin <250 pmol/L were used as decision limits for deficiency (de Benoist, Reference De Benoist2008; Bjørke-Monsen and Lysne, Reference Bjørke-Monsen and Lysne2023). Serum folate was categorised into Very low (< 5 nmol/L), Low (5–10 nmol/L), Normal (10–15 nmol/L) and High (≥ 15 nmol/L).
Statistical analyses
The results are presented as median and interquartile range (IQR; 25th and 75th percentile) and compared using nonparametric Mann–Whitney U test. Categorical data was analysed using Pearson’s Chi-Square test. P values <0.05 was considered statistically significant. The data were analysed by the software “Statistical Package for the Social Sciences” (SPSS) version 29.
Results
Patient characteristics
Patient demographics are given in Table 1. Admitted patients with mental disorders were significantly younger, with a higher proportion of men compared to the control group (Table 1). Significant differences between the two groups were seen for all tested biochemical parameters (Table 1). Median serum creatinine was slightly higher, but median eGFR was lower in the admitted mental disorders group compared to the controls, possibly due to higher percentage of men and a lower median age among the admitted patients with mental disorders compared to the controls.
Table 1. Demographics and biochemical parameters for patients admitted with mental disorders (n = 981) and patients from primary health care (Controls, n = 32,201)
1 Mann–Whitney U test.
2 Pearson’s Chi-Square test.
3 Serum folate data were available for n = 100 patients with mental disorders and n = 30,508 outpatients.
4 Serum cobalamin data were available for n = 978 patients with mental disorders and n = 30,619 outpatients.
5 Serum creatinine and estimated glomerular filtration rate (eGFR) data were available for n = 976 patients with mental disorders and n = 28,317 outpatients.
Folate status
In 2022 serum folate was requested in 10% of the patients admitted to the psychiatric acute unit. Their median serum folate level was 18% lower (Table 1) and a significantly higher percentage (43%) had folate deficiency (<10 nmol/L) compared to 27% of the controls (p < 0.001) (Table 2). Median tHcy concentration in admitted patients with mental disorders was 69% higher in the Very low folate group and 30% higher in the Low folate group compared to the controls (p < 0.001) (Table 2, Fig. 1).
Figure 1. Total homocysteine (mean and 95% confidence interval) from patients admitted with mental disorders (plasma tHcy) and patients from primary health care (serum tHcy) in relation to folate categories.
Table 2. Demographics and biochemical parameters according to serum folate categories for patients admitted with mental disorders (n = 100) and patients from primary health care (Controls, n = 30,508)
No differences were observed for serum cobalamin levels according to folate categories (p > 0.5) and apart from a significantly higher eGFR in patients admitted with mental disorders with High serum folate, there were no significant differences in renal function parameters between admitted patients with mental disorders and controls in each of the four folate categories (p > 0.08) (Table 2).
The prevalence of folate deficiency was 31% based on routine test panels performed in patients admitted with mental disorders in 2024.
Cobalamin status
Median serum cobalamin level was 11 % higher in patients admitted with mental disorders compared to the controls. Cobalamin deficiency (<250 pmol/L) was seen in 14% in the controls. The prevalence of cobalamin deficiency was 11% in 2022 and 6% in 2024 in patients admitted with mental disorders based on routine test panels.
Discussion
Median serum folate was 18% lower and median serum cobalamin was 11% higher in patients admitted with mental disorders compared to controls from primary health care. Folate deficiency was associated with higher tHcy levels among patients admitted with mental disorders compared to the controls.
The prevalence of folate deficiency was 31% and of cobalamin deficiency 6% in patients admitted to a psychiatric unit in a Norwegian hospital in 2024.
The definition of low folate and cobalamin levels differ substantially between studies (Green et al., Reference Green, Allen, Bjørke-Monsen, Brito, Guéant, Miller, Molloy, Nexo, Stabler, Toh, Ueland and Yajnik2017b; Bjørke-Monsen and Lysne, Reference Bjørke-Monsen and Lysne2023; Bjørke-Monsen and Ueland, Reference Bjørke-Monsen and Ueland2023), and this must be taken into account when prevalence patterns are discussed. Depending on the chosen cut-off, the prevalence of folate deficiency among psychiatric inpatients is reported to be 10–33%, and the prevalence of cobalamin deficiency 5–30% (Skerritt, Reference Skerritt1998; Silver, Reference Silver2000). The chosen definition in our study was below 10 nmol/L for serum folate and below 250 pmol/L for serum cobalamin deficiency and based on this, the prevalence for folate deficiency was 31% and for cobalamin deficiency 6% in patients admitted to a psychiatric unit in 2024. The prevalence for cobalamin deficiency was reduced from 11% in 2022 to 6% 2024, something which might be due to an increased awareness of the importance of adequate vitamin status, both among patients and health professionals during this period.
Low B vitamin levels have repeatedly been linked to mental disorders, particularly depression (Skerritt, Reference Skerritt1998; Lerner et al., Reference Lerner, Kanevsky, Dwolatzky, Rouach, Kamin and Miodownik2006; Clement et al., Reference Clement, Boylan, Miller, Rockwell and Allred2007; Liwinski and Lang, Reference Liwinski and Lang2023). Our patients admitted with mental disorders had lower serum folate, but higher median serum cobalamin compared to the controls. B-vitamin status is mainly a function of diet (Allen, Reference Allen2008), and patients with serious mental disorders are on the group level reported to have a poor diet characterised by a low consumption of fibre and fruit (Dipasquale et al., Reference Dipasquale, Pariante, Dazzan, Aguglia, Mcguire and Mondelli2013). An inverse association is also reported between intake of fruits and the likelihood of severe depression, anxiety and psychological distress symptoms (Shams-Rad et al., Reference Shams-Rad, Bidaki, Nadjarzadeh, Salehi-Abargouei, De Courten and Mirzaei2022). A low intake of fruits will reduce serum folate, but not serum cobalamin, a vitamin only found in animal foods.
The prevalence of cobalamin deficiency was lower (6%) among patients admitted with mental disorders, compared to 14% in requested test in controls from primary health care. One would expect requested tests to have a higher prevalence of vitamin deficiency. In addition, the prevalence of cobalamin deficiency is reported to increase with age (Green et al., Reference Green, Allen, Bjørke-Monsen, Brito, Guéant, Miller, Molloy, Nexo, Stabler, Toh, Ueland and Yajnik2017b), and the controls were older than patients admitted with mental disorders. A 6% prevalence of cobalamin deficiency was found in young women based on a serum cobalamin <220 pmol/L (Al-Musharaf et al., Reference Al-Musharaf, Mcternan, Hussain, Aleisa, Alnaami, Wani, Saravanan and Al-Daghri2020) compared to 14.5% in a population of outpatients above 65 years (Pennypacker et al., Reference Pennypacker, Allen, Kelly, Matthews, Grigsby, Kaye, Lindenbaum and Stabler1992).
Among patients with folate deficiency, the median tHcy level was 54% higher in patients admitted with mental disorders compared to controls. Several factors known to affect tHcy concentrations might explain this observation (Schneede et al., Reference Schneede, Refsum and Ueland2000).
Both folate and cobalamin deficiency impair homocysteine remethylation and increase plasma tHcy levels (Schneede et al., Reference Schneede, Refsum and Ueland2000). We did not measure vitamin B6 status, but plasma tHcy is reported to be more strongly influenced by folate and vitamin B12 than vitamin B6 status and moderate B6 deficiency is reported to cause only a mild increase in plasma tHcy concentration (Green et al., Reference Green, Allen, Bjørke-Monsen, Brito, Guéant, Miller, Molloy, Nexo, Stabler, Toh, Ueland and Yajnik2017b).
The tHcy concentration tend to increase with age and reduced renal function (Schneede et al., Reference Schneede, Refsum and Ueland2000). However, the patients admitted with mental disorders tended to be younger than the controls, and there were no significant differences in renal function parameters between the groups in each of the four folate categories.
Patients admitted with mental disorders included more men than the controls and as tHcy concentrations are reported to be higher in men than in women, this might be relevant for the observed higher tHcy concentrations in patients admitted with mental disorders. The tHcy difference between the genders is however not large, the reported geometric mean was 14.8 (SD 6.2) µmol/L in adult men compared to 13.4 (SD 4.8) µmol/L in adult women (de Bree et al., Reference De Bree, Verschuren, Blom and Kromhout2001), indicating a 10% difference between men and women.
In addition, some medications, such as valproate and lamotrigine, commonly used mood stabilisers for the treatment of bipolar disorder, can potentially interfere with folate metabolism, thereby increasing tHcy concentrations (Baek et al., Reference Baek, Bernstein and Nierenberg2013). As this study was based on anonymous data, we were unable to control for medications.
Preanalytical factors may have an impact on tHcy levels. Due to local laboratory routines, tHcy was measured in EDTA-plasma for 95% of the admitted patients with mental disorders, and in serum for 99.5% of the controls from primary health care. The concentration of tHcy in whole blood increases at room temperature, because of a continuous production and release of homocysteine from blood cells. The artificial increase is reduced if the blood sample is placed on ice (Ueland et al., Reference Ueland, Refsum, Stabler, Malinow, Andersson and Allen1993). In the hospital laboratory, the EDTA sample is placed in ice-water and plasma is separated after maximum two hours. This optimal handling of the blood sample results in lower tHcy concentrations in plasma compared to serum (Vester and Rasmussen, Reference Vester and Rasmussen1991; Ueland et al., Reference Ueland, Refsum, Stabler, Malinow, Andersson and Allen1993; Pfeiffer et al., Reference Pfeiffer, Caudill, Gunter, Bowman, Jacques, Selhub, Johnson, Miller and Sampson2000). So according to preanalytical factors, one might expect the tHcy concentration to be lower in patients admitted with mental disorders compared to controls, not the opposite as we observed.
We did only observe a difference in tHcy concentration between the two groups when serum folate concentrations were low. The MTHFR polymorphism is the most common genetic cause of increased tHcy concentrations, but it only affects tHcy levels if serum folate is low (Bagley and Selhub, Reference Bagley and Selhub1998; Liew and Gupta, Reference Liew and Gupta2015). Individuals with a homozygous MTHFR polymorphism are therefore recommended to maintain a serum folate level >15 nmol/L to improve folate metabolism (Huang et al., Reference Huang, Qin, Yang, Liu, Jiang, Zhang, Jiang, Bao, Su, Li, He, Song, Zhao, Yin, Wang, Zhang, Li, Yang, Wu, Hong, Wu, Chen, Sun, Li, Tang, Wang, Cai, Hou, Huo, Wang, Wang and Cheng2018).
In our study low folate concentrations were more common among patients admitted with mental disorders compared to controls from primary health care. Among requested blood tests in 2022, folate deficiency was found in 43% of patients admitted with mental disorders, which was significantly more compared to the 27% observed in the controls. This can be due to a poorer diet among patients admitted with mental disorders, however, the C677T MTHFR polymorphism is also per se associated with lower serum folate concentrations (Nishio et al., Reference Nishio, Goto, Kondo, Ito, Ishida, Kawai, Naito, Wakai and Hamajima2008).
Our data indicate that folate deficiency might have a greater impact on homocysteine metabolism in some patients with mental disorders, and this might be related to genetic variants, such as polymorphisms in the MTHFR gene. This hypothesis is supported by several publications reporting significant associations between the C677T polymorphisms and mental disease, like schizophrenia, major depression and bipolar disorder (Sazci et al., Reference Sazci, Ergul, Kucukali, Kara and Kaya2005; Gilbody et al., Reference Gilbody, Lewis and Lightfoot2007, Peerbooms et al., Reference Peerbooms, van Os, Drukker, Kenis, Hoogveld, de Hert, Delespaul, van Winkel and Rutten2011, Zhang et al., Reference Zhang, Yang, Gai, Cheng, Guo, Sun and Hu2022). We did however not do any genetic testing for MTHFR polymorphism among patients with mental disorders, which is a major limitation of the study.
Folate and cobalamin play crucial roles in central nervous system metabolism, and deficiency may cause neurologic and psychiatric symptoms (Bottiglieri, Reference Bottiglieri1996; Hutto, Reference Hutto1997). Optimising vitamin status by ensuring that patients have a serum folate above 15 nmol/L and a serum cobalamin above 250 pmol/L may improve psychiatric symptoms (Roffman et al., Reference Roffman, Lamberti, Achtyes, Macklin, Galendez, Raeke, Silverstein, Smoller, Hill and Goff2013; Allott et al., Reference Allott, Mcgorry, Yuen, Firth, Proffitt, Berger, Maruff, O’regan, Papas, Stephens and O’donnell2019; Lam et al., Reference Lam, Long, Saad, Lim, Doery, Griffin R.C. and Galletly2022).
Conclusion
Folate deficiency is more prevalent than cobalamin deficiency in Norwegian patients admitted with mental disorders. Among patients with folate deficiency, patients admitted with mental disorders have significantly higher tHcy concentrations compared to unselected patients from primary health care, indicating an impaired folate metabolism, which might be related to polymorphisms in the MTHFR gene. Optimising vitamin status, ensuring a serum folate above 15 nmol/L and serum cobalamin above 250 pmol/L, is recommended for patients with mental disorders.
Funding statement
The study did not receive any financial support.
Competing interests
The authors declare no conflicts of interest.
Ethical standard
The Regional Committee for Medical and Health Research Ethics, Region Western Norway, ref. number 2023/630455 evaluated and confirmed that the study ensured anonymity of the laboratory data and waived the need for informed consent.
Open access
