Identification process for eligible studies. MTHFR indicates methylenetetrahydrofolate reductase.
A, Kaplan-Meier survival curve of treatment vs no treatment in all 36 patients (P < .001). B, Kaplan-Meier survival curve of early vs no early treatment in all 36 patients (P = .17).
Comparison of patients with early vs delayed treated and psychomotor development (PMD). The sample size was 34 instead of 36 because PMD after treatment was not described in 2 patients.aFor early vs delayed treatment (excluding untreated patients), P < .001.
eFigure. Kaplan-Meier survival-curve of treatment vs no treatment in patients with deceased siblings (P?=?.002).
Diekman EF, de Koning TJ, Verhoeven-Duif NM, Rovers MM, van Hasselt PM. Survival and Psychomotor Development With Early Betaine Treatment in Patients With Severe Methylenetetrahydrofolate Reductase Deficiency. JAMA Neurol. 2014;71(2):188-194. doi:10.1001/jamaneurol.2013.4915
The impact of betaine treatment on outcome in patients with severe methylenetetrahydrofolate reductase (MTHFR) deficiency is presently unclear.
To investigate the effect of betaine treatment on development and survival in patients with severe MTHFR deficiency.
MEDLINE, EMBASE, and Cochrane databases between January 1960 and December 2012.
Studies that described patients with severe MTHFR deficiency who received betaine treatment.
Data Extraction and Synthesis
We identified 15 case reports and case series, totaling 36 patients. Data included the following: (1) families with 2 or more patients with severe MTHFR deficiency, of whom at least 1 received betaine, or (2) single patients with severe MTHFR deficiency treated with betaine. To define severe MTHFR deficiency, methionine, homocysteine, MTHFR enzyme activity in fibroblasts, or mutations (in the MTHFR gene) had to be described as well as the effect of treatment (survival and/or psychomotor development). We compared the outcome in treated vs untreated patients and early- vs late-treated patients. Sensitivity analysis was performed to address definition of early treatment. To further assess the impact of treatment on mortality, we performed a subanalysis in families with at least 1 untreated deceased patient.
Main Outcomes and Measures
Survival and psychomotor development.
Eleven of 36 patients (31%) died. All deaths occurred in patients who did not receive treatment or in patients in whom treatment was delayed. In contrast, all 5 early-treated patients survived. Subgroup analysis of patients with deceased siblings—their genotypically identical controls—revealed that betaine treatment prevented mortality (P = .002). In addition, psychomotor development in surviving patients treated with betaine was normal in all 5 early-treated patients but in none of the 19 surviving patients with delayed treatment (P < .001).
Conclusions and Relevance
Early betaine treatment prevents mortality and allows normal psychomotor development in patients with severe MTHFR deficiency, highlighting the importance of timely recognition through newborn screening.
Deficiency of (5,10)-methylenetetrahydrofolate reductase (MTHFR) (OMIM 236250) is a frequent remethylation disorder. The clinical consequences vary greatly depending on the degree of enzyme deficiency. Carriers of the prevalent mild thermolabile variant are at risk for moderate hyperhomocysteinemia, a putative risk factor for cardiovascular disease, but do not have neurological symptoms.1 Severe MTHFR deficiency is rare. Presentation is in infancy or early childhood with failure to thrive, hypotonia, and progression into lethargy, seizures, and coma.2 The inability to remethylate homocysteine to methionine induces profoundly elevated homocysteine and decreased methionine concentrations in blood. Together, these biochemical abnormalities are held responsible for the ensuing clinical picture. Untreated, patients either die or—if they survive—exhibit severe developmental delay and a life-long dependence on care.2- 7
Betaine serves as a methyl donor for betaine homocysteine S-methyltransferase, an enzyme also capable of recycling homocysteine to methionine. Treatment with betaine may rescue some of the consequences of severe MTHFR deficiency. Indeed, betaine treatment has been shown to normalize methionine levels and decrease—but not normalize—homocysteine levels. Case reports indicate that betaine treatment may prevent further cognitive decline in surviving symptomatic patients and may even prevent cognitive damage and death if treatment is started early.8- 10 Similarly positive effects were reported in an animal model for severe MTHFR deficiency.11 In this study, MTHFR−/− mice were treated with betaine prenatally and 3 weeks after birth. On betaine treatment, 76% of mice survived compared with 17% in untreated MTHFR−/− mice. In addition, betaine treatment ameliorated disturbances in proliferation and differentiation in the cerebellum and hippocampus.
To study the effect of betaine treatment on development and survival in patients with severe MTHFR deficiency, we performed a systematic review and included the retrieved clinical cases in a meta-analysis of case reports.
We searched the MEDLINE, EMBASE, and Cochrane databases for studies published between January 1960 and December 2012 with the search terms “methylene-tetra-hydrofolate-reductase” or “MTHFR” or “methylenetetrahydrofolate” or “MTHF” or “methyltetrahydrofolate” and “deficient” or “deficiency,” together with “betain” or “betaine.” Additionally, we manually checked the reference list of included articles for further suitable studies. Animal studies were excluded (Figure 1).
We included all studies that described the following: (1) single patients with severe MTHFR deficiency treated with betaine, or (2) families with 2 or more patients with severe MTHFR deficiency, of whom at least 1 received betaine. Severe MTHFR deficiency was considered proven if at least 1 of the following criteria were present: low methionine concentration (<0.22 mg/dL; to convert to micromoles per liter, multiply by 67.02) and high total homocysteine concentration (>50 μmol/L; to convert to milligrams per liter, divide by 7.397); MTHFR enzyme activity in fibroblasts below 10%; 2 mutations (in the MTHFR gene); or a positive nitroprusside test.
Relevant clinical characteristics (treatment onset, follow-up time, sex, treatment dosage, other treatments, and psychomotor development) of all included patients were retrieved from the original case descriptions.
We first compared the outcome of treated and untreated patients. Outcome parameters were mortality and psychomotor development. Next, we assessed the effect of the timing of treatment on outcome. For this purpose, early treatment was defined as treatment initiated at an age before any untreated patient had died (<16 days). All other patients were grouped as having delayed treatment or no early treatment (delayed plus no treatment). To further assess the impact of treatment on mortality, we performed a subanalysis in families with at least 1 untreated deceased patient.
To assess whether betaine treatment is beneficial for late-treated patients, we reviewed the case descriptions. A beneficial effect was defined as improvement or stabilization of psychomotor development as described by the author(s) of the relevant article.
Kaplan-Meier curves were used to calculate survival of untreated vs treated and early-treated vs not-early-treated groups, ie, those who dropped out prematurely were included for as long as they participated (Figure 2). Statistical significance was calculated using Fisher exact tests. In the subanalyses, Fisher exact test was used to test the efficacy of early treatment vs delayed treatment on the following: (1) mortality in the families with at least 1 untreated deceased patient, and (2) psychomotor development (Figure 3). Prism software version 5.0 (GraphPad Software) was used for statistical analysis.
The systematic review revealed 15 case reports and case series, including 36 patients, that were eligible for inclusion (Table 1 and Table 2). Twenty-six of the 36 patients were treated, of whom 5 were treated early. The median age at treatment initiation was 13 months (range, 0-23 years). Eleven of the 36 patients (31%) died; the median age at death was 42 days (range, 16 days to 2.5 years). In 8 of the 11 deceased patients, hypotonia was followed by recurrent apnea and ultimately respiratory failure with coma and death.
All treatment regimens consisted of betaine at a dosage of 100 mg/kg/d or more. Betaine treatment was clearly associated with increased survival: whereas 9 of 10 untreated patients died, only 2 of 26 died after initiation of treatment (P < .001) (Figure 2A). Of note, 1 of 2 patients who died despite introduction of betaine exhibited neurological deterioration and respiratory distress at the time treatment was initiated. None of the 5 early-treated patients died as compared with 11 of the 31 patients (33%) in the no-early-treatment group (P = .17) (Figure 2B).
In 6 families with 1 or more deceased siblings—indicating a severe genotype—9 of 9 patients died without treatment vs 0 of 6 patients with treatment (P = .002) (eFigure in Supplement).
Figure 3 shows the effect of betaine treatment on psychomotor development of all patients. Psychomotor development was described in 34 of 36 patients (5 treated early, 19 with delayed treatment, 10 untreated), with a median follow-up time of 2 years. Early treatment was associated with normal psychomotor development in all 5 patients. In contrast, none of the 19 patients in the delayed-treatment group (untreated patients excluded) had normal psychomotor development (P < .001) (Figure 3). Nevertheless, stabilized or improved psychomotor development was reported in all 17 surviving patients in whom betaine treatment was initiated later (Table 2).
We evaluated more lenient definitions of early treatment. Treatment initiation before ages 30 and 90 days resulted in 3 deaths in 8 patients and 7 deaths in 11 patients, respectively.
In this study, we evaluated the effects of betaine treatment on survival and psychomotor development in patients with severe MTHFR deficiency by performing a meta-analysis of case reports. The results clearly indicate that betaine treatment improves survival, prevents further cerebral damage, and allows normal development if treatment is initiated early.
To our knowledge, this is the first combination of case reports to determine the beneficial effects of early betaine treatment in this group of patients. Evaluating outcome and treatment modalities is always cumbersome in rare diseases. A systematic review of the literature and a meta-analysis of all the data available provide important insight even in disorders with a very low incidence. It is likely that our approach is one of the few methods available. Nevertheless, some potential limitations of this approach should be discussed.
First, pooling data from case studies to get insight into treatment efficacy is considered hazardous.12 This is based on the assumption that no standardization has been applied: the study population is inhomogeneous, controls are absent, the trial is often not blinded or placebo controlled, and treatment dosage can vary. However, in case of inborn errors, most of these hazards can be addressed: (1) clear inclusion criteria can be set that allow for a homogeneous population; (2) appropriate (genotypically identical) controls are available, namely siblings of index patients as often described in case reports; (3) the genetic defect determines to a great extent the natural course, so blinding of treatment would not significantly influence perceived outcome; and (4) although dosages were different, all treated patients received betaine at a dosage greater than 100 mg/kg/d, suggesting that this dosage is adequate. Second, selection bias (particularly publication bias) may have affected the exact composition of the study population, as some cases may have remained unreported. This may influence the effect size but is unlikely to change its direction. Theoretically increased survival may be due to other factors, eg, parents with a previous affected sibling may somehow learn how to handle the disease. Arguing against this is the observation that in some families several children died and only the betaine-treated patient did well. Third, the observed effect could be attributed to other supplements, including folic acid or 5-MTHF, as previously suggested.6 However, betaine was the only compound given to all patients. In one study, survival was attributed to early treatment with a combination of folic acid, vitamin B6, vitamin B12, and carnitine, but this treatment was not associated with normal psychomotor development.6,13 Addition of methionine to treatment with betaine has been described only once. This rapidly deteriorating patient (with 1 deceased sibling) received methionine treatment soon after betaine and vitamin B treatment because the patient did not respond quickly enough to betaine treatment.14 With methionine, the clinical condition of the patient improved remarkably. Methionine may therefore be of value as an add-on treatment in severely symptomatic patients. Fourth, assessment of psychomotor development was based on the limited information provided in the case descriptions.
As ultra-rare disorders are usually based on private mutations and each mutation differs with respect to its residual enzyme activity and disease course, the patient should ideally serve as his or her own control.15 An affected sibling (both sharing the same genotype and the overall genetic and socioeconomic context) approximates this ideal. Combining data from case reports is possible if patients can be stratified according to the expected phenotypic severity. This is possible in case of a strong genotype-phenotype correlation, as in severe MTHFR deficiency.
The beneficial effect of early betaine treatment evidenced in this meta-analysis supports the addition of MTHFR deficiency to newborn screening programs. Only newborn screening has the potential to allow timely recognition and treatment. Methods for high-throughput detection of homocysteine—pivotal for timely detection of this disorder—have recently become available.16 Deficiency of MTHFR thus meets the Wilson and Jungner criteria17 for (newborn) screening. Other treatable related disorders (such as cystathionine-β-synthase deficiency) will benefit from homocysteine screening as well.16
In summary, our systematic review and meta-analysis delineates the efficacy of early betaine treatment in reducing mortality and in allowing normal development in patients with severe MTHFR deficiency. Given the narrow window of opportunity, early recognition through newborn screening is essential.
Early betaine treatment prevents mortality and allows normal psychomotor development in patients with severe MTHFR deficiency. This highlights the importance of timely recognition through newborn screening.
Corresponding Author: Peter M. van Hasselt, MD, PhD, Wilhelmina Children’s Hospital, University Medical Centre Utrecht, Lundlaan 6, KC 03.063.0, 3584 EA, Utrecht, the Netherlands (firstname.lastname@example.org).
Accepted for Publication: August 27, 2013.
Published Online: December 9, 2013. doi:10.1001/jamaneurol.2013.4915.
Author Contributions: Dr van Hasselt had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Diekman, van Hasselt.
Acquisition of data: Diekman, de Koning.
Analysis and interpretation of data: All authors.
Drafting of the manuscript: Diekman, van Hasselt.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Diekman, Rovers, van Hasselt.
Study supervision: van Hasselt.
Conflict of Interest Disclosures: None reported.