SEARCH

SEARCH BY CITATION

Keywords:

  • anticonvulsive agents;
  • adverse effects;
  • heredity;
  • weight gain

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Objective: To evaluate whether genetic factors contribute to weight gain associated with valproate (VPA) therapy.

Research Methods and Procedures: We retrospectively and prospectively evaluated five pairs of monozygotic twins concordant for epilepsy and treated with VPA regarding weight course.

Results: In all twin pairs, both twins showed similar weight courses under therapy with VPA.

Discussion: These results suggest that genetic factors may have an influence on the weight change induced by VPA. Further studies are necessary to obtain heritability estimates regarding this effect of VPA therapy and to identify the relevant genes.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Valproate (VPA)1 is a commonly used anticonvulsive drug. It is known to induce weight gain in ∼50% of patients (1, 2, 3, 4, 5, 6, 7). This weight gain can predispose the patient to other health problems and can lead to non-compliance or discontinuation of anticonvulsive therapy (4, 5, 8, 9, 10, 11, 12, 13).

The exact mechanism causing the weight gain is not known. However, an association between VPA-related obesity and hyperinsulinemia has been shown (2, 11, 14, 15, 16, 17). As in other types of obesity, serum leptin concentration increases proportionally with weight gain (10, 14).

At present, no reliable risk factors for VPA-related weight gain are known. However, for clozapine, which can also induce weight gain, an influence of genetic factors on weight gain is suspected (18, 19). The extent to which genetic factors determine VPA-related weight gain is unknown.

This study was performed to determine whether genetic factors contribute to weight gain associated with VPA therapy.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

To identify twins concordant for VPA therapy, we contacted each of the 5600 family physicians, neurologists, and neurological hospital departments in Germany that were listed in the address database of the Department of Neurology at Philipps-University Marburg. We asked them to report all known monozygotic (MZ) and dizygotic twins concordant for VPA therapy. Six pairs of MZ twins were identified; one pair refused participation. No dizygotic twins concordant for VPA therapy were reported, which may be related to the lower concordance rate of epilepsy in dizygotic twins.

All twins were evaluated regarding their weight changes during VPA therapy, history of epilepsy, current and former anticonvulsive medication, and other factors possibly contributing to weight change, such as intake of other weight-influencing drugs, eating habits, physical activity, and gastrointestinal and metabolic disorders, including thyroid disease. In most cases, weight data were collected retrospectively from children examination books, reports of physicians, or anamnestic specifications. In some patients, current data were collected prospectively. Because in children and adolescents, age- and sex-specific developmental changes in BMI occur, SD scores (SDSs) for BMI based on a German reference population (20) were used in addition to BMI for description of the weight course under the age of 16 years. Zygosity was determined using polymerase chain reaction—based variable number of tandem repeats analysis (Promega Powerplex 16 kit; Promega Corporation, Madison, WI). Twin genotype concordance in the 15 autosomal systems tested was regarded as proof of MZ.

All probands (if they were minors, the parents also) gave written informed consent. The study was approved by the Ethics Committee of the Philipps University of Marburg, Germany.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Pair 1

Pair 1 are male MZ twins (born in 1990) who both developed febrile seizures at the age of 2.5 years and unprovoked absence seizures 6 months later. VPA (mean dose 600 mg/d) was started at this age in both twins. The weight of twin 1A increased from 11.8 kg at 90 cm to 17.0 kg at 107 cm in a period of 2 years and to 35 kg at 147 cm over 9 years of therapy with VPA. Twin 1B increased his weight from 11.8 kg at 89.5 cm to 17.0 kg at 107 cm over 2 years and to 33.8 kg at 145 cm in a period of 9 years of therapy with VPA (Table 1; Figure 1A). BMI SDSs similarly increased slightly during this period in both twins (twin 1A: −1.33 to −0.51; twin 1B: −1.17 to −0.63).

Table 1. . Weight change, mean VPA dose, initial and final weight/height, age at start, and duration of therapy in the five pairs of monozygotic twins
TwinAge at start of VPA (years)Mean VPA dose (mg/d)Duration of therapy (years)Initial weight/height (kg/cm)Final weight/height (kg/cm)Weight change under VPA (kg)BMI/SDS change under VPA
  1. In twin 2B, the short period of weight loss and recovery was not considered.

1A2.5600911.8/9035.0/147[UPWARDS ARROW] 23.2[UPWARDS ARROW] 0.82 SDS
 B2.5600911.8/9033.8/145[UPWARDS ARROW] 22.0[UPWARDS ARROW] 0.54 SDS
2A17900468/17390/173[UPWARDS ARROW] 22.0[UPWARDS ARROW] 7.4 BMI
 B17900493/180116/180[UPWARDS ARROW] 23.0[UPWARDS ARROW] 7.1 BMI
3A310802.511.1/8218.3/108[UPWARDS ARROW] 7.2[DOWNWARDS ARROW] 0.14 SDS
 B310802.511.8/8720.3/113[UPWARDS ARROW] 8.5[UPWARDS ARROW] 0.71 SDS
4A34800467/15867/158= 0.0= 0.0 BMI
 B3713000.582/16082/160= 0.0= 0.0 BMI
5A1810000.7565/17572/175[UPWARDS ARROW] 7.0[UPWARDS ARROW] 2.3 BMI
 B1810000.7569/17572/175[UPWARDS ARROW] 3.0[UPWARDS ARROW] 1.0 BMI
image

Figure 1. BMI course and VPA mean dosage in the five pairs of monozygotic twins (a—e). In twin 2B, the short period of weight loss and recovery was not considered.

Download figure to PowerPoint

Pair 2

The birth weights of the male MZ twins of pair 2 (born in 1983) differed markedly (twin 2A: 1500 grams; twin 2B: 2300 grams). Twin 2A stayed 6 weeks in an incubator. However, in contrast to pair 3, they showed no weight convergence. Twin 2B remained heavier and larger until the age of 16 years (twin 2A: 68 kg/173 cm; twin 2B: 93 kg/ 180 cm). We could not identify any plausible cause for this weight difference. Eating habits and physical activity were similar in both twins. They received no weight-influencing drugs until the age of 16 years. No gastrointestinal disorders could be identified. Nevertheless, this weight difference is likely caused by environmental factors, including possibly a feto-fetal transfusion syndrome. At the age of 17 years, twin 2B showed a short period of weight loss caused by an episode of depression (therapy: olanzapine for 4 weeks) followed by a fast recovery of weight. As both twins developed epileptic seizures, therapy with VPA (mean dose, 900 mg/d) was started at the age of 17 years. With VPA, twin 2A gained 22 kg in 4 years. Because VPA was started in twin 2B during his period of weight loss, only a part of the resulting weight gain can be attributed to VPA. Therefore, we used the predepression weight to calculate the weight gain for comparison with twin 2A. In this period, twin 2B gained 23 kg (Table 1; Figure 1B).

Pair 3

The male MZ twins of pair 3 (born in 1995) had markedly different birth weights and lengths (twin 3A: 1300 g/36 cm; twin 3B: 1800 g/46 cm). At the age of 2 years, height and weight had somewhat converged (twin 3A: 11.1 kg/82 cm; twin 3B: 11.8 kg/87 cm). Both were diagnosed as suffering from myoclonic astatic epilepsy shortly before their third birthday, and therapy with VPA (mean dose, 1080 mg/d) was started. Initially, twin 3A was also treated with carbamazepine for a few weeks. One and one-half years after starting VPA, the BMI of twin 3A (16.8 kg/102 cm) decreased somewhat, and the BMI of twin 3B (17.4 kg/ 105 cm) increased slightly (Table 1; Figure 1C). There was a slight increase of BMI SDSs in both twins (twin 3A: 0.32 to 0.49; twin 3B: −0.40 to 0.25). One year later, twin 3A (18.3 kg/108 cm) showed a small decrease of BMI and BMI SDS (0.18), whereas a minor increase of BMI and BMI SDS (0.31) was seen in twin 3B (20.3 kg/113 cm).

Pair 4

Pair 4 are female MZ twins (born in 1965). Twin 4A was initially diagnosed with idiopathic generalized epilepsy (IGE) at the age of 34 years. Therapy with VPA (mean dose, 800 mg/d) was started. At this age, she participated in a “weight watchers course.” During the course of 4 years, her weight remained virtually unchanged under therapy with VPA (67 kg/158 cm; Table 1; Figure 1D). Twin 4B was slightly less heavy than her sister at the age of 34 years. However, she gained weight up to 82 kg (160 cm) at the age of 37 years. She attributed this weight gain to less restricted eating habits compared with her sister. In addition, latent hypothyroidism has been diagnosed earlier in twin 4B (normal triiodothyronine/thyroxine, increased thyroid-stimulating hormone). She was initially diagnosed with IGE at the age of 37 years, and VPA (mean dose, 1300 mg/d) was started. After 6 months of follow-up, no change in her weight was observed.

Pair 5

Both female MZ twins of pair 5 (born in 1985) were initially diagnosed to have IGE with generalized tonic-clonic seizures precipitated by intermittent light stimulation at the age of 18 years. At this time, twin 5A weighed 65 kg and twin 5B weighed 69 kg (both 175 cm). Both twins received VPA (mean dose, 1000 mg/d). After 4 months of therapy, both had a weight of 71 kg and attributed this weight gain to an increased appetite. However, because both were studying for their graduation, they were physically less active. After 9 months using VPA, the weight of both increased to 72 kg (Table 1; Figure 1E).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

To our knowledge, this is the first report regarding the similarity of weight change associated with VPA therapy in MZ twins. Except for pair 3, the weight course within each twin pair was similar during therapy with VPA. The twins of pair 3 with markedly different birth weight underwent small converging changes of body weight in opposite directions during VPA therapy. MZ twins share one placenta and, therefore, frequently have different birth weights. The body weight typically becomes more concordant over time. This convergence effect is well documented and provides evidence for the genetic determination of body weight (21, 22, 23, 24, 25). In the twins of pair 3, VPA was started at the age of 3 years when the convergence was in progress but not completed. Subsequently, the lower BMI in twin 3B slightly increased, whereas the higher BMI in twin 3A slightly decreased during VPA therapy. Thus, the convergence effect but not a different response to VPA likely underlies the minor difference in the weight courses of these twins.

In all twin pairs, the effect of VPA on body weight was similar. Two of the three pairs who gained weight showed a comparable amount of weight gain (pairs 1 and 2). The other pair reached the same final weight, although VPA was started at different initial weights (pair 5). The increase was somewhat higher in the twin with the lower starting weight. Similarly, it has been reported in one study that weight gain under VPA is higher in patients with lower starting weight (12).

As the phenotype “VPA-induced weight gain” is not categorical but continuous, environmental influences on the amount of weight gain during VPA therapy likely occur (26). For that reason, slight differences in the amount of weight gain presumably can be attributed to non-shared environment. Nevertheless, all five pairs of twins showed very similar effects of VPA on body weight. It thus seems likely that genetic factors have an influence on the weight course during therapy with VPA.

The analysis was complicated by different initial weights. These differences are remarkable because BMI is strongly influenced by genetic factors. Numerous and large-scaled twin studies have repeatedly estimated heritability of BMI at ∼0.7 (27). We cannot exclude the possibility that, in twins concordant for epilepsy, BMI is less heritable.

In summary, the similarity of changes in body weight during VPA therapy observed in these five monozygotic twin pairs suggests that genetic factors may have an influence. Larger twin studies are warranted to obtain heritability estimates regarding this effect of VPA therapy.

Acknowledgement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

This study was supported by the Bundesministerium für Bildung und Forschung (National Genome Research Net 01GS0118) and the Ulran-Professorship for Neurology/ Epileptology.

Footnotes
  • 1

    Nonstandard abbreviations: VPA, valproate; SDS, SD score; MZ, monozygotic; IGE, idiopathic generalized epilepsy.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References