- Email: [email protected]
Chronology and evolution of temporal lobe epilepsy and endocrine reproductive dysfunction in women: Relationships to side of focus and catameniality
Epilepsy & Behavior 11 (2007) 185–191 www.elsevier.com/locate/yebeh
Chronology and evolution of temporal lobe epilepsy and endocrine reproductive dysfunction in women: Relationships to side of focus and catameniality Vladimir V. Kalinin *, Elena V. Zheleznova Department of Brain Organic Disorders and Epilepsy, Moscow Research Institute of Psychiatry, Ministry of Health and Social Development, 3, Poteshnaya Street, 107076 Moscow, Russian Federation Received 19 September 2006; revised 7 April 2007; accepted 13 April 2007 Available online 18 June 2007
Abstract The current study was carried out to determine the possible associations between side of focus, catameniality, reproductive dysfunction (RD), and chronology variables for epilepsy and concomitant RD in women. Eighty women of childbearing potential with temporal lobe epilepsy were included in the study. Catamenial epilepsy was observed mainly in women with left-sided foci, and a noncatamenial pattern in women with right-sided foci. Left-sided foci were associated with polycystic ovary syndrome, and right-sided foci with hypogonadotropic hypogonadism. Catamenial epilepsy with right-sided foci was associated with longer duration of epilepsy (P = 0.021), trend toward earlier age at onset of epilepsy, and trend toward longer interval between onset of epilepsy and onset of RD compared with catamenial epilepsy with left-sided foci. On the other hand, noncatamenial epilepsy with right-sided foci was characterized by a shorter interval between onset of epilepsy and onset of RD in comparison with noncatamenial epilepsy with left-sided activity (P = 0.03). In addition, comparison of patients with right-sided foci with catamenial and noncatamenial patterns of seizures revealed earlier age of epilepsy onset (P = 0.049), longer duration of epilepsy (P = 0.017) and of RD (P = 0.036), and longer interval between onset of epilepsy and onset of RD (P = 0.048) in patients with catamenial epilepsy. From an evolutionary point of view, catamenial epilepsy with right-sided focal activity seems to be the oldest subtype. Ó 2007 Elsevier Inc. All rights reserved. Keywords: Female epilepsy; Catameniality; Reproductive dysfunction; Laterality of foci; Age at epilepsy onset
1. Introduction Endocrine and psychiatric disorders are thought to be the most frequent comorbid pathology among women with epilepsy. These disorders continue to draw the attention of many research groups, because they remain utterly crucial for women in relation to childbearing and antiepileptic drug (AED) intake, which, in turn, have influence on endocrine and gynecological systems. Data on the effect of valproate (VPA) on reproductive function in women with epilepsy are ambiguous and rather controversial. Based *
Corresponding author. E-mail addresses: [email protected], [email protected] (V.V. Kalinin). 1525-5050/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.yebeh.2007.04.014
on the work of Isoja¨rvi et al. [1–3], the relationship between VPA use and polycystic ovary syndrome (PCOS) and obesity in women with epilepsy is regarded as well established, although not all research groups could confirm their findings, and the findings of Murialdo et al. [4] are somewhat contradictory. Murialdo and co-workers reported that PCOS does not occur more frequently under VPA treatment than under treatment with other AEDs. Moreover, they observed PCOS in 12% of women on phenobarbital, in 21% on carbamazepine, and in no women on VPA monotherapy [4], although in a subsequent controlled study [5], these authors reported increased body mass index, higher androgen levels, and higher frequency of ovulatory dysfunction in 63.5% of women treated with VPA in comparison to women treated with CBZ or phenobarbital
186
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
or to controls. Thus, the association between VPA administration and high androgen level, ovulatory dysfunction, and weight gain was again confirmed. The aforementioned discrepancy in data may be the result not only of the AEDs themselves, but of differences in the types of epilepsy, seizure semiology, and other neurobiological factors, such as side of foci. Herzog [6], Friedman and Herzog [7], and Herzog et al. [8] have proved the role of focus lateralization in the genesis of certain gynecological disorders, and PCOS has been observed in association with left-sided foci, whereas hypothalamic hypogonadism (HH) prevails in women with temporal lobe epilepsy with right-sided foci. The frequency of left-sided foci is believed to be, on average, threefold higher than the prevalence of right-sided foci [9,10], suggesting that PCOS should be more prevalent than HH, if there exists a strict relationship between focus lateralization and type of gynecological pathology. Our previous study established the existence of certain relationships between endocrine and gynecological pathology, catameniality, and psychopathological syndromes in women with epilepsy. Thus, a reciprocal interrelationship between thyroid and reproductive pathology was revealed, and catameniality was associated much more with reproductive dysfunction than with thyroid pathology in women with epilepsy [11]. Much less is known about temporal sequences of comorbid endocrine and gynecological pathology in relation to focus lateralization and catameniality in localizationrelated epilepsies, although this might shed light on the evolution of epilepsy with respect not only to ontogenesis, but to phylogeny too. According to the widely known biogenetic rule of Haeckel [12], the temporal sequence of acquired characteristics in ontogeny is quite similar to that in phylogeny in a condensed manner. In other words, the signs, symptoms, disorders, and pathology that occur during an individual’s life represent the time of their appearance in the phylogeny of the species. On the basis of this rule, it may be predicted that early occurrence of signs, symptoms, and disorders in individual Homo sapiens reflect their early origin in evolution and, conversely, their occurrence late in life indicates they are novel evolutionarily. It has also been suggested that hemispheric lateralization of different functions indicates the time of their occurrence in phylogeny [13]. Thus, the right hemisphere is responsible for Gestalt analysis, which is regarded as ancient brain function. Conversely, the left hemisphere determines analytical and verbal processing and is responsible for speech function. The latter brain functions are thought to be the prerogative strictly of Homo sapiens and were acquired in the later stages of human evolution [13–15], although this hypothesis seems somewhat speculative and should be confirmed in special studies. Nevertheless, taking all these data together, we suggest that any characteristic or symptom of neuropsychiatric disorders associated with the left hemisphere should be regarded as having occurred later in evolution
than those linked with the right hemisphere [13,14]. Thus, PCOS, because it is associated with left-sided foci in epilepsy, should be considered a relatively new disorder in terms of evolutionary origin than HH, which is associated with right-sided foci [6–8]. This assumption, however, remains strictly speculative. The principal objective of the present study was to determine the relationship of chronological variables (age at epilepsy onset and concomitant endocrine reproductive disorders) to side of foci and catameniality in women with epilepsy. In addition, the possible relationship between side of foci, on the one hand, and catameniality and noncatameniality, on the other hand, was studied. 2. Materials and methods The study was carried out on 80 women of childbearing age with temporal lobe epilepsy (TLE). Women with a definite diagnosis of TLE were intentionally selected. These women were recruited from all patients with epilepsy admitted for treatment at the Department of Brain Organic Disorders and Epilepsy, Moscow Research Institute of Psychiatry. Patients diagnosed with idiopathic generalized epilepsies and localization-related epilepsy of nontemporal origin were excluded from the study. Among all the women studied, left-sided foci were detected in 37 patients, and right-sided foci in 43 patients. Side of focus was detected strictly by looking at the EEG; the data on ictal semiology were not used for this purpose. MRI scans were performed on all patients prior to the study. Based on the results of the MRI study, the symptomatic form of TLE was diagnosed in 17 women, and the cryptogenic form in 63 women. Among the women with symptomatic TLE, cysts of traumatic origin were detected in 12 patients and mediotemporal sclerosis in 5 patients. Patients ranged from 16 to 45 years in age (mean ± SD: 26.13 ± 7.12 years). Thirteen patients had the so-called catamenial form of epilepsy, and 67 had the noncatamenial form. Here, catamenial epilepsy is understood to be the occurrence of seizures strictly in relation to certain days of the menstrual cycle. According to the classification of catamenial epilepsy proposed by Herzog [16], most of the women (12 patients, 92%) had the so-called perimenstrual pattern of catamenial epilepsy, that is, with seizures in interval from Day 3 to Day +3 of the menstrual cycle. All women were properly assessed by a gynecologist and an endocrinologist before inclusion in the current study. In addition, the ovaries of each woman were scanned by ultrasound to make precise diagnoses. Seizure semiology data are summarized in Table 1, and include mean seizure frequency for each type of seizure per month with respect to catameniality and subtype of RD. Data covered the time to inclusion in the study and through treatment with AEDs. Analysis of seizure dynamics during AED treatment was not considered the principal aim of the current study. Nevertheless, there were no statistically significant discrepancies in frequency of seizures between women with catamenial and noncatamenial seizures and between women with different subtypes of RD. Among the women studied, the diagnosis of refractory epilepsy was confirmed in 11 (13.8%), although epilepsy intractability was not considered the principal criterion for inclusion in or exclusion from the study. Here intractable epilepsy is defined as seizures that do not respond to any of three first-line AEDs usually administered for the given epilepsy syndromes [17]. The duration of follow-up under AED treatment in the current study varied, but usually was not less than 3 months. Mean seizure frequency before and during treatment was analyzed, and if seizure reduction was less than 50%, the epilepsy was regarded as intractable [17]. Obviously, seizure frequency was higher in patients with treatment-resistant seizures than in patients with a favorable response to AED therapy. The proportions of women with refractory epilepsy in the catamenial and noncatamenial groups were similar: 15 and 13% (2 and 9 patients, respectively).
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
187
Table 1 Mean seizure frequency per month relative to catameniality/noncatameniality and reproductive dysfunction Type of seizures
Catamenial epilepsy
Noncatamenial epilepsy
PCOSa
HH
Normal reproductive function
Simple partial Complex partial Secondarily generalized All seizures
66.5 ± 197.8 0.2 ± 0.8 8.3 ± 15.1 75.0 ± 199.0
34.2 ± 92.4 5.1 ± 22.6 7.7 ± 23.0 47.1 ± 105.4
39.5 ± 102 5.0 ± 9.2 19.3 ± 51.0 63.8 ± 151.0
46.4 ± 137.1 1.2 ± 4.3 6.6 ± 12.7 54.1 ± 137.2
29.1 ± 98.6 8.2 ± 33.0 5.2 ± 7.9 42.5 ± 104.9
a
HH, hypothalamic hypogonadism; PCOS, polycystic ovary syndrome.
All patients continued to take their AEDs. The AEDs being taken by the women prior to inclusion in the study are listed in Table 2. As can be seen, four first-line AEDs were usually used. Serum concentrations were not obtained for all patients, and so these data are not part of the current study. Nevertheless, it should be pointed out that in cases in which these levels were available, mean serum concentrations of all drugs were within the therapeutic windows. In line with the principal aim of this study, age at onset of epilepsy and concomitant endocrine reproductive pathology were determined, and the interval between onset of epilepsy and onset of endocrine RD (so-called delta disease) was calculated with the formula: age at reproductive disorder onset age at epilepsy onset. The study design was strictly retrospective, and specific information from the women’s diaries was selected and analyzed for this purpose (occurrence of seizures during the previous 2 years, and their possible relationship to the menstrual cycle). Semiology and frequency of seizures were also taken into account for further analysis. Because measurement of sex hormone levels was not part of the main objective of the current study, these levels were not assessed in all patients. Nevertheless, such data were obtained for 24 patients (9 with catamenial and 15 with noncatamenial seizures). Venous blood samples for assessment of serum levels of luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol, progesterone, testosterone, and prolactin were obtained on the seventh day of the menstrual cycle. Hypothalamic amenorrhea (i.e., HH), as defined by Herzog et al. [19], was diagnosed in 34 women. This diagnosis usually was apparent in women with menstrual dysfunction in the absence of any cystic alterations in ovaries. In addition, patients with HH were also characterized by lack of hirsutism, acne, and weight gain. Results of hormone level assessments were obtained for only 12 women with HH and, when compared with reference values [18], revealed low serum LH (1.25 ± 1.75 mIU/mL; normal range, 3–20 mIU/mL), low serum FSH (3.16 ± 2.38 mIU/ml; normal range, 6–23 mIU/mL); near-normal serum estradiol (242.37 ± 340.74 pg/ mL; normal range, 10–200 pg/mL); elevated serum progesterone (9.48 ± 17.64 ng/mL; normal range, 0.1–1.5 ng/mL), and slightly elevated serum free testosterone (2.61 ± 0.76 ng/dL; normal range, 0.3–1.9 ng/dL). On the contrary, PCOS was diagnosed in 12 women. Ultrasound findings constituted the main criterion for this diagnosis, and in most of these women, no fewer than 10 cysts were detected. Additional clinical criteria for PCOS used in the current study included: presence of ovulatory dysfunction, such as polymenorrhea, oligomenorrhea, or amenorrhea; and clinical signs of hyperandrogenism, such as hirsutism [16,19,20]. Laboratory findings in these patients revealed normal LH (5.64 ± 1.69 mIU/
mL), normal FSH (5.15 ± 2.38 mIU/mL), low estradiol (0.45 ± 0.49 pg/ mL), normal progesterone (1.2 ± 0.98 ng/mL), and elevated free testosterone (5.75 ± 2.19 ng/dL) levels. In the remaining 34 women, no reproductive disorder was diagnosed, and all had normal menstrual cycles. It should be stressed that results of endocrine tests were not available for all women to confirm the diagnosis of PCOS or HH. Therefore, the ultrasound scan coupled with clinical data was the principal method for diagnosis of RD.
2.1. Statistics All data were processed with the Statistica program (sixth version) on a personal computer. v2 statistics and Fisher’s exact test were used for the analysis of possible association between so-called nominal variables (i.e., type of endocrine pathology, catameniality, and side of focus). On the other hand, Student’s t test was used to analyze quantitative variables (onset of epilepsy, onset of RD, and the interval between).
3. Results The main results are summarized in Tables 3–7. Table 3 outlines the relationship between catameniality and side of focus in the women studied. As can be seen, catamenial seizures were observed more frequently in women with leftsided foci, whereas the noncatamenial pattern was observed in patients with right-sided foci. Fisher’s exact test (P = 0.016) and v2 statistics (v2 = 4.49, P = 0.034) detected statistically significant discrepancies between the Table 3 Relationship between side of focus and catameniality in women with epilepsy Type of epilepsy
Left-sided foci
Right-sided foci
Total
Catamenial Noncatamenial Total
10 27 37
3 40 43
13 67 80
v2 = 4.49, P = 0.034 with Yates correction; Fisher exact test: P = 0.016 (two-tailed); odds ratio x = 4.9.
Table 2 Mean daily AED doses (mg) in relation to catameniality and reproductive dysfunction AED
Catamenial epilepsy
Noncatamenial epilepsy
PCOS
HH
Normal reproductive function
PHB VPA PHT CBZ
128.6 ± 36.6 (n = 7) 750.0 ± 250.0 (n = 3) 234.0 ± 0.0 (n = 4) 333.3 ± 115.5 (n = 3)
179.0 ± 110.3 (n = 25) 500.0 ± 223.6 (n = 7) 175.5 ± 62.5 (n = 8) 453.3 ± 190.7 (n = 30)
129.2 ± 78.1 (n = 6) 300 ± 0.0 (n = 1) 175.5 ± 82.7 (n = 2) 400.0 ± 163.3 (n = 4)
162.5 ± 87.5 (n = 14) 710 ± 274.8 (n = 5) 175.5 ± 67.5 (n = 4) 430.8 ± 179.7 (n = 13)
184.1 ± 124.1 (n = 12) 475.0 ± 125.8 (n = 4) 234.0 ± 0 (n = 6) 460.0 ± 213.1 (n = 16)
n, Number of patients on concrete AED.
188
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
Table 4 Relationship between side of focus and diagnosis of reproductive dysfunction in women with epilepsya Reproductive dysfunction diagnosis
Left-sided foci
Right-sided foci
Total
HHb PCOS Total
13 11 24
21 1 22
34 12 46
a 2 v = 8.12, P = 0.0044 with Yates correction; Fisher exact test: P = 0.0019 (two-tailed); odds ratio: x = 17.8. b HH, hypothalamic hypogonadism; PCOS, polycystic ovary syndrome.
Table 5 Catameniality and diagnosis of reproductive dysfunction in women with epilepsya Reproductive dysfunction diagnosis
Catamenial epilepsy
Noncatamenial epilepsy
Total
HHb PCOS Total
8 2 10
26 10 36
34 12 46
a b
v2 = 0.01, P = 0.93; Fisher exact test: P = 1.0 (nonsignificant). HH, hypothalamic hypogonadism; PCOS, polycystic ovary syndrome.
catamenial and noncatamenial groups. The odds ratio (x = 4.9) indicates that compared with women with right-sided foci, patients with left-sided foci, on average, are 5-fold more likely to have catamenial than noncatamenial seizures. On the contrary, compared with women with left-sided foci, women with right-sided foci are 0.2-fold more likely to develop noncatamenial epilepsy. The data on the relationship between side of focus and type of RD are much more impressive, as seen in Table
4. Statistically significant associations between PCOS and left-sided foci and between HH and right-sided foci (v2 = 8.12, P = 0.0044) were also observed by Herzog [6], Friedman and Herzog [7], and Herzog et al. [8]. Taking into account the odds ratio (x = 18.1), the probability of development of PCOS in women with left-sided foci is on average 18 times higher than that of developing HH, compared with females with right-sided foci. In other words, left-sided foci are associated with catamenial epilepsy (less frequently) and PCOS (more frequently), whereas right sided foci are linked with noncatamenial epilepsy (less frequently) and HH (more frequently). Considering all the data together, we suggest that there exists an association between catamenial epilepsy and PCOS, on the one hand, and between noncatamenial epilepsy and HH, on the other hand. Contrary to our expectations, no statistically significant associations between catameniality and type of RD were revealed (Table 5), implying that catameniality is not a characteristic of PCOS alone, and may occur in HH too. Although left-sided focal activity is associated with both PCOS and catameniality, it is obvious that additional, unknown factors (probably AEDs) should have impact on the mechanisms underlying development of PCOS. To test this hypothesis at the next stage, the mean daily doses of the AEDs used were compared with respect to catameniality and type of RD. The main findings of this comparison are summarized in the Table 2. There were no statistically significant differences in the mean daily doses of any AED among patients with PCOS, HH, and normal reproductive function. This implies that in the cohort studied, type of AED does not determine type of gynecological pathology.
Table 6 Comparison of chronological variables of epilepsy and comorbid reproductive dysfunction with respect to side of focus Chronological variable
Age at epilepsy onset Duration of epilepsy Age at RD onset Duration of RD Delta disorders
Catamenial epilepsy
Noncatamenial epilepsy
Left focus
Right focus
Left focus
Right focus
13.67 ± 4.33 11.33 ± 5.83 16.56 ± 4.88 8.44 ± 4.19 3.29 ± 3.71
8.3 ± 4.51 21.0 ± 3.46 P = 0.0213a 17.75 ± 7.68 11.75 ± 8.5 9.25 ± 5.1 P = 0.045a
13.71 ± 8.68 12.0 ± 10.3 20.14 ± 7.95 6.0 ± 5.01 6.43 ± 8.97
16.9 ± 0493 10.08 ± 7.44 P = 0.0165b 19.1 ± 6.1 6.28 ± 3.86 P = 0.0357b 1.57 ± 6.21 P = 0.0304a P = 0.044b
a
Statistically significant difference for comparison between left- and right-sided foci within catamenial and noncatamenial groups separately. Two-tailed Student t test was used. b Statistically significant difference for comparison between catamenial and noncatamenial groups separately for left- and right-sided foci. Two-tailed Student t test was used.
Table 7 Comparison of chronological variables of epilepsy with type of comorbid reproductive dysfunction and catamenialitya Chronological variable
Catamenial HH
Catamenial PCOS
Noncatamenial HH
Noncatamenial PCOS
Age at epilepsy onset Duration of epilepsy Age at RD onset Duration of RD Delta disorders
11.4 ± 5.2 13.6 ± 8.7 16.4 ± 5.9 8.6 ± 6.4 6.0 ± 6.7
16.0 ± 1.7 7.0 ± 5.6 16.0 ± 5.2 7.0 ± 2.6 0 ± 3.5
14.8 ± 7.2 9.5 ± 6.7 18.0 ± 6.0 6.4 ± 4.3 2.8 ± 6.9
10.6 ± 7.4 13. ± 12.5 17.0 ± 5.5 6.6 ± 4.5 6.1 ± 9.7
a
Comparisons did not reveal any statistically significant differences.
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
Table 6 outlines the chronological variables of epilepsy and concomitant gynecological pathology with respect to side of focus. Two-way analysis was employed: between left- and right-sided focal activity separately within catamenial and noncatamenial epilepsy and between catamenial and noncatamenial epilepsy separately for left- and rightsided foci. Thus, women with catamenial seizures with right-sided foci had a longer duration of epilepsy (21.0 ± 3.46 vs 11.33 ± 5.83, P = 0.021), a longer interval between onset of epilepsy and onset of RD (9.25 ± 5.1 vs 3.29 ± 3.71, P = 0.045), and a trend towards earlier age at epilepsy onset (8.3 ± 4.51 vs 13.67 ± 4.33, n.s.) in comparison with women with catamenial epilepsy with left-sided foci. On the other hand, in the noncatamenial group, the interval between onset of epilepsy and onset of RD was longer in patients with left-sided foci than in those with right-sided foci (6.43 ± 8.97 vs 1.57 ± 6.21, P = 0.030). In other words, the opposite relationship between catamenial and noncatamenial epilepsy seems to exist in terms of epilepsy and comorbid endocrine pathology. This may be determined by interaction between focus laterality and endocrine function in women with epilepsy. Comparison of the catamenial and noncatamenial groups separately for patients with left- and right-sided foci reveals discrepancies strictly in those with right-sided foci, but not those with left-sided foci. Thus, patients with noncatamenial epilepsy and right-sided foci have a later age at epilepsy onset (16.9 ± 7.2 vs 8.3 ± 4.51, P = 0.049), a shorter duration of epilepsy (10.08 ± 7.44 vs 21.0 ± 3.46, P = 0.017), a shorter duration of RD (6.28 ± 3.86 vs11.75 ± 8.5, P = 0.036), and a shorter interval between onset of epilepsy and onset of RD (1.57 ± 8.71 vs 9.25 ± 6.,6, P = 0.044) compared with patients with catamenial epilepsy with right-sided foci (Table 6). Similar discrepancies between chronological variables of catamenial and noncatamenial subtypes for patients with left-sided foci were not seen; that is, the left-sided foci in women with epilepsy did not correlate with differences in age at onset and duration of epilepsy and RD. Therefore, catamenial epilepsy associated with rightsided foci should be considered the oldest subtype, and catamenial epilepsy with left-sided activity and noncatamenial epilepsy with right-sided activity should be considered younger subtypes, according to Haeckel’s [12] and Geodakian’s [13] rules. It explains the longer duration of epilepsy and longest interval between onset of epilepsy and onset of RD in patients with catamenial epilepsy and right-sided foci. Notably, there were no statistically significant differences in age between the subgroups with left- and right-sided foci within the catamenial group (25.0 ± 6.93 vs 29.5 ± 6.4, n.s.) and within the noncatamenial group (25.82 ± 8.63 vs 26.26 ± 6.14, n.s.). Moreover, comparison of age in the catamenial and noncatamenial groups separately for patients with left- and right-sided foci also did not reveal significant differences, implying that duration of epilepsy
189
and duration of RD did not depend on age in the women studied. In the final stage of the study, an attempt was made to determine possible discrepancies in the chronological variables of epilepsy and concomitant RD by taking into account the association between types of RD (PCOS or HH) and catameniality. These data are summarized in Table 7. Nevertheless, there were no statistically significant differences, suggesting that there is no interaction between catameniality and subtype of RD in terms of their influence on the chronological variables studied. 4. Discussion The current study was designed to analyze the influence of side of focus in women with TLE on catameniality, subtype of RD, and chronological variables of epilepsy and RD. The study has certain limitations, because age at RD onset can be established only approximately, as this variable reflects the age at diagnosis of RD, rather than the actual age of onset. Nevertheless, an a priori correlation should exist between the age at onset and age at diagnosis of RD, and first symptoms of RD were deliberately chronologically fixed in all the women studied. To our knowledge, this is the first attempt to analyze the effect of side of focus on catameniality and chronological variables in women with epilepsy and concomitant RD. The second and not less important limitation of the study concerns the fact that it was performed without data on serum AED concentrations and serum reproductive hormone levels; although hormonally based diagnosis of reproductive endocrine disorders would have been preferred, this was not possible for each patient. The apparent discrepancy in our definitions of PCOS and HH from the general literature may have been based on our strong reliance on the demonstration of ovarian cysts rather than hormonal levels to make the diagnosis of PCOS. In addition, differentiation of PCOS from HH in the women studied was not the principal aim of the current trial and may be fairly criticized. In other words, we followed the diagnostic criteria used in Europe, but not in the United States [21]. That is why the hormone level assessments were not included in the final analysis, and were used to supplement ultrasound findings and such clinical signs as infertility, menstrual irregularity, and hirsutism for diagnostic purposes. Our study confirms entirely the hypothesis of Herzog [6] and Friedman and Herzog [7] that left-sided foci are associated with the development of PCOS, whereas right-sided foci are associated mainly with HH. These data from the literature, in turn, indirectly support our diagnoses of PCOS and HH, respectively, in our study, because the aforementioned authors did not use hormone levels for diagnostic purposes. In addition to those data, our findings demonstrated that left-sided foci in women with TLE are associated with the catamenial pattern of seizures (77%), although the catamenial pattern may also exist in women
190
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
with epilepsy with right-sided foci (23%). The odds ratios indicate that risk for development of PCOS over HH is 18-fold higher in women with left-sided foci than in those with right-sided foci. However, the probability of catamenial seizures is about 5-fold higher in women with left-sided foci than in those with right-sided foci, and vice versa, the probability of noncatamenial seizures in women with left-sided foci is only 0.2-fold higher than that for women with right-sided foci. Dean et al. [9] and Labar et al. 10] demonstrated that left-sided foci are nearly three times more prevalent than right-sided foci in localization-related epilepsy. Interestingly, according to Friedman and Herzog [7], PCOS is also more prevalent than HH, as these disorders are present in 5.0 and 1.5% of the general population and in 20 and 12% of women with TLE, respectively, although in our study, PCOS was recognized as rarer than HH (15% vs 43%). This discrepancy may be partly explained by the unrepresentativeness of the data for the group studied, because only women admitted to our department were included, meaning a certain proportion of women with PCOS may have remained beyond the scope of the current study. As was mentioned in the Introduction, signs and symptoms associated with the right hemisphere are older, evolutionarily speaking, than those associated with the left hemisphere [13–15]. In accordance with this paradigm, it may be speculated that catamenial seizures with right-sided foci and catamenial seizures with left-sided foci represent two subtypes of epilepsy that differ with respect to chronological variables, evolution, and prognosis. The age at onset of catamenial epilepsy with right-sided foci is the youngest, and this form seems to be the oldest in evolutionary terms in comparison with right-sided noncatamenial epilepsy. Less impressive discrepancies exist between women with left- and right-sided activity within the catamenial group. Here there is only a trend toward a discrepancy, particularly towards later age at epilepsy onset in patients with left-sided foci, compared with those with right-sided activity. Nonetheless, within the catamenial group, there was definitely a difference in the interval between onset of epilepsy and onset of RD between women with left- and right sided foci: this interval was significantly shorter for women with left-sided foci than for those with right-sided foci. Certainly, the small number of patients within the catamenial group was the main cause of the lack of stochastically significant differences in other variables between patients with left- and right-sided foci. Obviously, the primary hypothesis of the current study has only partly been confirmed, and it concerns strictly the difference between catamenial epilepsy with right-sided foci (the oldest in evolutionary terms) and noncatamenial epilepsy with right-sided foci (younger in evolutionary terms). In other words, in chronological terms, only subtypes within the right, but not the left, hemisphere may be contrasted, because left-sided foci were not associated with any discrepancies in chronological variables between catamenial and noncatamenial groups.
Principally, catamenial epilepsy with right-sided foci also was characterized by the longest interval between onset of epilepsy and onset of RD; that is, in terms of concomitant endocrine pathology, this form of epilepsy has more favorable course, because the onset of gynecological pathology is strictly delayed in comparison with catamenial epilepsy with left-sided activity. In other words, catamenial epilepsy with left-sided foci and noncatamenial epilepsy with right-sided foci represent not only forms of epilepsy with later onset, but also an unfavorable course in terms of rapid development of endocrine pathology. In such cases, PCOS prevails over HH in epilepsy with left-sided foci, and conversely, HH prevails over PCOS in epilepsy with right-sided foci, although an association between catameniality and subtype of RD was not uncovered in our study. Thus, prediction of type of RD on the basis of catameniality/noncatameniality is impossible. Obviously, in terms of the prognosis for concomitant reproductive disorders, catamenial epilepsy with left-sided foci and noncatamenial epilepsy with right-sided foci are characterized by the most unfavorable course. In contrast, catamenial epilepsy with right-sided foci and noncatamenial epilepsy with left-sided foci seem to have a more favorable course, although the prognosis in such subtypes is usually worse than in epilepsy without concomitant endocrine gynecological pathology. Certainly, further research is needed to differentiate epilepsy with comorbid endocrine pathology from epilepsy without such pathology. The lack of association between catameniality and RD subtype, at first glance, seems contradictory and unexplainable. Other, still unknown, factors may exist that have additional influence. AEDs are candidates for this role, although our data could not confirm this suggestion, and higher dose levels of VPA or other AEDs should not be regarded as a risk factor for the development of PCOS. Moreover, among the patients studied, only one woman with PCOS was taking VPA (at a daily dose of 300 mg); the other patients taking VPA had been diagnosed with HH (n = 4) or did not have RD (n = 6). On the other hand, there was a trend toward differences in the mean daily doses of VPA and PHT between catamenial and noncatamenial epilepsy, but these differences did not reach a statistically significant level, and should not be exaggerated, because catamenial or noncatamenial patterns of seizures usually preceded the start of treatment. Obviously, side of focus and catameniality/noncatameniality should be taken into account in predicting the course of epilepsy and concomitant gynecological pathology in women with TLE. In accordance with this rule, women with epilepsy with catamenial seizures and leftsided foci and women with epilepsy with noncatamenial seizures and right-sided foci seem to represent the patients at highest risk of developing endocrine gynecological pathology, whether PCOS or HH, in the shortest time after epilepsy onset. Further studies are required to differentiate more precisely these epilepsy subtypes based not only on catameniality/noncatameniality, but taking into account
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
other characteristics, such as serum reproductive hormone levels and serum AED concentrations in women with epilepsy. References [1] Isoja¨rvi JIT, Laatikainen TJ, Pakarinen AJ, et al. Polycystic ovaries and hyperandrogenism in women taking valproate for epilepsy. N Engl J Med 1993;329:1383–8. [2] Isoja¨rvi JIT, Laatikainen TJ, Knip M, et al. Obesity and endocrine disorders in women taking valproate for epilepsy. Ann Neurol 1996;39:579–84. [3] Isoja¨rvi JIT, Tauboll E, Pakarinen A, et al. Altered ovarian function and cardiovascular risk factors in valproate-treated women. Am J Med 2001;111:290–6. [4] Murialdo G, Galimberti CA, Magri F, et al. Menstrual cycle and ovary alterations in women with epilepsy on antiepileptic therapy. J Endocrinol Invest 1997;20:519–26. [5] Murialdo G, Galimberti CA, Gianelli MV, et al. Effects of valproate, phenobarbital, and carbamazepine on sex steroid setup in women with epilepsy. Clin Neuropharmacol 1998;21:52–8. [6] Herzog AG. A relationship between particular reproductive endocrine disorders and the laterality of epileptiform discharges in women with epilepsy. Neurology 1993;43:1907–10. [7] Friedman MN, Herzog AG. Epilepsy and disorders of reproduction. In: Ettinger AB, Devinsky O, editors. Managing epilepsy and coexisting disorders. Boston: Butterworth/Heinemann; 2002. p. 305–15. [8] Herzog AG, Coleman AE, Jacobs AR, et al. Interictal EEG discharges, reproductive hormones and menstrual disorders in epilepsy. Ann Neurol 2003;54:625–37. [9] Dean A, Solomon G, Harden C, et al. Left-hemisphere dominance of epileptiform discharges. Epilepsia 1997;38:503–5.
191
[10] Labar D, Dilone L, Solomon G, et al. Epileptogenesis: left or right hemisphere dominance? Preliminary findings in hospital-based population. Seizure 2001;10:570–2. [11] Kalinin VV, Zheleznova EV. Interface between endocrinology and psychiatry in women with epilepsy: analysis of relationship with comorbid pathology. Epilepsy Behav 2006;8:192–200. ¨ ber die Entstehung und den Stammenbaum des [12] Haeckel E. U Menschengeschlechts. Berlin; 1870; [In German]. [13] Geodakian VA. Asynchronous asymmetry (sexual and lateral differentiation as a consequence of asynchronous evolution). J High Nervous Act 1993;43:543–61. [In Russian]. [14] Crow TJ. Introduction. In: Crow TJ, editor. The speciation of modern Homo sapiens. London: The British Academy, Oxford Univ. Press; 2002. p. 1–20. [15] Corballis MC. Laterality and human speciation. In: Crow TJ, editor. The speciation of modern Homo sapiens. London: The British Academy, Oxford Univ. Press; 2002. p. 137–52. [16] Herzog AG, Klein P, Ransil BJ. Three patterns of catamenial epilepsy. Epilepsia 1997;38:1082–8. [17] Sisodiya SM. Mechanisms of drug resistance in epilepsy. In: Shorvon S, Perucca E, Fish D, Dodson E, editors. The treatment of epilepsy. Ames, IA: Blackwell; 2004. p. 84–8. [18] Michelson D, Gold PW. Laboratory evaluation of neuroendocrine systems. In: Wolkowitz OM, Rothschild AJ, editors. Psychoneuroendocrinology. The scientific basis of clinical practice. Arlington, VA: American Psychiatric Press; 2003. p. 469–88. [19] Herzog AG, Seibel MM, Schomer DL, et al. Reproductive endocrine disorders in women with partial seizures of temporal lobe origin. Arch Neurol 1986;43:341–6. [20] Harden CL. Polycystic ovaries and polycystic ovary syndrome in epilepsy: evidence for neurogonadal disease. Epilepsy Curr 2005;5(4):142–6. [21] Duncan S. Polycystic ovarian syndrome in women with epilepsy: a review. Epilepsia 2001;42(Suppl. 3):60–5.
Chronology and evolution of temporal lobe epilepsy and endocrine reproductive dysfunction in women: Relationships to side of focus and catameniality Vladimir V. Kalinin *, Elena V. Zheleznova Department of Brain Organic Disorders and Epilepsy, Moscow Research Institute of Psychiatry, Ministry of Health and Social Development, 3, Poteshnaya Street, 107076 Moscow, Russian Federation Received 19 September 2006; revised 7 April 2007; accepted 13 April 2007 Available online 18 June 2007
Abstract The current study was carried out to determine the possible associations between side of focus, catameniality, reproductive dysfunction (RD), and chronology variables for epilepsy and concomitant RD in women. Eighty women of childbearing potential with temporal lobe epilepsy were included in the study. Catamenial epilepsy was observed mainly in women with left-sided foci, and a noncatamenial pattern in women with right-sided foci. Left-sided foci were associated with polycystic ovary syndrome, and right-sided foci with hypogonadotropic hypogonadism. Catamenial epilepsy with right-sided foci was associated with longer duration of epilepsy (P = 0.021), trend toward earlier age at onset of epilepsy, and trend toward longer interval between onset of epilepsy and onset of RD compared with catamenial epilepsy with left-sided foci. On the other hand, noncatamenial epilepsy with right-sided foci was characterized by a shorter interval between onset of epilepsy and onset of RD in comparison with noncatamenial epilepsy with left-sided activity (P = 0.03). In addition, comparison of patients with right-sided foci with catamenial and noncatamenial patterns of seizures revealed earlier age of epilepsy onset (P = 0.049), longer duration of epilepsy (P = 0.017) and of RD (P = 0.036), and longer interval between onset of epilepsy and onset of RD (P = 0.048) in patients with catamenial epilepsy. From an evolutionary point of view, catamenial epilepsy with right-sided focal activity seems to be the oldest subtype. Ó 2007 Elsevier Inc. All rights reserved. Keywords: Female epilepsy; Catameniality; Reproductive dysfunction; Laterality of foci; Age at epilepsy onset
1. Introduction Endocrine and psychiatric disorders are thought to be the most frequent comorbid pathology among women with epilepsy. These disorders continue to draw the attention of many research groups, because they remain utterly crucial for women in relation to childbearing and antiepileptic drug (AED) intake, which, in turn, have influence on endocrine and gynecological systems. Data on the effect of valproate (VPA) on reproductive function in women with epilepsy are ambiguous and rather controversial. Based *
Corresponding author. E-mail addresses: [email protected], [email protected] (V.V. Kalinin). 1525-5050/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.yebeh.2007.04.014
on the work of Isoja¨rvi et al. [1–3], the relationship between VPA use and polycystic ovary syndrome (PCOS) and obesity in women with epilepsy is regarded as well established, although not all research groups could confirm their findings, and the findings of Murialdo et al. [4] are somewhat contradictory. Murialdo and co-workers reported that PCOS does not occur more frequently under VPA treatment than under treatment with other AEDs. Moreover, they observed PCOS in 12% of women on phenobarbital, in 21% on carbamazepine, and in no women on VPA monotherapy [4], although in a subsequent controlled study [5], these authors reported increased body mass index, higher androgen levels, and higher frequency of ovulatory dysfunction in 63.5% of women treated with VPA in comparison to women treated with CBZ or phenobarbital
186
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
or to controls. Thus, the association between VPA administration and high androgen level, ovulatory dysfunction, and weight gain was again confirmed. The aforementioned discrepancy in data may be the result not only of the AEDs themselves, but of differences in the types of epilepsy, seizure semiology, and other neurobiological factors, such as side of foci. Herzog [6], Friedman and Herzog [7], and Herzog et al. [8] have proved the role of focus lateralization in the genesis of certain gynecological disorders, and PCOS has been observed in association with left-sided foci, whereas hypothalamic hypogonadism (HH) prevails in women with temporal lobe epilepsy with right-sided foci. The frequency of left-sided foci is believed to be, on average, threefold higher than the prevalence of right-sided foci [9,10], suggesting that PCOS should be more prevalent than HH, if there exists a strict relationship between focus lateralization and type of gynecological pathology. Our previous study established the existence of certain relationships between endocrine and gynecological pathology, catameniality, and psychopathological syndromes in women with epilepsy. Thus, a reciprocal interrelationship between thyroid and reproductive pathology was revealed, and catameniality was associated much more with reproductive dysfunction than with thyroid pathology in women with epilepsy [11]. Much less is known about temporal sequences of comorbid endocrine and gynecological pathology in relation to focus lateralization and catameniality in localizationrelated epilepsies, although this might shed light on the evolution of epilepsy with respect not only to ontogenesis, but to phylogeny too. According to the widely known biogenetic rule of Haeckel [12], the temporal sequence of acquired characteristics in ontogeny is quite similar to that in phylogeny in a condensed manner. In other words, the signs, symptoms, disorders, and pathology that occur during an individual’s life represent the time of their appearance in the phylogeny of the species. On the basis of this rule, it may be predicted that early occurrence of signs, symptoms, and disorders in individual Homo sapiens reflect their early origin in evolution and, conversely, their occurrence late in life indicates they are novel evolutionarily. It has also been suggested that hemispheric lateralization of different functions indicates the time of their occurrence in phylogeny [13]. Thus, the right hemisphere is responsible for Gestalt analysis, which is regarded as ancient brain function. Conversely, the left hemisphere determines analytical and verbal processing and is responsible for speech function. The latter brain functions are thought to be the prerogative strictly of Homo sapiens and were acquired in the later stages of human evolution [13–15], although this hypothesis seems somewhat speculative and should be confirmed in special studies. Nevertheless, taking all these data together, we suggest that any characteristic or symptom of neuropsychiatric disorders associated with the left hemisphere should be regarded as having occurred later in evolution
than those linked with the right hemisphere [13,14]. Thus, PCOS, because it is associated with left-sided foci in epilepsy, should be considered a relatively new disorder in terms of evolutionary origin than HH, which is associated with right-sided foci [6–8]. This assumption, however, remains strictly speculative. The principal objective of the present study was to determine the relationship of chronological variables (age at epilepsy onset and concomitant endocrine reproductive disorders) to side of foci and catameniality in women with epilepsy. In addition, the possible relationship between side of foci, on the one hand, and catameniality and noncatameniality, on the other hand, was studied. 2. Materials and methods The study was carried out on 80 women of childbearing age with temporal lobe epilepsy (TLE). Women with a definite diagnosis of TLE were intentionally selected. These women were recruited from all patients with epilepsy admitted for treatment at the Department of Brain Organic Disorders and Epilepsy, Moscow Research Institute of Psychiatry. Patients diagnosed with idiopathic generalized epilepsies and localization-related epilepsy of nontemporal origin were excluded from the study. Among all the women studied, left-sided foci were detected in 37 patients, and right-sided foci in 43 patients. Side of focus was detected strictly by looking at the EEG; the data on ictal semiology were not used for this purpose. MRI scans were performed on all patients prior to the study. Based on the results of the MRI study, the symptomatic form of TLE was diagnosed in 17 women, and the cryptogenic form in 63 women. Among the women with symptomatic TLE, cysts of traumatic origin were detected in 12 patients and mediotemporal sclerosis in 5 patients. Patients ranged from 16 to 45 years in age (mean ± SD: 26.13 ± 7.12 years). Thirteen patients had the so-called catamenial form of epilepsy, and 67 had the noncatamenial form. Here, catamenial epilepsy is understood to be the occurrence of seizures strictly in relation to certain days of the menstrual cycle. According to the classification of catamenial epilepsy proposed by Herzog [16], most of the women (12 patients, 92%) had the so-called perimenstrual pattern of catamenial epilepsy, that is, with seizures in interval from Day 3 to Day +3 of the menstrual cycle. All women were properly assessed by a gynecologist and an endocrinologist before inclusion in the current study. In addition, the ovaries of each woman were scanned by ultrasound to make precise diagnoses. Seizure semiology data are summarized in Table 1, and include mean seizure frequency for each type of seizure per month with respect to catameniality and subtype of RD. Data covered the time to inclusion in the study and through treatment with AEDs. Analysis of seizure dynamics during AED treatment was not considered the principal aim of the current study. Nevertheless, there were no statistically significant discrepancies in frequency of seizures between women with catamenial and noncatamenial seizures and between women with different subtypes of RD. Among the women studied, the diagnosis of refractory epilepsy was confirmed in 11 (13.8%), although epilepsy intractability was not considered the principal criterion for inclusion in or exclusion from the study. Here intractable epilepsy is defined as seizures that do not respond to any of three first-line AEDs usually administered for the given epilepsy syndromes [17]. The duration of follow-up under AED treatment in the current study varied, but usually was not less than 3 months. Mean seizure frequency before and during treatment was analyzed, and if seizure reduction was less than 50%, the epilepsy was regarded as intractable [17]. Obviously, seizure frequency was higher in patients with treatment-resistant seizures than in patients with a favorable response to AED therapy. The proportions of women with refractory epilepsy in the catamenial and noncatamenial groups were similar: 15 and 13% (2 and 9 patients, respectively).
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
187
Table 1 Mean seizure frequency per month relative to catameniality/noncatameniality and reproductive dysfunction Type of seizures
Catamenial epilepsy
Noncatamenial epilepsy
PCOSa
HH
Normal reproductive function
Simple partial Complex partial Secondarily generalized All seizures
66.5 ± 197.8 0.2 ± 0.8 8.3 ± 15.1 75.0 ± 199.0
34.2 ± 92.4 5.1 ± 22.6 7.7 ± 23.0 47.1 ± 105.4
39.5 ± 102 5.0 ± 9.2 19.3 ± 51.0 63.8 ± 151.0
46.4 ± 137.1 1.2 ± 4.3 6.6 ± 12.7 54.1 ± 137.2
29.1 ± 98.6 8.2 ± 33.0 5.2 ± 7.9 42.5 ± 104.9
a
HH, hypothalamic hypogonadism; PCOS, polycystic ovary syndrome.
All patients continued to take their AEDs. The AEDs being taken by the women prior to inclusion in the study are listed in Table 2. As can be seen, four first-line AEDs were usually used. Serum concentrations were not obtained for all patients, and so these data are not part of the current study. Nevertheless, it should be pointed out that in cases in which these levels were available, mean serum concentrations of all drugs were within the therapeutic windows. In line with the principal aim of this study, age at onset of epilepsy and concomitant endocrine reproductive pathology were determined, and the interval between onset of epilepsy and onset of endocrine RD (so-called delta disease) was calculated with the formula: age at reproductive disorder onset age at epilepsy onset. The study design was strictly retrospective, and specific information from the women’s diaries was selected and analyzed for this purpose (occurrence of seizures during the previous 2 years, and their possible relationship to the menstrual cycle). Semiology and frequency of seizures were also taken into account for further analysis. Because measurement of sex hormone levels was not part of the main objective of the current study, these levels were not assessed in all patients. Nevertheless, such data were obtained for 24 patients (9 with catamenial and 15 with noncatamenial seizures). Venous blood samples for assessment of serum levels of luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol, progesterone, testosterone, and prolactin were obtained on the seventh day of the menstrual cycle. Hypothalamic amenorrhea (i.e., HH), as defined by Herzog et al. [19], was diagnosed in 34 women. This diagnosis usually was apparent in women with menstrual dysfunction in the absence of any cystic alterations in ovaries. In addition, patients with HH were also characterized by lack of hirsutism, acne, and weight gain. Results of hormone level assessments were obtained for only 12 women with HH and, when compared with reference values [18], revealed low serum LH (1.25 ± 1.75 mIU/mL; normal range, 3–20 mIU/mL), low serum FSH (3.16 ± 2.38 mIU/ml; normal range, 6–23 mIU/mL); near-normal serum estradiol (242.37 ± 340.74 pg/ mL; normal range, 10–200 pg/mL); elevated serum progesterone (9.48 ± 17.64 ng/mL; normal range, 0.1–1.5 ng/mL), and slightly elevated serum free testosterone (2.61 ± 0.76 ng/dL; normal range, 0.3–1.9 ng/dL). On the contrary, PCOS was diagnosed in 12 women. Ultrasound findings constituted the main criterion for this diagnosis, and in most of these women, no fewer than 10 cysts were detected. Additional clinical criteria for PCOS used in the current study included: presence of ovulatory dysfunction, such as polymenorrhea, oligomenorrhea, or amenorrhea; and clinical signs of hyperandrogenism, such as hirsutism [16,19,20]. Laboratory findings in these patients revealed normal LH (5.64 ± 1.69 mIU/
mL), normal FSH (5.15 ± 2.38 mIU/mL), low estradiol (0.45 ± 0.49 pg/ mL), normal progesterone (1.2 ± 0.98 ng/mL), and elevated free testosterone (5.75 ± 2.19 ng/dL) levels. In the remaining 34 women, no reproductive disorder was diagnosed, and all had normal menstrual cycles. It should be stressed that results of endocrine tests were not available for all women to confirm the diagnosis of PCOS or HH. Therefore, the ultrasound scan coupled with clinical data was the principal method for diagnosis of RD.
2.1. Statistics All data were processed with the Statistica program (sixth version) on a personal computer. v2 statistics and Fisher’s exact test were used for the analysis of possible association between so-called nominal variables (i.e., type of endocrine pathology, catameniality, and side of focus). On the other hand, Student’s t test was used to analyze quantitative variables (onset of epilepsy, onset of RD, and the interval between).
3. Results The main results are summarized in Tables 3–7. Table 3 outlines the relationship between catameniality and side of focus in the women studied. As can be seen, catamenial seizures were observed more frequently in women with leftsided foci, whereas the noncatamenial pattern was observed in patients with right-sided foci. Fisher’s exact test (P = 0.016) and v2 statistics (v2 = 4.49, P = 0.034) detected statistically significant discrepancies between the Table 3 Relationship between side of focus and catameniality in women with epilepsy Type of epilepsy
Left-sided foci
Right-sided foci
Total
Catamenial Noncatamenial Total
10 27 37
3 40 43
13 67 80
v2 = 4.49, P = 0.034 with Yates correction; Fisher exact test: P = 0.016 (two-tailed); odds ratio x = 4.9.
Table 2 Mean daily AED doses (mg) in relation to catameniality and reproductive dysfunction AED
Catamenial epilepsy
Noncatamenial epilepsy
PCOS
HH
Normal reproductive function
PHB VPA PHT CBZ
128.6 ± 36.6 (n = 7) 750.0 ± 250.0 (n = 3) 234.0 ± 0.0 (n = 4) 333.3 ± 115.5 (n = 3)
179.0 ± 110.3 (n = 25) 500.0 ± 223.6 (n = 7) 175.5 ± 62.5 (n = 8) 453.3 ± 190.7 (n = 30)
129.2 ± 78.1 (n = 6) 300 ± 0.0 (n = 1) 175.5 ± 82.7 (n = 2) 400.0 ± 163.3 (n = 4)
162.5 ± 87.5 (n = 14) 710 ± 274.8 (n = 5) 175.5 ± 67.5 (n = 4) 430.8 ± 179.7 (n = 13)
184.1 ± 124.1 (n = 12) 475.0 ± 125.8 (n = 4) 234.0 ± 0 (n = 6) 460.0 ± 213.1 (n = 16)
n, Number of patients on concrete AED.
188
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
Table 4 Relationship between side of focus and diagnosis of reproductive dysfunction in women with epilepsya Reproductive dysfunction diagnosis
Left-sided foci
Right-sided foci
Total
HHb PCOS Total
13 11 24
21 1 22
34 12 46
a 2 v = 8.12, P = 0.0044 with Yates correction; Fisher exact test: P = 0.0019 (two-tailed); odds ratio: x = 17.8. b HH, hypothalamic hypogonadism; PCOS, polycystic ovary syndrome.
Table 5 Catameniality and diagnosis of reproductive dysfunction in women with epilepsya Reproductive dysfunction diagnosis
Catamenial epilepsy
Noncatamenial epilepsy
Total
HHb PCOS Total
8 2 10
26 10 36
34 12 46
a b
v2 = 0.01, P = 0.93; Fisher exact test: P = 1.0 (nonsignificant). HH, hypothalamic hypogonadism; PCOS, polycystic ovary syndrome.
catamenial and noncatamenial groups. The odds ratio (x = 4.9) indicates that compared with women with right-sided foci, patients with left-sided foci, on average, are 5-fold more likely to have catamenial than noncatamenial seizures. On the contrary, compared with women with left-sided foci, women with right-sided foci are 0.2-fold more likely to develop noncatamenial epilepsy. The data on the relationship between side of focus and type of RD are much more impressive, as seen in Table
4. Statistically significant associations between PCOS and left-sided foci and between HH and right-sided foci (v2 = 8.12, P = 0.0044) were also observed by Herzog [6], Friedman and Herzog [7], and Herzog et al. [8]. Taking into account the odds ratio (x = 18.1), the probability of development of PCOS in women with left-sided foci is on average 18 times higher than that of developing HH, compared with females with right-sided foci. In other words, left-sided foci are associated with catamenial epilepsy (less frequently) and PCOS (more frequently), whereas right sided foci are linked with noncatamenial epilepsy (less frequently) and HH (more frequently). Considering all the data together, we suggest that there exists an association between catamenial epilepsy and PCOS, on the one hand, and between noncatamenial epilepsy and HH, on the other hand. Contrary to our expectations, no statistically significant associations between catameniality and type of RD were revealed (Table 5), implying that catameniality is not a characteristic of PCOS alone, and may occur in HH too. Although left-sided focal activity is associated with both PCOS and catameniality, it is obvious that additional, unknown factors (probably AEDs) should have impact on the mechanisms underlying development of PCOS. To test this hypothesis at the next stage, the mean daily doses of the AEDs used were compared with respect to catameniality and type of RD. The main findings of this comparison are summarized in the Table 2. There were no statistically significant differences in the mean daily doses of any AED among patients with PCOS, HH, and normal reproductive function. This implies that in the cohort studied, type of AED does not determine type of gynecological pathology.
Table 6 Comparison of chronological variables of epilepsy and comorbid reproductive dysfunction with respect to side of focus Chronological variable
Age at epilepsy onset Duration of epilepsy Age at RD onset Duration of RD Delta disorders
Catamenial epilepsy
Noncatamenial epilepsy
Left focus
Right focus
Left focus
Right focus
13.67 ± 4.33 11.33 ± 5.83 16.56 ± 4.88 8.44 ± 4.19 3.29 ± 3.71
8.3 ± 4.51 21.0 ± 3.46 P = 0.0213a 17.75 ± 7.68 11.75 ± 8.5 9.25 ± 5.1 P = 0.045a
13.71 ± 8.68 12.0 ± 10.3 20.14 ± 7.95 6.0 ± 5.01 6.43 ± 8.97
16.9 ± 0493 10.08 ± 7.44 P = 0.0165b 19.1 ± 6.1 6.28 ± 3.86 P = 0.0357b 1.57 ± 6.21 P = 0.0304a P = 0.044b
a
Statistically significant difference for comparison between left- and right-sided foci within catamenial and noncatamenial groups separately. Two-tailed Student t test was used. b Statistically significant difference for comparison between catamenial and noncatamenial groups separately for left- and right-sided foci. Two-tailed Student t test was used.
Table 7 Comparison of chronological variables of epilepsy with type of comorbid reproductive dysfunction and catamenialitya Chronological variable
Catamenial HH
Catamenial PCOS
Noncatamenial HH
Noncatamenial PCOS
Age at epilepsy onset Duration of epilepsy Age at RD onset Duration of RD Delta disorders
11.4 ± 5.2 13.6 ± 8.7 16.4 ± 5.9 8.6 ± 6.4 6.0 ± 6.7
16.0 ± 1.7 7.0 ± 5.6 16.0 ± 5.2 7.0 ± 2.6 0 ± 3.5
14.8 ± 7.2 9.5 ± 6.7 18.0 ± 6.0 6.4 ± 4.3 2.8 ± 6.9
10.6 ± 7.4 13. ± 12.5 17.0 ± 5.5 6.6 ± 4.5 6.1 ± 9.7
a
Comparisons did not reveal any statistically significant differences.
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
Table 6 outlines the chronological variables of epilepsy and concomitant gynecological pathology with respect to side of focus. Two-way analysis was employed: between left- and right-sided focal activity separately within catamenial and noncatamenial epilepsy and between catamenial and noncatamenial epilepsy separately for left- and rightsided foci. Thus, women with catamenial seizures with right-sided foci had a longer duration of epilepsy (21.0 ± 3.46 vs 11.33 ± 5.83, P = 0.021), a longer interval between onset of epilepsy and onset of RD (9.25 ± 5.1 vs 3.29 ± 3.71, P = 0.045), and a trend towards earlier age at epilepsy onset (8.3 ± 4.51 vs 13.67 ± 4.33, n.s.) in comparison with women with catamenial epilepsy with left-sided foci. On the other hand, in the noncatamenial group, the interval between onset of epilepsy and onset of RD was longer in patients with left-sided foci than in those with right-sided foci (6.43 ± 8.97 vs 1.57 ± 6.21, P = 0.030). In other words, the opposite relationship between catamenial and noncatamenial epilepsy seems to exist in terms of epilepsy and comorbid endocrine pathology. This may be determined by interaction between focus laterality and endocrine function in women with epilepsy. Comparison of the catamenial and noncatamenial groups separately for patients with left- and right-sided foci reveals discrepancies strictly in those with right-sided foci, but not those with left-sided foci. Thus, patients with noncatamenial epilepsy and right-sided foci have a later age at epilepsy onset (16.9 ± 7.2 vs 8.3 ± 4.51, P = 0.049), a shorter duration of epilepsy (10.08 ± 7.44 vs 21.0 ± 3.46, P = 0.017), a shorter duration of RD (6.28 ± 3.86 vs11.75 ± 8.5, P = 0.036), and a shorter interval between onset of epilepsy and onset of RD (1.57 ± 8.71 vs 9.25 ± 6.,6, P = 0.044) compared with patients with catamenial epilepsy with right-sided foci (Table 6). Similar discrepancies between chronological variables of catamenial and noncatamenial subtypes for patients with left-sided foci were not seen; that is, the left-sided foci in women with epilepsy did not correlate with differences in age at onset and duration of epilepsy and RD. Therefore, catamenial epilepsy associated with rightsided foci should be considered the oldest subtype, and catamenial epilepsy with left-sided activity and noncatamenial epilepsy with right-sided activity should be considered younger subtypes, according to Haeckel’s [12] and Geodakian’s [13] rules. It explains the longer duration of epilepsy and longest interval between onset of epilepsy and onset of RD in patients with catamenial epilepsy and right-sided foci. Notably, there were no statistically significant differences in age between the subgroups with left- and right-sided foci within the catamenial group (25.0 ± 6.93 vs 29.5 ± 6.4, n.s.) and within the noncatamenial group (25.82 ± 8.63 vs 26.26 ± 6.14, n.s.). Moreover, comparison of age in the catamenial and noncatamenial groups separately for patients with left- and right-sided foci also did not reveal significant differences, implying that duration of epilepsy
189
and duration of RD did not depend on age in the women studied. In the final stage of the study, an attempt was made to determine possible discrepancies in the chronological variables of epilepsy and concomitant RD by taking into account the association between types of RD (PCOS or HH) and catameniality. These data are summarized in Table 7. Nevertheless, there were no statistically significant differences, suggesting that there is no interaction between catameniality and subtype of RD in terms of their influence on the chronological variables studied. 4. Discussion The current study was designed to analyze the influence of side of focus in women with TLE on catameniality, subtype of RD, and chronological variables of epilepsy and RD. The study has certain limitations, because age at RD onset can be established only approximately, as this variable reflects the age at diagnosis of RD, rather than the actual age of onset. Nevertheless, an a priori correlation should exist between the age at onset and age at diagnosis of RD, and first symptoms of RD were deliberately chronologically fixed in all the women studied. To our knowledge, this is the first attempt to analyze the effect of side of focus on catameniality and chronological variables in women with epilepsy and concomitant RD. The second and not less important limitation of the study concerns the fact that it was performed without data on serum AED concentrations and serum reproductive hormone levels; although hormonally based diagnosis of reproductive endocrine disorders would have been preferred, this was not possible for each patient. The apparent discrepancy in our definitions of PCOS and HH from the general literature may have been based on our strong reliance on the demonstration of ovarian cysts rather than hormonal levels to make the diagnosis of PCOS. In addition, differentiation of PCOS from HH in the women studied was not the principal aim of the current trial and may be fairly criticized. In other words, we followed the diagnostic criteria used in Europe, but not in the United States [21]. That is why the hormone level assessments were not included in the final analysis, and were used to supplement ultrasound findings and such clinical signs as infertility, menstrual irregularity, and hirsutism for diagnostic purposes. Our study confirms entirely the hypothesis of Herzog [6] and Friedman and Herzog [7] that left-sided foci are associated with the development of PCOS, whereas right-sided foci are associated mainly with HH. These data from the literature, in turn, indirectly support our diagnoses of PCOS and HH, respectively, in our study, because the aforementioned authors did not use hormone levels for diagnostic purposes. In addition to those data, our findings demonstrated that left-sided foci in women with TLE are associated with the catamenial pattern of seizures (77%), although the catamenial pattern may also exist in women
190
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
with epilepsy with right-sided foci (23%). The odds ratios indicate that risk for development of PCOS over HH is 18-fold higher in women with left-sided foci than in those with right-sided foci. However, the probability of catamenial seizures is about 5-fold higher in women with left-sided foci than in those with right-sided foci, and vice versa, the probability of noncatamenial seizures in women with left-sided foci is only 0.2-fold higher than that for women with right-sided foci. Dean et al. [9] and Labar et al. 10] demonstrated that left-sided foci are nearly three times more prevalent than right-sided foci in localization-related epilepsy. Interestingly, according to Friedman and Herzog [7], PCOS is also more prevalent than HH, as these disorders are present in 5.0 and 1.5% of the general population and in 20 and 12% of women with TLE, respectively, although in our study, PCOS was recognized as rarer than HH (15% vs 43%). This discrepancy may be partly explained by the unrepresentativeness of the data for the group studied, because only women admitted to our department were included, meaning a certain proportion of women with PCOS may have remained beyond the scope of the current study. As was mentioned in the Introduction, signs and symptoms associated with the right hemisphere are older, evolutionarily speaking, than those associated with the left hemisphere [13–15]. In accordance with this paradigm, it may be speculated that catamenial seizures with right-sided foci and catamenial seizures with left-sided foci represent two subtypes of epilepsy that differ with respect to chronological variables, evolution, and prognosis. The age at onset of catamenial epilepsy with right-sided foci is the youngest, and this form seems to be the oldest in evolutionary terms in comparison with right-sided noncatamenial epilepsy. Less impressive discrepancies exist between women with left- and right-sided activity within the catamenial group. Here there is only a trend toward a discrepancy, particularly towards later age at epilepsy onset in patients with left-sided foci, compared with those with right-sided activity. Nonetheless, within the catamenial group, there was definitely a difference in the interval between onset of epilepsy and onset of RD between women with left- and right sided foci: this interval was significantly shorter for women with left-sided foci than for those with right-sided foci. Certainly, the small number of patients within the catamenial group was the main cause of the lack of stochastically significant differences in other variables between patients with left- and right-sided foci. Obviously, the primary hypothesis of the current study has only partly been confirmed, and it concerns strictly the difference between catamenial epilepsy with right-sided foci (the oldest in evolutionary terms) and noncatamenial epilepsy with right-sided foci (younger in evolutionary terms). In other words, in chronological terms, only subtypes within the right, but not the left, hemisphere may be contrasted, because left-sided foci were not associated with any discrepancies in chronological variables between catamenial and noncatamenial groups.
Principally, catamenial epilepsy with right-sided foci also was characterized by the longest interval between onset of epilepsy and onset of RD; that is, in terms of concomitant endocrine pathology, this form of epilepsy has more favorable course, because the onset of gynecological pathology is strictly delayed in comparison with catamenial epilepsy with left-sided activity. In other words, catamenial epilepsy with left-sided foci and noncatamenial epilepsy with right-sided foci represent not only forms of epilepsy with later onset, but also an unfavorable course in terms of rapid development of endocrine pathology. In such cases, PCOS prevails over HH in epilepsy with left-sided foci, and conversely, HH prevails over PCOS in epilepsy with right-sided foci, although an association between catameniality and subtype of RD was not uncovered in our study. Thus, prediction of type of RD on the basis of catameniality/noncatameniality is impossible. Obviously, in terms of the prognosis for concomitant reproductive disorders, catamenial epilepsy with left-sided foci and noncatamenial epilepsy with right-sided foci are characterized by the most unfavorable course. In contrast, catamenial epilepsy with right-sided foci and noncatamenial epilepsy with left-sided foci seem to have a more favorable course, although the prognosis in such subtypes is usually worse than in epilepsy without concomitant endocrine gynecological pathology. Certainly, further research is needed to differentiate epilepsy with comorbid endocrine pathology from epilepsy without such pathology. The lack of association between catameniality and RD subtype, at first glance, seems contradictory and unexplainable. Other, still unknown, factors may exist that have additional influence. AEDs are candidates for this role, although our data could not confirm this suggestion, and higher dose levels of VPA or other AEDs should not be regarded as a risk factor for the development of PCOS. Moreover, among the patients studied, only one woman with PCOS was taking VPA (at a daily dose of 300 mg); the other patients taking VPA had been diagnosed with HH (n = 4) or did not have RD (n = 6). On the other hand, there was a trend toward differences in the mean daily doses of VPA and PHT between catamenial and noncatamenial epilepsy, but these differences did not reach a statistically significant level, and should not be exaggerated, because catamenial or noncatamenial patterns of seizures usually preceded the start of treatment. Obviously, side of focus and catameniality/noncatameniality should be taken into account in predicting the course of epilepsy and concomitant gynecological pathology in women with TLE. In accordance with this rule, women with epilepsy with catamenial seizures and leftsided foci and women with epilepsy with noncatamenial seizures and right-sided foci seem to represent the patients at highest risk of developing endocrine gynecological pathology, whether PCOS or HH, in the shortest time after epilepsy onset. Further studies are required to differentiate more precisely these epilepsy subtypes based not only on catameniality/noncatameniality, but taking into account
V.V. Kalinin, E.V. Zheleznova / Epilepsy & Behavior 11 (2007) 185–191
other characteristics, such as serum reproductive hormone levels and serum AED concentrations in women with epilepsy. References [1] Isoja¨rvi JIT, Laatikainen TJ, Pakarinen AJ, et al. Polycystic ovaries and hyperandrogenism in women taking valproate for epilepsy. N Engl J Med 1993;329:1383–8. [2] Isoja¨rvi JIT, Laatikainen TJ, Knip M, et al. Obesity and endocrine disorders in women taking valproate for epilepsy. Ann Neurol 1996;39:579–84. [3] Isoja¨rvi JIT, Tauboll E, Pakarinen A, et al. Altered ovarian function and cardiovascular risk factors in valproate-treated women. Am J Med 2001;111:290–6. [4] Murialdo G, Galimberti CA, Magri F, et al. Menstrual cycle and ovary alterations in women with epilepsy on antiepileptic therapy. J Endocrinol Invest 1997;20:519–26. [5] Murialdo G, Galimberti CA, Gianelli MV, et al. Effects of valproate, phenobarbital, and carbamazepine on sex steroid setup in women with epilepsy. Clin Neuropharmacol 1998;21:52–8. [6] Herzog AG. A relationship between particular reproductive endocrine disorders and the laterality of epileptiform discharges in women with epilepsy. Neurology 1993;43:1907–10. [7] Friedman MN, Herzog AG. Epilepsy and disorders of reproduction. In: Ettinger AB, Devinsky O, editors. Managing epilepsy and coexisting disorders. Boston: Butterworth/Heinemann; 2002. p. 305–15. [8] Herzog AG, Coleman AE, Jacobs AR, et al. Interictal EEG discharges, reproductive hormones and menstrual disorders in epilepsy. Ann Neurol 2003;54:625–37. [9] Dean A, Solomon G, Harden C, et al. Left-hemisphere dominance of epileptiform discharges. Epilepsia 1997;38:503–5.
191
[10] Labar D, Dilone L, Solomon G, et al. Epileptogenesis: left or right hemisphere dominance? Preliminary findings in hospital-based population. Seizure 2001;10:570–2. [11] Kalinin VV, Zheleznova EV. Interface between endocrinology and psychiatry in women with epilepsy: analysis of relationship with comorbid pathology. Epilepsy Behav 2006;8:192–200. ¨ ber die Entstehung und den Stammenbaum des [12] Haeckel E. U Menschengeschlechts. Berlin; 1870; [In German]. [13] Geodakian VA. Asynchronous asymmetry (sexual and lateral differentiation as a consequence of asynchronous evolution). J High Nervous Act 1993;43:543–61. [In Russian]. [14] Crow TJ. Introduction. In: Crow TJ, editor. The speciation of modern Homo sapiens. London: The British Academy, Oxford Univ. Press; 2002. p. 1–20. [15] Corballis MC. Laterality and human speciation. In: Crow TJ, editor. The speciation of modern Homo sapiens. London: The British Academy, Oxford Univ. Press; 2002. p. 137–52. [16] Herzog AG, Klein P, Ransil BJ. Three patterns of catamenial epilepsy. Epilepsia 1997;38:1082–8. [17] Sisodiya SM. Mechanisms of drug resistance in epilepsy. In: Shorvon S, Perucca E, Fish D, Dodson E, editors. The treatment of epilepsy. Ames, IA: Blackwell; 2004. p. 84–8. [18] Michelson D, Gold PW. Laboratory evaluation of neuroendocrine systems. In: Wolkowitz OM, Rothschild AJ, editors. Psychoneuroendocrinology. The scientific basis of clinical practice. Arlington, VA: American Psychiatric Press; 2003. p. 469–88. [19] Herzog AG, Seibel MM, Schomer DL, et al. Reproductive endocrine disorders in women with partial seizures of temporal lobe origin. Arch Neurol 1986;43:341–6. [20] Harden CL. Polycystic ovaries and polycystic ovary syndrome in epilepsy: evidence for neurogonadal disease. Epilepsy Curr 2005;5(4):142–6. [21] Duncan S. Polycystic ovarian syndrome in women with epilepsy: a review. Epilepsia 2001;42(Suppl. 3):60–5.