Effects of polydipsia–hyponatremia on seizures in patients with epilepsy


Mitsutoshi Okazaki, MD, Department of Psychiatry, National Center of Neurology and Psychiatry, Musashi Hospital, 4-1-1 Ogawahigashi-cho, Kodaira, Tokyo 187-8551, Japan. Email: okazakim@ncnp.go.jp


Abstract  Aggravation of seizures due to hyponatremia was investigated in five patients with epilepsy and polydipsia–hyponatremia. They experienced marked increases in the frequency of their complex partial seizures with a decrease in the serum sodium level to 118–127 mEq/L. In all cases, the serum sodium level returned to normal through restriction of fluids, and the clinical seizures improved. All patients had shown intellectual impairment and/or psychotic episodes, and all had been given antipsychotics. Hyponatremia caused by polydipsia appears to be a risk factor for aggravation of habitual seizures in patients with epilepsy.


Psychogenic polydipsia refers to excessive fluid intake without an identifiable underlying medical cause. Sudden and/or severe hyponatremia following polydipsia often causes various neurologic and psychiatric conditions such as headache, vomiting, lethargy, psychosis, seizures and even death. There have been many reports of polydipsia–hyponatremia, referred to as ‘water intoxication’, mainly in chronic psychiatric patients;1–6 however, there have been no reports of the effect of polydipsia–hyponatremia on seizures in patients with epilepsy. Here, we report five cases of epilepsy with polydipsia–hyponatremia.


We encountered five patients (four males and one female) in whom there appeared to be a close relation between their polydipsia–hyponatremia and aggravation of epileptic seizures. The clinical profile of each patient is shown in the Table 1. In all patients, a remarkable degree of polydipsia was confirmed by the medical staff. Their complex partial seizures were aggravated when the serum sodium level decreased to 118–127 mEq/L. When the serum sodium level returned to 130–140 mEq/L through restriction of fluids, the clinical seizures improved. No patient had a family history of epilepsy or any other neuropsychiatric disorder. There was no liver, heart, kidney, adrenal, or thyroid disease as a possible cause of the polydipsia-hyponatremia. The onset of seizures and diagnosis of epilepsy preceded the episode of polydipsia by more than 15 years. Carbamazepine (CBZ) had been given to four of the patients (patients 1, 2, 4 and 5). Four patients (patients 1–3 and 5) had shown intellectual impairment and four (patients 2–5) had suffered from psychotic episodes, and antipsychotics had been administered to all patients. The chlorpromazine equivalent doses were 150–2250 (mean 1090) mg/day.

Table 1.  Clinical summary of patients
Pt noAge (years)SexSz onset (years)EtiologySz typeInterictal EEGCT/MRI findingsAEDsAntipsychoticsSerum sodium level (mEq/L)
During polydipsiaAfter fluid restriction
  • † 

    Data were obtained immediately after increased CPS in each patient.

  • Pt no, patient number; M, male; F, female; Sz, seizure; CPS, complex partial seizure; GTC, generalized tonic-clonic seizure; L, left; R, right; Bil, bilateral; AED, antiepileptic drug; PHT, phenytoin; CBZ, carbamazepine; CLB, clobazam; VPA, sodium valproate; PP, propericyazine; HPD, haloperidol; LP, levomepromazine; RIS, risperidone; QUE, quetiapine.

139M15Closed head injury
(traffic accident)
L fronto-temporal sharp wavesL hippocampal sclerosis
R frontal traumatic injury
R temporal spikeDiffuse cerebral atrophyPHT
L temporal sharp wave slow waveBil hippocampal atrophyPHTHPD
439M12EpendymomaCPSL temporal sharp wave slow waveL medial temporal mass with calcificationCBZ
No definite
Epileptic discharge
R hippocampal atrophyPHT

Each patient gave us a documentary informed consent to be quoted as a case in this study.


Hyponatremia caused by polydipsia (i.e. water intoxication) has often been reported, especially in psychiatric patients,1–6 but the symptomatology of water intoxication is not well known in patients with a clinical history of epilepsy. To our knowledge, this is the first reported evaluation of polydipsia–hyponatremia-induced aggravation of seizures in patients with epilepsy.

Our patients with symptomatic partial epilepsy showed an increased frequency of complex partial seizures as an initial symptom of the hyponatremia, while reported patients with water intoxication often presented with generalized seizures.1,4 The mechanism of the increase in seizure frequency would most likely be the cerebral edema that emerges rapidly following water loading.7–9 Some studies have shown that the severity of symptoms of hyponatremia is related to the rate of decrease in the serum sodium level, particularly if the decrease is >0.5 mEq/L/h, as well as to the absolute serum sodium concentration.4,10 Four of our five patients experienced an increased number of seizures even when their serum sodium level was mildly decreased (around 125 mEq/L). We suggest that the aggravation of seizures caused by mild hyponatremia in our patients depended on the combination of rapid cerebral edema and a coarse brain lesion (closed head injury, brain tumor or postencephalitic sequlae).

It has been reported that CBZ can lead to hyponatremia in patients with epilepsy and psychiatric disorders and that CBZ-induced hyponatremia is asymptomatic when the serum sodium level is >125 mEq/L.11 It is possible that CBZ promoted hyponatremia in our patients, but the influence of CBZ alone does not explain the polydipsia.

Antipsychotics are thought to contribute both to the risk of polydipsia2–4,6 and to seizure aggravation.12 The type of antipsychotic (for instance, serotonin-dopamine antagonist (SDA) or non-SDA) might have had some influence on the clinical course in our patients, depending on the pharmacological action of the drug. We did not find an intimate relation between specific antipsychotics and the event reported here. However, the question cannot be fully resolved in our small study group.

It is also a matter of concern whether polydipsia is a risk factor for seizure aggravation in all types of epilepsy. Again, because our study group is small, we do not have sufficient information from which to draw a conclusion. Seizure aggravation could occur in relation to risk factors such as pronounced polydipsia, refractory seizures, and the regimen of antiepileptic drugs and antipsychotics.

In conclusion, polydipsia can be a risk factor for aggravation of habitual seizures in patients with epilepsy. In our patients, fluid restriction proved extremely useful, not only for improving the serum sodium level but also for decreasing the number of seizures. Polydipsia can be overlooked or not treated appropriately in patients with epilepsy because it is less commonly known than that in other psychiatric patients treated with antipsychotics. The recognition and proper management of polydipsia–hyponatremia in patients with epilepsy are important in controlling seizures.