Seizure: European Journal of Epilepsy
Volume 19, Issue 6 , Pages 359-362, July 2010

Occipital lobe seizures related to marked elevation of hemoglobin A1C: Report of two cases

  • Wan-Ling Hung

      Affiliations

    • Division of Neurology, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan, ROC
    • ,
  • Peiyuan F. Hsieh

      Affiliations

    • Division of Neurology, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan, ROC
    • School of Medicine, National Yang-Ming University, Taipei, Taiwan, ROC
    • Graduate Institute of Biomedicine and Biomedical Technology, National Chi Nan University, Nantou Hsien, Taiwan, ROC
    • Corresponding Author InformationCorresponding author at: Division of Neurology, Department of Internal Medicine, Taichung Veterans General Hospital, 160, Sec. 3, Taichung-Kang Rd., Taichung 407, Taiwan, ROC. Tel.: +886 4 2359 2525x3021; fax: +886 4 2358 4403.
    • ,
  • Yi-Chung Lee

      Affiliations

    • Division of Neurology, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan, ROC
    • School of Medicine, National Yang-Ming University, Taipei, Taiwan, ROC
    • ,
  • Ming-Hong Chang

      Affiliations

    • Division of Neurology, Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan, ROC
    • School of Medicine, National Yang-Ming University, Taipei, Taiwan, ROC

Received 12 July 2009; received in revised form 26 January 2010; accepted 20 May 2010. published online 17 June 2010.

Article Outline

Abstract 

Occipital lobe seizures caused by nonketotic hyperglycemia (NKH) have been reported in only a few cases and are not fully characterized. We report two cases of NKH-related occipital lobe seizures with high hemoglobin A1C (HbA1C), epileptiform electroencephalograph (EEG) and MRI abnormalities. Both patients had moderate hyperglycemia (310–372mg/dl) and mildly elevated serum osmolarity (295–304mOsm/kg) but markedly elevated HbA1C (13.8–14.4%). One patient had a clinico-EEG seizure originating from the right occipital region during sleep. The other patient had an interictal epileptiform discharge consisting of unilateral occipital beta activity in sleep. None of the previously reported cases fulfilled the criteria of a nonketotic hyperglycemic hyperosmolar (NKHH) state, or showed any interictal beta paroxysms, spikes, sharp waves, or spike/sharp-slow wave complexes. We suggest that prolonged exposure to uncontrolled hyperglycemia, as indicated by HbA1C, rather than an acute NKHH state is crucial in the development of this peculiar seizure. We also suggest clinicians look for the presence of interictal focal beta paroxysms in addition to the usual epileptiform discharges while reading the EEG of these patients.

Keywords: Hyperglycemia, Occipital lobe seizure, HbA1C, Electroencephalograph, Magnetic resonance imaging

 

Back to Article Outline

1. Introduction 

Nonketotic hyperglycemic hyperosmolar state (NKHH) is a clinical syndrome of severe hyperglycemia (>600mg/dl), hyperosmolarity (>330mOsm/l), and intracellular dehydration without ketoacidosis.1 The term nonketotic hyperglycemia (NKH) is often used to include situations in which an NKHH state is not strictly fulfilled. Seizures are a common manifestation in patients with NKH.2 Most are partial motor seizures with or without secondary generalization.3 Occipital lobe seizures caused by NKH have been reported in only a few cases.4, 5, 6, 7, 8

It is likely that prolonged exposure to uncontrolled hyperglycemia rather than an acute NKHH state plays an important role in the development of occipital lobe seizures. We present the HbA1C, EEG and MRI abnormalities of two patients with NKH-related occipital lobe seizures and compare the biochemical and EEG findings of our patients with nine previously reported cases.4, 5, 6, 7, 8

Back to Article Outline

2. Case report 

2.1. Patient 1 

A 30-year-old previously healthy man began to episodically see green-colored flashing lights in the left-sided visual field followed by gaze and head deviation to the left with or without fine convulsion over all four limbs, 1 week prior to admission. Each episode lasted for around 1min and occurred many times a day. His consciousness was clear throughout the whole period. He visited a local hospital where mild hyperglycemia (200–300mg/dl) was noted.

On the day of admission, he was afebrile, normotensive, alert, and fully orientated. A neurologic examination revealed partial left homonymous hemianopsia that was further confirmed by automated perimetry. On hospital day 2, he developed illusion with distortion of left-sided images. In the following days, he had frequent attacks and fencing posturing during one episode.

Hyperglycemia (372mg/dl) with calculated osmolarity 304mOsm/kg and without ketoacidosis was noted on admission. Hemoglobin A1C (HbA1C) was 13.8%. Other laboratory data, including hemogram, electrolytes, liver function, renal function, thyroid function, C-reactive protein, and cerebrospinal fluid analysis were normal. A genetic screen for hot spots of point mutation of MELAS (mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes) was negative. Electroencephalography (EEG) performed on hospital day 2 showed a clinico-EEG seizure originating from the right occipital region during drug (chloral hydrate)-induced light sleep (Fig. 1A). Brain MRI performed on hospital day 3 showed increasing thickness and signal intensity over the cortex of the right occipital lobe and right mesial temporal lobe and underlying subcortical hypointensity in the T2-weighted image (Fig. 1B). Tc-99m hexamethylpropylene-amine-oxime (HMPAO) single-photon emission computed tomography (SPECT) performed on hospital day 5 revealed hyperperfusion in the right occipital lobe (Fig. 1C).

  • View full-size image.
  • Fig. 1. 

    EEG, brain MRI, and SPECT of patient 1. (A) EEG on hospital day 2. A seizure starts as right occipital β activity of gradually increasing amplitude and decreasing frequency (arrow, upper segment). The end of the seizure is shown in the lower segment. The clinico-EEG seizure was recorded during drug-induced light sleep. (B) Brain MRI on hospital day 3, with T1-weighted image, T2-weighted image, and fluid-attenuated inversion-recovery image (FLAIR) in sequence. There are increased thickness and signal intensity over the cortex of the right occipital lobe and right mesial temporal lobe as well as underlying subcortical hypointensity in the T2-weighted image. (C) Tc-99m HMPAO SPECT showing hyperperfusion in the right occipital lobe on hospital day 5.

Phenytoin was given on hospital day 1. A continuous infusion of midazolam was administered from hospital day 3 to day 4 for frequent seizures. Hyperglycemia subsided with oral antidiabetic agents. The seizures subsided gradually and ceased after 5 hospital days. The visual disturbances, including left homonymous hemianopsia, also recovered after 7 hospital days.

2.2. Patient 2 

A 52-year-old woman presented with hypertension without regular control. She presented at our hospital due to episodic visual hallucinations for 2 weeks with increasing frequency. She saw rotating flowers in the air or people gathered in groups in the left-sided visual field. Each episode lasted 2–3min and the frequency was greater than 10 times a day.

On the day of admission, she was afebrile but hypertensive (182/91mmHg). A neurologic examination was normal.

Hyperglycemia (310mg/dl) with calculated osmolarity 295mOsm/kg and without ketoacidosis was noted on admission. HbA1C was 14.4%. Other laboratory data, including hemogram, electrolytes, liver function, renal function, C-reactive protein, and cerebrospinal fluid analysis were normal. EEG performed on hospital day 1 showed an interictal right occipital epileptiform discharge consisting of a two-second's medium-voltage beta paroxysm of 15–16Hz during chloral hydrate-induced light sleep (Fig. 2A, lower segment). Brain MRI performed on hospital day 2 revealed a relatively low signal in the T2-weighted image over the subcortical area of the right occipital lobe (Fig. 2B).

  • View full-size image.
  • Fig. 2. 

    EEG and brain MRI of patient 2. (A) EEG on hospital day 1. The upper segment shows interictal normal awake EEG. The lower segment recorded during drug-induced light sleep shows an interictal epileptiform discharge consisting of a medium-voltage beta activity for two seconds in the right occipital region (arrow). (B) Brain MRI on hospital day 2, with T1-weighted image, T2-weighted image, and FLAIR in sequence. There is a relatively low signal over the subcortical area of the right occipital lobe in the T2-weighted image.

Phenytoin was given on hospital day 4. Hyperglycemia subsided with oral antidiabetic agents and the visual hallucinations ceased after 4 hospital days.

Back to Article Outline

3. Discussion 

Both of our cases were in patients with newly diagnosed diabetes with markedly elevated HbA1C (13.8% and 14.4%, respectively). They had only moderate hyperglycemia at diagnosis, thus NKH-related occipital seizures were not among our initial working diagnoses. Other common etiologies of occipital seizures including encephalitis, tumor, cortical malformation, vascular event, mitochondrial disorders, and reversible posterior leukoencephalopathy were excluded by imaging, cerebrospinal fluid analysis, and a genetic study. In addition, both cases had focal subcortical T2 hypointensity in their brain MRI, which is thought to be a characteristic finding of NKH-related seizures.4, 9 We made a diagnosis of occipital seizures related to prolonged moderate hyperglycemia as there was markedly elevated HbA1C in conjunction with characteristic findings in brain imaging.

These two cases suggest that an NKHH state is not necessary for development of occipital seizures. We reviewed 9 previously reported cases (Table 1). None fulfilled the criteria of NKHH. All of the cases had only mildly elevated blood osmolarity. Serum sodium was below 130mg/dl in only two cases. Our patients had relatively lower blood glucose levels compared with those of other patients. HbA1C was mentioned in only case 5 (9.4%) and our cases. Eight of 11 cases had newly diagnosed diabetes mellitus. Case 5 had had diabetes mellitus for 1 year, but had not taken any hypoglycemic agent. Our two cases were newly diagnosed diabetes mellitus patients with markedly elevated HbA1C. Most of the other reported cases also had undiagnosed diabetes mellitus before seizures. Our patients and the previously reported cases suggest that prolonged uncontrolled hyperglycemia may play a more important role than acute extreme hyperglycemia in the development of occipital seizures. Our observation is in accordance with a recent study showing that diabetic patients with poor glycemic control (HbA1c>9%) had a significantly higher risk of seizure recurrence and clustering.10 However, the patients in that study had seizure types other than occipital seizures and more than half had old brain insults.

Table 1. Case reports of hyperglycemia-related occipital seizures or visual symptoms.
CaseAge/genderPresentationsBlood glucose (mg/dl)Serum Na/K (mEquiv./l)Serum osmolarity (mOsm/kg)EEG Abn.T2WI Abn.
1a28/MEpisodic flashing red and green lights in the left visual field, left homonymous hemianopsia371136/4.7306END
2a67/FBlurring of the left visual field, with intermittent green and red flashing lights in that field452131/5.4310END
3a59/MStereotyped head deviation to the left with leftward eye version, illusions484135/4.4311SND
4a43/FFlashing lights after gazing left, left homonymous hemianopsia499NA308S–
559/FComplex visual hallucinations and illusions535NA309E+
6a39/MBlurred vision, progressive left hemianopsia, left visual field hallucinations503129/4.1313S+
754/NAIntermittent blurred vision, visual hallucinations, staring spells, generalized seizure426125/3.5282E+
8a69/NABlurred vision and visual hallucinations in the right field487132/3.2305E+
942/FVisual hallucinations, right homonymous hemianopsia, complex partial seizures324142/4.0317E+
10a30/MFlashing lights in the left visual field, head and eyes turning to the left, left homonymous hemianopsia, illusions372136/3.7304E+
11a52/FVisual hallucinations310134/3.7295e+

Abbreviations: Abn, abnormality; E, ictal epileptiform discharges; S, focal slow waves; e, interictal epileptiform discharges; T2WI Abn, focal subcortical hypointensity on T2-weighted image; NA, not available; ND, not done. Serum osmolarity: [2(Na+K)]+[BUN/2.8]+[glucose/18]=mOsm/kg. Cases 1–3 [8], case 4 [7], case 5 [6], cases 6–8 [5], case 9 [4], cases 10–11: our patients.

aNewly diagnosed diabetes mellitus.

Singh and Strobos reported 21 patients with NKH-related continuous partial motor seizures (epilepsia partialis continua). However, there was no HbA1C data and most of their cases were before the era of computed tomography (CT). Thirteen patients had considerable evidence of a structural lesion on the appropriate side of the brain. Three patients had blood glucose less than 400mg/dl, serum sodium ranging from 130 to 134mEquiv./l, and serum osmolarity 293–296mOsm/kg. Two of these 3 patients had evidence of a structural lesion on the appropriate side of the brain, and the third patient did not have an EEG, CT, radionuclide scan, or angiogram examination.11 Their data cannot be extrapolated to the clinical settings of our patients because they had different types of seizures, evidence of structural lesions or incomplete laboratory studies.

The role of hyperosmolarity and hyponatremia in NKH-related seizures is inconsistent between studies.3, 11 An experimental study suggested that glucose itself is a pro-convulsant in diabetes mellitus.12 Previous studies have also disclosed seizure susceptibility even in only moderate degrees of hyperglycemia.3, 11 Our first case showed symptoms while his blood glucose was 200–300mg/dl. Further studies are needed to clarify how prolonged hyperglycemia contributes to seizures and in whom hyperglycemia provokes seizures.

As regards to interictal EEG, none of previously reported patients showed interictal beta paroxysms, spikes, sharp waves, or spike/sharp-slow wave complexes. Instead, our second patient had an interictal occipital epileptiform discharge consisting of a medium-voltage beta paroxysm of 15–16Hz. The paroxysmal focal beta activity is unlikely to be a marker of focal brain abnormality, which usually presents as continuous or intermittent focal slow waves. More clinical experience is required to answer the question whether paroxysmal beta activity is the predominant interictal epileptiform pattern in the patients with NKH-related occipital seizures.

From Table 1, we also found that all cases with brain MRI except case 4 had characteristic focal subcortical T2 hypointensity. The mechanism contributing to transient focal subcortical T2 hypointensity remains uncertain. Three mechanisms have been proposed. First, it may result from an accumulation of free radicals or iron related to a hyperglycemia-induced hypoxic–ischemic state.9 This theory of iron deposition goes against an animal study which showed iron deposition only in the cortex,13 the fact that iron deposition was not reversible, and a normal T1 image. Second, it may be a transient physiological response to the seizures due to complete resolution of the changes.4 This is not a typical finding with partial seizures (typical findings include hyperintensity in gray and subcortical white matter on T2 images attributed to edema or gliosis). Third, this may be due to intracellular dehydration of glial and supporting tissue.5

Phenytoin has deleterious effects on glucose homeostasis.14 Therefore, it is not a good choice in hyperglycemia-related seizures. In accordance with the literature, our first patient still had frequent seizures with phenytoin treatment during the first few days of hospitalization.

In conclusion, occipital lobe seizures can be the first manifestation of diabetes mellitus. Hyperglycemia, even moderately elevated serum glucose levels, should be considered as the cause of occipital lobe seizures, especially in conjunction with markedly elevated HbA1C and characteristic focal subcortical T2 hypointensity. We also suggest that while reading the EEG of these patients, clinicians should pay more attention to not only usual epileptiform discharges but also interictal focal beta paroxysms.

Back to Article Outline

References 

  1. Venkatraman R, Singhi SC. Hyperglycemic hyperosmolar nonketotic syndrome. Indian J Pediatr. 2006;73:55–60
  2. Stahlman GC, Auerbach PS, Strickland WG. Neurological manifestations of non-ketotic hyperglycemia. J Tennessee Med Assoc. 1988;81:77–80
  3. Tiamkao S, Pratipanawatr T, Tiamkao S, Nitinavakarn B, Chotmongkol V, Jitpimolmard S. Seizures in nonketotic hyperglycaemia. Seizure. 2003;12:409–410
  4. Raghavendra S, Ashalatha R, Thomas SV, Kesavadas C. Focal neuronal loss, reversible subcortical focal T2 hypointensity in seizures with a nonketotic hyperglycemic hyperosmolar state. Neuroradiology. 2007;49:299–305
  5. Lavin PJ. Hyperglycemic hemianopsia: a reversible complication of nonketotic hyperglycemia. Neurology. 2005;65:616–619
  6. Wang CP, Hsieh PF, Chen CC, Lin WY, Hu WH, Yang DY, et al. Hyperglycemia with occipital seizures: images and visual evoked potentials. Epilepsia. 2005;46:1140–1144
  7. Duncan MB, Jabbari B, Rosenberg ML. Gaze-evoked visual seizures in nonketotic hyperglycemia. Epilepsia. 1991;32:221–224
  8. Harden CL, Rosenbaum DH, Daras M. Hyperglycemia presenting with occipital seizures. Epilepsia. 1991;32:215–220
  9. Seo DW, Na DG, Na DL, Moon SY, Hong SB. Subcortical hypointensity in partial status epilepticus associated with nonketotic hyperglycemia. J Neuroimaging. 2003;13:259–263
  10. Huang CW, Tsai JJ, Ou HY, Wang ST, Cheng JT, Wu SN, et al. Diabetic hyperglycemia is associated with the severity of epileptic seizures in adults. Epilepsy Res. 2008;79:71–77
  11. Singh BM, Strobos RJ. Epilepsia, partialis continua associated with nonketotic hyperglycemia: clinical and biochemical profile of 21 patients. Ann Neurol. 1980;8:155–160
  12. Schwechter EM, Veliskova J, Velisek L. Correlation between extracellular glucose and seizure susceptibility in adult rats. Ann Neurol. 2003;53:91–101
  13. Palmer C, Menzies SL, Roberts RL, Pavlick G, Connor JR. Changes in iron histochemistry after hypoxic–ischemic brain injury in the neonate rat. J Neurosci Res. 1999;56:60–71
  14. Banner W, Johnson DG, Walson PD, Jung D . Effects of single large doses of phenytoin on glucose homeostasis—a preliminary. J Clin Pharmacol. 1982;22:79–81

PII: S1059-1311(10)00110-X

doi:10.1016/j.seizure.2010.05.006

Seizure: European Journal of Epilepsy
Volume 19, Issue 6 , Pages 359-362, July 2010

Article Tools

* Image Usage & Resolution