|
|
|
Diabetes and Development of Tardive Dyskinesia
Margaret G. Woerner, Ph.D., Bruce L. Saltz, M.D., John M. Kane, M.D., Jeffrey A. Lieberman, M.D., and Jose M.J. Alvir, Dr.P.H.
(American Journal of Psychiatry, 1993; Volume 150: pages 966-968)
Diabetes mellitus has been identified as a possible risk factor for tardive dyskinesia. The authors examined 160 elderly individuals who were beginning neuroleptics treatment; 24 had diabetes and 136 did not. After 43 weeks of neuroleptic exposure, the cumulative incidence rates of tardive dyskinesia were 54.1% (SE=5.6%) for the diabetics and 25.6% (SE=16.1%) for the nondiabetics. (American Journal of Psychiatry 1993; 150:966-968)
Increasing age has been the most consistently identified risk factor for tardive dyskinesias (1-3). Preliminary data from an ongoing prospective study of the development of tardive dyskinesia (4) showed the cumulative incidence of tardive dyskinesia after 43 weeks of cumulative neuroleptic exposure in individuals over age 55 to be 31% (SE=11%), six times as high as that seen in younger subjects. Although the prevalence of spontaneous (non-neuroleptic-induced) dyskinesias found in this group (5%) was slightly higher than the reported rates of 1%-3% in younger subjects, this would not account for the greater incidence of abnormal movements. Variables that may contribute to higher risk in the elderly include altered drug metabolism and increased vulnerability of the striatonigral dopamine system to perturbation.
Two investigations (5, 6) have suggested diabetes mellitus as a risk factor for development of tardive dyskinesia; this may contribute to the high rates seen in the elderly. Mukherjee et al. (5) reported diabetes mellitus rates considerably higher than the expected population prevalence in two groups of patients with persistent tardive dyskinesias. In addition, the fasting blood sugar levels for 23 patients with persistent tardive dyskinesias were significantly higher than those for 33 patients without tardive dyskinesias.
Ganzini et al. (6) evaluated 38 neuroleptic- treated diabetic psychiatric patients and equivalent numbers of matched comparison subjects (neuroleptic-treated non-diabetics and neuroleptic-naïve diabetics) for abnormal involuntary movements. Abnormal movements were observed in significantly more neuroleptic-treated diabetics (79%) than neuroleptic-treated nondiabetics (53%). The higher prevalence of tardive dyskinesia was not accounted for by other potential risk factors. However, the prevelance of abnormal movements resembling tardive dyskinesia in the group of neuroleptic-naïve diabetics was 21%. This unusually high rate of spontaneous dyskinesia could account for the excess tardive dyskinesia found in neuroleptic-treated diabetics over the rate for neuroleptic-treated comparison subjects.
We investigated this issue further by examining the medical histories of subjects participating in an ongoing prospective study of the development of tardive dyskinesia in elderly patients undergoing neuroleptic therapy for the first time (4), and we compared the rates of tardive dyskinesia in the diabetic and nondiabetic subjects.
Received March 9, 1992; revision received Sept. 8, 1992; accepted Sept. 28, 1992. From the Department of Psychiatry, Hillside Hospital, Long Island Jewish Medical Center, and the Department of Psychiatry, Albert Einstein College of Medicine, Bronx, N.Y. Address reprint requests to Dr.Woerner, Hillside Hospital, Glen Oaks, NY 11004. The authors thank David Ryan, Melissa Smith, and Linda Pestreich for their participation in the data collection. Supported by NIMH grants MH-40015, MH-32369, and MH-41960 and contract 278-81-0032. Copyright 1993 American Psychiatric Association.
METHOD
The method and preliminary results of the ongoing parent study have been described in detail elsewhere (4). Neuroleptic-naïve patients aged 55years and over were evaluated at the time of initiation of drug treatment, weekly for the first month, and at 3-month intervals thereafter. Tardive dyskinesia was rated by trained examiners using a standard rating scale. Extrapyramidal symptoms, psychopathology, and cognitive functioning were also assessed. For this study we analyzed data on the first 160 patients on the data base who had been followed for at least 3 weeks. Follow-up ranged from 3 to 119 weeks; the number of patients who were examined at 4, 13, 26, and 52 weeks, respectively, were 141, 106, 82, and 46. Individuals with chart diagnosis of diabetes mellitus were identified, and those receiving oral hypoglycemic or insulin medication were categorized as diabetes positive for the purpose of this analysis.
Life table analysis using the Kaplan-Meier method was used to estimate survival curves for the diabetics and the nondiabetics. The equality of these curves was tested by nonparametric Mantel-Cox and Breslow rank tests. Cox proportional hazards regression analysis was used to test for the independent effect of diabetic status on the risk of tardive dyskinesia when covariates (age, gender, race, etc.) were controlled for (7, 8).
RESULTS
Of the 160 subjects, 24 were categorized as having diabetes and 136 were not. The mean ages of the diabetic and nondiabetic subjects were 76.3 (SD=8.4) and 77.2 (SD=9.1) years, respectively, and their mean numbers of concurrent medical illnesses were 2.4 (SD=2.0) and 2.8 (SD=1.6), respectively. The mean Mini-Mental State score of the diabetic subjects was 18.5 (SD=6.0), and the score for the nondiabetics was 19.1 (SD=7.4). Table 1 presents other characteristics of the two groups. There were no significant differences in baseline characteristics between the two groups except in the numbers of nonpsychotropic medications (excluding medications for diabetes): the mean number was 4.7 (SD=2.2) for the diabetics and 3.6 (SD=2.2) for the nondiabetics (t=2.05, df=130, p<0.05; 95% confidence interval for difference=0.04-2.28). The mean number of psychotropic medications was 1.8 (SD=0.8) for the nondiabetics.
Eight diabetics and 20 nondiabetics developed tardive dyskinesias during follow-up. Life table analyses indicated greater vulnerability to tardive dyskinesia among the diabetics than among the nondiabetics (Mantel-Cox statistics=3.20, df=1, p=0.07). After 43 weeks of cumulative drug exposure, the cumulative incidence rates were 54.1% (SE=5.6%) for the subjects with diabetes and 25.6% (SE=16.1%) for the subjects without diabetes. The hazard ratio for diabetic status was constant over time. When age and gender were controlled statistically in a Cox proportional hazards model, the association between diabetes and tardive dyskinesia was stronger; the risk ratio of 2.38 (95% confidence interval=1.04-5.46) indicated a significantly greater risk of tardive dyskinesia among the diabetics. Controlling for race, psychiatric diagnosis, organic mental syndrome diagnosis, extrapyramidal symptoms at baseline and within the first month, history of ECT, and number of nonpsychotropic medications and did not alter the effect of diabetic status.
The neuroleptic dose in milligrams of chlorpromazine equivalents at study was slightly but not significantly lower for the diabetics (means=99.7, SD=119.0) than for the nondiabetics (means=125.8, SD=215.3) (95% confidence interval for difference=87.2-35.1). Controlling for starting dose did not alter the effect of diabetic status on risk of tardive dyskinesia. Nonsignificantly more diabetics (N=13, 54%) than nondiabetics (N=50, 37%) were not taking neuroleptics at the time of tardive dyskinesia diagnosis or at the last follow-up, raising the possibility that the higher rate of tardive dyskinesia among the diabetics was due to a greater likelihood of detecting covert dyskinesias. However, limiting the analysis to the 97 patients who were taking neuroleptics at the time of tardive dyskinesia diagnosis or at last follow-up resulted in an even greater risk ratio for diabetic status (risk=4.15; 95% confidence interval=1.04-16.50).
DISCUSSION
Our finding of greater vulnerability to tardive dyskinesias among elderly neuroleptic-treated diabetics than among nondiabetics is consistent with the reports of Mukherjee et al. (5) and of Ganzini et al. (6) and suggest that diabetes may contribute to the higher rates of tardive dyskinesias among the elderly. This result does not appear to be accounted for by other potential tardive dyskinesia risk factors. The high prevalence of spontaneous dyskinesia (21%) found among the non neuroleptic-treated diabetic comparison group in the Ganzini et al. study raises the possibility that diabetes may not be a risk factor for tardive dyskinesia per se, but rather for spontaneous abnormal involuntary movements resembling tardive dyskinesias. Our findings, based on a prospective study, make this alternative hypothesis less tenable. Our subjects were free of abnormal movements before neuroleptic treatment, but the diabetes was present at baseline for the diabetic group.
We do not know what it is about diabetes or just treatment that increases the susceptibility to tardive dyskinesia of older individuals. In neuroleptic-treated animals the presence of diabetes may be more likely to impair straital dopamine transmission and produce striatal dopamine receptor hypersensitivity, with increased tardive dyskinesia as a possible consequence (9). Alternatively, microvascular (10) and macrovascular complications of hyperglycemia may cause decreased efficiency of blood-brain burrier, enhancing susceptibility of the brain to neurotoxins. Finally, since diabetes is a risk factor for cerebrovascular disease and basal ganglia infarction may be a cause of dykinesia, ischemia may be the mechanism for the connection between diabetes and abnormal movements (11, 12). Whether there is a specific interaction between diabetes and neuroleptic toxicity that may enhance the vulnerability of diabetics to tardive dykinesia is unknown.
REFERENCES
1. Kane JM, Smith J: Tardive dyskinesia: prevalence and risk factors. Arch Gen Psychiatry 1982; 39: 473-481
2. Kane JM, Woerner MG, Lieberman JA, Kinon BJ: Tardive dykinesia and drugs, Drug Development Research. Edited by Cornfeldt ML, Shutske GM. New York, Alan R Liss, 1986
3. Woerner MG, Kane JM, Lieberman JA, Alvir J, Bergmann KJ, Borenstein M, Schooler NR, Mukherjee S, Rotrosen J, Rubinstein M, Basavaraju N: The prevalence of tardive dyskinesia. J Clin Psychopharmacol 1991; 11:34-42
4. Saltz BL, MG, Kane JM, Lieberman J, Alvir J, Bergmann KJ, Blank K, Koblenzer J, Kahaner K: prospective study of tardive dyskinesia incidence in the elderly. JAMA 1991; 266: 2402-2406
5. Mukherjee S, Wisniewski A, Bilder R, Sackheim HA: possible association between tardive dyskinesia and altered carbohydrate metabolism (letter). Arch Gen Psychiatry 1985; 42:205
6. Ganzini L, Heintz RT, Hoffman WF, Casey DE: The prevalence of tardive dyskinesia in neuroleptic treated diabetics. Arch Gen Psychiatry 1991;48:259-263
7. Benedetti G, Yuen K, Young L: Life tables and survival functions, in BMPD Statistical Software. Edited by Dixon WJ. Berkeley, University of California Press, 1985
8. Hopkins A: P2L-survival analysis with covariates-Cox models. Ibid
DIABETES/ TARDIVE DYSKINESIA
9. Saller C, Chido L: Glucose suppresses basal firing and haloperidol induced increases in the firing rate or central dopaminergic neurons. Science 1980; 210:1269-1271
10. McMillan D: Pathophysiology diabetic macro and microvascular disease, in Diabetes Mellitus: Theory and Practice, 3rd ed. Edited by Ellenberg M, Rifken H. New Hyde Park, NY, Medical Examination, 1983
11. Kase C, Maulsby G, DeJaun E, Mohr JP: Hemichorea-hemiballism and lacunar infarction in the basal ganglia. Neurology 1951;31:452-455
12. Saris S: Chorea caused by caudate infarction. Arch Neurol 1983; 40:590-591
back to list
|
|
|
|
|
