Revisiting Thyroid Hormones in Schizophrenia

Table 1

Literature review of studies on thyroid hormone levels in schizophrenic patients.

Serum analysis							Other parameters
TT4	FT4	TT3	FT3	rT3	TSH	Antiperoxidase, Tab, or TBG	Number of patients and controls	Diagnosis criteria
No significant difference between controls and drug-free. Decreased after chlorpromazine, trifluoperazine, or clozapine treatments	No significant difference between controls and drug-free. Decreased after chlorpromazine and clozapine treatments						41 drug-free (for at least 3 weeks) SZ female (28) and SZ male (13) patients; 24/41 treated (6 weeks), random attribution, with chlorpromazine (10), trifluoperazine (9), or clozapine (5)	Feighner criteria 1972 and BPRS
Notes and Conclusions. FT4I calculated from multiplication of serum T4 and radioactive T3 uptake. Calls for investigation of serum TH levels before and after neuroleptic treatments (Rinieris et al., [54]).
					Decreased		10 acutely-ill SZ patients, 10 healthy controls
Notes and Conclusions. Levels of prolactin, and L-thyroxine decreased in SZ patients, whereas dopamine was elevated. Authors conclude that increased dopaminergic activity affects pituitary secretory function, and may result in decreased TSH (Rao et al., [55]).
Basal within normal, posttreatment normal	Basal within normal, posttreatment normal				At 14 h, but not at 24 h, higher after treatment but within normal range		25 males drug-free SZ patients; treated (6w) with chlorpromazine (14) or fluspirilene (11)	Gorham criteria 1962 and DSM-III
Notes and Conclusions. Prolactin values higher (above normal range) posttreatment. Diagnosis of subclinical hypothyroidism in neuroleptic-treated SZ patients. Higher TSH basal level, and augmented TSH response to TRH, compared to pretreatment (Martinos et al., [56]).
75% paranoid-SZ increased when compared with total 4% change (decrease) in the other groups	50% paranoid-SZ increased when compared with total 14% change (decrease) in other groups						29 males: 8 paranoid-SZ, 6 undifferentiated-SZ, 7 bipolar, and 8 major depression (chronic or subchronic patients)	SADS interview
Notes and Conclusions. Hospitalized patients. Measurement at admission and every 2 weeks thereafter (average 4 samples/patient). Analysis of variation within groups. Calls for longitudinal assessments of thyroxine trends even when within normal range levels (Mason et al., [57]).
Overall falling levels during recovery. A subgroup with initial low levels that increased to middle-range values.	Falling levels during recovery. A subgroup with initial low levels that increased to middle-range values.						80 males: 6 undifferentiated-SZ, 9 paranoid-SZ, 18 schizoaffective, 15 bipolar disorder, 16 major depressive, 9 posttraumatic stress disorder, 7 others (chronic or subchronic)	BPRS and DSM III
Notes and Conclusions. Measurement of TT4 and FT4 at admission and at 2-week intervals. Indication that the existence of a change in TT4 and FT4, irrespectively of the direction, seems to be the most important parameter in recovery. No difference between diagnostic groups or treatments (Southwick et al., [58]).
No significant difference between drugnaïve, drug-withdrawn, and controls		No significant difference between drug-naïve and drug-withdrawn; drug-naïve less than controls			No significant difference between drug-naïve, drug-withdrawn, and controls		23 drug-naïve SZ patients, 67 SZ with 2-3 days drug withdrawn, 67 SZ with neuroleptics; 90 controls	Schneider criteria 1987
Notes and Conclusions. Drug-naïve and drug-withdrawn groups combined due to similar hormone levels. Smokers removed only from controls. Dopamine increased in drug-free SZ patients compared to controls. Norepinephrine, epinephrine, and prolactin higher in neuroleptic-treated, compared to drug-free and controls. Agrees with hypothesis of dopaminergic overactivity in schizophrenia (Rao et al., [59]).
Increased on day of hospitalization	Increased on day of hospitalization	Increased on day of hospitalization	Increased on day of hospitalization (negative correlation with BPRS)		Normal	TBG: Normal	15 male SZ and 34 female SZ patients; age-matched controls, 19 males and 34 females	DMS-III, BPRS and Montgomery-Asberg Depression scale
Notes and Conclusions. Measurements on day of hospitalization and after. Levels decreased later. Suggests that increases are due to increased T4 secretion by the thyroid (Roca et al., [60]).
							26-year old SZ male, hospitalized 5 times with normal thyroid function and at a 6th with severe hypothyroidism. Haloperidol treatment.
Notes and Conclusions. Thyroid function normalized with increased doses of medication. Managed hypothyroidism leads to decreased hospitalization. Hypothyroidism as a concurrent illness in SZ (Walch et al., [61]).
Significantly lower in those with low or high TSH	FT4I significantly lower in those with low or high TSH	Low in 23% of patients with normal TSH values	FT3I significantly lower in those with low or high TSH		60% normal, 5% elevated, 17% low	TAb: 20% of total patients (28% SZ female to 13% control; 14% SZ-male to 7% control)	249 patients with chronic SZ (136 males, 113 females; median age 36 years old)
Notes and Conclusions. Spectrum of thyroid function test abnormalities, but interpretation not clear (Othman et al., [33]).
Normal		Mesor (daily mean): drug free lower than controls			Mesor (daily mean): SZ drug-free equal to treated and both lower than controls. Acrophase (daily higher), less than half in both SZ groups compared to control		89 drug-free SZ patients (21 never received drugs, remaining over 3days free), and 25 typical neuroleptic-treated SZ (for at least 5 days); 34 controls	Schneider criteria 1987 and Huber criteria 1987
Notes and Conclusions. Three independent psychiatrists. No age difference in TH measurements. TSH decrease might be caused by the hyperdopaminergic state of these patients as the group reported before. None on Li+; no more information on medication. Suggests involvement of the noradrenergic receptor system (Rao [62]).
Increased in the acute SZ, normalized with perazine with 4-week treatment. Normal in all other SZ patients.		Normal		Normal	Normal		31 acutely ill SZ patients, 19 in remission without drugs, 20 in remission with different drugs, 24 with residual-SZ (negative symptoms); 24 controls	DMS-III revised and BPRS
Notes and Conclusions. Does not differ female from male. The higher T4 the higher the severity of illness, and the better the response to treatment. Concludes that function of T3 is altered as a consequence of neuroleptic therapies (Baumgartner et al., [63]).
	Higher in patients with mild or major syndrome		Higher in patients with mild or major syndrome					BPRS
Notes and Conclusions. No clinical thyroid illness. 34% of patients with thyroid function tests abnormalities, and no correlation found with neuroleptic use. Calls for careful interpretation of thyroid function tests in SZ patients (Sim et al., [28]).
		Higher in RP compared to controls	Higher in RP compared to controls		Higher in NRP compared to RP. Blunted response rate (TRH test) in RP group higher than NRP and control groups.		58 SZ patients, divided into a “remitted (RP)” (30) and “nonremitted (NRP)” (28) groups; 30 healthy controls	Mental deterioration battery (at regular intervals for 1 year)
Notes and Conclusions. Basal prolactin higher in RP, but not NRP, compared to controls. Basal growth hormone (GH) higher in NRP than RP SZ patients. Dexamethosone Suppression Test (DST) nonsuppression higher in RP than NRP and controls. Indicates that higher basal TSH and GH levels may be related with poorer treatment response. Higher TT3 and FT3, and blunted TSH response to TRH and nonsuppression in DST indicate better response in SZ patients (Yazici et al., [64]).
2/22 (9%) baseline abnormal vales			4/22 (18%) baseline abnormal values		4/30 (13%) baseline abnormal values		38 treatment-resistant SZ patients receiving 6-weeks quetiapine, risperid,one or fluphenazine treatment	DSM-IV and BPRS
Notes and Conclusions. Little change in thyroid function except for a significant decrease in TT4 in quetiapine treatment possibly due to UDP-glucuronosyltransferase competition with the drug. No hypothyroidism noted. At baseline no TH differences between treatment groups. No control group. No monitoring of TH function in patients receiving quetiapine recommended. Probably no effect of risperidone or fluphenazine on TH (Kelly and Conley [65]).
Normal	Normal	Normal		Normal	Normal	Antiperoxidase: normal	1077 individuals, 60–90 years old, enrolled in a MRI study. No dementia in the beginning. Serum collected at enrollment. 6-year followup for dementia and MRI scan. No further TH status assessment.	Minimental, Geriatric Mental Scale Schedule and DMS-III
Notes and Conclusions. Even though atrophy is present in demented patients, the presence of atrophy does not necessarily lead to dementia. Lower TSH with anti-TPO associated with 4 times higher risk of dementia but 𝑃>0.05. No association of serum TSH, FT4, TT3, rT3, anti-TPO with dementia. Higher FT4 and rT3 related with higher hippocampus and amygdala atrophy (de Jong et al., [66]).

BPRS: Brief Psychiatry Rating Scale; DSM: Diagnostic and Statistical Manual of Mental Disorders; SADS: Schedule for Affective Disorders and Schizophrenia; TTR: transthyretin; TBG: thyroxine-binding globulin; TAb: thyroid antibodies; SZ: schizophrenia.