Transcript Population-based Thyroidology

Update on the Thyroid
Douglas C. Bauer, MD
UCSF Division of General Internal Medicine
No Disclosures
Cases
• 68 yr old female with new atrial fibrillation
and no other findings except TSH=0.04,
normal free T4
• 79 yr old man with 1 yr of fatigue and
lassitude and no findings except TSH=9.0,
anti-TPO positive
• 45 yr old women, enlarged thyroid with
dominant nodule since 1999, FNA benign.
On T4 suppession ever since, TSH=0.1
Thyroid Tests: sTSH
• Very sensitive to circulating thyroid hormone levels
• Excellent correlation with TRH stimulation (sTSH < 0.1)
• Requires intact pituitary-hypothalamic axis;
4-6 weeks to equilibrate
• Falsely low: severe illness, corticosteriods, dopamine
• Normal range 0.5-4.8 mU/L; $58
Thyroid Tests: Free Thyroxine
•
•
•
•
•
Measures unbound hormone
Replacing “index” assays
Gold standard: Equilibrium dialysis
Other immunoassays: Improving
Normal range, 9-24 pmol/L; $64
Are Both sTSH and Free T4 Necessary?
• American Thyroid Association: Yes
• Others recommend sTSH first
• UCSF outpatient data
– Results when both tests ordered on the
same specimen (N=3143)
– Each test classified as low, normal or high
Diagnostic Redundancy of sTSH and Free T4
sTSH (mU/L)
< 0.5
0.5 - 5.5
<9
Free T4
(pmol/L)
> 5.5
4
16
49
9 - 24
536
2024
309
> 24
174
30
1
Subclinical Thyroid Disease
• Subclinical hypothyroidism
“Abnormally high sensitive TSH and
normal thyroid hormone levels”
• Subclinical hyperthyroidism
“Abnormally low sensitive TSH and
normal thyroid hormone levels”
Suggested Testing Strategy
• If sTSH is normal, STOP
• If sTSH is low, measure T4, consider T3
• If sTSH is high, measure T4, consider
TPO antibodies
Thyroid Antibodies
• Anti-thyroperoxidase, TPO (titer<100, $78)
– Similar to “anti-microsomal”
– Most sensitive thyroid autoantibody
– Specificity a problem
• TSH receptor antibody (absent, $112)
– Causes Grave’s disease
– Rarely found in normal individuals
Thyroid Scans
• Technetium 99 ($450)
– Low radiation, quick
– Useful for nodules in some circumstances
– Useful to determine cause of hyperthyroidism
• A. High uptake: Grave’s, toxic nodule
• B. Low uptake: thyroiditis, thyroxine use
Hyperthyroidism: Epidemiology
• Etiology:
– Iatrogenic
• A. Over replacement (30-50% given rx)
• B. Suppression of CA, goiters, and nodules
– Autoimmune (Grave’s disease): thyroid
stimulating autoantibodies
– Autonomous nodule(s). Occasionally T3
– TSH secreting tumors
Hyperthyroidism: Prevalence
• Population based prevalence of
suppressed TSH:
Author
age
men
women
Bagchi (1990)
Falkenberg (1991)
Parle (1991)
Bauer (1993)
>55
>60
>60
>55
1.8%
2.7%
1.9%
6.3%
5.8%
5.5
Crook’s Index*
Symptom/Sign
Present
Absent
Palpitation
+2
0
Cold prefer.
+5
0
Hyperkinetic
+4
-2
Weight loss
+3
0
Lid lag
+1
0
*hyperthyroid if 10 or more
Hyperthyroidism in the Elderly
• Weight loss, palpitations, and
nervousness less common
• Tachycardia, exophthalmos, tremor
less common
• Atrial fibrillation more common
• 8-10% are completely asymptomatic
Subclinical Hyperthyroidism: Cardiac Effects
• Systolic time intervals shortened
– Clinical significance uncertain
• Reduced exercise tolerance
• Increased incidence of atrial fibrillation
Swain, 1994
Prospective cohort, N = 2000
RR = 3.1 (1.7, 5.5) if sTSH < 0.1
Subclinical Hyperthyroidism: Skeletal Effects
• Florid hyperthyroidism causes fractures
• Effect on BMD, bone loss controversial
• Increased fracture risk (Bauer, 2001)
- Prospective study, 9407 older women
- TSH < 0.1 vs. normal
- Hip fracture: RR = 3.6 (1.0, 12. 9)
- Vertebral fracture: RR = 4.5 (1.3, 15.6)
• Effect of accelerated bone turnover?
Subclinical Hyperthyroidism: Natural History
• Exogenous: Well established
• Endogenous: Little longitudinal data
Parle, 1991
50 untreated individuals >60
1 developed overt hyperthyroidism
After 1 year, sTSH normal in half!
Who Should Be Treated?
• Exogenous (iatrogenic)
–Dose reduction unless contraindicated
• Endogenous (subclinical)
–Follow if uncomplicated
–Consider treatment if atrial fibrillation or
osteoporosis present
• Endogenous (overt)
–Rule out thyroiditis
–Tx everyone else with beta blocker and...
Hyperthyroidism: Treatment
• Anti-thyroid drugs (PTU and methimazole)
– Remission: 30-50% after 12-18 mo
– Side effects: rash, fever, arthritis,
agranulocytosis (all rare)
• Radioiodine
– Best treatment for hot nodules
– Remission: everyone
– Side effects: transient thyroiditis (rare),
hypothyroid (50%), worsening exophthalmous
Radioiodine and Mortality
• Franklyn, 1998
- 7209 hyperthyroid pts, 15 yr follow-up
- All cause mortality: 13% higher than
age and sex matched populations
- CV deaths increased, but not cancer
• Mechanism unknown, clear dose-response
• Unable to adjust for other potential
confounders
Hypothyroidism: Epidemiology
• Etiology
–Autoimmune (Hashimoto’s)
–Iodine deficiency
–Iatrogenic
A. Radioiodine/ surgery
B. Drugs (lithium, amiodarone)
–Pituitary/ hypothalamic disease
Hypothyroidism: Prevalence
• Population based prevalence of
elevated TSH:
Author
age men women
Tunbridge(1977)
Bagchi(1990)
Parle(1991)
Bauer(1993)
>65
>55
>60
>55
6.0%
1.8%
2.9%
10.9%
2.7%
11.6%
5.4%
Billewicz Index*
Symptom/Sign
Present
Absent
Bradykinesia
+11
-3
Cold interance
+4
-5
Coarse skin
+7
-7
Pulse <75
+4
-4
Delayed AJ
+15
-6
*hypothyroid if > 30
Overt Hypothyroidism in the Elderly
• “Classic” features often missing
• Neuropsychiatric complaints common:
depression, weakness, memory loss
• Other clues: hypercholesterolemia,
elevated CK, pleural effusion
Subclinical Hypothyroidism: CV Outcomes
• Observational studies
– Total cholesterol unchanged, but higher LDL
and lower HDL?
• What about atherosclerosis?
• Rotterdam population-based study (Hak, 2000)
–
–
–
–
1149 women, mean age 70
Subclinical hypo (TSH > 4, nl T4) in 10.8%
Aortic atherosclerosis RR = 1.7 (1.1, 2.6)
History of MI RR = 2.3 (1.3, 4.0)
Meta Analysis of Subclinical Hypothyroidism and CHD
P for heterogeneity: 0.12
Summary OR
1.65 (1.28-2.12)
Subclinical Hypothyroidism: Other Outcomes
• Observational studies of neuropsychiatric
symptoms
– Conflicting evidence
• Four small double blinded trials, sTSH > 5-7
– Randomized to thyroxine or placebo
– No significant change in weight, lipids, other
laboratory values
– Psychometric testing: Treated felt better and
had better memory scores
Subclinical Hypothyroidism: Natural History
• Many good studies
• Spontaneous resolution infrequent
• Antibodies strongly influence outcome
– If TPO positive, overt hypothyroidism
5%/yr
Hypothyroidism: Treatment
• Replace with thyroxine (T4)
– T3 + T4 benefit unproven
• Typical replacement dose 1.6 mcg/kg
–Elderly or CAD: start low (0.025-0.05
mg/d), gradually increase dose
• Maintain TSH within the normal range
–Wait 6 weeks after dose change
• Monitor yearly (noncompliance, reduced
T4 clearance)
What About Treatment of Symptomatic but
Euthyroid Patients? Forget It.
• Symptoms of hypothyroidism common
– Real but not detected by usual tests?
• Double blind RCT (Pollock, 2001)
– 25 “symptomatic”, 18 “controls”
– All euthyroid
– 3 mo of T4 (0.1/d) or placebo, cross-over
– TSH fell with T4 tx but no difference in
cognitive or psychological function
Thyroid Nodules: Epidemiology and Evaluation
• Nodules are common (and cancer is rare)
– 90% women over age 60 have one or more
thyroid nodules at autopsy
• Risk factors for cancer: neck irritation, FH
• Evaluation: FNA first
– 75% benign, 20% suspicious, 5% malignant
– Best centers: false negative 2%
false positive 1%
Thyroid Nodules: Treatment
• Cancer
- Histology is important (papillary best)
- Surgery and 131I ablation
- Suppression with T4? TSH = 0.1-0.4
• Benign nodules
- Many shrink spontaneously
- Meta analysis of T4 suppression
Smaller: 26% vs. 12% (NNT=7)
Larger: 8% vs. 17% (NNT=11)
- T4 doesn’t prevent new nodules
Screening Cost-effectiveness
• Danese and Sawin, 1995
– Cost-utility analysis, sTSH-based screening
– Modeled progression, symptoms and CAD
– Screening every 5 year from 35-65:
$9,223 per QALY in women
$22,595 per QALY in men
– Sensitivity analysis: cost of TSH key ($25)
Screening for Subclinical Thyroid Disease
• US Preventive Task Force, 1996
“Routine screening is not recommended.
Insufficient evidence for high risk patients,
including elderly.”
• ACP, 1998
“It is reasonable to screen women older
than 50 years of age for unsuspected but
symptomatic thyroid disease.”
Screening Cost-effectiveness
• Effects on HDL, fractures not included.
Cost of testing overestimated ($3/TSH)
• Published analyses underestimate
cost-effectiveness
• Other unresolved issues:
– Age to start screening?
– Optimal frequency?
Summary Take Home Points
• sTSH is best test
• Subclinical thyroid disease is common,
associated with morbidity, and treatable
• Low threshold to treat subclinical hypo
• Treatment threshold for subclinical hyper
less certain
• Screening with sTSH is cost-effective
Cases
• 68 yr old female with new atrial fibrillation
and no other findings except TSH=0.04,
normal free T4
• 79 yr old man with 1 yr of fatigue and
lassitude and no findings except TSH=9.0,
anti-TPO positive
• 45 yr old women, enlarged thyroid with
dominant nodule since 1999, FNA benign.
On T4 suppession ever since, TSH=0.1