The Tyranny of False Assumptions



The Clinical Diagnosis and Treatment of HypothyroidismSummary:The current TSH-based approach to thyroidology is illogical and ineffective, distorting the research, the interpretation of the research, and the free T4 and free T3 reference ranges. Thyroidology must be based upon the assessment of T3 effects in all tissues as indicated by the patient’s clinical criteria (signs and symptoms). Combination T4/T3 therapy is required to produce optimal T3 effects in all tissues while minimizing negative effects.Thyroid hormone effects cannot be understood in isolation from the rest of the endocrine system; in particular the patient’s cortisol status. The ultimate test of whether a patient is experiencing the effects of too much or too little thyroid hormone is not the measurement of hormone concentration in the blood but the effect of thyroid hormones on the peripheral tissues.The Tyranny of False AssumptionsHypothyroidism is the state of inadequate T3 effect in some or all tissues of the body. T3 is arguably the most powerful molecule in biology; it increases mitochondrial energy production, and thus improves the function of every cell, tissue and organ in the body. It has many other direct and indirect effects that we are only beginning to understand. T4 is the inactive thyroid prohormone, but is much more abundant in the blood and is the source of most of the T3 in the body. Any degree of T3 deficiency degrades our physical and mental functioning and our long-term health. The symptoms and signs of hypothyroidism are so many and varied (See Table 1.) that hypothyroid patients often receive many other diagnoses. The ability to diagnose and effectively treat every form and degree of hypothyroidism is central to medical practice, and, I will argue, the key to understanding the endocrine system as a whole. In particular, one must be able to diagnose and effectively treat hypothyroidism in order to understand cortisol deficiency (adrenal insufficiency). T3 and cortisol have very powerful interactions and deficiencies of one or both of these hormones cause most of the unexplained fatigue, depression, and pain that affect so many people. What guidance do physicians now receive regarding the diagnosis and treatment of hypothyroidism? Table 1: Signs and Symptoms of HypothyroidismFatigue, excessive need for sleepCold intoleranceWeight gain, cannot lose weightConstipation, poor digestionDry skin, itchingEsophageal refluxMuscle aches, cramps, stiffnessMyxedema of the face, lower legsPoor concentration, memoryDepression or anxietyVoice changes, hoarsenessHeadache, often upon awakeningElevated total and LDL cholesterolAtherosclerosisHypertensionCarotenemia, yellowing of skinDry hair and/or hair lossSlow heart rate, palpitationsInsomnia, daytime restlessnessHeavy menses or amenorrheaInfertilityCarpal tunnel syndromeSleep apneaClinical medicine is defined as “the study and practice of medicine in relation to the actual patient; the art of medicine as distinguished from laboratory science”. The American Association of Clinical Endocrinology (AACE) and American Thyroid Association (ATA) claim to provide “evidence-based clinical guidelines for the clinical management of hypothyroidism”, but instead provide a TSH-T4 laboratory reference-range scheme that ignores evidence and dismisses clinical criteria (the patient’s signs and symptoms) as irrelevant to diagnosis and treatment. The guideline’s authors define “euthyroidism” not as a physiological state of optimal T3 effect, but as having a thyroid stimulating hormone (TSH) level and/or free T4 (FT4) level anywhere within the laboratory’s reference ranges. Now TSH is not a thyroid hormone, and its level is an indirect and fallible indicator of a person’s thyroid hormone status. So they do state that the diagnosis of hypothyroidism must always be “biochemically confirmed” by a low FT4. (I use “normal” and “low” throughout to refer to reference ranges only, with no clinical implication.) The authors say nothing about how FT4 reference ranges are determined; so they have delegated the diagnosis of hypothyroidism to some laboratory scientists. They assert that the free T3 (FT3) level, the level of the active thyroid hormone in the blood, is of no consequence, even if low during treatment, and should not be tested. Abbreviations:HP hypothalamic-pituitaryTSH thyroid stimulating hormoneFT3 free triiodothyronine levelFT4 free thyroxine levelTFTs thyroid function testsTRT thyroid replacement therapyTSHT4Rx TSH-normalizing T4 therapyThe authors assume that almost all hypothyroidism is primary, due to failure of the end-hormone producing gland. This requires the unstated assumption that any degree of central hypothyroidism, due to hypothalamic-pituitary dysfunction and inadequate TSH production, is rare. They are thus assuming that, absent any obvious hypothalamic-pituitary (HP) disease or damage, TSH production is perfectly vigorous and will always assure optimal thyroid levels/effects for every individual. So they believe that dysfunctional central hypothyroidism does not exist; a belief for which no evidence can possibly be cited as support. It is a belief that, however, underlies their thyroidology. When it comes to treatment of primary hypothyroidism, which the TSH is usually elevated, they add yet another assumption: that the individual’s perfectly vigorous TSH secretion also reacts to once-daily oral T4 therapy exactly as it does to endogenous thyroidal production. Thus they believe that all they must do is prescribe enough of the inactive T4 hormone, levothyroxine, to lower the elevated TSH back to within its population range. They unthinkingly equate “euthyroidism” with a normal TSH, treated or untreated. The resultant thyroidology scheme is utterly simplistic: a normal TSH is euthyroidism; a high TSH is hypothyroidism and is treated with enough levothyroxine (T4) to normalize the TSH. Both the clinical effect of the treatment on the patient and the patients’ actual thyroid hormone levels (FT4 and FT3) are ignored as irrelevant because both lack “sufficient specificity to serve as therapeutic endpoints”. The AACE/ATA TSH-T4 reference range scheme is a closed system of assumptions and conclusions that is completely insulated from clinical reality. When it doesn’t work, the patient is presumed to have some other medical or psychological problem. The scheme cannot be “disproved” by any research because it is used to both produce the research and interpret it. It thus continues to persist in spite of its abject failure in scientific studies and in physicians’ and patients’ daily experiences. It is no coincidence that since it began to be adopted in the 1970s, there has been an explosion in the number of people diagnosed with chronic fatigue syndrome, fibromyalgia, depression and other disorders that can be caused by hypothyroidism. T4-treated patients are increasingly vocal about their dissatisfaction with their treatment. in many websites and forums laypersons are helping each other to obtain diagnosis and effective treatment according to an alternative thyroidology. Some patients have taken this issue to their legislature. To deal comprehensively with this problem I must thoroughly expose the false assumptions underlying TSH-T4 thyroidology. I will contrast them with the available evidence and then describe an approach to clinical thyroidology based upon the available research and my own experience. TSH-based thyroidology Let us state the assumptions of the ATA/AACE diagnosis-and-treatment scheme explicitly:Thyroid stimulating hormone (TSH) secretion is always perfect, assuring thyroid sufficiency for each person, unless there is obvious HP damage/disease. Almost all hypothyroidism is primary and detectable by an elevated TSH. (Immaculate TSH)Primary hypothyroidism is sufficiently treated by normalizing the TSH with replacement therapy. I will discuss the use of the TSH to guide T4 therapy below. First I will address the Immaculate TSH assumption. It is not just improbable; it is known to be false. HP function is highly complex and variable. The most direct and reliable laboratory tests of thyroid status are the thyroid hormone levels, FT4 and FT3. This is acknowledged by the AACE/ATA as they describe the “pitfalls” of interpreting the TSH and are careful to define hypothyroidism as a low FT4. However the authors believe that they can consider HP dysfunction to be rare, limited to cases with obvious intracranial pathology. They believe that they can avoid the “pitfalls” so well that they can substitute the TSH test for the FT4 and FT3 levels. This is illogical; it is analogous to insisting that one’s home-heating thermostat always works perfectly even as the house gets colder and colder. HP function is not immaculate; it is fallible. The stimulatory pituitary hormone level is always an indirect test. It is not a reliable inverse indicator of end-hormone levels or effects throughout the body. In no other endocrine system do we assume HP perfection. We do not use luteinizing hormone (LH), follicle-stimulating hormone, or adrenocorticotropin levels as guides for diagnosis or treatment. Dysfunctional LH hyposecretion, without any gross pathology, is nearly universal in aging males and contributes to the age-related decline in testosterone levels. TSH production also declines with age. Children and young adults have more vigorous TSH secretion and higher thyroid hormone levels. Between the ages of 20 and 80 FT3 declines by 30% and the TSH response to low FT4 levels declines by 75%. In thyroidology as in the rest of endocrinology, logic requires us to first determine that a hormone deficiency exists based upon symptoms and hormone levels, and then check the pituitary hormone to see if the deficiency is primary or central. It is easy to draw false inferences from population correlations. The level of the stimulatory pituitary hormone will be usually be elevated when the primary gland is failing and suppressed if the gland is overactive. Due to the presence of primary hyper- and hypothyroidism in the population there is an inverse population correlation between TSH and FT4/FT3 levels. However, there is no correlation within the normal TSH range of 0.5 to 3.0mIU/L. The slope in this region is flat; meaning that a TSH anywhere in this range coexists with the same broad range of FT4 values, from low to high. This fact attests to the variability of HP function. HP dysfunction exists; a normal TSH does not imply thyroid hormone sufficiency. Even outside this range the TSH is not diagnostic: an elevated TSH can be due to thyroid hormone resistance, adrenal insufficiency, or a TSH-secreting adenoma producing hyperthyroidism. Even an elevated TSH cause by thyroid gland pathology does not imply hypothyroidism; if TSH secretion is vigorous it may be compensatory, maintaining thyroid sufficiency. A low TSH can be associated with either low or high thyroid levels (i.e., central hypothyroidism or primary hyperthyroidism). Some attempt to defend the AACE/ATA’s reliance on the TSH test because it is a “sensitive” test. The latest generation TSH test is indeed more sensitive to lower TSH levels. The TSH level is sensitive too in that it responds in a logarithmically-amplified way to changes in the serum FT4 level in any given person. These facts do not imply that HP function is perfect or that the TSH is the right test for diagnosis or treatment. In my experience, dysfunctional central hypothyroidism (CH) is more common than primary hypothyroidism. This should be expected as the HP system is part of the brain and thus far more complex than the thyroid gland. It is much more likely to be dysfunctional. The hypothalamus is affected by inputs from many regions of the brain and by many neurotransmitters, environmental chemicals, drugs, illnesses and other factors. Dysfunctional central hypothyroidism has been associated with a number of mutations and other molecular disorders, including the secretion of an inactive form of TSH.,, There are reports of central hypothyroidism with normal imaging studies;,, these are the tip of an iceberg. Even if we had some independent way of knowing that HP function and TSH secretion were perfectly vigorous in a given patient, the TSH level still reflects only the HP system’s response to circulating T4 and T3, not T3-effect in other tissues. The HP system is more sensitive than other tissues to circulating T4; the brain and pituitary gland have higher levels of the deiodinase D2 causing them to convert T4 to T3 more avidly., The central nervous system has no D1, but in other tissues D1 is a major determinant of T3 production. The production and activity of D1, D2, and D3 are variously affected by many factors. There are also 4 different thyroid hormone receptors, and at least 10 different active transport systems with variable tissue distribution.,, All of these proteins are subject to single nucleotide polymorphisms., Peripheral thyroid hormone resistance may be more common than previously realized. Even the serum FT4 and FT3 levels are indirect indicators of intracellular T3 levels-effects. To reduce all of thyroidology to TSH or even T4 management is to ignore its complexities—known and unknown. The misuse of the TSH distorts the FT4 reference rangesDiagnostic assays produced by different manufacturers can vary significantly in the results they produce; this is particularly true of FT4 immunoassays. Each assay manufacturer provides a FT4 range based upon “apparently healthy” non-patients as described above. They may be screened with TSH and thyroid antibody tests to exclude PH, but are not screened for hypothyroid symptoms; so severely hypothyroid persons can be included. Each laboratory must “validate” a reference range for their use of the assay with their population. They do so by combining the values from the manufacturer’s range, the published literature and tests that they perform. Here again the immaculate TSH doctrine leads to error. Assuming that a normal TSH means euthyroidism, laboratories save time and money by including FT4 and FT3 values from physician-ordered thyroid panels they’ve performed in which the TSH was normal. The laboratories’ inclusion of both symptomatic undiagnosed and T4-treated hypothyroid clinic patients in their FT4 and FT3 ranges and this is evident in the breadth of the FT4 ranges that they report. Published studies of non-patient populations, without symptom screening, consistently yield relatively narrow 2 S.D. FT4 ranges of around 1.0 to 1.65ng/dL (12.9–21.3pmol/L).,,,, (The actual values vary with the immunoassay.) However, most laboratories using similar assays report considerably broader FT4 ranges with lower limits of only 0.6-0.8ng/dL and upper limits of 1.7-2.2ng/dL (7.7-10.3pmol/L to 23.2-28.4pmol/L). I submit that the lower limit is reduced by the inclusion of symptomatic clinic patients with TSH-normal central hypothyroidism, and the upper limit raised by the inclusion of T4-treated primary hypothyroidism patients. The lower FT4 limit attests to the prevalence of dysfunctional central hypothyroidism in the patient population, and explains why many cases of central hypothyroidism due to HP damage-disease have low-normal FT4 levels.,, Indeed, in my experience, dysfunctional central hypothyroidism is far more prevalent than primary hypothyroidism. A study of patients with depression found normal TSH levels but low or low-normal FT4 and FT3 values. The mean FT4 was only 11.43 (range: 10.6–19.40 pmol/l). The mean FT3 was only 4.45 (±0.81) pmol/l (range: 4.00–8.3 pmol/l). While the TSH levels were normal, 16% of the FT4 and 30% of the FT3 values were below the reference ranges. If laboratories were to report FT4 results with the unscreened, non-patient range of 1.0 to 1.65ng/dL (12.9–21.3pmol/L), the impact on medical practice would be immense. Many symptomatic patients have normal TSHs and FT4s below 1.0ng/dL. They would all be diagnosed with central hypothyroidism. Yet 1.0ng/dL is still not a diagnostic lower limit. With careful screening for signs and symptoms of hypothyroidism the lower limit would be higher, probably around 1.2ng/dL (15.5pmol/L); yet this is still just a 2 S.D. population statistic. Persons differ in their need for thyroid hormone,, their conversion of T4 to T3, their sensitivity to T3, their cortisol levels, and in other physiological parameters that affect thyroid hormone action. The validity of FT4 assays also falls far short of ideal. Therefore, a FT4 at the 50th percentile of a meaningful reference range may not be sufficient for some persons, while a low FT4 may be adequate for others. There is simply no laboratory substitute for clinical thyroidology.FT4 and FT3 Levels MatterThe excessive breadth of the FT4 and FT3 ranges is also evidenced by many studies that show detrimental effects with lower levels and beneficial effects with higher levels within the ranges, and that show benefits with treatment for patients with low-normal levels. Lower FT4 levels, even within the range, in the first trimester of pregnancy are associated with lower neonatal neurobehavioral scales, impaired psychomotor development and autism. In middle-aged adults, higher FT4 levels are associated with lower all-cause mortality, and higher FT3 levels with lower cancer mortality.Hypothyroidism is known to cause hypercholesterolemia and to increase atherosclerosis. Carotid artery intimal thickness is inversely associated with FT4 levels within the range., Persons with FT3s in the upper third of the range have half the incidence of severe atherosclerosis as those with FT3s in the lower third. Higher T4 doses prevent the progression of coronary artery atherosclerosis, whereas lower doses allow progression []. A lower FT4 within the range is associated with hypercoagulability. Subclinical hypothyroidism (SH), where FT4 levels are generally lower within the range, is associated with cardiovascular disease and mortality; the risk rises with higher TSH levels. SH is associated with higher blood pressure. Patients with SH have elevated lipid levels compared to controls and to eliminate the difference requires TSH-suppressing T4 doses. Lowering the TSH to under 2.0 mIU/L with T4 therapy, rather than just normalizing it, produces lower cholesterol, homocysteine, and CRP levels. The negative health and quality-of-life consequences of obesity are well-documented. In untreated persons, lower FT4 values within the range are associated with greater body mass index, weight gain, subcutaneous fat, and with four of the five components of the metabolic syndrome. Lower FT3 levels within the range are also associated with lower metabolic rate and weight gain. PCOS can be caused by hypothyroidism and can resolve with thyroid replacement therapy.Optimal, not just normal thyroid levels are also beneficial for cognitive function, mood, and well-being. Higher T4 levels within the range are associated with better cognitive function, and lower risk of cognitive decline. Lower thyroid hormone levels and/or higher TSH levels within the ranges have been associated with depression or a worse prognosis for remission of depression.,,,,, T3 therapy alleviates depression in persons with normal thyroid function tests (TFTs).,,,,,, Those who respond are more likely to have T4 levels in the lower third of the range. In T4-treated primary hypothyroidism patients, higher FT4 and lower TSH levels within the ranges are associated with psychological well-being. Persons with SH and lower FT4 levels complain more of myalgias and weakness and have lower muscle strength on testing. 80% of persons with symptoms but normal TFTs experienced better mood and energy on 125mcg T4 daily.In spite of such evidence, physicians claim that those with normal TFTs who experience subjective improvement on thyroid replacement therapy must have “thyrotoxic euphoria”. This ad hoc diagnosis illustrates the closed nature of the TSH-T4 reference range scheme; all clinical evidence that contradicts it is rejected. In fact, people usually feel worse, not better when they have excessive thyroid hormone levels-effects. Patients with endogenous subclinical hyperthyroidism have the same negative, undesirable symptoms as hyperthyroid patients. Women on TSH-suppressive therapy with excessive T4 doses have lower quality-of-life and psychometric functionality scores. Asymptomatic controls given 100mcg of T4 daily can suffer thyrotoxic symptoms initially. Excessive T3-for-T4 substitution therapy also reduces quality of life. (Appendix) T4 and T3 are not drugs. Thyroid supplementation that makes an individual feel and function better and produces no signs or symptoms of excess must be considered necessary and beneficial until proven otherwise. T3 and the ineffectiveness of TSH-T4-normalizing T4 therapyThese assumptions are also contradicted by the evidence. Oral T4 monotherapy is an unphysiological intervention into a highly complex system. Studies of TSH-normalizing T4 therapy (TSHT4Rx) in primary hypothyroidism consistently show that it does not restore clinical euthyroidism in most persons, at any TSH or FT4 level within the range. Patients receiving TSHT4Rx display significant impairment in psychological well being, health status, and cognitive function compared to controls.,, They have more depression and anxiety and their symptoms are worse with higher TSH levels within the range. They are twice as likely to be taking anti-depressant medications. They have higher hypothyroid index scores and body-mass indices, and 21% greater fat mass than controls. They have persistent endothelial dysfunction [] and an increased risk of cardiovascular morbidity. After thyroid ablation, patients on TSHT4Rx gain weight (avg. 4kg) whereas those on TSH-suppressive therapy do not. The use of the TSH to guide to T4 therapy was definitively tested when four experienced clinicians adjusted the T4 doses of 148 hypothyroid patients based on clinical criteria, using signs and symptoms as quantified by the Wayne index. After dose adjustment, for patients they judged to be clinically euthyroid, the treated TSH 2 S.D. range was <0.1-13.7mIU/L (conventional range: 0.35-5.0mIU/L). The treated FT4 range was 50% higher than the conventional range (12-36pmol/L vs. 9-25pmol/L). Only the treated FT3 range was similar to the conventional range (3.0-8.6 vs. 2.9-8.9pmol/L). With T4 therapy, the TSH was the least accurate measure of euthyroidism and the FT3 the most accurate. As this study suggests, the explanation for the inadequacy of TSHT4Rx is found in the T3 levels. Serum T3 reflects the amount of T4-to-T3 conversion throughout the body and thus the T3 levels and action in the tissues. Patients with untreated primary hypothyroidism are much less symptomatic if their FT3 is normal rather than low. TSHT4Rx produces higher FT4 levels, but lower T3 levels than in controls.,,,,,,, T4-treated patients can have the same 24-hour urine T3 levels as untreated hypothyroid patients. After thyroidectomy the restoration of pre-operative T3 levels requires T4 doses that either suppress the TSH or produce T4 levels 40% higher than before surgery. On TSH-suppressive T4 therapy, a FT4 that is 66% higher than controls produces the same T3 level and no symptoms of hyperthyroidism. Patients on T4 therapy feel better when their TSH is suppressed below its range and their FT4 and FT3 are in the upper half of their ranges. In SH the TSH can be normalized with T4 doses of only 25 to 50mcg, whereas the average replacement dose is 145mcg/day. These low T4 doses do not improve symptoms or raise T3 levels; they often lower the patient’s T3 level. T4 therapy given to “euthyroid” women to lower their TSH to the bottom of the range had no effect at all on parameters of thyroid status. Raising or lowering patient’s TSH-normalizing T4 dose by 25mcg also had no effect on symptoms. The ineffectiveness of TSHT4Rx has led some investigators to recommend adjusting T4 therapy according to the T3 level., In a trial comparing T3-only and T4-only therapy titrated to the same normal TSH value, The TSH-normalizing T3-only therapy produced no negative effects but did produce an average weight loss of 2.1kg with a 5% decrease in total fat mass. T3 therapy and lowered total and LDL cholesterol levels by more than 10% and apoliprotein B by 18%.Why are T3 levels and effects so low in TSHT4Rx? There are many reasons. First, there is no guarantee that the treated patient has perfectly vigorous TSH production to start with. If there is any degree of central hypothyroidism, any lack of vigor in the TSH response, TSH normalization will produce undertreatment. The aging process reduces HP function, and middle-aged adults most often receive thyroid treatment and are most often included in studies. Even if TSH secretion is perfectly vigorous in a given patient, the HP feedback control system evolved to interact with the thyroid gland’s continuous production of T4 and T3 and with the various deiodinases. It did not evolve to tell physicians how much T4 a person should swallow every morning. T4 monotherapy must often produce supraphysiological serum T4 levels in order to normalize T3 levels within the HP system, and thus normalize the TSH level. This is true even though the HP system is more sensitive than other tissues to circulating T4. The thyrotrophs can increase D2 expression to maintain intra-pituitary T3 production at higher T4 concentrations,, whereas outside the HP system D2 is suppressed by higher FT4 levels. D2 in skeletal muscle is the major source of circulating T3. Therefore, the supraphysiological T4 levels that can normalize or suppress the TSH do not produce proportionate increases in serum FT3, which represents T3 production throughout the body. The higher FT4 levels on T4 therapy also induce D3 action, promoting the conversion of T4 to reverse T3 (RT3). RT3 is not only inactive, but also inhibits T4-to-T3 conversion. On TSHT4Rx, RT3 levels are often high; they are 50% higher than on T4/T3 therapy that produces the same TSH. In sum, normalizing the TSH produces “euthyroidism” in the HP system, but not in the rest of the body. Whereas endogenous T4 and T3 production are essentially constant over 24hrs, oral dosing releases the entire day’s hormone into the circulation within a few hours. The peak FT4 levels at 3hrs. after a T4 dose are 13% to 36% higher, and FT3 levels 8% higher than at the 24hr trough.,, Even so, little variation is seen in TSH levels on once-daily T4 therapy, indicating that the T4 peaks may have a stronger and longer-lasting effect on the HP system than the rest of the body. When hypothyroid patients are given a single dose of 50mcg T4, their TSH drops by 30% at 6 hrs. but their T4 and T3 levels do not change. In rats, rapid T4 infusions suppress the TSH for over 22 hours; only a continuous T4/T3 infusion produces tissue thyroid sufficiency without suppressing the TSH.Furthermore, the reduction in TSH with T4 therapy reduces thyroidal production of both T4 and T3; the amount of T4 supplied may not compensate for these reductions, especially when low T4 doses are given to normalize a mildly elevated TSH. Lowering the TSH with T4 therapy also reduces T3 production throughout the body. TSH stimulates D1 and D2 activity in the periphery, where approximately 75% of T3 in the serum is produced. In thyroidectomized dogs on T4 therapy, TSH-injections raise serum T3 levels by 40% while lowering T4 levels. The T3:T4 ratio in primary hypothyroidism with its high TSH is double that in CH.Some patients may be sufficiently treated TSHT4Rx, but in my experience careful questioning reveals persisting hypothyroid symptoms; and some are markedly hypothyroid. Their FT4s may be only mid-range or low-normal and FT3s low-normal or low. Those with high-normal or high FT4s usually have high-normal or high RT3s, reducing the effectiveness of the therapy. Consider that in CH, where the TSH cannot be used to guide treatment, merely normalizing the FT4 with oral T4 is known to be insufficient, leaving the FT3 low in one-half of patients. In central hypothyroidism most guidelines recommend keeping the FT4 above the middle of range. However, with no TSH to promote T4-to-T3 conversion, this too is inadequate. The nearly universal weight gain with central hypothyroidism (“hypothalamic obesity”) is iatrogenic; the addition of T3 to T4 therapy produces weight loss and eliminates hypothyroid symptoms. Some experts recommend keeping the FT4 near the upper limit and FT3 in the upper half of the range, while others recommend monitoring clinical indices of thyroid action. Persons with central hypothyroidism are sometimes grossly undertreated by physicians who try to keep their TSH within the ref. range, to escape the grip of TSH-based thyroidology. TSH-suppression on thyroid replacement therapy does not imply thyrotoxicosisBecause of the AACE/ATA’s endorsement of the TSH as the “best test”, physicians and even thyroid researchers assume that a low TSH level on thyroid replacement has the same implications as endogenous hyperthyroidism, and will cause bone loss, muscle wasting, cardiac dysfunction and atrial fibrillation. However, some experts disagree. The Royal College of Physicians states that sufficient T4 therapy may produce a below normal serum thyroid stimulating hormone, and senior thyroidologists have asserted that “Some patients achieve a sense of wellbeing only if free T4 is slightly elevated and TSH low or undetectable. The evidence that this exogenous form of subclinical hyperthyroidism is harmful is lacking… and it is not unreasonable to allow these patients to take a higher dose if T3 is unequivocally normal”. The TSH test is simply the wrong test, both for any patient and for scientific studies. It is not a measure of thyroid hormone levels or effects in the untreated state, and is even less relevant in the unphysiological treated state. Since TSHT4Rx is usually inadequate, sufficient treatment must produce a low or suppressed TSH in most persons. True, a suppressed TSH will be seen with overtreatment too. It is also true that persons with certain underlying conditions will have negative effects with otherwise optimal thyroid replacement therapy. What matters in every patient, whether untreated or treated, are the clinical status as indicated by signs and symptoms first, and the FT4 and FT3 levels second. There is plenty of evidence, for those who need it, showing that TSH-suppressive therapy (TSHSupRx) does not equal hyperthyroidism. Calorimetry studies in thyroidectomy patients on TSHSupRx show no increase in metabolism compared with the pre-surgical state. A large study of T4-treated patients grouped by TSH levels found that those with low but not suppressed TSH levels (0.04-0.4mIU/L) had no increase in cardiovascular disease, dysrhythmias, fractures, or mortality compared to those with normal TSH values. Only those with completely suppressed TSH levels had some increase in morbidity, as expected since some patients with suppressed TSH levels may be overtreated. Another study found no increase in morbidity with suppressed TSH levels; and the average T4 level was only mid-range. It’s the wrong test.The benign nature of a low or suppressed TSH with T4 therapy contrasts sharply with the thyrotoxicosis seen in persons with similarly low TSH values caused by endogenous overproduction, again revealing the disconnect between endogenous and supplemented TSH levels. Patients with endogenous subclinical hyperthyroidism have signs and symptoms of thyroid excess even though their TSH is only slightly low (avg. 0.15mIU/L). Their FT4 and FT3 are both in the upper thirds of their ranges, a pattern not seen with a similar TSH on T4 therapy. In endogenous hyperthyroidism, FT3 levels are uniformly high, a result that is hard to achieve with T4 therapy. TSHSupRx has been associated with increased bone loss in some studies—and most physicians believe that a low TSH on TRT heralds bone loss. However, thyroid hormone does not cause bone loss. Any increase in thyroid levels-effects increases metabolic activity throughout the body, including the rate of bone turnover. If a person is in a bone-catabolic state—is losing bone—higher thyroid levels will speed the bone loss and lower levels will slow the loss. Men and women begin losing bone at around age 30, and women suffer a 10-year period of accelerated bone loss around menopause. This bone loss with age is primarily due to reduced estradiol and testosterone levels. Menopausal women are thus most at risk for excess bone loss with TSHSupRx; which is preventable with estrogen replacement. Conversely, adolescent females, in a hormone-sufficient bone-anabolic state, have increased bone mineral density on TSHSupRx. Bone loss is not seen in men on TSHSupRx, as they retain significant testosterone and estradiol levels. The solution to bone loss is not to keep all patients hypothyroid, but to assure sufficient levels of bone-anabolic hormones and Vitamin D3. Muscle wasting is seen only in endogenous hyperthyroidism where FT3 and FT4 levels are 2 or more times the upper limit of their ranges. Due to the fast metabolism, more calories are needed. Muscle is broken down for gluconeogenesis if the diet does not supply sufficient calories. Anti-thyroid therapy that lowers the FT3 and FT4 to high-normal levels eliminates muscle breakdown, even though the TSH remains low. Again, this problem is due to high FT3 and FT4 levels, not a low TSH level.Thyroid hormone effects exist on a continuum in the cardiovascular system too. Hypothyroidism with its inadequate T3-effect produces high blood pressure, atherosclerosis, bradycardia, cardiac dysfunction and congestive heart failure. Hyperthyroidism, on the other hand, overstimulates the heart producing hyperkinesis, increased heart rate, excessive contractility, impaired diastolic relaxation, and thickening of the walls of the heart. These changes increase cardiac work and reduce exercise tolerance. Patients on TSHSupRx with an excessive T4 dose have echocardiographic and ergometabolic signs of thyroid hormone excess; especially when there is clinical evidence of thyrotoxicosis. Lowering the T4 dose eliminates the symptoms and cardiac abnormalities, even though the TSH remains low. A study of athyreotic patients on TSHSupRx found no cardiac symptoms and cardiovascular studies were similar to controls. The FT4 was high but the FT3 was identical to that of the controls. The authors concluded that in the absence of symptoms of thyrotoxicosis, patients treated with TSHSupRx may be followed clinically without specific cardiac laboratory studies. Again, the TSH level during TRT is irrelevant.Perhaps the main reason that physicians fear a low TSH during TRT is the risk of atrial fibrillation (AF). This is a valid concern, but must be understood in context. AF is common, affecting 1 in 25 of persons over 60 and 1 in 10 of persons over 80. By the age of 80, a person has a 22% lifetime risk of AF. Risk factors for AF are prevalent in the population. In addition to age they include family history, obesity, sleep apnea, diabetes, alcohol use and many medical disorders. It is true that any increase in thyroid hormone levels or effects increases automaticity and trigger activity in the pulmonary vein myocytes which initiate AF, and thereby increases the risk of AF in susceptible persons. The prevalence of AF rises with higher endogenous FT4 levels within the reference range: from only 3% near the bottom to 7% near the top of the range. Therefore any attempt to increase thyroid levels/effects with TRT entails a risk of triggering AF in susceptible patients; it does not require overtreatment but would be more likely with overtreatment. Logically, then, clinically-effective T4 or T4/T3 treatment that suppresses the TSH entails a higher risk of AF than less effective therapy that leaves the TSH within the normal range. The AACE/ATA TSH-based guidelines do minimize the risk of AF, but at the cost of universal underdiagnosis and undertreatment. This is neither medically nor ethically justifiable. The patient has the right to choose effective therapy. The physician and patient must weigh the risk of AF against the health and quality-of-life benefits of optimal thyroid levels and effects. The physician can resolve the ethical dilemma by obtaining informed consent for thyroid optimization therapy, with specific mention of the risk of AF. Fortunately, AF induced by TRT usually resolves with reducing the dose, except in older patients with significant underlying heart disease. The greater efficacy of T4/T3 combination therapy The elevation of the TSH in the face of thyroid gland dysfunction is a compensatory mechanism, stimulating more thyroidal production and greater T4-to-T3 conversion throughout the body. Lowering the TSH with TRT interferes with these mechanisms, reducing both thyroidal and peripheral T3 production. T3 production is reduced in central hypothyroidism due to inadequate TSH production. It is thus only logical that TRT should include T3 in addition to T4; especially when the TSH is low or suppressed. In addition, some 16% of persons have a genetic polymorphism of their D2 gene that impairs T4-to-T3 conversion. They have lower quality of life scores on T4 therapy and significant improvement with the addition of T3. TSH and TRH also directly induce mitochondrial biogenesis and activity;, higher T3 levels help to compensate for their absence. The effects of T4, T3 and T4/T3 combination therapy in various tissues and organs were revealed in a series of experiments with thyroidectomized rats. The investigators determined both serum and post mortem tissue levels of T4 and T3 in rats receiving continuous thyroid hormone infusions. A T3-only infusion failed to restore T3 levels in all tissues—illustrating the importance of T4-to-T3 conversion. A continuous T4-only infusion failed to restore T3 levels in the serum and some tissues until T4 levels were supraphysiological and the TSH suppressed, identical to our experience with oral T4 therapy in humans. A continuous infusion of T4 and T3 in the same ratio produced by the rat’s thyroid gland (6:1) allowed a normalization of both serum and tissue levels of both hormones without suppressing TSH. This implies that if we could give patients continuous infusions of T4 and T3 in the human thyroid’s 14:1 ratio we might need to only normalize the TSH in pure PH, where TSH production is vigorous (no partial CH). However, due to the peak levels with once-daily oral T4/T3 therapy, TSH production is over-suppressed, reducing thyroidal output and peripheral T3 production. To compensate and achieve optimal T3 effects in all tissues we need to include T3 in TRT. Many studies have compared various T4/T3 combinations with T4-only therapy. They provide abundant detailed clinical data on the effects of substituting various amounts of T3 for T4. In most studies, the doses were adjusted to keep the TSH normal; so they compared two forms of undertreatment. Still, in 4 of 14 studies, the authors concluded that T4/T3 therapy was superior. In the other 10 studies the differences generally favored T4/T3, for many if not most of the patients. Typically, authors dismissed non-significant trends favoring T4/T3 and the patients’ usual blinded preference for combination therapy.Statistical conclusions can be misleading; if one-half of the patients improve and one-half deteriorate, the result is no change. One must separate the two groups and determine why they reacted as they did. While oral T3 is generally 3 times more potent than T4 there can be a 3-fold range in their relative potency in various persons with treatment. Arbitrary substitutions will produce over-replacement in some persons and under-replacement in others. Doses that produce negative symptoms do not have to be excessive doses. The patient may have underlying hormonal or other medical conditions that make them intolerant of youthful/optimal thyroid hormone effects. We know that thyroid replacement worsens cortisol deficiency, so if the patient has any degree of hypocortisolism, more effective TRT that includes T3 will produce negative symptoms. Autoimmune diseases including autoimmune thyroid diseases are associated with lower cortisol levels.,, Many persons included in T4/T3 studies have Hashimoto’s or Grave’s disease, and in my experience these patients often cannot tolerate effective TRT due to a relative cortisol deficiency. From my review of the existing T4/T3 studies (see Appendix), I draw the following conclusions: Patients on TSH-normalizing doses of either T4 or T4/T3 have higher symptom scores than euthyroid controls, as noted in other studies. They are undertreated.At any given TSH level, T4/T3 combination therapy is more effective than T4 monotherapy, in objective scales and especially in subjective effect. Higher thyroid doses that produce low or suppressed TSH levels usually produce better clinical effects; both with T4 and T4/T3 therapy; but some patients do not tolerate such doses. Arbitrary substitutions of some amount of T3 for T4 can produce under- or over-replacement in a significant number of patients. Persons with low-normal or low TSH levels on T4 treatment experience the greatest improvement with adding T3 to their regimen; consistent with the fact that they have less thyroidal and extra-thyroidal T3 production.T3/T4 therapy is safe and causes no problems due to fluctuations in T3 levels, even with once-daily T3 doses. In sum, the T4/T3 combination studies support the hypothesis that the addition of T3 to T4 therapy is beneficial for most if not all patients, and also support the need for clinical thyroidology—for the adjustment of T4 or T4/T3 combination therapy to produce optimal clinical effects for each patient, without regard for the TSH level. What is the ideal proportion of T4/T3 for oral thyroid replacement? I do not know; it probably varies from patient-to-patient. Once-daily oral replacement certainly requires a lower T4/T3 ratio (more T3) than thyroidal production due to the factors mentioned above. This explains the popularity (with patients) and efficacy (in my and other clinicians’ experience) of natural desiccated thyroid (NDT) products. NDT has a T4:T3 ratio of around 4:1, similar to the optimal ratio found in the T4/T3 study by Taylor. This ratio provides much more T3 than thyroidal production; whether a different ratio would be best for most persons is a project for future research. NDT is the only T4/T3 combination product available in the USA at this time. In Europe, synthetic combination products are available with T4:T3 ratios varying from 4:1 to 10:1. NDT was the only form of TRT prescribed in the United States until the 1970s when synthetic T4 became available—it worked perfectly well for millions of patients for many decades. Pharmaceutical NDT brands in the US (Armour, Nature-Throid, NP Thyroid) are USP-certified, meeting the same standards for consistency as synthetic T4 and T3 products. They provide and convenient and inexpensive form of T4/T3 combination therapy. NDT may have advantages over synthetic T4/T3 combination therapy. NDT is composed of whole thyroid gland; it also contains T2 which has metabolic activity.,, 60mg (1 grain) of Armour Thyroid? contains 38mcg of T4 and 9mcg of T3. In spite of its decades of successful use, physicians often claim that it has not been studied. In fact, Armour Thyroid was recently compared to T4 therapy in a randomized, double-blind, crossover study. With doses adjusted to achieve the same normal TSH level, the clinical differences favored Armour and there were no adverse effects. Judged by its efficacy in lowering the TSH, 1mg of Armour Thyroid was found to be equivalent to 1.5mcg T4. So 60mg of NDT is similar in its effect on the TSH to 90mg of levothyroxine.The clinical diagnosis of hypothyroidismClinical thyroidology requires all the skills of clinical medicine: taking a history, gathering the symptoms, querying the patient for diagnostic clues, and performing a focused physical exam. The diagnosis of hypothyroidism requires clinical suspicion; the physician must be aware of the many symptoms that can be caused by hypothyroidism, (Table 1.) and must consider the diagnosis whenever there are no other medical explanations for the patients symptoms. While there are classical signs and symptoms of hypothyroidism, many persons have atypical presentations. The diagnosis of hypothyroidism, as of other diseases and disorders, is the clinician’s theory; the best theory that he/she can produce to explain the patient’s history, signs, symptoms and laboratory tests. The ultimate test of any diagnostic theory is a therapeutic trial. An often-overlooked problem that mimics hypothyroidism is iron deficiency. It is common in females due to menstruation. Iron deficiency even seems to reduce thyroid levels and effects. Non-anemic women with fatigue and ferritin levels under 50ng/ml experience improved energy with iron replacement therapy., Young women with ferritin levels under 20ng/ml experience improved energy and mental function with replacement. Anemia does not occur until ferritin levels are <5ng/ml or so. Thus a normal hemoglobin and hematocrit does not rule out symptomatic iron deficiency. So if a female’s ferritin is very low, I will usually try to see how much benefit will be obtained with optimizing her iron levels first; even when FT3 and FT3 levels are rather low. Initial thyroid testing should include a FT4, FT3 and TSH level. The TSH is, logically, unnecessary for assessing thyroid hormone status. However it is inexpensive, readily available, and provides immediate evidence regarding thyroidal and HP function. Elevated total and LDL cholesterol levels suggest hypothyroidism. A serum prolactin may be mildly elevated in hypothyroidism. Thyroid antibody testing is not relevant to either diagnosis or treatment of hypothyroidism, but does help determine the cause of PH. Persons with Hashimoto’s thyroiditis need not be treated, even if the TSH is elevated, as long as they feel and function well. As stated above, many of them have rather low cortisol levels/effects and may not tolerate T4/T3 optimization therapy. Small doses of levothyroxine that satisfy the physician’s desire to see a normal TSH on laboratory testing may make the patient hypothyroid or may reduce cortisol levels/effects so as to cause negative symptoms (heart racing, fatigue, anxiety, insomnia, etc.). I have many Hashimoto’s patients who did not require TRT and did not progress to overt hypothyroidism. When the patient has symptoms and/or signs that can be explained by hypothyroidism and has relatively low FT4 and/or FT3 levels, the physician should offer him/her a trial of thyroid optimization therapy. The greater the number of hypothyroid symptoms and signs and the lower the FT4 and FT3 levels, the more certain is the diagnosis and the response to therapy. Certainly, any person with unexplained chronic fatigue, myalgias, depression, and/or cognitive dysfunction and relatively low thyroid hormone levels should be offered a therapeutic trial. If the patient experiences few benefits or worsens, then either they do not have hypothyroidism or they have hypothyroidism plus cortisol deficiency or some other disorder(s). The TSH is irrelevant to diagnosis. In fact the TSH level is normal in the majority of patients who consult me with symptoms of hypothyroidism and low-normal FT4/FT3 levels, and who respond well to T4/T3 optimization therapy. They thus have a form of dysfunctional CH, mixed CH-PH, or some form of thyroid resistance. In untreated persons, the FT4 is the more sensitive test of thyroid status as the FT3 is usually maintained by enhanced T4-to-T3 conversion. Symptomatic patients usually have a FT4 level between 0.7 and 1.2ng/dL (with assays that have an upper limit of 1.6 to 1.8nd/dL) and a mid-range or low-normal FT3 level. A mid-range FT4 with a low-normal FT3 also suggests hypothyroidism; and if both FT4 and FT3 are low-normal, the patient can be markedly hypothyroid. Myxedema coma has been reported with such levels. A relatively high FT3 can compensate for a relatively low FT4, maintaining clinical euthyroidism. Again, the primary criterion must always be the clinical state of the patient. TSH-normal patients with hypothyroid symptoms and low-normal thyroid hormone levels usually respond very well to effective treatment; others have reported the same. The treatment of patients with normal TFTs is often dismissed by quoting a single study in which symptomatic persons were given a fixed dose of 100mcg T4. It decreased the TSH to a lower point within range, increased the free T4 slightly, and did not increase the FT3. Such subreplacement doses of inactive T4 cannot improve most patients’ thyroid status long term and can even make patients more hypothyroid, as discussed above. This was clearly not an inadequate trial. What does constitute an adequate trial of TRT?Clinical T4/T3 thyroid optimization therapyAt this point, I can only discuss what I have learned from my own experience in attempting to optimize thyroid hormone levels and effects with T4/T3 combination therapy in more than a thousand patients with hypothyroidism of various kinds and degrees. The therapeutic goal is simple: The elimination, as much as is possible, of all symptoms and signs of hypothyroidism (Table 1.) without producing any signs or symptoms of thyrotoxicosis (Table 2.). The optimal dose is ultimately a decision reached by the physician and patient together—in light of the FT4 and FT3 levels. This is clinical thyroidology; it cannot be reduced to a number, scheme, scale, or a rule.I prescribe NDT preferentially, and adjust the dose by clinical criteria first and by the free hormone levels second, without regard to the TSH. As the TSH is usually normal initially, it is usually suppressed with therapy, often with initial doses that are clearly insufficient. Even in pure PH, clinically-optimized T4/T3 therapy usually suppresses the TSH, but there are some patients who seem well-replaced whose TSH is normal. Thyroid optimization therapy is one of the most powerful medical interventions a physician can make; it has powerful effects in every tissue. The physician who attempts to provide optimal T4/T3 replacement therapy must understand the actions of T3 in the various tissues/organs of the body, the interactions of oral T4/T3 with the patient’s own thyroid production system, the first-pass effects of oral T4/T3 therapy on the liver, and the effect of higher thyroid levels on other endocrine and somatic systems. Every patient is different; one cannot apply a simple algorithm. With the initiation of oral T4/T3 therapy, the exogenous hormone initially adds to the patient’s endogenous levels until his/her TSH and endogenous production decline, necessitating a dose increase. Thus it’s best to start T4/T3 therapy at a low dose and increase it at weekly or bi-weekly intervals to a moderate replacement dose, then test after 6 to 8 weeks. In the first months, one tries to obtain some symptom resolution and good T4/T3 levels. Patience is required because clinical improvements can continue to occur after many months on a sufficient dose. I have found it most efficient to begin healthy patients on 30mg of NDT (19mcg T4 + 4.5mcg T3) daily upon awakening. Once-daily AM dosing with NDT generally works well, but some patients feel better with splitting the dose. In healthy patients who appear to be cortisol-sufficient, I increase the dose by 30mg each week up to 120mg or more depending on body weight. In ill or elderly patients or those with suspected cortisol deficiency I will use 15mg NDT tablets and the dose every two weeks up to only 90mg daily. I then adjust the dose by clinical criteria and the 24hr. trough FT4 and FT3 levels. I do not tell patients to abstain from eating breakfast or drinking beverages for 1 to 2 hrs after their morning dose. Such recommendations are impractical for most people. I tell them to take their dose immediately upon awakening and then go about their usual morning routine. If with that routine, with their usual timing of eating and drinking, they tend to absorb less thyroid hormone, I will increase the dose as needed. NDT is inexpensive and there is no need to suffer every morning in order to maximize its absorption. I do advise them to continue to follow the same routine. Blood should be drawn in the morning, prior to taking the thyroid dose. The timing of the test is important as after a daily dose of NDT the FT3 levels are supraphysiologic for several hours. This is necessary because NDT contains more T3 and less T4 compared to thyroidal production. T3 has a relatively short half-life of 17 to 24hrs, and its levels peak at 3 hrs. after a dose. T4 have a half-life in the serum of one week and much lower peaks after an oral dose. Most healthy persons who are well replaced on once-daily NDT therapy will have a 24 hr. trough FT4 that is low-normal, around 1.0 to 1.3ng/dL, and a FT3 that is in the upper half of range or even slightly high. The high FT3 levels through much of the day on NDT are of no concern and do not represent overdosing. FT4 circulates in amounts 4 times greater than FT3, and every molecule of T4 can be converted to T3. So a proportionately higher FT3 is needed to compensate for a lower FT4 on NDT. The high-normal or high 24 hr. avg. FT3 level compensates for the rather low level of FT4. As the TSH is usually suppressed, more T3 is needed to compensate for the reduced T4-to-T3 conversion. NDT thus is well-suited to the non-physiological process of TRT. Excessive NDT dosing should be suspected when the FT4 is high-normal and the FT3 high-normal or high at trough. RT3 levels are generally normal on NDT. A high RT3 may indicate overtreatment as the body protects itself by converting more T4 into RT3 instead of T3. A normal RT3 on NDT therapy provides some assurance that the dose is not excessive. However, FT4, FT3, and RT3 levels are only indirect indicators of end-organ effect. The primary determinant of optimal dosing must always be clinical criteria. A small number of my patients require FT4 and FT3 levels that are both high in order to feel well, and have no signs or symptoms of overdosing. We must treat the patient, not the numbers. Table 2: Signs and Symptoms of Excessive Thyroid DosingIncrease in malaise or fatigueHeat intoleranceExcessive sweatingIrritability, inability to relax Hand tremorLower exercise toleranceShortness of breathPressured speechPupillary dilatationInsomniaHigh systolic blood pressurePalpitations or rapid heart rateFrequent premature atrial or ventricular contractionsPerhaps the best way for physicians to begin prescribing T4/T3 therapy is to add 5mcg T3 tablets to the existing T4 regimen for their patients who remain symptomatic on TSHT4Rx. The T3 dose can be gradually increased by 5mcg every 6 weeks according to clinical criteria and FT4-FT3 testing. The TSH will become low or suppressed. The physician can be confident that the patient is not overtreated if there are no signs or symptoms of thyrotoxicosis and the trough FT4 and FT3 are normal with ratios of 5:1 to 10:1. In my experience, the 4:1 ratio of NDT works well for most persons, but a higher ratio may be better in general or for particular patients. The combination of a 25mcg tablet of T4 with a 5mcg of T3 is slightly higher dose than 30mg NDT. To use this 5:1 ratio, the healthy patient can start with one tablet each, and add one more of each every week or two up to 3 tablets each (75mcg T4 + 15mcgT3). A higher ratio may work well, and a few patients require a lower ratio or even T3-only therapy to eliminate their hypothyroid symptoms.,, On T3 monotherapy the FT3 must be high at all times, even at the 24hr. trough, in order to compensate for the complete absence of the more abundant T4. Any persistent signs or symptoms of thyrotoxicosis and any intolerance of the therapy require a reduction in the dose, at least temporarily. (Table 2.) It often happens that patients later need—and benefit from—a dose they were unable to tolerate earlier. As I have optimized thyroid hormone levels-effects with T4/T3 therapy that suppresses the TSH, I have seen a phenomenon not described in the medical literature. FT4 and FT3 levels on a given T4/T3 dose tend to fall in the first year or two and the dose needs to be raised several times. What I surmise is happening is that even with the suppression of TSH, the thyroid gland continues to produce significant amounts of T4 and T3. This endogenous production only gradually declines over a year or two. With TSH-suppressive therapy that is gradual atrophy of the thyroid gland. With TSH suppression there is also a gradual down-regulation of D2 and D1, reducing T4-to-T3 conversion. This process can even result in hypothyroidism with a suppressed TSH on T4/T3 combination therapy. So it is very important to monitor FT4/FT3 levels every 3 months or so for the first 18 months. After 2 years one can generally reduce the frequency of testing to once yearly. However, patients will sometimes change their morning routine and start eating or drinking sooner after their thyroid dose, causing a decline in their levels and a return of symptoms. After obtaining good serum levels and clinical effect, testing should be done every 3 to 6 months in the first 2 years. After that time the FT4 and FT3 levels generally remain stable on a given dose. The physician can then see the patient once per year to reassess the clinical criteria and FT4/FT3 levels. The physician and patient decide together whether to keep the dose the same, or try a higher or lower dose to see if a better clinical result can be obtained.References ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download