Tinkering With Levothyroxine: Does It Really Make a Difference?

Richard M. Plotzker, MD


August 06, 2018

People living in New York City have probably heard stories of the late Mayor Ed Koch riding the subways and randomly introducing himself to riders, asking, "How'm I doin'?" And people would tell him what they thought. A single personal story probably didn't have much of an impact on his city agenda, but common feedback did.

When patients return to the endocrinology office for their follow-up visit or are seen 2-3 days after a hospital consult, we ask, "How are you doing?" We ask in hopes of getting a response that either elevates our professional ego or shows us that the patient thinks well of our skills, and maybe even thinks well of us personally.

If you give insulin to someone with symptomatic hyperglycemia, you can count on that person being your friend forever. Patients who have had their hearts beating out of their chests from untreated hyperthyroidism can always tell the difference between how they felt before and after taking well-tolerated methimazole and a beta-blocker to restore them to good function, if not entirely good health. We are their heroes even though we did something generally straightforward, maybe almost foolproof. It's one of the joys of dealing with conditions that reverse in a predictable way and that patients notice.

Diseases of the laboratory, however, patients don't notice. Bisphosphonates are prescribed before fracture on the basis of diagnostic testing and historical risk, but fracture prevention is far from a sure thing, and our image becomes tarnished when the patient shows up in the emergency department with a T8 vertebral collapse. People at cardiac risk who were treated to reach accepted LDL-cholesterol targets may still have a myocardial infarction, and patients with diabetes who you thought were doing well may come back with a medicine- or foot-related misadventure. We use our best judgment, but the art of endocrinology has room for improvement.

Managing Hypothyroidism

No common endocrine disorder brings as many unhappy campers at follow-up visits than simple hypothyroidism, particularly among previously euthyroid patients who had their glands resected for tumors. Their TSH values may read therapeutic on levothyroxine, but those who abruptly become hypothyroid can tell that the medicine leaves them less than the completely functioning people they were before surgery.

Those who acquire hypothyroidism more gradually from radioiodine or Hashimoto thyroiditis also convey to the prescriber that they wish they felt better, though they usually don't have the sense that something is missing.

Endocrinologists and primary care providers who titrate levothyroxine to the desired TSH are invariably people of empathy who take patient symptoms seriously, and we wish we could do better. Some of us become micromanagers of TSH, tweaking the dose up or down by 12.5 or 25 µg in hopes of having a less symptomatic patient or at least a TSH that is a smidgen different from where it was during the last visit. Others are nihilists who just leave the TSH somewhere within the laboratory normal range of about 0.4-4.0 µIU/mL and do their best to cheer up their patients with reassurances that there is no threat to their longevity or function by keeping the TSH where it is.

Opinions based on experience or the latest conference can have a lasting imprint, but real data on this would be nice to have. Now, we just might have it.

A Study on TSH and Its Effect on Quality of Life

A recent study[1] conducted by a group in Oregon attempted to blind prescribers and patients to levothyroxine dosing while adjusting the doses up or down by 12.5-50 µg per visit to achieve one of three randomly assigned TSH outcomes: lower half of the laboratory's normal range, upper half of the normal range, or mild hypothyroidism (TSH, 5.6-12.0 µIU/mL). All pills were prepared by a single pharmacy with a gel coating that would obscure the characteristic dose-related tablet color.

Doses were titrated to the assigned level for each patient, and a variety of psychometric tests were administered along the course of the study. There might have been some slight evidence of decline in well-being in those who were targeted to the upper half of the normal range, but it was not consistent throughout the study. The investigators found that neither the target TSH nor the dose required to achieve the target TSH had any statistically significant effect on the various quality-of-life measurements used in the analysis.

One of the strengths of the study was its blinding of physicians and patients to the levothyroxine dose. This allowed patients—who were also blinded to their assigned target TSH—to select the interval of the study where they felt the best between visits. They tended to select the interval where they thought they were taking the highest dose, but what they thought was their highest dose was often incorrect.

Psychometric quality-of-life testing was essential to the conclusions, so I took it upon myself to see what was asked and what the variability might be. Many of the surveys asked about nondescript symptoms—those that would change from one visit to the next, particularly aches that might be transient—over a period as short as 7 days. For this reason, the decline in well-being in patients with a high-normal TSH during one part of the study seemed to resolve when multiple testing was considered.

My sense from this study is that the nihilist prescribers do better than those who frequently tinker with the dose. However, the paper also suggests that patients may have the perception of progress with a different dose, even though objective evidence is lacking. Overall, the science that has made TSH measurement one of the most accurate values that a lab can report may have lured prescribers into thinking we can do better than we really can.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.