Your labs are normal but you feel terrible. What is a thorough workup actually looking for?

My TSH is normal, so why am I so tired?

A normal TSH rules out the most common forms of overt thyroid disease. It does not measure how much active thyroid hormone is reaching the tissues, whether the thyroid is under autoimmune attack, or any of the many causes of tiredness that have nothing to do with the thyroid. TSH is the pituitary's signal to the thyroid, a request rather than a report of what arrived.

A full thyroid panel adds free T4, the storage form of the hormone; free T3, the active form that cells actually use; reverse T3, an inactive form the body makes more of during illness, calorie restriction and sustained stress; and TPO and thyroglobulin antibodies, the markers of autoimmune thyroid disease. Antibodies can be present for years before the diagnosis is made [PMID: 21715532], so elevated antibodies with a normal TSH is a finding worth following. A high reverse T3 with a free T3 low in its range suggests that conversion, not production, is the bottleneck; reverse T3 is best validated in hospitalized illness, so in an outpatient it is a clue to weigh, not a diagnosis.

When reference populations were screened rigorously for thyroid disease, more than 95 percent of healthy individuals had a TSH below 2.5 [PMID: 16148345], well under the upper limit most labs print, although whether the reference range should be narrowed is still debated among endocrinologists. A result in the upper part of the range, with symptoms, is a reason to run the full panel and repeat it.

When the whole panel is normal, the tiredness usually has another explanation. Ferritin, the protein that reflects stored iron, can be low while hemoglobin is normal; in a randomized trial, women with unexplained fatigue, normal hemoglobin and ferritin below 50 improved more on iron than on placebo [PMID: 22777991]. A serum B12 in the lower part of its range can hide a functional deficiency that methylmalonic acid, a compound that builds up when B12 is short, makes visible. Sleep apnea is underdiagnosed in women and in lean adults, and no blood test detects it. A single morning cortisol is one point on a daily curve, and fasting insulin rises years before fasting glucose does.

My hormone tests are normal, but the symptoms are real. Can a blood test even show perimenopause?

A normal hormone panel does not rule out perimenopause. The transition is defined by the pattern of a woman's cycles, not by a hormone level, and the levels swing widely from day to day and month to month during the years when symptoms are loudest. The staging system clinicians use for reproductive aging, STRAW+10, anchors the transition to changes in cycle length and bleeding pattern, with hormone markers as supporting information [PMID: 22344196].

A single draw captures one morning. Estradiol, the main form of estrogen, can read high, low or normal on any given day, because the ovary is still producing it in bursts. FSH, the pituitary's signal to the ovary, rises overall but fluctuates and can fall back into range in a month when ovulation happens. Progesterone is only meaningful in the second half of the cycle, roughly a week after ovulation; drawn on the wrong day it reads low in every woman, and drawn in a cycle without ovulation it reads low for a reason that matters. What no single result can show is variability, direction of change and whether ovulation occurred, which are the things that produce the symptoms.

A physician reading this picture starts with the history: cycle length and how it has changed, sleep, night sweats, mood, word-finding, joint aches. Draws are timed to the cycle when timing helps, repeated when one result is doing too much work, and read alongside thyroid, iron and blood sugar, which overlap with perimenopausal symptoms. Decisions about hormone therapy rest on symptoms, history and risk, not on a number crossing a line.

Is a vitamin D level of 30 low, or is it just not optimal?

A vitamin D result around 30 ng/mL is not deficient by any major guideline, and whether it is optimal is an open question. The Institute of Medicine concluded that a 25-hydroxyvitamin D level of at least 20 ng/mL, or 50 nmol/L in the units some labs use, meets the bone-health needs of nearly everyone [PMID: 21118827], and an international consensus group judged levels between 20 and 50 ng/mL safe and sufficient for skeletal health [PMID: 29851137]. Guidelines disagree about the minimal desirable level across a wide span [PMID: 28387318], and many labs set their lower flag above the Institute of Medicine's line, which is why a result in the high 20s often prints as insufficient.

Reference range and optimal range are different ideas. A lab's range describes where a tested population lands. An optimal range is a claim about the level at which a person does best. The Institute of Medicine noted that higher levels were not consistently linked to greater benefit and that some outcomes showed risk at both ends [PMID: 21118827]. The Endocrine Society's 2024 guideline found no clear evidence defining an optimal target for disease prevention in healthy adults and advised against routine testing without an established reason [PMID: 38828931]. Many precision physicians still prefer a result comfortably above the deficiency line, in the upper half of the lab's range, particularly with low bone density, autoimmune thyroid disease or poor absorption. That is a judgment, not a proven threshold.

What a physician does with a 29 or a 38 is mostly context: when it was drawn, since winter and summer values in the same person differ; whether there is a reason it matters more here, such as low bone density, malabsorption or kidney disease; and what calcium and parathyroid hormone, the hormone that rises when the body wants more calcium, say alongside it. Variation between assays is real enough to muddy the definition of low [PMID: 29851137], so a few points either side of a line rarely mean much on their own. The real question is whether this result plausibly explains how the person feels, or is a bystander that is easy to fix while the actual driver goes unmeasured.

Is a magnesium blood test accurate, and does a normal result mean anything?

A serum magnesium test measures accurately what it measures: the magnesium in the liquid part of the blood. Most of the body's magnesium sits inside cells and bone, and the body defends the blood level tightly, drawing on those stores when intake falls short. Serum magnesium is a feasible way to monitor changes in status, but it does not necessarily reflect total body magnesium content [PMID: 26069819]. A normal result rules out the severe deficiency that causes dangerous heart rhythms and muscle spasms; it does not rule out a milder, longer-standing shortfall.

The range most labs print was derived from the distribution of serum magnesium in a 1974 population survey rather than from clinical outcomes, and a subclinical deficiency can exist inside it [PMID: 28140318]. A value at the bottom of the range and one at the top are both normal on paper and are not the same finding; where a result sits, and whether it has drifted, says more than the flag.

RBC magnesium, measured inside red blood cells, is the test many people ask about next; some clinicians use it as a second look when serum is normal, since it samples an intracellular pool. Its limits deserve the same honesty: red cells are one compartment among many, and the evidence that it predicts symptoms or outcomes better than serum is limited. A thoughtful physician treats it as one more data point rather than the verdict.

The verdict comes from the whole picture: how much magnesium is coming in; whether the gut absorbs it, since long-term acid-suppressing medication, heavy alcohol use and inflammatory bowel conditions all reduce absorption; whether the kidneys are losing it, as happens with diuretic medication and poorly controlled blood sugar; and whether the symptoms people associate with low magnesium, cramps, palpitations, poor sleep, migraine, fit the rest of the story.

What does a thorough workup include?

A thorough workup for normal labs with real symptoms is broader than a screening panel. It is chosen from the history, so that every test has a reason, and it is read as a set. Fatigue that arrives with chest pain, breathlessness, fainting, sudden weakness, or unexplained weight loss and fevers is a different situation that warrants prompt care; chest pain or trouble breathing means calling 911 or going to the nearest emergency department.

• The full thyroid panel: TSH, free T4, free T3, reverse T3, and TPO and thyroglobulin antibodies
• Iron studies with ferritin, and a complete blood count read for more than anemia
• B12 with methylmalonic acid, folate, vitamin D and serum magnesium, with RBC magnesium in selected cases
• Fasting insulin alongside fasting glucose and HbA1c, the three-month average of blood sugar
• hs-CRP, a sensitive inflammation marker, and a comprehensive metabolic panel for liver and kidney function
• Cortisol measured across the day rather than once in the morning
• Sex hormones timed to the cycle, and a sleep evaluation when the history points at the night
• Screening for celiac disease and other causes of poor absorption when iron or vitamin levels keep falling

Two things separate a thorough workup from a long one: the trend, since the same marker across years of physicals says more than any single value; and the reading, since results at the unfavorable edges of several ranges at once form a pattern no individual flag would raise.

What kind of physician does this?

An internal medicine physician, an internist trained in adult medicine across the organ systems, is the natural home for a question that spans thyroid, iron, hormones, sleep and blood sugar. Specialists take over when the workup finds their territory: an endocrinologist for confirmed thyroid or pituitary disease, a gynecologist or menopause-focused clinician for the reproductive transition, a sleep physician when apnea is likely.

The title matters less than what happens with the result. The physician worth looking for reads each value against the person's history and earlier results, can give the reason for every test ordered, decides what would actually change the plan, takes responsibility for that plan, and follows up long enough to know whether it worked. That physician can work alongside a person's primary care physician and specialists, and coordinate with them.

The deeper picture

People who arrive after several normal workups share a pattern: every result was normal on its own, none was read together, and none was compared with the year before. Two markers at the edge of their ranges, one hormone drawn on the wrong day, one nutrient never measured, and a symptom that has been real the whole time. The opposite error is just as common: chasing every value toward an imagined optimum when only one or two of them ever explained anything. Telling those apart takes physician-level clinical rigor applied to a whole person rather than to a panel, and that reading is the daily work at Extend Medical.