Blood Tests During Menstruation: What Changes and What Doesn't

Oak
menstrual-cycle biomarkers lab-testing hormones longitudinal-data
A huge coiled translucent creature in a misty tidal basin, its segments glowing in sequence from red to blue, surrounded by ringed pools.

You can have blood drawn during menstruation, and for most of a standard panel it makes no difference. Complete blood count, comprehensive metabolic panel, liver enzymes, creatinine, TSH, HbA1c, and lipid panel are not meaningfully perturbed by the fact that you are bleeding. What menstruation does change is a narrow, well-characterized set of analytes whose values depend on where you are in the cycle, and for a few tests the early follicular phase is the only window in which the number means anything at all. The practical problem is less “is this safe” than “will this measurement be comparable to the one I took four months ago,” and that is a scheduling and metadata problem you can solve deterministically.

The tests that require early follicular timing

The reason clinics ask people to come in on cycle day 2 or 3 is that the hypothalamic-pituitary-ovarian axis is at its reference point then. Cycle day 1 is the first day of full flow, not spotting. On day 2 to day 4, estradiol and progesterone are at their nadir, so follicle-stimulating hormone (FSH) and luteinizing hormone (LH) are unsuppressed by negative feedback and reflect ovarian reserve rather than the state of a dominant follicle.

A day 3 panel typically includes FSH, LH, estradiol (E2), and often prolactin and TSH alongside. The FSH value is only interpretable when estradiol is also low, because an early-rising follicle can suppress FSH into the normal range and produce a falsely reassuring number. That is why the two are always ordered together. Anti-Müllerian hormone (AMH) is the exception in this group: it is produced by small antral follicles and is stable enough across the cycle that it can be drawn on any day, which is why it has displaced day 3 FSH as the first-line reserve marker in many settings.

Progesterone is the mirror image. It is measured roughly seven days after presumed ovulation, which is cycle day 21 in a 28-day cycle and later in a longer one. A day 21 progesterone drawn in a 35-day cycle is often measured before ovulation has occurred and reads low for a reason that has nothing to do with luteal function. If you track LH surge at home, count seven days from the surge rather than from day 1. Interpretation of any of these values belongs with a clinician, particularly if you are working through fertility questions.

Analytes that drift with cycle phase

Beyond the reproductive hormones, a second group moves enough across the cycle to matter for longitudinal comparison. C-reactive protein (CRP), the standard inflammation marker, varies systematically across the menstrual cycle, with higher values around menses and lower values near mid-cycle 1. If you are tracking high-sensitivity CRP in the 0.5 to 3 mg/L range, a menses draw compared against a mid-luteal draw can produce a difference that looks like a change in inflammatory status and is only a change in cycle day.

Cardiometabolic markers show the same pattern. A review of premenopausal cohort data argued that failing to account for cycle phase can lead to misinterpretation of lipids, glucose, insulin, and inflammatory markers, because estradiol and progesterone both influence these measurements across the cycle 2. Total and LDL cholesterol tend to be highest around the follicular phase and lowest in the luteal phase in these data. The magnitude is small relative to the difference between someone at 90 mg/dL LDL and someone at 190 mg/dL, and large relative to the 5 to 10 percent change you might be trying to detect after a diet intervention.

Hematology is the case where intuition most often misleads. People assume that losing blood during menses must drop hemoglobin on the draw. Typical menstrual blood loss is on the order of 30 to 40 mL per cycle, which is small compared with the roughly 5 L circulating volume and is replaced over days rather than showing up as an acute drop. A study of hematological athlete biological passport markers, which are among the most sensitivity-tuned longitudinal blood measurements in existence, examined hemoglobin, reticulocyte percentage, and derived passport scores across the menstrual cycle and found fluctuations modest relative to the individual reference limits the passport uses 3. For ordinary purposes, a CBC drawn during your period is a CBC.

Ferritin is the one iron marker worth thinking about, and not because of the bleeding itself. Ferritin is an acute-phase reactant, so it rises with inflammation independently of iron stores. Since CRP also moves with cycle phase 1, a ferritin value without a paired CRP is ambiguous in a way that compounds across the cycle. Order them together, and if the CRP is above about 5 mg/L, treat the ferritin as a lower bound on depletion rather than a measurement. Serum iron itself has a strong diurnal rhythm, often 20 to 30 percent higher in the morning, which swamps any cycle effect, so fix the draw time before you worry about the cycle day.

A protocol for your own longitudinal data

Our recommendation is to standardize on early follicular draws for anything you intend to trend, and to record cycle metadata on every sample regardless. Draw on cycle day 2 to 4, fasted 10 to 12 hours, between 07:00 and 09:00. That window gives you an interpretable FSH and estradiol at no cost, puts CRP at a consistent (if slightly elevated) point in its cycle, and eliminates the diurnal confound for iron and cortisol. If you draw quarterly, four early follicular draws per year are directly comparable to each other.

Store the phase with the sample rather than reconstructing it later. A minimal record that makes downstream analysis tractable:

{
  "draw_datetime": "2026-03-04T07:42:00-08:00",
  "last_menses_onset": "2026-03-02",
  "cycle_day": 3,
  "prior_cycle_length_days": 29,
  "lh_surge_date": null,
  "fasting_hours": 11,
  "hormonal_contraception": "none"
}

The hormonal_contraception field matters more than any of the others. Combined oral contraceptives suppress endogenous FSH, LH, and estradiol, so a day 3 panel drawn on the pill measures the pill. They also shift sex hormone-binding globulin substantially, which changes free testosterone estimates, and they affect several clotting and lipid parameters. Any model you fit across a period that includes starting or stopping hormonal contraception needs that as a covariate, not as a footnote.

For continuous glucose monitoring, the cycle is a real signal rather than noise. Segment your CGM traces by cycle day and compare the same phases across cycles before concluding anything about a dietary change, since luteal-phase differences in glucose handling are part of the same hormonal pattern that moves the cardiometabolic markers 2.

Menstrual blood as a specimen in its own right

A separate and more interesting question is whether the menstrual effluent itself can be used as the sample. Work comparing menstrual blood with paired venous serum has reported concordance for a number of analytes, including HbA1c and several reproductive hormones 4, following an earlier pilot comparison of serum and menstrual blood across a broader panel 5. Groups have built in-pad collection devices that detect ovarian health biomarkers directly from menstruation blood at home 6, with the same platform extended to a wider set of disease biomarkers 7.

The caveats are substantial and well documented. Menstrual blood is a mixture of endometrial tissue, cervicovaginal secretions, and peripheral blood, with variable hemolysis and local proteolytic activity, so analyte-by-analyte validation against venous reference is required rather than assumed 8. Proteomic mass spectrometry has been used to identify protein markers that distinguish menstrual blood from peripheral blood 9, and miRNA expression models can classify menstrual blood in forensic samples 10, which is direct evidence that the matrix is compositionally distinct. Treat menstrual-blood assays as promising and analyte-specific, not as a drop-in replacement for a venipuncture.

Questions people also ask

Can menstruation affect blood test results? For a narrow set of analytes, yes. Reproductive hormones vary by design across the cycle, CRP shows a systematic cycle pattern 1, and lipids and glucose-related markers drift enough that cycle phase should be recorded when comparing measurements over time 2. Routine chemistry, liver and kidney panels, TSH, and HbA1c are not meaningfully affected.

Is it okay to get a blood test while on your period? Yes. There is no safety concern and no contamination of a venous draw. If the order includes a day 21 progesterone, that specific test needs to be rescheduled to the luteal phase, and estradiol drawn during menses will correctly read low.

What do day 3 labs check for? FSH, LH, and estradiol, sometimes with prolactin and TSH. The early follicular window is chosen because estradiol and progesterone are at their lowest, so FSH reflects ovarian reserve rather than feedback from a growing follicle. AMH can be drawn on any cycle day.

Which blood test should be avoided during periods? None must be avoided, but a mid-luteal progesterone is uninformative during menses, and if you are trending hs-CRP or a lipid panel across months, pick one cycle phase and hold to it rather than mixing phases.

Can being on your period lower my hemoglobin reading? Not appreciably on the day of the draw. Typical per-cycle loss is small relative to circulating volume, and careful longitudinal hematology work found cycle-related fluctuation in hemoglobin and reticulocytes to be modest 3. Persistently low ferritin with heavy bleeding is a pattern to take to a clinician.

Oak builds longitudinal molecular profiles of individuals: whole-genome sequencing, RNA sequencing, proteomics, blood biomarkers, and continuous glucose data, integrated into one model of you. Build your profile.

Footnotes

  1. Katherine Wander, Eleanor Brindle, Kathleen A. O’Connor. C‐reactive protein across the menstrual cycle. American Journal of Physical Anthropology, 2008. https://doi.org/10.1002/ajpa.20785 ↩ ↩2 ↩3

  2. Enrique F. Schisterman, Sunni L. Mumford, Lindsey A. Sjaarda. Failure to Consider the Menstrual Cycle Phase May Cause Misinterpretation of Clinical and Research Findings of Cardiometabolic Biomarkers in Premenopausal Women. Epidemiologic Reviews, 2013. https://doi.org/10.1093/epirev/mxt007 ↩ ↩2 ↩3

  3. Jenny Mullen, Lasse Bækken, Helena Bergström, et al. Fluctuations in hematological athlete biological passport biomarkers in relation to the menstrual cycle. Drug Testing and Analysis, 2020. https://doi.org/10.1002/dta.2873 ↩ ↩2

  4. Sara Naseri, Maria I. Avrutsky, Carlo Capati, et al. Concordance of hemoglobin A1c and reproductive hormone levels in menstrual and venous blood. F&S Reports, 2024. https://doi.org/10.1016/j.xfre.2023.11.009 ↩

  5. Naseri S, Lerma K, Blumenthal PD. Comparative Assessment of Serum versus Menstrual Blood for Diagnostic Purposes: A Pilot Study. Journal of Clinical and Laboratory Medicine, 2019. https://doi.org/10.16966/2572-9578.130 ↩

  6. Lucas Dosnon, Thomas Rduch, Salma Sherif Azer, et al. At Home Detection of Ovarian Health Biomarker in Menstruation Blood. Advanced Materials Technologies, 2026. https://doi.org/10.1002/admt.71077 ↩

  7. Lucas Dosnon, Thomas Rduch, Charlotte Meyer, et al. A Wearable In‐Pad Diagnostic for the Detection of Disease Biomarkers in Menstruation Blood. Advanced Science, 2025. https://doi.org/10.1002/advs.202505170 ↩

  8. Amna Zaheer, Aqsa Komel, Mohammad Baraa Abu Bakr, et al. Potential for and challenges of menstrual blood as a non-invasive diagnostic specimen: current status and future directions. Annals of Medicine & Surgery, 2024. https://doi.org/10.1097/ms9.0000000000002261 ↩

  9. Heyi Yang, Erin Butler, Samantha Monier, et al. Identification of menstrual blood markers by proteomic mass spectrometry. International Journal of Legal Medicine, 2025. https://doi.org/10.1007/s00414-025-03655-3 ↩

  10. Erin K. Hanson, Mohid Mirza, Kamel Rekab, et al. The identification of menstrual blood in forensic samples by logistic regression modeling of miRNA expression. ELECTROPHORESIS, 2014. https://doi.org/10.1002/elps.201400171 ↩