Progesterone: When to Test in Your Cycle and What a Low Result Means

Progesterone is a steroid hormone produced in the second half of the cycle mainly by the corpus luteum - the structure formed from the ruptured follicle after ovulation. Its job is to convert the uterine lining into a state of 'readiness for implantation', to raise body temperature by about 0.3-0.5°C (hence basal body temperature charts) and to suppress further ovulations. If fertilisation does not occur, the corpus luteum regresses after about 12-14 days, progesterone falls sharply and menstruation begins. In the follicular phase (before ovulation) progesterone is low, usually below about 1 ng/ml.

This means progesterone is above all a biochemical trace of ovulation - 'proof after the fact' that the follicle ruptured and a corpus luteum formed. Without ovulation there is no corpus luteum, and therefore no clear rise in progesterone. That is why the test is a basic tool when cycles are irregular, infrequent or uncertain and a woman is trying to conceive. One reliable measurement at the right time is often enough to distinguish an ovulatory from an anovulatory cycle.

In practice it also matters that ovulation itself cannot be predicted exactly. In an ultrasound study of 50 women treated for infertility whose cycles were all ovulatory, the initial rise in serum progesterone occurred on the day of the LH surge in 30% of women, a day later in 40%, two days later in 22% and three or more days later in 8%. Follicle rupture was most often found 2 days after the LH surge. A spread of several days is why 'mid-luteal' is counted from ovulation, not from the first day of the period.

The luteal phase lasts 12-14 days in most women and is relatively constant, whereas the length of the follicular phase varies greatly. A more reliable method than counting from the first day of the cycle is therefore counting backwards from the expected period: the test falls about 7 days before it. If you know the day of ovulation (from an LH test, ultrasound or symptoms), test 7 days after it. With regular cycles that are longer or shorter than 28 days, adjust the date by this rule.

The blood draw does not require fasting, and time of day matters less than the timing in the cycle, although - as we will see in the next section - morning and afternoon results may differ. Hormonal contraception, a hormonal IUD and drugs containing progestogens confound interpretation: in women using hormones, measuring progesterone as proof of ovulation makes no sense.

Reference values differ between laboratories and methods (immunoassay versus mass spectrometry), and units may be given as ng/ml or nmol/l; the conversion is roughly 1 ng/ml = 3.2 nmol/l. The table below shows approximate values used in practice, but you should always follow the range on your own report.

Note: there is no single universal threshold for a 'normal' luteal phase. In clinical practice 10 ng/ml is often quoted as an approximate 'good' result in the mid-luteal phase, but this is a convention, not a threshold proven in prospective studies. Guidelines and textbooks give different cut-offs, and laboratories use their own ranges. A result of 8 or 12 ng/ml should therefore not be read as 'disease' or 'health' without the context of the cycle.

Progesterone is not secreted evenly in the luteal phase. A classic physiological study in which 15 healthy volunteers had blood drawn every 10 minutes for 24 hours showed that in the mid and late luteal phase progesterone concentrations can fluctuate sharply - from 2.3 up to 40.1 ng/ml, often within minutes - following LH pulses. The authors concluded explicitly that single progesterone measurements in the mid and late luteal phase do not accurately reflect the adequacy of the corpus luteum.

What does this mean for you? First, a borderline result (for example 6-9 ng/ml) in a single measurement says little - it may have fallen in a trough of a pulse. Second, a very low value (below 3 ng/ml) in a well-timed test is more reliable than a value 'a little below ideal', because it is hard to explain by fluctuations alone. Third, a better proof of ovulation than a single number is a set of clues: a positive LH test, ultrasound showing follicle rupture and a corpus luteum, a rise in basal temperature, and the next, correctly timed period.

Luteal phase deficiency (LPD) is the hypothesis that the corpus luteum produces too little progesterone or works for too short a time, so the uterine lining is not well prepared for implantation, which would explain infertility or miscarriage. The concept is popular online and in clinics but remains controversial in reproductive medicine. The most data come from a prospective study of 259 women aged 18-44 who for up to two cycles were assessed against two criteria: a shortened luteal phase (under 10 days - 'clinical' LPD) and luteal progesterone of 5 ng/ml or less ('biochemical' LPD).

Practical conclusion: a single episode of a 'low' luteal result is fairly common in women with regular cycles, while a recurrent one is rare. The two criteria overlapped only partly, which shows that there is no commonly accepted definition. The position of professional societies is cautious: according to the ASRM and SREI committee (2021), LPD is a clinical diagnosis associated with a luteal phase of 10 days or less, but it has not been proven to be an independent entity causing infertility or recurrent pregnancy loss, and controversy exists both over how to diagnose it and over whether treatment improves outcomes.

Before looking for a disease, make sure the low result is not due to a timing error or testing conditions. Only then is it worth considering true causes of missing ovulation or a short luteal phase. The most common are polycystic ovary syndrome (PCOS), in which ovulations are rare or irregular; functional hypothalamic amenorrhoea (too little energy intake, intense exercise, stress, low body weight); hyperprolactinaemia; and thyroid disorders. In perimenopause, anovulatory cycles become more and more frequent - we describe what that looks like in our article on the differences between perimenopause and menopause.

It is worth noting that LUF is an interesting caveat for the whole progesterone-based approach: in a study of 56 healthy women with regular cycles who had daily ultrasound, luteinised unruptured follicles grew faster and reached larger diameters than ovulating follicles, and hormone levels in those cycles differed from cycles with true ovulation. That does not mean LUF should always be suspected, but that with a long absence of pregnancy despite 'good' progesterone results, ovulation-monitoring ultrasound is worth ordering.

What this text does not cover: it is not a guide to infertility treatment, does not discuss progestogen dosing or luteal support in IVF cycles, and does not replace a consultation. The cited studies have limitations: Filicori's 1984 paper covered a small group of healthy volunteers, and the Schliep et al. study assessed women with regular cycles and does not prove that LPD criteria predict infertility. Progesterone thresholds differ between sources and laboratories, and their interpretation always needs clinical context.

Progesterone is one of the simplest and most underrated tests in the cycle - provided it is done at the right time. We recommend starting by recording cycles and taking LH tests, and only then drawing blood; a single measurement then answers a specific question: did ovulation occur. Results 'a little below ideal' in one cycle are better treated as a reason for observation and conversation than for immediate supplementation. We do not recommend hormonal creams and supplements promising to 'balance progesterone' without a medical indication.