After a positive test, hCG rises fast: about 5–428 mIU/mL at week 4, climbing a median 124% every 48 hours early on, then peaking around 25,700–288,000 mIU/mL at weeks 9–12. One number is rarely the whole story — the trend matters far more than any single value.
ACOG notes: a single hCG value cannot determine a pregnancy outcome; results have to be read against the 48-hour trend and an ultrasound.
Data comes from peer-reviewed literature (Daya 1987; Barnhart et al. 2004; Seeber et al. 2006; Clarke et al. 2026), published laboratory reference intervals, and ASRM and ACOG guidance, plus the Betabase community beta-hCG database — used only for singleton-versus-twin medians, which is self-reported community data rather than a clinical study. For falling levels after a loss, see our hCG after miscarriage guide.
What Is hCG and What Does It Do?
hCG — human chorionic gonadotropin — is the hormone your embryo releases once it implants. Its first job is to signal the corpus luteum to keep making progesterone until the placenta takes over around week 10.
ACOG notes: hCG’s main function in early pregnancy is to maintain the corpus luteum so progesterone secretion continues to support the uterine lining.
A hormone made by the pregnancy, not the mother
The hormone comes from pregnancy tissue rather than your body — the syncytiotrophoblast, the outer cell layer that becomes the placenta — Cole, 2012.
hCG only enters your blood after implantation, usually 6–10 days after ovulation naturally or 1–5 days after a blastocyst transfer; before that, there is nothing to detect.
Why clinics call it “beta hCG”
hCG has two subunits: an alpha shared with LH, FSH and TSH, and a beta unique to hCG. Home tests and lab assays target the beta piece — hence “beta hCG” on your report — Cole, 2012.
Blood and urine measure the same hormone: blood is quantitative, urine usually qualitative — so clinics judge a progressing pregnancy on blood.
hCG Levels by Week: The Normal Range Chart
Typical published ranges, week 3 through delivery. Read them as a corridor, not a bullseye — the gap between lowest and highest healthy values is enormous every week.
These are gestational weeks counted from the last menstrual period (LMP), which is how clinicians chart hCG. That means the table runs about two weeks ahead of how far along the embryo actually is, and further still if you ovulated late in the cycle. Check dating against your scan before concluding you are “off the chart.”
| Status | Typical hCG range (mIU/mL) |
|---|---|
| Not pregnant | <5 |
| 3 weeks | 5–50 |
| 4 weeks | 5–428 |
| 5 weeks | 18–7,340 |
| 6 weeks | 1,080–56,500 |
| 7–8 weeks | 7,650–229,000 |
| 9–12 weeks | 25,700–288,000 |
| 13–16 weeks | 13,300–254,000 |
| 17–24 weeks | 4,060–165,400 |
| 25–40 weeks | 3,640–117,000 |
If your lab report says IU/L instead of mIU/mL, that is the same number: 1 IU/L = 1 mIU/mL, and the unit change does not move you on this chart.
How should you read this chart?
Look at week 4: both 5 and 428 mIU/mL count as normal — an 80-fold spread inside one row, which is why a single number tells you almost nothing. Labs run different assays, so these ranges matter less than your clinic’s reference comment.
Weeks 4–5: the fastest-moving numbers
Early hCG climbs exponentially — the ranges roughly double week to week. A week-4 value of 200 mIU/mL can be perfectly normal, and so can 4,000 at week 5 — which is why clinics retest after 48 hours: two points define a slope, one point defines nothing.
Weeks 6–8: why the range explodes
By week 6 the span between low and high already exceeds fifty thousand mIU/mL, and clinics lean on ultrasound — once a gestational sac and heartbeat are visible, imaging carries more information than bloodwork. A reading of 1,200 mIU/mL at week 6 sits below the typical range but is not automatically a failure: the next 48-hour reading decides.
Weeks 9–12 and beyond: peak, then decline
hCG hits its highest levels at weeks 9–12, commonly quoted at 25,700–288,000 mIU/mL. After the peak it falls — sometimes to a small fraction of the maximum — then plateaus in the tens of thousands.
ASRM patient education explains: hCG normally declines after peaking at weeks 9–12; that fall is physiological and does not indicate a problem with the pregnancy.
A second-trimester draw that looks “low” against a first-trimester memory causes needless panic; after the first trimester, going down is normal.
When Does hCG Stop Doubling?
In most pregnancies hCG stops doubling by weeks 8–11, when it peaks between 25,700 and 288,000 mIU/mL, then declines slowly and plateaus.
How much rise should you expect in 48 hours?
“Doubles every 48 hours” is usually read backwards: people treat it as a passing grade when it is really the average performance of a healthy early pregnancy. The measured curve, not the slogan, is what clinics work from.
Barnhart and colleagues followed 287 patients whose pregnancies were ultimately confirmed as viable intrauterine, taking 861 serial measurements, and redefined what a normal rise looks like (Obstetrics & Gynecology, 2004):
| Interval | Median rise | Fastest | Slowest that still ended normally |
|---|---|---|---|
| 1 day | 1.50× (+50%) | 1.81× | +24% |
| 2 days (48 hours) | 2.24× (+124%) | 3.28× | +53% |
| 4 days | 5.00× | 10.76× | — |
Two things fall straight out of that table. A 48-hour median of 124% is more than doubling, so a 60% rise is not a failure — it sits below the middle and comfortably above the floor. And the floor itself is low: in the authors’ words, “the slowest or minimal rise for a normal viable intrauterine pregnancy was 24% at 1 day and 53% at 2 days.”
The same analysis carries a second finding that undercuts a popular chart. In these viable pregnancies the rise was linear on a log scale (“a linear increase in log hCG best described the pattern of rise”), meaning the daily percentage gain stays roughly constant rather than tapering off as numbers climb. The widely copied “under 1,200 mIU/mL doubles in 48–72 hours, over 6,000 takes 96+” tier table has no counterpart in this dataset, so we do not reproduce it here.
The doubling framing does expire on its own: past roughly 6–7 weeks, and certainly after the peak, healthy pregnancies stop doubling, and measuring yourself against a first-trimester rulebook manufactures panic.
So where should the alarm line sit? A follow-up analysis by the same group answered it:
“Use of a more conservative cutoff for minimal rise in hCG, one as slow as 35% over 2 days, to characterize a potentially viable gestation would minimize potential interruption of a desired pregnancy.” — Seeber BE et al., Fertility and Sterility, 2006, PubMed
Hence the widely quoted 35% rise in 48 hours — a deliberately conservative floor. Rising slower does not diagnose failure but triggers closer monitoring; in Seeber’s analysis fewer wanted pregnancies were interrupted, at the cost of about 12% of ectopics missed.
Earlier work by Daya, tracking 29 normally advancing pregnancies dated by basal body temperature, had already shown the rise divides into three gestational phases with different doubling times (American Journal of Obstetrics & Gynecology, 1987) — the reason no single “doubling time” fits the whole first trimester.
The peak: weeks 8–11
Once production and clearance balance out — usually weeks 8–11 — levels stop doubling: compare later draws with the second-trimester range (about 3,640–117,000 mIU/mL) or your previous value, not first-trimester charts.
How Does IVF Change the Way You Read These Numbers?
IVF gives what most pregnancies lack: an exact fertilization date. Gestational week comes from the transfer day, not a guessed last period, making chart comparisons meaningful and letting clinics read your beta earlier.
Converting your transfer date to a week
After a Day 5 blastocyst transfer, gestational age on transfer day is 2 weeks 5 days; each day past transfer (DPT) adds one day.
Gestational age = (DPT + 19) ÷ 7 for Day 5 transfers; use (DPT + 17) ÷ 7 for Day 3.
So 9 DPT after a Day 5 transfer is (9 + 19) ÷ 7 = exactly 4 weeks. Our IVF week-by-week pregnancy calculator does the conversion; the implantation timeline guide confirms the rule.
Your day-9 beta: what the number says about the odds
Most clinics draw the first beta around day 9 after a Day 5 transfer (9dp5dt). A 2026 study of 6,410 single euploid embryo transfer cycles with a positive day-9 beta (>2.5 mIU/mL) turned that value into a ladder (Fertility and Sterility, 2026):
| Day-9 beta (mIU/mL) | Live birth outcome |
|---|---|
| 2.5 to <11 | 1.6% (lowest band) |
| 11 to <25 | 6.76× the band below (95% CI 3.02–15.11) |
| 25 to <50 | 3.08× (2.27–4.19) |
| 50 to <75 | 1.80× (1.59–2.04) |
| 75 to <100 | 1.16× (1.08–1.25) |
| ≥100 | 87.8% (highest band; 1.15× the band below) |
Read three things off it. 71.0% of all transfers with a positive beta ended in a live birth — a low start drags the odds down, it does not zero them. The four middle rows are adjusted relative risks between adjacent bands, which is what the authors reported; they are not separate percentages, and we have not converted them into any. And the ROC analysis puts the predictive accuracy of the day-9 value at an AUC of 0.83, with an optimal threshold of 75.7 mIU/mL — that is where the reassuring “what number should I be at?” figure actually comes from, rather than the round numbers copied around forums.
One caveat worth knowing: outcomes were similar across blastocyst biopsy day (5, 6 or 7) except in the 25 to <50 band, where Day 5 blastocysts came in lower (28.8%) than Day 6 and 7 (42.7% and 41.9%) — so read the ladder next to your own embryo report.
A separate 2025 study of 18,491 frozen blastocyst transfers found outcomes predictable as early as day 7–8, with live-birth cutoffs of 31.05, 46.6, 72.05, 95.4 and 159.9 mIU/mL for days 7 through 11 (F&S Reports, 2025). These odds describe groups, not individuals.
Why “normal” is wider than the chart suggests
Two people at the same week can differ by more than fiftyfold and both be carrying healthy pregnancies — in the chart above, week 4 runs from 5 to 428 mIU/mL, an 80-fold span inside a single row.
Most of that spread comes from where you started, not how fast you climb. Since the rise is roughly log-linear in early pregnancy (about +50% per day at the median), implanting two days earlier or later is enough to put two equally healthy pregnancies in different rows — and a last-menstrual-period due date cannot tell you which day implantation happened. It is the same mechanism behind a sonographer’s “baby measures a week small,” which usually means late ovulation rather than slow growth.
IVF patients are counterintuitively better positioned: your transfer date approximates implantation, so the one variable driving most of the spread is already pinned down. As for twins: in the Betabase community database (11,000+ reported pregnancies) twin medians run about 40–54% higher from 6dp5dt onward — roughly 232 versus 159 mIU/mL at 9dp5dt — but the ranges overlap so heavily that bloodwork cannot confirm twins; singleton values at that same draw span 2 to 8,270 mIU/mL. The twin hCG chart has the day-by-day medians.
What If My Numbers Are Higher or Lower Than the Chart?
A single value outside the range is not a verdict. Normal hCG differs 10- to 100-fold between healthy pregnancies at any given week, so clinics read your numbers as a trend alongside ultrasound — never against the chart alone.
Lower than expected: what doctors consider
When a number reads low, the first question is almost never “is it failing” but “is the dating right” — implantation off by a few days shifts everything, because early hCG compounds daily. A week-5 value that looks low may simply be a week-4 pregnancy.
If dating is correct and the rise is slow, expect a repeat test in 48 hours. One low value with a normal rise gets followed; a falling value starts a different conversation. Doctors also keep biochemical pregnancy in mind — a positive test whose numbers fall before any ultrasound could see one.
Higher than expected: does it mean twins?
Not reliably. In the Betabase community database twin pregnancies average 40–54% higher than singletons, but the ranges overlap enormously — twins can start in the “singleton” zone and singletons can sit at the top of the chart.
Only ultrasound, usually at weeks 6–7, can confirm.
Unusually high values occasionally reflect other conditions affecting placental tissue — hence the early scan clinics add when numbers climb faster than expected; that scan is routine diligence, not a diagnosis.
When should you call your clinic?
Numbers can wait for an appointment; these signals cannot: a rise slower than 35% over 48 hours, two consecutive plateauing or falling values in early pregnancy, or any sharp one-sided pain, shoulder pain, dizziness, or heavy bleeding. That last cluster can signal an ectopic pregnancy, a medical emergency — the ectopic pregnancy guide details the symptoms.
When in doubt, call the nurse line.
What Can Serial Testing Tell You — and What Can’t It?
Serial beta tests sort early pregnancies into three trajectories: rising adequately, plateauing, or falling. That is a blood test’s full reach — it suggests whether a pregnancy is progressing, not where it is or why.
The three patterns
A rise above 35% in 48 hours suggests a developing pregnancy; a plateau — two nearly identical values — is suspicious in early pregnancy; and a fall in the first trimester is the pattern most clearly linked to loss, though how fast it falls determines what happens next.
None of the three names the cause: a slow rise can come from a failing intrauterine pregnancy, an ectopic, or a viable pregnancy that started late.
The 29% of ectopics that look “normal”
In confirmed ectopic pregnancies, about 29% had hCG curves mimicking a normal pregnancy or completed miscarriage — 20.8% rose like a slow viable pregnancy and 8% fell like a completed loss — Obstetrics & Gynecology, 2006.
Numbers alone can never rule an ectopic in or out — not a reason to distrust the lab, but why ultrasound stays mandatory before any final call.
Ultrasound: the confirmatory step
The timing of ultrasound is pegged to hCG: the classic “discriminatory zone” — where a normal intrauterine pregnancy should be visible on scan — sits at 1,500–3,500 IU/L, the same value as mIU/mL (1 IU/L = 1 mIU/mL). A multi-platform comparison tested whether that number survives contact with your lab: readings did differ significantly between assays (71% of platform pairs diverged), yet five of seven platforms placed the zone within 9% of the target concentrations, and only one drifted as far as 40%. The zone is usable on nearly every common platform — but read it alongside the assay that produced your number, not as a universal constant — Fertility and Sterility, 2014.
Transvaginal ultrasound detects ectopic pregnancies with 87–94% sensitivity and 94–99% specificity — Clinics, 2019. Above the zone with no visible intrauterine sac, your clinic keeps looking — tubal pregnancy, or an early loss not yet cleared.
Patient Stories: Three Betas, Three Different Endings
These stories were shared with consent; names and identifying details are changed to protect privacy.
Maya, 34 — Bangkok, Thailand
Maya, 34, transferred one Day 5 blastocyst in Bangkok at a cycle cost of about $11,000. Her 9dp5dt beta of 62 mIU/mL rose 111% to 131 mIU/mL in 48 hours, a week-7 ultrasound showed one sac with a heartbeat, and she delivered a healthy girl at 38 weeks.
“The night I got 62, I decided it was over. I read every forum post in three languages. When the nurse called with 131, I sat on the kitchen floor and cried — then ate my first full meal in two days.”
Rachel, 37 — California, USA
Rachel, 37, started at 118 mIU/mL after a frozen transfer in California at a cycle cost of about $15,000. The 48-hour repeat rose just 38%, to 163 mIU/mL (above the 35% floor, below the 53% mark), the next barely moved at 193 mIU/mL, and an ultrasound found no intrauterine sac — a tubal pregnancy, treated with laparoscopic surgery the next day.
“The numbers were never bad enough to end the hope. That was the cruelest part — I stayed optimistic an extra week while the embryo was in the wrong place the whole time. If I could tell anyone one thing: when the rise stalls, the answer is in the ultrasound, not the next blood test.”
Daniela, 33 — Mexico City, Mexico
Daniela, 33, saw a 34 mIU/mL first beta rise only 21% in 48 hours, to 41 mIU/mL, during her second cycle (about $9,000) in Mexico City; a third test fell to 38 mIU/mL, and her clinic confirmed a biochemical pregnancy — one that never progressed, resolved before week 6. After a cycle off she restarted with a modified protocol.
“My doctor drew a tiny line at the bottom of the page and said: this is not a pregnancy we can save, and nothing you did caused it. I repeated that sentence to myself for weeks. It was still grief — but grief with the door left open.”
FAQ
Q: What is a normal hCG level at 4 weeks?
ACOG patient education states: a single hCG value cannot determine a pregnancy’s outcome; results must be read against the 48-hour trend and ultrasound findings.
Published week-4 ranges run 5–428 mIU/mL — any number in that band can accompany a healthy pregnancy, so clinics read it against dating and the 48-hour trend, not the chart alone.
Q: Can hCG levels be too high?
The most common explanations are simply a further-along pregnancy or twins — a “high” value alone rarely means trouble.
The healthy corridor is wide enough to absorb almost anything: 5-428 mIU/mL at week 4 alone.
Q: How often should hCG be tested after IVF?
First beta 9–14 days after transfer, then every 48–72 hours while the trend matters — typically 2–4 draws total; testing more often adds anxiety, not information.
Q: Can a healthy pregnancy start with low hCG?
Yes — starts of 2.5–11 mIU/mL carried a 1.6% live-birth rate in the 2026 cohort of 6,410 single euploid transfers, but odds climb steeply with each band and 71% of positive transfers ended in live birth. A low start means follow-up, not an ending.
Q: When does hCG peak in pregnancy?
Around weeks 9–12, typically 25,700–288,000 mIU/mL; the gradual fall afterwards is expected physiology, not a warning.
Q: Do twins have higher hCG?
On average yes — in the Betabase community database twin medians sit about 40–54% above singletons from 6dp5dt onward (232 versus 159 mIU/mL at 9dp5dt) — but overlapping ranges mean hCG can neither confirm nor exclude twins: singletons at that same draw range from 2 to 8,270 mIU/mL. Only ultrasound at weeks 6–7 settles it; see the twin hCG chart.
Q: What does it mean if hCG stops doubling?
Past week 8, that is simply the normal peak and plateau. Before weeks 6–7, a stalled pattern deserves a clinic call — it can accompany a problem or a slow-starting viable pregnancy.
Q: Can hCG drop and the pregnancy continue?
After weeks 10–12, yes — falling is normal; early in pregnancy a genuine drop is uncommon in viable pregnancies, so clinics usually add an ultrasound to look directly.
How to Follow These Numbers Without Losing Sleep
hCG is a signal, not a scoreboard. Five practical rules:
- Get your own clinic’s reference range first — platforms differ; only their comment applies to you.
- Read the trend, not the value. Two numbers 48 hours apart tell a story no single result can.
- Convert your dates correctly. After IVF, anchor to the transfer day — our week-by-week calculator does it.
- Learn your clinic’s thresholds and follow-up plan in advance, so a slow rise brings information, not panic.
- Do not sit on the red flags listed earlier — they mean calling now, not at the next appointment.
Choosing a clinic? Compare programs and follow-up styles in our hospital directory or contact our team; for which test to use and when, see the pregnancy testing after IVF guide.
Written by the ProIVF Medical Editorial Team and reviewed by the ProIVF Medical Advisory Board, based on peer-reviewed PubMed studies, published laboratory reference intervals, and ASRM and ACOG patient guidance.
Medical disclaimer: This content is for informational purposes only and is not medical advice. hCG patterns vary widely even between healthy pregnancies, and only your care team can interpret your results — always consult a licensed reproductive specialist about your specific situation.
Last updated: September 16, 2026
