There is an old rumor that the fetal heart rate is predictive of the gender of the baby. The 1980’s were an amazing decade for fetal sonography. Real-time and endovaginal sonography came into general use and provided views of normal pregnancies that had never been readily available before. At the same time personal computers were just coming into wide use. The juncture of these two technologies resulted in a measuring frenzy as excited sonographers documented everything that could be measured or that anyone could imagine measuring.
Austin, Texas was a buzzing with technonerds, still is. The Microelectronics and Computer Technology Corporation (MCC) located in Austin. Soon everyone else went to Austin: IBM, Texas Instruments, Dell, CompuADD, Advanced Micro Devices, Applied Materials, Inc., Sun Microsystems, 3M, Motorola, Novell, V-Tel, and even Apple (eventually). Everyone who was anyone in computers had an office in the South San Francisco Bay Silicon Valley area and in the Silicon Hills of Austin. There are two non-stop flights a day between Austin and San Jose. Go figure, they are called the "Nerd Birds". As soon as they get off the ground, everybody pops a laptop and goes to work.
Being in Austin as a sonographer was exciting. There was a particular interest in computers and sonography at Austin Radiological Association (ARA) in those days. Austin was growing, ARA was growing, everyone was busy. We were interested in fetal size and age, and began computerizing the measurements in 1984. In order to figure out how accurate our measurements were, we had to know the outcomes of the pregnancies... how much did they weigh?, when were they born?, were they normal or not?, and we figured we might as well see if our predictions of gender were accurate. So ARA gave 924 pregnant women return addressed, stamped postcards requesting birthing information. Slightly over fifty-one percent of the cards were returned (474/924)
One of the first questions asked, and easiest answered, was, "Does the fetal heart rate predict the sex of the baby?" We performed a t-Test of the means for females and males and found no significant difference (mean for females = 149.26, males = 149.38; p = 0.929; +/-1SD: females = 13.8, males = 13.1). The means were less than a quarter of a beat per minute (BPM) different, with two standard deviations being approximately +/- 26 BPM for both genders. A quarter of a BPM with that much variation is nothing.
As a part of the analysis we plotted the heart rate by gender against the age, and again, could not perceive any difference between the rates of girls and boys at any time during pregnancy. And while we did not find any differences between the heart rates for girls and boys, it was obvious that the heart rate varied with age. The following graph shows the heart rates from that original 1984-1987 data.
Pink circles = female heart rates. Blue squares = male heart rates. Cross-sectional data.
As more and more obstetrical examinations were done at ARA we gathered more and more data. By the end of 1993 we had over 10,000 sets of fetal parameters in a single database comprised of 50 fields for each record. There were 2457 beating hearts measured in 1993 alone. The following are the 1993 heart rates plotted against gestational age.
The very linear acceleration phase of early pregnancy meant that we could perform regression analysis on the embryonic heart rate. The following is what we found.
The human embryonic heart rate (EHR) accelerates at 3.3 beats per minute per day during the first month of beating. Beginning at a rate near the mother’s, at approximately 80-85 B/M at the start of the 5th week, the heart rate accelerates to a mean of 175 B/M (+/-20 = 2SD) at 9.2 LMP weeks. It then abruptly begins to decelerate. This increase is approximately 10 beats every 3 days until the early 9th week.
M-mode of 2457 heart rates recorded by eleven trained sonographers (ARDMS) are shown. Of this population, 274 embryonic heart rates were recorded by M-mode sonography before 9.2 weeks after the last menses (LMP).
Analysis of the crown-rump length (CRL) age before 9.2 LMP weeks and the EHR produced the following regression: AGE in DAYS = 11.5+EHR(0.27), (p < 0.001, r2 = 0.78). A modification of this regression formula yields the following simple relationship1:
EMBRYONIC AGE in DAYS = EHR(0.3)+6
+/- 2SD (95 %) = +/- 7 days; valid before 9.2 LMP weeks (64 LMP days)
|Follow-up found that 5 of 6 EHR ages that had fallen more than 7 days (- 2SD) behind the CRL ages ended in 1st trimester demise even when the measured heart rate had been > 95 beats/minute.2|
Twins with both EHRs recorded by M-mode. EHRs corresponded to CRL ages, respectively. About 6 weeks.
The fetal heart rate is not predictive of the gender of the neonate. The embryonic heart rate correlates well with age before 9.2 LMP weeks. In this population 5 out of 6 pregnancies ended in 1st trimester miscarriage in those cases in which the age by embryonic heart rate was a week or more behind the CRL age. This may mean the shape of the feed-back damping curve of the heart rate could be indicative of normal or abnormal cardiac development; however, more research is needed.