I think I may have figured it out. You may remember that back in May a report was released (which I blogged about), saying that the US infant mortality rate (5/1,000) is noticeably higher than that of many developed countries, eg Europe (3-4/1,000). Today I was forcing myself to read through the tedious chapter in Gabbe about fetal ultrasounds (I really want to be reading the part about operative deliveries, but I’m making myself read in order, because I’ll never come back to this if I once skip it).

Up to this point, there has been discussion about how, now that we’ve reduced infection and hemorrhage as major causes of maternal and fetal death, the biggest remaining contributor to perinatal mortality is birth defect. The authors at the end of the chapter on ultrasound discuss the value of “routine” fetal ultrasounds, as opposed to only those which are indicated (for one of 28 possible reasons, including inaccurate maternal dates, possibility of harm to fetus if dates are not very correct, suspected fetal death, suspected uterine anomaly, follow-up on screening tests suggesting fetal anomaly, determination of fetal presentation, etc etc). They note that in European countries, pregnant women routinely receive two ultrasounds, unlike in the US, where only 58% of women had at least one ultrasound per pregnancy. A survey of ACOG fellows (in 1989; and my edition of Gabbe is ten years old, 1996) reported that only 15% routinely scanned every pregnant patient.

So does this major difference in US vs. European practice have an impact on perinatal mortality? The authors cite a couple of studies showing (as one would expect) that routine sonography tends to give better dating (making intervention decisions based on preterm or postterm more accurate and more helpful), tends to diagnose twins earlier, thus allowing better observation, and tends to diagnose more placenta previas. Those are all clearly situations in which earlier diagnosis would tend to reduce mortality, or at least morbidity.

The most significant study cited is by Saari-Kemppainen in Finland in 1990 ( Ultrasound screening and perinatal mortlity: controlled trial of systematic one-stage screening in pregnancy. Lancet 336:387, 1990). 9,000 Finnish women in their second trimester were randomized to routine ultrasound screening vs. sonography only for the usual indication. Routine ultrasounds resulted in fewer clinic visits, better detection of twins, and better identification of placenta previa. The screened group had a perinatal mortality of only 4.6/1,000, while the control group had a perinatal mortality rate of 9.0/1,000. (Gabbe argues that this study is superior in design, ie more representative patient population, than an American study which showed no difference.)

Here’s the punchline, from Gabbe: “This reduction by almost one-half was due primarily to early detection of major malformations that led to induced abortion.” [emphasis mine]

Conclusion: The reason the US has a higher infant mortality rate is because we’re not killing as many deformed fetuses as the Europeans are. (And yes, I know they have a lower overall abortion rate.) So, if we want to drop our infant mortality rate to a less “embarassing” level, and become equal to the developed countries in that study, we need to implement routine ultrasounds (which has a price tag), and then abort the babies who are found to have lethal or very dangerous defects. Then, they won’t be counted under “perinatal” deaths, and we’ll look good – to men, though not to God.

(Just for a complete picture, in the chapter on preconception/prenatal care, Gabbe states: “[T]here are great international differences in the way live births are classified, as some countries exclude infants weighing less than 1kg and those with fatal anomalies.”)