R. M. Pauli, M.D., Ph.D.

Most stillbirths occur a considerable time prior to delivery. A small minority, however, arise intrapartum. In this series, for example, of 409 stillborns, 45 were "unexpected". (Frankly I am very surprised that the number is even this high — our experiences would suggest otherwise, but perhaps this is because these babies are being successfully resuscitated out in the community and so we never learn of those events.) These authors ask a very important question: When a baby is born with an Apgar score of 0 (at one minute) what should be done? They attempt to answer this by assessing the outcomes of attempted resuscitation in this group of 45.

Initially 42 were successfully resuscitated. Of these, 62% survived to be discharged home. Overall, of the survivors, 38% had significant disability but, quite remarkably, then, 16 were free of any significant impairment through followup ranging from 20 m to 8 y. The authors indicate that poor outcome (non-survival, presence of serious sequelae) is most strongly correlated with late return of heart beat/respiration (among other factors). In fact, their tabulated data show a nearly perfect correlation that is of great clinical utility: No child with an Apgar score of 0 at five minutes had an even reasonably positive outcome. Most died. Two survived but had severe quadriplegia and severe disability. In contrast, an Apgar score of 1 or more at 5 minutes predicted a 75% probability of survival and a 66% chance of a good outcome long term. This suggests that babies who are unexpectedly stillborn secondary to intrapartum events should be resuscitated but, if the Apgar score remains 0 at 5 minutes, further resuscitative efforts are futile.

Pregnancy after perinatal loss: The relationship between anxiety and prenatal attachment. Armstrong D, Hutti M. JOGNN 27:183-189, 1998.

Sixteen currently pregnant women who had experienced a previous late miscarriage, stillbirth or neonatal death were assessed and compared with a group of primiparous women. The authors wished to assess whether a previous loss affected levels of anxiety in the women and, in addition, whether there were differences in apparent prenatal attachment to their now-expected babies. Although this is a tiny sampling, as the authors emphasize, it is understandable since finding their study population was challenging — only women in their second or third trimester of a subsequent pregnancy were eligible.

Not surprisingly, there was an increase in pregnancy-specific anxiety in the loss group. In addition, and perhaps more worrisome, this group showed less prenatal attachment. While the authors found no significant relationship between increased anxiety and decreased prenatal attachment, this could well be a problem of sample size. Decreasing anxiety level is probably an important part of prenatal care in these families. It may even prove crucial in establishing normal feelings of attachment.

Placental anatomy, fetal demise and therapeutic intervention in monochorionic twins and the transfusion syndrome: New hypotheses. van Gemert MJC, Major AL, Scherjon SA. Eur j obstet gynecol 76:53-62, 1998.

This article contains no new data. Why then is it of any importance? That importance derives for the most part from its rigorous assessment of how twin-twin transfusions result in fetal death. Likewise it provides similar analysis of the expected physiologic events following interventions. This is particularly crucial, for example, in understanding the limited success rate of laser interruption of vascular anastomoses between twins.

The authors postulate that there are fundamentally three kinds of vessel anastomoses in monochorionic twins: 1. Unidirectional arteriovenous connections. These result in slow transfer of blood from the arterial (donor) to venous (recipient) side and so for the most part cause potentially severe size discrepancy. 2. AV connections with what the authors term compensating anastomoses. Here discordance of size will be limited to whenever the higher mean vessel pressure of the recipient results in equalization of return flow. This will result in quite variable size differences. 3. Artery-artery or vein-vein anastomoses plus unequal placental sharing. Here, while there is initial discordance, the vessel communications actually help since they allow blood from the ‘overfed’ twin to reach the one who is initially ‘underfed’.

Now things get complicated! What happens if one twin dies? That depends on the specific kinds of anastomoses that are present. What happens if laser therapy is initiated to remove any connections between the twins? That, too, depends on understanding details of the anastomotic connections. For example, in the third mechanism mentioned above, interruption of the communication could actually be quite harmful.

There is lots more in the article. Anyone who has interest or involvement in intrauterine care of twin pregnancies would do well to study it.

Trauma in pregnancy: Maternal and fetal outcomes. Shah KH, Simons RK, Holbrook T, Fortlage D, Winchell RJ, Hoyt DB. J trauma injury infec crit care 45:83-86, 1998.

Severe injury during pregnancy is uncommon but not rare — with about 1 in every 300 pregnant women requiring hospitalization because of trauma.

This is a ten year case-control study of 114 injured pregnant patients treated at one institution. Most were injured in motor vehicle crashes (the second most common cause being personal violence). Both maternal consequences and fetal consequences were assessed.

There was little difference between the pregnant and nonpregnant groups regarding the woman’s survival and recovery. There were more abdominal injuries in the pregnant group (and less chest and head trauma). Mortality and morbidity were virtually identical.

Fifteen perinatal deaths occurred. Most (13/15) occurred prior to presentation to the hospital or within 2 hours. Of the two late losses, one likely was unrelated to the mother’s injuries. Fetal deaths, not surprisingly, correlated with each measure of severity of maternal injury. The only other predictor of fetal death was presentation with vaginal bleeding (presumably because of massive abruption associated with the injury). All abruption occurred within 6 hours of injury. Overall, then, this study concludes that maternal trauma only affects the fetus in the short term. If the fetus survives the trauma itself, if there is no massive abruption and if there is no premature rupture of membranes, probability for intact survival is likely no different than in those without such a prenatal experience.

Action of anticardiolipin and antibodies to b2-glycoprotein-I on trophoblast proliferation as a mechanism for fetal death. Chamley LW, Duncalf AM, Mitchell MD, Johnson PM. Lancet 352:1037-1038, 1998.

It is thought that antiphospholipid antibodies are important in certain intrauterine deaths. One class of these — anticardiolipin antibodies — are somehow associated with both recurrent early losses and later fetal deaths. Because these antibodies induce thrombosis, it has been assumed that the early and late intrauterine deaths are secondary to a thrombotic mechanism. This never has been quite satisfactory, however, in particular because placentae of women with anticardiolipin antibodies may, or may not, show infarction and thrombosis.

These authors asked, instead, whether anticardiolipin antibodies might directly affect trophoblast cells of the placenta. What they show in a rather neat set of experiments is that anticardiolipin antibodies inhibit trophoblast cell proliferation in vitro. Furthermore, this inhibition seems to be mediated through binding to an intrinsic trophoblast protein called ß2-glycoprotein I. How is this of relevance to antiphospholipid disease? This would suggest that only when extra-villous trophoblasts invade maternal vessels will exposure to the anticardiolipin antibodies occur and, when it does, this proliferation would be inhibited. Then, with inhibition of the trophoblasts in this region the number of ‘co-mingled’ maternal and fetal vessels would be limited. Finally, that suggests that these antibodies may act, not through thrombosis, but by inhibition of trophoblast proliferation resulting in inadequate nutritional exchange.

This is of potential clinical relevance. Most treatments (e.g. aspirin, heparin), both for early and late losses associated with anticardiolipin antibodies, assume that control of thrombosis will be helpful. This study suggests that at least for recurrent early losses this might not be at all rational.

Placental pathology in patients using cocaine: an observational study. Mooney EE, Boggess KA, Herbert WN, Layfield LJ . Obstet gynecol 91:925-929, 1998.

Many studies suggest an increased risk of placental abruption in pregnancies of women using cocaine. It is assumed that this results from the vasoconstrictive effects of cocaine. However, even in cocaine complicated pregnancies the frequency of abruption is only about twice that seen in pregnancies overall. Finding antecedents of abruption in cocaine exposed pregnancies might help figure out why some, but only some, end with placental abruption. These authors specifically sought evidence for chorionic villus edema and villus hemorrhage (thought perhaps to be antecedents to abruption). No significant differences were found. The relationship between cocaine use and abruption remains an enigma.

Beyond infant mortality: Gender and stillbirth in reproductive mortality before the twentieth century. Hart N. Population stud 52:215-229, 1998.

Off the beaten trail — how many of you read the journal, Population Studies? Me neither. This article does have some value. It emphasizes that stillbirth can be a valuable indicator of overall public health status. It attempts to provide historical estimates of the impact of stillbirth through the uncharted previous centuries. On the other hand it uses, to me, remarkably questionable assumptions to estimate those numbers, utterly ignores fetal causes of intrauterine death and is written, it seems, from a political rather than purely scientific perspective. In addition, I am bothered by authors who uses terms like ‘heroic’ and ‘ambitious’ to describe their own undertaking.

On the basis of the (very questionable) methods the author uses, stillbirth rates are estimated to have been around 100-150 per 1,000 births in the sixteenth and seventeenth centuries (10-15%!) and still around 50 per 1,000 by the mid-nineteenth century. These compare with rates of less than 6 or 7 per 1,000 now (all estimates being for intrauterine death after 28 weeks gestation rather than the usual, current definition of 20 weeks and beyond). What a different world it must have been for parents and families!

WiSSPers newsletter cover page
WiSSP home page