History…and future of perinatal asphyxia research

14 Nov

I have added a history of perinatal asphyxia research to the placenta textbook. The association of birth asphyxia with cerebral palsy was first made by Dr. Little, a surgeon who developed a procedure to release the tendon in a spastic limb. The connection was not made by obstetricians not surprisingly because they were not in a position to accumulate the number of cases that Dr. Little saw in referral. It is hardly a novel idea that someone with a different perspective can provide a new insight into a problem outside their own special area of interest, but Dr. Little presents an archetypical example. Not nearly so dramatic, but I think pathology also has a useful perspective to offer obstetrics. For example, I have performed around 2000 perinatal autopsies, but an obstetrician will probably see far fewer than 100 such deaths in a lifetime. To me it is obvious that unexplained intrauterine asphyxia independent of labor is a major cause of intrauterine death. Yet, most obstetrical studies of stillbirth concentrate on risk factors but not on elucidating potential mechanisms of asphyxia. I will be presenting the evidence of the significance of asphyxia in stillbirth on the website, and hopefully I can interest someone.

Historically the most important insights inyo perinatal asphyxia were made by experimentalists. The foremost of these is Dr. Ron Myers. He inherited studies of asphyxia in which a healthy newborn monkey was totally asphyxiated at birth demonstrated a predictable course and onset of neurologic injury. Dr. Myers was also a neuropathologist which enabled him see that the neuronal lesions produced in the newborn monkey acute asphyxia experiments did not correspond to the devastating white and gray matter lesions seen in human cerebral palsy. More importantly he discovered the precise conditions under which the lesions of cerebral palsy can be produced in the fetal monkey. If the monkey has a partial or intermittent asphyxia prior to complete asphyxia, then there is injury to the endothelial blood brain barrier leading to cerebral edema and extensive brain necrosis. It is this one two punch that produces the devastating brain lesions seen with cerebral palsy. Clinically a fetus rescued from a sudden asphyxial event such as prolapsed cord or uterine rupture corresponds to the model producing only neuronal necrosis, and except in extreme prolongation of the asphyxia, will likely suffer less significant brain injury. On the other hand, the fetus with repetitive late decelerations who develops cardiovascular collapse is at risk of severe brain injury even if promptly rescued. The best prevention would be to reverse the mechanism causing fetal hypoxia before the severe asphyxia event. Research to that end would need to start by identifying the mechanisms. Of course some causes can be identified such as uterine tetany, maternal hypoxia, or placental abruption, but there is still a large unknown category.

In the monkey, Myers found numerous antecedent causes of the fetal acidosis including severe maternal stress, but he often manipulated the fetal monkey asphyxia by umbilical cord compression. Many other experimentalists have since used similar balloon cuffs around the umbilical cord to create patterns of partial, intermittent and complete asphyxia. Obviously balloon cuffs are not the natural cause of fetal asphyxia. There remains a role for obstetricians to discover the anatomic mechanisms underlying fetal asphyxia in many patients. It also wouldn’t hurt if research centers maintained and valued sheep physiologists with the expertise to test obstetrical hypotheses about the natural mechanisms of asphyxia. I believe understanding these asphyxia mechanisms is key to lowering Cesarean section rates, reducing cerebral palsy incidence, and preventing some stillbirths.

Long cord = short cord

14 Nov

A long umbilical cord is associated with an increased risk of neurologic injury (see Placenta: cord length on this web site for a review of the evidence)
I believe this is because a long cord paradoxically is evidence of a functionally short cord that in turn is responsible for fetal asphyxia. There are three steps to reach this conclusion.
1) Last year I was asked to participate in a gross pathology lab showing placentas. My first question to the medical students was, “How long does the umbilical cord have to be?”. Of course this was mean, since they were thinking I was asking about a number they should have read about. In my defense, the question does get them thinking (I hope). The answer of course is “long enough for the baby to get out”, approximately 40-50 cm. This assumes that the placenta stays attached to the uterine wall and the head and chest get free for respiration. The key concept here is that has to be a strong mechanism to be sure that the cord in fact is long enough for delivery.
2) There is ample clinical and experimental evidence that the cord does not lengthen normally if the fetus is paralyzed. This observation has led to the idea that fetal tension on the cord is needed to stimulate longitudinal growth of the umbilical cord. There is at least a plausible molecular mechanism for tension to stimulate growth in that Wharton’s jelly is full of growth factors that are inactive because they are bound in the glycosaminoglycan matrix of the jelly. Physically disrupting the jelly could conceivably release them to interact with the stromal cells in the jelly. There isa quantitative problem with this idea since the fetus is not able to pull the whole length needed, so there must be an overshoot to the stimulus. A little tug generates growth, and perhaps with active fetal movements this occurs even late in gestation. Imagine then that the infant wraps the cord around a body part very close to the placental insertion. The distance between the wrapped cord and the placenta could become just a few centimeters. If our mechanism for lengthening is correct, fetal movement should now create enough cord tension to lengthen the cord segment between the wrapped end and the placenta to a length needed for delivery. The cord will be long because is it will be the new added length plus the old length of the cord. Thus a long cord was at least for a period of time a short cord. There are a lot of gaps in understanding how physical force is transduced into growth, but this idea makes sense and is in agreement with evidence that infants with extensive wrapping have long cords. The alternative theory that long cords lead to wrapping has not been disproved, but there is little evidence for it. The second key concept then is that a long cord occurs because wrapping effectively created a short cord that grew long enough to permit safe delivery of the infant.
3) The cord can be too short. A medical student, Sean Brown, and I perfused the vein of umbilical cords with 50 cm of water pressure and wrapped them around pipes. The critical parameter was how much cord was left between the wrapping point and the end. There was a linear relationship between this length and the degrees of twisting needed to stop the flow in the vein ( I will get around to posting the actual data). We attributed this to the simple mathematical fact that twisting a short length increased the pitch more than the same degree of twisting did for a longer segment. If an umbilical cord is wrapped around the fetus in such a way that there is only a short segment between the beginning of the wrapped cord and the placenta, it should behave like our experimental model. Fetal twisting should increase the pitch of torsion on the vessels in the short segment. Likewise the fetus should be able to put more tension on the short segment of cord than on a long segment, and possibly even kink the cord to stop flow. This is exactly the scenario we expect in a long cord since the short segment will lengthen creating a long cord at delivery. Until that happens, the fetus is potentially in jeopardy. Thus, the last concept that a long cord implies that earlier in gestation there was the equivalent of a short cord, that is the segment between the wrapping and the placenta.
This logic explains why a long cord could be associated with a risk of neurologic injury in the fetus because it is actually evidence that there was a prior functionally short cord!

The clinical importance of this reasoning depends on proving that a functionally short cord is in fact a major cause of fetal asphyxia. I can only suggest the logic of this proposition, since we never receive the critical parameter of the distance from the point of wrapping of a cord to the placenta, even at delivery let alone from in utero measurements. The only exception is the notation of a tight nuchal cord, which of course is tight because the cord is now functionally short with expulsion of the fetus.

New Venue

12 Nov

I will be transferring some of the material from http://www.pediatricperinatalpathology.com to this WordPress site. I will concentrate on using it for now as a blog on obstetrical pathology rather than an online textbook. I encourage others to participate and start some conversations. I will be commenting on the published literature, on placenta cases, and on perinatal autopsies.