Congenital abnormalities and cervical ribs
Galis et al. (2006) in a retrospective study addressed whether selection against less than 7 cervical vertebrae in humans can be attributed to other, more harmful, congenital abnormalities that might occur simultaneously with 7cv ribs. They used records for fetuses and infants that had died up to age 1 that had been autopsied, including full-body x-rays, in an Amsterdam hospital over an eight-year period (598 cases). The x-rays were reexamined independently by two observers for vertebral changes at the cervical/thoracic and thoracic/lumber boundaries. Cases where the fetus died before the 14th week of gestation were not used because prior to this time ossification of the vertebrae would not be sufficient to distinguish the presence of ribs. More cases were eliminated because of poor image quality, disagreement between the two observers, other bones obscuring the regions of interest, or poor ossification even past the 14th week. About a fourth of the cases could not be used.
In 54% of the remaining cases, ribs were observed on the 7th cervical vertebra, indicating an anterior shift in the cervical/thoracic boundary. Other vertebral anomalies were much rarer, rudimentary 1st thoracic ribs (~2%), no or rudimentary 12th thoracic rib (~3%), rib on first lumbar vertebra (~6%). For the two types of anomalies at the thoracic/lumbar boundary, they each occurred less than 1% of cases when a 7th cervical rib was observed, so there was no indication that they occurred together. The reported incidence of 7cv ribs reported in adults is much lower, typically around 1%.
Based this decrease in incidence, the authors estimated what the survival rate to birth and 1 year old was when a 7th cervical rib occurs. Their calculations were based on the assumptions that for the entire European population 85% of pregnancies result in live births, with an 84% survival through the first year, with negligible additional deaths until adulthood. They used an estimate of 2% of adults with a 7th cervical vertebra, which was the upper confidence limit from a collection of studies on adult skeletons.
To produce the observed prevalence of 7th cervical ribs at birth, 1 year, and adulthood, they calculated persons with 7th cervical vertebra ribs have a survival rate of 22% at birth, and 17% at 1 year, while for those without have a survival rate of 92% at birth and only slightly lower at 1 year. The initial incidence of 7th cervical vertebra ribs would be about 11%. Cervical ribs are associated with about a 9-fold increased risk of death. There is thus strong selection against 7cv ribs.
In persons with 7cv ribs that survive beyond 1 year there is a several fold increase in some childhood cancers ( ). However, since this is a relatively rare occurrence it likely does not have as great an selective effect as earlier deaths and likely would not greatly alter the estimate of the initial rate of 7cv ribs.
The autopsies also recorded abnormalities. The authors classified the observed abnormalities as either minor or major, likely to be fatal. In only 11% of the cases were there no observed abnormalities, and the incidence of 7cv ribs was 31%. For cases with only minor abnormalities, the incidence of 7cv ribs was 44%. For cases in which at least one major abnormality was observed (58% of cases), the incidence of 7cv rib was 61%. There thus appears to be a connection between 7cv ribs and both minor and major congenital abnormalities.
Fourteen classes of abnormalities were each observed in more than 2% of the cases, with the largest class being bleeding disorders, which may include many disparate causes. In none of the other classes was the incidence of 7cv ribs substantially different from the overall incidence. This may seem surprising giving the many times lower survival if 7cv ribs are present. This is likely case of Simpson’s paradox, which can occur when two groups are much different sizes are studied. Although the presence of 7cv ribs is associated with an about a 9-fold increase in the probability of an early death, there are initially about 8 times as many embryos without 7cv ribs as with. Even if a particular abnormality is much more likely with 7cv ribs, none occur frequently enough to stand out.
Thus, in this study no particular abnormality that occurs solely in association with 7cv ribs occurred at a high enough rate to stand out. For the other abnormalities, their individual increases in the presence of 7cv ribs compared to no 7cv ribs were similar enough that again none stood out.
The authors attribute the association of 7cv ribs with lethality to pleiotropy effects from mutations that affect both anterior/posterior patterning at the cervical/thoracic vertebra boundary and deleterious consequences in other organs or physiological processes. As discussed in another section, this is often attributed to a higher degree of interactivity occurring during the so-called phylotypic p period of embryonic development.
This period occurs around the ? week of development in humans. Since in the Galis study was only possible for fetuses that survived at least until week 14, it did not include miscarriages occurring earlier. A frequent estimate of 30% is given for the fraction of conceptions that end prior to birth, higher than the15% figure Galis used, so it is possible that the mutations that result in 7cv are selected against even more strongly that this study suggests.
A better understanding of why 7cv ribs are associated with abnormalities is likely to occur only if some of these pleiotropic mutations can be identified. Presumably for ethical or privacy issues, studies such as Galis et al. (2006) have not reported much information on the medical circumstances which led to autopsies beng performed. It is thus difficult to determine whether how representative this study was of deaths in the general population. It was stated that the study included fetuses from abortions for medical reasons. Nevertheless, in another section I will discuss some possibilities for the genes from which these pleiotropic effects might arise.
Of possible note, while only 10 cases (~2%) were observed of rudimentary 1st thoracic vertebral ribs, indicative of an anterior shift, all were associated with major abnormalities.
Eight of these were observed to have a ventricular septum defect of the heart, which is often fatal in newborns if not treated. The article did not provide information if this abnormality had been detected and treated successfully at this hospital. This abnormality occurred along with cervical ribs no more frequently than expected if there was no association between them.
Posterior shifts over evolutionary time would produce species with more cervical vertebrae. Since near constant 7 vertebrae is much near the lower limit in birds ( ) than the upper limit ( ), the selection against posterior shifts seems yet to be explained.
One limitation of this and similar studies is that they are limited to fetal deaths after the 14th week, because the cervical ribs do not become sufficiently developed until this time. The commitment to produce a rib occurs early in the phylotypic stage (Galis and Metz 2001). Circulation begins in human embryos on the 4th week. Thus, if the 7th cervical vertebra is associated with death from heart or circulatory system defects, or defects in other organs that are required to function prior to the 14th week, these studies would have underestimated related deaths.
In all these studies, it is not clear whether the anterior shift, per se, is the direct cause (or raises the probability) of lethal abnormalities. However, the alternative would be that every mutation that would produce a favorable anterior or posterior shift would by associated with lethality by some other pathway. Otherwise, there would be no selection maintaining the 7cv constraint. Studies in mice of mutations in Hox and other genes sheds some light on this question.