Wednesday, 31 October 2007

More abortion

Via Badscience, the Commons Science and Technology Committe report on abortion is out today, (as is the minority report):
"The Committee concludes that while survival rates at 24 weeks (the current upper limit for abortion) and over have improved since 1990, survival rates (viability) have not done so below that gestational point. The Committee concludes that there is no scientific basis - on the grounds on viability - to reduce the upper time limit.

The Committee supports the removal of the requirement for two doctors signatures before an abortion can be carried out. The Committee is concerned that the requirement for two signatures may be causing delays in access to abortion services and found no evidence of its value in terms of safety.

Nurses and midwives with suitable training and professional guidance, should not be prevented by law from carrying out all stages of early medical and early surgical abortion. The Committee says that it found there is no evidence that this would compromise patient safety or quality of care.

On the issue of foetal pain, the Committee says the evidence suggests that while foetuses have physiological reactions to stimuli, this does not indicate that pain is consciously felt, especially not below 24 weeks. It further concludes that these factors may be relevant to clinical practice but do not appear to be relevant to the question of abortion law.

While new 4D imaging techniques are a useful tool in diagnosis of foetal abnormality, there is no evidence they provide any scientific insights on the question of foetal sentience or viability."

I haven't had time to read the whole report yet, but I was naturally drawn by morbid curiosity to the Minority Report (one of the two MPs writing it has a blog on it):
"We may never know for certain when foetuses first start to feel pain and there is no clear consensus amongst experts in the field. There are two main schools of thought. The first, represented to this enquiry byFitzgerald, Derbyshire and the RCOG, is that foetuses cannot feel pain until 26 weeks gestation, because that is the stage of development at which mature neural connections between the thalamus and cerebral cortex are first present. The second view, expounded in a review article by Anand et al published in Seminars in Perinatology in October 2007 (and also presented by the same author to the US Congress in 2005), is that foetuses feel pain using different neural mechanisms than adults and that these are present at earlier than 20 weeks gestation. Both schools are however agreed that conscious perception of pain cannot e inferred from observing anatomy, stress hormone levels and movements alone.
The alternative view supported by Anand et al argues that the more traditional Fitzgerald/Derbyshire/RCOG view ignores significant evidence, specifically that: a) sensory processing in the human brain develops well before birth;b)the subplate zone is functional well before the cerebral cortex develops; c) the key mechanisms of consciousness are located below the cortex (in areas that develop in early gestation); d) fetal behaviors suggest memory and learning as the highest-order evidence for perceptual function; and e) other lines of emerging evidence in the field of neuroscience."
Note that the two main schools of thought basically represent Anand on one side (published on the 31st of October, today, the first time this view has been published explicitly), and everyone else in the field on the other side, to claim that "there is no clear consensus amongst experts in the field" is simply false on this issue. I'm particularly intrigued by the claim, central to Anand's paper, and the minority report, that the mainstream view "..presupposes that cortical activation must be necessary for fetal pain perception. This reasoning, however, ignores clinical data that ablation or stimulation of the somatosensory cortex does not alter pain perception in adults, whereas thalamic ablation or stimulation does. If cortical function is not a necessary standard for adult pain perception, why must fetal pain be held to a higher standard?"

Note that it is obviously possible, as the thalamus represents the input tract to the cortex for thalamic damage to alter cortical activity and to cause pain even if the conscious perception of pain is mediated by the cortex (so called thalamic pain, an intractable and severe neuropathic pain) - after all, we know that damage to peripheral nerves can modulate pain too, but we don't think that pain is therefore perceived in peripheral nerves!

We also know that cortical pain circuits consist of much more than simply primary somatosensory cortex, including the insula (where direct stimulation can cause pain), and cingulate (where stimulation can relieve pain) and we know that cortical stimulation/inactivation, particularly to the motor cortex can relieve pain, including neuropathic and thalamic pain. So the whole point is simply invalid, the existence of pain/pain relief caused by thalamic damage does not prove that this is where conscious pain perception is mediated, nor does the failure of primary somatosensory cortex damage to modulate pain perception prove that it is not mediated by the cortex. This is such fundamental misunderstanding of the neuroscience that you have to wonder why they even bother making the point at all!

Here's some more:
"The RCOG in response to comments by Anand in a Channel Four Dispatches programme has issued a press release claiming they keep a 76 Scientific Developments Relating to the Abortion Act 1967 ‘watching brief on new scientific developments and advancements in fetal medicine, and continue to examine emerging evidence from the international scientific community about fetal awareness and fetal pain’ but are ‘unaware of the work of Dr Anand or any other work that contradicts the basic findings of (their) review’.
For the RCOG to report the studies of researchers who share their own official position, whilst ignoring research published by other leading researchers with contrary views, is at the very least misleading and at worst a serious abuse of power. It seems bizarre that the RCOG has not made more of an effort to find out more about contrary evidence before making such a bold public statement. It surely owes both Anand and Parliament a formal apology and explanation of why it has apparently ‘cherry picked’ the scientific evidence to support its opposition to a lowering of the 24 week upper limit for abortion."
Anand has not published any research on the issue of preterm foetal pain at all as far as I can see, although he has been an advocate of foetal pain in the media. Note that I have pretty much reported the entire foetal pain section of the Minority Report here, and the only reference is Anand's review, so it is unclear what this research that the RCOG has ignored is. Hopefully they don't think that Anand's utterly unsupported theorising on putative foetal specific pain circuits constitutes evidence?

Here's some more from Anand's review to get an idea of the quality, after claiming that cortex is not needed to experience pain:
"Some argue that activation of the sensory cortex is a necessary criterion for pain “perception” to occur in the fetus, citing the lack of evidence for pain-specific thalamocortical connections in fetal life...If cortical activity is not required for pain perception in adults, why should it be a necessary criterion for fetuses?"
As analysed above, they go on to say:
"Despite this caveat, robust cortical activity occurs in preterm neonates exposed to tactile or painful stimuli,30 which may be correlates of sensory content or its context and certainly imply conscious perception."
Given that we know that those scientists saying that cortex is necessary to percieve pain argue that it is not until about 28-29 weeks that this is possible surely the claim here must be that cortical activity is associated with pain before then? Nope, follow the link, "haemodynamic changes associated with activation of the primary somatosensory cortex. Forty preterm neonates at 28-36 weeks of gestation (mean=32.0)". I'm not sure how they reconcile the claim that somatosensory activity "certainly impl[ies] conscious perception" with their argument a few lines before that "cortical activity is not required for pain perception" but I suppose it fits with their next line of argument that:
"The subplate zone appears earlier in the somatosensory than in the visual area and reaches four times the width of the somatosensory cortex in the human fetus (2:1 in the monkey), implying that this embryonic structure expanded during evolution to subserve important sensory functions."
So, while somatosensory cortex is not necessary for pain perception, the existence of subplate cells in it proves that foetuses perceive pain? Ah well, I guess consistency isn't Anand's strong point. I think I fairly accurately summed up the subplate position before:
"the claim (based on currently no evidence at all) that the subplate mediates foetal pain, and then dies off in development, and the cortex then takes over pain perception (so foetuses have a special bit of the brain that only lasts a few months but is designed so they can feel pain whilst in the womb, but that then largely disappears to be replaced by a whole other neural system to serve the same purpose when the baby is actually born). So there is no reason to believe that this is the case, unless you happen to need to believe that there must be a mechanism that underlies foetal pain because you already believe that foetuses feel pain - can you say post hoc rationalisation?"

Belatedly here's a link to Ben Goldacre's Badscience column covering the provenance of some of the figures about preterm viability:
"I don’t blame Prof Wyatt, but the figure has taken on a life of its own. There may have been yet another mistake here, about the denominator. I don’t know. I’m quite prepared to believe that UCL may have unusually good results. But science is about clarity and transparency, especially for public policy. You need to be very clear on things like: what do you define as a “live birth”, how do you decide on what gestational age was, and so on. Even if this data stands up eventually, right now it is non-peer reviewed, non-published, utterly chaotic, personal communication of data, from 1996 to 2000, with no clear source, and with no information about how it was collected or analysed. That would be fine if it hadn’t suddenly become central to the debate on abortion."


Anonymous said...

Nice work pj.

Anonymous said...

Surprise, surprise. Perhaps pj, you could tell me what viability has to do with when you can and can't abort. I'm not sure what one has to do with the other.

pj said...

Nor am I. But it seems to be rather central to the debate in some quarters. And, since we're interested in science around here, it is always good to consider where the figures come from and how they're constructed.

You may have noticed that my main focus is on questions of foetal pain?

Anonymous said...

I noticed. i wasn't questioning your motives, but those of the people who were going to determine the policy.

pj said...

Ok. I think the viability thing is a hang up from when the law was drafted - and it was central to reducing from 28wks to 24wks in 1990(?). I keep intending to have a look at why that is, maybe I'll get the time to look at some of the reasoning from the time at some point.

Anonymous said...

It just seems to me that the minority report was simply trying to point out that the select committe was unfairly dismissive of the conflict of EXPERT opinion - concerning upper limits, foetal pain, health risks etc etc. I would have to agree that in matters of life and death - surely the precautionary principle cannot be regarded as an entirely ridiculous idea?!