The following paper was written in 1996, in response to the debate about whether fetuses feel pain. Comments or questions about it can be sent / e-mailed to the author at the addresses below.

Stuart W. G. Derbyshire, Ph.D. Asst. Professor
University of Pittsburgh Medical Center,
PET Facility, Room B-938 PUH,
200 Lothrop Street, Pittsburgh,
PA 15213-2582.
Phone: 412-647-0736
Fax: 412-647-0700
email:sderbysh@ucla.edu

The Science and Politics of Fetal pain - Doing the Wrong Thing?

In 1987, the Lancet published an article unequivocally demonstrating that neonates receiving fentanyl anaesthesia in preparation for surgery had improved clinical outcome as compared with neonates who only received nitrous oxide and curare (1). This research, and subsequent studies, (2)(3) led to a major reconsideration of analgesic practice with regard to neonates. In 1992, theNew England Journalran an editorial calling on clinicians to 'Do the Right Thing' concluding that 'it is our responsibility to treat pain in neonates and infants as effectively as we do in other patients'(4). Since then it has become common place to assume that neonates feel pain (5)(6). The assumption that neonates feel pain has led inevitably to speculation that the fetus may also experience pain (7). While the discussion about neonatal pain remained largely confined to the pages of medical texts, the discussion around fetal pain has attracted the attention of several major British newspapers and led the British parliament to discuss the curtailing of abortion (8)(9). Given the sensitivity of this issue in the United States (10), it is surely only a matter of time before this issue crosses the Atlantic. This article evaluates the evidence for and against fetal and neonatal pain and considers the implications for current clinical practice, abortion procedure and the contemporary understanding of pain.

The Evidence that the Fetus or Neonate can Feel Pain

Anand's seminal work on the use of fentanyl with neonates undergoing surgery demonstrated that the major hormonal response to invasive practice could be significantly reduced with fentanyl added to the anaesthetic regimen. Specifically it was demonstrated that plasma adrenalin, noradrenaline, glucagon, aldosterone, corticosterone, 11-deoxycorticosterone and 11-deoxycortisol levels were significantly greater in the non-fentanyl group than the fentanyl group up to 24 hours after surgery. The reduction of the 'stress response' to surgery by fentanyl was considered to be responsible for the improved clinical outcome of the fentanyl group who required less post-surgical ventilatory support and had reduced circulatory or metabolic complications. Anand and his colleagues later advanced these important and impressive findings in a report indicating that neonates receiving deep anaesthesia during surgery had improved post-operative morbidity compared with those neonates who received lighter anaesthesia. The reduced hormonal response and improved clinical outcome following invasive surgery in conjunction with anaesthetics used for pain relief in adults led naturally to the conclusion that the neonate could feel pain and that this pain needed to be controlled.

Dovetailing with the work of Anand and his colleagues was that of Fitzgerald. Fitzgerald has examined the developing nervous system of the rat and human fetus with special regard to the developmental neurobiology of pain (11)(12)(13). Fitzgerald has reviewed the biological development of the fetus and examined the possibility of fetal pain at each stage of development. The impression that a fetus experiences sensation is apparent at 7.5 weeks gestation when reflex responses to somatic stimuli begin. At this point touching the peri-oral region results in a contralateral bending of the head. The palms of the hands become sensitive to stroking at 10.5 weeks and the rest of the body and hindlimbs become sensitive at approximately 13.5 weeks. Shortly after the development of sensitivity, repeated skin stimulation results in hyperexcitability and a generalized movement of all limbs. This hyperexcitability has been interpreted as evidence for the presence of a functional pain system, reflecting an immature but intact pain response with early hypersensitivity to stimulation (14). This is not a view which is widely accepted, however, and is rejected by Fitzgerald herself. Prior to 26 weeks the thalamocortical fibres have not yet penetrated the cortical plate (15), and it seems unlikely, therefore, that the cortical structures considered necessary for pain are responding to noxious stimulation. The evidence for cortical involvement post 26 weeks is enhanced by behavioral studies which have demonstrated that the response to noxious stimulation becomes more focused and organized and can be better discriminated from other distress responses after 26 weeks (16). As with the hormonal response to surgery, the behavioral responses can be reduced with the use of appropriate anaesthetic adding support to the suggestion that these responses are related to pain perception (17).

Having established that the necessary neurobiology for pain is in place after 26 weeks and that behavioral responses to noxious stimulation are present in very premature babies of approximately 26 weeks gestation, it is logical to suggest that a fetus of 26 weeks gestation or more will launch a similar hormonal response to invasive practice as that observed in the neonate undergoing surgery. In 1994 Giannakoulopoulos and his colleagues from the Queen Charlottes Hospital in London, England successfully demonstrated that intrauterine needling to obtain a blood sample from fetuses of 20-34 weeks gestation resulted in a hormonal stress response analogous to that seen by Anand et al seven years previously (18). They demonstrated that needling the innervated intra-abdominal portion of the umbilical vein rather than the placental cord, which is not innervated, resulted in increased cortisol and []-endorphin concentrations in fetal plasma. If this group can now demonstrate that the hormonal and neural 'stress response' can be prevented with the use of appropriate anesthetics then they will have mirrored the criteria which have led to the widespread acceptance of 'neonatal pain'.

The Evidence Against

The undisputed discovery that the neonate and fetus launch a hormonal and neural response to invasive practice can not be considered definitive proof that there is anexperienceof pain. Anexperienceimplies that sensations have been interpreted in a conscious manner. Even when combined with the observations of behavior and improved clinical outcome when using anesthetics, there is still no proof that there is anexperienceof pain. Although all of these phenomena are associated with the notion of 'pain', none of them adequately describe or explain the phenomenological experience of 'pain'. These phenomena may exist independently of conscious experience. The relationship between the physiological responses of nociceptors, the hormonal and other responses of the CNS and the behavioral outcome of these changes to the psychological response has yet to be determined (19).

Unless it can be reasonably suggested that the fetus has a conscious appreciation of pain post 26 weeks gestation, then the responses to noxious stimulation post 26 weeks are still essentially just behaviorally complexreflexresponses, similar to the responses prior to 26 weeks. Despite the importance of providing evidence for the conscious appreciation of pain, the fetal and neonatal literature largely tries to ignore this issue. Anand, for example, highlighted the clinical findings with neonates as being of greater importance than 'any philosophical view on consciousness and 'pain perception''. Giannakoulopoulos et al distanced themselves from any implied fetal pain experience with the statement 'a hormonal response cannot be equated with the perception of pain'. In a report for the British Department of Health (Foetal pain: an update of current scientific knowledge. A paper for the Department of Health May 1995) Fitzgerald even went so far as to say that 'true pain experience [develops] postnatally along with memory, anxiety and other cognitive brain functions' leaving confusion as to what the 'untrue' pain experience of a fetus may be. More recently Lloyd-Thomas and Fitzgerald have suggested that if feeling and pain are properly understood then the fetus cannot be said to feel pain (20).

Such equivocation is perhaps not surprising in view of the general failure of material interpretations, i.e. interpretations which focus specifically upon the biological properties of human beings (21), to deliver a coherent account of human consciousness (22). Nevertheless, if a proper assessment of neonatal and fetal pain is to be undertaken, then we should examine the structure of the psychological experience 'pain', as the biological structures have been examined, and then work backwards to the fetus and neonate to decide whether it is likely or possible that these psychological structures are in place.

As Fitzgerald has identified, pain experience is now widely seen as a consequence of an amalgam of cognition, sensation and affective processes, this amalgam is commonly described under the rubric of the 'biopsychosocial' model of pain (23). Pain is no longer regarded as merely a physical sensation of noxious stimulus and disease, but is seen as a conscious experience which may be modulated by mental, emotional and sensory mechanisms and includes both sensory and emotional components. The whole biopsychosocial concept emphasizes the multidimensional nature of illness, injury and pain rather than emphasizing pain as purely a physical fact of illness or injury. Pain has been described as a multidimensional phenomena for some time (24) and this understanding is reflected in the current IASP (International Association for the Study of Pain) definition of pain as 'an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage' (25).

If this 'multidimensionality' is the basis of conscious pain experience, it seems unlikely that we can attribute this experience to the neonate or unborn fetus, which is naive as to all the cognitive, affective and evaluative experiences necessary for pain awareness. This is accepted in the current definition of pain that is further extended to state: 'pain is always subjective. Each individual learns the application of the word through experiences related to injury in early life.' Pain does not, so to speak, spring forth 'from the depths of the person's mind' prior to any experience, but is gradually formed as a consequence of general conscious development.

A further reason to doubt the viability of fetal pain post 26 weeks gestation is the development of the fetal cortex. Although it is true that the thalamocortical fibres penetrate the cortical plate at approximately 26 weeks gestation, the cortical regions which have been identified as important in processing the suffering components of pain (26)(27) do not become fully responsive until after birth (28). These structures, especially the anterior cingulate cortex (29), have a plasticity which allows for learning and adaptation and therefore retain the capacity to have a more dynamic relationship with conscious awareness. Interestingly, this capacity is not shared by the structures associated with sensory detection, such as the somatosensory cortex, which develop earlier but are less likely to have an involvement in the processes associated with pain experience. The concept of pain as a product of learning and adaptation is pursued in more detail underImplications for Pain Research.

The Implications for Current Clinical Practice

The debate about fetal pain need not affect clinical practice involving the fetus or neonate. Evidence that the stress response, which the fetus and new-born launches in response to physical insult, has known detrimental consequences is acceptable even to those who do not accept that pain is experienced. New borns who have been operated on without analgesia show increased mortality compared with new-borns who receive analgesia. There is also evidence that early physically stressing experiences may produce detrimental changes in later responses to potentially painful experience, such as inoculation (30). Therefore, in the cases of invasive practice where there is a clear clinical rational for the use of anesthetics, which does not rely upon the additional diagnosis of pain, the withholding of analgesia for fetus' and neonates should remain an unethical practice. As exciting advances in fetal surgery are being made continuously, it is imperative that similar clinically orientated research be carried out with the fetus.

Where the clinical advantage of anaesthetic is less clear, however, it is probably wise to avoid potentially dangerous procedures for the fetus or neonate, and probable uncomfortable procedures for the mother. There are those who argue that, while there is no consensus on this issue, clinicians should act according to the precautionary principle of assuming that pain is experienced until it is conclusively proved otherwise. However, as few clinical procedures are entirely risk free it seems more appropriate to reserve any form of intervention for occasions when it is known to be necessary.

A consideration of vaginal childbirth helps to place the discussion about fetal pain into some context. Childbirth is known to give rise to the hormonal and neural stress response which has been used as evidence for fetal and neonatal pain, this has led at least one popular British newspaper to run an article questioning whether babies feel pain during childbirth (31). It seems unlikely, however, that a process which the very vast majority of people have passed through is having long term detrimental consequences, and there is evidence which suggests the contrary to be the case (32)(33). It is still unknown exactly what the consequences of a hormonal stress response may be both for adults (34) and newborns. Under these circumstances an increase in relatively problematic anaesthetic, or even surgical practices such as caesarean section, to avoid fetal/neonatal 'pain' seems unnecessary and irrational (35)(36).

The Implications for Abortion Procedure

Fetal pain is obviously an important issue for those carrying out fetal operations and other invasive practices, but it is also of interest for those involved in abortion procedure and for those motivated to restrict the current abortion legislation. The broadly accepted conclusion that recorded responses to noxious stimulation prior to 26 weeks gestation arereflexresponses, not dependent on conscious appreciation, is important as it eliminates much of the generated concern regarding abortion. In 1994 just 94 abortions, out of more than 160,000 carried out in the UK, were later than 24 weeks (37). If it is accepted that consciousness is essential to the pain experience, and that consciousness is contingent upon psychological development, it would follow that even after 24 weeks gestation it would be more appropriate to describe fetal reaction to stimuli as reflex responses than as pain.

Guidelines on the termination of pregnancy for fetal abnormality issued by the UK Royal College of Obstetricians and Gynaecologists (RCOG) draws on the work of Fitzgerald (1995) which suggests strongly that the immaturity of the fetal nervous system prevents conscious awareness of pain before 26 weeks gestation. The document argues that 'it follows that up to this gestation the method of abortion should be selected to minimise the physical and emotional trauma to the woman' (38).

Regardless of one's own views on whether late term induced abortions may cause pain to the fetus, the issue warrants special attention because almost all late terminations are of wanted pregnancies where the putative parents may be emotionally vulnerable. Often, during counselling, they express concern about what the fetus may 'feel' during an abortion. In these circumstances it is common for the putative parents to think of the fetus as a 'baby' and to attribute to it the qualities that they anticipated their child would have were it to be born. In such cases good sympathetic clinical practice would require steps to be taken to reduce the concerns of the woman.

In the UK, the RCOG recommends that measures to stop the fetal heart should be taken in all terminations after 21 weeks gestation. This is to ensure that there is no possibility of the abortion resulting in a live birth. After 26 weeks the guidelines suggest that it is not possible to know the extent to which the fetus is aware and so after this gestation it is suggested that 'methods used during abortion to stop the fetal heart should be swift and involve a minimum of injury to fetal tissue.' Even if the fetus is not aware, as we suggest, these guidelines would be appropriate to avoid unnecessary distress to the woman.

The paramount interests of the woman in abortion procedures is an important principle. Arguments that with viability the fetus becomes a patient and the doctors' responsibilities towards the woman need to be balanced against those of the fetal patient remain controversial (39)(40). The view that the pregnant woman is the patient while the fetus is cared for on behalf of the woman endures among many clinicians and is in my view the ethical stance (41)(42).

Concern about fetal suffering is raised by those who oppose abortion in principle as a reason to restrict some methods of abortion. In both the US and the UK legislative changes have been proposed which would outlaw a late abortion procedure known by gynaecologists as intact dilation and evacuation and by opponents of abortion as 'partial-birth abortion'. In both countries the method was defended by the medical establishment on the grounds that there may be circumstances when such practice was in the interests of the woman. It was for this reason that President Clinton exercised his right of presidential veto in respect of the Partial Birth Abortion Ban Bill of 1995 (H.R. 1833/S 939) which had been approved by Congress. Clinton correctly stated that: 'By refusing to permit women in reliance on their doctors' best medical judgement, to use this procedure when their lives are threatened or when their health is put in serious jeopardy, Congress has fashioned a Bill that is neither consistent with the Constitution nor with sound public policy' (43).

The Implications for Pain Research

The implications of accepting the notion of neonatal and fetal pain for pain research are profound. The consequence of such a view is to undermine the current theoretical outlook of most pain researchers, namely the 'biopsychosocial' model of pain, the undermining of the current definition of pain (44)(45), and the return of ideas more closely resembling the discredited ideas of 'specificity' theory(46).

In the absence of any conceptual framework to account for a fetal/neonatal experience of pain, the fetal literature is drawn inexorably towards the discredited ideas of 'specificity' and 'pain centers'. Within the discussion of fetal pain, pain fibres (or peptides or neurotransmitters) are proposed to be stimulated and relay information to suggested pain centers somewhere in the brain. As for specificity, a painful stimulus therefore becomes that which activates the pain center, and pain becomes activity in the pain center. Specificity theory, however, has long been rejected because the definition of pain based on a direct relationship between stimulus and response has failed to resolve many of the major issues in pain research. Interpretations of injury based on a direct relationship between stimulus and pain cannot account for the variable link between stimulus and pain experience. This variable link is well documented (47)(48)(49)(50), and is a consequence of the fact that pain experience is a multidimensional phenomena contingent upon processes involved in general conscious awareness, namely evaluative, emotional and cognitive processing. The biopsychosocial model of pain has also encouraged a less 'specificity biased' view of central pain neurology which has long been dogged by specificity theorists searching for pain centers (51). Classical neurology has viewed the central projection to the somatosensory cortex as essentially a pain center, a region necessary and sufficient for the experience of pain (52). The information about noxious stimuli that travels via the spinothalamic tract to excite the lateral group of thalamic nuclei interconnected with somatosensory cortex, undergoes few alterations between the spinal cord and cortex (53). Excitatory responses in monkey somatosensory cortex are generally restricted to both innocuous and noxious mechanical and thermal stimuli. Somatosensory neurons have receptive fields that are small or at least confined to one limb and always contralateral (54). Such a system is ideal for providing detailed information about the location and characteristics of particular noxious stimuli but is not well suited for processes associated with affective and cognitive responses to noxious stimuli. The conscious appreciation of pain cannot be explained within this system, instead a 'neuromatrix' (55) of regions, incorporating anterior cingulate, prefrontal and insula cortices which show a plasticity with learning and development, is proposed as necessary for the experience of pain. Functional imaging studies have now demonstrated that a number of cortical regions are activated in response to pain which conform to the concept of a neuromatrix (56)(57).

While the neuromatrix is an important step away from specificity, a step which is threatened by the concept of neonatal and fetal pain, so long as the neuromatrix is seen as sufficient for pain experience it will fall foul of the same problems that the materialist accounts of consciousness face and can ultimately be reconciled with neonatal and fetal pain. The only way to avoid the failings of materialism, avoiding the view that the higher mental functions are fixed a priori or that consciousness is a product of metaphysical forces (58), is to see consciousness, and within it the experience of pain, as a consequence of developmental processes which the fetus and newborn baby are yet to pass through. According to one developmental model of pain, stimulus information is eventually organized and elaborated in the central nervous system with respect to three hierarchical mechanisms (59). The first two mechanisms in the hierarchy are perceptual-motor processing followed by schematic processing. Both these mechanisms are considered preconscious. Perceptual-motor processing involves the activation of an innate set of expressive motor reactions to environmental stimuli. Schematic processing involves the automatic encoding in memory of the experience to produce a categorical structure representing the general informational and sensory aspects of pain experiences. A set of conscious abstract rules about emotional episodes and associated voluntary responses is proposed to arise over time as a consequence of self observation and conscious efforts to cope with aversive situations. While rather mechanistic and far from ideal, this model outlines how the pressure of interacting with others gradually forces the subordination of our instinctual, unconscious, biology to our developing conscious will.

The response of fetuses and neonates to invasive practice is a valuable research area that should lead to better clinical practice in the future. Basing this research upon the assumption that there is pain experience, however, could lead to the hasty introduction of unnecessary and possibly detrimental anaesthetic procedures as well as increasing the distress faced by those women who seek abortion. In addition, the focus on fetal pain is likely to result in a considerable challenge on the current understanding of pain - a challenge which will push back the past 30 years of pain research and undermine the contemporary conceptual framework for understanding pain. Such changes do not appear to be advantageous and may even be damaging to the pain field in general and to the treatment and understanding of nociceptive responses in the fetus and newborn baby.

References

(1) Anand KJS, Sippel WG, Aynsley-Green A. Randomised trial of fentanyl anasthesia in preterm babies undergoing surgery: effects on the stress response.
Lancet 1987; 1: 243-248.

(2) Anand KJS, Hickey PR. Halothane-morphine compared with high dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery.
N Engl J Med 1992; 326: 1-9.

(3) Fitzgerald M. Pain and analgesia in neonates. Trends Neurosci 1987; 10: 344-346.

(4) Rogers MC. Do the right thing: Pain relief in infants and children.
N Engl J Med 1992; 326: 55-56.

(5) Fitzgerald M. Developmental biology of inflammatory pain.
Br J Anaesth 1995; 75: 177-185.

(6) Anand KJS, Hickey PR. Pain and its effects in the human neonate and fetus.
N Engl J Med 1987; 317: 1321-1329.

(7) Richards T. Can a fetus feel pain?
BMJ 1985; 291:1220-1221

(8) Hansard 1995; 236 (136): 906-914

(9) Derbyshire SWG. Comment: Do fetuses feel pain during abortion?
Abortion Review 1995; 57: 1-2.

(10) Annas GJ, Caplan A, Elias S. Sounding Board: The politics of human-embryo research - avoiding ethical gridlock. N Engl J Med 1996; 334: 1329-1332.

(11) Fitzgerald M. Spontaneous and evoked activity of foetal primary afferents in vivo.
Nature 1987; 326: 603-605.

(12) Fitzgerald M. The prenatal growth of fine diameter afferents into the rat spinal cord - a transganglionic study.
J Comp Neurol 1987; 261: 98-104.

(13) Fitzgerald M. Neurobiology of fetal and neonatal pain. In Wall P, Melzack R, eds. Textbook of Pain. Churchill Livingstone, 1994: 153-163.

(14) Barr RG. Pain experience in Children. In Wall P, Melzack R, eds Textbook of Pain. Churchill Livingstone, 1994: 739-765.

(15) Mrzljak L, Uylings HBM, Kostovic I, van Eden CG. Prenatal development of neurons in prefrontal cortex: a qualitative Golgi study. J Comp Neurol 1988; 271: 355-386.

(16) Craig KD, Whitfield MF, Grunau RVE, Linton J, Hadjistavropoulos HD. Pain in the preterm neonate: behavioural and physiological indices. Pain 1993; 52: 287-299.

(17) Fitzgerald M, Millard M, McIntosh N. Cutaneous hypersensitivity following peripheral tissue damage in newborn infants and its reversal with topical anaesthesia. Pain 1989; 39: 31-36.

(18) Giannakoulopoulos X, Sepulveda W, Kourtis P, Glover V, Fisk NM. Fetal plasma
cortisol and ?-endorphin response to intrauterine needling. Lancet 1994; 344: 77-81.

(19) Wall PD, McMahon SB. The relationship of perceived pain to afferent nerve impulses. Trends Neurosci 1986; 9: 254-255.

(20) Lloyd-Thomas AR, Fitzgerald M. Reflex responses do not necessarily signify pain.
BMJ 1996.

(21) Crick F. The Astonishing Hypothesis: The Scientific Search for the Soul.
Simon & Schuster, 1994.

(22) Chalmers DJ. Facing up to the problem of consciousness. JCS 1994; 1: 1-16.

(23) Waddell G. A new clinical model for the treatment of low-back pain.
Spine 1987; 12: 632-644.

(24) Melzack R, Casey KL. Sensory, motivational and central control determinants of pain. In Kenshalo D, ed. The Skin Senses. Springfield Ill: Thomas 1968: 423-443.

(25) Merskey H. The definition of pain. Eur J Psychiatry 1991; 6: 153-159.

(26) Jones APK, Brown WD, Friston KJ, Qi LY, Frackowiak RSJ. Cortical and subcortical localization of response to pain in man using positron emission tomography. Proc R Soc Lond 1991; 244: 39-44.

(27) Derbyshire SWG, Jones AKP, Devani P et al. Cerebral responses to pain in patients with atypical facial pain measured by positron emission tomography. J Neurol Neurosurg Psychiatry 1994; 57: 1166-1173.

(28) Chugani HT, Phelps ME. Maturational changes in cerebral function in infants determined by 18FDG positron emission tomography. Science 1986; 231: 840-843.

(29) Gabriel M. Functions of anterior and posterior cingulate cortex during avoidance learning in rabbits. In Uylings H, Van Eden C, De Bruin J, Corner M, Feenstra M eds. Progress in Brain Research. NY: Academic press 1990; 85: 467-483.

(30) Taddio A, Goldbach M, Ipp M, Stevens B, Koren G. Effect of neonatal circumcision on pain responses during vaccination in boys. Lancet 1995; 344: 291-292.

(31) Feger H. Babies feel the pain of childbirth, say doctors.
Sunday Express 1996; Jan 28: 17.

(32) Sangild PT, Hilsted L, Nexo E, Fowden AL, Silver M. Secretion of acid, gastrin, and cobalmin-binding proteins by the fetal pig stomach: developmental regulation by cortisol. Exp Physiol 1994; 79: 135-146.

(33) Wenderlein JM, Ritz-Schafer R. Is moderate labor stress for newborn infants an advantage? Pilot study of postpartum weight gain of 791 newborn infants. Geburtshilfe Frauenheilkd 1994; 54: 65-68.

(34) Salmon P. Anxiety and stress in surgical patients. Br J Hosp Med 1992; 48: 531-532.

(35) Glasser M. Cesarean section: science or ritual surgery?
Med Hypotheses 1991; 34: 73-80.

(36) Francome C, Savage W. Caesarean section in Britain and the United States 12% or 24%: is either the right rate? Soc Sci Med 1993; 37: 1199-1218.

(37) Abortion Statistics Series AB 1995; 19 (HMSO).

(38) Royal College of Obstetricians and Gynaecologists, Termination of Pregnancy for Fetal Abnormality in England, Wales and Scotland. 1996; pp. 12.

(39) McCullough LB, Chervenak FA. Ethics in Obstetrics and Gynaecology. Oxford University Press, New York 1994.

(40) Chervenak FA, McCullough LB, Campbell S. Is third trimester abortion justified? Br J Obstet Gynaecol 1995; 103: 187-189.

(41) Sirisena J. Correspondence. Is third trimester abortion justified? Br J Obstet Gynaecol 1996; 103: 187-189.

(42) Derbyshire, SWG. Locating the beginnings of pain. Bioethics 1999; 13: 1-31.

(43) Furedi A. Clinton vetoes ban on abortion methods. Abortion Review 1996; 59: 7.

(44) Anand KJS, Craig KD. New perspectives on the definition of pain. Pain 1996.

(45) Derbyshire SWG. A response to Anand and Craig on new perspectives on the definition of pain. Pain 1996.

(46) Wall PD. Why the definition of pain is crucial. In Wall P, Melzack R, eds. Textbook of Pain. Churchill Livingstone 1989: 1-18.

(47) Beecher HK. Measurement of Subjective Responses. Oxford University Press, New York 1959.

(48) Kosambi DD. Living prehistory in India. Sci Ameri 1967; 216: 105-114.

(49) Carlen PL, Wall PD, Nadvorna H, Steinbach T. Phantom limbs and related phenomena in recent traumatic amputations. Neurology 1978; 28: 211-217.

(50) Melzack R, Wall PD, Ty TC. Acute pain in an emergency clinic: Latency of onset and descriptor patterns. Pain 1982; 14: 33-43.

(51) Jones AKP. Do 'pain centres' exist? Br J Rheum 1994; 31, 290-292.

(52) Albe-Fassard D, Berkley KJ, Kruger L, Ralston HJ, Willis WD. Diencephalic mechanisms of pain sensation. Brain Res Rev 1985; 9: 217-296.

(53) Vogt BA, Sikes RW, Vogt LJ. Anterior cingulate cortex and the medial pain system. In Vogt BA, Gabriel M eds. Neurobiology of cingulate cortex and limbic thalamus: A comprehensive treatise. Birkhauser, Boston 1993: 330-360.

(54) Kenshalo DR, Isensee O. Responses of primate S1 cortical neurons to noxious stimuli.
J Neurophysiol 1983; 50: 1479-1496.

(55) Melzack R. Phantom limbs, the self and the brain: The D.O. Hebb memorial lecture. Can Psychol 1989; 30: 1-16.

(56) Hsieh JC, Belfrage M, Stone-Elander S, Hansson P, Ingvar M. Central representation of chronic ongoing neuropathic pain studied by positron emission tomography. Pain 1995; 63: 225-236.

(57) Vogt BA, Derbyshire SWG, Jones AKP. Pain processing in four regions of human cingulate cortex localized with coregistered PET and MR imaging. Eur J Neurosci 1996; 8: 1461-1473

(58) Eccles JC. How the Self Controls its Brain. Springer Verlag, 1994.

(59) Leventhal H. A perceptual-motor theory of emotion.
Adv Exp Psychology 1984; 17: 117-175.