Journal of the Royal Society of Medicine Volume 83 July 1990
433
Atlantoaxial instability in Down syndrome - guidelines for screening and detection
M Roy MRCPsych Monyhull Hospital, Kings Norton, Birmingham B30 3QB M Baxter MBChB Chelmsley Hospital, Marston Green, Birmingham B37 7HL A Roy MRCPsych North Birmingham Community Mental Handicap Team, 66 Anchorage Road, Sutton Coldfield, West Midlands B74 2PH Keywords: atlantoaxial instability; Down syndrome; community survey; clinical radiological correlations; ethical implications
Summary A community survey was conducted in all adults with Down syndrome living in three health districts to see if there was any correlation between radiological and neurological abnormalities which could indicate the presence of atlantoaxial instability. There was no difference in the proportion of individuals with neurological abnormalities in the group with radiological abnormalities suggestive of atlantoaxial instability (6/14) compared with individuals with normal X-rays (50/123) as determined by the chi square test (0.01463: not significant). The clinical and ethical implications for screening of people with Down syndrome living in the community are discussed in view of these findings. Introduction Atlantoaxial instability in people with-Down syndrome has been extensively researched in the last two decades. Earlier workers reported an association between the two conditions14. In a survey of 158 children with Down syndrome, Pueschel5 noted the vast preponderance of asymptomatic cases. In a prospective study6 he emphasized the need for early diagnosis and treatment of symptomatic cases especially in the early stages. In a survey of mostly institutionalized adults, Alvarez7 found a reasonable degree of correlation between clinical and radiological abnormalities. He was able to ascribe a possible aetiological factor in neurologically abnormal but radiologically normal individuals. He then set out a clear statement of policy for screening in his hospital. There has been considerable interest in the nature of atlantoaxial instability itself. In a prospective study Burke et aL8 found progression of the lesion in children with Down syndrome over a period of time and warned against the increased risks of neurological deficits in boys over the age of 10 years. This feature of progression over time was to some extent refuted by Pueschel's study9 who said that the lesion progressed in some cases but regressed in some. Uncertainties about the nature of the lesion have made it difficult to arrive at a clear set of guidelines on screening and treatment10. A range of craniocervical abnormalities in Down syndrome have been described11 as have abnormalities specifically associated with atlantoaxial instability'2. As a result the Committee on Sports Medicine of the American Academy of Pediatrics published their guidelines13 which included among other suggestions radiological examinations on prospective participants with restrictions on sports 'that involve trauma to the head and neck'. British guidelinesl4 were broadly similar
and additionally cautioned about the need for support to the head of a person with Down syndrome while
travelling. In a recent review, Davidson15 felt that keeping in mind the rarity of cases of atlantoaxial instability that proceed to dislocation and cord compression, depriving large numbers of people with Down syndrome from various sporting activities based on radiological abnormalities may be too restrictive. He described a case of cord compression precipitated by strenuous activity in a person with previously normal cervical X-rays. He noted the high frequency of onset of neurological signs and symptoms prior to dislocation. A recent editorial in the Lancetl6 also highlighted the importance of neurological symptoms in people with Down syndrome. A separate factor to be taken into account was the changing nature of atlantoaxial instability itself as demonstrated in follow up studies89. Thus radiologically normal individuals who were then not X-rayed could develop a dangerous degree of instability unknown to those around them. Similarly individuals who had been restricted as a result of radiological abnormalities could become radiologically normal and be deprived of certain activities needlessly. In a CT scan studyl7 Miller demonstrated a correlation between radiologically diagnosed atlantoaxial instability and flattening of the cord on scanning. Since neurological examinations were not performed it was not possible to see the relationship between neurological symptoms and signs and cord compression. A community survey of adults with Down syndrome was undertaken to see if there was any correlation between neurological and radiological abnormalities and thus if more people with radiological evidence of atlantoaxisl instability had neurological abnormalities on clinical examination.
Subjects and methods A register for adults with Down syndrome was created in three health districts from all the existing day care and residential facilities in the mental handicap service. A letter was sent to their next of kin giving information about atlantoaxial instability and asking if they wanted their relative to have a clinical neurological examination and a radiological examination to detect the instability. The response rate was 92%. A proportion of people had already been X-rayed in response to parental concern and in preparation for Special Olympics. These people were examined neurologically with the examiner (MR) being unaware of their X-ray results. No objections were received
0141-0768/90/
070433-03/$02.00/0 © 1990 The Royal Society of Medicine
434
Journal of the Royal Society of Medicine Volume 83 July 1990
for the proposed examinations. The neurological examination consisted of the following: (a) tendon reflexes (biceps, triceps, knee, ankle) (b) plantar reflex (c) clonus (patellar, ankle) The tendon reflexes were graded: (i) not elicited; (ii) present; (iii) exaggerated The plantar reflexes were rated: (i) downgoing; (ii) upgoing Clonus was rated: (i) absent; (ii) present Clonus was rated present only if it was sustained throughout the period that stretch was applied at the patella and the ankle21. The age and sex of each subject were recorded. Patients with one or more exaggerated reflex and/or upgoing plantars and/or clonus were rated as having a neurological abnormality. Patients with an anterior atlanto odontoid distance of 3 mm or more on lateral cervical X-ray in flexion were diagnosed to have radiological evidence of atlantoaxial instability18-20. This was done by taking lateral cervical X-rays with the neck held in the neutral position and in flexion. The data was examined to determine the following: (i) the prevalence of radiological abnormalities (ii) the prevalence of neurological abnormalities in the two subgroups with and without radiological abnormalities. (iii) the sex distribution of the two subgroups (iv) the mean age ofthe individuals in the two subgroups Every individual with Down syndrome residing in the three health districts was included in the study. Six individuals were hospital inpatients at the time of the study. The chi-squared test was used to determine if the proportions of people with neurological abnormalities were different between the two radiologically distinct groups.
Results The number of people with Down syndrome who had radiological abnormalities suggestive of atlantoaxial instability was 14 out of a total population of 137 (10.2%). The distribution of neurological abnormalities is shown in Table 1. Nine men and five women (average age 30.5 years) had abnormal X-rays, of whom four men and two women had neurological abnormalities and five men and three women did not. Sixty-seven men and 56 women (average age 33.2 years) had normal X-rays. No individual had sustained clonus or upgrading plantar response. Discussion The prevalence of radiologically evident instability in this study was slightly lower than figures quoted Table 1. Association between radiological and neurological abnormalities
Abnormal Normal X-ray X-ray
Neurological abnormality No neurological abnormality Total
6 (43%) 8 (57%) 14
Chi-squared test 0.01463; not significant
50 (41%) 73 (59%) 123
in other surveys'0. There was a preponderance of men (9:5) over women. This was at variance with other surveys7 9 although the numbers were small. There was a predominance of men in the total sample. In the absence of true clonus and extensor plantar response, exaggerated reflexes were the sole indicator of neurological abnormality. Anxiety alone could not account for exaggerated reflexes and the ill-sustained clonus that were elicited. Since the extensor plantar responses is considered pathognomonic of an upper motor neurone lesion, the presence of other abnormalities could not be attributed to cord compression. It is possible that people with Down syndrome display exaggerated reflexes and ill-sustained clonus due to some other, as yet unknown, reason. This needs to be kept in mind while performing neurological examina-tions. There was no difference in the proportion of people with Down syndrome having exaggerated reflexes in the radiologically abnormal group compared with the normal group. This meant that abnormal neurological examination could not distinguish between the radiologically distinct groups and there was thus no correlation between neurological and radiological examiations. The -central difficulty facing clinicians and policy makers is that there is a need to be able to pick out those people with Down syndrome who are likely to develop atlantoaxial dislocation and cord compression. This has to be done without being alarmist and over restrictive. The most common assumption made is that a finding of anterior atlanto odontoid distance of greater than 3 mm means that the individual has the instability which can be the precursor to a full dislocation with its attendant complications. The unfortunate consequence of this assumption is that nearly one sixth of people with Down syndrome face restrictions on their activities16 the benefits from which could far outweigh the risks especially if it is accepted that 'there is no evidence that the current roentgenographic criteria of atlantoaxial instability are predictive of a tendency to dislocation'15. The changing nature of the lesion over time8'9 further reduces the predictive value ofthe X-ray abnormality. There appears to be a need to produce a set of sensitive and specific criteria which are flexible to take into account clinical, radiological and demographic correlates of an individual with Down syndrome. The first step in this process is to recognize the limitations of uniaxial parameters (such as abnormal X-rays) used currently and to perform a large clinico-radiological Down syndrome survey along with CT scans of the neck to confirm the findings of earlier studies'7 showing a correlation between radiologically suspeced instability and flattening of the cord. It would also clarify the association between abnormalities in the neurological examination (such as exaggerated reflexes) and varying degrees ofcord flattening, if any. If there was a correlation between the two then the presence of neurological symptoms such as pain in the neck, torticollis, gait abnormalities and neurological signs could help to establish the degree of cord flattening (or compression) that may exist. Alvarez7 reported that in his subgroup of radiologically normal but clinically abnormal cases he could find evidence of other pathology which could explain this anomaly. Abnormal neurological signs seem, therefore, to be rather nonspecific and could be present because of a variety of unrelated conditions. This needs to
Journal of the Royal Society of Medicine Volume 83 July 1990
be considered with Davidson's15 finding that in a significant number of cases atlantoaxial dislocation was preceded by the appearance of new neurological symptoms. It seems likely that symptoms are probably of much greater diagnostic significance than signs. It would then be possible to use a combination of symptoms, signs and X-ray findings to establish the risk of the individual developing atlantoaxial dislocation. All these clinical correlates would need to be matched up with demographic factors such as the age of the individual, his life style, likes and dislikes and the likelihood of him consenting to a series of examinations over a long period of time. This study was conducted in the community where the persons with Down syndrome exercised much more autonomy over their lives than their institutional counterparts and their active cooperation was essential at every stage. It was not sufficient to get agreement from the carers alone. In one adult training centre the trainees committee made it clear to the management that some of them did not feel that they had been adequately consulted on an individual basis prior to being subjected to the neurological and X-ray examinations. They also did not see the point of the exercise when they were, in fact, feeling well. This occurred in spite of obtining written agreement from carers prior to the study. This implies that any set of recommendations that relies on repeated examinations at frequent intervals is unlikely to be practical, acceptable and hence useful in community screening. It became obvious during this study that people with Down syndrome had to compete with other groups of patients for use of the radiology departments who were themselves unsure of the cost benefit ratio ofthis mass screening. In many instances it became clear the lifestyles of many of the participants in the study were so sedentary that the presence of any X-ray abnormalities in the absence of symptoms made no difference to their lives casting doubt on the need for total screening for Down syndrome populations. In conclusion, it may be worthwhile employing a stepwise approach to the process of detecting atlantoaxial instability and possible dislocation in people with Down syndrome living outside institutions: (i) Identification of the entire population from educational, family, residential and day care settings. (ii) obtaining agreement from the carers and if possible from the individuals themselves after as complete an explanation as possible. (iii) obtaining a neurological history from individuals and their carers to detect signs and symptoms of recently changing neurological status. (iv) examining the lifestyle ofthe individual to asse the need to undergo the testing procedure. (v) performing the clinical and radiological emmations. (vi) identifying the subgroup with both abnormalities as being at the highest risk of dislocation especially if displaying symptoms and signs of a recent onset.
Acknowledgments: The authors would like to thank the staff at various training centres, hostels and other establishments, relatives of participants, DaArkell, Lowe and Talbot (D)epartments of Radiology, Dudley, Telford and North Birmingham), Dr Oliver (Chelmsley Hospital), Dr Butler
(Ridge Hill), Dr Whiting (General Practitioner, Cornwall), Professor Corbett (Birmingham), Professor Simon (Birmingham), Profesor Roberts (Nottingham) and Mr Stanley (Bradford) and Miss J Rushton for encouragement and help with the project and with the manuscript.
References 1 Tishler JM, Martel W. Dislocation of the atlas in mongolism: preliminary report. Radiology 1965;84: 904 2 Dzenitis AJ. Spontaneous atlanto-axial dislocation in a mongoloid child with spinal cord compression. J Neurosurg 1966;25:458-60 3 Curtis BH, Blank S, Fisher RL. Atlantoaxial dislocation in Down's syndrome: report of two cases requiring surgical correction. JAMA 1968;205:464-5 4 Hungerford GD, Akkaraju V, Rawe SE, Young GF. Atlanto-occipital and atlanto-axial dislocation with spinal cord compression in Down's syndrome: a case report and review of the literature. Br J Radiol 1981;54:758-61 5 Pueschel SM, Scola FH, Perry CD, Pezzullo JC. Atlantoaxial instability in children with Down syndrome. Pediatr Radiol 1981;10:129-32 6 Peuschel SM, Herndon JH, Gelch MM, Senft KE, Scola FH, Goldberg MJ. Symptomatic atlantoaxial subluxation in persons with Down syndrome. JPediatr Orthop 1984;4:682-8 7 Alvaraz N, Rubin L. Atlantoaxial instability in adults with Down syndrome: a clinical and radiological survey. Appi Res Mental Retardation 1986;7:67-78 8 Burke SW, French HG, Roberts JM, Johnston CE II, Whitecloud TS m, Edmunds JO Jr. Chronic atlantoaxial instability in Down syndrome. J Bone Joint Surg 1985;67:1356-60 9 Pueschel SM, Scola FH. Atlantoaxial instability in individuals with Down syndrome: epidemiologic, radiographic and clinical studies. Pediatrics 1987;80:555-60 10 Collacott RA. Atlantoaxial instability in Down's syndrome. Br Med J 1987;294:988-9 11 Coria F, Quintana F, Villalba M, Rebollo M, Berciano J. Craniocervical abnormalities in Down's syndrome. Dev Med Child Neurol 1983,25:252-5 12 Onari K, Izawa T, Kuroki Y. Clinical and radiological evaluation of atlanto-axial instability in children with Down's syndrome. Orthop Traumatic Surg 1981;24: 619-24 13 Committee on Sports Medicine. Atlantoaxial instability in Down syndrome. Pediatrics 1984;74:152-4 14 Acheson ED. Atlanto-axial instability in people with Down's syndrome. London: Department of Health and Social Security, 1986:CMO(86)9 15 Davidson RG. Atlantoaxial instability in individuals with Down syndrome: a fiesh look at the evidence. Pediatrics 1988;81:857-65 16 Editorial. Atlantoaxial instability in Down syndrome. Lancet 1989;i:24 17 Miller JDR, Grace MDR, Lampard R. Computer tomography of the upper cervical spine in Down syndrome. J Comput Assist Tomogr 1986;10:589-92 18 Keats TE, Lusted LB. Atlas of roentgenographic measurement. 5th edn. Chicago. Yearbook Medical, 1985:113 19 Hincke VC, Hopkins CE. Measurement ofthe Atlantodental interval in the adult. AJR 1960;84:945 20 Jayson MIV, Grennan DM. Clinical features of rheumatoid arthritis. In: Weatherall DJ, Ledinlgham JGG, Warrell DA, eds. Oxford tetook of medicine. Oxford: Oxford University Press, 1987;16:6 21 Mawdsley C. The nervous system. In: Macleod J, ed.
Clinical exaination. Edinurgh: Churchill Livingsoe 1983:271,290,293 (Accepted 17 October 1989. Correspondence to Dr A Roy)
435
433
Atlantoaxial instability in Down syndrome - guidelines for screening and detection
M Roy MRCPsych Monyhull Hospital, Kings Norton, Birmingham B30 3QB M Baxter MBChB Chelmsley Hospital, Marston Green, Birmingham B37 7HL A Roy MRCPsych North Birmingham Community Mental Handicap Team, 66 Anchorage Road, Sutton Coldfield, West Midlands B74 2PH Keywords: atlantoaxial instability; Down syndrome; community survey; clinical radiological correlations; ethical implications
Summary A community survey was conducted in all adults with Down syndrome living in three health districts to see if there was any correlation between radiological and neurological abnormalities which could indicate the presence of atlantoaxial instability. There was no difference in the proportion of individuals with neurological abnormalities in the group with radiological abnormalities suggestive of atlantoaxial instability (6/14) compared with individuals with normal X-rays (50/123) as determined by the chi square test (0.01463: not significant). The clinical and ethical implications for screening of people with Down syndrome living in the community are discussed in view of these findings. Introduction Atlantoaxial instability in people with-Down syndrome has been extensively researched in the last two decades. Earlier workers reported an association between the two conditions14. In a survey of 158 children with Down syndrome, Pueschel5 noted the vast preponderance of asymptomatic cases. In a prospective study6 he emphasized the need for early diagnosis and treatment of symptomatic cases especially in the early stages. In a survey of mostly institutionalized adults, Alvarez7 found a reasonable degree of correlation between clinical and radiological abnormalities. He was able to ascribe a possible aetiological factor in neurologically abnormal but radiologically normal individuals. He then set out a clear statement of policy for screening in his hospital. There has been considerable interest in the nature of atlantoaxial instability itself. In a prospective study Burke et aL8 found progression of the lesion in children with Down syndrome over a period of time and warned against the increased risks of neurological deficits in boys over the age of 10 years. This feature of progression over time was to some extent refuted by Pueschel's study9 who said that the lesion progressed in some cases but regressed in some. Uncertainties about the nature of the lesion have made it difficult to arrive at a clear set of guidelines on screening and treatment10. A range of craniocervical abnormalities in Down syndrome have been described11 as have abnormalities specifically associated with atlantoaxial instability'2. As a result the Committee on Sports Medicine of the American Academy of Pediatrics published their guidelines13 which included among other suggestions radiological examinations on prospective participants with restrictions on sports 'that involve trauma to the head and neck'. British guidelinesl4 were broadly similar
and additionally cautioned about the need for support to the head of a person with Down syndrome while
travelling. In a recent review, Davidson15 felt that keeping in mind the rarity of cases of atlantoaxial instability that proceed to dislocation and cord compression, depriving large numbers of people with Down syndrome from various sporting activities based on radiological abnormalities may be too restrictive. He described a case of cord compression precipitated by strenuous activity in a person with previously normal cervical X-rays. He noted the high frequency of onset of neurological signs and symptoms prior to dislocation. A recent editorial in the Lancetl6 also highlighted the importance of neurological symptoms in people with Down syndrome. A separate factor to be taken into account was the changing nature of atlantoaxial instability itself as demonstrated in follow up studies89. Thus radiologically normal individuals who were then not X-rayed could develop a dangerous degree of instability unknown to those around them. Similarly individuals who had been restricted as a result of radiological abnormalities could become radiologically normal and be deprived of certain activities needlessly. In a CT scan studyl7 Miller demonstrated a correlation between radiologically diagnosed atlantoaxial instability and flattening of the cord on scanning. Since neurological examinations were not performed it was not possible to see the relationship between neurological symptoms and signs and cord compression. A community survey of adults with Down syndrome was undertaken to see if there was any correlation between neurological and radiological abnormalities and thus if more people with radiological evidence of atlantoaxisl instability had neurological abnormalities on clinical examination.
Subjects and methods A register for adults with Down syndrome was created in three health districts from all the existing day care and residential facilities in the mental handicap service. A letter was sent to their next of kin giving information about atlantoaxial instability and asking if they wanted their relative to have a clinical neurological examination and a radiological examination to detect the instability. The response rate was 92%. A proportion of people had already been X-rayed in response to parental concern and in preparation for Special Olympics. These people were examined neurologically with the examiner (MR) being unaware of their X-ray results. No objections were received
0141-0768/90/
070433-03/$02.00/0 © 1990 The Royal Society of Medicine
434
Journal of the Royal Society of Medicine Volume 83 July 1990
for the proposed examinations. The neurological examination consisted of the following: (a) tendon reflexes (biceps, triceps, knee, ankle) (b) plantar reflex (c) clonus (patellar, ankle) The tendon reflexes were graded: (i) not elicited; (ii) present; (iii) exaggerated The plantar reflexes were rated: (i) downgoing; (ii) upgoing Clonus was rated: (i) absent; (ii) present Clonus was rated present only if it was sustained throughout the period that stretch was applied at the patella and the ankle21. The age and sex of each subject were recorded. Patients with one or more exaggerated reflex and/or upgoing plantars and/or clonus were rated as having a neurological abnormality. Patients with an anterior atlanto odontoid distance of 3 mm or more on lateral cervical X-ray in flexion were diagnosed to have radiological evidence of atlantoaxial instability18-20. This was done by taking lateral cervical X-rays with the neck held in the neutral position and in flexion. The data was examined to determine the following: (i) the prevalence of radiological abnormalities (ii) the prevalence of neurological abnormalities in the two subgroups with and without radiological abnormalities. (iii) the sex distribution of the two subgroups (iv) the mean age ofthe individuals in the two subgroups Every individual with Down syndrome residing in the three health districts was included in the study. Six individuals were hospital inpatients at the time of the study. The chi-squared test was used to determine if the proportions of people with neurological abnormalities were different between the two radiologically distinct groups.
Results The number of people with Down syndrome who had radiological abnormalities suggestive of atlantoaxial instability was 14 out of a total population of 137 (10.2%). The distribution of neurological abnormalities is shown in Table 1. Nine men and five women (average age 30.5 years) had abnormal X-rays, of whom four men and two women had neurological abnormalities and five men and three women did not. Sixty-seven men and 56 women (average age 33.2 years) had normal X-rays. No individual had sustained clonus or upgrading plantar response. Discussion The prevalence of radiologically evident instability in this study was slightly lower than figures quoted Table 1. Association between radiological and neurological abnormalities
Abnormal Normal X-ray X-ray
Neurological abnormality No neurological abnormality Total
6 (43%) 8 (57%) 14
Chi-squared test 0.01463; not significant
50 (41%) 73 (59%) 123
in other surveys'0. There was a preponderance of men (9:5) over women. This was at variance with other surveys7 9 although the numbers were small. There was a predominance of men in the total sample. In the absence of true clonus and extensor plantar response, exaggerated reflexes were the sole indicator of neurological abnormality. Anxiety alone could not account for exaggerated reflexes and the ill-sustained clonus that were elicited. Since the extensor plantar responses is considered pathognomonic of an upper motor neurone lesion, the presence of other abnormalities could not be attributed to cord compression. It is possible that people with Down syndrome display exaggerated reflexes and ill-sustained clonus due to some other, as yet unknown, reason. This needs to be kept in mind while performing neurological examina-tions. There was no difference in the proportion of people with Down syndrome having exaggerated reflexes in the radiologically abnormal group compared with the normal group. This meant that abnormal neurological examination could not distinguish between the radiologically distinct groups and there was thus no correlation between neurological and radiological examiations. The -central difficulty facing clinicians and policy makers is that there is a need to be able to pick out those people with Down syndrome who are likely to develop atlantoaxial dislocation and cord compression. This has to be done without being alarmist and over restrictive. The most common assumption made is that a finding of anterior atlanto odontoid distance of greater than 3 mm means that the individual has the instability which can be the precursor to a full dislocation with its attendant complications. The unfortunate consequence of this assumption is that nearly one sixth of people with Down syndrome face restrictions on their activities16 the benefits from which could far outweigh the risks especially if it is accepted that 'there is no evidence that the current roentgenographic criteria of atlantoaxial instability are predictive of a tendency to dislocation'15. The changing nature of the lesion over time8'9 further reduces the predictive value ofthe X-ray abnormality. There appears to be a need to produce a set of sensitive and specific criteria which are flexible to take into account clinical, radiological and demographic correlates of an individual with Down syndrome. The first step in this process is to recognize the limitations of uniaxial parameters (such as abnormal X-rays) used currently and to perform a large clinico-radiological Down syndrome survey along with CT scans of the neck to confirm the findings of earlier studies'7 showing a correlation between radiologically suspeced instability and flattening of the cord. It would also clarify the association between abnormalities in the neurological examination (such as exaggerated reflexes) and varying degrees ofcord flattening, if any. If there was a correlation between the two then the presence of neurological symptoms such as pain in the neck, torticollis, gait abnormalities and neurological signs could help to establish the degree of cord flattening (or compression) that may exist. Alvarez7 reported that in his subgroup of radiologically normal but clinically abnormal cases he could find evidence of other pathology which could explain this anomaly. Abnormal neurological signs seem, therefore, to be rather nonspecific and could be present because of a variety of unrelated conditions. This needs to
Journal of the Royal Society of Medicine Volume 83 July 1990
be considered with Davidson's15 finding that in a significant number of cases atlantoaxial dislocation was preceded by the appearance of new neurological symptoms. It seems likely that symptoms are probably of much greater diagnostic significance than signs. It would then be possible to use a combination of symptoms, signs and X-ray findings to establish the risk of the individual developing atlantoaxial dislocation. All these clinical correlates would need to be matched up with demographic factors such as the age of the individual, his life style, likes and dislikes and the likelihood of him consenting to a series of examinations over a long period of time. This study was conducted in the community where the persons with Down syndrome exercised much more autonomy over their lives than their institutional counterparts and their active cooperation was essential at every stage. It was not sufficient to get agreement from the carers alone. In one adult training centre the trainees committee made it clear to the management that some of them did not feel that they had been adequately consulted on an individual basis prior to being subjected to the neurological and X-ray examinations. They also did not see the point of the exercise when they were, in fact, feeling well. This occurred in spite of obtining written agreement from carers prior to the study. This implies that any set of recommendations that relies on repeated examinations at frequent intervals is unlikely to be practical, acceptable and hence useful in community screening. It became obvious during this study that people with Down syndrome had to compete with other groups of patients for use of the radiology departments who were themselves unsure of the cost benefit ratio ofthis mass screening. In many instances it became clear the lifestyles of many of the participants in the study were so sedentary that the presence of any X-ray abnormalities in the absence of symptoms made no difference to their lives casting doubt on the need for total screening for Down syndrome populations. In conclusion, it may be worthwhile employing a stepwise approach to the process of detecting atlantoaxial instability and possible dislocation in people with Down syndrome living outside institutions: (i) Identification of the entire population from educational, family, residential and day care settings. (ii) obtaining agreement from the carers and if possible from the individuals themselves after as complete an explanation as possible. (iii) obtaining a neurological history from individuals and their carers to detect signs and symptoms of recently changing neurological status. (iv) examining the lifestyle ofthe individual to asse the need to undergo the testing procedure. (v) performing the clinical and radiological emmations. (vi) identifying the subgroup with both abnormalities as being at the highest risk of dislocation especially if displaying symptoms and signs of a recent onset.
Acknowledgments: The authors would like to thank the staff at various training centres, hostels and other establishments, relatives of participants, DaArkell, Lowe and Talbot (D)epartments of Radiology, Dudley, Telford and North Birmingham), Dr Oliver (Chelmsley Hospital), Dr Butler
(Ridge Hill), Dr Whiting (General Practitioner, Cornwall), Professor Corbett (Birmingham), Professor Simon (Birmingham), Profesor Roberts (Nottingham) and Mr Stanley (Bradford) and Miss J Rushton for encouragement and help with the project and with the manuscript.
References 1 Tishler JM, Martel W. Dislocation of the atlas in mongolism: preliminary report. Radiology 1965;84: 904 2 Dzenitis AJ. Spontaneous atlanto-axial dislocation in a mongoloid child with spinal cord compression. J Neurosurg 1966;25:458-60 3 Curtis BH, Blank S, Fisher RL. Atlantoaxial dislocation in Down's syndrome: report of two cases requiring surgical correction. JAMA 1968;205:464-5 4 Hungerford GD, Akkaraju V, Rawe SE, Young GF. Atlanto-occipital and atlanto-axial dislocation with spinal cord compression in Down's syndrome: a case report and review of the literature. Br J Radiol 1981;54:758-61 5 Pueschel SM, Scola FH, Perry CD, Pezzullo JC. Atlantoaxial instability in children with Down syndrome. Pediatr Radiol 1981;10:129-32 6 Peuschel SM, Herndon JH, Gelch MM, Senft KE, Scola FH, Goldberg MJ. Symptomatic atlantoaxial subluxation in persons with Down syndrome. JPediatr Orthop 1984;4:682-8 7 Alvaraz N, Rubin L. Atlantoaxial instability in adults with Down syndrome: a clinical and radiological survey. Appi Res Mental Retardation 1986;7:67-78 8 Burke SW, French HG, Roberts JM, Johnston CE II, Whitecloud TS m, Edmunds JO Jr. Chronic atlantoaxial instability in Down syndrome. J Bone Joint Surg 1985;67:1356-60 9 Pueschel SM, Scola FH. Atlantoaxial instability in individuals with Down syndrome: epidemiologic, radiographic and clinical studies. Pediatrics 1987;80:555-60 10 Collacott RA. Atlantoaxial instability in Down's syndrome. Br Med J 1987;294:988-9 11 Coria F, Quintana F, Villalba M, Rebollo M, Berciano J. Craniocervical abnormalities in Down's syndrome. Dev Med Child Neurol 1983,25:252-5 12 Onari K, Izawa T, Kuroki Y. Clinical and radiological evaluation of atlanto-axial instability in children with Down's syndrome. Orthop Traumatic Surg 1981;24: 619-24 13 Committee on Sports Medicine. Atlantoaxial instability in Down syndrome. Pediatrics 1984;74:152-4 14 Acheson ED. Atlanto-axial instability in people with Down's syndrome. London: Department of Health and Social Security, 1986:CMO(86)9 15 Davidson RG. Atlantoaxial instability in individuals with Down syndrome: a fiesh look at the evidence. Pediatrics 1988;81:857-65 16 Editorial. Atlantoaxial instability in Down syndrome. Lancet 1989;i:24 17 Miller JDR, Grace MDR, Lampard R. Computer tomography of the upper cervical spine in Down syndrome. J Comput Assist Tomogr 1986;10:589-92 18 Keats TE, Lusted LB. Atlas of roentgenographic measurement. 5th edn. Chicago. Yearbook Medical, 1985:113 19 Hincke VC, Hopkins CE. Measurement ofthe Atlantodental interval in the adult. AJR 1960;84:945 20 Jayson MIV, Grennan DM. Clinical features of rheumatoid arthritis. In: Weatherall DJ, Ledinlgham JGG, Warrell DA, eds. Oxford tetook of medicine. Oxford: Oxford University Press, 1987;16:6 21 Mawdsley C. The nervous system. In: Macleod J, ed.
Clinical exaination. Edinurgh: Churchill Livingsoe 1983:271,290,293 (Accepted 17 October 1989. Correspondence to Dr A Roy)
435
