with hypothyroidism who progressed to dose of 300 jig/day during trial.Benchimol and Dimond et al in their study of the apex cardiogram in ischaemic heart disease made no reference to abnormalities of the IRT.'1 Our results show that the IRT is prolonged in hypothyroid patients and reverts to normal after thyroxine replacement treatment. Since there was no radiographic or clinical evidence of pericardial effusion in any of our patients, we may have been measuring the effect of myxoedematous infiltration of the muscle fibres (a finding first reported in 188012), and possibly this infiltration no longer causes detectable change in the IRT when the patient is rendered euthyroid.It is particularly interesting that there was no reduction in the IRT when the dose of thyroxine was increased above 150 ug/day, as this finding correlates well with work showing that most hypothyroid patients became euthyroid, using the serum thyrotrophin level as a criterion, on 100-200~ig/day of thyroxine and may be hyperthyroid when on the accepted full replacement regimen of 300 ,tg!day of thyroxine."There is at present no readily available clinical or biochemical test that indicates the gradual return to normal body metabolism with increasing doses of thyroxine and, in particular, none that show normal end-organ function on less than the full replacement dose of 300 jig/day. The IRT as an index of end-organ function in hypothyroidism therefore seems to have great potential as both a research and a diagnostic tool in thyroid disease.We are grateful to the department of cardiology for their technical help and for use of their equipment.Requests Relation between blood pressure, weight, and plasma sugar and serum insulin levels in schoolchildren aged 9-I2 years in Westland, Holland C DU V FLOREY, S UPPAL, C LOWY British MedicalJournal, 1976, 1, 1368-1371 Summary In 2388 schoolchildren aged 9-12 years who took part in a study of cardiovascular risk factors in Westland, Holland, plasma sugar concentrations were found to be positively correlated with systolic and diastolic blood pressure, independently of weight. Serum cholesterol levels were also related to systolic blood pressure in boys, but much less strongly than plasma sugar levels. The relation between serum insulin and blood pressure, independent of plasma sugar, was weak. sugar and systolic pressure existed for both sexes and regardless of whether measurements were made in the morning or afternoon; its association with diastolic pressure was weaker, and was not so consistent over all groups. These findings suggest that the relations between risk factors for coronary heart disease that exist in adults are already evident in childhood. IntroductionThe epidemiology of coronary heart disease (CHD) has been studied extensively in adults. People with raised serum cholesterol levels, blood pressure, and probably body weight are more likely than those with normal levels to have evidence of myocardial ischaemia at the time of the measurement or to suffer overt clinical CHD in the future. Rais...