Ghucocorticoid receptors in depression - NCBI

0 downloads 0 Views 560KB Size Report
April, p %69) catalogues the non-haematological detrimental consequences ... Dudley Road Hospital,. Birmingham B18 .... WJONATHAN BOYCE. Northampton ...
1334

BRITISH MEDICAL JOURNAL

skeletal muscle, white blood cells, and brain in animal studies and children. He cites a recent study from California' in which iron deficiency was more common in infants with a history of recurrent infections and adds that it is uncertain whether the infections preceded or followed the iron deficiency. He fails to indicate, however, that a fall in plasma iron concentration is an integral part of the response to infection of any nature and persists during the septic period. This mechanism, leading to sequestration of iron in the liver and promoted by endogenous mediators of activated phagocytic cells, is important in host defence,2 probably by SIR,-We accept the points made by Mr J E A depriving micro-organisms- of a trace element Wickham in his letter of 26 April (p 1134) about essential for- growth, multiplication, and toxin the importance of clinical experience in the evalua- production.3 tion of new technology. However, clinical exThere is good evidence of adverse effects on perience is only one component of a rigorous clinical outcome of routine iron supplementation assessment; it is complementary to the special skill to iron deficient children with infection4 or malof evaluation. nutrition.5 A degree of caution should be exercised We are criticised for our lack of "hands on" before recommending "blind" provision or iron experience with lithotripsy. We were aware of this supplements to all children with iron deficiency. in designing the evaluation and therefore sought the advice of urologists in selecting outcome DAVID P TAGGART measures. Department of Peripheral Vascular Surgery, By contrast, we welcome the interest of Mr Royal Infirmary, Wickham in evaluation and the paper in which he Glasgow G31 2ER and his colleagues compare extracorporeal shock I Reeves JD, Yip R, Kiley VA, Dalhnan PR. Iron deficiency in wave lithotripsy with the alternatives (29 March, infants: the influence of mild antecedat infection. J Paediatr 1984;105:874-9. p 879). We would, however, like to point out a few 2 WR. Metabolic effects of infection. Progress in Food Bresel methodological weaknesses and errors in the NutrinoScience article. Over and above the inappropriate use of 3 Weinberg ED. Metal1984;8:43-75. starvation of pathogens by hosts. Bioscience historical controls the authors use different end1975;25:314-8. 4 Murray MJ, Murray AB, Murray MB, Murray CJ. l he adverse points to compare success of lithotripsy and the effect of iron repletion on the course of certain infections. Br alternative treatments. Simple statistical analysis MedJ 1978;ii: 113-5. of the results would have shown that the dif- 5 McFarlane H, Reddy S, Adcock KJ, Adeshina H, Cooke AR, Akene J. Immunity, transferrin and survival in kwashiorkor. ferences in the success rates -given could have BrMedJ 1970;iv:268-70. arisen purely by chance. In the costing there are a number of errors. Among the most important is the use of average inpatient cost in table IV as a measure of hotel AUTHOR'S REPLY,-I think it is difficult to draw expenses. In fact this figure includes both treat- conclusions about children, especially those in ment and hotel costs; treatment costs are therefore developed industrial societies, from a study of counted twice in the pricing of open and per- adult Somali nomads.' In Ibadan death was more cutaneous surgery. The use of average instead of common in children with kwashiorkor soon after marginal costs further exaggerates the bias in starting treatment if they had a low serum transfavour of lithotripsy, with its shorter length of stay ferrin value.' The causes of death in these children' were unknown and iron was only part of their compared with the alternatives. We hope that these points will demonstrate the treatment. It was assumed, without evidence, that' need for clinicians and those with skill in evalua- high'circulating concentrations of free iron had tion to collaborate in assessing new technology, contributed to the deaths. The evidence about iron and infection is equiso each can benefit from others' "hands on" vocal and many believe that iron deficiency may be experience. SABRI CHALLAH associated with increased susceptibility.3 Reeves NICHOLAS MAYS and Yip found that giving iron to 1 year old infants was not associated with- adverse effects.4 In my Department of Community Medicine, St Thomas's Hospital, opinion the weight of evidexice is greatly in favour London SEI 7EH of treating iron deficiency in'children in almost all circumstances. D P ADDY SIR,-Mr C R Charig (3-May, p 1199) misunder- Department of Paediatrics, stands my concern about "shillings and pence." I Dudley Road Hospital, entirely agree with him that the decision must be Birmingham B18 7QH taken on "benefits for the patient": in my opinion this requires a controlled trial. I regret that Mr 1 Murray MJ, Murray AB, Murray MB, Murray CJ. The adverse effects of iron repletion on the course of certain infections. Br Charig has been misinformed: Bloomsbury Health MedJ 1978;ii:1113-S. Authority assures me that the unit cost statistic for 2 McFarlane H, Reddy S, Adcock KJ, Adeshina H, Cooke AR, 1984-5 is £181-8 and that the statistic includes the Akene J. Immunity, transferrin and survival in kwashiorkor: BrMedJ 1970;iv:268-70. cost of surgical procedures. 2 Spodick DH. The randomized controlled clinical trial: Scientific and ethical bases. AmJMed 1982;73:420-5. 3 Spodick DH. Revascularization of the heart-numerators in search of denominators. Am HearzJ 1971;81:149-57. 4 DeWood MA, Spores J, Notske RN, et al. Medical and surgical management of myocardial infarction. Am3' Cardiol 1979;44: 1356-64. 5 DeWood MA, Heit J, Spores J, et al. Anterior transmural myocardial infarction: effects of surgical coronary reperfusion on global and regional left ventricular function. Journal of the American College ofCardiology. 1983;1: 1223-34. 6 DeWood MA. Reply to Spodick DH. Joumal of the American College of Cardiology. 1983;2:1240-1. 7 Spodick DH. Randomize the first patient: scientific, ethical and behavioral bases. AmJf Cardiol 1983;51:916-7.

JJJ JONES Department of Community Medicine, Leicester LEI 6TP

3 Oski FA. The non haematological effects ofiron deficiency. AmJ

Dis Child 1979;133:315-22. JD, Yip R. Lack of adverse side-effects of iron ferrous sulphate therapy in I year old infants. Pediatrics 1985;75: 352-5.

4 Reeves

Happiness is: iron

SIR,-The leading article by Dr A P Addy (12 April, p

%69) catalogues the non-haematological

Ghucocorticoid receptors in depressionSIR,-Dr L J Whalley anld colleagues are to be congratulatedl on their excellent article (2-9 March,

detrimental consequences of iron deficiency, with or without anaemia, on the metabolic function of p 859). Apart from the main finding of a reduced

VOLUME 292

17 m,Ay 1986

number ofglucocorticoid receptors in the lymphocytes of patients with major depression (research diagnostic criteria "endogenous" and "psychotic" subgroups) in comparison with a group of chronic schizophrenics and normal controls, several points merit further discussion. Firstly, the methods describe blood as being taken between 1130 and 1230 (presumably the middle of the day) for assessment of cortisol values. There were no significant differences in cortisol values between the depressed group ofpatients and the other groups. It would have been informative to know whether or not the depressed patients actually had hypercortisolaemia. This would have been more likely to have been found in blood specimens taken later in the day, between 1300 and 1600. There is a high correlation between mean 24 hour plasma cortisol values and those in the early afternoon taken over this three hour subperiod. Shortening this subperiod to two hours causes a sharp decrease in the association. ' Secondly, the object of having a group of chronic schizophrenic patients for comparison was to deal with unknown variables related to hospital admission. This raises the question when admission occurred, and when the investigations occurred in relation to it. It is a new admission and its acute aspects (and presumably the anxiety produced) that produce changes in cortisol profiles.2 Thirdly, Dr Whalley and colleagues quite rightly point out in their discussion that the similar finding of reduced numbers of glucocorticoid receptors in the lymphocytes of patients with anorexia nervosa may be due to a common factor of reduced food intake and weight loss. However, the research diagnostic criteria, although stringent, do not require appetite and weight loss as essentials for diagnosing endogenous or psychotic depression. Some of the patients in the depressed group might therefore have had unchanged food intake and weight, or indeed increased food intake and weight (the authors do not say). If this were the case it would be of great interest to know whether there was a significant difference in glucocorticoid receptors between subgroups showing marked increases and marked decreases in food intake and weight. This difference might not be picked up on the ordinary correlation matrix which they have used.

BRIAN HARRIS Department of Psychological Medicine, University of Wales College of Medicine, Cardiff CF4 4XN

ROGER THOMAS Sully Hospital, South Glamorgan I Halbreich U, Asnis GM, Shindledecker MA, Zumoff B, Nathan

S. Cortisol secretion in endogenous depression. Arch Gen Psyckiaty 1985;42:904-8. 2 Berger M, Pirke KM, Doerr P, Krieg JC, Von Zerssen P. The

limnited utility of the dexamethasone suppression test for the diagnostic process in psychiatry. Br J Psychiatr 1984;145: 372-82.

AUTHoR's REPLY-Drs Harris and Thomas are correct to emphasise the importance ofserial blood sampling in the study of cortisol secretion. Previously we examined plasma cortisol concentrations over 17 hours in newly admitted psychiatric patients and, like many others, detected the greatest differences between patients and controls during the afternoon period.' The hypercortisolaemia of depressive illness was best distinguished, however, during the late evening (2300-2400). We related this to the hypothesis that the hypercortisolaemia of depression is linked to a phase advancement of the early morning increase in cortisol secretory activity. In our subsequent study of glucocorticoid receptor numbers in depressed patients we reasoned that the most informative time of sampling would be during the late evening.

BRITISH MEDICAL JOURNAL

VOLUME 292

In the light of our previous results, we could not justify protracted, serial night time blood sampling in severely ill psychiatric patients simply to establish the already well known fact that hypercortisolaemia commonly occurs in depressive illness. We decided therefore to examine a single late morning sample in a larger number of depressed patients, with the advantages of increased patient compliance and more efficient use of assay resources. In our previous study plasma cortisol concentrations were closely correlated throughout the day, and the use of the late morning sample would not substantially reduce the likelihood of detecting a relation between hypercortisolaemia and glucocorticoid receptor number. The question of temporal stability of our findings was examined in an alternative study of two volunteers over 24 hour periods. Since we had previously shown that hypercortisolaemia did not depend on length of stay in hospital we did not include a recently admitted, non-depressed comparison group.' Instead, we studied the effects of medication on glucocorticoid receptor number in a group of chronic schizophrenic patients treated with neuroleptic drugs. The effects of weight change and duration of hypercortisolaemia on glucocorticoid receptor number in depressed patients will be examined in our present and future longitudinal studies on the temporal stability of the decrease in glucocorticoid receptor number. L J WHALLEY MRC Brain Metabolism Unit, Royal Edinburgh Hospital, Edinburgh EHJO 5HF I Christie JE, Whailey Ll, Dick H, Blackwood DHR, Blackburn IM, Fink G. Raised plasma cortisol concentrations a feature of drug-free psychotics and not specific for depresaion. Br J

Psychiatry 1986;148:58-5.

Impotence: science and sciencibility

1335

17 MAY 1986

tolamine. Some of those with psychogenic causes certainly did. At a time when newer modalities for treating erectile impotence are emerging, including penile revascularisation and intracavernosal administration of vasoactive and a blocking agents, it is vital that centres dealing with this problem standardise both the terminology and investigative methods so that results may be compared. This will necessarily involve the use of sophisticated tests, but unless this happens confusion will continue to prevail.

K DE&u J C GINGELL Department of Urology, Southmead General Hospital, Bristol BSIO SNB 1 Lue TF, Hricak H, Harick KW, Tanagho EA. Vulogenic impotence evaluated by high resolution uhrasonography and pulsed Doppler spectrum analysis. Radioy 1985;I55:

777-81.

Role of dru in fractures of the femoral neck SIR,-Mr Saifudin Rashiq's and Dr Richard F A Logan's unexpected finding (29 March, p 861) that patients with fractures of the femoral neck are less likely to be being prescribed drugs than are controls in the community is strongly at variance with our own unpublished data. We recorded the drugs being prescribed to 139 patients immediately before the fracture and compared these with the drugs prescribed for 139 age and sex matched controls from the same general practice lists (table). The 139 patients were selected from a consecutive series of 333 on their ability to pass a standard mental state questionnaire, the purpose being to ensure accurate histories. The same questionnaire was applied to the controls. Compared with the 139 patients who passed the questionnaire, the remainder were older, less mobile, and less independent with activities of daily living. We would expect these patients to have had even higher levels of medication. We obtained information on prescribed drugs by interviewing the patients and when possible checking against the hospital notes (95%), and by interviewing the controls and checking against the drug containers (97%) (the controls were interviewed in their own homes). Although both these methods of checking drug use may give rise to under-recording of the usually prescribed medication, this is much more likely to be true of the information obtained from patients. In other words, the effect of recording errors would have been to mask any real excess of prescribing to patients with hip fractures. The explanation of these seemingly irreconcilable results may lie in the methods used to select cases from among patients with hip fractures. We have described our use of a mental state questionnaire. Mr Rashiq and Dr Logan selected from a series of 227 patients who were operated on,

excluding 85 because they lived more than 10 km from the hospital and a further 40 because the general practitioner was not recorded in the hospital notes or refused permission. We cannot comment on the possible bias introduced by the geographical exclusion, but patients with no recorded GP would very likely be from institutions (most hip fracture series record about 20% as coming from institutions but Mr Rashiq and Dr Logan make no mention of this). They would thus be older, more dependent, and more likely to be taking medication than the 102 patients who made up their final series. When these are taken in conjunction with the patients who were excluded by virtue of dying early or being treated conservatively, it seems possible that Mr Rashiq and Dr Logan have inadvertently excluded about 60 patients who would be likely to be being prescribed higher than average levels ofmedication. This may well be sufficient to account for their strange results. The real difficulty with their result is how to explain it. We cannot accept their hypothesis that "before their injury most patients with fracture of the femoral neck either are fitter or regard themselves as fitter than their peers." Such a suggestion is contrary to all evidence. The large case-control studies of Baker' and Brocklehurst2 (J C Brocklehurst et al, unpublished report), and our own data, show that such patients are less mobile, less independent, lighter, more demented, more likely to have chronic illnesses, and more likely to be under medical care than their peers. They are decidedly less fit, not more. We are forced back to the conclusion that these results cannot be sensibly explained and must be the consequence of the biases we have outlined. W JONATHAN BOYCE Northampton Health Authority, Northampton NNI 5DN

MARTIN P VESSEY Department ofCommunity Medicine and General Pratice,

SIR,-Dr W J Jeffcoate's leading article on erectile Oxford University, impotence (22 March, p 783) gave a timely and well Oxford balanced review of a subject that is in need of I Baker MR. The epidemioo andaeoiogyorjoffemwalneckfracnwe. objective evaluation. The subsequent letter by MD Thesis. University of Newcastle upon Tyne, 1980. Mr E A Kiely and others (26 April, p 1137) 2 Brocklehurst JC, Exton-Smith AN, Barber SML, Hunt LP, MK. Fracture of the femur in old age: a two centre Palmer denouncing "high tech" investigation of this probstudyof associated clinical factors and the cause of the fall. Age lem warrants clarification. Aing 1978;7:7-15. They state that all 25 of their organically impotent men achieved erections with an intraSIR,-Mr Saifudin Rashiq and Dr Richard F A cavernosal injection of papaverine and phentolaLogan (29 March, p 861) address a subject which mine. In our experience, and also that of others,' we have investigated similarly, but we also used men with vasculogenic impotence respond only interviews and hospital notes as well as general partially, with varying degrees of tumescence but practitioners' records. The authors found a fracrarely a full erection. Can one assume from their tional relative risk for patients with hip fractures results that the organic group was composed over controls for all drugs except steroids and mainly of men with neurogenic impotence? If this antibiotics and concluded that most of those with is true then it reflects a surprisingly high incidence hip fractures were either fitter or regarded themof neuropathy in their impotent patients. It is selves as fitter than their peers. remarkable that over half of their patients diagThis inference seems -extraordinary and stems nosed as having organic impotence subsequently from looking at drug consumption both in isolation reported spontaneous erections in the ensuing month. One clearly needs to know the diagnostic criteria that were used to define organicity in their Companson of numbers ofdrugs taken by patients withfractures of the femoral neck in nto studies patients. In our cohort of 35 impotent men, who were Boyce and Vessey Rashiq and Logan 1986 comprehensively assessed by nocturnal penile Controls Cases Controls Cases No of drugs tumescence monitoring, intracavernosal papaverine injection complemented by computerised 54 34 78 61 penile Doppler arterial wave form analysis, and 0 34 27 28 10 neurophysiological tests including sacral reflex 21 16 27 37 14 15 20 33 9 latency measurements, an organic cause was 3 18 12 23 4 4 considered to be present in 16. Arterial insuffi8 13 5 4 5 ciency was contributory in 12 of these, neuropathy accounting for only four. None of our men with organic causes reported return of spontaneous erections after a test dose of intracavernosal papaverine either alone or in combination with phen-

102

204

xz= 15-89, 5 df, p