killing of bacteria due to hypochlorhydria, allows large numbers of the organisms to reach the intestine. What are the possible ill-effects of endogenous bacteria in the stomach? In post-gastrectomy patients with biliary reflux, intragastric faecal organisms may degrade bile acids to compounds which in laboratory animals act as co-carcinogens in chemically induced cancer. 12-14 Moreover, bacterial overgrowth results in raised nitrite levels,7,15 presumably from reduction of salivary and dietary nitrate by nitro-reductase active bacteria such as E. coli and micrococci. Bacteria, in the presence of nitrite, catalyse the nitrosation of amines in gastric juice to N-nitrosamines in vitro,and raised concentrations of these carcinogens have been found notably in pernicious anaemia and gastric carcinoma, after partial gastrectomy, and in patients receiving
with less efficient
know whether N-nitrosamines cause gastric cancer in man, but if concentrations are raised in certain conditions with an increased risk of gastric cancer, this is a hint which must not be ignored. On firmer ground is the association between overgrowth of stomach bacteria and postoperative sepsis.6,18 In 73 postoperative patients with duodenal ulcer, peptic ulcer, and gastric carcinoma GATEHOUSE et a1.6 reported that 2 -5%, 25%, and 44%, respectively, had wound infections due to organisms previously isolated from gastric aspirates. With gastric aspirate counts of 5 x 106 or more organisms per ml, wound sepsis developed in 93% of individuals, compared with 16% when the counts were less than this. Other workers have found large numbers of bacteria in the stomachs of patients being treated with cimetidine.19,20 In one study20 45% of aspirates, from 44 fasting patients taking cimetidine 1 g daily, contained more than 106 bacteria per ml 2-4 h after the last dose, compared with 4% of 51 comparable patients not receiving the drug. Sixteen different organisms were isolated from a total of 136 aspirates and 94% of the 192 organisms isolated were from the 49 aspirates with a pH of 4 or more. MUSCROFT et al .20 suggest two possible
F. Contraction of Salmonella gastroenteritis following previous operation the stomach. Acta Med Scand 1962; 171: 783-90. Giannella RA, Broitman SA, Zamcheck N. Salmonella enteritis: I. Role of gastric secretion in pathogenesis. Digest Dis 1971; 16: 1000-06. Hill MJ The role of colon anaerobes in the metabolism of bile acids and steroids, and its relation to colon cancer. Cancer 1975; 36: 2387-2400. Reddy BS Role of bile metabolites in colon carcinogenesis. Animal models. Cancer 1975; 36: 2401-06. Raicht RF, Cohen BI, Fazzini E, Sarwai A, Takahashi M. Effects of bile acids on induced colon cancer in rats. Gastroenterology 1978; 75: 981. Ruddell WSJ, Bone ES, Hill MJ, Blendis LM, Walters CL Gastric juice nitrite: a risk factor for cancer in the hypochlorhydric stomach? Lancet 1976; ii: 1037-39.
12. 13. 14 15.
PI, Smith PLR, Haines K, House FR, Walters CL. Gastric juice N-nitrosamines health and gastroduodenal disease. Lancet 1981; ii 550-52. Reed PI, Smith PLR, Haines K, House FR, Walters CL. Effect ofcimetidine on gastric juice N-nitrosamine concentration. Lancet 1981; ii: 553-56. Nichols RL, Miller B, Smith JW. Septic complications following gastric surgery: relationship to the endogenous gastric microflora. Surg Clin North Am 1975; 55:
strategies to counter the enhanced risk of postoperative wound infections-withdrawal of cimetidine 24 h before operation; or antibiotic cover for patients whose gastric contents have a pH of 4 or more preoperatively. Some say that the increasingly common practice of giving cimetidine preoperatively, to reduce the risk of acid-aspiration syndrome,2’ may likewise increase the risk of postoperative infection. Others doubt whether this practice gives the organisms time to accumulate in sufficient numbers. What is clear is that we need to know more about what cimetidine and other H2 blockers do to the gastric flora.
CHRONIC COUGH PRECISE diagnosis of chronic cough is not- always easy. When dealing with this common symptom we are sometimes forced to leave the aetiology undiscovered, or vaguely guessed at in terms such as persistent bronchitis. The vigour with which’ we pursue investigation of chronic cough (arbitrarily defined as troublesome cough lasting more than three weeks) is largely determined by our clinical assessment of the severity of symptoms and the likelihood of important underlying disease. Before deciding to investigate, it is therefore helpful to have some idea of the likely yield from various tests, and of the chances of finding a specific effective therapy as a result of full aetiological diagnosis. A paper originally presented at the 1979 American Thoracic Society Meeting by Irwin and his colleagues’ proved illuminating on these points, the main conclusion being that the cause of chronic cough can be established in virtually all patients, if one is prepared to go far enough. Moreover, since specific effective treatment is available for most causes of chronic cough the diagnostic exercise is well worth while. The causes of cough must be approached logically. In man, coughing can be most readily provoked by mechanical or chemical stimulation of the larynx, carina, trachea, and main bronchi, in that order.2 Direct evidence about cough receptors at these sites in man is difficult to obtain, but in the cat the receptors are fine non-myelinated nerve fibres lying beneath tight junctions in the airway epithelium.3These are the principal receptors for the vagally mediated cough reflex, but other receptor sites exist in the pleura, pericardium, stomach, and external auditory meatus. Receptors may be stimulated by foreign bodies, irritant gases, parasites, and neoplasms; they probably become "exposed" and more sensitive in inflammatory conditions of the respiratory tract, including laryngitis, acute and chronic bronchitis, and asthma. This heightened tendency to cough is often termed bronchial hyperirritability. The cause of cough in primarily parenchymal conditions such as pulmonary oedema or fibrosing alveolitis is less obvious, but probably depends upon deformation of airways. This is also the mechanism of cough in space-occupying mediastinal conditions-nodes, thymomas, aneurysms, and the like. Outside the thorax,
1367-72. 19. Ruddell WSJ, Axon ATR, Findlay JM, Bartholomew BA, Hill MJ. Effect of cimetidine on the gastric bacterial flora. Lancet 1980; i: 672-74 20 Muscroft TJ, Youngs D, Burdon DW, Keighley MRB. Cimetidine and the potential risk of postoperative sepsis. Br J Surg 1981; 68: 557-59.
21. Editorial. Cimetidine and the 1. Irwin
acid-aspiration syndrome. Lancet 1980;
Corrao WM, Pratter MR. Chronic persistent cough in the adult: the spectrum and frequency of causes and successful outcome of specific therapy. Am Rev Resp Dis 1981; 123: 413-17 2. Widdicombe JG. Mechanism of cough and its regulation. Europ J Resp Dis 1980; 61:
suppl 110. 11-20. 3. Widdicombe JG. Receptors 71-104.
the trachea and bronchi of the
J Physiol 1954;
forgotten, nor must (for example, by a auditory hair in the ear). Finally, firmly bottom of the list, comes cough without organic origin (habit cough), which should be diagnosed only after exclusion of other possibilities. The most common causes of persistent cough are probably
irritations of the external
chronic bronchitis and bronchial asthma. These should be diagnosed in most cases without difficulty from the history, backed by simple pulmonary function tests where asthma is suspected. A chest X-ray will help with several other common causes, including carcinoma. But what of the many cases in which none of these is helpful? This was the situation in most of the forty-nine patients described by Irwin and his colleagues, and the final list of diagnoses established by intensive investigation is instructive: chronic postnasal drip (29%), bronchial asthma (25%), asthma plus postnasal drip (18%), chronic bronchitis (12%), gastro-oesophageal reflux (10%), together with a few miscellaneous conditions, including sarcoidosis, metastatic carcinoma, and left ventricular failure. Postnasal drip is not an entirely satisfactory diagnosis, since it depends upon a subjective sensation for which there may be several causes. However, the message is clear that, in cases of chronic cough, history-taking should include careful inquiry about catarrh, sinusitis, rhinitis, the need to clear the throat frequently, and an actual sensation of postnasal drip. Physical examination should include the nasopharynx, with sinus X-rays where appropriate. Bronchial asthma can also be an elusive entity; in the American study a standard methacholine bronchial challenge4 was used in cases of doubt. Several patients with a positive methacholine response had absolutely normal baseline pulmonary function, an observation which may perhaps resurrect arguments about the definition of asthma. Should it be modified to include patients with no wheeze and normal pulmonary function, but with a chronic cough relieved by anti-asthmatic medication?5 Gastro-oesophageal reflux needs to be considered where the respiratory tract seems faultless; in the reported series, four patients had upper gastrointestinal endoscopy and all proved to have free reflux. Left ventricular failure with pulmonary congestion may seem a surprising cause of chronic cough, and in the study was diagnosed only by cardiac catheterisation; but what experienced clinician has not let such a case of disguised cardiac failure slip by? On the question of treatment, it is clear that most of these causes of chronic cough are amenable to specific treatment. This is particularly so in hidden asthma, where cough may be greatly improved by bronchodilators, plus corticosteroids when necessary. Indeed it might be argued that a trial of corticosteroids is reasonable in all cases where there is no other obvious cause for a chronic cough, and where diagnostic bronchial provocation tests for asthma are not feasible. Postnasal drip will often be improved by an intensive regimen of antihistamines and decongestants, with antibiotics in cases of sinusitis. Even the cough of chronic bronchitis may be expected to disappear within about six weeks of stopping smoking in about 70% of patients.6 Only those patients with persistent, useless cough, often due to malignancy, will require non-specific treatment with 4. Chai
H, Farr RS, Luz A, Froehlich LA, Matheson DA, McLean JA, Rosenthal RR, Sheffer AL, Spector SL, Townley RG. Standardisation of bronchial inhalation challenge procedures. J Allergy Clin Immunol 1975; 56: 323-27. 5. Corrao WM, Braman SS, Irwin RS. Chronic cough as the sole presenting manifestation of bronchial asthma. N Engl Med J 1979; 300: 633-37. 6 Irwin RS, Rosen MJ. Cough: a comprehensive review Arch Intern Med 1977; 137: 1186-91.
antitussive agents. After full investigation, Irwin and his colleagues were able to provide successful and sustained relief for 90% of their previously problematical patients-a fact which should encourage us to work hard to achieve a complete diagnosis in cases of chronic cough.
PUERPERAL PSYCHOSIS PUERPERAL
regarded by Hippocratesi
specific disorder, but this view is no longer held. Psychotic illnesses arising in the puerperium are generally believed to belong, in terms of psychopathology and prognosis, to the major psychoses. Thus both the International Classification of Diseases2 and the American Diagnostic and Statistical Manual of Mental Disorders3 recommend that illnesses in the puerperium be coded on the basis of the presenting symptoms only.
Nevertheless, the term, if not the concept, is tenacious and carries a special meaning for many clinicians. Psychoses in the puerperium follow a distinct life event and their onset is usually sudden and can therefore be dated accurately. On the basis of work done on life events in relation to depressive illnesses,4,5 and Brown’s concept of "brought-forward time",5 it has been suggested that childbirth is not just a life event but a specific stress and a true causal factor in the genesis of the psychosis. This notion is supported by the greater incidence of psychotic disorders after childbirth than at any other time of life.6,7 The increased risk after first pregnancies lends further weight to the crucial role of psychological factors.8,9 Whether the pregnancy ends in stillbirth or an illegitimate birth does not seem important, and one study pointing to an increased risk when there is marital disharmony10 has not been replicated. Some workers have found an increased risk after obstetric complications9,11 and have interpreted this as indicating a primary biological cause. Others have suggested that the increased incidence of psychotic illness after childbirth is a catching-up effect because pregnancy protects the mother for the nine months of gestation.6 The epidemiological data offer no support for this idea and the frequency of puerperal psychoses is much higher than expected even with allowance for the delaying effect of pregnancy. Some workers have suggested that the predisposition may be cyclical8’12 and that the manifestation of illness depends on a chance hormonal imbalance interacting with heightened susceptibility. This could explain why a psychotic illness follows one pregnancy and not another. In addition, seasonal variation in the month of conception12 1. Hamilton JA. Post-partum psychiatric problems. St Louis: Mosby, 1962. 2. World Health Organisation. Mental disorders - Glossary and guide to their classification in accordance with the 9th revision ofthe International Classification of Diseases. Geneva: W.H.O., 1978. 3. D S M III. Diagnostic & statistical manual of mental disorders, 3rd ed. American Psychiatric Association, 1978. 4. Paykel ES, et al. Life events and depression. Arch Gen Psychiatry 1969; 21: 753-60 5. Brown GW. Life events and psychiatric disorder. Psychol Med 1973; 3: 159-76. 6. Pugh TF, et al. Rates of mental disease relating to childbearing. N EngJ Med 1963; 268: 1224-28. 7. Kendell RE, et al. The influence of childbirth on psychiatric morbidity. Psychol Med 1976; 6: 297-302. 8. Protheroe C. Puerperal psychosis. A long term study 1927-61. Br J Psychiatry 1969, 115: 9-30. 9. Kendell RE, et al. The social and obstetric correlates of psychiatric admission in the puerperium. Psychol Med 1981; 11: 341-50. 10. Tod EDM. Puerperal depression - A prospective epidemiological study. Lancet 1964. ii: 1264-66. 11. Poffenbarger RS. Epidemiological aspects of post-partum mental illness. Br J Prev Soc Med 1964; 18: 189-95. 12. Grundy PF. Observations on the epidemiology of post-partum mental illness. Psychol Med 1975; 5: 286-90.