1069 water,

placed

on

coated

stained with

grids, negatively

ammonium molybdate, and examined with

a

Philips EM

200 electron microscope. Stools were obtained from 6 neonates within the first week of life, 12 normal children, 29 adult controls, and 16 adults with chronic tropical sprue.

RESULTS

found in the stools from 6

neo-

However, pleomorphic fringed particles

were

Particles nates.

were not

ally related studies

are

these intestinal abnormalities. in progress. to

We thank Prof. C. K. Job and the leprosy mission for electron-microscope facilities; Prof. I. H. Holmes for helping us establish the techniques used in this study; and Dr June Almeida, Dr D. A. J. Tyrrell, F.R.S., and Dr D. Taylor-Robinson for their help and advice. Requests for reprints should be addressed to S. J. B., Wellcome Research Unit, Christian Medical College Hospital, Vellore 632004, Tamil Nadu, India. REFERENCES

found in stools from 19 of the 12 children, in stools from 27 of the 29 adult controls, and in stools from 14 of the 16 patients with tropical sprue. The particles varied in shape. Round (fig. a), oval or kidney shaped (see fig. b and c), and bizarre elongated forms (see fig. d) were seen. Particles of widely differing sizes and shapes were present in the same specimen. They ranged in size from about 100 nm. for the smallest rounded particles up to about 400 nm. for some of the larger ones. The elongated forms were about 100 nm. in width and up to 800 nm. long. the particles showed various morphological patterns. The most common pattern was made up of round or oval shaped knobs attached to the main body with a thin stalk (fig. a). In some Tshaped projections were attached peripherally to the knobs (see fig. b), while some others had a spike or a second small spherical structure similarly arranged (fig. c). The total width of the fringe in different particles ranged from 20 to 46 nm. The

fringes

on

Bishop, R. F., Davidson, G. P., Holmes, I. H., Ruck, B. J. Lancet, 1973, ii, 1281. 2. Bishop, R. F., Davidson, G. P., Holmes, I. H., Ruck, B. J. ibid. 1.

1974, ii, 149. 3. Flewett, T. H., Bryden, A. S., Davis, H. ibid. 1973, ii, 1497. 4. Middleton, P. J., Szymanski, M. T., Abbott, G. D., Bortolussi, R., Hamilton, J. R. ibid. 1974, i, 1241. 5. Tan, G. S., Townley, R. R. W., Davidson, G. P., Bishop, R. F., Holmes, I. H., Ruck, B. J. ibid. p. 1109. 6. Cruickshank, J. G., Axton, J. H. M., Webster, O. F. ibid.

p. 1353. Holmes, I. H., Mathan, M., Bhat, P., Albert, M. J., Swaminathan, S. P., Maiya, P. P., Pereira, S., Baker, S. J. ibid. 1974, ii, 658. 8. Wyatt, R. G., Kapikian, A. Z., Thornhill, T. S., Sereno, M. M., Kim, H. W., Chanock, R. M. J. infect. Dis. 1974, 130, 523. 9. Flewett, T. H., Bryden, A. S., Davies, H. J. clin. Path. 1974, 27,

7.

603. 10.

Baker, S. J., Mathan, V. I. Am. J. clin. Nutr. 1972, 25, 1047.

Hypothesis

In the same stool specimen particles with some or all of these types of fringe could often be seen. No special type of particle was associated with the patients with tropical sprue. Attempts at growing these particles in tissue-culture and human fetal intestinal organ culture have not so far been successful.

DISCUSSION

of these particles is not yet known. They morphological similarities to the and some to the coronaviruses. orthomyxoviruses the fringe morphology is not identical with However, electron micrographs of either of these groups of viruses. The possibility of their being mycoplasmas cannot at present be excluded, but on morphological The

exact nature

show

Further

PATHOGENESIS OF NONKETOTIC HYPEROSMOLAR DIABETIC COMA B. I. JOFFE L. H. KRUT

R. B. GOLDBERG H. C. SEFTEL

Carbohydrate-Lipid Research Unit, Department of Medicine, University of the Witwatersrand Medical School, Johannesburg, South Africa

some

grounds it seems improbable. These submicroscopic particles are common in the community and are acquired after birth. Since they were found in nearly all stool specimens they are presumably being excreted most of the time by all the subjects examined. They do not seem to be associated with any specific disease state, since they were found in healthy children and adults and in patients with chronic tropical sprue. Flewett et al., in their studies in the U.K., did not find particles like this either in patients with gastroenteritis or in controls. The absence of these particles in neonates and their widespread distribution in children and adults reflects the frequency of intestinal morphological abnormalities and malabsorption in the symptomfree population of southern India 1° This raises the question whether these particles are in any way caus-

Two concepts are advanced to explain some of the puzzling biochemical features found in nonketotic hyperosmolar diabetic It is firstly suggested that an insulinised liver coma. (reflecting residual beta-cell secretory activity) coexists with a diabetic periphery, thereby inactivating intrahepatic oxidation of incoming free fatty acids, which are directed largely along nonketogenic metabolic pathways such as triglyceride synthesis. This could account for the lack of hyperketonæmia. Secondly, it is hypothesised that within the liver enhanced neoglucogenesis occurs, due to the prevailing portal-vein ratio of glucagon to insulin, and is mainly responsible for the development of massive

Sum ary

hyperglycæmia. IN

the

nonketotic

hyperosmolar diabetic

syn-

drome 1-3 there has been much speculation about the absence of significant ketosis, since this is the cardinal biochemical feature differentiating it from ketoacidotic coma. An early hypothesis H linked the absence of hyperketonaemia with decreased plasma-

1070

free-fatty-acid (F.F.A.) concentrations consequent on suppressed lipolysis in adipose tissue. Suppression of lipolysis was thought to result either from a raised endogenous-plasma-insulin, or from decreased amounts of lipolytic hormones such as cortisol and growth hormone combined with an antilipolytic effect of hyperosmolarity itself. Although this explanation may apply in certain situations, a number of clinical observations argue against its acceptance as an all-embracing concept. Thus raised plasma-F.F.A. levels have been reported in the acute phase of hyperosmolar nonketotic diabetic stupor 1; plasma-insulin levels, in general, are similar to those found in the ketotic variety 6,9; and glucagon, a potential lipolytic hormone especially in hypoinsulinaemic diabetics has lately been shown to be strikingly increased." Furthermore, hyperosmolarity may be considerably increased in the ketotic state 12 without suppression of plasma-F.F.A. levels. EXPERIMENTAL DATA

Experimental reproduction of the nonketotic hyperosmolar diabetic syndrome in rats 13,14 has provided additional information about its pathogenesis. Increased hepatic-glycogen and adequate portal-vein-insulin levels in these animals, together with peripheral insulinopenia and massive hyperglycsemia, make the concept of an "insulinised" liver coexisting with a "diabetic" periphery attractive. Incoming F.F.A. of varying plasma concentrations would then be directed largely along nonketogenic pathways-particularly esterincation to triglyceride (and phospholipid) and secretion in very-low-density lipoprotein.15 Some support for this concept comes from the experimental observation of normal hepatic triglyceride concentrations and raised serum triglyceride levels.16 This is compatible with the notion of accelerated synthesis and secretion of very-low-density-lipoprotein triglyceride, possibly combined with defective peripheral triglyceride clearance by lipoprotein lipase, consequent on insulino-

penia.17 THE ROLE OF NEOGLUCOGENESIS

Foster 18 has focused attention

on the importance of a To insulin milieu. hepatic explain the extreme often he found, postulated that neoglucohyperglycaemia was enhanced in and that this (as ketoacidosis) genesis enhancement continued for a considerable time before clinical presentation. This implies that an insulinised liver may be antiketogenic yet may allow neoglucogenesis (thought to be suppressed by insulin 19). Moreover, ketogenesis and neoglucogenesis are usually directly related processes.20 The interpretation of this paradox may lie in some lately published work on the importance of ,the ratio of glucagon to insulin in determin.-ing the direction of intermediary metabolic pathways in the liver.21,22 Since peripheral plasma-glucagon levels are raised in nonketotic hyperosmolar diabetic coma," and since the peripheral glucagon level closely reflects its portal concentration,23 it is tempting to postulate that a raised portal-vein glucagon exerts a major stimulatory effect on hepatic neoglucogenesis which cannot be overcome by insulin, whereas insulin serves to control ketogenesis by modulating F.F.A. oxidation. We have found tentative evidence in the rat model of nonketotic hyperosmolar diabetes to support the thesis of enhanced neoglucogenesis; the plasma-level of the important neoglucogenic aminoacid alanine was subnormal, suggesting its enhanced hepatic uptake (unpublished observations).

critical

HYPOTHESIS

A hypothetical explanation for some of the puzzling biochemical features in the nonketotic hyperosmolar diabetic syndrome can be summarised as follows: 1. Absence

of hyperketonaemia

Critical portal-vein insulin, diverting incoming substrates: (a) Massively increased glucose-+glycogen. (b) Variably increased F.F.A.-triglyceride. 2. Massive

hyperglyccemia (a) Continuing neoglucogenesis due to prevailing portalvein glucagon/insulin ratio. (b) Contributory factors: increased intake, decreased peripheral utilisation, and reduced renal excretion of glucose.

3.

Frequent hypertriglyceridæmia (a) Accelerated hepatic synthesis and secretion of very-lowdensity-lipoprotein triglyceride. (b)Defective peripheral clearance.

The absence of hyperketoneemia depends largely on a small but critical amount of insulin reaching the liver from the pancreas. As a consequence, fatty-acid oxidation is not activated, the mechanism possibly involving inhibition of acylcarnitine transferase, which is necessary for the transport of F.F.A. into the mitochondria of liver cells. The marginally sufficient portal-vein-insulin concentration probably reflects a lesser degree of beta-cell deficiency than occurs in diabetic ketoacidosis; the nonketotic syndrome characteristically develops in mild maturity-onset diabetics or previously healthy subjects in whom insulin secretion is temporarily reduced 25 We suggest that substantial utilisation of insulin by the liver of a nonketotic hyperosmolar diabetic patient could then lead to peripheral insulinopenia of a degree found in ketoacidosis, where portal-vein insulin levels are also very low. (In rats with experimental nonketotic hyperosmolar diabetes the mean portal-vein-insulin concentration was 33 AU per ml. and the peripheral concentration was 18 //U per ml., whereas in ketotic animals the mean insulin concentration was 13 jU.U per ml. in both portal and peripheral plasma samples 1.) Several factors contribute to the massive hyperglycaemia of which the most important is continuing hepatic gluconeogenesis due to the prevailing portal glucagon/ insulin ratio. Hyperglucagonæmia conceivably arises in response to the same stimuli that reduce insulin secretion.22 In some instances hyperglycaemia may be aggravated by an increased glucose intake whether oral,26 intravenous,2’ or even intraperitoneal from dialysis 28; by reduced peripheral utilisation of glucose ; and at a late stage by decreased renal excretion of glucose3 due to severe dehydration and renal hypoperfusion. The hypertriglyceridæmia often found 7,8 indicates accelerated hepatic synthesis and secretion of very-low-density-lipoprotein triglyceride from incoming F.F.A., with possibly defective peripheral clearance as well. We thank Prof. D. W.

Foster, University of Texas SouthMedical School, and Prof. A. H. Rubenstein, University of Chicago, for helpful advice and criticism; the South African Medical Research Council, the Atomic Energy Board, and Witwatersrand University Council for financial assistance; and Mrs R. E. Joffe for help with the typing. western

1071

Reviews of Books Medical Oncology Medical Aspects of Malignant Disease. Edited by K. D. BAGSHAWE, F.R.C.P., Charing Cross Hospital, London. Oxford: Blackwell. 1975. Pp. 588. E13.50. SPECIALISTS of many types will be interested

to see

this

contribution to the rapidly developing specialty of oncology.

opening section few will fail to find the insight given biology stimulating. The chapter on immunoand malignant disease is outstanding for its balance and logy clarity, while the contributions on vascular structure, cell kinetics, and haematological, metabolic, and endocrine disturbances provide a breadth of well-referenced material. The chapter on genes and chromosomes in cancer fails to achieve the same clinical orientation, while the topic of In the

to tumour

steroid hormones, discriminant functions, and cancer is pursued to such a depth that it is not easy to maintain a perspective. The section on diagnosis offers much of interest to the clinician. The chapters on neurological and cutaneous manifestations of systemic cancer are not complemented by a general outline of the paraneoplastic syndromes. The diagnostic aids discussed (arteriography, lymphangiography, thermography, and radioisotope scanning) appear in isolation without an attempt to integrate them into a clinical setting with other important investigative techniques. The principles of therapy, including the mechanism of action of cytotoxic drugs and a well-referenced introduction to the clinical pharmacology of cytotoxic drugs, provide a solid third section to the book. The section on treatment of specific cancers is disappointing. Unfortunately, the standard in the concise account of trophoblastic tumours and in the balanced chapter on the

DR

JOFFE

AND OTHERS: REFERENCES

Halmos, P. B., Nelson, J. K., Lowry, R. C. Lancet, 1966, i, 675. Jackson, W. P. U. Acta diabet. lat. 1969, 6, 759. 3. Arieff, A. I., Carroll, H. J. Medicine, 1972, 51, 73. 4. Arieff, A. I., Carroll, H. J. Metabolism, 1971, 20, 529. 5. McCurdy, D. K. Med. Clins N. Am. 1970, 54, 683. 6. Gerich, J. E., Martin, M. M., Recant, L. Diabetes, 1971, 20, 228. 7. Bewsher, P. D., Petrie, J. C., Worth, H. G. J. Br. med. J. 1970, iii, 82. 8. Vinik, A., Seftel, H., Joffe, B. I. Lancet, 1970, ii, 797. 9. Watkins, P. J., Hill, D. M., Fitzgerald, M. G., Malins, J. M. Br. med. J. 1970, iv, 522. 10. Liljenquist, J. E., Bomboy, J. D., Lewis, S. B., Sinclair-Smith, B. C., Felts, P. W., Lacy, W. W., Crofford, O. B., Liddle, G. W. J. clin. Invest. 1974, 53, 190. 11. Lindsey, C. A., Faloona, G. R., Unger, R. H. J. Am. med. Ass. 1974, 229, 1771. 12. Fulop, M., Tannenbaum, H., Dreyer, N. Lancet, 1973, ii, 635. 13. Bavli, S., Gordon, E. E. Diabetes, 1971, 20, 92. 14. Joffe, B. I., Seftel, H. C., Goldberg, R., Van As, M., Krut, L., Bersohn, I. ibid. 1973, 22, 653. 15. Mayes, P. A., Felts, J. M. Nature, 1967, 215, 716. 16. Goldberg, R., Joffe, B. I., Seftel, H. C., Krut, L., Van As, M., Wolfe, E. Unpublished. 17. Bagdade, J. D., Porte, D., Bierman, E. L. Diabetes, 1968, 17, 127. 18. Foster, D. W. Adv. intern. Med. 1974, 19, 159. 19. Felig, P. Med. Clins N. Am. 1971, 55, 821. 20. Garber, A. J., Menzel, P. H., Boden, G., Owen, O. E. J. clin. Invest. 1974, 54, 981. 21. Parrilla, R., Goodman, M. N., Toews, C. J. Diabetes, 1974, 23, 725. 22. Unger, R. H., Orci, L. Lancet, 1975, i, 14. 23. Felig, P., Gusberg, R., Hendler, R., Gump, F. E., Kinney, J. M. Proc. Soc. exp. Biol. Med. 1974, 147, 88. 24. McGarry, J. D., Foster, D. W. J. clin. Invest. 1973, 52, 877. 25. Pyke, D. A. in Disorders of Carbohydrate Metabolism (edited by G. K. McGowan and G. Walters); p. 62. London, 1969. 26. Macaulay, M. B. Postgrad. med. J. 1971, 47, 191. 27. Rosenberg, S. A., Brief, D. K., Kinney, J. M., Herrera, M. G., Wilson, R. E., Moore, F. D. New Engl. J. Med. 1965, 272, 931. 28. Boyer, J., Gill, G. N., Epstein, F. H. Ann. intern. Med. 1967, 67, 568. 1. 2.

chemotherapy of other solid tumours is not maintained. The leukaemias are given less space than befits their importance in medical oncology, and the layout of the two succeeding chapters makes reference difficult. The contribution on the lymphomas seems strongly oriented towards American work, and ignores British contributions to chemotherapy. Space is given to unusual paediatric tumours, few of which of this size and nature. The unsatisfor non-trophoblastic gynaecological tumours is very evident from that chapter. These criticisms should not detract from the relevance of many parts of this book to clinical medicine, but in its present form it is unlikely to become a standard text.

are

relevant in

factory

state

a text

of

chemotherapy

Hypertension A Practitioner’s Guide to Therapy. JAMES C. HUTCHISON, M.D., Abington Memorial Hospital, Abington, Pennsylvania. Flushing, New York: Medical Examination Publishing. London: Lewis. 1975. Pp. 295.$10; £5.

IN the mind’s eye, the author appears a busy, enthusiastic, and highly experienced practitioner, widely read in hypertension and some of its related topics, and with files bursting with reprints, drug lists, work-up forms, and clinical abstracts. To distil from these a " practical guideline for the busy physician " is a demanding task; in the attempt, this book suffers from Dr Hutchison’s apparent desire to be all-inclusive at the expense of providing the unique insights of a community practitioner. The author works in a community hospital, and this could have given rise to very useful discussions of how to handle the emotional (as well as pharmacological) aspects of therapy, when to seek consultation, how to detect and deal with the patient who refuses to take his medications as prescribed, whether to turn over the routine care of hypertensives to the office nurse, and of the background to some of his highly personalised approaches to treatment. Instead, he has opted for a format which parallels other, more comprehensive, texts. Although its sections on drugs and drug combinations are good, the book attempts to explain too many things in too few words and is further marred by repetition (blank work-up forms keep reappearing), fragmentation (the reader must consult eight different headings to learn about thiazides), the inclusion of extraneous information (epidemiological data and a long report of a local screening project), and an inadequate index. The busy practitioner will continue to have to look elsewhere for practical advice on the management of hypertension.

Urinary Incontinence Edited by K. P. S. CALDWELL, Devon and Exeter London: Sector Publishing. 1975. Pp. 172. C6.

Hospital.

THE seven distinguished contributors to this elegant but rather expensive volume have covered in an authoritative manner almost all aspects of the diagnosis and management of incontinence. The incidence of this condition is fully reviewed, with emphasis on the enormity of the problem in geriatrics. It is surprising to find, in the editor’s preface, the old notion that incontinence in the elderly has more to do with decay of the caudal musculature than of the cranial neurons. Unfortunately, a decreasing proportion of incontinent patients will be suitable in the future for surgical cure, and the detailed account of palliative care given in this book is opportune. The scientific chapters reflect the uncertainty in the physiological realm which has resulted from the rapid accumulation of a mass of neuropharmacological and urodynamic data. Even the classic innervation of the external urethral sphincter is now in doubt. Greater emphasis on the revolutionary concept introduced by G. Enhörning (Acta