455

anterior lobe of the pituitary. I suggested that further information on this role might become available if anterior-lobe hormones in serum were assayed and the pituitary gland was examined at necropsy in such cases.4 The hypothesis that thyroid-stimulating hormone (TSH) was the sole stimulus reponsible for exophthalmic ophthalmoplegia, pretibial myxoedema, and thyroid acropachy has not been confirmed; neither has the suggestion that there might be a separate exophthalmos-producing substance. Weetman and Fells have brought us right up to date in their assessment of thyroid-associated eye disease. Most, if not all, patients with ophthalmopathy have autoimmune disease. Raised TSH levels support the clinical diagnosis. Patients with dermopathy (pretibial myxoedema) similarly have a subgroup of TSH-receptor antibodies that are unusually potent stimulators of thyroid-cell growth and protein synthesis.5 Am I correct in the assumption that the cause of thyroid acropachy remains as obscure as it did over 30 years ago? Meadow Rise, 3 Lakeside,

Swindon, Wiltshire SN3 1QE, UK

ANTHONY G. FREEMAN

1. Freeman AG. Gross digital clubbing and exophthalmic ophthalmoplegia in thyroid disorders. Lancet 1958; ii: 57-70. 2. Gimlette TMD. Thyroid acropachy. Lancet 1960; i: 22-24. 3. Gimlette TMD. Pretibial myxoedema. Br Med J 1960; ii: 346-51. 4. Freeman AG. The cause of digital clubbing. Lancet 1959; i: 888-89. 5. Tao T-W, Leu S-L, Kriss JP. Biological activity of autoantibodies associated with Graves’ dermopathy. JClin Endocrinol Metab 1989; 69: 90-99.

Cholera in Peru SiR,—The epidemic of cholera in Peru has prompted renewed interest in preventive control measures. The Peruvian Ministry of Health has recommended the exclusive use of boiled water for drinking.’ The boiling of water for 10 min to ensure proper inactivation is advocated. Kerosene, one of the most popular fuels, is used very sparingly because of its cost. This investigation was requested by the US Centers for Disease Control, to establish the minimum boiling time needed to ensure inactivation and thus conservation of fuel. Three strains of Vibrio cholerae 01 Inaba, El Tor, were investigated. These included a Peruvian epidemic strain (CDCC6706), anon-haemolytic US gulf coast strain (CDC-216478),and an American Type Culture Collection strain (ATCC 14033). A recent Ohio river isolate of Escherichia coli was also included in the

boiling experiments. Phosphate-buffer3

washed cells were suspended in 300 ml volumes of autoclaved river water, lake water, and dechlorinated tap water, resulting in turbidities of 28, 27, and 15 nephelometric turbidity units, respectively. Initial cell titres were between 10 and 107 colony-forming units (CFU)jml. Cell suspensions were brought to a rolling boil (99’5OC) over an open flame for 30 s at an altitude of 0.23 km. Cell suspensions were cooled and the full volume was assayed by the membrane filtration procedure with TCBS agar for Vcholerae and m-Endo LES agar for E coli. Results from triplicate experiments showed that boiling for 30 s at 99-5°C inactivated all strains of V cholerae and E coli. The thermal death point (TDP) (temperature required to totally inactivate a specific organism in 10 min) was also measured for the three toxigenic V cholerae strains suspended in autoclaved dechlorinated tap water. TDP measurements were done in triplicate at each temperature in a constant temperature oscillating water bath.’ V cholerae was assayed by membrane filtration with TCBS agar. TDPs for the V cholerae strains were between 60 and 62°C:

Thus, heating with

a

full boil, even in very turbid conditions safety factor of 1 min, will be sufficient to

water to a

conservative

vibrios and indicator organisms. Similar has been recommended for other microorganisms.5 Water boils at lower temperatures at higher altitudes (about 90°C at 3 km), but on the basis of a TDP of V cholerae between 60-62°C increasing the boiling time to 3 min would more than adequately compensate for the decreased atmospheric pressure conditions at higher altitudes. It should be noted that these suggested intervals (11 or 3 rnin) refer to the total time the water is held at a rolling boil and should not be confused with the first sign of bubbles being liberated in the heating process. inactivate

pathogenic

treatment

Microbiological Treatment Branch, Drinking Water Research Division, Risk Reduction Engineering Laboratory, US Environmental Protection Agency, Cincinnati, Ohio 45268, USA

EUGENE W. RICE CLIFFORD H. JOHNSON

1. Centers for Disease Control. Cholera: Peru, 1991. MMWR 1991; 40: 108-10. 2. Wachsmuth IK, Bopp CA, Fields PI, Carrillo C. Difference between toxigenic Vibrio cholerae O1 from South America and US gulf coast. Lancet 1991; 337: 1097-98. 3. American Public Health Association. Standard methods for the examination of water and wastewater, 17th ed. 1989. 4. Smith JL, Marmer BS, Benedict RC. Influence of growth temperature on injury and death of Listeria monocytogenes Scott A during a mild heat treatment. J Food Protection 1991; 54: 166-69. 5. Geldreich EE. Drinking microbiology: new directions toward water quality enhancement. Int J Food Microbiol 1989; 9: 295-312.

Childhood

cancer

in Kerala, India

SiR,—Cancer is an important cause of death in childhood, and in India paediatric cancers are receiving more attention with the introduction of control measures against infections and malnutrition. Kerala, on the south-western coast of India, with a population of 30 million, has the lowest infant and child mortality rates and the highest literacy rate and life expectancy in India.! We describe here the paediatric cancer pattern as revealed by the records of the hospital cancer registry, part of the National Cancer Registry Programme of India, at the Regional Cancer Centre, Trivandrum. Cases under age 15 years registered in 1983-89 were reviewed. In the absence of population cancer registration in Kerala, hospital statistics, with their selection bias, provide the only opportunity to study cancer patterns. Paediatric cancers constituted 4-3% (males) and 4-0% (females) of all cancers registered. In Britain, this figure would be about 05% but in India children constitute a higher proportion of the population. The sex ratio in Kerala was 1 27 to 1. 43% of male and 46% of female cases were under 5 years of age. The most common childhood malignant disease was leukaemia, accounting for almost one-third of cases in boys and one-quarter in girls. Two-thirds of the leukaemias were acute lymphatic (ALL). Malignant neoplasms of the brain and nervous system (CNS) were the next most common malignant disease, the leading types being astrocytomas and medulloblastomas. Most tumours of the sympathetic nervous system were neuroblastomas. Of the lymphomas (the third leading form of malignant disease) half were non-Hodgkin: other leading causes are summarised in the table. FREQUENCIES OF PAEDIATRIC MALIGNANT DISEASES

456

Virtually all kidney tumours of children were Wilms’ and there was striking preponderance in girls. 70% of eye tumours were retinoblastoma. The relative frequency ofretinoblastoma was much higher than it is in western countries. Data from the population-based registries at Bangalore, Bombay, and Madras indicate paediatric cancer frequencies varying from 37% to 4% 23 In Dibrugarh in north-eastern India and Chandigarh in the north-west the frequency was 2-48%. International comparisons of cancer incidence are potentially fraught by variability in diagnosis, classification, and coding practices, by competing causes of death, by differential access to medical care, and by incomplete registration, so patterns need cautious interpretation. Leukaemia is the leading type of cancer and malignant tumours of the CNS are the second most common cancer in childhood in several countries.2,-u, Lymphomas are the commonest cancer (59%) among Nigerian children, with Burkitt’s lymphoma accounting for 87% of all lymphatic malignancies.’ A a

low incidence of Wilms’ tumour has been reported from China.8 Variations in the population distribution of these cancers suggest differences in aetiology. Parental recall of exposure histories has a major role in retrospective studies of risk factors in children, and several factors (genetic factors, birth characteristics, environmental, infectious) have been identified. The establishment of committed paediatric cancer registries in India would contribute usefully to clinical and epidemiological research. P. KUSUMAKUMARY R. SANKARANARAYANAN

Regional Cancer Centre, Trivandrum-695 011, India

G. PADMAKUMARY CHERIAN VARGHESE S. RAJEEV KUMAR M. KRISHNAN NAIR

1. Health

profile of Kerala. Ministry of Health and Family Welfare, Government of Kerala, Trivandrum, 1989. 2. Parkin DM, Stiller CA, Draper GJ, Bieber CA, Terracini B, Young JL, eds. International incidence of childhood cancer. IARC Sci Publ 1987, no 87. 3. Annual reports of the National Cancer Registry Programme of India. New Delhi: Indian Council of Medical Research, 1982-85. 4. Muir C, Waterhouse J, Mack T, Powell J, Whelan S, ed. Cancer incidence of five continents: vol V. IARC Sci Publ 1987, no 88. 5. Gloeckler Ries LA, Hankey BF, Edwards BK. Cancer statistics review 1973-1987. Bethesda: National Cancer Institute, 1988. 6. Parkin DM, Stiller CA, Draper GJ, Bieber CA. The international incidence of childhood cancer. Int J Cancer 1988; 42: 511-20. 7. Williams CKO, Folami AO, Laditan AAO, et al. Childhood acute leukaemia in a hospital population. Br J Cancer 1982; 46: 89-94. 8. Tu J, Li FP. Incidence of childhood tumours in Shangai, 1973-1977. JNCI 1983; 70: 589-92.

LeVeen shunt with

wandering tip

SiR,—The LeVeen peritoneovenous shunt has been used for nearly 20 years in the treatment of intractable ascites due to chronic liver disease. However, the technique has a high failure rate due to shunt occlusion.1 An uncontrolled study has suggested that the incidence of shunt occlusion due to thrombosis may be reduced if a titanium tip is used (fig 1).2 This tip is not part of the standard catheter assembly but is attached to the venous end at the time of insertion using silicone adhesive. We report here a complication resulting from our first experience of this modification. The patient was a 52-year-old man with alcoholic cirrhosis diagnosed 1 year previously. Despite total abstinence he had tense ascites, unresponsive to fluid restriction and diuretics. Twiceweekly paracentesis meant that the patient was unable to resume normal activities and he was reluctant to leave hospital. A LeVeen shunt was inserted, draining to the left internal jugular vein. Function seemed satisfactory at first but ascites reaccumulated within a few days. Contrast studies confirmed shunt patency but scintigraphy, 3 weeks after insertion of the shunt, indicated very slow clearance of radiolabelled albumin from the abdomen. A week later the shunt was removed without difficulty and a new shunt with a titanium tip was positioned in the right internal jugular vein. Despite reverse Valsalva exercises and a surgical corset, the ascites once again reaccumulated. Liver transplantation was felt to be the sole remaining option, and the second shunt was removed after 1 month. The rectus incision

Fig 1- titanium tip to Length 3 cm.

LeVeen shunt.

Fig 2-X-ray views of detached tip. was reopened and the abdominal portion of the shunt was retrieved. The venous limb was delivered by gentle traction, although more resistance was evident than with the previous untipped shunt. The titanium tip was missing. Palpation of the right supraclavicular fossa revealed a firm mass. After X-ray screening (fig 2) the site was explored and the tip was found to be incarcerated at the jugular venotomy site by a ring of fibrous tissue. It was recovered without further mishap. This case highlights the potential dangers of "self-assembly" of biocompatible implants, especially those positioned intravascularly. Detachment and embolism seem a real risk. If the titanium tip is to be widely adopted it might be wise to have it bonded to the peritoneovenous shunt at the time of manufacture.

J. R. NOVELL Hepatobiliary and Liver Transplantation Unit, Royal Free Hospital, London NW3 2QG, UK

P. A. MCCORMICK A. K. BURROUGHS K. E. F. HOBBS

1. Moskovitz M. The peritoneovenous shunt: expectations and reality. Am J Gastroenterol 1990; 85: 917-29. 2. Hillaire S, Labianca M, Smadja C, Grange D, Franco D. Improving peritoneovenous shunting in cirrhosis: results of a prospective study. Gastroenterol Clin Biol 1988; 12: 681-86.

Lymphokine-activated killer-cell traffic in metastatic melanoma SIR,-Mr Swift and colleagues (June 22, p 1511) report the successful imaging of colorectal metastases with I IIndium-labelled tumour-activated killer lymphocytes (TAK). They mention the advantages of TAK cells over tumour-infiltrating lymphocytes (TIL)-in that they are not restricted by tumour histology or stage, are easily handled, and are independent of interleukin-2 support in vivo. We have lately reported the imaging of melanoma metastases with radiolabelled lymphokine-activated killer (LAK) cells in four of six patients. Additional interleukin-2 infusion was not necessary.’ Metastases have so far been demonstrated by LAK-cell scintigraphy in 8 (57%) of 14 patients. These cells can be produced quickly and easily. However, only 1 x l0a to 1 x 109 cells (which were used in our protocol) can be obtained with ease by leucapheresis. In contrast to TAK, LAK cells were able to image lymph-node, bone, skin, and gastrointestinal metastases, but not

parenchymatous metastases. patients gave informed consent for biopsy of scintigraphically-positive metastases, and we could therefore analyse the peritumoural infiltrate with a panel of monoclonal antibodies (APAAP-staining of cryosections). Activated T-helper (CD3, CD4, CD25) lymphocytes were identified as the main population in the reactive infiltrate in these 2 patients. No natural Two

Childhood cancer in Kerala, India.

455 anterior lobe of the pituitary. I suggested that further information on this role might become available if anterior-lobe hormones in serum were...
369KB Sizes 0 Downloads 0 Views