PSEUDOBACTERAEMIA ASSOCIATED WITH CONTAMINATED SKIN CLEANING AGENT

671

POSTOPERATIVE ANALGESIA WITH INTRAVENOUS INFUSION OF ALFENTANIL

SiR,—The satisfactory control of pain after major surgery still remains elusive. The short, terminal half-life (90 min) of alfentanil theoretically makes it the most suitable narcotic analgesic available 1

for intravenous infusion. Yate et al2 infused alfentanil postoperatively in ventilated patients. We have now established the infusion rate required for analgesia, in patients breathing spontaneously for their first three days after major surgery. Alfentanil 20 g/ml was started in the recovery ward, infused via a burette (60 drops/ml) via a separate intravenous cannula, and controlled with an IVAC pump. 10-20 fig/kg was infused over 10 min until patients were pain-free, followed by 10-20 g/kg/h. The rate was then frequently altered until a satisfactory analgesic infusion rate had been determined. After 2-20 hours in the recovery ward, patients returned to the general ward. Every half-hour for the first 12 hand 2-4 hourly thereafter, pulse, blood pressure, and respiratory rate were recorded and patients were asked if they felt sick or had pain, which was scored as none, some, or severe. Most of the measurements were done by nurses, who recorded their own opinion of the efficacy of analgesia, and altered the infusion rate according to the patient’s discomfort, with a maximum allowed infusion rate, after returning to the general ward, of 18 tAg/kg/h. Seven women and six men aged 32-75 yr (mean 62±12 SD) and weighing 51-92 kg (65±11) received infusions of alfentanil for 50-130 h (77-6±19-5h) after surgery for Ivor Lewis

oesophagogastrectomy (five), thoracotomy (three), laparotomy radical vulvectomy (one), thoracic laminectomy (one), and debridement and skin grafting to 50% burns (one). The mean infusion rate was 8 - 2±2 - 5 g/kg/h, with a range of 3’ 8-12 - 3 jg/kg/h. At these rates, patients reported no pain for 74% of the observations and some pain for 25%; one patient only reported severe pain for a short time. The nurses scored analgesia as excellent for 26% of the time, adequate for 70%, and inadequate for 4%. In three patients, the lowest respiratory rate was 12/min, in the other ten the rate always exceeded 14/min. Although two patients vomited and four others were nauseated at some time, emesis was not a problem. Our mean infusion rate of8’2 2 Ilg/kg/h is just under half that known to produce a 50% depression of the carbon dioxide response curve,3 and the respiratory depression seen after higher doses4 did not occur. Despite our long periods of infusion, none of the patients appeared to develop tolerance. Two of them had further infusions after 12 and 14 days (2nd thoracotomy and 2nd debridement of burns) for 65and 44 hours respectively, and showed no increased need for the drug. Rapid alterations in analgesia can be achieved easily with alfentanil, and its use for long-term infusion looks promising (though the product licence for alfentanil does not yet include its use for postoperative infusion).

(two),

Department of Anaesthetics, Plymouth General Hospital, Plymouth, Devon PL4 8QQ

C. J. H. ANDREWS J. A. ROBERTSON J. M. CHAPMAN

Niemegeers CJE, Vanden Bussche G. The development of alfentanil. Br J Anaesth 1983; 55: 147S-56S. 2 Yate PM, Thomas D, Sebel PS. Alfentanil infusion for sedation and analgesia in 1. Cookson RF,

intensive care. Lancet 1984; ii: 396. 3 Andrews CJH, Sinclair M, Prys-Roberts C, Dye A. Ventilatory effects during and after continuous infusion of fentanyl and alfentanil. Br J Anaesth 1983, 55: 211S-16S. 4 Sebel PS, Lalor JM, Flynn PJ, Simpson BA. Respiratory depression after alfentanil infusion. Br Med J 1984; 289: 1581-82.

CHEESE, WINE, AND ISONIAZID

SIR,-Despite the fact that isoniazid is closely related to the monoamine oxidase inhibitors few cases of reaction to food rich in tyramine have been published. 1-3 We describe here a patient who had an unusual reaction to cheese and red wine while taking isoniazid. The patient

was a

43-year-old physician with no special medical

history. In March, 1985, a routine tuberculin test was positive for the first.time. She was prescribed isoniazid/rifampicin (100

mg/150

mg) three

tablets daily and pyridoxine 40 mg daily. Physical examination and laboratory data were normal and at a follow-up 3 weeks later. In mid-April, half an hour after eating pasta with parmesan cheese washed down with burgundy, she noticed

palpitations, severe general flushing, conjunctival suffusion, headache, dyspnoea, tightness of the chest, moderate tachypnoea, and sweating. She came to the intensive care unit when symptoms had worsened for an hour. At that time, she noted tachycardia (130/min) and a small increase from her normal systolic blood pressure (from 100 to 130 mm Hg). The symptoms resolved within an hour after intravenous prednisolone 100 mg. The isoniazid/rifampicin was stopped and the patient was discharged the

evening. urinary 5-hydroxyindoleacetic acid (to exclude carcinoid tumour) was 27 mol (normal 10-42). 3 weeks later she took the same quantity of cheese and wine again without symptoms. 4 weeks later her oxidative phenotypewas found to be "intermediate metaboliser", her metabolic ratio being 3-3("normal" below 1, "poor metaboliser" above 12 - 6). This case is evocative of the cheese reaction in patients on monoamine oxidase inhibitors. A deficient oxidation rate in this patient probably permitted the overloading of tyramine contained in large amounts in both parmesan and red wine, leading to the symptoms described. An unusual reaction to cheese in a patient taking isoniazid may be an indication for a debrisoquine test of oxidation phenotype. Poor or intermediate oxidative metabolism may explain the diminished capacity of the mono-oxidase system to neutralise large amounts of absorbed tyramine. In such patients the choice lies between dietary restriction or withdrawal of isoniazid if an unpleasant reaction is to be prevented. same

24 hour

Department of Medicine, Hôpital des Cadolles, CH-2002 Neuchâtel, Switzerland *Present address. USA.

MOHAMED TOUTOUNGI ROBERTA L. A. CARROLL* JEAN-F. ENRICO LUCIEN PEREY

Department of Radiology, Massachusetts General Hospital, Boston,

DS, Lovenberg W, Keiser H, Sjoerdsma A. Effects of drugs on human blood, platelet and plasma amine oxidase activity in vitro and in vivo. Biochem Pharmacol

1. Robinson

1968; 17: 109-19. 2. Smith CK, Durack DT. Isoniazid and reaction to cheese. Ann Intern Med 1978; 88: 520-21. 3. Lejonc JL, Schaeffer A, Brochard P, Portos JL. Ann Med Int (Paris) 1980; 136: 346-48. 4. Mahgoub A, Idle JR, Dring LG, Lancaster R, Smith RL. Polymorphic hydroxylation of debrisoquine in man. Lancet 1977; ii: 584-86.

PSEUDOBACTERAEMIA ASSOCIATED WITH CONTAMINATED SKIN CLEANING AGENT

SiR,—Dr Ispahani and colleagues (Aug 17, p 383) describe pseudobacteraemia arising from inappropriate procedure when

taking blood cultures. They were able to absolve the skin preparation, but this is not always so. 1,2 We recently obtained positive blood cultures in two patients who were clinically suspected of having bacterial endocarditis. In one patient two out of eight sets of blood cultures yielded a nonaeruginosa pseudomonas, which was identified by the API 20NE test strip as Pcepacia. One week later the same organism was isolated from another patient on the same ward. Although a positive blood culture was not unexpected in these patients, the organism isolated made an environmental origin more likely, and we attempted to the source. On the ward the storage of blood culture bottles was satisfactory and specimens had been collected by different members of the junior medical staff. They said they used ’Hibiscrub’ for initial cleansing of the venepuncture sites before taking blood for culture. Hospital disinfection policy recommends chlorhexidine in alcoholic solution. Hibiscrub contains chlorhexidine in a detergent base, and the container does not state the exact contents. The widespread use of hibiscrub for handwashing ensured that it was within easy reach of the medical staff. The containers hold 500 ml, and two containers with about 60 ml remaining were taken from the ward to attempt to isolate the organism in question. Both bottles were shown to contain P cepacia in numbers in excess of 105 colony forming units/ml and trace

672 the same API 20NE profile as the original blood culture isolates. We therefore feel confident that the organism isolated from the blood cultures was introduced at the time of venepuncture from the contaminated antiseptic. Although care is needed to ensure scrupulous asepsis in the collection of blood for culture this incident emphasises that the agent used for achieving cleansing of the venepuncture site should not only be appropriate but also that any recommendations made should be monitored to ensure that they are being followed.

having

Public Health Laboratory, Northern General Hospital, Sheffield S5 7AU 1. Kaslow 2

P. E. GOSDEN P. NORMAN

RA, Kackel DC, Mallison GF. Nosocomial pseudobacteremia: Positive blood

cultures due to contaminated benzalkonium antiseptic. JAMA 1976; 236: 2407-09. Coyle-Gilchrist MM, Crewe P, Roberts G. Flavobacterium meningosepticum in the hospital environment. J Clin Pathol 1976;

29: 824-26.

HEAT INACTIVATION OF SERUM MAY INTERFERE WITH HTLV-III/LAV SEROLOGY

SiR,-Commercial kits have lately been introduced for the detection of antibodies to human T-lymphotropic virus type III/lymphadenopathy-associated virus (HTLV-III/LAV) in serum. Not all kits licensed for the screening of blood donors can be used for routine diagnostic purposes in other circumstances. The US Centers for Disease Control recommends that sera from individuals suspected or known to be at high risk of HTLV-III/LAV infection be heated at 56°C for 30 min before testing by enzyme immunoassay. Martinet al2 state that sera can be heated without loss of antibody activity. However, we have found that this procedure may interfere with subsequent serological testing. We have tested 15 sera from 15 healthy laboratory workers by the Organon Teknika ’Vironostika’ anti-HTLV-III micro-ELISA and by the Abbott HTLV-III enzyme immunoassay before and after heat treatment. Both kits have been licensed by the US Food and Drug Administration for blood donor screening. All 15 unheated sera were negative in both tests (median ratio of extinction value to cut-off: Organon 0 -4, range 0 -2-0’7; Abbott 0 -4, range 0 -2-0’5). After heating at 560C for 30 min all sera were still negative in the Organon test (median 0 -4, range 0 - 2-0 5) but had become positive in the Abbott test (median 2 - 1, range 1. 2-34). Although the Abbott kit is recommended by the manufacturer for blood donor testing only, for which purpose sera are not routinely heated, this test is also used for diagnostic work with heated sera. The directions enclosed in this kit do not warn against heat treatment.

Rijksinstituut voor Volksgezondheid en Milieuhygiene, 3720 BA Bilthoven, Netherlands 1.

2.

R.

VAN DEN

AKKER

A. C. HEKKER A. D. M. E. OSTERHAUS

Tsang VCW, Hancock K, Wilson M, Palmer DF, Whaley S, McDougal JS, Kennedy S. Instructions for a course in western blot technique. Atlanta, Georgia: Centers for Disease Control, 1985. Martin LS, McDougal JS, Loskoski SL. Disinfection and inactivation of human T lymphotropic virus type III/lymphadenopathy-associated virus. J Inf Dis 1985; 152: 400-03.

TRANSMISSION OF AIDS VIRUS AT RENAL TRANSPLANTATION

The recipient of the other kidney had miliary tuberculosis. Immunosuppressive drugs had been discontinued in December, 1984, but graft function remained normal. This patient was a 52-year-old heterosexual man with no history of drug abuse. In June, 1985, he was HTLV-III antibody positive (Abbott ELISA) with a total lymphocyte count of 1110/1 (OKT4:OKT8 ratio 0’2). Serum from both transplant recipients obtained immediately before transplantation for cross-matching, was available in-the tissue typing laboratory. Assay for HTLV-III antibodies was negative in both. One of the patients did not receive blood transfusion, during or after the transplant surgery. We conclude that both patients were infected by HTLV-III probably transferred in the kidney grafts from the donor. We recommend that potential cadaver donors, especially homosexuals or haemophiliacs, should be screened for HTLV-III infection. CARLOS A. PROMPT MIRIAN M. REIS FERNANDO M. GRILLO JAIME KOPSTEIN ELENICE KRAEMER ROBERTO C. MANFRO MARCELO H. MAIA JAIME B. COMIRAN

Nephrology Service, Hospital Maia Filho and Hospital de Clinicas de Porto Alegre, 90000 Porto Alegre RS, Brasil; and Section of Immunology, Laboratono Weinmann, Porto Alegre

FAMILY ANTHROPOMETRY: A NEW STRATEGY FOR DETERMINING COMMUNITY NUTRITION

SIR,-Anthropometric measurements are generally socially acceptable, have good levels of test/retest and inter-observer reliability, and can be done readily in the field. Indices derived by comparing the measurements with reference data are widely used to show the type and timing of nutritional disorders in children. 1,2 In developing countries, the indices in the children are often used as indicators of the nutritional

indirect

status

of the whole

community ;3In Western countries, anthropometry is increasingly used in adults to measure nutritional status and so help estimate the risk of cardiac and other diseases.4,5 Much more information can be gained by studying the anthropometric patterns of nutrition and malnutrition within and between families rather than in individuals. Weight-for-height is the best anthropometric indicator of present nutritional status and it can be used at all ages. In children the WHO reference sethas been recommended for international use. Weightfor-height can be determined easily.A cut-off point of 90% standard has been widely accepted as the lower limit of normalIn adults, body mass index (BMI, weight/height2) is generally used.s BMI is commonly used to determine the upper limit of normality, so the lower acceptable limit is not well defined. 9, 10 A value of 19 or 20 is generally accepted as a reasonable lower limit, but data from Thailand (unpublished) suggests that 18 might be a better cut-off level for developing countries. The results of weight-for-height indices can be tabulated on a family-by-family basis (see table). This table has been simplified by entering " +" for individuals above the cut-off point, "0" for those below this level, and "-" for no family member. The patterns of malnutrition shown in the table are not exhaustive, but show the main family distributions that can be

expected: SOME PATTERNS OF UNDERNUTRITION IN FAMILIES

SIR,-In May, 1985, a 42-year-old man with chronic renal failure admitted to our hospital with a 6-month history of generalised rash, fever, and malaise. In February, 1984, he had received at another hospital a cadaver kidney graft from a haemophiliac donor who had died of cerebral haemorrhage. Because of irreversible loss of renal function immunosuppressive drugs had been discontinued one month before his May, 1985 admission. He was acutely ill, febrile, and had a widespread vesiculopapular rash with ulcers. His .total lymphocyte count was 1000/1 (OKT4:OKT8 ratio 0-9) and his serum was strongly positive for HTLV-III antibodies (Abbott ELISA). Skin biopsy revealed vasculitis with no evidence of Kaposi was

sarcoma.

2 months after discontinuation of

lymphocyte

count was

immunosuppression his 1200/1 (OKT4:OKT8 ratio 0-97).

+means

weight-for-height satisfactory, family member.

- means no

0

means

weight-for-height below critical level,