886

bleeding in 1. The attempt to instil hyperosmolar glucose had to be abandoned and salpingectomy or salpingostomy had to be done to extend the 5 mm site of tubal rupture and to remove the products of conception. Patients should be monitored after transvaginal ultrasound guided puncture of tubal pregnancy, so that pelviscopic intervention, if necessary, can be done without delay. We feel that puncture of ectopic pregnancies should not be an outpatient procedure. Department of Obstetrics and Gynaecology, University of Graz, A-8036 Graz, Austria

W. HONIGL P. LANG

W. URDL P. A. M. WEISS

for a range of 23 tests, 99% of urgent and non-urgent specimens are analysed within 30 and 90 min, respectively, of specimen receipt. Although these analysers are complex and need supervision by skilled individuals, the routine operation is simple allowing high quality analytical data to be easily obtained by workers who are not fully trained in the wider aspects of laboratory work; this will undoubtedly help laboratories to meet the recruitment problems expected to result from demographic changes in the 1990s. Department of Clinical Chemistry, Western General Hospital, Edinburgh EH4 2XU, UK 1.

1. Robertson DE, Smith W, Craft I. Reduction of ectopic pregnancy by ultrasound methods. Lancet 1987; ii: 1524. 2. Timor-Tritsch I, Baxi L, Peisner DB. Transvaginal salpmgocentesis: a new technique for treating ectopic pregnancy. Am J Obstet Gynecol 1989; 160: 459-61. 3. Leeton J, Davison G. Nonsurgical management of unruptured tubal pregnancy with intra-amniotic methotrexate: preliminary report of two cases. Fertil Steril 1988; 50: 167-69. 4. Feichtinger W, Kemeter P. Treatment of unruptured ectopic pregnancy by needling of sac and injection of methotrexate or PG E2 under transvaginal sonography control. Arch Gynecol Obstet 1989; 246: 85-89. 5. Lang PF, Weiss PAM, Mayer HO, Haas JG, Honigl W. Conservative treatment of ectopic pregnancy with local injection of hyperosmolar glucose solution or prostaglandin-F2x: a prospective randomised study. Lancet 1990; 336: 78-81. 6. Lmdblom B, Enk L, Hahlin M, et al. Non-surgical treatment of ectopic pregnancy. Lancet 1988; ii: 1403.

Dry-slide clinical chemistry S:R,—Clinical laboratories face mounting pressures. The spread of HIV in the unidentified

community has resulted in more identified and "high-risk" specimens than before. Their safe handling places greater demands on resources at a time when funding is increasingly restricted and poor salaries impede the retention of trained staff. Furthermore, the drive for clinical efficiency makes it essential that laboratories provide a high-quality, timely, and cost-effective service. Despite technological advances the efficiency of many biochemistry laboratories is impaired because samples continue to be processed at different work stations, even for automated tests. Further inefficiencies arise when paediatric, urgent, and known high-risk samples are analysed either on separate, expensive dedicated instruments or by labour-intensive procedures. However, the application of photographic layering technology to biochemical analysis! now offers a viable, highly efficient alternative to traditional "wet" chemistry techniques. This technology not only eliminates liquid reagents, and therefore minimises the risk of analyser contamination with potentially infectious materials, but also greatly simplifies the analysis. The instrumentation needs little daily maintenance and the inherent stability of analytical slides means calibration is necessary only every three months. This analytical approach has important practical and economical advantages. Firstly, the inherent safety makes the technology highly suitable for laboratories serving populations with a high prevalence of HIV infection, including enzymes, and ensures the availability of a wide range of tests both to known HIV positives and to most other high-risk patients; separate expensive equipment and labourintensive procedures for the processing of such specimens are therefore not needed. Secondly, because daily calibration is unnecessary and there is no start-up time, a wide range of tests, including enzymes, can be rapidly performed with little effort, both

G. BLUNDELL

Shirery TL. development of a layered-coating technology for clinical chemistry Clin Biochem 1983; 16: 147-55

Aprotinin and reduction of blood loss and transfusion requirements in orthotopic liver transplantation SiR,—Orthotopic liver transplantation (OLT) is frequently complicated by severe coagulopathy and blood loss. Accelerated fibrinolysis has been identified as an important component of the haemostatic disorders that contribute to perioperative bleeding.’ Aprotinin has been reported to reduce blood loss in OLT; however, Hunt et aP have shown no such reduction, despite evidence of striking inhibition of fibrinolytic activity by aprotinin. This is contrary to our experience. In a pilot study we examined the effects of aprotinin on transfusion requirements and coagulation profiles in 12 patients undergoing OLT for end-stage liver disease and in an equal number of consecutive transplants in patients with similar pathology who did not receive aprotinin. All transplants were done by one surgeon and with a standard anaesthetic technique. No form of extracorporeal bypass was used in any patient. Aprotinin was given as a loading dose of 2000 000 Kallikrein inhibitory units (K1U), followed by an infusion of 500 000 KIU/h until the patient’s return to the intensive care unit. In addition 70 000 KIU were given with each unit of blood transfused. Intraoperative transfusion of blood products in the aprotinin group was less than a third of that in the controls. Operative times were also reduced and, as would be expected, the improved perioperative course resulted in a shorter stay in intensive care

(table). Our haematological data confirm that aprotinin causes profound inhibition of fibrinolysis, which is likely to be the main mechanism by which blood loss is reduced. In 4 of 6 controls thrombelastography revealed in the anhepatic stage severe fibrinolytic changes, which accelerated in 3 after reperfusion. The degree of fibrinolysis was related to the amount of blood loss. By contrast, in the aprotinin group only 1 of 12 patients showed fibrinolytic activity on thrombelastography in the anhepatic period and there was no evidence of fibrinolysis in any patient following reperfusion. This inhibition of fibrinolysis was reflected in a striking reduction in transfusion requirements during the post anhepatic stage (mean 0-8 units, range 0-2). This is in direct contrast to the usual pattern of transfusion during OLT, where there is generally a similar transfusion requirement during the pre-anhepatic, TRANSFUSION REQUIREMENTS, OPERATIVE TIME, AND ITU STAY IN OLT PATIENTS RECEIVING APROTININ AND IN CONTROLS*

within and outside usual working hours; this enhances the quality of service to the clinical user and ensures the proper continuity of analytical data irrespective of the time of day. Internal laboratory

efficiency is improved not only by elimination of the requirement for separate equipment and staff to process urgent requests but also by removal of the need to reprocess "on call" samples for additional but non-urgent tests. Thirdly, because the volume of sample needed is small (10-11 ltl) the technology may also be used for children’s samples (including those from newborn babies). Finally, the wide range of tests that can be continuously available on a single analytical system ensures the complete and efficient processing of nearly all samples at a single work station. For example, in this department

J. P. ASHBY

I

*Mean

(range) tp < 0 002, Mann WhitneyU

test RBC=red blood cells, FFP=fresh frozen plasma

887

anhepatic,

and

post-anhepatic stages. Moreover,

if there is poor

initial graft function, there is often a pronounced increase in blood loss in the post-anhepatic stage. Following reperfusion, in the patients treated with aprotinin the operated area was remarkably dry and haemostasis where needed was rapidly and easily achieved, resulting in reduced operation time. Our findings have implications for the costs of liver transplantation, but more importantly for patient outcome.

Liver Transplantation Unit,

Royal Free Hospital and School of Medicine, London NW3 2QG, UK

S. V. MALLETT D. COX A. K. BURROUGHS K. ROLLES

1. Porte RJ, Bontempo FA, Knott EAR, et al. Tissue type plasminogen activator associated fibrinolysis in orthotopic liver transplantation. Trans Proc 1985; 21: 3542 2. Neuhaus P, Bechstein WO, Lefebre B, et al. Effect of aprotinin on intraoperative bleeding and fibrinolysis in liver transplantation. Lancet 1989; ii: 924-25. 3. Hunt BJ, Cottam S, Segal H, et al. Inhibition by aprotinin of tPA-mediated fibrinolysis during orthotopic liver transplantation. Lancet 1990; 335: 381.

this protocol. Adjuvant radiotherapy has been discontinued by the UKCCSG on the basis of this randomised clinical trial which indicated that it provided no advantage over chemotherapy alone.2 No genetic conditions predisposing to leukaemia were diagnosed in these six patients. Pui et al conclude that the absolute risks of acute myeloid leukaemia were low after solid tumours of childhood. In Britain this conclusion should be modified: while the absolute risk is low after solid tumours other than NHL, among patients treated for NHL during the early 1980s the risk was somewhat high. Childhood Cancer Research Group,

University of Oxford, 57 Woodstock Road, Oxford OX2 6HE, UK 1. Hawkins MM. Second 2.

3.

M. M. HAWKINS cancer

in relation to treatment for childhood cancer in Britain

(abstract). J Cancer Res Oncol 1990; 116 (suppl): 982. Mott MG, Eden OB, Palmer MK. Adjuvant low dose radiation in childhood non-Hodgkin’s lymphoma. BrJ Cancer 1984; 50: 463-69. Ingram L, Mott MG, Mann JR, et al. Second malignancies in children treated for non-Hodgkin’s lymphoma and T-cell leukaemia with the UKCCSG regimens. Br J Cancer 1987; 55: 463-66.

Cuff size and blood pressure SiR,—Your editorial on the effect of cuff size on blood pressure readings (Aug 18, p 410) states that most sphygmomanometers are provided with the standard cuff, which gives higher readings than those obtained with the larger cuff. The large cuff is recommended because this would avoid treatment for many, supposedly normotensive, subjects. You fail to recognise that the standard cuff used to obtain most of the data that determine blood pressure levels for which treatment is indicated. If the larger cuff is used for screening, the criteria defming hypertension should be lowered by the differences in readings due to cuff size. If this is not done, those who warrant treatment based on epidemiological data obtained with the standard cuff will be falsely identified as normotensive and will not get that treatment.

was

4077 Olive Hill Drive, Claremont, California 91711, USA

SYLVAN GOLLIN

Risks of myeloid leukaemia in children treated for solid tumours SIR,-Professor Pui and colleauges (Aug 18, p 417) report the risks myeloid neoplasia (acute myeloid leukaemia [AML] or myelodysplastic syndrome) after treatment for solid tumours in of

childhood. I have reported the risks of second primary leukaemia within a cohort of children (aged under 15 years) diagnosed with cancer in Britain between 1962 and 1983 and surviving at least 1 year.! The number surviving 1 year (and the number of AMLs) after each type of solid tumour were: Hodgkin’s disease 1046

Flow cytometry for measurement of antibodies to spermatozoa in subfertile men SIR,-Prednisolone treatment has been shown to result in a decrease in seminal antibody titres and improved fertility.! However, unreliable tray agglutination and mixed antiglobulin tests were used to measure the antibodies. We suggest the use of flow cytometry to monitor the density of antibody bound on spermatozoa while evaluating patients on steroid treatment. The method can be used to monitor such antibody levels in individuals, or to compare different samples under various flow cytometer laser power/photo multiplier tube voltage settings. To estimate the density of spermatozoan antibody binding, spermatozoa from subfertile patients were labelled with propidium iodide (PI) and fluoresceinisothiocyanate (FITC) conjugated F(ab’), IgG or IgA antibodies. On dual colour analysis the percentage of antibodypositive live cells were noted together with their mean channel numbers. From the mean channel numbers the fluorescein equivalents per cell were read off a standard curve (figure). The average number of antibody molecules per spermatozoa is obtained by division of the number of fluorescein equivalents by the fluorescence/protein ratio (provided by the manufacturer) of the FITC-antibody conjugate. This gives the density of antibody binding on spermatozoa. We have used this method on 6 spermatozoa-antibody-positive patients, in 1 of whom the complete disappearance of IgG

(3), non-Hodgkin lymphoma (NHL) 699 (6), astrocytoma/ medulloblastoma 2258 (5), neuroblastoma 712 (1), and other solid tumours 7036 (0). The Kaplan-Meier cumulative risks of AML during 10 years’ follow-up after surviving 1 year were 0-4% (SE 0-2%) for Hodgkin’s disease, 1-4% (0-5%) for NHL, and 0-3% (01%) for astrocytoma/medulloblastoma. The average follow-up beyond 1-year survival was 7-7 years. The absolute risks estimated from these data are similar to those

reported by Pui et al, although the risk after Hodgkin’s disease may be smaller in the present study; however, myelodysplastic syndrome was not included. The largest cumulative risk of AML was seen after NHL; most patients were treated with

chemotherapy, with or without radiotherapy. The cumulative risks of AML at 5 years from 1-year survival after chemotherapy for NHL in the 1960s, 1970s, and 1980s were 0%, 0-7%, and 5-6%, respectively (test for trend p 0-0068). Among those treated in the 1980s with chemotherapy alone, or with chemotherapy and radiotherapy, the cumulative risks at 5 years from 1-year survival were 2-5% and 82%, respectively. All six patients in whom AML developed were entered into a UK Children’s Cancer Study group (UKCCSG) clinical trial,z and Ingram and co-workers3 have reported the high risk of AML among patients treated according to =

-j

Mean fluorescence channel number Comparison of fluorescence intensity between samples. ’Immuno-Brite’ (Coulter) beads with 31 000 (I), 115 000 (II), 460 000 (III), and 1155 000 (IV) fluorescein equivalents per bead were run on an ’FACS-Star’flow cytometer with low (A) and high (B) photo multiplier tube voltages Mean channel numbers for any bead (I-IV) differ in curves A and B, but log fluorescein equivalents are the same