Summary. Splenectomized patients are at risk of overwhelming infection and are advised to take life-long prophylactic oral penicillin. Compliance studies have not been published for adults in this situation. We used a standard biological assay to detect penicillin in the urine of 58 splenectomized patients. 24 (42%) patients had evidence of penicillin in their urine. Patients' sex, age, years from splenectomy and underlying diagnosis were not important factors in identifying good or poor compliance. Since 58% of patients did not take their penicillin on the day studied, we need to consider alternative strategies of antibiotic use and patient education.Keywords: splenectomy, penicillin, compliance, infection, prophylaxis.Following splenectomy, patients have a lifelong risk of lifethreatening sepsis (Deodar et al, 1993). Infections are due usually to the encapsulated organisms Streptococcus pneumoniae, Haemophilus influenzae type B (HIB) and Neisseriae meningitidis. Some protection may be provided by immunization, but since 1989 we have been advising patients who are not allergic to penicillin to take prophylactic penicillin 250 mg 12-hourly. This practice has more recently been reinforced by U.K. guidelines (Working Party Task Force, 1996). In view of earlier studies showing poor compliance with long-term penicillin prophylaxis in both adolescents with cancer (Festa et al, 1992) and sickle cell disease (Cummins et al, 1991) we aimed to identify oral penicillin compliance in a heterogenous group of splenectomized adults.
SUBJECTS AND METHODSWe wrote to 88 patients on our register who were known to have had a splenectomy, requesting a urine sample. The letter stated our aim of monitoring and improving patient care without mentioning compliance to penicillin. A contact telephone number was supplied as well as a sample bottle and self-addressed packaging. Samples were returned by post and stored at ¹25ЊC for batch testing. Using urineimpregnated filter paper discs and standard plate-agar a biological assay was performed with Bacillus subtilus, an organism exquisitely sensitive to penicillin. Suitable positive and negative controls were employed. The commonest indication for splenectomy was immune thrombocytopenia (28 cases) followed by Hodgkin's disease (12 cases). The remaining patients had a mixture of malignant and nonmalignant haematological disease. The patients were aged from 24 to 72 years (median 48 years) and had undergone splenectomy a median of 7 years earlier (range 1-40 years). The results were examined by the Chi-square test. In patients where penicillin was not detected by the urine assay a supplementary questionnaire was subsequently sent seeking to establish the reasons for the absence of penicillin.
RESULTSFifty-eight patients responded (66%). Of these, only 24 (42%) had evidence of penicillin in their urine. There was a trend for male patients to be more compliant than females (48% detected v 30% respectively, x 2 ¼3·81; P ¼ 0·06). Age, interval from splenectomy and underlying diagnosis were not i...
