Exposure to cytotoxic drugs in wards : contamination sources and risk assessment
15 novembre 2010W. Fransman Institute for Risk Assessment Sciences, Pays-Bas
Several studies have shown that exposure to antineoplastic drugs can cause toxic effects on reproduction as well as carcinogenic effects. Presence of these drugs in the urine of hospital personnel has been widely studied, but the actual pathway through which exposure occurs was still unknown. Dermal exposure has been suggested to be the main route of exposure, but had never been accurately measured among oncology nurses in their daily work situation.
In a recently conducted study 1, we assessed dermal exposure to cyclophosphamide (a frequently used antineoplastic drug) in four Dutch hospitals during five tasks (preparation, decanting urine, washing the patient, removing bed sheets, and cleaning the toilet) on hands, forearms and forehead. Pharmacy technicians, oncology nurses and cleaning personnel appeared to be exposed via the skin to cyclophosphamide during performance of their daily duties 1. Exposure occurred predominantly on the hands and sporadically on the forehead and forearms. Surprisingly, exposure levels detected on the hands during nursing tasks in wards were higher than those measured in the hospital pharmacy during preparation of cyclophosphamide. Even when nurses used gloves during nursing tasks, skin underneath gloves was repeatedly contaminated. Results of tests on bulk and surface contamination samples confirmed that patients intravenously treated with cyclophosphamide excrete the unmetabolized drug. A clear relationship was found between dermal exposure levels and direct sources of exposure for all tasks 1.
Time trends in exposure
To study the impact of the introduction of guidelines and regulations, we investigated time trends in occupational exposure to antineoplastic drugs among nurses. Data from three cross-sectional surveys were pooled 2. Nurses’ urine samples were analyzed, surface contamination was determined and gloves used during preparation or while handling patients’ urine were collected. The percentage of urine samples with detectable cyclophosphamide decreased 4-fold between 1997 and 2000. Median cyclophosphamide levels in the positive urine samples were 3-fold lower in 2000 than in 1997. Surface and glove contamination also decreased significantly between 1997 and more recent years. The study showed that oncology nurses in the Netherlands are still being exposed to cyclophosphamide, but their exposure decreased considerably between 1997 and 2000/2002 2. Despite this decrease in task-based exposure levels, recent findings show that the median weekly exposure to antineoplastic drugs of the total population of oncology nurses in the Netherlands has not changed between 1997 and more recent years 3, because of an increase in the frequency of task performance related to antineoplastic drugs. This is in accordance with the growing number of patients treated with chemotherapy in the last decade. The identification of the patient as a source of contamination and exposure 1 makes it very complicated to further reduce exposure to antineoplastic drugs of nurses.
Toxic effects on reproduction
In an epidemiological study 4, self-administered questionnaires were completed by exposed and non-exposed nurses in the Netherlands employed between 1990 and 1997. Questions were asked on pregnancy outcome, work-related exposures, and lifestyle. Exposure to antineoplastic drugs was estimated using task-based dermal exposure measurement results and self-reported task frequencies. The results indicated an increased risk of a prolonged time to pregnancy (on average one-month) among nurses with relatively high exposure to antineoplastic agents (>0.74 μg/week) compared with referent nurses. Positive relations were apparent between exposure to antineoplastic drugs and premature delivery and low birth weight. Spontaneous abortion, stillbirth, congenital anomalies, and sex of offspring appeared not to be related to prevailing levels of exposure to antineoplastic drugs 4.
Cancer risk assessment
In addition to the toxicity to reproduction, antineoplastic drugs have been found to have carcinogenic potential. Nine antineoplastic drugs (including cyclophosphamide) have been classified by the International Agency for Research on Cancer (IARC) to be carcinogenic to humans. We recently assessed the carcinogenic risk of occupational exposure to cyclophosphamide. Average task frequencies from the total population of oncology nurses in the Netherlands and dermal exposure intensities for each task were used to calculate oncology nurses’ dermal exposure levels. A dermal absorption model in combination with a physiologically based pharmacokinetic model was used to accurately assess the delivered dose of cyclophosphamide and its active metabolites in the primary target organ : bone marrow. This delivered dose was subsequently related to pharmacodynamic information from a study with cyclophosphamide treated patients to estimate the risk of leukaemia for oncology nurses. The leukaemia risk of an oncology nurse after 40 years of dermal exposure to cyclophosphamide was estimated to be on average 0.27 (range = 0 - 40) extra cases per million oncology nurses. This study concluded that the risk of an oncology nurse exposed to average levels cyclophosphamide is below the target risk of one extra cancer case per 10 million per year, but that this level may be exceeded in a worst-case scenario. We need to emphasize that this estimate is for cyclophosphamide only and in a worst-case exposure scenario the estimated leukaemia risk exceeded the target risk, so the cancer risk among oncology nurses exposed to antineoplastic drugs should not be ignored.
1. Fransman W, Vermeulen R, Kromhout H. Dermal exposure to cyclophosphamide in hospitals during preparation, nursing and cleaning activities. Int Arch Occup Environ Health 2005 ;78:403-412.
2. Fransman W, Peelen S, Hilhorst S, Roeleveld N, Heederik D, Kromhout H. Trends in exposure to antineoplastic drugs among nurses. (submitted) 2006.
3. Meijster T, Fransman W, van Hemmen J, Kromhout H, Heederik D, Tielemans E. A probabilistic assessment of the impact of interventions on the exposure to antineoplastic agents of oncology nurses. Occup Environ Med 2006 ;63:530-537.
4. Fransman W, Peelen S, Roeleveld N, Kromhout H, Heederik D. Nurses with dermal exposure to antineoplastic drugs - Reproductive outcomes. Epidemiology 2007 ;(in press).