Abstract
Introduction
Antidotes can be used to reverse the pharmacological effects of a particular poison, to displace a poison from its target organ receptor site, or to deactivate the poison by binding irreversibly to affected molecules. 1 Not all toxic substances have antidotes, thus making the use of such a technique is considered a specific treatment. 2 Some patients can recover fully without the use of antidotes. 1 If an antidote is required, its administration should be carried out as early as possible. 3
In certain circumstances, specific antidotes may significantly reduce the requirement for medical interventions. In addition, they can reduce morbidity, mortality, and the length of hospitalization. This is particularly necessary in countries in which high rates of poisonings exist. In these situations, antidotes may be considered essential and life-saving pharmaceuticals. 4 Over the past 25 years, many studies have demonstrated that antidotes are essential and until now insufficiently stocked in health care facilities. 5 –18 The exact causes of this problem are unknown but appear to be associated with limited hospital resources, cost and possible unfamiliarity with antidotes.
Several regional poison control centres in the United States and some textbooks have developed recommendation lists for antidote stocking. 10,19,20 In 2000, the first evidence-based agreement guidelines for stocking antidotes in the United States were published. These guidelines recommended that 16 essential antidotes should be stocked in each health care facility that treats acutely poisoned patients. 10 Antidotes were considered essential if they were both effective and necessary within the first hour of patient presentation. The quantity of antidote recommended was based on the dose necessary to treat one or two 70-kg acutely poisoned patients for the first 4 hours after exposure. 10,19 The WHO guidelines for antidotes and their availability have divided antidotes into those that should be immediately available in the emergency department and those that should be available in the hospital for use as symptomatic treatment of poisoning. 20
There has been no study addressing antidote availability in Malaysian hospitals. Since poisoning remains a serious problem in Malaysia and various antidotes are frequently understocked, we initiated a countrywide survey to describe the current antidote availability and the anticipated requirements in Malaysia. The purpose of this study was to determine the availability of antidotes at hospitals, based on the published guidelines for antidote stocking, and to evaluate the impact of the hospital types on the availability of antidotes for the management of acute toxic exposures and poisonings in Malaysia. The availability of antidotes in each hospital was compared to the 16 antidotes considered essential in the guidelines of Dart et al. for stocking antidotes. 10 Furthermore, the guideline by which we compared the stocking of antidotes were that implemented in the United States, as the nature of poisoning cases in Malaysia might not be widely different than those in the United States. 19,21,22 This is in an attempt to encourage the finding of an appropriate antidote storage and sharing network in this country. The survey results allow us to communicate with the health administrators in Malaysia to define reasonable budgets for an antidote distribution system.
Methods
Design of study
This is a cross-sectional survey study using a questionnaire that involves descriptive and comparative analysis.
Study area and sample size
The data collection was conducted from April to September 2008 and was intended to cover all government hospitals (including different states and hospital levels). The primary target was the Accident and the Emergency Department from each hospital. Hospital category (type) involved general and district hospitals. A list of government hospitals was extracted from Ministry of Health website (http://www.moh.gov.my/Mohportal/govhospPublic.jsp, accessed 28 February 2008). A total of 127 hospitals were identified and the addresses and contact information for each one was extracted and recorded. According to convenient sampling techniques for descriptive studies, 20% of the total would be sufficient. In order to overcome the drop-out rate, another 20% of the total was added to the sample. 23
In Malaysia, they are two types of government hospitals run by the Malaysian Ministry of Health and distributed in each of the 13 states in the country, namely the general hospitals and the district hospitals; the distance between the general hospital and the district one is not more than 100 km. The general hospitals are the biggest government hospitals in each state of Malaysia. They provide health care and emergency treatment for all illnesses and accidents. District hospitals are distributed in the districts of each state. Generally, the large district hospitals have specialist (i.e., medical doctors who have further training in a specific specialty area of medicine). Irrespective of whether the specialist is available, these hospitals accept all cases but may refer cases to a general hospital within the state (http://moh.gov.my/, accessed 6 December 2010).
Study tool: The questionnaire
A semi-structured questionnaire was developed for the purpose of the current study. The questionnaire was designed by the National Poison Centre at Universiti Sains Malaysia (USM) in Penang. It was prepared in English and designed to be self-administered by the heads of the Emergency Departments of each hospital, accompanied by an official document explaining the purpose and importance of the survey. The heads of the Emergency Departments were asked to assign a specialist to complete the questionnaire.
A list of commonly required antidotes and essential drug use for treating poisoning complications was compiled from the WHO and published guidelines for antidotes stocking. 10,20 Participants were asked to indicate which antidotes they had available in their hospital. The questionnaire asked about the epidemiological data pertaining to the number of beds in the hospital and the types and number of poisoning cases admitted during the last year. All aspects of the study protocol were authorized by the Ministry of Health and local health authorities before the initiation of this study.
The survey
On the first week of April 2008, all questionnaires were sent by registered mail to the mailing addresses of the 127 hospitals. A return stamped envelope was enclosed with each questionnaire to facilitate respondents’ return of the answered questionnaire to the office of the National Poison Centre. Feedback from the hospitals was received approximately 1 month later and some responses were received in the second month of the survey. At the end of May, the hospitals who had not responded were given reminders by telephone calls. In July 2008, another reminder was given. The survey was called to a close at the end of September 2008 and deemed completed. Those hospitals that did not provide a feedback were considered as non-respondents.
Statistical analysis
Data extracted from the questionnaires was keyed into the Statistical Package for Social Sciences (SPSS version 16) software. Data were analyzed descriptively as frequencies and percentages and by using comparative analysis. For comparative analysis, the chi-square was used to test the differences between groups. One-way ANOVA with the Tukey post hoc test was used to test differences in the means between categories. A p value of less than 0.05 was considered statistically significant.
Results
Responding hospitals and their general characteristics
Though our sample size was 52 hospitals, we managed to send the questionnaire to all 127 hospitals. Seventy-five hospitals returned the surveys, a response rate of 59.06%. From the total responses received, one questionnaire was discounted as the hospital returned it without filling in the required information. Of the remaining 74 hospitals, 44 were district hospitals without specialists, 19 district hospitals with specialists, and 11 were general hospitals. On average, the general hospitals have the highest mean number of beds (mean: 704.8; median: 718; range: 244–1090), followed by district hospitals with specialists (mean: 311.7; median: 272; range: 94–900). District hospitals without specialists have the lowest number of beds (mean: 94.8; median: 82; range: 24–275). This difference is significant with a p value < 0.001. The type of hospitals with the highest average number of poisoning cases during the last year is the general hospitals (mean: 260.6; median: 196; range: 24–681), followed by district hospitals with specialists (mean: 197.6; median: 88; range: 27–698), while the lowest are district hospitals without specialists (mean: 90; median: 25; range: 2–798). This difference is significant with a p value < 0.037.
Table 1 shows the top 10 toxic agents most frequently reported by the Accident and Emergency Departments to have been treated at least once during the last year at their hospital. Among the responded hospitals, paracetamol poisoning was in the first rank as it is reported by 86.5% of hospitals, followed by snake bites (82.4%) and bee stings (81.1%). Furthermore, 47.3% of hospitals reported at least one case of toxic alcohol exposure, and thus made that toxic agent to be on the tenth ranking place.
Top 10 toxic agents most frequently reported by the Accident and Emergency Departments to have been treated at least once during the last year (number of hospitals = 74)
a The proportion of hospitals reporting at least one treated poisoned patient from the toxic agents during the last year.
Availability of antidotes and essential drugs
Analysis of the collected data, compared to the the guidelines of Dart et al. for stocking 16 emergency antidotes, showed that the availability of specific antidotes ranged from zero (for fomepizole) to more than 90% (for atropine sulphate, calcium gluconate, naloxone, sodium bicarbonate). However, five antidotes were severely deficient in hospitals (ie, available in less than 20% of all hospitals). Those included digoxin immune Fab, dimercaprol, ethanol (100%), methylene blue, and a cyanide kit. None of the responding hospitals stocked all of the antidotes on the list. In relation to hospital type, there is great variability in the availability of antidotes (there were significant differences between hospitals for six items). The availabilities of most items are far better in the general hospitals and the district hospitals with specialists compared to district hospitals without specialists. Calcium gluconate, sodium bicarbonate, atropine sulphate, naloxone, and pyridoxine were available at all general hospitals. Atropine sulphate and naloxone were available at all district hospitals with specialists (Table 2 ).
Percentage of hospitals stocking the antidotes compared according to recommended guidelines a for antidote stocking in emergency departments stratified by hospital type
DHS: district hospital with specialists, DH: district hospital without specialists, GH: general hospital.
aThe availability of antidotes in each hospital types was compared to the 16 antidotes considered essential in guidelines of Dart et al. for stocking antidotes. 10
Availability of other antidotes varied widely from 5.4% (for calcium disodium edentate) to 95.9% (for vitamin K). The availabilities of other antidotes were far better in general hospitals and district hospitals with specialists compared to district hospitals without specialists. Flumazenil and vitamin K were available at all general hospitals (Table 3 ).
The availability of other antidotes and essential drugs in hospitals stratified by hospital type
DHS: district hospital with specialists, DH: district hospital without specialists, GH: general hospital, NSAIDs: nonsteroidal anti-inflammatory drugs.
By hospital type, the availability of essential drugs varied substantially. Nevertheless, these differences did not reach statistical significance in all cases. Overall, the availability of most of the items was excellent, ranging from more than 80% to 100%, except for oxazepam and thiamine (Table 3).
Discussion
In the present work, the response rate was 59.06%, this rate is within the previous ranges reported for studies conducted in accident and emergency department through mailed questionnaires. 24,25 The response rate is higher in this study compared to rates reported for mailed questionnaires for other countries in the Asia Pacific region. 12,26 According to the official reports from the Ministry of Health Malaysia, there is at least one general hospital in each state. The distribution of the hospitals that are involved in the present study represents the distribution of government hospitals according to the type all over Malaysia. For example, the proportion of general hospitals is equal among different regions. This provided good opportunity to have more representative evaluation of the hospitals’ preparations.
Our results show that a large percentage of antidotes were not available in hospitals in Malaysia. None of the responding hospitals stocked all of the antidotes on the list, other than calcium gluconate, sodium bicarbonate, atropine sulphate, naloxone, flumazenil, vitamin K, and pyridoxine, which were available at all general hospitals. Atropine sulphate and naloxone were available at all district hospitals with specialists. However, eight items were severely deficient in hospitals (i.e., available in less than 20% of all hospitals). Those included digoxin immune Fab, dimercaprol, ethanol (100%), methylene blue, isoproterenol, physostigmine salicylates, and a cyanide kit. The antidote fomepizole was not available in any hospitals. Since the timely use of antidotes is potentially lifesaving in certain toxic exposures, maintaining a sufficient stock of antidotes is the responsibility of any hospital that provides emergency care. If a poisoned patient needs a certain antidote that is not stocked at a particular hospital, then either the patient must be transferred or the antidote must be acquired from another hospital.
Alcohol toxicity (ethanol, methanol, and ethylene glycol) has been found to be among the top 10 toxic agents reported by hospitals in Malaysia. Fomepizole, which is used as an antidote for alcohol toxicity, was not available in all hospitals. Ethanol, an alternative antidote, was found to be very scarce among all hospitals (available in <20% of hospitals).
Among hospitals, 86.5% reported paracetamol toxic exposure followed by snake bites in 82.4% and organophosphates in 67.6%. Unfortunately, availability of antidotes used for the treatment of toxic doses of paracetamol, snake bite, and organophosphate exposure were less than the actual proportion of hospitals that reported these cases (74.3% for N-acetylcysteine, 75.7% for polyvalent antivenom, and 45.9% for atropine sulphate). A recent Australian study found that the antidote stocked most frequently at sufficient levels was acetylcysteine (68.1%), which is appropriate given the high incidence of paracetamol overdose in Australia. 18,27
Patients with severe cholinergic syndrome from organophosphate poisoning are likely to die from respiratory failure without the early administration of atropine and/or pralidoxime. Interestingly, atropine availability was 97.3% which is higher than the actual proportion of hospitals (67.6%) that reported organophosphate poisoning. As has been previously reported, 12,16,27 antidotes used to treat conditions other than poisonings were more frequently stocked, such as atropine, dopamine, dobutamine, calcium gluconate, diazepam, and sodium bicarbonate.
Allergic reactions to bee sting are treated like other allergic reactions. Mild reactions are treated with antihistamines. If a more severe reaction develops, epinephrine should be administered. 28 Among hospitals, epinephrine was available in 79.7%, which has been noticeable to be equal to the actual proportion of hospitals (81.1%) that reported bee sting. However, that notice did not confirm whether the hospitals that treated bee sting were the same hospitals that stocked epinephrine. Moreover, since anaphylaxis, delayed allergic reaction and serum sickness have been reported after administration of the antivenom for the snake venom poisoning. 19 Treatment with corticosteroids, H1 blocking agents, and intravenous or subcutaneous epinephrine may be needed if immediate and delayed hypersensitivity reactions occur and should be readily available at the time of antivenom initiation. 29
The availability of other essential drugs did not vary by hospital type. Bronchodilators, corticosteroids, antihistamines, dextrose, diazepam, phenytoin, morphine, and nonsteroidal anti-inflammatory drugs were available in the majority of hospitals, regardless of type. As has been previously reported, 12,16,30 antidotes used to treat conditions other than poisonings were more frequently stocked.
To our knowledge, this is the first publication of antidote stock levels in Malaysian government hospitals. Our findings are consistent with studies from multiple countries that reported variable and inadequate antidote stocking levels. 5 –18 Also, the results obtained were similar to those from developing countries, which suggest that antidote stocking is often inadequate. 16,17 A recent study in South Africa found that seven antidotes were not stocked by any of their hospitals, including digoxin-specific antibody fragments, dimercaprol, calcium disodium edetate, fomepizole, intravenous pyridoxine, silibinin, and succimer. 17 A similar study in north Palestine showed that digoxin-specific Fab antibodies, ethanol, fomepizole, glucagon, penicillamine, physostigmine, succimer, and thiamine were not available at any of the surveyed hospitals. 16
A recent Australian study designed to determine the stock levels of 13 antidotes in Queensland hospitals showed that no hospital had sufficient stock of al 13 antidotes. 18 The proportion of hospitals with sufficient stocks varied from 0% (pyridoxine) to 68.1% (acetylcysteine). Larger hospitals had a higher frequency of sufficient antidote stocks. Only 16% of hospitals claimed to be able to acquire an antidote from another facility within 30 min. 18 Furthermore, another study in Taiwan revealed inadequate stocking of 20 selected antidotes. The study found that physostigmine, cyanide antidote kit, BAL (dimercaprol), EDTA (calcium disodium edentate), methylene blue, Vipera Russell formosensis antivenin, and botulism antitoxin were not available in most (>90%) hospitals. 12
Bed capacity and number of poisoning cases were the only suggested factors associated with higher antidote stock levels. General hospitals with a larger bed capacity stocked more antidotes, consistent with the findings of previous studies. 6,8,10,12,15,18 Also, general hospitals with the highest average number of poisoning cases stocked more antidotes. Another recent study in South Africa found that tertiary hospitals stock 46% of the antidotes while secondary and private hospitals stock 37% and 35%, respectively. 17
A recent Australian study suggested that possible reasons for availability of antidotes include budget constraints in smaller hospitals, perceived lack of antidote need, short antidote expiry dates, or the assumption that patients or antidotes could be transferred quickly enough if required. However, smaller hospitals can still expect to receive patients with acute poisoning, so a degree of preparedness is required. Decisions regarding the stocking of antidotes are complex from a pharmaco-economic perspective. 18 However, in the Australian study, there was no relationship between antidote cost and the frequency with which it is stocked. For example, atropine is used for treatment of anticholinesterase poisoning and was the cheapest antidote to stock. All hospitals stocked some atropine, but only 27% stocked it sufficiently to treat a 70-kg patient. 18
Although this study is the first of its type, there were some limitations to this study. The objective of the study was only to document the availability of antidotes and to evaluate the impact of hospital types on the availability of antidotes for the management of acute toxic exposures and poisonings in Malaysian government hospitals. We did not study other factors that may determine appropriateness of antidote stocking. In addition, this study relied upon self-reported data. No direct observations or location visits were conducted; data collected depended upon knowledge and stock responsiveness of respondents. This study was conducted 2 years ago, thus the results might not reveal current stocks in Malaysian government hospitals, although we are not aware of any changes in strategy that might influence antidote stocking.
Conclusions
Most Malaysian government hospitals stocked certain important antidotes. In relation to hospital type, there was great variability in the availability of antidotes. The availabilities of most antidotes were far better in the general hospitals and the district hospitals with specialists compared to district hospitals without specialists. Raising awareness of the importance of antidotes by instruction, regular evaluation of antidote stocking, distribution strategy, and suitable legislation might provide solutions. Coordination between Malaysian hospitals and the National Poison Centre at Universiti Sains Malaysia is also important.
Footnotes
Acknowledgement
The assistance of the Accident and Emergency department staff is gratefully acknowledged.
The authors would like to thank USM for the financial support provided for their research.
We would like to declare that there is no conflict of interests in conducting this research.