Patients with acute poisoning presenting to an urban emergency department of a tertiary hospital in Tanzania

Background Poisoning is a major public health concern in sub-Saharan Africa, affecting patients of all age groups. Poisoned patients often present to the emergency department (ED) and prompt evaluation and appropriate management are imperative to ensure optimal outcomes. Unfortunately, little is known about the specific presentations of poisoned patients in East Africa. We describe the clinical and epidemiological features of patients presenting to the Muhimbili National Hospital (MNH) ED with suspected toxicological syndromes. Methods This prospective study enrolled a consecutive sample of ED patients who presented with a suspected toxicological syndrome from March 2013 to June 2013. Trained investigators completed a structured case report form (CRF) for each eligible patient, documenting the suspected poison, demographic information, the clinical presentation, and the ED outcome and disposition. The study data were analyzed and summarized with descriptive statistics. Results Of 8827 patients, who presented to ED-MNH, 106 (1.2%) met inclusion criteria, and all were enrolled. Among those enrolled, the median age was 28 years (interquartile range [IQR] 16 years), and 81 (76.4%) were male. Overall 55 (52%) were single, and 28 (26.4%) had professional jobs. 60 (56.6%) patients were referred from district hospitals, 86.8% of which were in Dar es Salaam. Only 13 (12.3%) of patients presented to the ED within 2 h of the toxic exposure. The etiology of poisoning included alcohol in 42 (50%), a mixture of different medications in 12 (14.3%), and snakebite in 6 (11.3%). Most exposures were intentional (63 [59.4%]) and were via the oral route (88 [83%]). The most common abnormal physical findings were altered mental status (66 [62.3%]) and tachypnoea (68 [64.2%]). One patient died in the ED and 98 (92.5%) required hospital admission. Conclusions Most patients presenting to the ED with a toxicological syndrome were adult males with intentional exposures. The most common toxic exposure was alcohol (ethanol) intoxication and the most common abnormal findings were altered mental status and tachypnoea. More than three-quarter of patients presented after 2 h of exposure. Almost all patients were admitted to the hospital.


Background
Acute poisoning is a major medical emergency that carries significant morbidity and mortality in patients of all age groups across the world [1][2][3][4]. The World Health Organization (WHO) estimates that poisoning results in an annual loss of 7.4 million years of healthy life (disability adjusted life years) globally. However, this burden disproportionally impacts low and middle-income countries, where over ninety percent of deaths due to unintentional poisoning occur [5].
Acute poisoning may be intentional or unintentional, and studies have shown that intentional poisoning is more common in adults and accidental poisoning is more Open Access BMC Research Notes *Correspondence: hendry_sawe@yahoo.com 1 Emergency Medicine Department, Muhimbili University of Health and Allied Sciences, Dar es Salaam, P.O. Box 54235, Tanzania Full list of author information is available at the end of the article common in children [6]. Patients with acute poisoning have highly variable clinical presentations, making diagnosis difficult, especially in resource limited settings [7,8]. The variability of the clinical presentation is further compounded by the fact that substances involved in poisonings vary by age group, intention, geographic region, and level of economic development [6,[9][10][11]. In the United States, the American Association of Poison Control Centers reports that the top five substance classes most frequently involved in human exposures are analgesics, cosmetics, household cleaning substances, sedatives/hypnotics/antipsychotics, and foreign bodies [8]. European data indicates a different pattern, reflective of the fact that most poisoned patients are adults with suicidal intention, with illicit drugs (mostly benzodiazepines), sedatives/hypnotics/antipsychotics, alcohol, and carbon monoxide as the most common exposures [10].
In Sub Saharan Africa, there is a general paucity of Emergency Department (ED) data describing patterns of acute poisoning, with only a few hospital and mortuary based studies [12][13][14]. In Tanzania, ED based studies reporting on acute poisoning are non-existent, but poisonings have been documented in other clinical settings. In one study of suicides conducted in the mortuary of the main tertiary referral hospital, poisoning was the most common method suicide, mainly from ingesting antimalarials and pesticides [15]. Another autopsy based surveillance study conducted in southern Tanzania documented 300 accidental cyanide poisoning deaths from consumption of bitter cassava during a country-wide drought in 2002-2003 [16]. Poisonings, have also been documented in mining operations (occupational exposures) and among affected pregnant women presenting to a tertiary referral hospital [17][18][19].
In acute settings, early recognition and appropriate management have been known to improve outcomes of poisoned patients [20]. Delayed recognition and suboptimal management may lead to increased morbidity and mortality [21]. The case fatality rates vary by poison and region. In Africa and Asia, fatalities have been shown to be highest from pesticides, followed by medications and household products, while paracetamol was found to be a significant cause of poisoning in other regions of the developing world [6].
In high resource settings, poison control centres are common [22] and most emergency physicians managing poisoned patients have the benefit of real time consultation with the expert toxicologists staffing those centres. Thus, in most EDs in the developed world, physicians are able quickly identify the type of poisoning and institute appropriate management, increasing the likelihood of satisfactory outcomes [23]. Unfortunately, there are currently no formal Poison Control Centres or toxicological consultation services readily available to providers in most Sub Saharan Africa countries, including Tanzania. Recognizing toxic cases is still a challenge in Tanzania because there are relatively few acute care providers with the training necessary to detect the subtle and variable presentations of toxicological syndromes. This issue is compounded by lack of access to laboratory toxicological analysis.
Therefore, we aimed to describe the population of patients presenting to the ED of a large public tertiary care hospital with suspected acute poisoning or other toxic exposure, characterizing their presenting signs and symptoms, and reporting on suspected toxicological agents involved. This was hoped to increase providers' knowledge of patients with suspected toxicological syndromes presenting to EDs and other acute intake areas in Tanzania, thereby informing the development of diagnostic and management protocols and guiding educational and clinical initiatives to reduce the morbidity and mortality from toxicological disease.

Study design
This was a prospective observational study of a consecutive sample of patients presenting to the MNH-ED from 15th March 2013 to 17th June 2013.

Study setting and population
MNH is a tertiary referral hospital in Dar es Salaam, Tanzania and the main clinical training site for the Muhimbili University of Health and Allied Sciences (MUHAS). The ED was opened in 2010, and it is the site for the only emergency medicine residency program in the country. The MNH-ED receives high acuity patients from within Dar es Salaam and the surrounding regional and district hospitals and served 50,000 patients in 2013. The top five patient presentations are trauma, infectious disease, mental health, neoplasm and pregnancy related complications [24].
The department is staffed by locally trained emergency physicians, who provide clinical care, supervision and teaching to interns (fresh graduates from medical school), registrars (generalists) and emergency medicine residents.
In this study, all patients presenting with suspected poisoning were eligible for enrollment. Suspicion of poisoning was determined by the treating physician based on the history, suggestive clinical picture or presence of the containers or wrappers of the poisonous agent or medications.

Study protocol Operational definition
Toxicological syndrome was defined as a group of signs and symptoms constituting the basis for a diagnosis of poisoning.

Inclusion criteria
All patients presenting to MNH ED with provider impressions including but not limited to: alcohol intoxication, recreational or prescription drug overdose, toxic inhalation, chemical skin exposure, accidental or intentional poisoning, and those who had a snakebite or envenomation.
The treating physicians screened and enrolled patients consecutively with support monitoring by one of the study authors (GM). Data collection was done 24 h everyday and 7 days per week. Providers obtained informed consent from all eligible patients or their accompanying eligible proxy. A standardized case report form (CRF) with explicitly defined variables was used to prospectively collect the study data, including age, sex, occupation and other demographic data, as well as substance(s) suspected, details of exposure, prior psychiatric history, household medications, clinical presentation, ED outcomes and disposition.

Data analysis
The study data recorded in the CRF were transferred into an Excel database (Microsoft corporation, Redmond, WA), cleaned, and analyzed with Excel and Stata (version 13, StataCorp LP, Texas, USA). Continuous variables were summarized as means and standard deviations (SD) or medians and interquartile ranges (IQR), depending upon the data distribution. Categorical variables are summarized as counts and percentages.

Study population demographics and referral status
Out of 8827 patients who presented to ED-MNH, 106 (1.2%) met inclusion criteria and all were enrolled.

Route of exposure and intent for all age groups
Overall, nearly 63 (59.4%) of the toxic exposures were intentional, and intentional poisoning was most common in adults 59 (95.2%). Most common route of exposure was oral (88 [83.0%]), followed by bites or stings by venomous animal/insects (11 [10.4%]) ( Table 5).

Patient disposition and outcome
Overall, most of the patients (98 [92.5%]) were admitted to the hospital, 7 (6.6%) were discharged home from the ED, and one patient (0.9%) died in the ED.

Discussion
Most of the patients in our study population were from Dar es Salaam city, reflecting similar patterns of referral to the MNH ED shown in previous studies [25]. We found that the majority of acute poisoning among patients presenting to our ED in Tanzania are males, with a 3:1 male to female ratio. These findings are similar to other studies done in Sub Saharan Africa, where acute poisoning is known to be common among young males [26]. In our study population, single, adult males with a limited level of education represented nearly one-third, which is consistent with a study by Ndosi et al. on fatal self-poisoning patients in Tanzania in which most of the patients were single, with a primary level of education [15]. However, contrary to observations made in other studies in similar settings [15,26], over one-quarter of our patients had professional jobs (including nurse,  engineer, lawyer, businessman, and accountant), with most of these professionals presenting with intentional, recreational poisonings.
Contrary to observations made in most high-income countries [27,28], alcohol taken for recreational purposes was found to be the most common toxic agent in our study population. These findings are different from earlier studies of poisoning done in Tanzania which showed that the most common substances used for fatal self-poisoning were antimalarials and pesticides. This difference may be due to our inclusion of all patients with suspected toxic exposure regardless of intentionality or suicidality, while both previous studies evaluated deceased patients or those with suicidal intention.
In evaluating children as a subset of our study population, we found that most common toxic exposure was snakebite, followed by accidental ingestion of kerosene or rodenticides. Contrary to findings of previous studies in similar settings [26] our findings of snakebites are interesting and suggests the need for a larger multicenter study which can lead to dedicated protocols of snakebite management. We were unable to identify the exact contents of the rodenticides but the most common products available in Tanzania include warfarins and related anticoagulants, zinc or aluminium phosphide, and organophophates. Despite having no reported deaths in the group of children who had rat poisoning, the diversity of the contents and serious potential toxicity continue to pose a significant challenge.
Nearly two-thirds of the toxic exposures occurred at home or in recreational areas, and over half of the toxic exposures were intentional. Intentional poisoning was five times more common in adults than in children. Most children had their toxic exposure at home, similar to other studies from sub-Saharan Africa [29], which have suggested that the culture of leaving children unattended is one important risk factor for domestic based accidents,  including poisoning [29,30]. Most of the intentional toxic exposures were due to recreational use of alcohol and only about a quarter were due to suicidal intention. Community based behavioral interventions [31] may be part of the solution to tackling acute alcohol poisoning incidents in Tanzania. Clinical presentations of toxic exposures depend on the specific agent involved, the quantity absorbed, and the intent of exposure. The most common presenting clinical feature in our study population was altered mental status, which occurred in over 60% of patients. The differential diagnoses and management strategies of patients with altered mental status are widely variable [32], and health providers should always consider poisoning as a potential cause for altered mental status in patients presenting to acute intake areas in Tanzania.
Conservative management-including intravenous fluids and anti-pain-was the main stay of treatment for our patients, and over 90% of these patients were admitted to the hospital, three of who were intubated and admitted to intensive care units (ICU). In our setting mechanical ventilation and ICU care is significantly limited by availability of acute care beds and ventilator machines [33]. As such, we believe it is inappropriate to use ICU admission or intubation as markers of illness severity in this setting. Less than 10% of patients were discharged from the ED, half of who had a discharge diagnosis of alcohol intoxication. All patients who were discharged had stable vital signs; however, due to the nature and timing of the study, and the available resources, we were unable to follow up the discharged patients to investigate their outcomes after discharge. In this study, one patient died in the ED and the toxic substance that was involved was unknown and, as is common in our setting, no postmortem examination was performed.

Limitations
This was a single site descriptive study with a small sample size over a short period of time, at the ED of a large public tertiary referral centre, hence our results may not necessarily be generalizable to other EDs with different patient populations in difference settings. Furthermore, the potential for seasonal variations might not be ascertained within our findings. We relied on the treating doctors' clinical impression to identify eligible patients. These doctors may have variable ability to diagnose toxicological syndromes depending on their level of training and experience, and we have no laboratory analysis capabilities for toxicology screening. As such, we suspect that many toxicology patients pass through the MNH-EMD undiagnosed.

Conclusions
The majority of patients presenting to our ED with poisoning are adult male with intentional exposures, mostly after recreational use of alcohol. Substances used in adult suicidal ingestions included rodenticides, acid, antimalarials, organophosphates, tricyclic antidepressants, haloperidol and opioids. Substances involved in childhood poisonings included kerosene, rodenticides, antimalarials, antibiotics and ethanol. Future studies should focus on larger multicentre patient population and laboratory confirmation of the suspected poisoning agents. Authors' contributions GM, TR, HS contributed to the conception and design of the study, acquired, analyzed and interpreted the data, and drafted and revised the manuscript. SL, JM, JS contributed to the design of the study, data acquisition and entry and also revised the manuscript. KO, VM, MR contributed to the conception and assisted in the initial design of the study and critically revised the manuscript. GM, HS, TR contributed to the conception and assisted in the initial design of the study, data interpretation and critically revised the manuscript, contributed to the conception and assisted in the initial design of the study, analyzed and interpreted the data and critically revised the manuscript. All authors read and approved the final manuscript.