The aim of this study was to review the aspects of malaria at a Canadian pediatric hospital and to identify gaps in management. Thirty-eight cases were diagnosed in patients with an average age of 8.4 years, the majority of which were due to Plasmodium falciparum. Two required intensive care, but survived. The majority of patients were immigrants/refugees, and accounted for almost all cases over the last 8 years of review.
Almost 1 million Canadians travel annually to malaria endemic areas, with several hundred cases reported each year. Travelers who visit friends and relatives (VFRs) are well known to be at increased risk for malaria.[2, 3]
Anecdotally, cases of malaria at Winnipeg Children's Hospital (WCH) appeared to be increasing over time. The aim of this study was to review the aspects of malaria at WCH in both travelers and immigrants, and to identify possible gaps in management.
Charts for all cases of malaria in children (≤18 years), identified by ICD-9 code and hematology lab record review and confirmed by positive thick or thin smears, as reviewed by a hematopathologist, presenting to WCH from January 1, 1989, to December 31, 2008, were retrospectively reviewed. Data were collected by way of a collection tool. Our hospital is the only tertiary pediatric center for the province of Manitoba, northwestern Ontario, eastern Saskatchewan, and southern Nunavut. There are an estimated 50,000 outpatient visits to the emergency department (ED) per year. Data were analyzed using Microsoft Excel (Microsoft, Seattle, WA, USA). The review was approved by the University of Manitoba Bannatyne Research Ethics Board. Statistical comparisons were done using Fisher's exact test.
From 1989 to 2008, 38 cases of pediatric malaria were identified in patients presenting to WCH. The mean age of cases was 8.4 ± 4.6 years, and 50% were male. Most cases occurred in older children, with 24 cases (63%) > 6 years of age. On average, two cases of malaria were identified per year. Twelve cases occurred in pediatric travelers from malaria non-endemic areas (11 from Canada, 1 from UK), 11 of which were among VFRs (children born in Canada or overseas, returning to family's nation of origin to visit friends and relatives). Six VFRs traveled to India, and five to sub-Saharan Africa. One child traveled to the Solomon Islands with family on business. The mean time from date of return to Canada to diagnosis was 123.3 days. The remaining 26 cases occurred among new immigrants and refugees, with a mean time from arrival in Canada to diagnosis of 92.3 days. Only 4 immigrants emigrated from India or Pakistan, while 22 emigrated from sub-Saharan Africa (Nigeria and Mozambique most commonly). All but two (93%) of the immigrant/refugee cases presented from 2000 onwards, whereas only four (33%) of the travel-related cases occurred in the same time period (Figure 1). From the traveler's group, information about pre-travel counseling was available for 10 patients of whom 6 consulted a clinician prior to travel, none via a travel clinic. Only one child was prescribed appropriate malarial prophylaxis for the area of travel, and the parents of that child forgot to administer it overseas (two cases not specified, one given nothing, two given chloroquine inappropriately).
Common symptoms at presentation included fever (63%), chills (34%), headache (32%), anorexia (29%), fatigue (26%), abdominal pain (24%), emesis (21%), diarrhea (16%), and cough (13%). Of the children who were afebrile, 1 presented with gastroenteritis, and 13 were diagnosed after a family member was recently diagnosed with malaria, and were relatively asymptomatic. There were no significant differences in presenting symptoms between those < 6 years and ≥ 6 years of age (p = 0.07).
The mean peak parasitemia was 2.2% (range 0.01%–19.3%), and was 2.5% in those with Plasmodium falciparum infection. Severe malaria with a parasitemia of >5% occurred in three cases, all in immigrants <6 years of age from Mozambique. There were no mortalities. Two children required admission to the intensive care unit. The causative species of Plasmodium in the 38 cases were most commonly P falciparum alone (29%) or a mixed infection with P falciparum and Plasmodium vivax (29%). The remainder included P vivax alone (26%), P falciparum with non-P falciparum species (10%), P falciparum with Plasmodium ovale (3%), and P ovale alone (3%). Among the children who had traveled, P falciparum was the most commonly identified species (7/11, 63%). P vivax was seen in 100% of cases from India/Pakistan, but in only 37% of those from Africa.
Nineteen cases (50%) were admitted to hospital for an average of 2.6 ± 1.9 days. In 20 cases, there was documentation that the child was seen by an offsite clinician before presentation to WCH. Only half the children (55%) had a malaria smear performed at an outside facility, and 80% (16/20) had more than a 24-hour delay from the time of initial assessment to the time of presentation at WCH. Of the cases involving P falciparum, all but one was treated with a quinine-containing regimen. For cases with only P vivax or P ovale, treatment information was available for 9 of 11 cases, with 4 receiving a regimen of quinine/doxycycline/primaquine and 5 receiving chloroquine/primaquine. At WCH, the mean time from smear collection to initiation of antimalarials was 6.8 hours (range 1.3–10 h); however, documentation was available only for 10 cases (26%). Intravenous antimalarials were used in two ICU cases (quinine), and no exchange transfusions were performed.
Pediatric malaria presenting to Canadian tertiary care centers has been the subject of a limited number of reports from very large urban centers.[4, 5] In a series of 40 cases from Vancouver, the majority (71.4%) occurred in travelers, with only 28.6% in immigrant or refugee children, and P falciparum was identified in only 7% of cases overall. Goldfarb and colleagues described 58 pediatric cases (81% were P falciparum) at the Children's Hospital of Eastern Ontario in the setting of changes in malaria management in the emergency room, but did not distinguish between infections in travelers versus immigrants/refugees. Active surveillance of 100 cases of imported malaria in Toronto, including both adults and children, found that 83% had been Canadian residents for >2 years. By contrast, the vast majority of cases in our study were recent immigrants or refugees, with an average time from arrival to diagnosis of ∼92 days.
Changes in immigration patterns in Manitoba likely influenced the results of our study. Reports from the Government of Manitoba reveal increasing immigration rates from 2002 (<5,000) to 2008 (>11,000). Top source nations were the Philippines, Germany, and India. Ethiopia was the highest ranked African source nation. In 2008, 29% were refugees, family class, or economic migrants, with the top source nations for refugees being the Democratic Republic of the Congo, Ethiopia, Afghanistan, Myanmar, and Sudan. Seventy-one percent applied via the provincial nominee program, an economic stream for skilled workers. For this category, Manitoba received the largest percentage in Canada (35.5%). Our numbers, although small and limited by the nature of a retrospective chart review, seem to parallel this increasing trend in immigration to Manitoba from malaria endemic countries. Of immigrants to Manitoba in 2008, over 7,600 were from Southeast Asia or Africa. The high percentage of cases with P falciparum and P vivax in our study appears to reflect the expanding demographics of immigrants and refugees to Manitoba. Canadian guidelines do not recommend routine screening of asymptomatic immigrants and refugees for malaria. A recent study from Canada has shown that polymerase chain reaction (PCR)-based testing detects Plasmodium DNA (including that of P vivax and ovale) in some asymptomatic recently arrived refugees. Our study did demonstrate a higher proportion of mixed infections than others.[4, 10] Nucleic acid-based detection was not routinely available at our center during the study period, and there may have been variability in skill level between hematopathologists which may have changed through the study period. No cases occurred where P falciparum was misidentified as non-falciparum species on the initial smear. Access to nucleic acid-based testing would allow for a clearer understanding of the epidemiology of imported malaria over time.
Current Canadian recommendations for the treatment of malaria in children are similar to those in adults. For severe P falciparum infection, parenteral artesunate is the therapy of choice, available through the Canadian Malaria Network. For uncomplicated P falciparum acquired in a chloroquine-resistant area, oral therapy with atovaquone/proguanil (Malarone) or quinine and a second drug (such as doxycycline, or clindamycin if doxycycline is contraindicated) is recommended. The WHO recommends oral combination therapy with artemesinin derivatives as first-line choice, but these agents are not yet available in Canada. Our study spanned a period prior to the widespread availability and use of Malarone in Canada, which is now the first-line therapy for uncomplicated P falciparum at WCH.
Prompt diagnosis and treatment of malaria are key to good outcomes.[1, 10] In this study many children had a significant delay from their initial clinical assessment before presentation to our facility. Options to decrease time to therapy once malaria is suspected include stocking antimalarials in the ED, access to rapid diagnostic tests in rural areas, and possible presumptive antimalarial therapy. This study reinforces that clinicians need to consider malaria in the diagnosis of a febrile child with an appropriate travel history, and to utilize appropriate resources for timely diagnosis and therapy. Immigration to regional Manitoba communities has been increasing, with 23.3% more immigrants settling outside of Winnipeg from 2007 to 2008; therefore, clinicians in both urban and rural communities may encounter children with malaria. Our study would seem to indicate that frontline clinicians and residents in Manitoba may require ongoing education and formal academic teaching (resident academic days, province-wide Pediatric Grand Rounds) on the diagnosis and management of clinical malaria, rather than a focus on screening and presumptive treatment of migrants. Ongoing reinforcement could include communication via the bulletin of the provincial medical college sent to all physicians, done by our group initially. As pre-travel services are not covered by provincial health plans in Canada, the associated costs may be a barrier for travelers obtaining appropriate advice regarding malaria prevention, especially VFRs. Clinicians in Canada should advocate for the coverage of pre-travel care, especially for children.
Declaration of Interests
S. T. F. was supported by a clinical postdoctoral fellowship from the Manitoba Institute of Child Health. The other authors state they have no conflicts of interest to declare.