Malaria is a serious and potentially fatal infection. It is

Transcription

1 A Regional Centralized Microbiology Service in Calgary for the Rapid Diagnosis of Malaria Deirdre L. Church, MD, PhD, FRCPC; Angelika Lichtenfeld, MLT, ART; Sameer Elsayed, MD, FRCPC; Susan Kuhn, MD, FRCPC; Daniel B. Gregson, MD, FRCPC Context. A regional centralized laboratory service for the rapid diagnosis of malaria was implemented 3 years ago in May 1999 within the Division of Microbiology, Calgary Laboratory Services. Objective. To describe the design and performance of this unique microbiology laboratory service. Design. Blood specimens must arrive at the central laboratory within 2 hours of collection. Thin blood smears are read and reported from suspected acute cases within 1 hour of receipt, 24 hours per day, 7 days a week, by trained and experienced microbiology technologists. All positive malaria smears are reviewed by a medical microbiologist and confirmed by polymerase chain reaction at a reference laboratory. Setting. Calgary Laboratory Services provides integrated laboratory services to the Calgary Health Region, an urban area of more than 1 million people. Main Outcome Measures. Performance of the service has been continuously monitored by measuring preanalytic and analytic test turnaround times, test accuracy, clinical relevance, and the results of proficiency testing. Results. More than 90% of blood specimens for malaria from community locations have consistently arrived within 2 hours of collection, and hospitals have reached this target within the past year. Although polymerase chain reaction was more sensitive at detecting the presence of malaria, the expert microscopists were as accurate at determining the type of Plasmodium infection. More than 95% of all positive smear results are consistently reported within 2 hours of receipt of a blood specimen. Conclusions. Implementation of a regional centralized microbiology service has improved our ability to make a rapid and accurate diagnosis of malaria in this region. (Arch Pathol Lab Med. 2003;127: ) Malaria is a serious and potentially fatal infection. It is caused by the coccidian protozoan parasite of the genus Plasmodium (ie, Plasmodium falciparum, Plasmodium vivax, Plasmodium ovale or Plasmodium malariae), which may be carried in tropical and subtropical areas by female Anopheles species mosquitoes. 1,2 In Canada, the rate of imported cases continues to be 3 to 10 times the per capita rate of the United States. 3 The number of reported cases of malaria in Canada has steadily increased throughout the 1990s, reaching a peak of 1036 cases in ,4 Due to the underreporting of malaria cases to public health agencies, this figure may be a gross underestimate of the true number of infections being diagnosed and treated each year. Infection with P falciparum in particular can be fatal, owing to the high parasitemia level that occurs because all stages of red blood cells (RBCs) in the circulation may be infected by the parasite. 2 The case fatality rate varies from 0.6% to 3.8% for less severe P falciparum infections, but may exceed 20% in patients who require admission to Accepted for publication December 27, From Calgary Laboratory Services, Calgary, Alberta (Drs Church, Lichtenfeld, Elsayed, and Gregson); and the Departments of Pathology & Laboratory Medicine (Drs Church, Elsayed, and Gregson), Medicine (Drs Church and Gregson), and Pediatrics (Dr Kuhn), University of Calgary, Calgary, Alberta. Reprints: Deirdre L. Church, MD, PhD, FRCPC, Calgary Laboratory Services, th Ave SW, Calgary, Alberta, Canada T3C 0J5 ( deirdre.church@cls.ab.ca). an intensive care unit. 1,2 There have been an increased number of recent malaria deaths due to P falciparum infections in Canada, with 7 fatalities reported between 1997 and ,5 Management of malarial infections is becoming more complex owing to the increasing resistance of P falciparum to chloroquine and other antimalarials. 6,7 Since delays in recognition and diagnosis of acute malarial infection are associated with increased morbidity and mortality, clinicians must have a high level of suspicion for malarial infection in anyone presenting with a febrile illness who has recently returned from an endemic area. A rapid, accurate diagnosis must also be provided by the laboratory s malaria service. Prior to 1996, when laboratory services underwent restructure in Calgary, malaria diagnostic services were decentralized in the hospital as well as private community laboratories, analytic procedures were not standardized, and there was no routine monitoring of either the amount of time taken to transport blood specimens or to report a smear result. 8 Malaria diagnosis relied on the microscopic examination of blood smears by laboratory medical staff. General pathologists read all malaria smears in private community laboratories, while hematopathologists performed this function in acute care hospitals. Only at the Foothills Medical Center, the major academic tertiary-care hospital in Calgary, did an expert group of highly trained microbiology technologists read all malaria smears from patients attending the Travel Clinic at the University of Calgary. Although laboratory medical staff were most of- Arch Pathol Lab Med Vol 127, June 2003 Regional Malaria Diagnostic Service Church et al 687

2 ten able to recognize a malarial infection, frequently the report provided either no parasite-species identification or an incorrect one. In addition, it usually took several days (on average 4 days from receipt of a blood specimen) for a malaria smear to be read and reported by the laboratory medical staff. Because of the variable proficiency and timeliness of laboratory medical staff in accurately reading malaria smears, a centralized regional microbiology service where all smears are read by an expert group of highly trained microbiology technologists has been implemented during laboratory restructure in Calgary. This article describes and evaluates the performance of this unique regional malaria diagnostic service model. MATERIALS AND METHODS Laboratory Setting Calgary Laboratory Services (CLS) is a large integrated medical laboratory company that provides diagnostic testing to the Calgary region, an urban health care region of more than 1 million people. Overall, CLS transports and analyzes nearly 1 million patient specimens each month. The majority of microbiology services from hospitalized and ambulatory patients are consolidated into a single centralized laboratory located in the community; this laboratory provides service 24 hours a day, 7 days a week. The central microbiology laboratory receives approximately 2200 patient specimens each day. Calgary Laboratory Services operates its own regional fleet of courier vehicles to deliver patient specimens to the central laboratory every 1 to 1.5 hours from each of 4 hospitals and 20 Patient Service Centers (PSCs) strategically located in the community. The processes and procedures used recently to restructure microbiology services in the Calgary region have been outlined previously by Church and Hall. 8 In May 1999, CLS implemented PathNet Classic (Cerner Corporation, Kansas City, Mo) as the regional laboratory information system. Malaria Diagnostic Service Design Laboratory restructuring in Calgary created the opportunity to develop and implement a centralized regional service for malaria diagnosis. To have highly trained personnel immediately available to read malaria smears, the laboratory service was transferred from the Divisions of Pathology and Hematology to the Division of Microbiology. A service delivery model was then designed for an excellent and clinically relevant system for the rapid, accurate diagnosis of malaria that would meet the existing published standards. 1,2 The following laboratory procedures outline the essential elements of that service. 1. Collection Procedures. Phlebotomists collected a minimum of 5 ml of blood for malaria diagnosis into a tube containing EDTA. The ordering physician was required to complete a CLS Malaria History Form and Microbiology Requisition, and both forms accompanied the blood sample to the laboratory. The clinical history requests information about travel to a country or geographic region where malaria is endemic and about recent clinical symptoms, such as fever. The test will not be done if the patient is asymptomatic and has never traveled to a malaria-endemic area. 2. Transportation of Blood Specimens. Priority transportation is arranged from each collection site for blood samples for malaria diagnosis, since exposure to EDTA causes significant deterioration in the morphology of the malaria parasite. 9 Hospital rapid-response laboratories (RRLs) and community PSCs immediately call the central microbiology laboratory after collecting blood for malaria testing to inform them that the sample is being sent. The microbiology technologist records the patient s name and the hospital RRL/PSC site and makes sure that a clinical history has been obtained and that the specimen is being sent stat. The Division of Microbiology routinely monitors the transport time of blood specimens for malaria in the Calgary region and expects that the central laboratory will receive a minimum of 90% of the blood samples within 2 hours of collection. A preanalytic turnaround time analysis could only be readily performed for the first quarter of 2000 (January to April) and each subsequent year (ie, May 2000 April 2001 and May 2001 April 2002) since CLS keeps only 2 years of current microbiology data on the laboratory information system. Laboratory staff in rural regions outside Calgary have been instructed on how to make thin and thick smears, which are air dried and immediately transported along with the rest of the blood specimen to the CLS central microbiology laboratory. 3. Smear Preparation. Multiple buffy-coat and thin and thick smears are immediately prepared on receipt of the blood specimen using standard procedures outlined elsewhere Buffy-coat smears concentrate the parasites infecting enlarged RBCs into the top layer of RBCs and eliminate the majority of platelets, which are a major source of false-positive smears. 10 The buffy-coat smears are stained using a rapid Wright-Giemsa or Camco Stat- Quick stain in order to rapidly confirm a diagnosis of malaria in all symptomatic patients. 11 If the slides are positive for malaria, then another 2 thin and 2 buffy-coat smears are stained with the Giemsa stain and used for identifying the type of malarial infection. Quantitation (malarial count) is only performed from the Giemsa-stained thin smears using standard procedures Reading of Smears. A group of highly trained microbiology technologists performs and reads all malaria smears. Although an expert technologist is available each day to supervise the reading of malaria smears, a larger group of technologists routinely screen all malaria smears. As a minimum, 1 technologist reads 1 thin and 1 buffy-coat slide before calling preliminary results as negative. Technologists spend a minimum of 10 to 15 minutes reading each of these primary slides. All observations made in reading a malaria smear are recorded on a specially designed worksheet. All areas of the slide where RBCs are not as heavy are scanned for malarial parasites, and areas of the slide where RBCs are not distorted are used to identify the type of malarial infection. For malarial counts, areas where the RBCs do not overlap and are not distorted are used. A different technologist reads the thick smear. If P ovale or P malariae infection is suspected by the primary technologists, an expert technologist always reviews the smears, owing to the low numbers of these types of infections found. Other blood parasites that may be incidentally found when reading malaria smears (ie, microfilaria, Trypanosoma species, Babesia species) are also reported. In addition, one of the expert malaria technologists checks all positive malaria smears before a final report is issued. The medical microbiologist on call is immediately informed of all positive malaria results by the technologist reading the smears. The medical microbiologist also reviews all positive malaria smears the same day. 5. Reporting of Results. All malaria specimen results and malaria identifications are phoned to the attending physician as soon as the preliminary slides are read. The results are also entered into the laboratory information system (ie, PathNet 3.0 Classic) to generate a preliminary report. The Division of Microbiology routinely tracks the analytic turnaround time of malaria tests and expects that a minimum of 90% of reports will be finalized within 24 hours of receipt (ie, it takes a minimum of 4 6 hours to dry the thick smears before staining/reading). Interpretive comments are routinely placed on all positive reports that instruct physicians to repeat malaria smear results until the patient tests negative for parasites. For P falciparum infections, it is recommended that malaria smears are repeated every 8 hours for the first 48 hours and then daily until the patient s parasitemia clears. 9,10 Plasmodium infections caused by other species should have malaria smears repeated daily until the patient is free of parasites. 9,10 A final report is not generated until the thick smears have been read. An expert technologist reviews all positive malaria specimen results. If the patient has been hospitalized, a copy of the final report for all positive malaria cases is faxed to the Acute Care Regional Infection Prevention & Control Office, Cal- 688 Arch Pathol Lab Med Vol 127, June 2003 Regional Malaria Diagnostic Service Church et al

3 gary Health Region. A copy of all final positive malaria reports is sent to the Medical Officer of Health, Calgary Health Region. 6. Medical Microbiologist Consultation. All patients with confirmed malaria are monitored by the medical microbiologist on call. The microbiologist on call contacts the attending physician as part of the laboratory s clinical consultation to determine if the patient is receiving appropriate treatment or has been referred to the Infectious Diseases Service for immediate management. Quality Assurance Procedures Calgary Laboratory Services has designed a competency program to ensure the maintenance of a high level of proficiency in the technologists reading malaria smears. The following components form the key elements of the malaria technologists training program. 1. Technologist Training. All technologists assigned to the malaria diagnostic service undergo extensive training prior to being placed on the expert and screening malaria technologist lists. 2. Proficiency Testing. Calgary Laboratory Services subscribes to the College of American Pathologists (CAP) surveys that include blood parasite identification challenges. Both the Parasitology (P-A) and the Blood Parasite Survey (P-1) from the CAP are performed. The CLS microbiology laboratory receives 4 surveys per year, each containing 5 blood parasite slide challenges in the P-1 survey, and at least 1 of the challenges in the P-A survey is a malaria smear. Survey slides are handled in the same manner as any other malaria smear and are read by the technologist(s) assigned to the bench whenever proficiency-testing slides are received. A variety of positive malaria smears are also reviewed as part of the CLS malaria reading competency procedure. Routine (screening) technologists must read 1 positive malaria specimen every 2 months chosen from the following sources: (1) a positive malaria case in which they were the first technologist to find and identify the organism, (2) CAP survey slides, or (3) in-house unknown malaria smears that are set up every 2 to 3 months. Expert technologists must read more malaria smears to maintain their expert status. They must read a positive malaria case a minimum of once a month from any of the 3 sources listed and must independently read the entire CAP survey specimen set. Technologists assigned to the malaria bench each day must scan 1 unknown thick smear per week before reading the clinical cases. These smears are generated for the malaria bench each week, and each technologist must document their results on the CLS Thick Smear QC form. Comparison to Polymerase Chain Reaction A portion of the blood sample from all positive malaria cases is referred to a reference laboratory within the Tropical Disease Unit, The Toronto Hospital, University of Toronto (K. Kain, MD) for confirmation of malaria detection and species identification using a polymerase chain reaction (PCR) assay as previously described. 4,12,13 The PCR result provides an additional measure of the quality of service being provided, and in particular, documents that the technologists are not missing dual infections. Data Analysis Data were entered into a Microsoft Excel worksheet (Version 6.0) and analyzed using standard statistical methods to calculate confidence intervals (CI), standard deviation around the mean (SEM), and probability (P value). A P value less than.05 was considered to be statistically significant. Test performance was calculated using standard methods to determine the sensitivity and positive predictive value of microscopy compared to PCR. Specificity and the negative predictive value could not be calculated because the true negatives were not determined because blood specimens that were negative for Plasmodium by microscopy were not confirmed by PCR testing. RESULTS Epidemiology of Malaria in Calgary A total of 95 positive malaria tests were reported from 52 new cases of infection in the Calgary Health Region from May 1999 through April Of the 52 new cases of malaria, 26 (50%) were P falciparum infections, 22 (42.3%) were P vivax infections, 3 (5.8%) were P ovale infections, and 1 (1.9%) was a P malariae infection. No mixed-species malarial infections were diagnosed during the study period. Sixty-five percent of all new malaria cases were diagnosed during the fall and winter months. Most of the malaria cases (44 [84.6%]) occurred in Calgarians who had recently traveled to a malaria-endemic region. Of these 44 travelers, 23 (52.3%) had gone to Africa and 10 (22.7%) to Southeast Asia, with the remainder returning from the Middle East, South America, and India or Pakistan. Another 7 (13.5%) of the 52 new cases were diagnosed in people who had recently immigrated to Calgary from a malaria-endemic region, and 1 patient was visiting from Uganda and came down with P vivax malaria. All malaria cases during the study period occurred in adults with the following typical symptoms: fever in 50 patients (96.2%); chills in 49 (94.2%); headache in 26 (50%); myalgia in 22 (42.3%); and nausea, vomiting, or malaise in 23 (44.2%). No differences occurred in the age of men versus women acquiring infection. The mean age ( SEM) for men was years (range, years; 99% CI 33.8, 46.9), while the mean age ( SEM) for women was years (range, years; 99% CI 25.3, 54.4). There were 36 (69.2%) infections in men and only 16 (30.8%) in women, yielding a 2:1 predominance in males. Plasmodium falciparum infections, in particular, occurred more commonly in men (84.6%; 22/26) than in women (15.4%; 4/26) (P.001). Nineteen (36.5%) of the 52 patients had had a prior episode of malaria either in the distant past (ie, several years beforehand) or recently during travel and had been given antimalarial treatment overseas. Only 6 (11.5%) of the patients with malaria reported taking prophylaxis despite travel to a malaria-endemic region, and of those 6, 3 (50%) took an inadequate dose and or duration of their prescribed antimalarial therapy (ie, mefloquine). Time to Clinical Diagnosis The time between the patient s return to Calgary from travel abroad and the onset of symptoms was highly variable and dependent on the type of malarial infection acquired. On average, symptoms of malaria presented within 7.9 SEM 9.9 days (mean SEM; range, 1 39 days; 99% CI 3.2, 12.6) of returning from a malaria-endemic region in travelers. Longer incubation times occurred in patients who had recently taken antimalarial drugs and/ or who had had a previous episode of infection. However, 7 cases of malaria presented 3 or more months after returning to Canada; 1 case was likely a reactivation of prior P ovale infection, and the rest were reactivation of prior P vivax infections. Patients frequently delayed seeking medical attention for several days, despite the onset of significant clinical symptoms. On average, days (mean SEM; range, 1 12 days; 99 CI 2.6, 4.8) lapsed between the time patients initially started having symptoms and were as- Arch Pathol Lab Med Vol 127, June 2003 Regional Malaria Diagnostic Service Church et al 689

4 sessed by a physician. Delays in seeking medical attention occurred regardless of the type of malarial infection. Patients were initially assessed at various health care locations in the region. Twenty-four (46.2%) of the patients presented to the emergency department at an acute-care hospital, and of these, 6 required admission to hospital; 2 cases of P falciparum were admitted directly to the intensive care unit because of life-threatening infection. Another 18 (34.6%) patients were seen at a community physician s office or outpatient clinic, and only 7 (13.5%) were initially assessed by a clinic specializing in travel medicine. Most patients were not receiving antimalarial treatment at the time that the first blood smears for malaria were ordered by the attending physician. Only 9 (17.3%) of the 52 patients with laboratory-confirmed malaria had had recent treatment or had empiric treatment started at the time of the initial blood smear. Some of these patients also had been prescribed the wrong medication for the type of malarial infection subsequently diagnosed (ie, chloroquine was prescribed for empiric antimalarial treatment for a P falciparum infection). Monitoring Laboratory Performance The performance of the centralized regional malaria diagnostic service has been continuously monitored against expected quality indicators over the past 2 years (March 1999 through April 2002). Figure 1 outlines the transportation time (ie, preanalytic turnaround time) of malaria specimens from January 2000 forward, according to the location of specimen collection (ie, hospital RRLs vs PSCs in the community). There has been a steady improvement in the efficiency of transport of malaria specimens from each hospital. During the last year (Figure 1, C), more than 90% of malaria specimens have consistently arrived from the hospital RRLs within 2 hours of collection, and almost all specimens arrived within 3 hours of collection. Delays in transportation of hospital specimens were mainly due to delayed transport of the blood tube from the ward to the RRL. More than 90% of malaria specimens from community locations have consistently arrived in the central laboratory within 2 hours of collection during the past 2 years, and almost all specimens arrived within 3 hours of collection. Delays in transportation of community specimens were mainly due to delayed transport of the blood tube from the physician s office to the central laboratory. All specimens drawn at CLS PSCs in the community arrived at the central laboratory within the expected time. Preliminary results of thin smears are reported as soon as the slides have been read, and on average this result is phoned to the attending physician within hours (mean SEM; 99% CI 0.5, 5.2) of receipt of the blood sample. Figure 2 shows the average analytic turnaround time for a positive and negative finalized malaria test result during the past 2 years. On average, it took hours (mean SEM; 99% CI 5.2, 27.8) to finalize a positive malaria test and hours (mean SEM; 99% CI 19.1, 23.9) to finalize a negative result. More than 90% of malaria smear reports have been consistently finalized within 24 hours, and all tests have been reported within 30 hours of receipt of a blood specimen. The malaria expert technologists have accurately read and reported all of the blood malaria challenges on the CAP Blood Parasite (P-1) and Parasitology (P-A) surveys during the study period. Performance of the expert microbiology technologists in accurately diagnosing malarial infection using standard microscopy methods has also been routinely compared with a reference PCR method. Comparison of the results between microscopy and PCR for the 3 patient samples in which there was a discrepancy demonstrated the following: (a) In 2000, PCR did not detect 1 patient with typical morphologic features of P ovale infection on smear. (b) In 2001, PCR testing detected a P ovale infection in a patient who had been diagnosed with P vivax on smear. Review of the smear did not demonstrate typical morphologic features of P ovale infection, although there can be considerable overlaps in the appearance of these 2 species. Finally (c), in 2002, a patient who had been recently treated with a full course of halofantrine had persistently negative malaria smears, but subsequent PCR testing of these blood samples was positive for P falciparum infection. Although PCR was more sensitive in detecting the presence of malarial infection (100%) compared to microscopy (98.1%), the expert microscopists could more accurately distinguish the specific species of malaria causing infection, with sensitivity and positive predictive values of 98.1% and 100%, respectively, compared to 96.2% and 98.1%, respectively, for PCR. In particular, the current PCR method used by the reference laboratory had difficulty detecting some P ovale infections. Despite interpretive comments being reported on every malaria test, physicians do not consistently order followup smears as recommended. Sixty-eight (71.6%) blood specimens were submitted from the 24 patients with P falciparum infection. All of these patients, therefore, had between 1 and 3 repeat blood specimens submitted to the laboratory after therapy was started. Twenty-two (23.2%) blood specimens were submitted from the 13 patients with P vivax infection, with only 9 (69.2%) having a repeat malaria smear after therapy was started. Patients with P ovale infection had a repeat malaria blood sample submitted, but no follow-up testing was done on the patient with P malariae infection. COMMENT Since Calgary is the major urban business center for the petroleum industry in Canada, many people travel abroad for their work or can afford to vacation abroad, so that malarial infection is not uncommon despite this being an area of nonendemicity. On average, the laboratory has confirmed 1 to 2 cases each month during the study period, mainly in adults of working age. Although all types of Plasmodium species infections have been found, there has been an increased trend over the previous 3 years toward diagnosing more P falciparum infections. In the past year, P falciparum was by far the most common type of malarial infection found, and most of these cases occurred in men. A potential explanation for this observation is the predominance of men in Calgary who work in the petroleum industry as long-term overseas workers, particularly in parts of Africa. Men may also have a potentially greater risk for acquiring malaria because they do not seeking pretravel care and/or do not use protective measures or take prophylaxis; however, further study of these factors is required. Our study confirms the remarkable consistency in the symptoms and duration of illness in patients before seeking care. 1,14 Although most patients had onset of typical symptoms of malaria (ie, fever, chills) within a week of 690 Arch Pathol Lab Med Vol 127, June 2003 Regional Malaria Diagnostic Service Church et al

5 Figure 1. A through C, Preanalytic turnaround time (TAT) for malaria blood specimens in the Calgary Health Region. Asterisk indicates the transportation time between blood draw and processing of the specimen by the Calgary Laboratory Services centralized microbiology laboratory. Arch Pathol Lab Med Vol 127, June 2003 Regional Malaria Diagnostic Service Church et al 691

6 Figure 2. Average analytic turnaround time (TAT) for malaria thin and thick smears in the Calgary Health Region. Asterisk indicates the time taken by the Calgary Laboratory Services centralized microbiology laboratory to read malaria thin and thick smear results and to provide a report to the physician. Target 90% of specimens reported within 24 hours of receipt. returning from travel abroad, several days lapsed before they sought medical attention at either the emergency department, their physician s office, or the travel clinic. In general, longer incubation times for primary malarial infections occurred in patients who had recently taken antimalarials and/or had had a previous episode of infection. Also, patients with reactivation of either P ovale or P vivax infections presented several months after travel abroad or immigration from a malaria-endemic area. Because of the delay in presentation of patients with potentially life-threatening malarial infections presenting for care in our region, access to timely and accurate laboratory diagnostic services is paramount in expediting appropriate therapy. Malaria test turnaround time and the accuracy of microscopic diagnosis performed in laboratories without specialized tropical medicine training has been previously reported to be below an acceptable standard. 4,15 In Toronto, where there is a decentralized model for malaria diagnostic services and where laboratory medical staff are responsible for reading and reporting smear results, Kain et al 4 documented that 45% of laboratories did not report an accurate malaria species identification, and it took a mean of 7.6 days after a patient presented to a physician before treatment was started. The transportation of blood smears for malaria to the laboratory is also not being routinely monitored. In Calgary, a similar laboratory service model existed prior to restructure in the mid-1990s. The Foothills Medical Center, the main academic teaching hospital, was the only site at which an expert service had been put in place for the rapid diagnosis of malarial infection. In our prior experience working in a decentralized laboratory system, it was difficult to justify implementation of standardized processes at all sites in the region to ensure the prompt transport and reading of malaria smears, despite the potential for a less than ideal patient outcome if an inaccurate or delayed diagnosis of malarial infection occurred. 1,2 Regional standardization was not only impeded by the independence of hospital and community laboratories, but also by the inordinate expense of implementing an efficient courier system for such a low volume and relatively esoteric test like malaria. However, restructure of laboratory services into CLS provided us with a unique opportunity to expand the expert technologist malaria service at the Foothills Medical Center to become regional in scope as part of consolidating all clinical microbiology testing in Calgary into a central laboratory located in the community. Calgary Laboratory Services routinely transports thousands of patient s laboratory specimens for other types of tests (ie, high-volume chemistry, hematology, and microbiology) from hospitals and community sites to the central laboratory each day that also have an urgent clinical priority. It is therefore cost-effective within our regional laboratory transportation system to handle blood specimens for malaria within the standards described herein. Since the CLS central microbiology laboratory is already performing a wide variety of other types of tests 24 hours a day, 7 days a week, it is feasible for us to read and report malaria smears without incurring additional labor costs. Continuous monitoring of the performance of this service shows that current malaria diagnostic laboratory standards have consistently been met or exceeded since appropriate processes and procedures were implemented 3 years ago by CLS. 1,2 Blood specimens for malaria diagnosis are routinely transported to the centralized laboratory within 2 hours after collection, regardless of whether the initial patient assessment occurs in a physician s office in the community or at one of the hospitals. Analytic performance of the expert group of microbiology technologists to rapidly and accurately provide a microscopic diagnosis of malaria to the species level is excellent compared with molecular identification by the reference laboratory. Although previous studies have shown that PCR is better than expert microscopy at detecting mixed infections, 4,12,13 this possibility could not be evaluated during the study period since no infections occurred with more than 1 Plasmodium species. Preliminary results of thin smears are routinely reported directly to the attending physician within 2 to 3 hours of receipt of a blood specimen, regardless of the time of day, and the vast majority of final malaria reports are available within 24 hours. Although the time between reporting of a preliminary malaria test result and treatment of patients was not documented, expert management advice is immediately provided by the medical microbiologist on call, who contacts the patient s attending physician to determine if appropriate treatment has been started. The use of interpretive comments on all malaria test reports that outline when repeat smears should be done also improves the laboratory follow-up of patients with malarial infection. However, by providing reminders to practitioners either by telephone or through electronic mail messages, the laboratory could further enhance compliance of physicians in ordering follow-up malaria smears. Patient compliance in actually having repeat malaria smears done, particularly when they become asymptomatic, also needs to be studied. Our experience in providing a quality malaria diagnostic service to the Calgary region documents the critical role that the clinical microbiology laboratory plays in expediting appropriate care of patients once infection is suspected and confirmed. Laboratories performing malaria tests must put in place processes and procedures that ensure that an accurate and timely diagnosis of the type of malaria is routinely available. 692 Arch Pathol Lab Med Vol 127, June 2003 Regional Malaria Diagnostic Service Church et al

7 The authors thank the reference laboratory within the Tropical Diseases Unit, The Toronto Hospital, University of Toronto, Toronto, Ontario, for performing polymerase chain reaction tests for malaria identification on all microscopically positive samples. References 1. Committee to Advise on Tropical Medicine and Travel (CATMAT), Laboratory for Disease Control. Canadian recommendations for the prevention and treatment of malaria among international travelers. Can Commun Dis Rep. 2000;26(suppl 2): World Health Organization Expert Committee on Malaria. Twentieth Report. Geneva, Switzerland: World Health Organization; 2000: Centers for Disease Control and Prevention. Summary of notifiable diseases United States. MMWR Morb Mortal Wkly Rep. 2000;49: Humar A, Sharma S, Zoutman D, Kain KC. Fatal falciparum malaria in Canadian travelers. Can Med Assoc J. 1997;156: Kain KC, Harrington MA, Tennyson S, Keystone JS. Imported malaria: prospective analysis of problems in diagnosis and management. Clin Infect Dis. 1998;27: Murphy GS, Basri H, Purnomo, et al. Vivax malaria resistant to treatment and prophylaxis with chloroquine. Lancet. 1993;341: Kain KC. Chemotherapy for drug-resistant malaria. Can J Infect Dis. 1996; 7: Church DL, Hall P. Centralization of a regional clinical microbiology service: the Calgary experience. Can J Infect Dis. 1999;10: Garcia LS. Malaria and babesiosis. In: Diagnostic Medical Parasitology. 4th ed. Washington, DC: American Society for Microbiology Press; 2001: Ontario Medical Association. Laboratory proficiency testing program: recommendations for examination of blood films for malaria parasites. Toronto, Ontario: Ontario Medical Association; 1996;3: Anthony RL, Bangs MJ, Anthony JM, Purnomo. On-site diagnosis of Plasmodium falciparum, P. vivax and P. malariae by using the Quantitative Buffy Coat system. J Parasitol. 1992;78: Brown AE, Kain KC, Pipithkul J, Webster HK. Demonstration by the polymerase chain reaction of mixed Plasmodium falciparum and P. vivax infection undetected by conventional microscopy. Trans R Soc Trop Med Hyg. 1992;86: Sethabutr O, Brown AE, Panyim S, Kain KC, Webster HK, Echeverria P. Detection of Plasmodium falciparum by polymerase chain reaction in a field study. J Infect Dis. 1992;166: Svenson JE, MacLean JD, Gyorkos TW, Keystone J. Imported malaria: clinical presentation and examination of symptomatic travelers. Arch Intern Med. 1995;155: Milne LM, Kyi MS, Chiodini PL. Accuracy of routine laboratory diagnosis of malaria in the United Kingdom. J Clin Pathol. 1994;47: Arch Pathol Lab Med Vol 127, June 2003 Regional Malaria Diagnostic Service Church et al 693