Incidence of Malaria in Kuala Lumpur, Malaysia: A Single-In- stitution Retrospective Review from 2005 to 2020

While there has been a tremendous decline in malaria disease burden in the remote parts of the Malaysia, little is known about malaria incidence in its urban localities. This study aimed to analyse trends of malaria cases in urban Kuala Lumpur, Malaysia. All suspected cases presented to a university hospital in Kuala Lumpur from January 2005 to December 2020 were examined by microscopy. Infection status was analysed using descriptive statistics and curve estimation analysis. Of 3105 blood films examined, 92 (3%) were microscopically confirmed malaria cases. Plasmodium vivax infections accounted for the majority (36.9%) of all malaria cases. Nearly half (47.8%) of cases were found among foreign cases (P<0.001). The majority of foreign cases were males (86.4%) and came from Southeast Asian countries (65.9%). Curve estimation analysis showed significant decreases of malaria cases due to P. vivax (R2 = 0.598; P<0.001) and Plasmodium falciparum (R2 = 0.0259, P = 0.029), but increase for Plasmodium knowlesi (R2 = 0.325, P = 0.021) during the 16 years. This study revealed that malaria incidence in urban Kuala Lumpur is low and has remained stable since 2005. However, P. knowlesi played a significant role in the increase of overall malaria in the area, highlighting the importance of continued vigilance and improved surveillance.


Introduction
In the Western Pacific region categorised by the World Health Organization (WHO), there are 753 million people in 10 countries that are currently at risk of infections with malaria [1]. Malaysia, which is included in this region, is in the pre-elimination phase and continues to progress towards elimination, reporting only 85 cases of indigenous human malaria cases in 2017 [1]. Even though malaria control activities have significantly reduced human malaria incidence in Malaysia, the resurgence of the monkey malaria parasite Plasmodium knowlesi remains a main public health problem in the less developed areas of the country, especially in Malaysia Borneo [2][3][4] and among hard-to-reach populations of indigenous people (i.e. Orang Asli) in Peninsular Malaysia [5][6][7][8]. About one-third (32%) of total malaria cases occur in Peninsular Malaysia, and the majority of these are found in the central, south-eastern and northern coastal regions [9]. The remaining 68 percent of cases are found in Malaysian Borneo, primarily in the states of Sabah and Sarawak [10].
Malaysia reoriented its intent from malaria control to elimination in 2011, with a phased goal of achieving zero local transmission in Peninsular Malaysia by 2015, and in Sabah and Sarawak by 2020. Malaysia is vulnerable to malaria importation, primarily from Indonesian and Filipino migrant workers seeking employment in Malaysia's growing economy [10,11]. In addition, many documented and undocumented migrants from Myanmar, Bangladesh, Nepal, Indonesia and Thailand also enter Peninsular Malaysia to serve the low-skilled and semi-skilled sectors of the economy, especially in the urban areas. In 2014, imported cases accounted for 20 percent of all cases in Malaysia [12]. In countries approaching elimination, imported cases tend to make up most of the recorded cases and threaten the re-establishment of malaria transmission in receptive areas [13]. As the country is dependent on foreign labour that comes from regional countries, the malaria elimination goal in Malaysia may be at risk.
Despite the significant decline in malaria disease burden in remote parts of Malaysia, the overall trend of malaria incidence based on passive case detection is not well-documented in urban localities. To our knowledge, there were few published data available on passive case detection of malaria in Malaysia since 2003. Epidemiological data such as trends of malaria positivity rates at public institutions and hospitals are essential to design appropriate interventions. Therefore, this study aims to describe the more recent epidemiological and trend of malaria cases diagnosed in the tertiary care referral and teaching hospital of Kuala Lumpur, Malaysia. As the hospital is located in the capital city, it serves as a proxy measure for the trend of malaria in the urban area which may contribute to evidence-based decisions on malaria control activities.

Study Area
This study was conducted at Hospital Canselor Tunku Muhriz (HCTM), a tertiary care referral and teaching hospital of the National University of Malaysia (UKM). The hospital is a major medical centre located in the capital city of Kuala Lumpur, Malaysia.

Study Design
This retrospective laboratory record review study was carried out to determine 16 years (January 2005 -December 2020) malaria cases. This study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Research and Ethics Committee of UKM (Reference no: JEP-2018-055). The need for informed consent was waived by the committee considering the retrospective nature of the study.

Data Collection
The study participants were all malaria suspected individuals who had a complaint of febrile illness at HCTM during the study period. Sociodemographic and laboratory data regarding malaria were extracted from the electronic-based reporting system of HCTM and yearly laboratory log-books from the Department of Parasitology and Medical Entomology, Faculty of Medicine in UKM. Malaria diagnosis was based on microscopic examination of Giemsa-stained thick and thin blood smears. In HCTM, microscopic examination is the gold standard diagnostic method for detection and species identification of Plasmodium parasites. Peripheral smear examination of a well-prepared and well-stained thick and thin blood films were used to diagnose malaria in the laboratory. The hospital strictly follows Malaysia's standard operating procedures in all phases of the quality control for capillary blood sample collection, smear preparation, staining and blood film examination for malaria parasite detection. Blood films were fixed and stained with 3% Giemsa stain and examined under oil emersion (10x100 magnification) by trained microscopists. Blood films were defined as negative if no parasites were found after examining 100 high-power microscopy fields. For all positive samples, malaria species were identified and asexual parasite forms were counted against 500 leukocytes. Parasite density was estimated from parasite counts, assuming that there were 8,000 leukocytes per microliter (µL) of blood. For the remaining blood sample of positive cases, multiple blood smears were made and kept for educational purposes and future research.

Statistical Analysis
All the data was merged, cleaned and cross-checked using a Microsoft Excel spreadsheet. The data were analysed using STATA/SE version 13.1 (StataCorp, Texas, USA) and GraphPad Prism version 5.03 (GraphPad Software Inc., California, USA). A descriptive analysis was performed in order to assess yearly distribution, gender, age group, ethnicity and nationality. The Pearson's Chi-square test, Fisher's exact test and Kruskall-Wallis test were used to describe the association of variables. Curve estimation analysis was used to evaluate the trends of the data. A P<0.05 was considered statistically significant.

Results
Over a period of 16 years (2005 -2020), 3105 blood films were requested for malaria diagnosis at HCTM, of which 92 (3%; 95% confidence interval [CI]: 2.4 -3.6]) were microscopically confirmed malaria cases ( Table 1). The median age of the malaria cases was 30 (interquartile range [IQR]: 25 -39) years. The majority (71.7%, 95% CI: 61.4 -80.6) of the infected were young adults (18 -40 years) and the age distribution differed significantly between the infected and non-infected groups (P = 0.007). Individuals of Bumiputera ethnicity accounted for most of the malaria cases, but there was no significant difference in ethnic distributions between the infected and the non-infected groups. When compared to the non-infected group, malaria infections were significantly more common in males (P<0.001) and among Malaysian (P<0.001). The trend of malaria cases is summarized in Table 2. Malaria cases were reported in all years except 2015, with the highest prevalence reported in 2018 (10.3%; 3/29). Overall, Table 2. Trend of malaria incidence reported at University Hospital of the National University of Malaysia (UKM) in 2005-2020  Total number tested, N  3105  276  256  214  281  305  255  277  245  230  166  168  160  135  29  18  90 No. of positive malaria cases, n (%)      19 of all malaria cases, respectively. No Plasmodium ovale infections were observed. The 11 20 cases of mixed-Plasmodium spp. infections were P. falciparum/P. vivax (n = 3), P. falcipa-21 rum/P. malariae (n = 3), P. vivax/P. malariae (n = 1), P. falciparum/P. knowlesi (n = 1), P. malar-22 iae/P. knowlesi (n = 2) and P. falciparum/P. malariae/P. knowlesi (n = 1). Plasmodium knowlesi 23 cases were first detected in 2010 and peaked in 2020, accounting for the majority of the 24 cases that year. 25 Despite the apparent fluctuation over the 16-year period (Table 2), no significant dif-26 ference was observed between years (P = 0.082). Similarly, the proportion of malaria cases 27 from local and foreign patients was not significantly different between years (P = 0.096), 28 with foreign cases contributed to essentially half (47.8%; 44/92) of all positive cases. Inter-29 estingly, out of 44 foreign cases, 38 (86.4%) were males and 29 (65.9%) came from South-30 east Asian countries. There were a total of 3 cases contributed solely by foreign patients 31 in 2014 and a total of 10 cases solely by local patients between 2017 and 2020. 32 Albeit not significant, curve estimation analysis using linear models showed a slight 33 increment of overall malaria positivity rates as well as in local cases from 2005 to 2020 34 (Figure 1a). In contrast, a significant reduction was observed in foreign cases from 2005 to 35 2020 (R 2 = 0.313, P = 0.024). With regards to Plasmodium species (Figure 1b), interestingly, 36 significant reductions were observed for malaria due to P. vivax (R 2 = 0.598; P<0.001) and 37 P. falciparum (R 2 = 0.0259, P = 0.029), but not for P. knowlesi (R 2 = 0.325, P = 0.021) during 38 the 16 years. As far as these models were used, P. malariae and mixed-infections did not 39 show a significant increase across the years.

44
Malaysia aims to achieve malaria elimination by the year 2020, and indeed the drop 45 (98.4%) between 2010 and 2017 from a total of 4731 to 77 recorded indigenous cases for P. 46 vivax and P. falciparum is highly promising [1]. However, the drop in the number of im-47 ported cases over the same period were less encouraging from a total of 831 to 423 rec-48 orded cases [1] and detected zoonotic P. knowlesi infections in remote parts of Malaysia 49 have steadily increased [4,[14][15][16]. In this study, we examined the available record data 50 from a referral and teaching hospital of UKM located in the capital Kuala Lumpur, Pen-51 insular Malaysia from 2005 to 2020. Over the 16 years, the number of malaria cases diag-52 nosed annually at our hospital has remained relatively low and stable. The present study 53 also revealed that the overall slide positivity rate of malaria was low (i.e. 3%), but high 54 among males and non-Malaysian. This is similar to that of studies conducted in the same 55 setting in 2003 [17,18]. Malaria species-specific data showed that P. vivax was the most 56 dominant species particularly in foreign cases with the estimated incidence showed a sig-57 nificant reduction over time. Moreover, the emergence of P. knowlesi infections in 2010 58 among local cases signify the alarming threat of zoonotic malaria in the country and may 59 hinder malaria elimination efforts. 60 As one of the fastest-growing cities in Asia, where thousands of foreign workers ar-61 rived every year, Kuala Lumpur represents a likely hotspot for malaria importation in 62 Malaysia. Our hospital in Kuala Lumpur received 44 (47.8%) confirmed malaria cases 63 from foreigners over the period of 16 years (January 2005 to December 2020) of which 64 86.4% were males and 65.9% from neighbouring Southeast Asia countries. This finding 65 was in line with the studies conducted in other Asian countries that highly rely on foreign 66 workers such as Singapore [19,20] [24], Qatar [25], and United Arab Emirates [26]. In 2017, imported cases in Malaysia ac-68 counted for 10.3% of all cases in the country [1]. There were more than 1.8 million regis-69 tered migrants/foreign workers working in Malaysia in 2020. These migrants/foreign 70 workers come from 12 different countries in Asia, and about 1.4 million of them were 71 males (Department of Labour Peninsular, Ministry of Human Resources Malaysia, 2020). 72 Rapid development in the city has led to an influx of low-and semi-skilled foreign work-73 ers and many of whom have come illegally or without work permits. In addition, there 74 are significant numbers of displaced people in Kuala Lumpur with no nationality that 75 arrive from malaria-endemic countries in Asia, particularly from Myanmar. Our finding 76 also revealed that 16 out of 29 confirmed malaria cases from Southeast Asia were from 77 Myanmars patients. As of January 2021, of the approximately 164,620 refugees and asy-78 lum-seekers registered with the United Nation High Commissioner for Refugees (UN-79 HCR) in Malaysia, 86.5% were from Myanmar, 67% were males, and 16.8% of them had 80 resettled in Kuala Lumpur [18]. As Malaysia moves toward elimination, malaria will 81 begin to cluster among certain high-risk groups, including migrants and displaced popu-82 lations. Improved surveillance, collaboration with key industries and other government 83 agencies, and cross-border cooperation with neighbouring endemic countries are critical 84 for addressing the ongoing threat of malaria importation and to achieve elimination. 85 In the present study, P. vivax was the most prevalent Plasmodium species detected, 86 similar to what was found in the retrospective studies conducted in Peninsular Malaysia 87 [7,17,[27][28][29][30]. Plasmodium vivax has been the main cause of human malaria in Malaysia for 88 the past 10 years and remains a health concern today [1,28]. In 2010, of the 5819 reported 89 cases, 58.2% were due to P. vivax [1]. The ability of P. vivax to remain dormant in the liver 90 as hypnozoites that can cause relapse following a primary infection, greater asymptomatic 91 asexual carriage, and early gametocyte production provide far greater challenges for ma-92 laria elimination in the country. Nevertheless, no case of P. vivax was recorded in our hos-93 pital from 2015 to 2020 and based on the curve estimation analysis model, there was a 94 significant reduction of P. vivax cases over the 16 years. This declining trend is a testament 95 to the commitment of the government and other parties in Malaysia. The Malaysian Gov-96 ernment launched the National Malaria Elimination Strategic Plan 2011-2020 with the ul-97 timate goal of stopping locally-acquired malaria (except P. knowlesi) in Peninsular Malay-98 sia by 2015 and in East Malaysia by 2020 [10]. The national strategic malaria elimination 99 plan currently outlines seven key actions to achieve the elimination goal, including 100 strengthening malaria surveillance system through an online system, intensifying control 101 activities by indoor residual spray (IRS) and insecticide-treated nets (ITN), ensuring early 102 case investigation, prompt treatment and management of outbreaks as well as enhancing 103 community awareness and knowledge of malaria [31]. All these efforts have resulted in a 104 significant reduction in overall malaria incidence in general and P. vivax cases in particular 105 over the last decade. 106 Our work has provided insight into P. knowlesi cases in an urban area. Although the 107 greatest number of P. knowlesi cases has been reported in remote areas in East Malaysia 108 [2, 9,12,32], the infection is also the predominant cause of malaria in Peninsular Malaysia 109 [6,8]. It is unlikely that patients admitted to our hospital, acquired the P. knowlesi infection 110 in the capital Kuala Lumpur (Federal Territory), which is considered a malaria-free area. 111 However, it is interesting to note that Kuala Lumpur is located within the State of Selan-112 gor, a malaria-endemic area in Peninsular Malaysia. In Selangor, local malaria transmis-113 sion is still being reported from a few districts that adjoin sub-urban and forest range areas 114 with rapid development and deforestation [30,33]. A similar link of deforestation and P. 115 knowlesi malaria transmission has been observed recently in the State of Sabah, East Ma-116 laysia [34][35][36]. Except for a comprehensive case-control study by Grigg et al. on individual-117 level risk factors of acquiring P. knowlesi in villages in East Malaysia [37], no study has 118 been conducted in Peninsular Malaysia. Furthermore, human behavioural factors may 119 also be associated with acquiring knowlesi malaria. Activities appealing to urban popula-120 tions such as jungle tracking, camping as well as waterfall picnic and fishing may increase 121 exposure to environmental factors conducive to zoonotic transmission of knowlesi ma-122 laria. More detailed evidence about the risk of transmission in urban settings is required 123 to design appropriate interventions. 124

125
Collectively, the malaria positivity rate in the study area is low and declining. The 126 declining trend of the overall rate could be due to the significant decline of human malaria 127 cases, particularly due to P. vivax and P. falciparum infections. However, malaria cases re-128 main a public health concern in the urban setting with the influx of migrant workers and 129 the increasing number of cases with P. knowlesi infections. Therefore, malaria case notifi-130 cation and interventions in Malaysia should be strengthened and reinforced to achieve 131 elimination. Furthermore, improved surveillance of malaria clusters among certain high-132 risk groups displaced populations and migrant workers will require mutual collaboration 133 with key industries and other government industries and cross-border cooperation with 134 neighbouring endemic countries.