Plasmodium malariae infections as a cause of febrile disease in an area of high Plasmodium falciparum transmission intensity in Eastern Uganda

Plasmodium falciparum is responsible for the vast majority of (severe) clinical malaria cases in most African settings. Other Plasmodium species often go undiagnosed but may still have clinical consequences. Here, five cases of Plasmodium malariae infections from Eastern Uganda (aged 2–39 years) are presented. These infections were all initially mistaken for P. falciparum, but Plasmodium schizonts (up to 2080/µL) were identified by microscopy. Clinical signs included history of fever and mild anaemia. These findings highlight the importance of considering non-falciparum species as the cause of clinical malaria. In areas of intense P. falciparum transmission, where rapid diagnostic tests that detect only P. falciparum antigens are commonly used, non-falciparum malaria cases may be missed.


Background
In 2019, an estimated 229 million cases and 409,000 deaths due to malaria were reported globally [1]. Most of these occurred in sub-Saharan Africa, and were caused by Plasmodium falciparum. Thus, it is not surprising that P. falciparum remains the focus of malaria control programmes and most malaria research in Africa. Other Plasmodium species are endemic in sub-Saharan Africa, including Plasmodium ovale, Plasmodium vivax and Plasmodium malariae [2]. These non-falciparum species also cause clinical symptoms, and lead sometimes to severe consequences. Chronic infections with P. malariae may last for years, and can cause severe complications in approximately 3% of cases, including nephrotic syndrome, splenomegaly and anaemia [3,4]. Moreover, P. malariae infections have been observed following treatment of P. falciparum infections with artemisinin-based combination therapy (ACT) [5,6].
Different Plasmodium species can be differentiated by microscopy but this requires laboratory infrastructure and technical expertise. Infections with P. malariae are typically characterized by low parasitaemia [7] and the presence of rosette schizonts (daisy heads) and other parasite life-stages in the peripheral blood [8]. Plasmodium malariae is often missed or misdiagnosed by microscopy [9,10]. However, polymerase chain reaction (PCR) suggests a non-negligible prevalence of P. malariae infection, either alone or mixed with other species, in population Open Access  [11,12].

Malaria Journal
Rapid diagnostic tests (RDTs), which are simple to use and require minimal training, are increasingly being used to diagnose malaria. In Africa, most RDTs detect histidine-rich protein 2 (HRP2), an antigen that is specific for P. falciparum; use of these HRP2 RDTs is likely to contribute further to under-diagnosis of P. malariae infections.
Here, five cases are presented of symptomatic, uncomplicated P. malariae. These cases were coming from an area of intense P. falciparum transmission in Eastern Uganda and were initially misdiagnosed as P. falciparum by microscopy.

Case presentation
Patients described in this case report presented at either Masafu General Hospital or Nagongera Health Centre IV. Masafu General Hospital is a district referral facility in Busia District, an area in south-eastern Uganda that borders Kenya and Lake Victoria, where transmission of P. falciparum is perennial and intense [13]. Nagongera Health Centre IV in located in Nagongera sub-county in Tororo district. Nagongera sub-county is predominantly a rural setting, with traditionally very high malaria transmission [14]. Between 2019 and 2020 the total Outpatient Department treatment (OPD) attendance in Masafu General Hospital was 40,471, whilst in Nagongera Health Centre IV 18,624 [15].
In both settings, hospital and health centre, malaria diagnosis is routinely performed by HRP-2 based rapid diagnostic test (Standard Q Malaria Pf Ag, SD Biosensor, South Korea). However, as part of a research study on P. falciparum, clinic technicians all received formal microscopy training and thick and thin blood films were used for malaria diagnosis, instead of RDTs. Haemoglobin measurements were performed by Hemocue ™ (haemoglobinometer, Angelholm, Sweden).

Case-patient 1
On January 30th 2020, a 15-year-old male presented to the OPD with history of fever, headache and joint pain for 3 days with no evidence of severe malaria or danger signs. At initial presentation, he had an axillary temperature of 36.8 °C and his physical examination was within normal limits. Malaria parasites were reported on the thick and thin blood film. By thick blood smear, total parasite count was 896 parasites/μL; malaria schizonts were observed at 176 parasites/μL (and included in the total parasite density estimate). Haemoglobin concentration was 12.3 g/dL. The patient was diagnosed with P. falciparum malaria.

Case-patient 2
On February 19th 2020, a 20-year-old female presented to the OPD with history of fever for 3 days and headache, weakness and joint pain for 2 days. Her temperature on admission was 37.2 °C and her physical examination did not show abnormalities. After reading the thick blood film, malaria infection was diagnosed with a total of 2288 parasites/μL (2080 schizonts/μL). Mild anaemia was detected with haemoglobin of 10.8 g/dL. The patient was diagnosed with P. falciparum malaria.

Case-patient 3
On February 27th 2020, a 39-year-old female presented to the OPD with history of fever and joint pain for 2 days and headache, weakness and low back pain for 4 days. Her axillary temperature on admission was 37.0 °C and she was diagnosed with uncomplicated P. falciparum malaria based on a thick blood film with 336 parasites/ μL (272 schizonts/μL). Mild anaemia was detected with haemoglobin of 9.5 g/dL.

Case-patient 4
On March 3rd 2020, a 2-year-old boy presented to the outpatient department with persistent vomiting, anorexia and headache. He had an axillary temperature of 38.3 °C. Microscopy slides were interpreted as P. falciparum/P. malariae mixed infection based on presence of 32 ringstage parasites/μL and 520 schizonts/μL for P. falciparum and one trophozoite "band form" for P. malariae. Haemoglobin was 7.8 g/dL. The patient was diagnosed with uncomplicated mixed species malaria infection.

Case-patient 5
On April 29th 2020, a 14-year-old male with mild symptoms of fever, headache, joint pain and anorexia presented to Nagongera Health Center IV, Tororo district. He had an axillary temperature of 37.4 °C. Malaria was diagnosed by microscopy with 4016 parasites/μL (96 schizonts/μL). Haemoglobin was 11.1 g/dL. The patient was diagnosed with uncomplicated P. falciparum malaria.
All five cases were treated for uncomplicated P. falciparum infection according to national Ugandan treatment guidelines with artemether lumefantrine 20 mg/120 mg twice per day for 3 days and discharged home [16]. Although there are some reports of persisting P. malariae parasites after artemether-lumefantrine [5,17], suggestive of reduced treatment efficacy for this species, most available data support the use of artemether-lumefantrine for P. malariae [18,19]. Thick and thin smears were re-read by expert microscopists and representative images taken (Fig. 1). Parasite DNA was extracted from fixed Giemsa-stained thick and thin smears using a modified protocol for isolating genomic DNA from dried blood spots (QIAamp ® DNA Micro kit, cat numb. 56304) [20]. Plasmodium speciation was performed using modified methods described by Snounou et al. [21]. Plasmodium malariae parasite DNA was detected in material from all slides; P. falciparum was also detected for case 4 ( Table 1).

Discussion and conclusions
In these five symptomatic malaria cases, P. malariae was the most likely cause of the clinical symptoms. All cases were classified as uncomplicated malaria, although they presented with anaemia, a condition that is often associated with chronic infection with P. malariae [3,4]. In four cases, P. falciparum infection was initially diagnosed. Only after an expert microscopist reviewed the blood slides was P. malariae identified as the infecting species. The presence of symptomatic P. malariae monoinfections has implications for the use of HRP-2 RDTs. In both of the study settings, HRP-2 RDTs are routinely used to diagnose malaria. Since non-falciparum Plasmodium species do not express HRP-2 antigen, at least four of these cases would presumably have been missed in routine practice and would not have received malaria treatment.
he cases presented here all reported to the health facilities within a relatively short 13-week period. No effort is made to systematically investigate for non-falciparum malaria among individuals presenting with suspected malaria under standard care. This suggests that a considerable number of patients may present with uncomplicated non-falciparum malaria in this setting. The P. malariae cases reported here were older than typical P. falciparum cases in the area [22], which could be the consequence of a lower force of infection of non-falciparum malaria, older age at first infection, and higher average age at clinical presentation. However, other studies found P. malariae infections predominantly in children [23,24], further highlighting the need for research into the clinical burden of non-falciparum malaria in African settings.  Identifying the species of Plasmodium infections using molecular diagnostic techniques is needed to quantify the burden of malaria in areas where multiple species may be present.