Answer to Q10
It is common knowledge that 95% of the malaria in Sub-Saharan Africa is caused by Plasmodium falciparum, the most lethal form of malaria and responsible for the majority of deaths due to malaria in the world.
Non-falciparum malaria is known to occur in the region but not routinely looked for by either clinicians or laboratory technicians as it very scarce and seldom the cause of complications and death in otherwise well patients.
Rapid antigen tests that are single species specific will not detect non-falciparum malaria if intended solely for the detection of P falciparum. Rapid antigen tests that can detect multiple species are generally much less reliable and their interpretation prone to mistakes.
It is therefore of epidemiological and public health importance to confirm the diagnosis as this may be the index case highlighting the fact that there is non-falciparum malaria on site and that it might be prudent to change the malaria management protocol on site, assuming that this particular patient did not contract P ovale elsewhere in the world where P ovale is well known to occur.
Discussion
Plasmodium ovale malaria is not commonly diagnosed in South Africa
As Plasmodium falciparum malaria is the most common, as well as the potentially most lethal form of malaria in Sub-Saharan Africa, we tend to focus on the clinical diagnosis, laboratory investigation and management of acute uncomplicated and complicated falciparum malaria in travel health teaching and discussions in South Africa.
This case puts the focus on the other, somewhat ‘neglected’ malaria species on the continent.
In our travel health practice we consult approximately 3 500 travellers a year and, on remote sites in Sub-Saharan Africa, take care of roughly 600 expatriates and 5 700 national client employees on any given day of the year.
We annually diagnose about 280 malaria cases on the remote sites (implying that less than 1% of the total number of persons under our care in malaria endemic areas are diagnosed with malaria) and perhaps 10 in returned travellers and medevac cases in our Johannesburg based clinic.
We have been doing this for more than fifteen years.
This is the first case of non-falciparum malaria emanating from one of our sites in which it can be considered that we ‘missed’ the diagnosis. In total we are aware of about three cases of non-falciparum a year amongst our entire client base, including that of an international health insurance company for whom we provide medical cover. Most of these cases originated from Mozambique and the actual diagnosis of non-falciparum malaria could not be confirmed in more than one of them, illustrating the fact that the diagnosis of non-falciparum malaria is often not an easy one in a non-endemic area.
We are aware of one other case - the patient worked on a remote site in Liberia and was diagnosed with P ovale in Australia. We could however not say with certainty that he contracted the malaria on site in West Africa as he had travelled and worked in South East Asia, where P ovale is much more common, extensively.
Plasmodium ovale
Historical references istretching from the Philippines to the Gold Coast in West Africa, of what was likely Plasmodium ovale can be found in the literature from 1900 onwards. It had been reported from Indonesia and has been introduced to Vietnam, Thailand and India. Two sub-species have recently been documented: P. ovale curtisi and P. ovale wallikeri.ii
P ovaleiii is seldom seen except in Sub-Saharan Africa and some western Pacific islands (New Guinea and the Philippines). In SSA it makes up part of the less than five percent of non-falciparum malaria diagnosed annually, amongst a sprinkling of P malariae and some P vivax. P ovale has been documented in Senegal and The Gambia before. Prevalence rates in children in the Republic of Congo, Liberia, Cameroon and Gabon are in the region of 1,9% to 9% but this must be seen on the background of a prevalence of malaria of 24,5 to 30% amongst the same children. In a study in Dielmoiv , Senegal only three cases of P ovale was detected out of a total number of 162 malaria patients. All the others had P falciparum malaria and this in turn was responsible for 61% of febrile episodes recorded in the study. Other countries from which P ovale infestations had been documented include: Southern Sudan, Uganda, Zimbabwe, Ethiopia, Zambia, Tanzania and Kenya.
Sporozoites injected by the mosquito rapidly invade the liver where it matures in about 9 days. This corresponds with a prepatent / ‘incubation’ period of 12 to 20 days with a median in one study of 14 days. Hundreds of merozoites are released to invade young red cells. Some of the parasites remain behind in the liver in limbo to form hypnozoites. They may only develop into merozoites months to years later.
The merozoites in the red cells will erupt and invade a next batch of red cells after about 49 hours. After another 49 hours some of the merozoites become gametocytes.
Mosquitoes then take up these microgametocytes during a blood meal, so completing the developmental cycle in the mosquito. Sprozoites appear in the salivary glands weeks after the mosquito had become infested.
Fever spikes occur roughly every 49 hours.
P ovale malaria smears usually record low parasite counts in comparison to P falciparum and P vivax infections due to the parasites restriction to reticulocytes. This makes diagnosis by conventional smear as well as per rapid antigen test relatively difficult. Persons who have previously had P ovale may have reduced parasite levels and lower fever peaks.
Relapsing infection occurs in P ovale due to the presence of hypnozoites in the liver. These may be asymptomatic - posing a potential public health threat with the ‘silent’ introduction of the parasite in to a P ovale naïve community, such as a mine or construction camp.
A first relapse may occur 17 days after completion of treatment of the primary attack but has been recorded up to 255 days later.
Delayed primary attacks may occur up to four years after infestation of the liver due to the effect of malaria prophylaxis or treatment taking at the time of what would have been the usual primary attack.
The parasite makes use of a wide range of Anophelene mosquitoes as vectors, with A gambiae and funestus being the most likely.
The diagnosis of P ovale is normally done by microscopic examination of a Giemsa stained peripheral blood smear. It is difficult to differentiate P ovale from P vivax!
P ovale produces fewer merozoites in the red cell and does not distort / enlarge the erythrocyte as much as P vivax. It does however produce a higher percentage of ‘oval’ / elliptical red cells and more Schüffner’s stippling - for those of you that cant wait to rush out to a microscope!
Human malaria parasite species can be differentiated by making use of PCR - this should be reserved for difficult cases and use in research and is NOT intended as first in line for day to day diagnosis of malaria in the clinical setting.
Lessons Learned
The case under discussion provides an opportunity to revisit a number of well-known dictums in the diagnosis and treatment of malaria in Africa and adds a few new one’s to consider:
- Any patient who presents with a febrile / ‘flu-like illness in a malaria endemic area has MALARIA UNTIL PROVEN OTHERWISE.
- One negative blood smear / rapid antigen test does not exclude malaria: Repeat the test until malaria is diagnosed or another diagnosis explaining the fever and other signs and symptoms is conclusively made and laboratory confirmed.
- A good history and comprehensive physical examination remains the basis of good medical practice – however, in this case the history of ‘urological problems’ coupled with blood and leukocytes in the urine side-room test, served as a distractor. Haemolysis caused by malaria will produce a positive dipstick for blood / haemoglobin as well as bili- and urobilinogen. Malaria should not as a rule cause a positive dipstick for leukocytes or nitrates – which is why the clinicians taking care of the case ‘saw’ the urogenital infection in their findings.
- If the clinicians had access to the patient’s pre-deployment medical records they would have been able to compare the results of a pre-deployment medical examination with the present - this might have shown that the patient had leukocytes prior to deployment as part of an on-going chronic low-grade urogenital condition. This might have turned their attention back to another likely cause for the blood / haemoglobin and urobilinogen, e.g. malaria.
- A travel history, current vaccine cover and whether the patient has been compliant with malaria prophylaxis is always of paramount importance in any febrile traveller as this assists in compiling a list of more and less likely differential diagnosis.
- A comprehensive physical examination in which the clinician actively looks for signs that may explain the origin of a documented fever is essential. Important points to note include skin rashes, insect bites and eschars, lymphadenopathy, organomegaly and jaundice.
- Bedside / side room laboratory tests play an important role in any clinical setting, more so in areas with limited medical infrastructure, such as a remote mine or construction site. In addition to malaria rapid antigen tests, basic urine dipstick screening and blood glucose levels are invaluable in every setting.
- Malaria rapid antigen tests have revolutionized the diagnosis of malaria in individuals and the epidemiology of the disease across the African continent.
- A good quality rapid antigen test is more accurate and reliable than poor microscopy in both the malaria endemic developing world and in laboratories in the industrialised, non malaria endemic world, benefitting febrile inhabitants of malaria endemic countries as well as ill returned travellers in non-endemic industrialised countries.
- Rapid antigen tests have increased the standards of malaria diagnosis in the developing world, thereby contributing to the massive decrease in malaria reporting to the WHO in the last decade.
- Despite the advantages listed in the previous bullet point, the test is subject to certain limitations that need be borne in mind at all times:
- The rapid antigen test used must be well manufactured, correctly stored and used.
- There are two types of rapid tests commonly used in Africa:
- A test that detects the Histidine 2 Rich Protein of the parasite wall and,
- A test that detects LDH produced by living parasites.
- The single species H2RP test for P falciparum is highly sensitive and specific and the test of choice in Sub-Saharan Africa where P falciparum is rife and poses a high risk of death in non-immune travellers and immunocompromised individuals.
- The multi-species tests that detect P falciparum, non-falciparum and mixed infections are less sensitive and specific and are more difficult to interpret correctly.
- Because of the serious consequences that a missed P falciparum infestation may have, clinicians may choose to preferentially use the single species rapid antigen test for P falciparum only, but then run the calculated risk of missing a non-falciparum malaria, as happened in this case.
- It is noteworthy that the laboratory in the United Kingdom detected P ovale on the malaria smear, confirmed by relatively sophisticated PCR, but that two immunochromatic tests (Rapid antigen tests) for P ovale were in any event negative…
- Conversely, the rapid antigen tests done on the mine were intended for the diagnosis of P falciparum only; the test was repeated twice but, with hindsight not surprisingly, negative in both instances.
- PCR is not intended as a primary method of detecting malaria in the clinical setting.
- Look - and think! – twice whenever a patient who presents with a fever on the first day, looks and feels better on the second day: This is no reason for relief as malaria very often presents just like that. Fever on the second day is more likely to be influenza (or another infectious disease) whereas a normal temperature is all the more reason to check on the patient on the third day – when the likelihood of malaria rises exponentially if the fever returns…. As it did in this case.
- For those of us who operate in a closed community setting - as on a remote site and sometimes in small villages / communities: Beware the “Canteen consultation” - the patient who declares himself fit and looks so at a distance, may still be seriously ill. Only a proper consultation in a clinic setting will reveal a bounding, rapid pulse, actual fever as recorded with a proper thermometer or jaundice observed at close quarters in good light.
- Although not always typically so, recording a temperature on graph paper remains useful in the detection of a pattern - more so than simple inscriptions in a patient file. Doing so might have revealed a pattern reminiscent of malaria in this patient.
- Sending a febrile patient out to his room for the night if he had been ‘sleeping’ all day and is still pyrexial can be courting disaster. Cerebral malaria does not always present as full blown ‘coma’ but may have an insidious onset with somnolence and mild cognitive impairment as early ‘soft’ signs.
- The kidneys are target organs in malaria - a febrile patient with prominent lower back pain and concentrated urine on a remote site in the tropics is a disaster waiting to happen. Ensure adequate hydration - but don’t ‘drown’ the lungs! - And be extremely careful with the use of non-steroidal anti-inflammatories under such conditions. Paracetamol is far safer for the relief of pain and fever and equally effective.
- Any clinician working in an isolated setting with limited support must use good judgment when it comes to making decisions regarding medical evacuation to a higher level of medical care. It is often better to make use of an opportunity to refer a patient out by ‘routine’ means such as a regular roster charter than to have to become dependent on a dedicated medevac modality by road or air…. Making a call to refer earlier rather than later is often the better decision - as long as the clinician taking care of the patient is assured that he is indeed referring to a higher level of care and not just ‘dumping’ the patient. This sometimes implies that a patient who cannot be taken care of on a remote site will also not be better off in the capital of a developing country but needs to be medevaced all the way to a first class medical facility much further afield. This has several implications including logistics, safety and security issues, cost and time away from site.
- In this case the clinician on site decided to refer to Dakar but it is the patient who called for referral home. It would have been interesting to see whether the diagnosis of non-falciparum malaria would have been entertained in Dakar…
- In the absence of a definite diagnosis it is questionable whether it was ethically correct to allow the patient to return to the UK on a commercial airline – the risk to fellow passengers always needs to be considered. In this case the actual cause of the fever turned out to be malaria, which is not contagious in the airline setting. Pragmatism unfortunately often rules in situations like this and in this particular case did not harm any fellow travellers and afforded the patient a rapid, accurate diagnosis once back in the United Kingdom.
- The UK based doctor acted correctly in immediately asking for a malaria screen as part of the initial work-up of the patient - as much as thirty percent of febrile returned travellers from malaria endemic areas have indeed got some form of malaria. This is more the case in returned VFR’s (Visiting Friends and Relatives) than in expatriates and business travellers but nevertheless should never be omitted in any traveller from a malaria risk area.
- Typhoid remains part of the differential diagnosis of this patient in spite of the fact that he had been vaccinated against typhoid - the vaccine is only about 75% effective and other Salmonellosis should also be considered. The absence of diarrhoea does not exclude the diagnosis.
- The use of rapid malaria antigen tests should not just be limited to remote settings far from good laboratory services - they are also useful in a practice that sees febrile returning travellers on a regular basis as it helps in the triage process and to reach a quick decision regarding the likely diagnosis of malaria / not and whether a patient is likely to need urgent admission or not. The clinically very ill patient with a positive rapid test is likely to need immediate admission whereas one that is not acutely ill but still has a positive malaria antigen test would deserve a ‘second look’ regarding his immediate further management. Patients that are not acutely ill, have no obvious other diagnosis and a negative malaria antigen test can be sent home with more confidence than without the test.
- The use of artemether / lumefantrine in the treatment of P ovale malaria has recently been documented. It is at least as effective as treatment with chloroquine and simplifies decision making in the clinical setting where either ovale, falciparum or a mixed infestation is considered.
- Clinicians involved in travel health, work across borders and should be au fait with some of the important trade names they are likely to come across - hence the inclusion of the reference to “Riamet®” and “Coartem®” in the case study and highlighting the fact that it is exactly the same drug by a single manufacturer marketed under different trade names. This is sometimes cause for concern amongst travellers and clinicians alike, especially in view of the distrust of medicines in developing countries across the globe and the concern that colleagues in non malaria-endemic countries may “… not know how to treat malaria correctly...”
- The use of Primaquine to avoid hypnozoite induced relapse is essential not only for the patient but also to ensure that the patient does not perhaps introduce P ovale to the mosquito population on site on his return. (It is not patently clear that he did contract ovale on site - he had been travelling through Dakar and may have been visiting town and villages in the region during his time on the mine.)
- The argument that he will be returning to an area with P ovale and that he should therefor only receive Primaquine as a radical cure at the end of his contract does not hold true!
- Finally, it is important for the mine management to know whether there is a real risk for the existence of P ovale malaria on the mine or in its immediate environs as this would influence the way malaria is diagnosed on site and may also have a bearing on the existing malaria vector control programme.
- Surveillance needs to be set up and an attempt made to identify ovale infested mosquitoes as part of the existing and on-going malaria control programme.
- The clinicians on site need to consider adding multi-species malaria rapid antigen test kits to their existing side-room laboratory and to institute the use of the kits as part of a suspected malaria management protocol. In the light of just a single case of P ovale being detected in nearly ten years on the project it would not necessarily be wise to replace the single species, P falciparum rapid tests with multi-species tests in the light of the lesser sensitivity and specificity of the latter.
- Should the mine expand further it may be worthwhile considering a well trained, full time laboratory technician who can do malaria smears on site, but ONLY if a malaria smear quality control programme can be instituted as well.
Plasmodium ovale is seldom diagnosed in returned travellers in South Africa but some cases may be missed either because they are subclinical or because they are reported as P falciparum by less experienced laboratories in non-endemic areas. Treatment with the standard first line regime of artemether / lumefantrine will be effective in warding off the primary attack. Any person who has a relapse soon or weeks after an initial episode of malaria should be considered as perhaps harbouring P ovale or P vivax and a specific attempt made to confirm such as diagnosis to afford the patient a radical cure with Primaquine and avert the possible introduction of P ovale into a previously un-infested community when he returns to the malaria endemic area.
Thank you to everyone who took the time to work through this case study - the author would welcome any constructive remarks and comments.
I would like to thank Prof John Frean, NICD and University of the Witwatersrand for his input on the content of this case.
References
- Plasmodium ovale: PARASITE AND DISEASE. William E Collins, G M Jeffery in CLINICAL MICROBIOLOGY REVIEWS, July 2005, p 570 – 581
- Sutherland CJ, Tanomsing N, Nolder D, Oguike M, Jennison C, Pukrittayakamee S, Dolecek C, Hien TT, do Rosário VE, Arez AP, Pinto J, Michon P, Escalante AA, Nosten F, Burke M, Lee R, Blaze M, Otto TD, Barnwell JW, Pain A, Williams J, White NJ, Day NP, Snounou G, Lockhart PJ, Chiodini PL, Imwong M, Polley SD (2010). "Two nonrecombining sympatric forms of the human malaria parasite Plasmodium ovale occur globally". J Infect Dis 201 (10): 1544–50
- http://www.malariasite.com/malaria/africa.htm
- The Dielmo Project: A longitudinal study of natural malaria infection and the mechanisms of protective immunity in a community living in a holoendemic area of Senegal. Am.J.Trop Med. Hyg., 51(2) 1994, pp 123-137. ISSN 0002-9637
