Characterization of Plasmodium ovale spp. imported from Africa to Henan Province, China

As indigenous malaria has decreased over recent decades, the increasing number of imported malaria cases has provided a new challenge for China. The proportion of imported cases due to Plasmodium ovale has increased during this time, and the difference between P. ovale curtisi and P. ovale wallikeri is of importance. To better understand P. ovale epidemiology and the differences between the two subspecies, information on imported malaria in Henan Province was collected during 2010–2017. We carried out a descriptive study to analyze the prevalence, proportion, distribution, and origin of P. o. curtisi and P. o. wallikeri. It showed that imported P. ovale spp. accounts for a large proportion of total malaria cases in Henan Province, even more than that of P. vivax. This suggests that the proportion of P. ovale cases is underestimated in Africa. Among these cases, the latency period of P. o. curtisi was significantly longer than that of P. o. wallikeri. More attention should be paid to imported ovale malaria to avoid the reintroduction of these two subspecies into China.


Results
Prevalence of imported malaria in Henan Province, 2010-2017. A total of 1,372 confirmed imported malaria cases were reported in Henan Province during 2010 to 2017, and no indigenous malaria case has been reported since 2012. The number of imported malaria cases increased yearly and reached a peak in 2014 (n = 216). Among all of the cases, 77.2% were infected with P. falciparum (n = 1059), followed by P. vivax (n = 142), P. ovale (n = 126), and P. malariae (n = 36), accounting for 10.3%, 9.2%, and 2.6%, respectively. One patient was infected with P. knowlesi imported from Indonesia in 2017. Nine patients with mixed infections accounted for 0.7% of cases, mostly with the two species P. falciparum and P. ovale. There were four cases of P. falciparum coinfected with P. ovale, two of P. malariae with P. ovale, two of P. falciparum with P. vivax, and one of P. falciparum with P. malariae. Since 2013, the number of ovale malaria cases was more than that of P. vivax each year, and P. falciparum and P. ovale were the two main species of imported malaria in Henan Province (Fig. 1).

Confirmation of ovale malaria and identification of subspecies.
Including the 6 mixed infection cases of P. ovale with other species, there were a total of 132 ovale malaria cases. Three patients were diagnosed by microscopy without obtaining blood samples, and the other cases were all diagnosed by microscopy and nested PCR. No case of P. ovale was found in 2010, but after that, the number of P. ovale cases increased yearly and reached a peak in 2014 (n = 29). Fifty-two (52/132, 39.4%) cases were confirmed as P. o. curtisi and 77 were P. o. wallikeri (77/132, 58.3%); the other 3 cases could not be identified and are considered as P. ovale spp. Except in 2015 and 2017, the number of P. o. wallikeri cases was much more than that of P. o. curtisi (Fig. 2). As for the 6 mixed infection cases, 5 cases were P. o. wallikeri while the other one was P. o. curtisi.   For the 6 mixed infection cases of P. ovale, the 4 cases coinfected with P. falciparum were from Angola (n = 2), Equatorial Guinea (n = 1), and Cameroon (n = 1), and two cases coinfected with P. malariae were from Congo and Liberia (Fig. 3).
proportion of imported malaria in the country of origin. For the origin countries, the total number of imported malaria cases was different, so more cases of P. ovale malaria did not mean a higher proportion among the four Plasmodium species. To obtain the proportion of the 4 Plasmodium species in the origin countries, the top 12 countries with the most P. ovale malaria cases were analyzed, and among them Liberia had the highest proportion of P. ovale malaria, and the Republic of Guinea had the lowest. For all 12 of the countries, except for mixed infection, the proportion of P. falciparum malaria cases was largest, followed by P. ovale and P. malariae, and that of P. vivax was smallest or equal to P. malariae. Except for Uganda, imported P. malariae existed in the other 11 countries, but imported P. vivax was just found in Angola, Equatorial Guinea, Congo, and Guinea (Fig. 4).
temporal distribution of ovale malaria. There were ovale malaria cases reported every month, but the peaks appeared in April and October for all of the cases, accounting for 12.9% (n = 17) and 11.4% (n = 15), respectively. Compared with the other months, cases in January, July, and November were fewer. For P. o. curtisi, two peaks appeared in March and October, and cases in February, July, and November were fewer compared with the other months. For P. o. wallikeri, most cases were concentrated in February to October without an obvious peak, accounting for 85.7% (66/77) of the total (Fig. 5).

Interval between arrival in China and illness onset. Except for the 6 mixed infection cases of P. ovale
with other species, among the 126 imported ovale malaria cases, 120 (95.2%) had complete information about the date of arrival in China and illness onset, from which the interval days could be calculated. For the interval between arrival and disease onset, the shortest was 23 days before returning to China, the longest was 1265 days after arrival in China, the median was 59 days, and the interquartile range was (14-160) days. Three (   Interval between disease onset and diagnosis. Except for the 6 mixed infection cases of P. ovale with other species, information about the date of symptom onset and diagnosis was collected from the 124 imported ovale malaria cases, and this allowed for calculating the interval days. For the interval between disease onset and diagnosis, the shortest was 1 day, the longest was 59 days, the mode was 1 day, and the median and interquartile range was 4 days (2-7). In addition, 49.2% (61/124) of the patients were diagnosed within 3 days after disease   10,11,28,29 . However, there are relatively few studies about imported malaria that describe the proportion of the four human Plasmodium species in Chinese patients following African travel. Imported P. falciparum cases accounted for 83.4% of the total in Jiangsu Province during 2011-2014, although the proportion of P. ovale cases increased yearly to 3.7%, 9.6%, 8.8%, and 13.0%, respectively 28 . In the present study, the proportion of imported P. falciparum cases was 77.2% during 2010-2011 in Henan Province, which remained about 75.0% in these years, while imported P. ovale cases accounted for 0.0%, 1.4%, 6.4%, 13.2%, 12.0%, 13.0%, 10.6%, and 10.1%, respectively, which was similar to the results in Jiangsu Province 28 .
In this study, we found that P. ovale cases were imported from 20 countries in Africa, with Angola, Congo, Equatorial Guinea, Nigeria, and Liberia being the 5 top origin countries for the 132 imported P. ovale cases. P. ovale is widespread in Africa and fairly common. However, studies about the prevalence of P. ovale in Africa are relatively few. Until recently, species-level diagnosis of malaria has been difficult in some African countries, so the prevalence of ovale malaria may be underestimated. Also, differences in prevalence may be due to different characteristics of Chinese travelers compared to locals in African countries. In addition, factors, such as length of exposure, access to care, past treatment, usage of prophylaxis, and clinical presentation, could explain these differences, which need more study. There are some limitations to this study: the results were obtained without carefully planned sampling, so the largest number of imported malaria cases just showed which origin country was the most common in Henan Province, and the true prevalence of Plasmodium species in the origin country cannot be estimated from this proportion. The proportion of P. ovale malaria in cases travelling to Liberia was highest (25.9%), while that of Angola was just 10.3%. However, there were only 20 African countries analyzed in this study, which were the most common sites of origin with P. ovale in Henan Province, so more surveys about the prevalence of P. ovale in Africa should be performed.
Since P. ovale was classified into two different subspecies, P. o. curtisi and P. o. wallikeri 23 , more attention has been paid to the detection and genotype of the two subspecies [30][31][32][33][34][35] , and studies focusing on the prevalence and distribution of P. ovale spp. have been relatively few. Among 98 imported cases of P. ovale in Jiangsu Province, China, the proportions of P. o. curtisi and P. o. wallikeri were 48% and 52%, which are comparable 12 . In this study, P. o. curtisi and P. o. wallikeri accounted for 39.4% and 58.3%, respectively, with the number of P. o. curtisi cases being less than that of P. o. wallikeri in most countries. The proportion of the two subspecies is similar to that in Ethiopia and Senegal. In northwest Ethiopia, out of 9 P. ovale cases, 7 patients were infected with P. o. wallikeri while the other 2 were infected with P. o. curtisi 36 . In Senegal, out of 235 malaria cases, 4 patients were infected with P. ovale, 3 with P. o. wallikeri, and 1 with P. o. curtisi 37 .
For indigenous malaria, the peak has appeared around August and September in China. However, there is no obvious peak for imported malaria 8 . In this study, the peaks of total P. ovale and P. o. curtisi were similar and appeared in March and October, which might be related with the Chinese festivals, the Spring Festival in February, and National Day in October. Migrant workers prefer to return to China during these festivals. However, no obvious peak was observed for P. o. wallikeri. The differences in temporal pattern between P. o. curtisi and P. o. walliker deserve more study.
It was difficult to obtain the exact date of malaria infection in this epidemiological investigation. However, imported malaria cases are suitable for observing the latency period of malaria. The date of arrival in China was used as the date of malaria infection, and the intervals between the date of arrival and illness onset were calculated. Although imprecise, this can be used as a proxy to estimate the latency period, for which P. ovale malaria is mostly observed through imported cases 28,38 38 , whereas the latency period of the two subspecies showed no difference in the study of Cao et al. 28 . In the present study, the latency period of P. o. curtisi (97.5 days) was more than three times as long as that of P. o. wallikeri (31 days), which is similar to the work of Nolder et al. 38 . The longest latency period in this study was 1265 days, which happened in a case of P. o. curtisi; similarly, the longest was also a case of P. o. curtisi in the study of Nolder et al. (1083 days) 38 . The liver stage of malaria commonly exists for parasites of P. ovale and P. vivax; the hypnozoites in the liver last months or years after infection, which results in a long latency period. This makes it quite difficult to diagnose imported malaria cases in nonendemic countries because the presenting illness may be unrelated to recent travel abroad [39][40][41] . More studies about the latency period of P. ovale and P. vivax parasites will improve the diagnosis of imported malaria. Henan Province, located in the Middle East of China, has a warm temperate zone, subtropical, humid-semi-humid monsoon climate. The climate and environment are suitable for the breeding of Anopheles, including Anopheles sinensis and Anopheles anthropophagus, so the threat of malaria reintroduction remains. Currently, for imported malaria, timely diagnosis, standardized treatment, and preventing further transmission are the key strategies to eliminate sources of infection in China.

Conclusion
This study illustrates that imported P. ovale cases account for a large proportion of the total cases in Henan Province, even larger than P. vivax cases. The proportion of P. ovale cases is underestimated in the origin countries of Africa, which is large and exceeded only by P. falciparum. We confirmed that the latency period of P. o. curtisi is much longer than that of P. o. wallikeri. It is recommended that more attention be paid to imported P. ovale cases to avoid reintroduction in China.

Methods
Data collection. Information on imported malaria in Henan Province was collected from case investigation reports and the Disease Surveillance Information Report Management system of the China Centers for Disease Control and Prevention (CCDC), including gender, age, occupation, travel history, date of arrival in China, date of symptom onset, date of diagnosis, and reported institution.
Species and subspecies confirmation. All of the cases were initially diagnosed by blood smear microscopy and/or rapid diagnostic tests (RDT) at the basic level of the hospital or CDC, then all of them were confirmed by the Henan Provincial Reference Laboratory for Malaria Diagnosis using nested polymerase chain reaction (PCR) and/or blood smear microscopy. The microscopy was performed by microscopists through the examination of thick and thin blood film at 1000×magnification. Rapid diagnostic tests were used to detect malaria antigens following the manufacturer's instructions by laboratory staff. Nested PCR was performed using the methods published 42,43 . Subspecies of P. ovale were confirmed by nested PCR following sequencing according to previously published methods 23,30 . Data analysis. All of the case information was collected and input into SSPS v.21.0 (Statistical Product and Service Solutions), in which all of the calculations and analyses were performed. The differences were compared by Wilcoxon rank-sum test, Chi-square, or Fisher's exact tests, and a two-sided p value of <0.05 was considered statistically significant. Geographic distribution was conducted by geographic information system (GIS)-based spatial analysis, which was fulfilled by QGIS (Quantum GIS) v2. 18.19. ethics approval and consent to participate. All of the identifiers of patients were removed from the data. Informed consent was obtained from all of the patients. This study was approved by the Ethical Committee of the Henan Province Center for Disease Control and Prevention. All of the methods were performed in accordance with the relevant guidelines and regulations of the Henan Province Center for Disease Control and Prevention.

Data Availability
The datasets used and/or analyzed during the current study are available from the first/corresponding author on reasonable request.