High seroprevalence of IgM antibodies against chikungunya among patients with acute febrile illness seeking healthcare in a malaria-endemic area in the Afar Region,Northeast Ethiopia

Abstract

Objective:

Chikungunya is a disease caused by Aedes mosquito-borne chikungunya virus infection. This disease is becoming one of the emerging/re-emerging viral diseases in tropical and subtropical countries but is neglected by public health systems. This study assessed the seroprevalence of chikungunya virus infection among patients with acute febrile illness seeking healthcare in a malaria-endemic area in the Afar Region, Northeast Ethiopia.

Methods:

An institution-based cross-sectional study was employed from September 2022 to March 2023 to assess the seroprevalence of chikungunya virus infection among malaria-suspected febrile patients attending health institutes in the Afar Region. Sociodemographic, clinical features, and venous blood were collected from each study participant. Blood films were prepared and examined for plasmodium infection using microscopy. Sera were separated and screened for anti-chikungunya virus IgM and IgG antibodies using an enzyme-linked immunosorbent assay. Data were entered into Epi Data 3.1 and analyzed using Stata/SE 14.2.

Results:

In this study, 368 malaria-suspected febrile patients (55.4% female, aged 5–80 years old, mean ± SD = 27.28 ± 14.0) participated. The prevalence of anti-chikungunya virus IgM antibodies, indicating acute infection, was 47.8%, while the prevalence of IgG antibodies, indicating previous exposure, was 6.3%. Nonmarried participants were found to be more likely to have acute chikungunya virus infection (AOR = 2.34, 95% CI: 1.141–4.964), and back pain was associated with higher likelihood of acute infection (AOR = 1.785; 95% CI: 1.078–2.954). About 10.6% of the participants tested positive for Plasmodium parasite infection, with P. falciparum (84.6%) being the most common, followed by P. vivax (10.3%) and mixed (5.1%) infections.

Conclusion:

The study revealed a high seroprevalence of acute chikungunya virus infection among febrile patients in the Afar Region, highlighting the importance of regular screening for arbovirus infection in febrile patients to control and mitigate chikungunya spread.

Keywords

Chikungunya seroprevalence malaria-endemic acute febrile patients Ethiopia

Introduction

Chikungunya (CHIK) is a viral disease transmitted through the bite of the chikungunya virus (CHIKV) infected Aedes aegypti and Ae. albopictus mosquitoes.^1,2 CHIK has been spreading rapidly across tropical and subtropical regions since its first isolation in Tanzania in 1952.^3,4 The infection was reported in over 100 countries and caused outbreaks with irregular re-emergences every 2–20 years, affecting millions of people and causing clinical and epidemiological concerns since the 2000s.^5–9 This can be attributed to environmental factors that favor mosquito breeding, the emergence of alternative vectors and new ecological niches, and viral mutations.^8–10

Chikungunya is a type of acute febrile illness characterized by a body temperature of 37.5°C or higher, lasting for a week.¹¹ It primarily affects the musculoskeletal system, leading to severe joint pain, fever, and other symptoms such as headache, back pain, nausea, vomiting, rash, and conjunctivitis.^5,12 Diagnosis of CHIKV infection involves detecting virus or viral RNA in serum within 6 days of illness using molecular techniques like reverse transcriptase polymerase chain reaction.¹³ Serological tests such as enzyme-linked immunosorbent assays (ELISAs), indirect immunofluorescence assays, or rapid lateral flow tests can also detect CHIKV-specific immunoglobulin M (IgM) antibodies after 5 days of illness onset.^13,14 The IgM antibodies typically appear within 3–8 days of CHIKV infection illness onset and decrease within 2–3 months,¹⁵ making it a valuable preliminary diagnostic test for acute infection, especially from day 5 of symptom onset.¹⁶ Detection of IgG antibodies can provide evidence of previous exposure to the virus and offer epidemiological insight into the disease.^17,18

In Ethiopia, the identification of the yellow fever virus in 1960–1962 from unidentified febrile illness outbreak cases raised suspicion of other arboviral etiologies like CHIK, responsible for unknown febrile illness outbreaks across the country.^19,20 However, due to similar clinical symptoms and lack of laboratory facilities for the screening of arboviral infections, febrile cases have often been misdiagnosed as malaria or typhoid fever.¹⁷ In 2016, Ethiopia reported its first confirmed case of CHIKV infection during a nonmalaria febrile illness outbreak in the Dollo Ado district of the Somali Region.²¹ Since then, significant outbreaks have occurred in the eastern regions of the country, impacting many people primarily in the Dire Dawa City Administration, the Afar Region, and the Somali Region.^22,23

To effectively monitor and control CHIK and other arboviral diseases, continuous surveillance and routine screening of febrile patients are crucial, even during the non-outbreak period.²⁴ However, information on the burden and impact of CHIKV infection among the community during non-outbreak periods is limited in most tropical countries, including Ethiopia. A study was conducted to determine the seroprevalence of CHIKV infection among patients with acute febrile illness seeking healthcare in a malaria-endemic area in the Afar Region, Northeast Ethiopia. The findings of this study can expand our understanding of the public health impacts of CHIK and help facilitate early control and prevention measures.

Methods

Study design, period, and settings

A cross-sectional study was conducted among patients with acute febrile illness seeking healthcare in four public health facilities in Zone 3 of the Afar region, Ethiopia, from September 2022 to March 2023. The Afar region is characterized by its rich biodiversity, which includes various nonhuman primates and mosquito species as well as periodic flooding of the Awash River, increasing the risk of arbovirus transmission. The recent outbreaks of dengue fever, CHIK, and other febrile illnesses in the region and neighboring regions²² indicate that these diseases potentially remain transmitted among humans and primates through mosquitoes, leading to additional arboviral spillover.²⁵ Zone 3, located about 140 km from Semera, the region’s capital city, and 260 km from Addis Ababa, was selected purposively due to the previous experience of outbreaks of arboviral illnesses and its proximity. This Zone is located on the main international road that links Ethiopia and Djibouti and has predominantly tropical to subtropical climates. The rural populations of the Zone are pastoralists who depend on camels and small and large ruminants, while some agro-pastoralists cultivate crops using the Awash River’s irrigation. Meanwhile, the urban population comprises government employees, labor workers, and small-scale merchants.

Out of the eight districts in Zone 3,²⁶ the study focused on four public health centers (HCs) in three districts, selected based on their proximity and previous history of outbreaks, namely, Awash Arba HC and Sidafage HC from Amibara District, Andido HC from Harunka District, and Awash Sebat HC from Awash Sebat City.

Inclusion and exclusion criteria

The study included patients with acute febrile illness and a body temperature of ⩾37.5°C who were suspected of having malaria. Patients who did not exhibit symptoms of malaria or were suspected of having other diseases, pregnant women, and apparently anemic patients were excluded from the study. To minimize the difficulty of blood collection related to children in the laboratory, patients below 5 years old that were managed in the pediatric outpatient departments were excluded.

Sample size determination and sampling

To estimate the required sample size for our study, we used the prevalence of CHIKV-specific IgM from another area of Ethiopia, which was 22.5% among febrile patients in a recent health facility-based seroprevalence study.²⁷ We used the sample size estimation method outlined by Sullivan²⁸ and calculated a sample size of 294 at a 95% confidence level, with a 5% margin of error and a 10% nonresponse rate. To ensure reliable results, the sample size was rounded up to 368.

Study participants and data collection

The participants of the study were patients with acute febrile illness (auxiliary temperature ⩾ 37.5°C) aged ⩾ 5 years who sought healthcare and were suspected of having malaria by clinicians, regardless of their sociodemographic characteristics. At each public HC, outpatient department healthcare workers were trained and collected sociodemographic data, clinical features, and duration of symptoms using a structured questionnaire after obtaining written informed consent or assent. To avoid the duplication of participants, the data were collected following patient registration number during the data collection period. Trained medical laboratory personnel collected venous blood samples using a gel-containing serum separation transport vacutainer tube. Thick and thin blood films were prepared on-site to detect malaria parasite infection. The remaining blood was left to clot, and the serum was separated using centrifugation. The clear serum was stored at −70°C until further testing for CHIKV infection at the laboratory of the Aklilu Lemma Institute of Pathobiology of Addis Ababa University in Ethiopia. The tools were pretested on 5% of the sample size to evaluate and check its consistency at Werer HC of Amibara district.

Laboratory analysis

To detect malaria parasite infection, blood films were stained with Giemsa stain and tested on-site following the Ethiopia Ministry of Health (MOH) guidelines.²⁹

The Virion-Serion Diagnostic ELISA Classic IgM (Cat. No. ESR148M, Germany) and IgG (Cat. No. ESR148G, Germany) kits were used to detect anti-CHIKV IgM and IgG antibodies from patients’ sera according to the manufacturer’s instructions.³⁰ Briefly, to assess evidence of recent or acute CHIKV infection, ELISA CHIKV IgM with indicated a sensitivity of 91.7% and a specificity of >99% was used. To determine evidence of previous CHIKV infection, ELISA CHIKV IgG was employed with indicated sensitivity and specificity of >99%. The testing procedure for anti-CHIKV IgM antibodies was identical to that for IgG detection, except that the sera were pretreated with rheumatoid factor absorbent during the preparation of specimen dilutions.³⁰ The optical density (OD) of the sample was measured spectrophotometrically at 405 nm in a micro plate reader. The mean OD value of the measured standard serum of each kit, tested in duplicate, was used to fix the cut-off ranges for qualitative evaluation of results. Results were interpreted as positive, negative, or borderline (equivocal) according to the Quality Control Certificate provided by the company, but the borderline results were considered negative for this analysis.^30,31

Data entry and analysis

Data were entered into Epi Data v.3.1 (The EpiData Association, Odense, Denmark) and analyzed using Stata/SE 14.2 (College Station, Texas, USA). Frequencies and percentages were used to summarize sociodemographic characteristics and clinical features. The study estimated the seroprevalence of anti-CHIKV antibodies and Plasmodium malaria by dividing the number of participants with positive test results by the total number of study participants. Univariable logistic regression analysis was conducted to assess the association between rates of anti-CHIKV IgM antibodies and malaria parasites with sociodemographic characteristics such as gender, age, occupation, education, residence, marital status, ethnicity, religion, and districts, as well as clinical features of study participants. Similarly, multivariate logistic regression analysis was conducted on sociodemographic and clinical features to identify the main risk factor for the rates of anti-CHIKV IgM antibodies and Plasmodium infection while controlling for confounders at a statistically significant level. Odds ratios (OR) with 95% confidence intervals (CI) were used to measure the strength of the association between positivity and the variables. The analysis result with a p-value of <0.05 was considered statistically significant.

Ethical consideration

This study was ethically approved by the Institutional Review Board of Aklilu Lemma Institute of Pathobiology, Addis Ababa University (ALIPB IRERC/88/2014/22). Permission for the study was obtained from Amibara, Harunka, and Awash Sebat City Health Offices. Written informed consent was obtained from all participants aged 18 years and older, and from those under 12, informed consent was obtained from their legally authorized representatives. For participants between the ages of 12 and 17 years, written informed consent was obtained from the participants and their legally authorized representative. For participants without formal education, the consent or assent was read aloud in the presence of literate witnesses, and participants provided their mark with a thumbprint. Blood samples were collected under aseptic conditions by experienced laboratory technicians. Participants who tested positive for malaria were treated with appropriate antimalaria drugs according to malaria treatment guideline standards, while nonmalaria cases were treated with alternative antibiotics or analgesics/antipyretics based on the presumptive diagnosis of the clinician.

Results

General characteristics of the study participants

A total of 368 febrile patients (55.4% females, age ranged from 5 to 80, mean ± SD = 27.3 ± 14.0 years old) were involved in the study. The majority of respondents were between 21 and 30 years old (37.2%), married (60.6%), illiterate (45.7%), and urban residents (53.8%). Most of the patients were from Awash Arba HC (44.8%), followed by Awash Sebat HC (34.2%) (Table 1).

Table 1.

General characteristics of patients with acute febrile illness seeking healthcare in a malaria-endemic area in the Afar Region, Northeast Ethiopia (N = 368).

General characteristics	Number (%)
Sex
Male	164 (44.6)
Female	204 (55.4)
Age (years)
5–10	35 (9.5)
11–20	94 (25.5)
21–30	137 (37.2)
31–40	49 (13.3)
41–50	28 (7.6)
51 and above	25 (6.8)
Education status
Illiterate	168 (45.7)
Grade 1–8th	118 (32.1)
Grade 9–12th	62 (16.8)
College and above	20 (5.4)
Resident
Urban	198 (53.8)
Rural	170 (46.2)
Occupation
Agro/pastoralist	41 (11.1)
Government/private worker	129 (35.1)
Housewife	104 (28.3)
Students	94 (25.5)
Marital status
Married	223 (60.6)
Single	145 (39.4)
Religion
Muslim	226 (61.4)
Christians	142 (38.6)
Ethnicity
Afar	177 (48.9)
Others	185 (51.1)
Districts
Amibara	201 (54.6)
Awash Sebat	126 (34.1)
Harunka	41 (11.1)
Health institutes
Awash Arba	165 (44.8)
Awash Sebat	126 (34.2)
Sidafage	36 (9.8)
Andido	41 (11.1)

Clinical features reported in relation to CHIKV infections

During the study, participants were asked if they experienced any clinical features associated with the CHIKV infection besides fever. Headache (90.76%), joint pain (71.20%), and back pain (50.27%) were the most commonly reported symptoms. The duration of illness before seeking medical care ranged from 1 to 8 days, with an average of 2.97 ± 1.45 days. The majority of participants (93.75%) experienced the illness for 5 days or less (Table 2). None of the respondents reported any other clinical symptoms associated with arbovirus infections, such as mucosal bleeding, bloody vomit or stool, bleeding gums or nose, or petechiae.

Table 2.

Clinical features reported by patients with acute febrile illness seeking healthcare in a malaria-endemic area in the Afar Region, Northeast Ethiopia (N = 368).

Reported clinical features		Number (%)
Duration of the illness	⩽5 days	345 (93.7)
	>5 days	23 (6.3)
Headache		334 (90.8)
Joint pain		262 (71.2)
Back pain		185 (50.3)
Chills		167 (45.4)
Nausea/vomiting		161 (43.7)
Diarrhea		49 (13.3)
Malaise		81 (22.0)
Muscle pain		81 (22.0)
Sore throat		42 (11.4)
Rash		6 (1.6)
Retro-orbital pain		4 (1.1)

Seroprevalence of anti-CHIKV IgM antibody among malaria-suspected patients

The ELISA results revealed that almost half of the patients’ sera (47.8%) had anti-CHIKV IgM antibody suggesting recent or acute CHIKV infection. The patients’ sera found at the borderline for IgM antibody were 19.0%, while the remaining 33.2% sera found negative for anti-CHIKV IgM antibody. On the other hand, while only 6.3% of the patients’ sera tested positive for IgG antibody, 2.7% of patients’ sera showed the presence of both anti-CHIKV IgM and IgG antibodies (Figure 1(a)).

Figure 1.

Prevalence of anti-CHIKV antibodies (a) and Plasmodium parasite infections (b).

Prevalence of malaria

Out of all febrile patients screened for malaria parasites using microscopy, 10.6% of participants were diagnosed for Plasmodium parasite infections. The majority of whom were positive for P. falciparum (84.6%), followed by P. vivax (10.3%) and mixed infections (5.1%) (Figure 1(b)).

Out of all the participants, 16 (4.3%) were found to be positive for both anti-CHIKV IgM antibodies and plasmodium infection, with 15 patients positive for P. falciparum, and one patient for P. vivax (data not presented).

Seropositivity of anti-CHIKV IgM and associated sociodemographic and clinical features

The seropositivity of anti-CHIKV IgM antibodies was similar in males and females, with females having a slightly higher rate (49.0% vs 46.3%). The lowest seropositivity rate for anti-CHIKV IgM antibodies was found among individuals aged 51 and above (COR = 0.187, 95% CI: 0.059–0.587). However, this difference was not statistically significant in multivariable logistic regression analyses. The odds for single febrile patients to test positive for anti-CHIKV IgM antibodies were 2.3 times as large as those married (AOR = 2.340, 95% CI: 1.141–4.964). Univariable logistic regression analysis showed that the odds of participants from the Harunka district having positive test results for anti-CHIKV IgM antibodies were two times larger compared with those from Awash Arba HC (COR = 2.006, 95% CI: 1.010–3.986). Contrarily, multivariable logistic regression analysis showed that the odds for study participants from Sidafage HC (AOR = 4.927, 95% CI: 1.975–12.290) and Andido HC (AOR = 2.593, 95% CI: 1.122–5.993) to test positive for anti-CHIKV IgM antibodies were nearly 5 and 2.6 as large compared with those from Awash Arba HC. The odds for participants who reported back pain to be seropositive for anti-CHIKV IgM antibodies were about two times larger compared to those without back pain (AOR = 1.785; 95% CI: 1.078–2.954), while the odds of participants with a rash to test positive for anti-CHIKV IgM antibodies were about 10 times larger compared to those without these symptoms, although the statistical significance of this finding was marginally insignificant (AOR = 10.124; 95% CI: 0.954–107.405) (Table 3).

Table 3.

Seroprevalence, and univariable and multivariable logistic regression analyses of sociodemographic factors and clinical features for association with anti-CHIKV IgM antibody in patients with acute febrile illness seeking healthcare in a malaria-endemic area in the Afar Region, Northeast Ethiopia (N = 368).

Variable		N tested	N +ve (%)	COR (95% CI)	AOR (95% CI)
Sex	Male	164	76 (46.3)	1	1
Sex	Female	204	100 (49.0)	1.113 (0.738–1.681)	1.527 (0.885–2.636)
Age (years)	5–10	35	22 (62.9)	1	1
	11–20	94	52 (55.3)	0.732 (0.330–1.624	0.809 (0.315–2.074)
	21–30	137	63 (46.0)	0.503 (0.234–1.079)	0.827 (0.285–2.397)
	31–40	49	22 (44.9)	0.481 (0.198–1.169)	0.627 (0.177–2.222)
	41–50	28	11 (39.3)	0.382 (0.137–1.063)	0.427 (0.109–1.672)
	51 and more	25	6 (24.0)	0.187 (0.059–0.587)	0.271 (0.060–1.222)
Education status	Illiterate	168	80 (47.6)	1	1
	Grade 1–8th	118	58 (49.2)	1.063 (0.664–1.703)	0.540 (0.263–1.111)
	Grade 9–12th	62	30 (48.4)	1.031 (0.576–1.847)	0.818 (0.348–1920)
	College and above	20	8 (40.0)	0.733 (0.285–1.886)	0.223 (0.117–1.200)
Resident	Urban	198	94 (47.5)	1	1
Resident	Rural	170	82 (48.2)	1.031 (0.684–1.554)	0.971 (0.519–1.816)
Occupation	Agro/pastoralist	41	21 (51.2)	1	1
	Government/private worker	129	62 (48.1)	0.881 (0.436–1.780)	1.028 (0.427–2.476)
	House wife	104	42 (40.4)	0.645 (0.312–1.334)	0.602 (0.237–1.528)
	Students	94	51 (54.3)	1.130 (0.542–2.354)	0.922 (0.304–2.800)
Marital status	Married	223	90 (40.4)	1	1
Marital status	Single	145	86 (59.3)	2.154 (1.407–3.298)	2.340 (1.141–4.964)
Religion	Muslim	226	111 (49.1)	1	1
Religion	Christians	142	65 (45.8)	0.874 (0.574–1.332)	1.309 (0.663–2.586)
Ethnicity	Afar	177	89 (50.3)	1	1
Ethnicity	Others	185	85 (45.9)	0.840 (0.556–1.270)	0.700 (0.345–1.419)
Districts^a	Amibara	201	88 (43.8)	1
	Awash Sebat	126	63 (50.0)	1.284 (0.821–2.007)
	Harunka	41	25 (61.0)	2.006 (1.010–3.986)
Health Institute	Awash Arba HC	165	63 (38.2)	1	1
	Awash Sebat HC	126	63 (50.0)	1.619 (1.012–2.590)	1.657 (0.879–3.121)
	Sidafage HC	36	25 (69.4)	3.680 (1.694–7.992)	4.927 (1.975–12.29)
	Andido HC	41	25 (61.0)	2.530 (1.254–5.103)	2.593 (1.122–5.993)
Illness duration	⩽5 days	345	165 (47.8)	1	1
Illness duration	>5 days	23	11 (47.8)	1.00 (0.430–2.328)	0.781 (0.282–2.168)
Joint pain	No	106	43 (40.6)	1	1
Joint pain	Yes	262	133 (50.8)	1.510 (0.956–2.386)	1.488 (0.878–2.522)
Headache	No	34	18 (52.9)	1	1
Headache	Yes	334	158 (47.3)	0.798 (0.393–1.618)	0.921 (0.414–2.052)
Back pain	No	183	83 (45.4)	1	1
Back pain	Yes	185	93 (50.3)	1.218 (0.809–1.834)	1.785 (1.078–2.954)
Chills	No	201	89 (44.3)	1	1
Chills	Yes	167	87 (52.1)	1.368 (0.906–2.066)	1.121 (0.661–1.905)
Nausea/vomiting	No	207	98 (47.3)	1	1
Nausea/vomiting	Yes	161	78 (48.4)	1.045 (0.692–1.578)	1.056 (0.660–1.690)
Malaise	No	287	142 (49.5)	1	1
Malaise	Yes	81	34 (42.0)	0.739 (0.449–1.216)	0.800 (0.444–1.438)
Muscle pain	No	287	136 (47.4)	1	1
Muscle pain	Yes	81	40 (49.4)	1.083 (0.661–1.774)	0.980 (0.537–1.786)
Rash	No	362	171 (47.2)	1	1
Rash	Yes	6	5 (83.3)	5.584 (0.646–48.278)	10.124 (0.954–107.405)

COR: crude odds ratio; AOR: adjusted odd ratio; CI: confidence interval.

Excluded from multiple logistic regression model due to collinearity.

Factors and clinical features associated with malaria infection

The study found that among the malaria-suspected participants, the rate of P. falciparum was 11.0% in males versus 7.3% in females. Mixed infections were only found among males, with a rate of 1.2%. P. falciparum was detected in all age groups, with relatively high rates in the 31–40 years (14.3%) and 21–30 years (12.4%) age groups. P. vivax positivity rate was similar in both male and female sexes but was only detected among patients aged 20 years or younger.

The study revealed that malaria infection was diagnosed relatively more frequently in male patients (12.8%) than in female patients (8.8%). Additionally, patients aged 21–30 years (13.1%) and 31–40 years (14.3%) had a higher likelihood of malaria, but no statistically significant differences were observed in gender and age categories. The univariable logistic regression analysis demonstrated that the odds for febrile patients from Awash Sebat HC to test positive for malaria were about 0.2 times lower compared to those from Awash Arba HC (COR = 0.202, 95% CI: 0.076–0.540). Conversely, the multivariable logistic regression analysis showed that the odds for febrile patients from Andido HC were to have plasmodium infection were about 0.1 as low compared with those from Awash Arba HC (AOR = 0.135, 95% CI: 0.022–0.837). Regarding clinical features, the odds for participants with joint pain (AOR = 0.226, 95% CI: 0.093–0.551) and chills (AOR = 0.115, 95% CI: 0.033–0.398) having malaria infection were nearly 0.2 and 0.1 times lower compared to those participants without joint pain and chills, respectively (Table 4).

Table 4.

Prevalence and univariable and multivariable logistic regression analyses of sociodemographic factors and clinical features for association with malaria infection in patients with acute febrile illness seeking healthcare in a malaria-endemic area in the Afar Region, Northeast Ethiopia (N = 368).

Variables		N tested	N +ve (%)	COR (95% CI)	COR (95% CI)
Sex	Male	164	21 (12.8)	1	1
Sex	Female	204	18 (8.8)	0.659 (0.338–1.283)	0.493 (0.193–1.256)
Age (years)	5–10	35	2 (5.7)	1	1
	11–20	94	9 (9.6)	1.747 (0.358–8.516)	2.517 (0.380–16.688)
	21–30	137	18 (13.1)	2.496 (0.551–11.308	1.895 (0.242–14.821)
	31–40	49	7 (14.3)	2.750 (0.535–14.124)	3.730 (0.353–39.358)
	41–50	28	2 (7.1)	1.269 (0.167–9.628)	1.553 (0.113–21.285)
	51 and more	25	1 (4.0)	0.687 (0.059–8.026)	0.615 (0.028–13.695)
Education status	Illiterate	168	17 (10.1)	1	1
	Grade 1–8th	118	10 (8.5)	0.822 (0.362–1.866)	1.625 (0.488–5.402)
	Grade 9–12th	62	8 (12.9)	1.316 (0.537–3.223)	1.823 (0.489–6.793)
	College and above	20	4 (20.0)	2.221 (0.665–7.409)	3.084 (0.581–16.366)
Resident	Urban	198	17 (8.6)	1	1
Resident	Rural	170	22 (12.9)	1.583 (0.811–3.090)	1.041 (0.408–2.655)
Occupation	Agro/pastoralist	41	3 (7.3)	1	1
	Government/private worker	129	16 (12.4)	1.793 (0.495–6.494)	2.107 (0.434–10.222)
	House wife	104	12 (11.5)	1.652 (0.441–6.187)	2.342 (0.444–12.367)
	Students	94	8 (8.5)	1.178 (0.296–4.687)	0.418 (0.054–3.244)
Marital status	Married	223	22 (9.9)	1	1
Marital status	Single	145	17 (11.7)	1.213 (0.620–2.373)	1.691 (0.486–5.887)
Health institute	Awash Arba HC	165	28 (17.0)	1	1
	Awash Sebat HC	126	5 (4.0)	0.202 (0.076–0.540)	0.337 (0.085–1.339)
	Sidafage HC	36	4 (11.1)	0.611 (0.200–1.867)	0.904 (0.219–3.734)
	Andido HC	41	2 (4.9)	0.251 (0.057–1.100)	0.135 (0.022–0.837)
Duration of illness	⩽5 days	345	39 (11.3)	1
Duration of illness	>5 days	23	0 (0.0)	Omitted
Headache	No	34	1 (2.9)	1	1
Headache	Yes	334	38 (11.4)	4.236 (0.563–31.869)	3.262 (0.336–31.701)
Joint pain	No	106	19 (17.9)	1	1
Joint pain	Yes	262	20 (7.6)	0.378 (0.193–0.742)	0.226 (0.093–0.551)
General back pain	No	183	25 (13.7)	1	1
General back pain	Yes	185	14 (7.6)	0.517 (0.260–1.031)	0.479 (0.203–1.126)
Chills	No	201	34 (16.9)	1	1
Chills	Yes	167	5 (3.0)	0.152 (0.058–0.397)	0.115 (0.033–0.398)
Nausea/vomiting	No	207	23 (11.1)	1	1
Nausea/vomiting	Yes	161	16 (9.9)	0.883 (0.450–1.732)	0.549 (0.238–1.267)
Diarrhea	No	319	34 (10.7)	1	1
Diarrhea	Yes	49	5 (10.2)	0.952 (0.354–2.566)	1.079 (0.303–3.846)
Malaise	No	287	34 (11.8)	1	1
Malaise	Yes	81	5 (6.4)	0.489 (0.185–1.295)	0.371 (0.112–1.228)
Muscle pain	No	287	33 (11.5)	1	1
Muscle pain	Yes	81	6 (7.4)	0.616 (0.248–1.525)	2.308 (0.700–7.608)
Sore throat	No	326	38 (11.7)	1	1
Sore throat	Yes	42	1 (2.4)	0.185 (0.025–1.383)	0.276 (0.030–2.557)

COR: crude odds ratio; AOR: adjusted odd ratio; CI: confidence interval.

Discussion

This study presents the seroprevalence of CHIKV infection among febrile patients suspected of malaria in the Afar Region of Ethiopia during non-outbreak periods. In the study, over half (51.4%) of the participants were found to have IgM and/or IgG antibodies against CHIKV. The finding suggests that CHIKV is widespread in the area, and its infection has been increasingly causing febrile illness in the study area. The prevalence of CHIKV in this study is higher than studies conducted among febrile patients at Gonder Specialized Hospital in northern Ethiopia (23%).²⁷ Patients with clinical features suggesting arboviral infections were referred to Kassala Hospital, Eastern Sudan (43.7%).³² Patients suspected of CHIKV infection attended hospitals in West Bengal, India (36.9%),³³ and patients with febrile syndrome attended emergency services at the Jorge Cristo Sahium Hospital in the municipality of Villa del Rosario, Colombia (29.9%).³⁴ However, it is lower than the prevalence of CHIKV immediately after the end of the 2014 outbreak in HIV-infected people followed up in clinical cohorts at the university hospitals of Guadeloupe and Martinique of France’s Caribbean Islands (61.0%).³⁵ This difference in infection rates may be due to changing climate and weather patterns that facilitate the spread of vector-borne diseases³⁶ and differences in diagnostic methods and study participants. For example, the study in Brazil used specific and sensitive molecular diagnostic techniques. In contrast, the study in France’s Caribbean islands recruited HIV-positive adult patients and diagnosed only through anti-CHIKV IgG antibodies.

The results of the study showed that almost half of the participants (47.8%) had anti-CHIKV-specific IgM antibodies, indicating acute CHIKV infection. Additionally, the prevalence of anti-CHIKV IgG was only 6.3%, suggesting a low number of previous positive CHIKV infections. The detection of high IgM antibodies in participants of this study emphasizes the importance of considering CHIKV as one of the major emerging causes of febrile illness in the Afar region. This seroprevalence is higher than previously reported studies in Ethiopia (22.5%)²⁷ and Tanzania (12.9%).³⁷ Back pain was significantly associated with a high seroprevalence of IgM to CHIKV infection, consistent with previous reports from Tanzania and Ethiopia.^27,37 In contrast, only 10.6% of these malaria-suspected study participants were diagnosed with malaria parasite infection, despite the region being known as a malaria-endemic area of the country.³⁸ Our findings indicate that the prevalence of recent or acute CHIKV infection is higher than that of malaria infection among malaria-suspected patients. This low prevalence of malaria cases in the study area could be attributed to various ongoing preventive measures, such as the use of insecticide-treated bed nets, residual spraying, and the adoption of artemisinin-based combination therapies administered by the MOH to eliminate and control the spread of the disease.²⁹

CHIKV infection shares clinical symptoms with malaria, making it easy to be misdiagnosed in tropical countries, particularly countries with limited laboratory facilities. This misdiagnosis can lead to a lack of necessary prevention and control interventions for CHIKV and other arbovirus infections, ultimately contributing to the spread of the disease.³⁹ Despite various outbreaks of CHIKV infection being reported in this study area and the surrounding areas, CHIKV and other arbovirus infections have not been considered as a major public health problem in the study area or the country at large.^40,41 However, this study’s findings underscore the significance of CHIKV infection as a public health concern among febrile patients suspected of having malaria in the study area. The study also found that participants with joint pain and chills were less likely to test positive for malaria infections. Several previous studies indicated that joint pain is among the most common clinical symptoms of CHIKV infection.^42,43 These findings support the need for further studies to standardize clinical indicators for the diagnosis of CHIKV infection in countries like Ethiopia, where both malaria and CHIK are widespread but adequate laboratory diagnostic tools are often lacking.

The rates of acute CHIKV infection were fairly similar between males and females in this study area. The finding is in contrast to those from studies conducted in Gondar, Ethiopia, and Anyigba, Nigeria, which showed that males had twice as high a seroprevalence of anti-CHIKV IgM compared to females.^17,27 Furthermore, this study found a lower seropositivity rate for anti-CHIKV IgM antibodies among individuals aged 51 and above. The result aligns with a recent study conducted in India, which reported a higher prevalence among younger individuals.⁴⁴ However, it differs from a study conducted in Nigeria, which reported a higher anti-CHIK IgM antibody in older patients.¹⁷ The difference in the findings may imply that individuals in the age group below 50 years are more exposed to day-biting mosquitoes because of movement to forest areas for various activities like tending animals, in the case of Afar pastoralists. Previous studies have shown a negative association between age and antibody responses to mosquito saliva. The biomarkers for exposure to Aedes mosquito saliva to predict disease risk have shown a negative association between age and antibody responses to Ae. aegypti salivary gland extract (SGE).⁴⁵ Additionally, single individuals were more likely to test positive for anti-CHIKV IgM compared to those who were married individuals, which could be attributed to the age distribution within the single group, with 70% of them being under 18 years old. This result is supported by a report from Mato Grosso, Brazil, which stated that being single is a risk factor for CHIKV infection.⁴⁶ However, in contrast to this study’s findings, married individuals, either currently or previously (divorced or widowed), were associated with an increased risk of arboviral diseases, including CHIK and dengue.⁴⁷

Despite the similar prevalence of anti-CHIKV IgM antibodies between males and females and lower prevalence among older individuals in this study, other studies have highlighted that women and older individuals are at a higher risk of experiencing chronic symptoms like chronic joint pain, swelling, and headache after contracting the disease.^5,48 Therefore, it is crucial to pay attention to females and elders when implementing prevention and control interventions.

Among the study participants diagnosed with malaria, 84.6% exhibited P. falciparum infection, 10.3% had P. vivax, and 5.1% had mixed infections. This finding is in agreement with studies reporting a higher incidence of P. falciparum infection among febrile patients attending public health institutes in Ethiopia.^49,50 The prevalence of malaria cases in this area should not be ignored as low and insignificant, as it can contribute to the ongoing transmission of the disease and potentially cause future outbreaks in the community. P. falciparum, the most severe form of malaria, is widely distributed in the study area and its surroundings similar to previously reported studies.^40,41 The result calls attention to further strengthening malaria control and prevention programs in the study area and its surroundings. This study also found that co-infection with malaria and CHIKV infections was in 4.3% of study participants, indicating these independent mosquito-borne microorganisms coexist in the area. This may potentially exacerbate the illness and make it more difficult to clinically identify malaria and CHIK. Furthermore, patients with this co-infection may be at risk of developing severe malaria or severe CHIK with complications, as reported in cases of malaria and dengue co-infection.⁵¹

It is important to note that there are certain limitations to this study. First, due to shortages of available diagnostic resources, such as molecular techniques and reagents, the diagnosis of patients with febrile illnesses who were seropositive for anti-CHIKV IgM antibodies was not confirmed virologically. This made it difficult to identify the circulating lineage of CHIKV in Ethiopia. Second, patients with early onset illness might have false-negative results from ELISA for anti-CHIKV IgM antibodies because virus-specific IgM antibodies can be detectable from 3 to 8 days after the onset of symptoms. Third, the patient has a bias toward remembering the exact dates the illness started and providing that information. It has been observed that the community usually waits for illness to subside by its own or with the traditional healers and only seeks medical care when it gets severe. Fourth, of the patients in this study, only 210 (57.1%) participants were identified to have at least one of the aforementioned febrile illness agents. However, the remaining 158 (42.9%) febrile patients’ infections remained undiagnosed.

Conclusion

The study findings indicated a high prevalence of acute CHIKV infection (47.8%), low previous infections (6.3%), and a relatively low prevalence of malaria (10.6%) among patients with acute febrile illness seeking healthcare in the studied districts of the Afar Region. This suggests an ongoing silent CHIKV infection outbreak, emphasizing the need for intervention actions, further screening using sensitive diagnostic tools, and continuous surveillance in the study area and other potentially affected regions of the country. The findings also provide valuable information regarding the potential of some common clinical symptoms for the diagnosis of CHIKV infection in areas where malaria coexist, though further studies are necessary. Additionally, the study revealed that 54.1% of participants were diagnosed with acute CHIKV infection, malaria, or co-infection, underscoring the importance of determining the causative agents of febrile illness-related symptoms in the remaining 45.9% of patients.

Footnotes

Acknowledgements

We are grateful to study participants, data collectors, and Health Bureau of Amibara district, Harunka district, and Awash Sebat city administration and their respective health centers staff members. We are also grateful to Addis Ababa University of Ethiopia and LaRuke Development Inc. of USA for their financial and material supports.

Author contributions

Biruk Zerfu: Conceptualization, Formal analysis, Writing – original draft, Writing – review and editing. Gezahegne Mamo: Conceptualization, Writing-review and editing. Tesfu Kassa: Writing – original draft, Writing – review and editing. James W Larrick: Formal analysis, Writing-review and editing. Mengistu Legesse: Conceptualization, Formal analysis, Writing – original draft, Writing – review and editing.

Availability of data and materials

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

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was financially supported by Addis Ababa University and LaRuke Development Inc., USA. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Ethics approval

Informed consent

Written informed consent was obtained from all participants aged 18 years and older, and from those under 12, informed consent was obtained from their legally authorized representatives. For participants between the ages of 12 and 17 years, written informed consent was obtained from the participants and their legally authorized representative. For participants without formal education, the consent or assent was read aloud in the presence of literate witnesses, and participants provided their mark with a thumbprint. Blood samples were collected under aseptic conditions by experienced laboratory technicians. Participants who tested positive for malaria were treated with appropriate anti-malaria drugs according to malaria treatment guideline standards, while nonmalaria cases were treated with alternative antibiotics or analgesics/antipyretics based on the presumptive diagnosis of the clinician.

Trial registration

Not applicable.

ORCID iD

Biruk Zerfu

References

Selhorst

Makiala-Mandanda

De Smet

, et al. Molecular characterization of chikungunya virus during the 2019 outbreak in the Democratic Republic of the Congo. Emerg Microbes Infect 2020; 9(1): 1912–1918.

Powers

Brault

Tesh

, et al. Re-emergence of Chikungunya and O’nyong-nyong viruses: evidence for distinct geographical lineages and distant evolutionary relationships. J Gen Virol 2000; 81(2): 471–479.

WHO. Chikungunya. Key facts [Internet]. World Health Organization, https://www.who.int/news-room/fact-sheets/detail/chikungunya#:~:text=Chikungunya%20causes%20fever%20and%20severe,treatment%20for%20chikungunya%20virus%20infections (2022, accessed 19 March 2023).

Mason

Haddow

AJ.

An epidemic of virus disease in Southern Province, Tanganyika Territory, in 1952–53; an additional note on Chikungunya virus isolations and serum antibodies. Trans R Soc Trop Med Hyg 1957; 51(3): 238–240.

Silva

Dermody

TS.

Chikungunya virus: epidemiology, replication, disease mechanisms, and prospective intervention strategies. J Clin Invest 2017; 127(3): 737–749.

Cao

, et al. Recent progress on chikungunya virus research. Virol Sin 2017; 32(6): 441–453.

Powers

Logue

CH.

Changing patterns of chikungunya virus: re-emergence of a zoonotic arbovirus. J General Virol 2007; 88(9): 2363–2377.

Caglioti

Lalle

Castilletti

, et al. Chikungunya virus infection: an overview. New Microbiol 2013; 36(3): 211–227.

Mourad

Makhani

Chen

LH.

Chikungunya: an emerging public health concern. Curr Infect Dis Rep 2022; 24(12): 217–228.

10.

Ross

RW.

The Newala epidemic: III. The virus: isolation, pathogenic properties and relationship to the epidemic. J Hyg 1956; 54(2): 177–191.

11.

Moreira

Bressan

Brasil

, et al. Epidemiology of acute febrile illness in Latin America. Clin Microbiol Infect 2018; 24(8): 827–835.

12.

PAHO. Preparedness and response for Chikungunya virus: introduction in the Americas. Washington, DC: Pan American Health Organization, https://iris.paho.org/bitstream/handle/10665.2/4009/chikungunyavirus.pdf?sequence=1 (2011, accessed 19 March 2023).

13.

Johnson

Russell

Goodman

CH.

Laboratory diagnosis of chikungunya virus infections and commercial sources for diagnostic assays. J Infect Dis 2016; 214(suppl 5): S471–S474.

14.

Natrajan

Rojas

Waggoner

JJ.

Beyond fever and pain: diagnostic methods for chikungunya virus. J Clin Microbiol 2019; 57(6): e00350.

15.

Chelluboina

Robin

Aswathyraj

, et al. Persistence of antibody response in chikungunya. Virus Dis 2019; 30(3): 469–473.

16.

Pereira

Manuli

Coulon

, et al. Performance evaluation of VIDAS^® diagnostic assays detecting anti-chikungunya virus IgM and IgG antibodies: an international study. Diagnostics 2023; 13(13): 2306.

17.

Omatola

Onoja

Fassan

, et al. Seroprevalence of chikungunya virus infection in five hospitals within Anyigba, Kogi State of Nigeria. Braz J Infect Dis 2020; 24(1): 1–6.

18.

Kam

Lum

Teo

, et al. Early neutralizing IgG response to Chikungunya virus in infected patients targets a dominant linear epitope on the E2 glycoprotein. EMBO Mol Med 2012; 4(4): 330–343.

19.

Mekonnen

Kroos

Yellow fever and other arboviral diseases. In: Berhane

Haile Mariam

Kroos

(eds.) Epidemiology and ecology of health and disease in Ethiopia. 1st ed. Addis Ababa, Ethiopia: Shama Books, 2006, pp. 635–645.

20.

Serie

Andral

Lindrec

, et al. [Epidemic of yellow fever in Ethiopia (1960–1962). Preliminary study]. Bull World Health Organ 1964; 30: 299–319.

21.

Mengesha Tsegaye

Tayachew

Belay

, et al. A35 The first laboratory confirmation of chikungunya outbreak in Ethiopia. Virus Evol 2019; 5(Supplement_1): vez002.034.

22.

EPHI. Arboviral disease vectors surveillance and control guideline. Public Health Entomology Research Team. Ethiopia Public Health Institute, Addis Ababa Ethiopia, https://ephi.gov.et/wp-content/uploads/2021/12/Arboviral-Disease-Vectors-Surveillance-and-Control-Guideline.pdf (2021, accessed 18 March 2023).

23.

Geleta

Tesfaye

Ayigegn

Epidemiological description of chikungunya virus outbreak in Dire Dawa administrative city, Western Ethiopia, 2019. Int J Clin Exp Med 2020; 6(3): 41.

24.

Longbottom

Walekhwa

Mwingira

, et al. Aedes albopictus invasion across Africa: the time is now for cross-country collaboration and control. Lancet Global Health 2023; 11(4): e623–e628.

25.

Althouse

Guerbois

Cummings

DAT

, et al. Role of monkeys in the sylvatic cycle of chikungunya virus in Senegal. Nat Commun 2018; 9(1): 1046.

26.

Macro

Central statistical agency Addis Ababa, Ethiopia. Addis Ababa, Ethiopia: Central Statistical Agency, 2006.

27.

Ferede

Tiruneh

Abate

, et al. Evidence of chikungunya virus infection among febrile patients in northwest Ethiopia. Int J Infect Dis 2021; 104: 183–188.

28.

Sullivan

LM.

Estimation from samples. Circulation 2006; 114(5): 445–449.

29.

MOH. National Malaria Guidelines. Addis Ababa: Ethiopia. Federal Democratic Republic of Ethiopia Ministry of Health, 2012.

30.

Manufacturer’s Instructions for Use (IFU). SERION ELISA classic Chikungunya Virus IgG/IgM, V148.3. Institut Virion\Serion GmbH, Würzburg, Germany: Friedrich-Bergius-Ring. www.virion-serion.de

31.

Wittke

Robb

Thu

, et al. Extinction and rapid emergence of strains of dengue 3 virus during an interepidemic period. Virology 2002; 301(1): 148–156.

32.

Mohamed

Magzoub

Mohamed

REH

, et al. Prevalence and identification of arthropod-transmitted viruses in Kassala state, Eastern Sudan. Libyan J Medic 2019; 14(1): 1564511.

33.

Chattopadhyay

Mukherjee

Nandi

, et al. Chikungunya virus infection in West Bengal, India. Indian J Med Microbiol 2016; 34(2): 213–215.

34.

Carrillo-Hernández

Ruiz-Saenz

Villamizar

, et al. Co-circulation and simultaneous co-infection of dengue, chikungunya, and zika viruses in patients with febrile syndrome at the Colombian-Venezuelan border. BMC Infect Dis 2018; 18(1): 61.

35.

Curlier

Fagour

Herrmann-Storck

, et al. Seroprevalence of chikungunya virus infection among HIV-infected adults in French Caribbean Islands of Martinique and Guadeloupe in 2015: a cross-sectional study. PLoS Negl Trop Dis 2021; 15(4): e0009267.

36.

Fouque

Reeder

JC.

Impact of past and on-going changes on climate and weather on vector-borne diseases transmission: a look at the evidence. Infect Dis Poverty 2019; 8(1): 51.

37.

Kajeguka

Kaaya

Mwakalinga

, et al. Prevalence of dengue and chikungunya virus infections in north-eastern Tanzania: a cross sectional study among participants presenting with malaria-like symptoms. BMC Infect Dis 2016; 16(1): 183.

38.

National Malaria Control Team, Ethiopian Public Health Institute, World Health Organization, Addis Ababa University, the INFORM Project. An epidemiological profile of malaria in Ethiopia. A report prepared for the Federal Ministry of Health, Ethiopia, the Roll Back Malaria Partnership and the Department for International Development, UK, 2014.

39.

Irekeola

Engku Nur Syafirah

EAR

Islam

, et al. Global prevalence of dengue and chikungunya coinfection: a systematic review and meta-analysis of 43,341 participants. Acta Trop 2022; 231: 106408.

40.

Mehari

Zerfu

Desta

Prevalence and risk factors of human brucellosis and malaria among patients with fever in malaria-endemic areas, attending health institutes in Awra and Gulina district, Afar Region, Ethiopia. BMC Infect Dis 2021; 21(1): 942.

41.

Zerfu

Medhin

Mamo

, et al. Community-based prevalence of typhoid fever, typhus, brucellosis and malaria among symptomatic individuals in Afar Region, Ethiopia. PLoS Negl Trop Dis 2018; 12(10): e0006749.

42.

Manimunda

Vijayachari

Uppoor

, et al. Clinical progression of chikungunya fever during acute and chronic arthritic stages and the changes in joint morphology as revealed by imaging. Trans R Soc Trop Med Hyg 2010; 104(6): 392–399.

43.

Chang

Encinales

Porras

, et al. Frequency of chronic joint pain following chikungunya virus infection: a colombian cohort study. Arthritis Rheumatol 2018; 70(4): 578–584.

44.

Badoni

Gupta

, et al. Dengue-chikungunya infection in the tertiary care hospital of northern India: cross-sectional latent class cluster analysis in viral infection. Heliyon 2023; 9(3): e14019.

45.

Parker

Medina

Bohl

, et al. Determinants of exposure to Aedes mosquitoes: a comprehensive geospatial analysis in peri-urban Cambodia. Acta Tropica 2023; 239: 106829.

46.

Oliveira Ferreira

Atanaka

Espinosa

, et al. Sociodemographic and sanitary profile of chikungunya virus infection in medium-sized municipality in Mato Grosso, from January to March 2018, Brasil. PSM 2021; 18(2): 94–120.

47.

Power

Vaughan

Qiao

, et al. Socioeconomic risk markers of arthropod-borne virus (arbovirus) infections: a systematic literature review and meta-analysis. BMJ Glob Health 2022; 7(4): e007735.

48.

Zingman

Paulino

Payano

MP.

Clinical manifestations of chikungunya among university professors and staff in Santo Domingo, the Dominican Republic. Revis Panamer Salud Púb 2017; 41: 1.

49.

Negatu

Abebe

Yalew

WG.

Prevalence of malaria and associated factors among malaria-suspected patients attending Hamusit Health Center, Northwest Ethiopia: a cross-sectional study. J Parasitol Res 2022; 2022: 1–7.

50.

Belay

Gelana

Gebresilassie

Malaria prevalence, knowledge, attitude, and practice among febrile patients attending Chagni Health Center, Northwest Ethiopia: a cross-sectional study. Trop Dis Travel Med Vaccines 2021; 7(1): 20.

51.

Kotepui

Milanez

GDJ

, et al. Prevalence of and risk factors for severe malaria caused by Plasmodium and dengue virus co-infection: a systematic review and meta-analysis. Infect Dis Poverty 2020; 9(1): 134.