A Hospital-Based Study on Seroprevalence of Leptospirosis Among Febrile Cases In

International Journal of Infectious Diseases 17 (2013) e394–e397

Contents lists available at SciVerse ScienceDirect

International Journal of Infectious Diseases

jou rnal homepage: www.elsevier.com/locate/ijid

A hospital-based study on seroprevalence of leptospirosis among febrile cases in

northeastern Malaysia

a a, a a a a

A.A. Noor Raﬁzah , B.D. Aziah *, Y.N. Azwany , M. Kamarul Imran , A. Mohamed Rusli , S. Mohd Nazri ,

b c c a d

A. Mohd Nikman , I. Nabilah , H. Siti Asma’ , W.M. Zahiruddin , I. Zaliha

Department of Community Medicine, School of Medical Sciences, Universiti Sains Malaysia, 16150 Kubang Kerian, Kelantan, Malaysia

Department of Anaesthesiology, School of Medical Sciences, Universiti Sains Malaysia, Malaysia

Department of Medical Microbiology and Parasitology, School of Medical Sciences, Universiti Sains Malaysia, Malaysia

Population Health & Preventive Medicine Discipline, Faculty of Medicine, Universiti Teknologi MARA, Malaysia

A R T I C L E I N F O S U M M A R Y

Article history: Objective: To determine the seroprevalence of leptospirosis among febrile inpatient cases in

Received 8 May 2012

northeastern Malaysia.

Received in revised form 21 November 2012

Methods: A hospital-based cross-sectional study was conducted among 999 febrile cases admitted to 10

Accepted 14 December 2012

hospitals in northeastern Malaysia. A survey using a proforma sheet was used to obtain socio-

Corresponding Editor: Eskild Petersen,

demographic and occupational information. Serum samples were screened for leptospirosis by IgM

Skejby, Denmark

enzyme-linked immunosorbent assay test (IgM ELISA) and conﬁrmed by microscopic agglutination test

(MAT).

Keywords: Results: There was an equivalent distribution of males and females in the 999 respondents enrolled in

Leptospirosis

the study. The majority were Malay (94.7%) and their mean age was 39.4 (standard deviation 17.6) years.

Seroprevalence

The overall seroprevalence of leptospirosis was 8.4% (95% conﬁdence interval (CI) 6.8–10.3) (n = 84). The

Microscopic agglutination test

high-risk occupational group was found to have a higher seroprevalence, which was 56% (95% CI 45.3–

Febrile cases

66.1) (n = 47). The predominant serogroup was Sejroe (82.1%, 95% CI 72.6–88.8) (n = 69).

Conclusions: This study revealed a possible high seroprevalence of leptospirosis among febrile cases,

indicating the need to review the importance of adding leptospirosis to the case investigation of febrile

illness, especially among high-risk occupational groups in Malaysia, as well as in other endemic

countries.

1. Introduction were positive using the sensitized erythrocyte lysis test. High

antibody rates (23.2% to 32.6%) were found among oil palm estate

Leptospirosis is reported as a frequent cause of undifferentiated and rubber estate workers and hospital staff. Moderately high rates

febrile illness in developing countries. Multicenter investigations (13.0% to 17.9%) were observed in laborers, the army, tin miners (of

in Indian hospitals have indicated that leptospirosis accounts for open cast, dredge, and gravel pump mines), farmers, and paddy

12.7% of acute febrile illnesses. Leptospirosis is the cause of a planters. Moderate rates (11.6% to 12.2%) were found among shop

signiﬁcant proportion of cases of non-hepatitis A and E jaundice, owners, policemen, and veterinary staff, and low rates (1.5% to

non-malarial febrile illnesses, and non-dengue hemorrhagic fever 9.4%) were detected in school teachers, housewives, ofﬁce workers,

2 4

in Southeast Asia. and tin miners of underground or lode mines.

Leptospirosis was formerly considered to be a primarily Although leptospirosis cases have been reported in Malaysia

occupational disease, and it has been associated with activities since the 1920s, the actual disease burden in the country remains

such as mining, sewer maintenance, livestock farming and underestimated. This is partly because it was not on the list of

butchering, veterinary medicine, and military maneuvers. A notiﬁable diseases under the Prevention and Control of Commu-

survey for leptospiral antibodies was conducted throughout West nicable Diseases Act 1988 until recently. In addition, notiﬁed

Malaysia from 1961 through 1971 on 18 occupational groups. A cases only cover clinical cases, however a seroprevalence study

total of 4646 serum samples were tested, of which 592 (12.7%) can better estimate disease burden by detecting both clinical and

subclinical phases. Thirteen serogroups have reportedly been

identiﬁed in Malaysia, namely Australis, Autumnalis, Bataviae,

Canicola, Celledoni, Grippotyphosa, Hebdomadis, Icterohaemor-

* Corresponding author. Tel.: +60 (0)9 7676633; fax: +60 (0)9 7676654.

E-mail address: [email protected] (B.D. Aziah). rhagiae, Javanica, Pomona, Pyrogenes, Sejroe, and Tarassovi.

A.A.N. Raﬁzah et al. / International Journal of Infectious Diseases 17 (2013) e394–e397 e395

The objective of this study was to determine the seroprevalence of titer result was taken as the last dilution that showed <50% of free

leptospirosis among febrile cases in northeastern Malaysia. leptospires compared to control wells. A positive MAT was deﬁned

as a MAT titer of 1:400 in a single, acute specimen. This titer was

2. Methods speciﬁed by the IMR as the national reference laboratory, and it

was based on the background exposure and seroprevalence of

A hospital-based cross-sectional study was conducted in 10 leptospirosis in the country. This cut-off MAT titer has also been

7–13

hospitals in northeastern Malaysia. The proportionate sampling applied in several studies on leptospirosis.

method based on the proportion of the number of beds in each In this study, the sensitivity, speciﬁcity, positive predictive

study hospital was used, and a total of 1200 respondents were value, and negative predictive value between the MAT and ELISA

estimated. could not be calculated due to the fact that only indeterminate and

Cases were febrile adults aged 18 years and older, admitted to positive ELISA results (n = 111) were subjected to the MAT.

the medical wards with an oral temperature above 37 8C. HIV The statistical analysis was performed using PASW Statistics 18

patients, patients on immunosuppressive drugs, and patients with (formerly SPSS Statistics). All continuous variables are described

autoimmune diseases were excluded from this study. using the mean and standard deviation (SD), whereas for

Informed written consent was obtained from all subjects based categorical variables, frequencies and percentages are presented.

on the approved study protocol, by the Research Ethics Committee The seroprevalence of leptospirosis was computed using positive

(Human), Universiti Sains Malaysia (Reference No. USMKK/PPP/ MAT over total samples and is presented as a percentage with 95%

JEPeM (219.3.(03))) and the Medical Review and Ethics Committee, conﬁdence interval (CI).

National Institute of Health, Ministry of Health Malaysia (NMRR-

10-318-5389). A patient proforma sheet was used to collect

3. Results

sociodemographic and occupation data. For this study, occupation

types with high and moderate antibody rates were grouped into a

A total of 999 subjects were recruited to participate in this

high-risk occupation group and those with low antibody rates

study. The majority of study subjects were Malay, and their mean

were grouped into a low-risk occupation group, as stated in Tan.

age was 39.4 (SD 17.6) years. Thirty-six percent of the subjects

Single venous samples of 5–8 ml were collected from the study

were grouped into the high-risk occupational group (Table 1).

subjects. The samples were centrifuged and stored at À20 8C. The

Eighty-eight samples tested positive by IgM ELISA and 23

PanBio Leptospira IgM ELISA test (PanBio, Queensland, Australia)

samples were indeterminate. Eighty-four respondents were

was used for the qualitative detection of IgM antibodies to

conﬁrmed positive for leptospirosis and the overall seroprevalence

Leptospira. All serology samples were analyzed in the Microbiolo-

was 8.4% (95% CI 6.8–10.3) (Table 2). The seroprevalence of

gy Laboratory of Universiti Sains Malaysia. One operator-trained

leptospirosis in the high-risk occupational group was 56% (95% CI

laboratory technician performed the ELISA test to avoid bias. A

45.3–66.1) (n = 47) and in the low-risk group was 44% (95% CI

score of less than 9 units indicates a negative result or no

33.8–54.6) (n = 37).

detectable IgM antibodies, a score of 9–11 units indicates an

Table 3 shows the level and the frequency of the MAT titer from

indeterminate result, and a score of more than 11 units indicates a

a total of 111 IgM ELISA (indeterminate and positive) results. The

positive result, i.e. the presence of Leptospira-speciﬁc IgM

cut-off point for a positive MAT was a titer of 1:400, giving a total

antibodies.

number of 84 positive MAT cases. The MAT titer range was from

The microscopic agglutination test (MAT) was performed

1:400 to 1:1600.

among positive and indeterminate samples of the ELISA test.

The highest frequency of leptospirosis cases was found among

The frozen sera were sent to the Institute of Medical Research

agricultural workers, followed by housewives or unemployed and

(IMR), Malaysia, and all were tested for the presence of anti-

outdoor workers. None of the leptospirosis cases were noted

leptospiral antibodies using the MAT following standard methods.

among the retired. Figure 1 shows the frequency of leptospirosis

A panel battery of 18 live reference serovars, representing 18

cases based on type of occupation.

serogroups, were used as antigen. Leptospira reference cultures

Table 4 shows the distribution of serogroups that were

were obtained from the World Health Organization (WHO)

determined by the positive MAT among the 84 conﬁrmed

Collaborating Centre, Queensland, Australia, with three additional

leptospirosis cases. Among these, eight serogroups were identiﬁed

local pathogenic serovars (unidentiﬁed). The serovars included in

and the most common serogroup was Sejroe, which contributed

the antigen panel were Leptospira biﬂexa serovar Patoc, Leptospira

interrogans serovar Australis, L. interrogans serovar Autumnalis,

Leptospira borgpetersenii serovar Ballum, L. interrogans serovar Table 1

Sociodemographic characteristics of respondents (n = 999)

Bataviae, L. interrogans serovar Canicola, Leptospira weilii serovar

Celledoni, L. interrogans serovar Hebdomadis, Leptospira kirschneri Variable

serovar Cynopteri, L. interrogans serovar Grippotyphosa, L. inter-

Age, years

rogans serovar Icterohaemorrhagiae, L. borgpetersenii serovar

Mean (SD) 30.4 (17.6)

Javanica, L. interrogans serovar Pomona, L. interrogans serovar Range 18–94

Pyrogenes, L. borgpetersenii serovar Tarassovi, L. interrogans serovar Sex, n (%)

Female 456 (45.6)

Hardjo, L. borgpetersenii serovar Sejroe, and L. interrogans serovar

Male 543 (54.4)

Djasiman.

Race, n (%)

Live Leptospira cell suspensions representing the 18 serovars

Non-Malay 53 (5.3)

were added to serially diluted serum specimens in microtiter Malay 946 (94.7)

Occupational classiﬁcation, n (%)

plates and incubated at 30 8C for 2 h. Each serum was tested at

Low-risk occupational group 638 (63.9)

three dilutions, 1:25, 1:50, and 1:100, and when agglutination was b

High-risk occupational group 361 (36.1)

observed at the 1:100 dilution, it was titrated to determine the

SD, standard deviation.

highest agglutination titer. Agglutination was examined using a

Low-risk: indoor worker, teacher, student, housewife/unemployed, retired,

dark-ﬁeld microscopy at a magniﬁcation of Â100. Positive

production operator, and others.

agglutination was considered when the approximate numbers of b

High-risk: agriculture worker, military, medical personnel, and outdoor

free leptospires was <50% of leptospires in the control wells. The worker.

e396 A.A.N. Raﬁzah et al. / International Journal of Infectious Diseases 17 (2013) e394–e397

Table 2

ELISA vs. MAT results

ELISA MAT Total

Negative Positive

frequency (%) frequency (%)

Indeterminate 7 (30.4) 16 (69.6) 23 (100.0)

Positive 20 (22.7) 68 (77.3) 88 (100.0)

Total 27 (24.3) 84 (75.7) 111 (100.0)

ELISA, enzyme-linked immunosorbent assay; MAT, microscopic agglutination test.

82.1% (95% CI 72.6–88.8) (n = 69) of the leptospirosis cases in the

Figure 1. Occupation types of the leptospirosis cases conﬁrmed by MAT (n = 84).

northeastern state of Malaysia.

4. Discussion We found that agriculture and outdoor workers contributed

about 50% of the leptospirosis cases. This ﬁnding is in keeping with

9,20,21

This study of febrile inpatient cases from 10 hospitals in some previous studies. However, a study by Myint et al. in

northeastern Malaysia has an adequate coverage for publication, Thailand showed that more than 80% of conﬁrmed leptospirosis

and a hospital-based study is the best method to capture febrile cases were farmers and agricultural workers, which is higher than

cases. Seroprevalence data from a retrospective study of human in our study. They conducted passive surveillance during the

leptospirosis in Malaysia based on the MAT showed a 13% overall rainy season and paired serum samples were taken from the

prevalence of infection for the period 1983 to 1998; this has been suspected cases. Besides, more than 75% of the subjects worked in

5,14–16

repeatedly cited in recent reviews and studies. the agricultural sector and nearly 60% of the land of the province

In this study, leptospirosis was conﬁrmed in 84 febrile subjects was used for agriculture. In the current study, employment in the

(8.4%), which is in line with other tropical areas. A similar agricultural sector was 50%. Unexpectedly, 20% of the cases were

prevalence of leptospirosis among febrile patients has been in the group of housewives and unemployed. This could be the

reported previously in urban Bangladesh, which is also a tropical result of various factors such as walking barefoot, the presence of

climate area. In that study, the seroprevalence of leptospirosis was rats in the house, previous occupation, type of water supply, and

calculated based on deﬁnite and probable diagnoses in which a waste disposal, which were not explored in the current study.

deﬁnite diagnosis was deﬁned as patients with either a single MAT In this study, leptospirosis cases reacted to a broad range of

titer 1:800 for any serovar at either phase, or a serovar-speciﬁc Leptospira serogroups. All of them had positive MAT results for

four-fold increase from acute phase to convalescent phase with a serogroups previously identiﬁed as common in Malaysia, but

minimum convalescent titer of 1:400. Meanwhile a MAT titer overall, the most reactive serogroup in our study population was

1:100 for any serovar in either an acute-phase or convalescent- Sejroe, which accounted about 82.1% of the leptospirosis cases; this

24 7

phase serum sample was categorized as probable leptospirosis. In is higher than in other studies done in Vietnam, rural Thailand,

25 26

our study, the prevalence of leptospirosis was based on a single Reunion Island, and Bulgaria.

MAT titer of 1:400, as we used only a single serum sample, Our ﬁnding of the most prevalent serogroup was different from

and the subjects had an acute febrile illness as well as the presence the serogroups that have been reported elsewhere in Southeast

2 27 28

of background antibody as part of the population of an endemic Asia, the Western Paciﬁc, and Israel. In India, studies done

area. among acute undifferentiated febrile illness patients and during

However, the seroprevalence in this study was lower than leptospirosis outbreaks revealed the most common infecting

7 11,29

ﬁndings in a rural Thai population, which showed seropreva- serogroup to be Icterohaemorrhagiae. However, these two

lences of 36.9% and 17.5% among adults on the Thai–Myanmar studies used a lower MAT cut-off titer for diagnosis and fewer

border. Suttinont et al. used more stringent clinical criteria, with panel antigens compared to our study.

at least two serum specimens comprised of acute and convalescent Our study had a few limitations. Firstly, our seroprevalence

samples. Ellis et al. recruited both inpatients and outpatients with estimation was limited by the use of a single acute phase serum

a temperature of 38 8C over the previous 48 h and obtained paired sample, which might have led to an underestimation in febrile

samples. In contrast, a previous study by Shaheen and Shah cases. This was mainly due to the difﬁculty in tracing cases post-

reported that the prevalence of leptospirosis among cases with hospitalization in the large geographical area covered. At least two

pyrexia of unknown origin (PUO) in the Kashmir Valley was 2.8%. serum specimens should be collected from each patient (one

This lower ﬁnding could be due to the smaller sample size and ‘acute’ sample, on admission, and a ‘convalescent’ sample, 2–4

different diagnostic criteria. In addition, climatic factors might also weeks post-discharge). Secondly, the cut-off value of the MAT

have contributed to the ﬁnding. usually depends on the baseline in the community in a particular

Table 3 Table 4

MAT titer among ELISA indeterminate and positive cases (n = 111) Identiﬁed serogroups for leptospirosis cases (n = 84)

MAT titer Frequency % Serogroup Frequency

1:50 1 0.9 Sejroe 69

1:100 5 4.5 Pyrogenes 5

1:200 21 18.9 Icterohaemorrhagiae 3

1:400 31 27.9 Celledoni 2

1:800 30 27.0 Javanica 2

1:1600 23 20.7 Australis 1

Total 111 100 Bataviae 1

Autumnalis 1

ELISA, enzyme-linked immunosorbent assay; MAT, microscopic agglutination test.

Total 84

The highest frequency was 1:400, which was 28.0% of the total MAT.

A.A.N. Raﬁzah et al. / International Journal of Infectious Diseases 17 (2013) e394–e397 e397

4 13 9. Cumberland P, Everard CO, Wheeler JG, Levett PN. Persistence of anti-lepto-

geographical area and varies from laboratory to laboratory.

spiral IgM, IgG and agglutinating antibodies in patients presenting with acute

Currently, there is no consensus regarding the standard cut-off

febrile illness in Barbados 1979–1989. Eur J Epidemiol 2001;17:601–8.

titer in view of the local scenario. The cut-off titer of 1:400 was 10. Pradutkanchana J, Pradutkanchana S, Kemapanmanus M, Wuthipum N, Silpa-

pojakul K. The etiology of acute pyrexia of unknown origin in children after a

set by the IMR based on background exposure and seroprevalence

ﬂood. Southeast Asian J Trop Med Public Health 2003;34:175–8.

of leptospirosis in the country.

11. Bharadwaj R, Bal AM, Joshi SA, Kagal A, Pol SS, Garad G, et al. An urban outbreak

In conclusion, this study revealed a possible high seropreva- of leptospirosis in Mumbai, India. Jpn J Infect Dis 2002;55:194–6.

12. Thammakumpee K, Silpapojakul K, Borrirak B. Leptospirosis and its pulmonary

lence of leptospirosis among febrile cases, indicating the need to

complications. Respirology 2005;10:656–9.

review the importance of adding leptospirosis to the case

13. Sanders EJ, Rigau-Perez JG, Smits HL, Deseda CC, Vorndam VA, Aye T, et al.

investigation of febrile illness, especially among those in the Increase of leptospirosis in dengue-negative patients after a hurricane in Puerto

high-risk occupational group in Malaysia, as well as in other Rico in 1996 [correction of 1966]. Am J Trop Med Hyg 1999;61:399–404.

14. El Jalii IM, Bahaman AR. A review of human leptospirosis in Malaysia. J Trop endemic countries.

Biomed 2004;21:113–9.

15. Lim JK, Murugaiyah VA, Ramli A, Rahman HA, Mohamed N, Shamsudin N, et al. A

Acknowledgements case study: leptospirosis in Malaysia. WebmedCentral Infectious Diseases

2011;2:WMC002764.

16. Lim VK. Leptospirosis: a re-emerging infection. Malays J Pathol 2011;33:1–5.

The authors gratefully acknowledge the help and cooperation of 17. Kendall EA, LaRocque RC, Bui DM, Galloway R, Ari MD, Goswami D, et al.

Dr Fairuz binti Amran (IMR) and all the directors and staff of the Leptospirosis as a cause of fever in urban Bangladesh. Am J Trop Med Hyg

2010;82:1127–30.

participating hospitals. We also thank all the subjects who made

18. Ellis RD, Fukuda MM, McDaniel P, Welch K, Nisalak A, Murray CK, et al. Causes of

this study a success.

fever in adults on the Thai–Myanmar border. Am J Trop Med Hyg 2006;74:

Funding: The study was funded by the Research University 108–13.

19. Shaheen R, Shah A. Prevalence of leptospirosis among the PUO cases in a tertiary

Grant Universiti Sains Malaysia (1001/PPSP/812057) and was

care hospital of Kashmir Valley. Indian Journal for the Practising Doctor 2006;3.

ethically approved by Research and Ethic Committee of Universiti

20. Sethi S, Sharma N, Kakkar N, Taneja J, Chatterjee SS, Banga SS, et al. Increasing

Sains Malaysia. trends of leptospirosis in northern India: a clinico-epidemiological study. PLoS

Negl Trop Dis 2010;4:e579.

Conﬂict of interest: No conﬂict of interest to declare.

21. Russ AK, Jali IE, Bahaman AR, Tuen AA, Ismail G. Seroepidemiological study of

leptospirosis among the indigenous communities living in periphery of Croker

References

Range Park Sabah, Malaysia. ASEAN Review of Biodiversity and Environmental

Conservation (ARBEC); January–March 2003.

1. Sehgal S, Sugunan A, Vijayachari P. Leptospirosis: disease burden estimation 22. Myint KS, Gibbons RV, Murray CK, Rungsimanphaiboon K, Supornpun W,

and surveillance networking in India. South East Asian. J Trop Med Public Health Sithiprasasna R, et al. Leptospirosis in Kamphaeng Phet, Thailand. Am J Trop

2003;34(Suppl 2):170–7. Med Hyg 2007;76:135–8.

2. Laras K, Van CB, Bounlu K, Tien NT, Olson JG, Thongchanh S, et al. The importance 23. Malaysia, Kelantan Darul Naim. Information Malaysia 2002 yearbook. Kuala

of leptospirosis in Southeast Asia. Am J Trop Med Hyg 2002;67:278–86. Lumpur: Berita Publishing; 2002, p. 841–55.

3. Bharti AR, Nally JE, Ricaldi JN, Matthias MA, Diaz MM, Lovett MA, et al. 24. Van CT, Thuy NT, San NH, Hien TT, Baranton G, Perolat P. Human leptospirosis in

Leptospirosis: a zoonotic disease of global importance. Lancet Infect Dis 2003;3: the Mekong Delta, Viet Nam. Trans R Soc Trop Med Hyg 1998;92:625–8.

757–71. 25. Paganin F, Bourdin A, Dalban C, Courtin JP, Poubeau P, Borgherini G, et al.

4. Tan DS. Occupational distribution of leptospiral (SEL) antibodies in West Leptospirosis in Reunion Island (Indian Ocean): analysis of factors associated

Malaysia. Med J Malaysia 1973;27:253–7. with severity in 147 conﬁrmed cases. Intensive Care Med 2007;33:1959–66.

5. Bahaman AR, Ibrahim AL. A review of leptospirosis in Malaysia. Vet Res Commun 26. Taseva E, Christova I, Gladnishka T. Epidemiological, clinical and serological

1988;12:179–89. features of human leptospirosis in Bulgaria in 2005. 17 European Congress of

6. World Health Organization. Leptospirosis: laboratory manual. Geneva: WHO; Clinical Microbiology and Infectious Disease, Munich, Germany, March 31–

2007. April 4, 2007.

7. Suttinont C, Losuwanaluk K, Niwatayakul K, Hoontrakul S, Intaranongpai W, 27. Berlioz-Arthaud A, Kiedrzynski T, Singh N, Yvon JF, Roualen G, Coudert C, et al.

Silpasakorn S, et al. Causes of acute, undifferentiated, febrile illness in rural Multicentre survey of incidence and public health impact of leptospirosis in the

Thailand: results of a prospective observational study. Ann Trop Med Parasitol Western Paciﬁc. Trans R Soc Trop Med Hyg 2007;101:714–21.

2006;100:363–70. 28. Bishara J, Amitay E, Barnea A, Yitzhaki S, Pitlik S. Epidemiological and clinical

8. Dassanayake DL, Wimalaratna H, Agampodi SB, Liyanapathirana VC, Piyarath- features of leptospirosis in Israel. Eur J Clin Microbiol Infect Dis 2002;21:50–1.

naa TA, Goonapienuwala BL. Evaluation of surveillance case deﬁnition in the 29. Bal AM, Bharadwaj RS, Joshi SA, Kagal AS, Arjunwadkar VP. Common infecting

diagnosis of leptospirosis, using the microscopic agglutination test: a validation leptospiral serovar in and around Pune, Mahastra. Indian J Med Res 2002;115:

study. BMC Infect Dis 2009;9:48. 14–6.