Demographic, Clinical and Bacteriological Characteristics of Human Leptospirosis in Sri Lanka: A Retrospective Study

Research Article

J Bacteriol Mycol. 2016; 3(2): 1028.

Demographic, Clinical and Bacteriological Characteristics of Human Leptospirosis in Sri Lanka: A Retrospective Study

Karunanayake L1*, Karunanayake SAAP2, Perera KCR1, Senarath U3, Gunarathna HDNT1 and Peter D1

1National Reference Laboratory for Leptospirosis, Department of Bacteriology, Medical Research Institute, Colombo, Sri Lanka

2Department of Clinical Medicine, Faculty of Medicine, University of Colombo, Sri Lanka

3Department of Community Medicine, Faculty of Medicine, University of Colombo, Sri Lanka

*Corresponding author: Karunanayake L, Consultant Clinical Microbiologist, National Reference Laboratory for Leptospirosis, Department of Bacteriology, Medical Research Institute, Colombo, Sri Lanka

Received: June 13, 2016; Accepted: July 02, 2016; Published: July 06, 2016


The clinical and laboratory diagnosis of human leptospirosis is complex due to a variety of factors, but the Microscopic Agglutination Test (MAT) is probably the most specific laboratory test. We describe the demographic, clinical and bacteriological features of all MAT-confirmed cases in 2015 at the National Reference Laboratory in Sri Lanka. The majority of cases had acquired the infection through indirect transmission through high-risk occupations, which relate to the lower socio-economic groups, and a majority was from the economically productive age group. A second sample during follow-up was useful. The common prevalence of serogroups such as Tarassovi and Autumnalis implicate the buffaloe as an important carrier animal in Sri Lanka. There was no significant association between organ involvement and province, except for multi-organ dysfunction syndrome. Several significant associations between serovar and organ involvement were found which could be useful for future pathophysiological studies. Serogroups Australis and Cyanopteri were found significantly more commonly in Uva Province. Awareness of the disease among high risk groups and bovine vaccination programs with locally prevalent serovars should be urgently considered for this potentially fatal disease.

Keywords: Leptospirosis; Serogroups; AKI; MAT; Sri Lanka


AKI: Acute Kidney Injury; CFR: Case Fatality Rate; MAT: Microscopic Agglutination Test; MODS: Multi-Organ Dysfunction Syndrome; NHSL: National Hospital of Sri Lanka; NRL: National Reference Laboratory; SND: Standard Normal Distribution


Leptospirosis is a neglected tropical zoonosis caused by pathogenic spirochetes of the species Leptospira interrogans. It is distributed worldwide and is endemic in Southeast Asia. It is maintained by chronic renal infection in carrier animals, which excrete leptospira in their urine, contaminating the environment [1].

Human leptospirosis is an acute febrile illness with a wide spectrum of clinical manifestations, ranging from mild to severe. The infection occurs through direct or indirect exposure to urine or tissue of infected animals. Direct contact is important in specific occupations such as veterinarians, abattoir workers and animal handlers. Indirect contact is more common, and is the mode of transmission in the great majority of cases, especially in tropics [1]. Such exposure occurs in a contaminated environment especially with wet soil and water, as in rice farming, flooding after heavy rainfall, and recreational water sports [2].

Leptospirosis is a notifiable diseasein Sri Lanka [2]. It is highly endemic, with an annual incidence rate of >10/100,000 population. The Case Fatality Rate (CFR) ranges from 1.5-2.9% [2]. In 2008, Sri Lanka reported the largest outbreak of leptospirosis yet, with 7,406 notifications of suspected cases and 204 deaths, with an annual incidence rate of 35.7/100,000 population. The CFR was 2.7% [2,3].

The agricultural economy especially rice farming, warm tropical climate, and seasonal rainfalls with flooding provide an ideal environment for its indirect transmission in Sri Lanka. There are two main paddy cultivation seasons in the country, named ‘Yala’ and ‘Maha’. Yala, the minor growing season in the dry zone, occurs with the first monsoon rains in March-April. The major growing season, Maha, begins with the second inter-monsoon rain in October to November [4].

The first case of clinically suspected Weil’s disease in Sri Lanka (Ceylon) was reported in 1953 [5]. In 1959, Rajasuriyaetal reported a case series of leptospirosis with laboratory confirmation [6]. The cases were serologically confirmed at National Reference Laboratory (NRL), Medical Research Institute by complement fixation test and later isolated by guinea pig inoculation, and identified as Leptospira Icterohaemorrhagiae [7]. In 1971, Nityananda and Harvey examined the different serotypes and their reservoir hosts in Sri Lanka. Their findings suggested the existence of a diversity of serotypes maintained by different maintenance hosts such as rodents, domestic farm animals and dogs [5]. Since then, many new reference leptospira serovars from humans and animals have been added from Sri Lanka to the world literature [5,8].

In 1962, a serological survey of occupational groups was done at NRL using macro-agglutination test. Rice field workers, sewer workers, workers in coconut plantations and desiccated coconut mills, sugar cane workers, abattoir workers, fish market workers and river bathers were identified as at risk [9].

In Sri Lanka, dengue fever and leptospirosis are significant public health problems. They occur throughout the year with peaks related to rainfall, and show similar clinical manifestations in early illness. In comparison to dengue, however, leptospirosis carries a higher risk of death in Sri Lanka. In the first quarter of 2015, 1,130 clinically suspected leptospirosis cases were notified with 23 deaths (CFR 2.03%). In contrast, for the same period 12,035 clinically suspected cases of dengue fever/dengue hemorrhagic fever were notified, with 27 deaths (CFR 0.22%), showing a10-fold higher risk of death for leptospirosis [10]. Severe leptospirosis involves renal, hepatic, pulmonary, cardiac and pancreatic complications, often in combination leading to Multi- Organ Dysfunction Syndrome (MODS).

The Microscopic Agglutination Test (MAT) is the serological reference method and the most common serological technique used for confirmation [1,11]. It detects IgM and IgG agglutinating antibodies. Microscopic agglutinating antibodies appear at the end of first week of illness and reach peak levels during third to fourth week [11,12]. It is a complex test to maintain, perform and interpret, and the highest specificity is achieved by using live cultures of different serogroups of leptospires. Therefore, the use of this test is restricted to reference laboratories. While it has high specificity, the sensitivity in early illness is low especially if used for single specimens [11].

Niloofa et al analyzed the accuracy of the MAT in 255 clinically diagnosed patients, with a single acute sample, in the NRL using only the saprophytic species (Patoc -1) as the antigen. The results showed a very low sensitivity of 55.3%, specificity of 95.7%, positive predictive value of 0.95 and negative predictive value of 0.55 with a single acute sample [13]. In the same study, assuming that all tests were imperfect using the Bayesian Latent Class Modelling for a single sample, the sensitivities of MAT, IgM ELISA and Lepto check were 77.4%, 87.4% and 86.0%, and the specificities were 97.6%, 82.9% and 84.5%, respectively [13]. Accordingly, the MAT using live organisms was confirmed as the ‘serological reference’ test in our setting.

Limmathurotsakul et al (2012) questioned the accuracy of the MAT as a reference test in Thailand. In that study the MAT was not performed in Thailand [14]. The country-specific diversity of serovars and strain differences may have contributed to their final analysis. Although there are many limitations, with a pathogenic panel carefully selected by local laboratory experts with their knowledge on local patterns and strains, the MAT claims to be the best serological reference test in this complex disease.

Furthermore, MAT is the most appropriate test for epidemiological sero-surveys [11]. MAT data is useful to determine the diverse serogroups within a country. In 2012, a preliminary serological study by MAT was done using 8 pathogenic serogroups. Interestingly, Pyrogenes was the commonest serogroup, followed by Pomona, Autumnalis and Icterohaemorrhagiae [15]. In the same period leptospira cultures were identified as belonging to Pyrogenes and Autumnalis serogroups (unpublished data) [16]. MAT with a live pathogenic panel was introduced in our laboratory from 2014,and periodically the panel was changed to increase the positivity of the test.

An animal species can be a maintenance host to one serovar and an incidental host to another [17]. In rural Sri Lanka, water buffaloes are still utilized for paddy cultivation. They are used in land preparation before the monsoon. Serological studies in buffaloes in Sri Lanka had shown the presence of serogroups Autumnalis, Sejroe, Pyrogenes and Tarassovi [18,19]. Another study done in Sri Lankan murid rodents in wild, urban and urban-wild interface has shown the presence of serogroups Pomona, Autumnalis, Sejroe, and Javanica, mainly in rats and mongoose species [20]. In a local sero-survey, unvaccinated dogs were found to be positive for serogroups Canicola, Australis, Icterohaemorrhagiae and Djasiman [21]. The knowledge of the prevalent serovars and their maintenance hosts is useful to understand the epidemiology of leptospirosis and to select targeted prevention and control strategies.

Our objective is to contribute towards the knowledge of the epidemiology, clinical manifestations and laboratory aspects of human leptospirosis in Sri Lanka. We have analyzed the findings at the NRL, which is the reference center that performs MAT and processes free of charge specimens from state hospitals throughout the country. We would discuss the background prevalence of leptospira serotypes in the provinces and its implications.

Material and Methods

A retrospective study was carried out at the NRL using data for 2015. The target population consisted of clinically suspected patients from all 9 provinces in Sri Lanka whose specimens yielded a significant result for leptospirosis antibodies at the NRL during 2015. Out of the total number of 2,662 clinical records received in the laboratory, only 2,433 non-repetitive, single records were used for the analysis of data. Of these, 1,004 (41.3%) were serologically confirmed as leptospirosis.

Laboratory confirmation

MAT positivity or a significant result was based on a cut-off titre of ≥ 1/320, seroconversion, or four-fold rise in paired sera in any one of the serovars in the panel. When analyzing multiple samples in the same patient, the results of the second sample were used. To assess the prevalence of serogroups, a cut-off titre of ≥ 1/40 by MAT was used.

For the present study 13 serovars were included. The pathogenic panel included 12 serogroups:Australis, Autumnalis, Bataviae, Canicola, Cynopteri, Grippotyphosa, Hebdomadis, Icterohaemorrhagiae, Pomona, Pyrogenes, Sejroe, Tarassovi. The saprophytic serogroup was Semaranga.

Information on disease severity

The analysis of clinical information, including disease severity, was based on the information in the request form. Clinical information had been determined by specialists in internal medicine in hospitals where the patients were treated.

The hospitals belonged to different levels of care such as Teaching Hospitals, Provincial General Hospitals and Base Hospitals. The National Hospital of Sri Lanka (NHSL) is the premier healthcare provider in the country. Therefore, severely ill patients with complications are transferred there. Also, it is a popular station for patients with renal impairment. This is evident in a previous study that showed that 67.7% of severe leptospirosis with Acute Kidney Injury (AKI) was treated in the Colombo group hospitals [22].

Statistical analysis

Socio-demographic data of the patients were summarized as means or percentages, as appropriate. The clinical and serological data were entered in binary form (present=1, absent=0) according to the defined criteria, and presented as percentages. The crosstabulations were generated to compare clinical and serological data across the provinces, with SND test used for statistical significance between proportions. The association between serovars and the organ involved in severe leptospirosis, and the association between serovars and province were determined by chi-square test. Data were analyzed using SPSS software. A p value <0.05 was considered statistically significant.


In 2015, 4,402 clinically suspected leptospirosis cases were notified to the Epidemiology Unit [23] while the NRL received 2,433 specimens. Although it is not possible to make a direct comparison between notification numbers and laboratory-tested numbers as these do not refer to the same populations, it is noteworthy that the number of cases tested in 2015 constitutes over half (55%) the number of notifications.

Demographic and Laboratory results

Out of the total 2,433 non-repetitive records of clinically suspected cases of leptospirosis, 1,429 was negative and 1,004 (41.3%) yielded a significant result by MAT. For final analysis of demography and laboratory results, these 1,004 cases were analyzed.

Multiple samples were received in 179 patients (7.3%, n=2,433). In 106 patients the MAT titer was significantly higher in the second sample. Therefore, 59.2% of these patients were confirmed by MAT with the second sample.

The majority (78%) of the samples was received from wards and 19.2% from intensive care units. The rest (2.8%) were received from preliminary care units, dialysis units, clinics and postmortem samples. Fourteen (21%) postmortem blood samples were confirmed by MAT. Samples were received throughout the year. The number of samples was higher in March and in September to December compared to other months.

The distribution of cases according to province is shown in Table 1. The majority (n= 700, 69.7%) were from hospitals in Western Province and least from North Western Province (n=16, 1.6%). A majority were received from NHSL (32.7% n=328) and Colombo South Teaching Hospital (11.1% n=111).