Skip to main content
  • Loading metrics

Risk factors for melioidosis in Udupi District, Karnataka, India, January 2017-July 2018

  • Akhileshwar Singh,

    Roles Conceptualization, Formal analysis, Investigation, Methodology

    Affiliation Epidemic Intelligence Service Programme, National Centre for Disease Control, Delhi, India

  • Ashok Talyan,

    Roles Conceptualization, Data curation, Formal analysis

    Affiliation Epidemic Intelligence Service Programme, National Centre for Disease Control, Delhi, India

  • Ramesh Chandra,

    Roles Investigation, Methodology

    Affiliation Epidemic Intelligence Service Programme, National Centre for Disease Control, Delhi, India

  • Anubhav Srivastav,

    Roles Conceptualization

    Affiliation Epidemic Intelligence Service Programme, National Centre for Disease Control, Delhi, India

  • Vasudeva Upadhya,

    Roles Data curation

    Affiliation District Surveillance Office, Udupi, Karnataka, India

  • Chiranjay Mukhopadhyay,

    Roles Conceptualization, Data curation, Formal analysis, Writing – review & editing

    Affiliations Kasturba Medical College, Manipal Academy of Higher Education, Manipal, Karnataka, India, Center for Emerging and Tropical Diseases, Manipal Academy of Higher Education, Manipal, Karnataka, India

  • Shyamsundar Shreedhar,

    Roles Investigation

    Affiliation Kasturba Medical College, Manipal Academy of Higher Education, Manipal, Karnataka, India

  • Deepak Sudhakaran,

    Roles Investigation

    Affiliation Kasturba Medical College, Manipal Academy of Higher Education, Manipal, Karnataka, India

  • Suma Nair,

    Roles Conceptualization, Data curation, Investigation

    Affiliation Kasturba Medical College, Manipal Academy of Higher Education, Manipal, Karnataka, India

  • Mohan Papanna ,

    Contributed equally to this work with: Mohan Papanna, Rajesh Yadav, Tanzin Dikid

    Roles Conceptualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing

    Affiliations Division of Global Health Protection, US Centers for Disease Control and Prevention, Atlanta, United States of America, Huck Institute of Life Sciences, The Pennsylvania State University, PA, United States of America

  • Rajesh Yadav ,

    Contributed equally to this work with: Mohan Papanna, Rajesh Yadav, Tanzin Dikid

    Roles Conceptualization, Data curation, Formal analysis, Writing – original draft, Writing – review & editing (RY); (TD)

    Affiliation Division of Global Health Protection, US Centers for Disease Control and Prevention, Atlanta, United States of America

  • Sujeet Kumar Singh,

    Roles Conceptualization, Data curation

    Affiliation Epidemic Intelligence Service Programme, National Centre for Disease Control, Delhi, India

  • Tanzin Dikid

    Contributed equally to this work with: Mohan Papanna, Rajesh Yadav, Tanzin Dikid

    Roles Conceptualization, Formal analysis, Methodology, Supervision, Writing – original draft, Writing – review & editing (RY); (TD)

    Affiliation Epidemic Intelligence Service Programme, National Centre for Disease Control, Delhi, India


We initiated an epidemiological investigation following the death of a previously healthy 17 year-old boy with neuro-melioidosis. A case was defined as a culture-confirmed melioidosis patient from Udupi district admitted to hospital A from January 2013—July 2018. For the case control study, we enrolled a subset of cases admitted to hospital A from January 2017- July 2018. A control was resident of Udupi district admitted to hospital A in July 2018 with a non-infectious condition. Using a matched case-control design, we compared each case to 3 controls using age and sex groups. We assessed for risk factors related to water storage, activities of daily living, injuries and environmental exposures (three months prior to hospitalization), using conditional regression analysis. We identified 50 cases with case fatality rate 16%. Uncontrolled diabetes mellitus was present in 84% cases and 66% of cases occurred between May and October (rainy season). Percutaneous inoculation through exposure to stagnant water and injury leading to breakage in the skin were identified as an important mode of transmission. We used these findings to develop a surveillance case definition and initiated training of the district laboratory for melioidosis diagnosis.


Melioidosis is caused by gram-negative intracellular bacteria Burkholderia pseudomallei, which can infect both humans and animals. This environmental saprophyte is widely distributed in soil and fresh surface water in endemic regions of South East Asia, Northern Australia, the Indian subcontinent and areas of South America [14]. Up to 20% of community-acquired sepsis in the tropics is due to melioidosis and the overall mortality varies from 20–50% depending on the availability of healthcare services [58]. In 2015, the global burden of melioidosis was 4.6 million DALYs, which was higher than other common neglected tropical diseases [9]. Estimates suggest that the extent of melioidosis global distribution is widespread, and the cases are grossly under-reported in 45 countries currently reporting [10]. This disparity is partly due to under-recognition due to its diverse clinical manifestations and the inadequacy of conventional bacterial identification methods [5,6,8,9].

Studies from Australia and South East Asia indicate that environmental and host factors determine disease acquisition. The disease incidence increases during the rainy season and adverse weather conditions like tsunamis and cyclones; agricultural workers are commonly affected [6,8,11,12]. Host factors such as the presence of one or more preexisting conditions that alter immune response (such as long standing uncontrolled diabetes mellitus or chronic renal failure), severe or penetrating injury or near-drowning are favorable for the occurrence of melioidosis [8,12,13]. In India, cases of melioidosis have been recognized from different regions, however case identification is confined to few tertiary centres due to limited diagnostic facilities [1416]. The incidence of melioidosis in India is unknown but could be substantial due to the high burden of diabetes mellitus [17] and long coastline prone to extreme weather conditions.

On July 23, 2018, the death of a boy aged 17 years was reported to Moodabettu Primary Health Centre (PHC), Udupi district of Karnataka. On July 24, a team from the district disease surveillance office visited hospital A, where the deceased received treatment and had been diagnosed with melioidosis. The surveillance team also searched for similar cases in the village and conducted an awareness program for reporting sudden deaths. National Centre for Disease Control (NCDC) was notified, and Epidemic Intelligence Service Officers (EISOs) from NCDC joined the district investigation on August 1, 2018. We investigated to describe the epidemiology and identify risk factors to inform the initial public health responses.


I. Ethics statement

This investigation was undertaken as part of an emergency public health response to identify the cause of an outbreak for early intervention. All statutory permissions were obtained from NCDC and Integrated Disease Surveillance Programme. Ethical approval was exempted as the investigation was conducted consistent with applicable state and central government law (Epidemic Diseases Act no.3, 1897). Strict data protection protocols reviewed by NCDC were followed while collecting information from cases and controls.

II. Case investigation

To ascertain the cause of death, we interviewed the treating physicians and family members of the deceased. Information was collected on clinical presentation, the timeline of events before the death, treatment and laboratory results. The patient’s CSF specimen was plated on routine media as well as in BacT/ALERT automated culture system (bioMérieux, Marcy-L’Etoile, France). The patient’s blood was also collected to rule out septicemia and cultured in BacT/ALERT automated culture system. The preliminary Gram stain of the CSF sample revealed bipolar gram negative staining leading to a presumptive diagnosis, which was conveyed to the treating doctors to initiate immediate empirical treatment. The culture isolate from CSF after 48 hours of incubation at 37°C with 5% CO2 were examined by latex agglutination, matrix assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF), and type three secretion system (TTSS) polymerase chain reaction (PCR).

III. Descriptive epidemiology

Study site.

Udupi district has a population of 1,177,908 and a literacy rate of 83% [18]. The weather is hot and humid during summers and receives heavy rainfall from May-August. The population’s comprises of agricultural communities, and rice is the main crop grown in the region. Udupi district has 76 primary health centres, six community health centres and a district hospital in the urban area. Patients who are critically ill are referred from most of Udupi district, other parts of Karnataka state and nearby districts of Kerala state to hospital A, which is a tertiary care teaching hospital.

Case definition for descriptive analysis.

A case was defined as a diagnosis of culture-confirmed melioidosis (isolation of B. pseudomallei from any clinical sample and suggestive clinical features) in a patient from the Udupi district who was admitted to the ward/intensive care unit of hospital A from January 2013—July 2018.

Case search and data collection.

We reviewed inpatient records and laboratory registers from 2013–18 from hospital A. For patients meeting the case definition, we collected clinical and laboratory data and created a line list without patient identifiers. Clinical data was available for patients attending the hospital after 2016.

IV. Case-control study

We conducted a 1:3 matched case-control study to identify risk factors on hospitalized patients meeting the above case definition from January 2017 to July 2018 to limit recall bias. A control was defined as a resident of Udupi district admitted to hospital A in July (rainy season) 2018 with a non-infectious condition. We matched each case with three hospital controls by age and sex group (males age 15–43 years, males >50 years and females >49 years). All controls were selected from the hospital admission records (eligible controls were listed and selected by random number generation using MS Excel using RANDBETWEEN function). We excluded patients receiving antibiotic treatment for pneumonia or sepsis.

A semi-structured questionnaire was used to collect data on socio-demographics, housing conditions, water storage practices, daily living activities, injuries, and environmental exposures (three months before hospitalization) from cases and controls during interviews. For the deceased cases, we conducted proxy interviews with family members.

Data analysis.

The Epi Info software 7.2 was used to analyze frequencies and proportions. For the identified risk factors, crude odds ratio (OR) with a 95% confidence interval (CI) was calculated. The exact conditional logistic regression analysis was run to obtain matched odds ratios (mOR) and 95% CI considering the small sample size using SAS version 9.4


I. Case investigation

A 17-year-old boy from the Udupi district developed complaints of fever and vomiting on July 7, 2018. On July 8, his illness progressed by evening, and he developed high-grade fever, severe headache and dizziness. On July 9, the family noticed deviation at the angle of the mouth and consulted a local doctor. The boy’s condition deteriorated and he was taken to hospital A, where he was admitted on July 11 with difficulty in swallowing, change in voice and ataxia. He developed seizures and coma on July 14 and was transferred to intensive care. His condition deteriorated with neurological involvement, and he died on July 21 in hospital A. The patient had a history of fall on June 25 during a Kabaddi game (Kabaddi is a local team game of chasing the opponent team played barefoot on a muddy playground). No external injuries were apparent, but minor abrasions on limbs and hands were noted. The case had no history of diabetes, renal dysfunction, or other comorbidities. He was diagnosed with neuro-melioidosis by cerebrospinal fluid culture and PCR. His blood culture for bacteria including mycobacterium tuberculosis was negative. Contrast magnetic resonance imaging of the brain and spinal cord showed patchy T2 and FLAIR hyperintense lesions in the brainstem, middle cerebellar peduncles, and bilateral posterior limbs of internal capsule, in addition to few subcortical lesions. Post-contrast enhancement was seen in the brainstem lesions, along the trigeminal nerves, facial and abducens nuclei. This case was later reported as part of case series showing spectrum of nervous involvement in melioidosis with detailed clinical description [19].

II. Descriptive epidemiology of cases

A total of 50 cases of melioidosis were identified at hospital A from 2013–18. The median age of cases was 52 (range 17–83) years; 80% were males, and the case fatality rate was 16%. The cases were distributed (by residence) in all the three taluks (administrative subdivisions) of the Udupi district (Fig 1); 76% of cases occurred in the villages closer to the coastline, and 66% of cases occurred between May and October (Fig 2).

Fig 1. Geographical distribution of melioidosis cases from 2013–2018 Udupi District, Karnataka, India (n = 50).

Base map republished from [20] under a CC BY license, with permission from Karnataka State Remote Sending Application Centre (KSRC), original copyright KSRC, 2022.

Fig 2. Monthly distribution of average melioidosis cases and rainfall from 2013–2018, Udupi district, Karnataka, India (n = 50).

We analyzed the clinical presentation of all 19 cases reported from January 2017- July 2018. The most common presenting symptoms were fever (89%), cough (42%), joint pain (37%) and abscess (16%). Uncontrolled diabetes (HbA1c >7%) was documented in 84% (16) cases with overall median HbA1c of 9.5% (range 5.8% -13%). The other chronic comorbidities included chronic kidney disease with diabetes (10%), COPD with diabetes (10%), liver disease with diabetes (10%), tuberculosis (5%), and cancer (5%). Among 19 cases, the most common presentation was bacteremic melioidosis (58%), followed by skin and soft tissue (11%), septic arthritis (11%), pneumonia (5%), splenic abscess (5%), neuromelioidosis (5%) and no focus (5%). Out of 19 cases, five had lung involvement; of these, one had focal lung involvement and four others had lung involvement with bacteremia. All five cases with pneumonia were exposed either to soil or stagnant water. Among the 14 non pneumonia cases, 13 had history of exposure to soil or stagnant water.

III. Case-control study

We enrolled all 19 cases from 2017–18 and 57 hospital controls in the matched cases control analysis. The enrollment flowchart for study participants is in appendix page 1. Univariate analysis showed that melioidosis cases were more likely than controls to have an injury with breach of skin, contact with stagnant water, wet soil, and both. In matched analysis, injury with a breach in skin and contact with stagnant water had a significant association with illness. In addition, we also looked for activity-specific odds ratios for contact with stagnant water. The odds of exposure to swimming in stagnant water, working in paddy field, and walk in waterlogged areas were higher among cases compared to controls (Table 1).

Table 1. Risk factors associated with melioidosis in Udupi district 2017–2018.


The death of an adolescent due to neuromelioidosis led to a broader, multi-year retrospective investigation of admitted cases of melioidosis in one district. These cases were mostly males with uncontrolled diabetes. The results from the case-control study suggest that outdoor exposure to stagnant water and wet soil in rainy season are a risk factor for melioidosis in the Udupi district.

We observed that the majority (66%) of the cases occurred during the rainy session (May-October). This is similar to the seasonality reported from the west coastal region of India, Singapore and Australia [11,16,21]. The descriptive analysis showed most (76%) cases were from villages closer to the coast with paddy fields and low-lying areas that frequently flood during the rainy session. These environmental conditions could increase the chances of contact with contaminated wet soil and stagnant water during agricultural and non-agricultural activities associated with infection. In our case-control study, exposure to wet soil and stagnant water were significant risk factors for melioidosis (p<0.001). Additionally, we also found that sustaining cut injuries was an independent predictor in multivariate analysis. Melioidosis cases were six times more likely to be exposed to cut injuries compared to controls. These findings, combined with environmental conditions, indicate that percutaneous inoculation is an important transmission mode for melioidosis in rural India. Similar epidemiological risk factors were identified among melioidosis cases in rural Thailand and northern Australia [11,12]. In contrast, recent findings have also recognized inhalation of B. pseudomallei and eating food contaminated with soil or dust as other important transmission modes [21,22].

People with long-standing diabetes with poor glycemic control are known to be at increased risk of acquiring melioidosis [6,12], and 84% of the cases in this study detected between 2013–2018 had uncontrolled diabetes. A systematic review from India for the period 1991–2018 found diabetes mellitus to be a major predisposing condition in 70% of reported cases [16]. There are multiple reports of melioidosis from various parts of India [15,16,2325], but it is not a notifiable disease. Estimates suggest that melioidosis is endemic to India with an annual burden of ~52,006 cases and death count of 31,425 (13,405–75,601) cases, with further escalation of the mortality rate to 90% if the disease remains undiagnosed and untreated [10]. Given the high burden of diabetes in India, inadequate diagnostic facilities in the microbiology laboratories, especially in rural parts [26] and low awareness among physicians and the public, the actual burden of melioidosis in India is expected to be high. Therefore, priorities include establishing state and national referral laboratory networks, training for diagnosis and surveillance for melioidosis, and increasing awareness in the community and among physicians.

Though neuromelioidosis is a rare (4–5%) presentation of melioidosis [19,27,28], children and adolescents have poor outcomes, frequently resulting in either death or neurological impairment (37%) attributed to severe sepsis and its complications, resulting from delay in treatment [5,28]. Studies from Australia [29] and Cambodia [30] have documented a case fatality rate of 7–17%; notably children in the Australian study did not have any underlying comorbidities, similar to the young index patient in our study.

Our study has limitations, including recall bias for exposure due to the retrospective nature of the study design. We attempted to limit this by enrolling the most recent cases and asking for exposures from cases and controls within a three-month reference period. The enrollment period for cases and controls was different due to logistical issues. However, we ensured comparability of risks by enrolling controls during the rainy season, as more than half of the cases were reported during this season. Finally, the findings are from admitted cases from a single hospital were enrolled, potentially limiting the generalizability of our findings.

We recommended enhancing the education of medical doctors for early diagnosis and treatment of melioidosis among fever cases; and strengthening district public health labs in Karnataka state for melioidosis diagnosis. We also recommended communication campaigns targeting high-risk groups (diabetics in coastal areas, patients with chronic renal failures, agricultural workers and people with frequent soil exposure like daily labourers). These communication campaigns should focus on minimizing unnecessary exposures to soil and water through avoidance or protective clothing including use of footwear.

Following this investigation, the National Centre for Disease Control published an information bulletin to increase awareness among clinicians of melioidosis as a differential diagnosis in fever of unknown origin and community acquired pneumonia. To enhance surveillance, an expert group meeting was convened to develop surveillance case definitions, and district lab personnel training was initiated to strengthen melioidosis diagnostic capacity.

Supporting information

S1 Fig. Flow diagram showing selection of melioidosis cases, in the case control study, Udupi District, Karnataka.



  1. 1. Goodrick I, Todd G, Stewart J. Soil characteristics influencing the spatial distribution of melioidosis in Far North Queensland, Australia. Epidemiol Infect. 2018;146(12):1602–1607.
  2. 2. Aldhous P. Tropical medicine: melioidosis? Never heard of it. Nature. 2005; 434: 692–693. pmid:15815599
  3. 3. Stone R. Infectious disease. Racing to defuse a bacterial time bomb. Science. 2007; 317: 1022–1024. pmid:17717162
  4. 4. Currie BJ, Dance DA, Cheng AC. The global distribution of Burkholderia pseudomallei and melioidosis: an update. Trans R Soc Trop Med Hyg. 2008; 102 Suppl1: S1–4. pmid:19121666
  5. 5. White NJ: Melioidosis. Lancet. 2003, 361:1715–1722. pmid:12767750
  6. 6. Wiersinga WJ, Van der Poll T, White NJ, Day NP, Peacock SJ: Melioidosis: insights into the pathogenicity of Burkholderia pseudomallei. Nature Rev. 2006;4:272–282. pmid:16541135
  7. 7. Currie BJ, Fisher DA, Howard DM, Burrow JN, Lo D, Selva-Nayagam S, et al: Endemic melioidosis in tropical northern Australia: a 10-year prospective study and review of the literature. Clin Infect Dis. 2000;31:981–986. pmid:11049780
  8. 8. Hassan MR, Pani SP, Peng NP, Voralu K, Vijayalakshmi N, Mehanderkar R, et al. Incidence, risk factors and clinical epidemiology of melioidosis: a complex socio-ecological emerging infectious disease in the Alor Setar region of Kedah, Malaysia. BMC Infect Dis. 2010;10:302. pmid:20964837
  9. 9. Birnie E, Virk HS, Savelkoel J, Spijker R, Bertherat E, Dance DAB, etal. Global burden of melioidosis in 2015: a systematic review and data synthesis. Lancet Infect Dis. 2019; 19 (8):892–902. pmid:31285144
  10. 10. Limmathurotsakul D, Golding N, Dance DA, Messina JP, Pigott DM, Moyes CL, et al. Predicted global distribution of Burkholderia pseudomallei and burden of melioidosis. Nat Microbiol. 2016;1:15008.
  11. 11. Currie B. J. & Jacups S. P. Intensity of rainfall and severity of melioidosis, Australia. Emerg. Infect. Dis.2003; 9:1538–1542. pmid:14720392
  12. 12. Suputtamongkol Y, Chaowagul W, Chetchotisakd P, Lertpatanasuwun N, Intaranongpai S, Ruchutrakool T, et al. Risk factors for melioidosis and bacteremic melioidosis. Clin Infect Dis. 1999 Aug;29(2):408–13. pmid:10476750
  13. 13. Chierakul W. et al. Melioidosis in 6 tsunami survivors in Southern Thailand. Clin. Infect. Dis.2005; 41, 982–990. pmid:16142663
  14. 14. Saravu K, Mukhopadhyay C, Vishwanath S, Valsalan R, Docherla M, Vandana KE, et al. Melioidosis in southern India: epidemiological and clinical profile. Southeast Asian J Trop Med Public Health. 2010 Mar;41(2):401–9. pmid:20578524
  15. 15. Menon R, Baby P, Kumar V A, Surendran S, Pradeep M, Rajendran A, et al. Risk Factors for Mortality in Melioidosis: A Single-Centre, 10-Year Retrospective Cohort Study. Scientific World Journal. 2021 Jul 5;2021:8154810. pmid:34285680
  16. 16. Mukhopadhyay C, Shaw T, Varghese GM, Dance DAB. Melioidosis in South Asia (India, Nepal, Pakistan, Bhutan and Afghanistan). Trop Med Infect Dis. 2018 May 22;3(2):51. pmid:30274447
  17. 17. India State-Level Disease Burden Initiative Diabetes Collaborators. The increasing burden of diabetes and variations among the states of India: the Global Burden of Disease Study 1990–2016. Lancet Glob Health. 2018 Dec;6(12):e1352–e1362. pmid:30219315
  18. 18. Government of Karnataka. About district. Uduppi district. Available at (accessed on 04 July, 2022).
  19. 19. Chatterjee A, Saravu K, Mukhopadhyay C, Chandran V. Neurological Melioidosis Presenting as Rhombencephalitis, Optic Neuritis, and Scalp Abscess with Meningitis: A Case Series from Southern India. Neurol India. 2021 Mar-Apr;69(2):480–482. pmid:33904481
  20. 20. Karnataka State Remote Sensing Application Centre. Avaliable at (accessed on July 20, 2022).
  21. 21. Liu X, Pang L, Sim SH, Goh KT, Ravikumar S, Win MS, et al. Association of melioidosis incidence with rainfall and humidity, Singapore, 2003–2012. Emerg. Infect. Dis.2015; 21, 159–162. pmid:25531547
  22. 22. Limmathurotsakul D, Kanoksil M, Wuthiekanun V, Kitphati R, deStavola B, Day NP, et al. Activities of daily living associated with acquisition of melioidosis in northeast Thailand: a matched case-control study. PLoS Negl Trop Dis. 2013;7(2):e2072. pmid:23437412
  23. 23. Ganesan V, Sundaramoorthy R, Subramanian S. Melioidosis-Series of Seven Cases from Madurai, Tamil Nadu, India. Indian J Crit Care Med. 2019 Mar;23(3):149–151. pmid:31097893
  24. 24. Annamalai AK, Padmini K. Melioidosis. Indian J Med Res. 2019 Apr;149(4):561–562. pmid:31411183
  25. 25. Subramony H, Gunasekaran S, Paul Pandi VK. Disseminated melioidosis with native valve endocarditis: a case report. Eur Heart J Case Rep. 2019 Jun 1;3(2):ytz097. pmid:31449647
  26. 26. Central Bureau of Health Intelligence. National health profile 2019. Available at (accessed on December 30, 2020).
  27. 27. Saravu K, Kadavigere R, Shastry AB, Pai R, Mukhopadhyay C. Neurologic melioidosis presented as encephalomyelitis and subdural collection in two male labourers in India. J Infect Dev Ctries. 2015 Nov 30;9(11):1289–93. pmid:26623640
  28. 28. Wiersinga WJ, Currie BJ, Peacock SJ. Melioidosis. N Engl J Med. 2012 Sep 13;367(11):1035–44. pmid:22970946
  29. 29. Kandasamy Y, Norton R. Paediatric melioidosis in North Queensland, Australia. J Paediatr Child Health. 2008 Dec;44(12):706–8. pmid:19054292
  30. 30. Turner P, Kloprogge S, Miliya T, Soeng S, Tan P, Sar P, et al. A retrospective analysis of melioidosis in Cambodian children, 2009–2013. BMC Infect Dis. 2016 Nov 21;16(1):688. pmid:27871233