Scholarly article on topic 'Emerging clinico-epidemiological trends in melioidosis: analysis of 95 cases from western coastal India'

Emerging clinico-epidemiological trends in melioidosis: analysis of 95 cases from western coastal India Academic research paper on "Clinical medicine"

0
0
Share paper
OECD Field of science
Keywords
{Melioidosis / " Burkholderia pseudomallei " / Rainfall / Epidemiology / "Musculoskeletal melioidosis" / "Dental abscess"}

Abstract of research paper on Clinical medicine, author of scientific article — K. Vidyalakshmi, S. Lipika, S. Vishal, S. Damodar, M. Chakrapani

Summary Objectives To study the clinico-epidemiological trends in melioidosis, an emerging disease in the western coastal region of India. Methods Data of 95 patients with melioidosis in the western coastal region of India were retrospectively analyzed with respect to monthly rainfall, risk factors, clinical presentations, and outcome. Results A strong linear correlation was seen between average monthly rainfall and the occurrence of cases (p =0.002). Mortality was seen only in patients with bacteremia (p <0.001). Nine (40.9%) patients with septic shock died (p <0.001). Age ≥40 years and diabetes mellitus were seen in 75.8% of cases, each. Pneumonia was the most common clinical presentation (32.6%), followed by musculoskeletal disease (20%), melioidotic lymphadenopathy (7.4%), and dental abscess (6.3%). Only 36.8% of patients had exposure to wet soil/surface water. Conclusions Melioidosis is quite prevalent in the western coastal region of India, and is strongly associated with rainfall, age, and diabetes mellitus. Higher proportions of musculoskeletal, dental, and lymph node melioidosis were seen in this region as compared to endemic areas. Bacteremic melioidosis has a poorer prognosis than non-bacteremic melioidosis. The presence of septic shock is a strong predictor of mortality. Percutaneous inoculation may not be the main portal of entry for Burkholderia pseudomallei in this region.

Academic research paper on topic "Emerging clinico-epidemiological trends in melioidosis: analysis of 95 cases from western coastal India"

ELSEVIER

Contents lists available at SciVerse ScienceDirect

International Journal of Infectious Diseases

journal homepage: www.elsevier.com/locate/ijid

Emerging clinico-epidemiological trends in melioidosis: analysis of 95 cases from western coastal India

K. Vidyalakshmia'*, S. Lipikab, S. Vishalc, S. Damodard, M. Chakrapanid

a Department of Microbiology, Kasturba Medical College, Manipal University, Light House Hill Road, Mangalore, Karnataka, 575001, India b Department of Medical Microbiology, Post Graduate Institute of Medical Education and Research, Chandigarh, India c Department of Medicine, Kasturba Medical College, Manipal University, Manipal, Karnataka, India d Department of Medicine, Kasturba Medical College, Manipal University, Mangalore, Karnataka, India

ARTICLE INFO SUMMARY

Objectives: To study the clinico-epidemiological trends in melioidosis, an emerging disease in the western coastal region of India.

Methods: Data of 95 patients with melioidosis in the western coastal region of India were retrospectively analyzed with respect to monthly rainfall, risk factors, clinical presentations, and outcome. Results: A strong linear correlation was seen between average monthly rainfall and the occurrence of cases (p = 0.002). Mortality was seen only in patients with bacteremia (p < 0.001). Nine (40.9%) patients with septic shock died (p < 0.001). Age >40 years and diabetes mellitus were seen in 75.8% of cases, each. Pneumonia was the most common clinical presentation (32.6%), followed by musculoskeletal disease (20%), melioidotic lymphadenopathy (7.4%), and dental abscess (6.3%). Only 36.8% of patients had exposure to wet soil/surface water.

Conclusions: Melioidosis is quite prevalent in the western coastal region of India, and is strongly associated with rainfall, age, and diabetes mellitus. Higher proportions of musculoskeletal, dental, and lymph node melioidosis were seen in this region as compared to endemic areas. Bacteremic melioidosis has a poorer prognosis than non-bacteremic melioidosis. The presence of septic shock is a strong predictor of mortality. Percutaneous inoculation may not be the main portal of entry for Burkholderia pseudomallei in this region.

© 2012 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.

Article history: Received 25 October 2011 Accepted 14 February 2012 Corresponding Editor: William Cameron, Ottawa, Canada

Keywords: Melioidosis

Burkholderia pseudomallei

Rainfall

Epidemiology

Musculoskeletal melioidosis Dental abscess

1. Introduction

Melioidosis, which is endemic in Southeast Asia and northern Australia, is an emerging infection in other Asian regions, including India, China, and Sri Lanka, and also in Brazil, South America.1,2 With increasing awareness of melioidosis, more cases have now been reported from various regions of India, including Maharashtra, Kerala, Karnataka, Tamil Nadu, and Pondicherry.3-6 Recently, clusters of cases have been reported from the western coastal region of India.7,8

The epidemiology and risk factors of melioidosis have been extensively studied in endemic areas, from where high mortality rates have been reported, i.e., up to 50% in northeast Thailand and 14% in northern Australia.1,2 However, detailed studies on the epidemiology of melioidosis in India are lacking. The western coastal region of India seems to be an ideal setting for endemicity of this disease, with an annual rainfall of about 300 cm (tropical

* Corresponding author. Tel.: +91 824 2444590 ext. 5065/5063. E-mail address: k.vidyalakshmi@lycos.com (K. Vidyalakshmi).

monsoon climate), a high population of diabetics,9,10 and agriculture as the predominant occupation. In this study we retrospectively analyzed the data of cases of melioidosis presenting at our hospital to study the epidemiology, risk factors, and clinical presentations of this disease in the western coastal region of India.

2. Materials and methods

2.1. Study population

All patients with culture-confirmed melioidosis whose clinical specimens (blood, exudates, and urine) were received at the Microbiology Laboratory, Kasturba Medical College Hospital, Mangalore, during the period January 2005 to December 2010 were included in the study. Blood cultures were performed using the BacT-Alert automated blood culture system (bioMerieux, Marcy l'Etoile, France). All other specimens were cultured on sheep blood agar, chocolate agar, and MacConkey agar. Isolates were identified by standard microbiological techniques.11 Salient features of identification included: oxidase-positive, Gram-negative bacilli with bipolar staining; rough, wrinkled, pink colonies on

1201-9712/$36.00 - see front matter © 2012 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijid.2012.02.012

MacConkey agar; oxidative utilization of glucose, lactose, and maltose; lysine decarboxylase-negative; arginine dihydrolase-positive; reduced nitrates; positive gelatin liquefaction; growth at 42 °C positive; resistance to gentamicin (10 mg/disk) and polymyxin B (300 U/disk); and growth on Ashdown's medium12 positive, with formation of rugose colonies resembling cornflower heads, which take up neutral red dye from the medium. Identification was confirmed by the Mini API System (bioMerieux) for the initial isolates, and the later isolates were confirmed by the Vitek 2 Compact System (bioMerieux).

Case records of culture-positive patients were retrospectively analyzed. The month of the year in which the patient presented with illness was noted in order to determine if there was a correlation with the monsoon season. The monthly rainfall statistics for the past 6 years were obtained for two districts, namely Dakshina Kannada (DK) and Kasaragod, situated in the western coastal region of India.13

Primary clinical presentations and secondary clinical foci of infection were noted. Melioidosis was classified as bacteremic melioidosis (a positive blood culture with a single focus or no identifiable focus of infection), or localized melioidosis (one or more foci of infection and a negative blood culture). Clinical presentation was considered acute and from a recent infection if symptoms had been present for less than 2 months, and presentation was considered chronic when symptoms had been present for more than 2 months.14

The presence of risk factors and underlying comorbidities were recorded, including diabetes mellitus, alcohol intake, chronic renal disease, chronic lung disease, occupation (whether it involved contact with wet soil and water), age, and sex of the patient. The outcome of treatment was noted.

2.2. Definitions

Diabetes mellitus was defined as a fasting blood glucose level >126mg/dl (7.0 mmol/l) or postprandial blood glucose level >200 mg/dl (11.1 mmol/l).15 Excessive alcohol consumption was defined as more than 14 drinks per week or four drinks per occasion for men, and as more than seven drinks per week or three drinks per occasion for women, in accordance with the criteria established by the US National Institute on Alcohol Abuse and Alcoholism.16 Chronic kidney disease was defined as kidney damage or a glomerular filtration rate (GFR) <60 ml/min/1.73 m2 for >3 months, irrespective of cause.17 Chronic lung disease was defined as a documented diagnosis of chronic obstructive airways disease.14

2.3. Statistical analyses

The correlation coefficient was calculated for assessing the linear relationship between average monthly rainfall and the number of cases of melioidosis for those months. Correlation was assessed by non-parametric method to obtain the corresponding Spearman's correlation coefficient. Associations between categorical variables were analyzed by Fisher's exact test. A p-value of <0.05 was considered significant. Institutional ethics committee clearance for the study was obtained. Informed patient consent was not required as this was a retrospective analysis of case records.

3. Results

During the 6-year study period, 95 cases of melioidosis were confirmed by culture in our laboratory (Table 1). Of these, 37 (38.9%) were bacteremic melioidosis. Twenty-two (59.5%) of the bacteremic patients had septic shock. Nine of the 22 patients with septic shock died (40.9%), as compared to none of the patients without septic shock, who all (100%) survived (Fisher's exact test,

Table 1

Clinical presentations and outcomes of 95 cases of melioidosis

Clinical presentations Total cases Septic shock cases Deaths

n (%) n (%) n (%)

Bacteremic melioidosis (n = :37)

Pneumonia 19 (20) 11 (50) 7 (77.8)

Septic arthritis 5 (5.3) 2 (9.1) 0

Splenic abscess 2 (2.1) 0 0

Lymphadenopathy 1 (1.1) 1 (4.5) 0

Prostatic abscess 1 (1.1) 0 0

No focus identified 9 (9.5) 8 (36.4) 2 (22.2)

Non bacteremic melioidosis (localized melioidosis) (n = 58)

Pneumonia 12 (12.6) 0 0

Septic arthritis 10 (10.5) 0 0

Lymphadenopathy 6 (6.3) 0 0

Dental abscess 6 (6.3) 0 0

Psoas abscess 4 (4.2) 0 0

Skin nodules 3 (3.2) 0 0

Parotid abscess 2 (2.1) 0 0

Pericardial effusion 2 (2.1) 0 0

Gluteal abscess 2 (2.1) 0 0

Breast abscess 2 (2.1) 0 0

Splenic abscess 2 (2.1) 0 0

Liver abscess 2 (2.1) 0 0

Prostatic abscess 1 (1.1) 0 0

Pyopneumothorax 1 (1.1) 0 0

Lung abscess 1 (1.1) 0 0

Pyelonephritis 1 (1.1) 0 0

Meningitis 1 (1.1) 0 0

Total 95 22 (23.2) 9 (9.5)

p < 0.001). All nine patients who died were bacteremic; there was no mortality among non-bacteremic patients (Fisher's exact test, p < 0.001).

All our patients came from DK and Kasaragod districts, which are situated in the western coastal region of India. Sixty-eight patients (71.6%) presented during the heavy monsoon months (June to September, southwest monsoon). Figure 1 shows the relationship between the average monthly rainfall in DK and Kasaragod districts, and the occurrence of melioidosis cases, as well as bacteremia, septic shock, and deaths due to melioidosis. The number of melioidosis cases in a month was linearly correlated with the average monthly rainfall, and this correlation was found to be highly statistically significant (Spearman's correlation coefficient = 0.803; p = 0.002). Twenty-eight (75.7%) of the bacteremic cases, 18 (81.8%) septic shock cases, and six (66.7%) deaths occurred during the monsoon months.

Patients ranged in age from 20 days to 74 years (median 50 years, interquartile range 16 years). Seventy-two (75.8%) patients were aged >40 years. The maximum number of cases occurred in those between the ages of 40 and 70 years (69 patients, 72.6%); seven patients were children (age <12 years). Sixty-three (66.3%) patients were male, and the male-to-female ratio was 2.0:1.

3.1. Clinical presentations

The primary clinical presentations and outcomes of the 95 patients with culture-confirmed melioidosis are shown in Table 1.

Fever was the most consistent presenting complaint and was present in 92 (96.8%) patients. Of the three patients without fever, one was a 29-year-old pregnant woman with no defined risk factors who presented with cervical lymphadenopathy, the second was a 61-year-old male with impaired glucose tolerance who presented with a dental abscess from which Burkholderia pseudomallei was isolated, and the third was a 34-year-old diabetic woman with B. pseudomallei pneumonia.

The duration of presenting complaints ranged from 2 days to as long as a year (median duration 4 weeks, interquartile range 7 weeks). Sixty-eight (71.6%) patients had acute presentations (25 pneumonia, 11 septic arthritis, seven septicemia, five dental

1200 - 25

□ Average rainfall ^^"Total no. of cases > No. of cases with septic shock--No. of deaths

■No. of cases with bacteraemia

Figure 1. Relationship between average monthly rainfall and the number of cases of melioidosis, and of bacteremia, septic shock, and deaths due to melioidosis.

abscess, five lymphadenopathy, and 15 others); 27 (28.4%) had chronic presentations (six pneumonia, four septic arthritis, three psoas abscess, and 14 others). Forty-eight (70.6%) patients with acute presentations and 19 (70.4%) patients with chronic presentations presented during the monsoon season.

Pulmonary melioidosis was the most common clinical presentation, accounting for 33 (34.7%) cases; of these, 31 presented with pneumonia, which accounted for 32.6% of cases overall. Of the remaining two, one presented with pyopneumothorax and the other with a lung abscess. Twenty-five (80.6%) cases of pneumonia had an acute presentation, and of these 23 (92%) presented during the monsoon season. Even when considered separately among bacteremic and non- bacteremic patients, pneumonia was the most common presentation (51.4% and 20.7%, respectively) (Table 1). Seven patients with bacteremic pneumonia died.

Nineteen cases (20%) presented with musculoskeletal melioi-dosis (15 septic arthritis and four psoas abscess). Septic arthritis caused by B. pseudomallei affected only the large joints, most often the knee (12 cases), followed by the elbow joint (two patients), hip (one patient), and shoulder (one patient). One patient had involvement of both the knee and elbow joint. Eleven patients had acute presentations with symptoms of less than 8 weeks duration. Five patients were bacteremic and two had septic shock. All 15 patients were diabetic, and six of them had occupational exposure to wet soil/ground water. Thirteen patients had a raised erythrocyte sedimentation rate (ESR). Ten had a chest radiographic picture suggestive of tuberculosis in the form of consolidation, fibrocavitary lesion, or non-homogeneous upper lobe opacity, due to which anti-tubercular therapy (ATT) was started on clinical and radiological grounds. Surgical drainage was done, and B. pseudo-mallei was isolated from pus in 10 cases and from blood in five cases. Following this, ATT was withdrawn and appropriate therapy for melioidosis was given. All 15 patients recovered.

Spondylitis with psoas abscess was the presenting feature for four patients. The predominant clinical complaints were fever and backache. Three of them had a chronic presentation. All four patients had a raised ESR, and an X-ray of the spine showed spondylitis and a reduction in disc space. Hence, all four were initially treated for tuberculosis on clinical and radiological grounds. Following surgical drainage, B. pseudomallei was isolated from psoas abscess pus in all four patients. Interestingly, in one patient, the psoas abscess pus was also positive for Mycobacterium

tuberculosis by acid-fast bacillus (AFB) smear and by PCR. This patient was also Mantoux test positive. Specific antibiotic therapy was given to all four patients. The person who was AFB-positive was also simultaneously given ATT. All four patients recovered.

Lymphadenopathy was the presenting feature in seven (7.4%) patients. One patient was bacteremic and had involvement of the upper deep cervical nodes. This patient also had radiological evidence of liver abscess and non-homogeneous lung opacity. The other six patients were non-bacteremic and had involvement of cervical (n = 4) and supraclavicular (n = 2) nodes. In one patient with supraclavicular node involvement, a computed tomography scan revealed enlarged mediastinal lymph nodes and a small encysted pleural effusion. Six patients had a raised ESR, and in four patients fine needle aspiration cytology from the lymph nodes showed granulomatous reactions suggestive of tuberculous lymphadenitis. All were initially empirically treated with ATT. When B. pseudomallei was recovered from lymph node aspirate and/or blood culture, ATT was withdrawn and appropriate treatment for melioidosis was instituted. All patients recovered.

Six cases presented with dental abscesses, subsequent to dental manipulations. One was a 4-year-old male child who underwent filling of a cavity in the lower molar tooth. This was followed 4-5 days later by fever, pain, swelling, and abscess formation in the mandible just below the tooth. The other five patients were adults (four diabetic and one with impaired glucose tolerance) who had undergone root canal treatment, which was followed by abscess formation in the jaw just below the manipulated tooth. One patient also developed an enlarged submandibular lymph node. In all six cases the affected tooth was extracted and pus from the abscess was sent for culture; all grew B. pseudomallei. Following appropriate treatment for melioidosis all six patients recovered.

Other presentations included visceral abscesses (liver, spleen, prostate, parotid, and pericardium), skin and soft tissue abscesses, pyelonephritis, and meningitis (Table 1).

In addition to the primary clinical presentation, secondary clinical foci were subsequently evident during the course of clinical illness (Table 2) in 24 (25.3%) patients overall.

Three patients had culture-confirmed recurrent melioidosis following completion of antibiotic therapy and clinical improvement during the first presentation. One was a 17-year-old female juvenile diabetic in whom B. pseudomallei was isolated from blood during the initial presentation and from sputum during the

Table 2

Secondary clinical foci

Primary clinical No. No. with Secondary clinical focia

presentation secondary foci

Pneumonia 31 6 2 liver abscesses, 1 septic arthritis, 1 mediastinal mass, 1 pleural effusion, 1 peritonitis, 1 splenic abscess, 1 parotid abscess

Septic arthritis 15 4 1 pneumonia, 1 subcutaneous nodules, 1 mediastinal lymphadenopathy, 1 urinary tract infection

Lymphadenopathy 7 2 1 pneumonia, 1 pleural effusion, 1 liver abscess

Splenic abscess 4 2 1 pneumonia, 1 pyopneumothorax

Psoas abscess 4 1 1 peritonitis

Skin nodules 3 1 1 lymphadenopathy, 1 liver abscess, 1 splenic abscess

Gluteal abscess 2 1 pneumonia, 1 septic arthritis

Pericardial effusion 2 1 1 pleural effusion

Prostatic abscess 2 1 1 pneumonia

Liver abscess 2 1 1 perinephric abscess

Meningitis 1 1 1 prostatic abscess, 1 renal abscess

Pyopneumothorax 1 1 1 liver abscess

Pyelonephritis 1 1 1 lymph node enlargement

a Some patients had more than one secondary focus.

recurrence. The presentation during the primary infection was septicemia. The recurrence occurred 6 months later when she presented with pneumonia; an ultrasonogram showed right lung collapse, pleural effusion, and focal lesions in the liver. The second patient was a 55-year-old diabetic woman who presented with pneumonia during the primary episode. The second episode occurred after 6 months, and she presented with cellulitis of the lower limb. B. pseudomallei was isolated from blood during the primary episode and from pus during the recurrence. Both of these patients were given a second course of treatment for melioidosis, following which they recovered. The third was a 31-year-old diabetic man who presented initially with pneumonia and sepsis. The recurrence occurred after 3 months, and he presented with features of septic shock. B. pseudomallei was isolated from blood in both instances. The patient died of the infection during the second episode.

Table 3 shows the risk factors and associated bacteremia and mortality. Predisposing factors were present in 83 (87.4%) patients. Seventy-two (75.8%) of the 95 patients were diabetic and 72 (75.8%) were aged over 40 years. Eleven (11.6%) patients had chronic kidney disease (CKD). There was a statistically significant higher occurrence of bacteremia in patients with CKD compared to those without CKD (81.8% vs. 33.3%, Fisher's exact test p = 0.003). Thirty-five (36.8%) patients had occupational exposure to wet soil or surface water in the form of farming, agriculture, gardening, fishing, manual labor, land surveying, and building construction. The other major risk factors included alcoholism, immunosup-pression, and chronic lung disease. Twelve (12.6%) patients had no defined risk factors.

Table 4 shows a comparison of the epidemiology of melioidosis between the western coastal region of India and known endemic regions.

Table 3

Risk factors in 95 patients with melioidosis, associated bacteremia, and mortality

Risk factor No. of patientsa

Total (N =95) Bacteremic cases (n = 37) Deaths (n = 9)

n (%) n (%) n (%)

Diabetes 72 (75.8) 23 (62.2) 4 (44.4)

Age >40 years 72 (75.8) 27 (73.0) 5 (55.6)

Exposure to soil 35 (36.8) 9 (24.3) 1 (11.1)

Chronic kidney disease 11 (l1.6) 9 (24.3) 2 (22.2)

Alcoholism 10 (10.5) 3 (8.1) 1 (11.1)

Chronic lung disease 9 (9.5) 0 0

Immunosuppressive therapy/other immunosuppressive condition13 12 (12.6) 6 (16.2) 3 (33.3)

No defined risk factors 12 (12.6) 4 (10.8) 1 (11.1)

a Some patients had more than one risk factor.

b Four with steroid therapy, two with HIV infection, two with malignancy, one with systemic lupus erythematosus and steroid therapy, one had a splenectomy, one with

malnutrition, and one with cirrhosis.

Table 4

Comparison of epidemiology of melioidosis in India with other regions

Parameter This study Northern Australia14 Thailand Malaysia20

n = 95 n = 540 n = 6861, n = 20421 n =145

Median age (years) 50 49 52 50

M:F ratio 2.0:1 2.2:1 1.6:1 3:1

Bacteremic cases % 38.9 55 58 52

Mortality % 9.5 14 38-61 34

Diabetes mellitus % 75.8 39 60.3 57

Environmental exposure to Burkholderia pseudomallei % 36.8 81 81 NR

Pulmonary disease % 34.7 51.4 45 42.1

Musculoskeletal disease % 20 3.7 5 4.8

Dental abscess % 6.3 0 0 0

Prostatic abscess % (calculated in males only) 3.2 20.4 0.3 0.9

Neurological disease % 1.1 2.5 3 4.8

NR, not recorded.

3.2. Treatment

Upon receipt of laboratory reports patients were treated as per recommendations,1,2 with intravenous ceftazidime during the acute phase (2 weeks) and trimethoprim-sulfamethoxazole, doxycycline, and chloramphenicol during the eradication phase (12-20 weeks). Chloramphenicol was given only for the first 8 weeks and was withdrawn if the leukocyte count fell below 4 x 109/l. Nine patients died. Among those who recovered, the time taken for defervescence varied from 7 to 15 days.

4. Discussion

Detailed studies on the epidemiology and risk factors of melioidosis from the Indian subcontinent are lacking, although increasing numbers of cases have been reported from this region due to increasing awareness.3-8 During the span of 6 years from January 2005 to December 2010, we diagnosed a total of 95 cases of melioidosis by culture. Many similarities in the epidemiological features of melioidosis in this region were observed as compared to those of known endemic regions. However, some differences in epidemiology were also observed.

During the southwest monsoon (June to September) the western coastal region of India receives approximately 250300 cm of rainfall, along with heavy winds. The association between rainfall intensity and melioidosis is well documented from endemic regions, with 75% and 85% of cases presenting during the rainy season in northeast Thailand18 and northern Australia,19 respectively. Our study also shows a similar pattern, with 71.6% of patients presenting during the monsoon. B. pseudomallei is a soil saprophyte found in soil and surface water in endemic regions. Infection is believed to occur by inoculation through skin abrasions or by inhalation.1,2 The intensity of rainfall has been shown to be an independent risk factor for melioidosis pneumonia, septic shock, and death.19

A similar association was seen in our study, with 71.6% of melioidosis cases, 75.7% of melioidotic bacteremia, 81.8% of septic shock cases, and 66.6% of deaths due to melioidosis occurring during the monsoon months (Figure 1). Many possible explanations have been suggested for more severe disease following rainfall, including aerosolization of bacteria from surface water and soil during heavy winds and rain, larger bacterial inoculating dose, and infection with more virulent bacteria.19

A higher proportion of males were found to be affected by melioidosis (66.3%). A male preponderance has been documented in other studies from endemic regions. Males accounted for 69%, 75.2%, and 61.8% of patients, respectively, in studies from Australia14, Malaysia,20 and Thailand21 (Table 4). This is probably explained by the higher exposure of men to outdoor activities.

Melioidosis has been referred to as the ''remarkable imitator''22 and the ''mimicker of maladies'',3 because the disease can mimic pyogenic bacterial infection, Gram-negative sepsis, or tuberculosis. In our study there was no single presenting clinical feature that was typical of melioidosis, but fever was the most consistent complaint and was present in 96.8% of patients. Melioidosis is known to present as a febrile illness with protean clinical manifestations, ranging from acute fulminant septicemia to chronic debilitating localized infection, characterized by abscess formation. Evidence of recent trauma or an infected wound is usually not present, and it can affect almost any organ in the body.2,23 In our study too we found varied clinical presentations (Table 1).

Pneumonia as the most common clinical presentation (32.6%) correlates well with studies from endemic regions, although the incidence was less than that seen in endemic areas. In the Darwin study from northern Australia, pneumonia accounted for 51% of

cases.14 In Thailand1 and Malaysia,20 pneumonia accounted for 45% and 42.1% of cases, respectively (Table 4). Inhalation is believed to be the mode of acquisition, particularly of acute pulmonary melioidosis, especially during severe weather conditions like tropical cyclones and intense rainfall with winds. In our study, 92% of acute melioidotic pneumonia occurred during the monsoon. Chronic pulmonary melioidosis is thought to be acquired through cutaneous exposure followed by hematogenous dissemination to the lung.19,24 Pulmonary melioidosis often mimics pulmonary tuberculosis, both clinically and radiologically, and can be differentiated reliably only on isolation of B. pseudomallei in culture. This should particularly be kept in mind in regions where both diseases are prevalent.8

There were certain differences in the clinical presentations in this study when compared with studies from endemic regions, which included a higher proportion of musculoskeletal melioi-dosis, a higher frequency of dental abscesses following dental procedures, and a higher proportion of melioidotic lymphadenop-athy.

Musculoskeletal melioidosis is a well-recognized manifestation of the disease, but is relatively rare even in endemic areas. In the Darwin study from Northern Australia, musculoskeletal melioi-dosis accounted for 3.7% of all cases of melioidosis14 and in the Infectious Diseases Association of Thailand series, 5% of all cases of melioidosis presented with musculoskeletal disease.1 In contrast, musculoskeletal melioidosis accounted for 20% of all cases in our study (Table 4).

Septic arthritis due to B. pseudomallei usually involves large joints, most often the knee, followed by elbow, shoulder, and hip joints.25 The clinical signs and symptoms do not differentiate it from other causes of septic arthritis (the affected joint appears hot, red, tender, and swollen), except that the systemic features of illness are often more prominent.25 Radiographic pictures are indistinguishable from those of tuberculous bone involvement. Spondylitic melioidosis poses a diagnostic and therapeutic challenge. It is clinically and radiologically indistinguishable from tuberculous spondylitis.8 Since tuberculosis is endemic in India, chronic spondylitis with psoas abscess and a high ESR is empirically treated as tuberculosis, unless proved otherwise. Melioidosis of bone and joint mimicking a tuberculous infection has been reported previously.8 Isolation of the causative agent in culture is the only way to arrive at a specific diagnosis of melioidosis, and a high index of suspicion is required for an early and accurate diagnosis. Surgical drainage and long-term courses of intravenous antibiotics are required for successful treatment.

Although rare, lymph node involvement by B. pseudomallei has previously been reported. Previous studies have reported suppu-rative involvement of inguinal nodes and mediastinal nodes, as well as lymph nodes of the head and neck region by B. pseudomallei.26,27 Lymphadenitis due to B. pseudomallei can also mimic tuberculous lymphadenitis, especially in areas where tuberculosis is endemic.8 Therefore melioidosis should be considered in the differential diagnosis of suppurative or granulomatous regional lymphadenitis in this region.

Melioidosis following dental manipulation is rare. To the best of our knowledge there has only been a single report of disseminated melioidosis occurring after dental extraction; this involved a Japanese man after his return from a business trip to Singapore and Indonesia.28 Six patients in our study had a localized dental abscess following dental manipulation. None of them had a history of exposure to soil. Dental root canal treatment is usually done in three to four sittings, each sitting a few days apart. Between the sittings, the canal is left open for drainage. The causative agent might have been introduced through this opening during rinsing/ gargling while brushing teeth, and the domestic water supply could have been the source. Chlorination of water for domestic

supply is not practiced in many parts of India. Culture of the water supplies for B. pseudomallei should be done in this region to confirm this hypothesis.

Melioidosis presenting as abscesses (both skin and visceral) is not uncommon and may serve as a source of systemic infection, or result from hematogenous spread.1 It has previously been suggested that melioidosis should be considered in the differential diagnosis when abscesses are encountered at unusual sites, such as spleen, prostrate, and parotid, with chronic presentation.29 Liver abscess can be caused by Entamoeba histolytica or by enteric bacteria, in addition to B. pseudomallei. However splenic abscess with enteric bacteria or protozoa is much less common, and should therefore alert one to the possibility of melioidosis.2 In Thailand, 95% of splenic abscesses are due to B. pseudomallei.2 In our study we had four cases of splenic abscess. The incidence of prostatic abscess seems to be high in Australia (20% of male patients) as compared to Thailand (0.3%)1 and in our study (3.2%). It has been reported that prostatic abscesses require drainage along with appropriate treatment in contrast with other internal abscesses which may respond to medical therapy alone.30

Recurrent melioidosis after an apparent cure following the first episode is well documented and has been studied extensively in endemic areas.31-33 Reported rates of recurrent melioidosis are 16% in Thailand31 and 6% in northern Australia.14 A study from Thailand showed that 75% of recurrences were due to relapse of the original infecting strain, and relapse in these cases usually occurred within 12 months of the initial episode; 25% of recurrences were due to reinfection, where the strain isolated during the primary and recurrent episodes differed genotypically, and reinfection in these cases usually occurred 1 year after the initial episode.33 The most important determinants of relapse in a study from Thailand were the choice and duration of oral antimicrobial therapy. Oral therapy with trimethoprim-sulfamethoxazole and doxycycline for 12-16 weeks had a 90% reduced risk of relapse compared with patients treated for <8 weeks.31 Other documented risk factors for relapse include bacteremia on initial admission, multifocal distribution, and diabetes melli-tus.14,31 In our study all three patients with recurrent melioidosis were diabetic and had bacteremia on initial presentation. One patient had multifocal distribution of lesions. One patient with recurrent melioidosis died.

Table 2 shows that the development of secondary clinical foci of infection is not uncommon in melioidosis (25.3%), and this has also been reported in other studies.14

Diabetes mellitus is the most important risk factor for melioidosis and is reported to increase the relative risk of infection by up to a hundred-fold in certain age groups in northeast Thailand.18 Case-control studies from Thailand have documented diabetes mellitus, pre-existing renal disease, thalassemia, and occupational exposure to soil and water to be confirmed risk factors for melioidosis and bacteremic melioidosis.21 Excessive alcohol use, malignancy, and immunosuppression are also recognized risk factors for melioidosis.1,2

It has been suggested that melioidosis may be the presenting feature of diabetes, and in areas of endemic melioidosis, diabetics are at higher risk of developing melioidosis, especially systemic disease, than septicemia caused by other bacteria.21 In Thailand and Australia the estimated relative risk of melioidosis in diabetic patients has been reported to be 13.1 and 5.9, respectively.1,2 It has been established that competent neutrophil function and innate immunity are critical in providing host immunity to B. pseudomal-lei.30,34 Melioidosis has also been documented in patients with chronic granulomatous disease, in whom neutrophil function is impaired.35 A study from Thailand showed that polymorphonuclear leukocytes (PMNs) from diabetic subjects display impaired phagocytosis of B. pseudomallei, reduced migration in response to

interleukin-8, and inability to delay apoptosis, when compared to PMNs from non-diabetic persons.36 Neutrophil function is also impaired in alcoholics37 and in patients with chronic renal disease,38 and this probably explains their role as risk factors for melioidosis.

Endemic regions have reported varying rates of diabetes: 39% in the Darwin study,14 56.6% in a study from Malaysia,20 and 60.3% in Thailand21 (Table 4). In our study 75.8% of patients were diabetic, which appears to be the highest reported rate so far. Melioidosis could probably be looked upon as a specific opportunistic infection in a diabetic patient in this region. Age >40 years also appears to be a major risk factor for contracting melioidosis (75.8%). Chronic kidney disease as a significant risk factor for the development of bacteremic melioidosis (p = 0.003) correlates well with other studies,14 in which 78% of patients with CKD had B. pseudomallei bacteremia.

Melioidosis does not seem to be associated with HIV infection. A study from Thailand showed that the clinical presentation and outcome of melioidosis were similar in HIV-positive and HIVnegative patients.39 Therefore, cell-mediated immune mechanisms are probably not of critical importance in the immune response to B. pseudomallei. In our study only 2.1% of patients were HIV-positive.

The development of septic shock appears to be a strong predictor of mortality (p < 0.001) from melioidosis, and correlates well with other studies. In the Darwin study from Northern Australia, 50% of patients with septic shock due to B. pseudomallei died.14 An important observation in our study was that mortality was seen only in patients with bacteremia (p < 0.001). Non-bacteremic melioidosis appears to have a favorable outcome as compared to bacteremic melioidosis. This observation should help the physician to prioritize and be aggressive in the treatment of bacteremic melioidosis, which may be important in resource-poor settings. Such trends have also been observed in a previous study in which mortality was 20% in bacteremic patients and only 7% in non-bacteremic patients.14

Exposure to soil/water has been documented as a risk factor for the acquisition of melioidosis.1,2 Rice farmers constitute 81% of patients in Thailand. A study from Thailand showed that among diabetic patients, rice farmers had a six- to nine-times higher risk of developing melioidosis than non-diabetics and non-rice farmers. Impaired immunity due to diabetes and the extent of exposure to B. pseudomallei in the field probably act as synergistic risk factors for melioidosis.21 In northern Australia, 75% of patients had exposure to environmental B. pseudomallei through recreational activities and 18% through occupational activities.14 However in our study only 36.8% of patients had exposure to soil/water (Table 4). Hence, percutaneous inoculation may not be the main portal of entry for B. pseudomallei in this region.

Melioidosis is often looked upon as a latent infection, but if so, the question arises as to why there should be a seasonal variation even in those patients showing a chronic presentation, 70.4% of whom presented during the monsoon season. B. pseudomallei is an environmental organism that grows in a humid environment. Therefore it is likely that the majority of infections are not latent infections, but fresh infections. It has been proposed that the moist, humid conditions during the monsoon may have an exalting effect on pathogen virulence, and also increase the environmental B. pseudomallei load.19

Melioidosis is quite prevalent in this region and accounts for considerable morbidity and mortality. Melioidosis frequently presents along with comorbid illnesses. A high index of suspicion and good communication between physician and laboratory personnel is essential for a rapid and accurate diagnosis of melioidosis. Isolation in culture and subsequent identification take 5-6 days at the earliest. There is a definite need for rapid and reliable diagnostic tests for melioidosis to facilitate early diagnosis

and treatment, thereby reducing morbidity and mortality. PCR-based tests and serological tests developed from stock cultures of isolates from the present study are being planned to address this issue. Soil, water, and other environmental sources need to be investigated in order to develop optimal strategies for the control of melioidosis.

In conclusion, the epidemiology of melioidosis in the Indian subcontinent is similar to that in reported endemic areas with respect to the diverse clinical presentations, association with the monsoon season, and diabetes being a major risk factor. Differences include a higher proportion of musculoskeletal melioidosis, a higher frequency of dental abscesses following dental procedures, a higher proportion of melioidotic lymphade-nopathy, and fewer patients with a history of exposure to soil/ water in this study. The presence of septic shock is a strong predictor of mortality. Bacteremic melioidosis has a poorer prognosis than non-bacteremic melioidosis.

Acknowledgements

The authors wish to thank Manipal University and the Indian Council of Medical Research for funding the study. The funding agencies had no role in the study design, in the collection, analysis, and interpretation of data, in the writing of the manuscript, or in the decision to submit the manuscript for publication. Conflict of interest: No conflict of interest to declare.

References

1. Cheng AC, Currie BJ. Melioidosis: epidemiology, pathophysiology and management. Clin Microbiol Rev 2005;18:383-416.

2. White NJ. Melioidosis. Lancet 2003;361:1715-22.

3. JohnTJ. Melioidosis, the mimicker of maladies. Indian J Med Res 2002;119. vi-viii.

4. Jesudason MV, Anbarasu A, John TJ. Septicaemic melioidosis in a tertiary care hospital in south India. Indian J Med Res 2003;117:119-21.

5. Kanungo R, Padhan P, Bhattacharya S, Srimannarayana J, Jayanthi S, Swami-nathan RP. Melioidosis—a report from Pondicherry, South India. J Assoc Physicians India 2002;50:1438-9.

6. Rao PS, Dhawan R, Shivananda PG. Burkholderia pseudomallei infections. Trop Doct 2002;32:174-5.

7. Vidyalakshmi K, Shrikala B, Bharathi B, Suchitra U. Melioidosis: an under-diagnosed entity in western coastal India: a clinico-microbiological analysis. Indian J Med Microbiol 2007;25:245-8.

8. Vidyalakshmi K, Chakrapani M, Shrikala B, Damodar S, Lipika S, Vishal S. Tuberculosis mimicked by melioidosis. Int J Tuberc Lung Dis 2008;12:1209-15.

9. Ramachandran A, Snehalatha C, Baskar AD, Mary S, Kumar CK, Selvam S, et al. Temporal changes in prevalence of diabetes and impaired glucose tolerance associated with lifestyle transition occurring in the rural population in India. Diabetologia 2004;47:860-5.

10. Mohan V, Deepa R, Deepa M, Somannavar S, Datta M. A simplified Indian diabetes risk score for screening for undiagnosed diabetic subjects. J Assoc Physicians India 2005;53:759-63.

11. Gilligan PH, Lum G, Vandamme PA, Whittier S. Burkholderia, Stenotrophomonas, Ralstonia, Brevundimonas, Comamonas, Delftia, Pandoraea and Acidovorax. In: Murray PR, Baron EJ, Jorgensen JH, Pfaller MA, Yolken RH, editors. Manual of clinical microbiology. Washington DC: ASM Press; 2003. p. 729.

12. Ashdown LR. An improved screening technique for isolation of Pseudomonas pseudomallei from clinical specimens. Pathology 1979;11:293-7.

13. India Meteorological Department. Districtwise rainfall information. Available at: http://www.imd.gov.in/section/hydro/distrainfall/districtrain.html (accessed September 25, 2011 and September 1, 2010).

14. Currie BJ, Ward L, Cheng AC. The epidemiology and clinical spectrum of melioidosis: 540 cases from the 20 year Darwin prospective study. PLoS Negl Trop Dis 2010;4:e900.

15. Genuth S, Alberti KG, Bennett P, Buse J, Defronzo R, Kahn R, et al. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Follow-up report on the diagnosis of diabetes mellitus. Diabetes Care 2003;26:3160-7.

16. National Institute on Alcohol Abuse and Alcoholism. The physicians' guide to helping patients with alcohol problems. NIH publication 95-3769. Washington DC, USA: Government Printing Office; 1995.

17. Levey AS, Eckardt KU, Tsukamoto Y, Levin A, CoreshJ, Rossert J, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2005;67:2089-100.

18. Suputtamongkol Y, Hall AJ, Dance DA, Chaowagul W, Rajchanuvong A, Smith MD, et al. The epidemiology of melioidosis in Ubon Ratchatani, northeast Thailand. Int J Epidemiol 1994;23:1082-90.

19. Currie BJ, Jacups SP. Intensity of rainfall and severity of melioidosis, Australia. Emerg Infect Dis 2003;9:1538-42.

20. 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.

21. Suputtamongkol Y, Chaowagul W, Chetchotisakd P, Lertpatanasuwun N, Intar-anongpai S, Ruchutrakool T, et al. Risk factors for melioidosis and bacteremic melioidosis. Clin Infect Dis 1999;29:408-13.

22. Poe RH, Vassallo cL, Domm BM. Melioidosis: the remarkable imitator. Am Rev RespirDis 1971;104:427-31.

23. Loveleena, Chaudhry R, Dhawan B. Melioidosis; the remarkable imitator: recent perspectives. J Assoc Physicians India 2004;52:417-20.

24. Currie BJ. Melioidosis: an important cause of pneumonia in residents of and travellers returned from endemic regions. Eur Respir J 2003;22:542-50.

25. Raja NS. Melioidotic septic arthritis: a case report and literature review. J Microbiol Immunol Infect 2007;40:178-82.

26. Chlebicki MP, Tan BH. Six cases of suppurative lymphadenitis caused by Burkholderia pseudomallei infection. Trans R Soc Trop Med Hyg 2006;100: 798-801.

27. Lim WK, Gurdeep GS, Norain K. Melioidosis of the head and neck. MedJ Malaysia 2001;56:471-7.

28. Arakawa M, Mitsui T, Miki R, Yabuuchi E. Chronic melioidosis: a report of the first case in Japan. Kansenshogaku Zasshi 1993;67:154-62.

29. Mathew S, Perakath B, Mathew G, Sitaram V, Nair A, Lalitha MK, et al. Surgical presentation of melioidosis in India. Natl Med J India 1999;12:59-61.

30. 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;31:981-6.

31. Limmathurotsakul D, Chaowagul W, Chierakul W, Stepniewska K, Maharjan B, Wuthiekanun V, et al. Risk factors for recurrent melioidosis in northeast Thailand. Clin Infect Dis 2006;43:979-86.

32. Currie BJ, Fisher DA, Anstey NM, Jacups SP. Melioidosis: acute and chronic disease, relapse and re-activation. Trans R Soc Trop Med Hyg 2000;94:301-4.

33. Maharjan B, Chantratita N, Vesaratchavest M, Cheng A, Wuthiekanun V, Chier-akul W, et al. Recurrent melioidosis in patients in northeast Thailand is frequently due to reinfection rather than relapse. J Clin Microbiol 2005;43:6032-4.

34. Easton A, Haque A, Chu K, Lukaszewski R, Bancroft GJ. A critical role for neutrophils in resistance to experimental infection with Burkholderia pseudomallei. J Infect Dis 2007;195:99-107.

35. Dorman SE, Gill VJ, Gallin JI, Holland SM. Burkholderia pseudomallei infection in a Puerto Rican patient with chronic granulomatous disease: case report and review of occurrences in the Americas. Clin Infect Dis 1998;26:889-94.

36. Chanchamroen S, Kewcharoenwong C, Susaengrat W, Ato M, Lertmemongkol-chai G. Human polymorphonuclear neutrophil responses to Burkholderia pseu-domallei in healthy and diabetic subjects. Infect Immun 2009;77:456-63.

37. Brayton RG, Stokes PE, Schwartz MS, Louria DB. Effect of alcohol and various diseases on leukocyte mobilization, phagocytosis and intracellular bacterial killing. N Engl J Med 1970;282:123-8.

38. Heinzelmann M, Mercer-Jones MA, Passmore JC. Neutrophils and renal failure. Am J Kidney Dis 1999;34:384-99.

39. Chierakul W, Wuthiekanun V, Chaowagul W, Amornchai P, Cheng AC, White NJ, et al. Short report: disease severity and outcome of melioidosis in HIV coin-fected individuals. Am J Trop Med Hyg 2005;73:1165-6.