Summary
Objectives
To better understand the characteristics of patients with acute melioid community-acquired pneumonia (CAP) on emergency department (ED) arrival, and the risk factors in patients with acute melioid CAP that differ from those in patients with severe CAP of causes other than melioidosis.
Methods
This was a retrospective case–control study.
Results
During the study period, a total of 15 patients suffered from acute melioid CAP. Comparison with 60 patients with severe CAP of causes other than melioidosis, revealed that visit to the ED in the rainy season, shock on arrival, diabetes, poor sugar control with glycemia ≥250 mg/dl, chest radiograph with cavity formation, and poor clinical outcome, were significantly predominant in patients with acute melioid CAP. Multivariate logistic regression analysis indicated that poor sugar control with glycemia ≥250 mg/dl (odds ratio (OR) 38.3, 95% confidence interval (CI) 3.6–406.2; p < 0.01), visiting the ED during the rainy season (OR 13.7, 95% CI 2.3–80.9; p < 0.01), and shock on ED arrival (OR 18.7, 95% CI 1.8–192.8; p = 0.01) were independent risk factors for patients with CAP caused by Burkholderia pseudomallei.
Conclusions
Physicians in melioidosis endemic areas should administer antimicrobials covering B. pseudomallei to patients with CAP who visit the ED during the rainy season, who have poor sugar control with glycemia ≥250 mg/dl, and who are in shock on ED arrival, to facilitate timely, appropriate antibiotic therapy and lower the mortality rate.
Keywords
1. Introduction
Although melioidosis is an endemic disease in Southeast Asia and northern Australia, it has emerged as a global problem due to an increase in worldwide travel.
1
This infection usually occurs following percutaneous inoculation or inhalation of Burkholderia pseudomallei present in the soil and surface water in endemic regions.2
Due to the diverse clinical manifestations of melioidosis, it is nicknamed ‘the great mimicker’.3
, 4
Pulmonary melioidosis is the most common clinical presentation of melioidosis,2
, 5
, 6
and it usually presents as community-acquired pneumonia (CAP). The mortality rate of pulmonary melioidosis is high if not promptly diagnosed and treated with effective antibiotic therapy.5
, 6
, 7
, 8
, 9
It would be useful to determine whether patients are more likely to have acute melioid CAP at an early stage. Therefore, we conducted a retrospective study to better understand the demographics and clinical and laboratory characteristics of patients with acute melioid CAP arriving at the emergency department (ED).2. Patients and methods
Data were retrospectively collected for patients who visited the ED of Kaohsiung Chang Gung Memorial Hospital (KCGMH), whose microbiological cultures were positive for B. pseudomallei, during the period January 2001 to December 2010. KCGMH is a 2500-bed facility that serves as a primary care and tertiary referral center in southern Taiwan. The study was approved by our institutional review board (No. 97-0599B). The presence of at least one clinical criterion (fever, dyspnea, cough, leukocytosis), findings suggestive of an infiltrate on chest radiograph, and the disease contracted in the community, were needed to make a diagnosis of CAP. Controls were retrieved from a computer-aided selection of eligible individuals who had visited the ED with severe CAP due to causes other than melioidosis and who were admitted to the intensive care unit. Each subject was age-matched, and the ratio of cases of acute melioid CAP to control patients was 1:4.
Data retrieved from the medical records included: (1) demographics (i.e., age, sex, and whether the patient had visited the hospital during the rainy season), (2) coexisting conditions (i.e., neoplastic disease, chronic obstructive lung disease, diabetes mellitus (DM), renal disease, liver disease, excessive alcohol intake, and cerebrovascular accident), (3) physical examination (i.e., temperature, respiratory rate, heart rate, shock on arrival, and altered mental status), (4) laboratory tests (i.e., C-reactive protein (CRP), sodium, PaO2, hematocrit, blood urea nitrogen (BUN), and glucose), (5) chest radiograph (i.e., multilobar involvement, presence of cavity, and presence of pleural effusion), (6) antimicrobial treatments, and (7) final outcome. The rainy season in Taiwan occurs between June and September. Alcoholism was defined as more than two drinks (20 g) per day in women and more than three drinks (30 g) per day in men. Shock on arrival was defined as a systolic blood pressure <90 mmHg, or a decrease of more than 30 mmHg as compared to the baseline systolic blood pressure on arrival in the ED, coupled with clinical evidence of peripheral hypoperfusion. Altered mental status was defined as a new onset of confusion, delirium, obtundation, stupor, or coma. Empirical antibiotic therapy at adequate dosages was considered appropriate if the antibiotic was effective against B. pseudomallei,
2
which included imipenem, meropenem, ceftazidime, and amoxicillin–clavulanic acid. Early mortality was defined as death within 48 h after presentation to the ED, and before microbiological confirmation of the diagnosis. Overall mortality was defined as the sum of all patients with death as the final clinical outcome. The final clinical outcome was evaluated at the time of hospital discharge.The Fisher's exact test was used to assess differences in dichotomous variables between the two groups. Variables with a p-value of <0.1 in univariate analyses were separately entered into a multivariate logistic regression model to identify independent risk factors to differentiate acute melioid CAP from severe non-melioid CAP, and odds ratios (OR) and 95% confidence intervals (CI) were calculated. Variables with a two-tailed p-value of <0.05 in the multivariate logistic regression were considered statistically significant. Data were analyzed using SPSS for Windows version 10.0 (SPSS Inc., Chicago, IL, USA).
3. Results
A total of 36 patients with melioidosis, 28 males and eight females, all ≥18 years of age (mean age 58 years), were identified during the study period. Among them, 15 patients (42%), 13 males and two females with a mean age of 56 years, were diagnosed with acute melioid CAP. Demographics and clinical characteristics of the acute melioid CAP patients are shown in Table 1. All 15 patients had bacteremia, and the time lapses between obtaining blood and non-blood specimens for culture and the availability of results indicating B. pseudomallei infection ranged from 3 to 7 and 3 to 6 days, respectively. Ten patients died (overall mortality 66.7%), and seven (46.7%) died within 48 h of arrival at the ED. Inappropriate empiric antibiotic therapy was administered to 12 acute melioid CAP patients (80%). Only three patients (20%) received appropriate empirical antibiotics (two received meropenem and one received ceftazidime) before cultures were available, and the two who received meropenem as empirical therapy survived. A microbiological etiology of the severe non-melioid CAP was identified in 21 patients (35%). Streptococcus pneumoniae was the most common pathogen detected, and was found in eight (38.1%) of these 21 patients. Inappropriate empirical antibiotic therapy was administered to six of the 21 patients (28.6%). Blood cultures were performed in all these 60 patients, and 11 (18.3%) had positive cultures. A total of 23 patients died (overall mortality 38.3%), and 11 (18.3%) died within 48 h of arrival at the ED.
Table 1Demographic and clinical characteristics of 15 patients with acute melioid community-acquired pneumonia
| Number/gender/age (years)/month/year | Underlying disease/condition | Positive culture (days after admission) | Shock on arrival | Rainy season visit | Outcome (days of survival after ED arrival)/ mechanical ventilation |
|---|---|---|---|---|---|
| 1/M/60/June/2001 | DM, alcoholism | Blood (3), urine (4) | Yes | Yes | Died (2)/yes |
| 2/M/58/Aug/2001 | DM | Blood (7) | No | Yes | Died (47)/yes |
| 3/M/87/July/2002 | Stroke | Blood (6) | Yes | Yes | Survived/yes |
| 4/M/60/Sep/2002 | DM, LC, HCC | Blood (5) | Yes | Yes | Died (2)/yes |
| 5/M/50/Sep/2002 | DM, alcoholism | Blood (4), sputum (3) | No | Yes | Died (16)/no |
| 6/M/41/Feb/2003 | Alcoholism, CRF | Blood (6) | No | No | Died (1)/yes |
| 7/M/35/June/2004 | DM, LC, HCC | Blood (5) | Yes | Yes | Died (1)/yes |
| 8/M/59/Aug/2004 | DM | Blood (5), sputum (6) | No | Yes | Survived/no |
| 9/M/65/May/2007 | DM, CAD, hypertension | Blood (4) | Yes | No | Died (1)/yes |
| 10/F/52/June/2008 | DM | Blood (4) | No | Yes | Survived/no |
| 11/M/58/July/2008 | LC | Blood (3), sputum (3) | Yes | Yes | Died (2)/yes |
| 12/F/47/May/2009 | DM | Blood (4), urine (3) | Yes | No | Survived/no |
| 13/M/66/June/2009 | Alcoholism, CRF, COPD | Blood (4), sputum (3) | Yes | Yes | Died (1)/yes |
| 14/M/58/June/2010 | DM, LC | Blood (4) | Yes | Yes | Survived/yes |
| 15/M/51/Oct/2010 | DM | Blood (3) | No | No | Died (22)/no |
CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; CRF, chronic renal failure; DM, diabetes mellitus; ED, emergency department; F, female; HCC, hepatocellular carcinoma; LC, liver cirrhosis; M, male.
a Three patients received appropriate empirical antibiotics before cultures were available, and two survived.
b Early mortality.
Comparisons of the demographic characteristics, clinical presentations, coexisting conditions, laboratory test results, and chest radiograph findings on arrival at the ED between the two groups are shown in Table 2. Univariate analysis revealed that acute melioid CAP patients were more likely to: (1) have visited the ED during the rainy season, (2) have DM, (3) have poor sugar control with glycemia ≥250 mg/dl, (4) have shock on arrival at the ED, and (5) have evidence of cavity formation on chest radiograph. Inappropriate empirical antibiotic therapy with early and overall mortality was significantly greater in patients with acute melioid CAP. Multivariate logistic regression analysis indicated that glycemia ≥250 mg/dl (OR 38.3, 95% CI 3.6–406.2; p < 0.01), visiting the ED during the rainy season (OR 13.7, 95% CI 2.3–80.9; p < 0.01), and in shock on ED arrival (OR 18.7, 95% CI 1.8–192.8; p = 0.01) were independent risk factors for acute melioid CAP on ED arrival.
Table 2Comparisons of differences between patients with acute melioid community-acquired pneumonia and patients with severe community-acquired pneumonia of causes other than melioidosis, on arrival in the emergency department
| Variable | Patients with acute melioid CAP (n = 15), n (%) | Patients with severe CAP of causes other than melioidosis (n = 60), n (%) | p-Value |
|---|---|---|---|
| Demographics | |||
| Male gender | 13 (86.7) | 40 (66.7) | 0.21 |
| Rainy season visit a Multivariate logistic regression analysis indicated that poor sugar control and glycemia ≥250mg/dl (OR 38.3, 95% CI 3.6–406.2; p<0.01), visiting the ED during rainy season (OR 13.7, 95% CI 2.3–80.9; p<0.01), and in shock on ED arrival (OR 18.7, 95% CI 1.8–192.8; p=0.01) were independent risk factors for CAP caused by Burkholderia pseudomallei. | 11 (73.3) | 8 (13.3) | <0.01 |
| Coexisting conditions | |||
| Malignancy | 2 (13.3) | 7 (11.7) | 1 |
| COPD | 1 (6.7) | 19 (31.6) | 0.06 |
| DM | 11 (73.3) | 22 (36.7) | 0.02 |
| Poor sugar with glycemia ≥250 mg/dl a Multivariate logistic regression analysis indicated that poor sugar control and glycemia ≥250mg/dl (OR 38.3, 95% CI 3.6–406.2; p<0.01), visiting the ED during rainy season (OR 13.7, 95% CI 2.3–80.9; p<0.01), and in shock on ED arrival (OR 18.7, 95% CI 1.8–192.8; p=0.01) were independent risk factors for CAP caused by Burkholderia pseudomallei. | 10 (66.7) | 6 (10) | <0.01 |
| Renal disease | 2 (13.3) | 8 (13.3) | 1 |
| Liver disease | 4 (26.7) | 8 (13.3) | 0.24 |
| Alcoholism | 4 (26.7) | 9 (15) | 0.28 |
| Cerebrovascular accident | 1 (6.7) | 7 (11.7) | 1 |
| Physical examination | |||
| Temperature ≥40 °C or ≤35 °C | 10 (66.7) | 26 (43.3) | 0.15 |
| Respiratory rate ≥30 breaths/min | 6 (40) | 15 (25) | 0.34 |
| Heart rate ≥125 beats/min | 6 (40) | 22 (36.7) | 1 |
| Shock on arrival a Multivariate logistic regression analysis indicated that poor sugar control and glycemia ≥250mg/dl (OR 38.3, 95% CI 3.6–406.2; p<0.01), visiting the ED during rainy season (OR 13.7, 95% CI 2.3–80.9; p<0.01), and in shock on ED arrival (OR 18.7, 95% CI 1.8–192.8; p=0.01) were independent risk factors for CAP caused by Burkholderia pseudomallei. | 9 (60) | 14 (23.3) | 0.01 |
| Altered mental status | 3 (20) | 8 (13.3) | 0.68 |
| Initial laboratory data | |||
| Sodium <130 mmol/l | 4 (26.7) | 8 (13.3) | 0.24 |
| PaO2 <60 mm Hg | 7 (46.7) | 27 (45) | 1 |
| Band WBC >1% | 7 (46.7) | 16 (26.7) | 0.21 |
| Hematocrit <30% | 5 (33.3) | 10 (16.7) | 0.16 |
| BUN >30 mg/dl | 5 (33.3) | 16 (26.7) | 0.75 |
| CRP >100 mg/dl | 10 (76.9) | 20 (50) | 0.12 |
| Chest radiograph findings | |||
| Multilobar infiltrates | 10 (66.7) | 30 (50) | 0.39 |
| Presence of a lung cavity | 7 (46.7) | 8 (13.3) | <0.01 |
| Presence of pleural effusion | 2 (13.3) | 12 (20) | 0.72 |
| Inappropriate empirical antibiotic therapy | 12 (80) | 6 (28.6) | <0.01 |
| Clinical outcomes | |||
| Mechanical ventilation | 10 (66.7) | 32 (53.3) | 0.39 |
| Early mortality | 7 (46.7) | 11 (18.3) | 0.04 |
| Overall mortality | 10 (66.7) | 23 (38.3) | 0.02 |
BUN, blood urea nitrogen; CAP, community-acquired pneumonia; COPD, chronic obstructive pulmonary disease; CRP, C-reactive protein; DM, diabetes mellitus; WBC, white blood cell.
a Multivariate logistic regression analysis indicated that poor sugar control and glycemia ≥250 mg/dl (OR 38.3, 95% CI 3.6–406.2; p < 0.01), visiting the ED during rainy season (OR 13.7, 95% CI 2.3–80.9; p < 0.01), and in shock on ED arrival (OR 18.7, 95% CI 1.8–192.8; p = 0.01) were independent risk factors for CAP caused by Burkholderia pseudomallei.
b Number of patients available for analysis, n = 13.
c Number of patients available for analysis, n = 40.
d Inappropriate empirical antibiotic therapy was found in six (28.6%) of 21 patients with severe CAP from causes other than melioidosis in whom an etiology was identified.
4. Discussion
B. pseudomallei has been detected in Taiwan,
10
and it is probable that southern Taiwan is an area endemic for melioidosis. Su et al.11
reported that the annual incidence of melioidosis in the Er-Ren river basin (located in southern Taiwan) in 2005 was 70 cases/100 000 persons, which is higher than that in areas where melioidosis is endemic. In this study, we reported 15 sporadic cases of acute melioid CAP, and 80% of these patients were empirically treated with quinolones, β-lactam/β-lactamase inhibitors, or extended non-antipseudomonal cephalosporins, which are the recommended antimicrobial regimens for CAP according to the current treatment guidelines developed by the Infectious Diseases Society of America and the American Thoracic Society.12
Although amoxicillin–clavulanic acid is effective against B. pseudomallei in vitro,2
the clinical condition of the three patients who received amoxicillin–clavulanic acid therapy deteriorated rapidly, and they all died within 24 h after ED arrival. An inadequate dosage (1200 mg every 8 h) of amoxicillin–clavulanic acid might be one of the reasons for treatment failure. An increased dosage (amoxicillin–clavulanic acid 20/4 mg/kg intravenous every 4 h) for the treatment of melioidosis to ensure adequate levels of clavulanic acid has been suggested.13
The early mortality rate of our patients with acute melioid CAP was 46.7%, which is much higher than the 18.3% mortality rate of patients with non-melioid CAP, and the mortality rate reported in another study.
14
Pulmonary melioidosis is one of the common causes of CAP in Southeast Asia. In one study from Singapore, the incidence of CAP caused by B. pseudomallei was 12.9% among 31 patients with CAP in which organisms were identified.15
It might not be wise to follow the general guidelines for the treatment of CAP in patients with the risk factors for melioid CAP. It is difficult to differentiate CAP caused by B. pseudomallei from that caused by other pathogens in the early diagnosis on ED arrival, on the basis of clinical presentation and chest radiograph findings (Table 2). Similar to the cases reported in previous studies,1
, 2
, 5
, 16
DM was the most common underlying disease in patients with acute melioid CAP. It is hypothesized that patients with poor sugar control may have impaired human polymorphonuclear neutrophil migration and apoptosis, which explains why poor sugar control could be a risk factor for acute melioid CAP.17
It is postulated that B. pseudomallei migrates from the deeper soil layers to the surface with the rising water table in the rainy season, thus explaining the association of this season with the incidence of melioidosis.2
The organism is more common at soil depths of 30–60 cm than on the soil surface.10
It is possible that heavy rainfall with strong winds brings buried B. pseudomallei isolates to the soil surface, and thereafter they are more likely to be inhaled at high concentrations and cause subsequent pulmonary melioidosis. Thus, clinicians in melioidosis endemic areas should administer antimicrobials covering B. pseudomallei to patients with CAP who visit the ED during the rainy season, who have poor sugar control with glycemia ≥250 mg/dl, and who are in shock on ED arrival.Early diagnosis is vital if the patients with melioidosis are to receive effective treatment.
18
, 19
Definitive laboratory diagnosis of melioidosis is based on the isolation of B. pseudomallei. In our study, the mean duration of time that elapsed before obtaining culture results confirming B. pseudomallei infection was 5 days (range 3–7 days). It is noteworthy that 46.7% of the deaths in this study occurred before a confirmed culture was available. Direct microscopy using Gram staining may be helpful if typical slender, pale, bipolar-staining bacilli are observed; however, this technique lacks specificity.20
Rapid targeted microbiological work-up, such as direct microscopy with direct immunofluorescence21
and urinary antigen enzyme-linked immunosorbent assays,22
may allow a rapid provisional diagnosis, and thus enable administration of appropriate antimicrobial therapy. In 2008, we found that patients with poor sugar control who visited the ED during the rainy season were at risk of having acute melioid CAP. ED physicians were inclined to prescribe antibiotics against B. pseudomallei to CAP patients with hyperglycemia in the rainy season, and also to those with severe sepsis, until culture results were available. As a result, three (50%) of the six acute melioid CAP patients enrolled during the period from 2008 to 2010 received appropriate empirical antibiotics before a positive culture of B. pseudomallei was confirmed. Both overall (66.7% vs. 33.3%) and early (50% vs. 33.3%) mortality rates were higher in patients who received inappropriate empirical antibiotic therapy than in those who received appropriate empirical antibiotic therapy. This result is similar to those of several other studies19
, 23
in which inappropriate empirical antibiotic therapy has been clearly shown to be an independent risk factor for mortality in community-acquired as well as nosocomial-acquired bacteremia. Our report describes the outcomes of individuals presenting with acute melioid CAP and the consequences of a wrong antibiotics choice. It provides intriguing insights for assessing the probability that a patient has acute melioid CAP in a region where melioidosis is known to be endemic, albeit uncommon.Our study has several limitations. First, this paper describes the hospital course of individuals presenting with acute melioid CAP at a single institution. The sample size had adequate statistical power to analyze these risk factors, but CIs were large, which is likely to be due to the small number of patients. Second, the retrospective nature of the study inevitably rendered incomplete information collection regarding the time of diagnosis and the time of initiation of effective antimicrobial therapy in these patients. Nonetheless, this study provides intriguing insights for assessing the probability that a patient has acute melioid CAP early on ED arrival.
In conclusion, patients with a presentation of CAP who have concomitant poor sugar control, present during the rainy season, and are in shock upon ED arrival, are at an increased risk of having acute melioid CAP. Taking into account these findings allows appropriate therapy to be started, and may reduce the high mortality of this condition. We suggest that physicians in areas in which B. pseudomallei is endemic administer antibiotics for the treatment of melioidosis to patients with CAP who have the aforementioned risk factors, until culture results are available.
Conflict of interest: The authors have no conflict of interest to declare.
References
- Melioidosis as an emerging global problem.Acta Trop. 2000; 74: 115-119
- Melioidosis: epidemiology, pathophysiology, and management.Clin Microbiol Rev. 2005; 18: 383-416
- The radiology of melioidosis.Australas Radiol. 1996; 40: 244-249
- Melioidosis, the great mimicker: a report of 10 cases from Malaysia.J Trop Med Hyg. 1988; 91: 249-254
- Endemic melioidosis in tropical northern Australia: a 10-year prospective study and review of the literature.Clin Infect Dis. 2000; 31: 981-986
- Bacteraemic melioidosis pneumonia: impact on outcome, clinical and radiological features.J Infect. 2004; 48: 334-338
- Melioidosis: a major cause of community-acquired septicemia in northeastern Thailand.J Infect Dis. 1989; 159: 890-899
- Septicaemic melioidosis: a review of 50 cases from Malaysia.Trans R Soc Trop Med Hyg. 1992; 86: 683-685
- Clinical characteristics and outcome of severe melioidosis requiring intensive care.Chest. 2005; 128: 3674-3678
- Detection of Burkholderia pseudomallei in rice fields with PCR-based technique.Folia Microbiol (Praha). 2003; 48: 521-524
- Prevalence of melioidosis in the Er-Ren river basin, Taiwan: implications for transmission.J Clin Microbiol. 2007; 45: 2599-2603
- Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults.Clin Infect Dis. 2007; 44: S27-72
- The activity of amoxicillin/clavulanic acid against Pseudomonas pseudomallei.J Antimicrob Chemother. 1989; 24: 1012-1014
- Antibiotic therapy and 48-h mortality for patients with pneumonia.Am J Med. 2006; 119: 859-864
- Severe community-acquired pneumonia requiring intensive care: a study of 80 cases from Singapore.Singapore Med J. 2008; 49: 458-461
- Bacteremic melioidosis in southern Taiwan: clinical characteristics and outcome.J Formos Med Assoc. 2007; 106: 1013-1022
- Human polymorphonuclear neutrophil responses to Burkholderia pseudomallei in healthy and diabetic subjects.Infect Immun. 2009; 77: 456-463
- Pseudomonas bacteremia. Review of 108 cases.Am J Med. 1976; 60: 501-508
- Community-acquired bloodstream infection in critically ill adult patients: impact of shock and inappropriate antibiotic therapy on survival.Chest. 2003; 123: 1615-1624
- The laboratory diagnosis of melioidosis.Br J Biomed Sci. 1996; 3: 249-253
- Rapid immunofluorescence microscopy for diagnosis of melioidosis.Clin Diagn Lab Immunol. 2005; 12: 555-556
- Detection of Pseudomonas pseudomallei antigen in urine for the diagnosis of melioidosis.Am J Trop Med Hyg. 1994; 51: 627-633
- The influence of inadequate antimicrobial treatment of bloodstream infections on patient outcomes in the ICU setting.Chest. 2000; 118: 146-155
Article info
Publication history
Published online: June 22, 2011
Accepted:
April 28,
2011
Received:
March 31,
2011
Corresponding Editor: Timothy Barkham, Tan Tock Seng, Singapore.Identification
Copyright
© 2011 International Society for Infectious Diseases. Published by Elsevier Inc.
User license
Elsevier user license | How you can reuse 
Elsevier's open access license policy
Elsevier user license
Permitted
For non-commercial purposes:
- Read, print & download
- Text & data mine
- Translate the article
Not Permitted
- Reuse portions or extracts from the article in other works
- Redistribute or republish the final article
- Sell or re-use for commercial purposes
Elsevier's open access license policy
