Melioidosis with non-necrotising diffuse cellulitis and intramuscular abscesses of the trunk mimicking necrotising soft tissue infection: a case report

Abstract

Background

Melioidosis is endemic in Southeast Asia and northern Australia and can resemble various conditions. Soft tissue involvement is usually limited to localised cellulitis or superficial abscesses. Our case presented with diffuse cellulitis and deep intramuscular abscesses of the thoracoabdominal wall—an exceedingly rare manifestation mimicking necrotising soft tissue infection, a surgical emergency. Reporting such cases is essential to prevent misdiagnosis and unnecessary surgery in endemic settings.

Case presentation

A man in his sixties from southern Vietnam with previously undiagnosed diabetes mellitus presented with right upper quadrant and flank pain, high fever, and hypotension. Laboratory tests showed leukocytosis and elevated inflammatory markers. Ultrasound revealed gallstones and a positive Murphy’s sign. He was diagnosed with acute calculous cholecystitis and septic shock, and promptly started on intravenous meropenem before emergency laparoscopic cholecystectomy. However, his back pain progressively intensified after admission and became severe and disproportionate to examination findings. Within 30 hours, he developed rapidly spreading erythema over the flank, back, buttock, and thigh. Necrotising soft tissue infection (NSTI) was suspected, with a Laboratory Risk Indicator for Necrotising Fasciitis (LRINEC) score of 7. Despite concern, he remained haemodynamically stable and inflammatory markers declined. MRI revealed diffuse cellulitis and multiple intramuscular abscesses involving thoracoabdominal and thigh muscles, including a large abscess in the right posterior thoracoabdominal wall. Blood cultures grew Burkholderia pseudomallei, confirming melioidosis. Meropenem was continued for four weeks, and oral trimethoprim–sulfamethoxazole was introduced. No surgical drainage was required. By day 6, pain and erythema improved. Follow-up MRI on day 10 showed reduced abscess size and improved cellulitis. At two-week follow-up, the patient remained asymptomatic, bedside ultrasound confirmed resolution, and he continued oral therapy to complete the three-month course.

Conclusions

This case underscores the importance of considering melioidosis in patients from endemic areas presenting with rapidly progressive soft tissue infections that mimic necrotising soft tissue infection. Early recognition and a structured diagnostic approach can help distinguish non-necrotising forms and avoid unnecessary surgery.

Clinical trial

Not applicable.

Background

Melioidosis, caused by the environmental Gram-negative bacterium Burkholderia pseudomallei, is endemic in Southeast Asia and northern Australia—including parts of southern Vietnam—with an estimated global burden of up to 165,000 cases and 89,000 deaths annually [1, 2]. Known as “the great mimicker,” it presents with diverse clinical manifestations that often complicate diagnosis [1, 3]. Cutaneous involvement is typically limited to localised cellulitis or superficial abscesses following direct inoculation [4, 5]. Diffuse cellulitis with deep intramuscular abscesses of the thoracoabdominal wall is an exceedingly rare manifestation of melioidosis. This presentation may mimic necrotising soft tissue infection (NSTI), which requires urgent surgical debridement [6, 7]. In contrast, melioidosis-related soft tissue infections can respond well to antibiotics, infrequently progress to necrosis when treated promptly, and may not require drainage in selected cases [4, 8]. This case report aims to raise clinical awareness of this uncommon manifestation of melioidosis, emphasising its potential to mimic NSTI and lead to inappropriate surgical intervention. It also contributes to the broader understanding of melioidosis-related soft tissue infections in endemic regions.

Case presentation

A previously well man in his 60s, a manufacturing manager residing in southern Vietnam, presented to the emergency department with a two-day history of right upper quadrant and flank pain and high-grade fever. He denied any recent trauma, skin abrasions, or contact with floodwater. The patient reported occasional alcohol use but did not meet criteria for hazardous alcohol consumption. He had no known chronic medical conditions; however, his HbA1c on admission was 7.38%, suggestive of undiagnosed diabetes mellitus.

On arrival, he was febrile and hypotensive, requiring noradrenaline support after fluid resuscitation. Physical examination revealed right flank tenderness and a positive Murphy’s sign. Laboratory tests showed leukocytosis (WBC 13 × 10⁹/L; normal 3.6–10.2 × 10⁹/L, neutrophils 93%), elevated CRP (257 mg/L; normal < 5), and procalcitonin (1.42 ng/mL; normal < 0.5), with normal liver and renal function. Bedside abdominal ultrasound demonstrated a distended gallbladder (4 × 8 cm) with multiple gallstones, thin walls (3–4 mm), no pericholecystic fluid, and a positive sonographic Murphy’s sign. He was diagnosed with acute calculous cholecystitis, for which two sets of blood cultures were obtained, intravenous meropenem (1 g every 8 h) was initiated, and emergency laparoscopic cholecystectomy was performed. Intraoperatively, the gallbladder wall appeared thin with mild bile staining; turbid bile was aspirated but cultures were negative. The surgical site healed without complication, and there were no signs of postoperative infection.

Approximately 30 h after admission, the patient developed worsening right flank and back pain that was severe and disproportionate to examination findings. Analgesia was escalated to intravenous fentanyl. Mild erythema and warmth were noted over the right flank and buttock, which rapidly extended within four hours to the contralateral lower back, right buttock, and thigh (Fig. 1). The rapid progression and severe pain raised concern for NSTI. Accordingly, the LRINEC score was calculated and found to be 7, based on WBC 8.3 × 10⁹/L, CRP 151 mg/L, sodium 131 mmol/L, creatinine 70 µmol/L, haemoglobin 11.6 g/dL, and glucose 8 mmol/L. Creatine kinase was mildly elevated at 237 U/L (normal < 171). Empirical linezolid was administered as a single dose to cover toxin-producing Gram-positive organisms. A bedside soft tissue ultrasound showed subcutaneous oedema and suspected small fluid collections, but was inconclusive. Although local pain and erythema were worsening, the patient had improved systemically—haemodynamically stable, off vasopressors, and with declining inflammatory markers (WBC 13 × 10⁹/L → 8.3 × 10⁹/L; CRP 257 → 151 mg/L); therefore, a gadolinium-enhanced MRI was obtained. This revealed diffuse cellulitis and multiple intramuscular abscesses involving the paraspinal, thoracoabdominal, and thigh musculature, including a large abscess (34 × 69 × 60 mm) in the right posterior thoracoabdominal wall (Fig. 2). There was no evidence of fascial gas, fascial thickening, or abnormal tracking along fascial planes. Both sets of blood cultures became positive after 34 h of incubation, growing Burkholderia pseudomallei identified using the VITEK 2 compact system (bioMérieux, France; Excellent identification score). The result was confirmed by typical colony morphology (wrinkled, dry, metallic sheen), oxidase positivity, and a biochemical profile consistent with B. pseudomallei. Antimicrobial susceptibility testing (AST) was performed using the European Committee on Antimicrobial Susceptibility Testing (EUCAST) disc diffusion method, showing susceptibility to amoxicillin–clavulanate, trimethoprim–sulfamethoxazole (TMP–SMX), imipenem, and tetracycline.

Fig. 1.

Clinical photographs showing extensive skin involvement. (a–b) Erythema over the right flank (a) and right thigh (b), outlined with skin marker at 34 h post-admission. White arrows indicate the leading edge of erythema. The rapid expansion raised clinical concern for necrotising soft-tissue infection

Fig. 2.

Baseline MRI showing cellulitis and intramuscular abscesses. (a) Coronal T2 fat-suppressed image shows widespread hyperintense signal in the subcutaneous tissue and musculature of the right thoracoabdominal wall, consistent with diffuse cellulitis. White arrow indicates the area of subcutaneous oedema and muscle involvement. (b) Axial T2 image reveals a large intramuscular abscess in the posterior thoracoabdominal wall, approximately 34 × 69 × 60 mm. White arrow marks the abscess cavity. (c) Sagittal T2 image of the same lesion demonstrates rim enhancement and a non-enhancing central component. White arrow indicates the abscess wall showing peripheral enhancement. (d) Axial T2 image of the right thigh shows two additional rim-enhancing intramuscular abscesses. White arrows indicate both abscesses

The diagnosis of melioidosis with diffuse cellulitis and intramuscular abscesses was established. Meropenem was continued at the same dosage as on admission. Oral TMP–SMX was introduced using a graded dosing regimen. Neurological examination throughout the admission revealed no headache, altered mental status, neck stiffness, seizures, focal weakness, cranial nerve palsies, or other neurological signs. The patient had no respiratory symptoms, chest radiography was normal, and MRI showed no prostatic abscess. By day 4, pain control had improved and erythema had stabilised, despite extending to the right lower leg. By day 6, there was marked improvement with resolution of pain and regression of erythema and warmth. A follow-up MRI on day 10 showed a reduction in the size of the largest abscess (now 19 × 63 × 53 mm) and improvement in cellulitis (Fig. 3). The patient completed a 4-week course of intravenous meropenem and was discharged in stable condition with normalised inflammatory markers and resolution of symptoms. He was advised to continue oral TMP–SMX (160/800 mg, two tablets twice daily) for a planned 3-month eradication phase. No surgical drainage or debridement was required during hospitalisation. At a clinic review two weeks later, he had returned to work and remained asymptomatic; bedside ultrasound confirmed resolution of all soft-tissue abscesses. At one month, ultrasound again showed no residual fluid or oedema. He then completed a three-month eradication phase with TMP–SMX at full dose without adverse effects or recurrence. The patient’s clinical trajectory—including symptom evolution, investigations, imaging findings, and sequential management decisions—is summarised in Table 1.

Fig. 3.

Follow-up MRI on day 10 demonstrating resolution of abscesses and cellulitis. (a) Coronal T2 fat-suppressed image at the same level as Fig. 2a shows markedly reduced oedema in the right thoracoabdominal wall. White arrow indicates the site of previous oedema. (b) Axial image at the same level as Fig. 2b shows decreased size of the intramuscular abscess with minimal residual inflammation. White arrow indicates the shrunken abscess cavity

Table 1.

Timeline of clinical events and treatments

Timepoint	Clinical Findings	Investigations	Interventions
Admission (Hour 0)	Right upper quadrant and back pain; septic shock	Elevated WBC, CRP, PCT; abdominal ultrasound showed cholecystitis with gallstones	Laparoscopic cholecystectomy; meropenem started
Hour 30	Worsening back and buttock pain; warm, erythematous skin; vasopressors discontinued	Bedside ultrasound: cellulitis with possible microabscesses; WBC and CRP decreased; LRINEC score = 7	Surgery deferred; soft tissue MRI; single dose of linezolid
Hour 34	Pain intensified; erythema extended contralaterally and down right thigh	MRI showed extensive cellulitis and intramuscular abscesses; blood culture positive for B. pseudomallei	Meropenem continued; linezolid discontinued; TMP–SMX added
Day 6	Marked improvement: resolution of pain, reduced erythema and warmth	—	Continued intravenous meropenem and oral TMP–SMX
Day 10	Clinical recovery continued	MRI: abscess size reduced (19 × 63 × 53 mm); improved cellulitis	Continued intravenous meropenem and oral TMP–SMX
Week 4 (Discharge)	Stable and asymptomatic	Inflammatory markers normalized	Meropenem stopped; continued TMP–SMX
Week 6	No residual symptoms	Bedside ultrasound: no residual abscesses or focal fluid collections	TMP–SMX maintained to complete eradication phase

Discussion

This case describes a patient with melioidosis presenting with extensive soft tissue infection without involvement of typical organs such as the lungs, genitourinary tract, or central nervous system [3, 9, 10]. The diagnosis of soft tissue melioidosis was supported by classical clinical signs [7, 11], MRI findings, exclusion of other foci, and positive blood cultures. While the patient also had acute calculous cholecystitis, sterile bile cultures and concurrent involvement of the adjacent abdominal wall suggested the gallbladder was likely reactive [12] rather than the primary focus.

Diffuse cellulitis with deep intramuscular abscesses caused by melioidosis is exceedingly rare. In an Australian cohort of 486 cases, primary cutaneous involvement occurred in 12%, with cellulitis being rare [5]. Similarly, in the Darwin study of 540 cases, cellulitis and muscle abscesses were seen in only 0.6% and 1.5%, respectively [4]. A Malaysian cohort reported soft tissue involvement in 28% of patients, but all intramuscular abscesses were confined to the limb [13]. Rashidi Ahmad described anterior chest wall involvement [14], but to our knowledge, thoracoabdominal intramuscular abscesses with overlying diffuse cellulitis and contiguous spread to the thigh have not been previously reported. Cutaneous melioidosis is often attributed to direct inoculation [3, 5], whereas intramuscular abscesses more likely result from haematogenous dissemination [3]. In our case, both lesions coexisted in the same region without preceding trauma or skin breach, and it remains unclear whether one preceded the other.

Recent guidelines recommend classifying soft tissue infections based on the presence or absence of necrosis, which determines the need for surgery [6, 7]. This applies across all tissue layers—from skin and subcutaneous tissue to fascia and muscle [6, 7]. In melioidosis, necrosis is uncommon unless antibiotic therapy is delayed or inappropriate [14, 15] and drainage is generally not required except in cases of prostatic abscesses [4, 8, 16]. In this case, features concerning for NSTI included rapidly progressive erythema, disproportionate pain, and a LRINEC score of 7 (intermediate risk) [7]. MRI revealed no soft tissue gas, fascial thickening, or signal tracking—findings that strongly argued against NSTI [17]. However, with its imperfect sensitivity (~ 86.4%) [17], the decision to defer surgery and drainage was based on the combination of early, appropriate antibiotics, clinical stability, and imaging findings. No single factor alone was sufficient to exclude necrosis. Among the recognized risk factors for severe melioidosis—diabetes mellitus, hazardous alcohol use, chronic kidney disease, and chronic lung disease [4]—this patient had undiagnosed diabetes mellitus, but lacked the others.

The antimicrobial regimen followed Darwin melioidosis treatment guideline recommendations: meropenem for four weeks in the intensive phase for deep-seated collections [9], with TMP–SMX introduced during this phase using an eight-day graded-dose escalation (40/200 mg BD → 320/1600 mg BD) to reduce adverse effects, followed by a three-month eradication phase at full dose [9]. A single dose of linezolid was used empirically when NSTI was suspected [7], but was discontinued after microbiological confirmation of melioidosis.

Limitations

This is a single case report, which limits generalisability. The primary focus of infection could not be definitively established due to the absence of early soft tissue imaging. Earlier imaging beyond hepatobiliary ultrasound might have clarified whether cellulitis or intramuscular abscesses developed first. Microbiological identification relied on the VITEK 2 compact system (bioMérieux, France), which is not formally validated for Burkholderia pseudomallei; however, the result showed an Excellent identification score and was supported by compatible colony morphology, oxidase positivity, and biochemical characteristics. A further limitation is the absence of pus culture from the abscesses, as surgical drainage was not indicated and the diagnosis was already confirmed by blood culture and clinical–radiologic findings.

Conclusion

This case highlights the importance of considering melioidosis in patients from endemic regions presenting with soft tissue infections, even when clinical features mimic necrotising soft tissue infection. A structured diagnostic approach—incorporating clinical progression, microbiological confirmation, and advanced imaging—enabled timely recognition of a non-necrotising form and successful non-operative management. Early initiation of guideline-concordant antimicrobial therapy is essential for favourable outcomes and to avoid unnecessary surgical intervention.

Learning points

Melioidosis should be considered in patients presenting with soft tissue infections in endemic areas, especially when clinical features are atypical. It may present with uncommon patterns such as diffuse cellulitis or intramuscular abscesses of the thoracoabdominal wall. Deep-seated intramuscular abscesses in melioidosis may mimic necrotising soft tissue infection when located beneath areas of cellulitis, causing rapid progression and disproportionate pain. Accurate differentiation between necrotising and non-necrotising soft tissue infection requires careful integration of clinical progression, haemodynamic stability, and advanced imaging when feasible. In cases of melioidosis-related cellulitis and deep-seated abscesses, early and appropriate empirical antibiotic therapy is a key factor in successful non-surgical management.

Abbreviations

CRP

C-reactive protein

ICU

Intensive Care Unit

LRINEC

Laboratory Risk Indicator for Necrotising Fasciitis

MRI

Magnetic Resonance Imaging

NSTI

Necrotising Soft Tissue Infection

TMP–SMX

Trimethoprim–sulfamethoxazole

WBC

White Blood Cell Count

Author contributions

TPN conceived the report, collected clinical data, performed the literature review, and drafted the manuscript. HTTV, NNPN, and TTK contributed to clinical management, interpretation of the patient’s ICU course, and manuscript revision. TDT interpreted and contributed to the radiological findings, image selection, and figure preparation. PKNH provided input on antimicrobial management and verified the pharmacological aspects of treatment. All authors read and approved the final manuscript .

Funding

No funding was received for this study.

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

Not applicable (case report; observational data from routine clinical care).

Consent for publication

Written informed consent for publication of this case report and accompanying images was obtained from the patient.

Competing interests

The authors declare no competing interests.