Emphysematous osteomyelitis of the clavicle: a pleural process?

  1. Amira Elshikh 1,
  2. Niraj Gowda 1,
  3. Lisa Glass 2 and
  4. Robert B Maximos 3 , 4
  1. 1 Internal Medicine, George Washington University, Washington, District of Columbia, USA
  2. 2 Division of Pulmonary, Critical Care and Sleep Disorders Medicine, Department of Medicine, George Washington University, Washington, District of Columbia, USA
  3. 3 Pulmonary, Critical Care and Sleep Disorders Medicine, The George Washington University Hospital, Washington, District of Columbia, USA
  4. 4 Pulmonary and Critical Care, Department of Veterans Affairs, Washington, District of Columbia, USA
  1. Correspondence to Dr Amira Elshikh; aelshikh43@gwu.edu

Publication history

Accepted:27 Jun 2020
First published:28 Jul 2020
Online issue publication:28 Jul 2020

Case reports

Case reports are not necessarily evidence-based in the same way that the other content on BMJ Best Practice is. They should not be relied on to guide clinical practice. Please check the date of publication.

Abstract

Emphysematous osteomyelitis (EO) is a rare infection associated with intraosseous gas. EO is an often fatal disease with an estimated 34% mortality. We present a case of a 63-year-old man with sternoclavicular EO with pleural involvement and significant subcutaneous emphysema diagnosed by CT. Extension of intraosseous gas into the pleural cavity is an extremely interesting presentation that has not been previously reported. The patient underwent a multidisciplinary treatment approach with surgical debridement and an extended antibiotic course. Intraoperative cultures of the pectoralis muscle and bone biopsy grew pan-sensitive Escherichia coli. Prompt recognition and treatment are paramount to avoid a potentially fatal outcome. A review of the literature of the previous 46 cases of EO is presented for associated risk factors, the role of surgical management and antibiotic therapy.

Background

EO is an unusual infectious presentation first described in 1981.1 Intramedullary air in the extra-axial skeleton has been considered pathognomonic for EO.1 2 There have been a total of 46 reported cases, with no reported cases involving the pleural space or extension into the intrathoracic cavity.1–42 Gas formation extending into the pleural cavity is very rare. This presentation may be mistaken for a primary pulmonary process, which may result in treatment delay. In the past 5 years, the incidence of reported cases of EO has increased dramatically.1–42 Despite the increase in incidence, it remains a relatively obscure entity. Early identification and appropriate therapy is key in decreasing morbidity and mortality. Poor outcomes may occur as a result of delayed diagnosis and given the absence of guidelines to target therapy.

Case presentation

A 63-year-old man presented to the emergency department with right arm swelling and malaise of 3-week duration. Medical history included hypertension, squamous cell carcinoma affecting the oropharynx treated with radiation, diabetes complicated by chronic kidney disease and peripheral arterial disease. Initial evaluation was notable for normal vital signs without fever and significant pitting oedema of the entire right upper extremity. Additionally, a hard, erythematous mass without fluctuance measuring 4×5 cm was palpable over the medial third of the right clavicle (figure 1). Laboratory evaluation revealed leucocytosis to 28×109/L, elevated blood urea nitrogen to 180 mg/dL, creatinine 7 mg/dL and decreased serum bicarbonate of 14 mmol/L. Urinalysis was positive for the presence of leucocyte esterase. Urine and blood samples were sent for culture. Empiric broad spectrum antibiotic coverage was started. Right upper extremity duplex ultrasound was negative for deep venous thrombosis. CT scan of head, neck and thorax demonstrated diffuse soft-tissue gas involving the neck, anterior mediastinum, supraclavicular region and extending into the pectoralis muscle (figure 2). Intraosseous air was noted in the right clavicle. The soft-tissue gas was suspected to be arising secondary to a primary pleural process, and pulmonology was consulted. Ultrasound evaluation, physical examination and CT thorax did not support a pleural defect or a primary pulmonary process, such as necrotising pneumonia or malignancy, as the cause. Urine culture grew pan-sensitive Escherichia coli.

Figure 1

Right sided 4×5 cm subcutaneous erythematous mass over the medial third of the right clavicle.

Figure 2

(A) Axial bone view CT image demonstrating subcutaneous emphysema in the soft tissue adjacent to the right clavicle, with intramedullary air in the clavicle. (B) Coronal bone view CT image demonstrating subcutaneous emphysema in the right neck, axillary area and superior to the right clavicle, with multiple foci of air within the clavicular head. (C) Coronal lung view CT image with extensive subcutaneous emphysema tracking into the right neck with pleural involvement. (D) Axial lung view CT image showing subcutaneous emphysema with intrathoracic tracking.

Surgical drainage of the chest wall was performed. Intraoperative cultures of the pectoralis muscle and bone biopsy grew pan-sensitive E. coli. A diagnosis of emphysematous osteomyelitis (EO) was suspected with the presence of E. coli in the setting of intraosseous gas.

Differential diagnosis

The differential diagnosis of subcutaneous emphysema includes necrotising fasciitis (NF), clostridial myonecrosis, pneumonia complicated by bronchopleural fistula, and osteonecrosis. The potential impact of oropharyngeal radiation was considered due to the proximity of the radiation field to the clavicle and the potential for short-term and long-term complications secondary to radiation therapy. However, the radiation field did not include the right clavicle and therapy was completed several years prior to presentation. Prompt consideration and exclusion of NF are always appropriate. NF typically presents with a rapidly progressive course not consistent with our patient’s subacute presentation and haemodynamic stability.43 Typical findings of NF include grey appearance of the fascia, low purulence with exudate and easily separated tissues, which were not present on surgical exploration in support of this diagnosis.44 Clostridial myonecrosis was also considered, and discounted based on the absence of trauma in the patient’s reported history. Severe pneumonia with bronchopleural fistula formation and spread was another possible diagnosis. This diagnosis was not consistent with the patient’s lack of respiratory symptoms or CT thorax imaging showing lack of parenchymal involvement. Osteonecrosis was another diagnostic consideration in light of the presence of intraosseous air. However, the diagnosis was not supported by imaging, which lacked the pathognomonic subchondral radiolucency found in osteonecrosis.45 Finally, a vacuum phenomena can be observed in degenerative bone conditions, with an expected linear distribution of gas, not visualised in EO.31 Small et al reviewed CT imaging characteristics of EO, and described intermedullary foci of air usually associated with extraosseous air without cortical destruction, consistent with our patient’s findings.20

Treatment

Initial treatment consisted of empiric broad spectrum antibiotic coverage using vancomycin and meropenem. After identification of subcutaneous air on CT imaging, vancomycin was discontinued, and clindamycin was started. Haemodialysis was initiated for oliguric renal failure.

Surgical right chest wall drainage was performed with a multidisciplinary surgical team. The operation identified multiple pockets of purulence, soft-tissue necrosis with involvement of the pectoralis major and sternoclavicular joint head. His sternoclavicular joint was debrided, with partial removal of the clavicular head, remobilisation of the right pectoralis major muscle for coverage, followed by a partial skin flap. A right thoracostomy drain as well as a wound vacuum was placed. Intraoperative cultures grew pan-sensitive E. coli. A diagnosis of EO was confirmed. Antibiotic therapy was adjusted to cefazolin 2 grams post haemodialysis for a 6-week total antibiotic duration.

Outcome and follow-up

Follow-up CT imaging 4 weeks post presentation showed interval improvement with resolution of his subcutaneous emphysema and decrease in intraosseous gas. His pleural findings resolved. Post operatively, his course was complicated by poor wound healing of the pectoralis muscle flap, requiring two additional debridements without improvement and continued haemodialysis requirement. He developed severe sepsis and expired 5 months after his initial presentation.

Discussion

EO is a rare disease with high mortality. The case presented above represents an unusual presentation of an uncommon disease process. To our knowledge, this is the first reported case of pleural involvement in EO. He developed EO secondary to E. coli with evidence of a sternoclavicular joint abscess. His risk factor for EO was likely diabetes. Probable origins for EO in our patient include transient bacteraemia in the setting urinary tract infection or joint abscess noted on the sternoclavicular joint intraoperatively. Literature review was performed using the search term “emphysematous osteomyelitis” in PubMed, with English language results reviewed. Notable findings from our review of the literature are included below.

Review of the medical literature revealed no difference between female and male incidence. The average patient age was 53 years, with a range of 15–79 years.1–42 Many of the infected individuals were immunocompromised including those with malignancy, chronic corticosteroid usage and diabetes mellitus. Diabetes mellitus was the most commonly identified risk factor (41% of cases).3–5 8 13 14 16–26 Mortality estimates in reported cases were calculated to suggest a 30%–35% mortality rate.1 3–15 EO is thought to develop secondary to haematogenous spread; however, bacteraemia was documented in only 16 cases (35%).5–8 10 12 16 18 24 26 28 32 33 Identifiable abscess was reported in 25 cases (57%).1 3 5 6 9–15 19–21 24 26–31 The majority of identified abscesses were reported in anatomic proximity to EO; however, several cases showed distant involvement. One case, for example, reported liver abscess and osteomyelitis of L5-sacrum.5

Most cases of EO are caused by a single organism (84%), while some reported polymicrobial infection. E. coli is the most commonly identified bacteria (23% of cases)1 6 8 9 12–15 19 27 32 42 followed by Klebsiella pneumoniae (21% of cases).3–5 7 8 15 16 28 29 31 Antibiotic therapy commonly included meropenem, clindamycin or a cephalosporin. Empiric recommendations for antibiotic therapy are challenging, ever-changing and should be tailored to local antibiotic resistance patterns. Although frequently unreported, antibiotic duration was noted to be at least 6 weeks.2 4 6 16 24 27 28 36 42 Surgical management was used in 22 cases (49%).1 2 8–10 12 14–18 23 26 27 29–31 33 36 42

This is an uncommon disease process, with limited available information to guide a clinician’s management. It is important to consider the EO disease process when soft-tissue gas is present and proceed with appropriate diagnostic imaging and biopsy.

Learning points

  • Intraosseous air should increase suspicion for emphysematous osteomyelitis (EO) even in the presence of intrapleural disease.

  • Risk factors for EO include immune-compromised states, with diabetes mellitus as the most common risk factor.

  • Most common causative organisms of EO include Escherichia coli and Klebsiella pneumoniae.

  • Prompt therapy with tailored antibiotics and possible surgical intervention may be warranted in EO.

Footnotes

  • Contributors All authors made substantial contributions to this case presentation. AE and NG started the first draft and LG and RBM revised it critically for intellectual content. AE and NG collected data from prior case reports and interpreted the trends, which were confirmed by LG and RBM. All authors have read and approved the final draft and agree to be held accountable for all aspects of this report and its integrity.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.

References

Use of this content is subject to our disclaimer