Marked cachexia in probable invasive pulmonary aspergillosis with bronchopleural fistula

Abstract

A 49-year-old man with a medical history of diabetes and heavy smoking was admitted to intensive care with severe bilateral pneumonia associated with marked cachexia. He developed a complex right-sided bronchopleural fistula and was transferred to our tertiary centre for consideration of surgical intervention.

Despite escalation of antibiotic therapy, he did not improve and further investigations led to a diagnosis of invasive pulmonary aspergillosis. Definitive treatment plans required a right pneumonectomy; however, given the severity of cachexia, he remained unable to undergo such a large operation. This case demonstrates an atypical presentation of invasive pulmonary aspergillosis in a mildly immunodeficient individual. It highlights the challenges in assessment and management of critically ill patients’ nutrition as well as optimal timing for surgical intervention.

Background

Invasive pulmonary aspergillosis (IPA) is well documented in patients who are immunocompromised, such as those with haematological malignancies. Other risk factors for IPA include pre-existing lung disease, corticosteroid use, diabetes, liver cirrhosis and end-stage renal failure.¹ It is often challenging to diagnose and lung biopsy remains the gold standard for proven infection. However, lung biopsy is an invasive procedure that may not always be possible in critically ill patients.²

Diagnostic challenges of IPA mean that delays in initiation of antifungal therapy are common. In our case, this delay, in combination with severe cachexia, contributed to the patient’s deterioration and prolonged respiratory wean. This case demonstrates the difficulties in optimising critically ill patients for surgery, especially in those who have an ongoing consumptive process. We also discuss the challenges in assessing energy requirements and managing cachexia within this context.

Case presentation

A 49-year-old man was transferred to our tertiary cardiothoracic centre after admission to intensive care at his local hospital. He initially presented with a 1-week history of worsening breathlessness, productive cough and fevers on a background of several months of unintentional weight loss associated with night sweats. His medical history was significant for type 2 diabetes mellitus and he had a 40-pack-year smoking history. He worked as a landscape gardener and reported no recent travel.

On admission, he was diagnosed with bilateral pneumonia with type 2 respiratory failure and was admitted to intensive care for respiratory and cardiovascular support. He remained ventilator-dependent requiring tracheostomy and received multiple courses of antibiotics for chest sepsis. Imaging revealed a large right-sided bronchopleural fistula, for which he was transferred to our cardiothoracic centre for surgical intervention (figure 1).

Figure 1.

Chest radiograph showing large right-sided pneumothorax.

Investigations

During admission, the patient underwent multiple investigations to account for the severity of the lung pathology and the presenting cachexia. CT chest on admission to our tertiary centre showed parenchymal destruction of the right upper and middle lobes with a large air-filled pleural space and a bronchopleural fistula driving a pneumothorax (figure 2).

Figure 2.

CT chest showing multiple right-sided bronchopleural fistula and large pneumothorax, and bilateral consolidation.

Screening for HIV, hepatitis, autoimmune and malabsorptive processes, as well as alpha-1-antiptrypsin deficiency testing were all reported negative. Direct tuberculosis staining and cultures of bronchoalveolar lavage fluid (BALF) and sputum samples were also negative. Investigations with CT imaging and blood films did not show evidence of an underlying malignancy.

Serratia marcescens was isolated on several occasions from blood, sputum and BALF, and PCR of BALF showed influenza B infection. Fungal screen of BALF was positive for galactomannan antigen (test result of 4.2) although serum galactomannan antigen and BALF fungal culture were negative.

Treatment

The patient received several courses of intravenous antimicrobials including a prolonged course of meropenem for S. marcescens infection and two courses of oseltamivir following positive influenza B PCRs. Despite this, he continued to deteriorate and was empirically started on anidulafungin for presumed fungal infection which was changed to voriconazole after positive galactomannan antigen test of BALF. Following commencement of antifungal agents, the patient improved and was able to tolerate weaning of vasopressors.

From a respiratory point of view, he required tracheostomy following several failed extubation attempts, and had a slow respiratory wean. A combination of factors including the complex lung pathology and severe generalised weakness contributed to the poor respiratory function. Surgical management for bronchopleural fistula was thought to be the only definitive treatment and the plan was to optimise him medically prior to a pneumonectomy. Several interim surgical measures were performed: video-assisted thoracoscopic surgery with washout and debridement of right middle lobe and a modified Clagett procedure (figure 3). Both enteral and parenteral nutrition regimens were used in attempts to improve his nutrition status.

Figure 3.

Image of right lung parenchyma from video-assisted thoracoscopic surgery.

Outcome and follow-up

This patient remained in the intensive care unit for 6 months. During that time, he was unable to tolerate weaning of invasive ventilation; he received multiple courses of antibiotics and had daily physiotherapy and dietetics input. Despite this, he deteriorated with further weight loss and raised inflammatory markers and died before surgical repair of the large bronchopleural fistula was possible.

Discussion

Invasive pulmonary aspergillosis in an atypical patient

Classical features of IPA are not always present in critically ill patients making the diagnosis more challenging within this population.³ The gold standard of proven IPA is lung biopsy but many, as with our patient, remain too unwell to undergo a tissue biopsy.² The severity of our patient’s presentation should have prompted earlier treatment with antifungals; however, due to the protected nature of antifungal therapy within our institution, this was not possible until other causes were excluded or if criteria for initiation were met.

Three diagnostic criteria tools can be used to identify probable IPA without a lung biopsy using a combination of host factors, clinical data and mycological findings: the European Organisation for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG), the AspICU (Aspergillus in intensive care units)criteria and the Bulpa criteria. The EORTC/MSG is only validated in the immunocompromised³ so the AspICU criteria were developed to discriminate between putative IPA and aspergillosis colonisation within the critically ill population.⁴ With this tool, probable IPA can be diagnosed without lung biopsy, but when applying this to our patient only three of the four criteria were met. Finally Bulpa criteria focuses on diagnosis of IPA in critically ill patients with chronic obstructive pulmonary disease (COPD).⁵ An element of COPD was possible in our patient given his smoking history but it was unlikely to have been as severe as the Bulpa criteria require for diagnosis.

Our patient had a significantly positive BALF galactomannan antigen test result of 4.2 (>0.5 indicates positive result).¹ Though he did not entirely meet the aforementioned criteria, the combination of a positive BALF galactomannan result and the clinical improvement following antifungal therapy gave us confidence that IPA as a diagnosis was probable.

Previous research has found that galactomannan antigen sensitivity ranges between 46% and 80% and specificity between 21% and 94% with variations depending on the patient population examined.⁶ Despite this variation, including patients with a positive galactomannan antigen test within the diagnostic criteria could help improve IPA detection in the critical care population and reduce delays in treatment.¹ Shroeder et al¹ found that when comparing a group of Aspergillus-positive culture patients with a group of galactomannan antigen positive only, there were no statistically significant differences in outcome, baseline characteristics or scoring systems.¹ Furthermore, BALF fungal cultures are positive in less than 50% of cases of IPA and sensitivity is only 30%–40%. Cultures are also affected by timing of sample and previous antimicrobial therapy. This may have been reflected in our patient for whom fungal BALF cultures returned negative.^{1 7}

Management of bronchopleural fistula

Management of patients with bronchopleural fistula can be complex. Large air leaks with ineffective gas exchange create challenges in ventilation. The main principles of treatment are to drain the infected pleural space, treat infection, protect the lung and clear secretions from the remaining lung.⁸ Definitive treatment is surgical with interventions including endobronchial sealants and valves, Clagett procedure and pneumonectomy.⁸ One study looking at the management of bronchopleural fistulae found that patients underwent an average of 3.3 surgical procedures, demonstrating the complexity of this pathology.⁹ Definitive surgical repair is usually delayed until the patient’s condition is optimised and the remaining lung cavity is clean.⁸ Our patient was aggressively treated with antimicrobials and underwent two surgical procedures with debridement of lung tissue but unfortunately remained too unwell for more invasive surgical management.

Investigation and management of cachexia

Patients admitted to critical care are at high risk of developing malnutrition due to the catabolic state of severe illness. Malnutrition can increase the risk of infection and death and is therefore an important factor in the patient’s management.¹⁰ Changes in metabolism during severe illness make assessment of nutritional status and energy requirements complex. Traditional markers of nutrition, such as transferrin or albumin, become unreliable in the context of critical illness.¹¹ No consensus has been reached about the use of other markers, such as indirect calorimetry, with this group of patients.

The European Society for Parenteral and Enteral Nutrition guidelines recommend that patients with severe malnutrition receive enteral nutrition of up to 25–30 kcal/kg/day tailored according to disease progression and gut tolerance.¹² Parenteral nutrition is typically used in patients with gut dysfunction, but guidelines support the use of supplementary parenteral nutrition in those who do not reach target nutrition on enteral nutrition alone.¹² In our case we used caloric targets of 25 kcal/kg/day and protein targets of 2 g/kg/day. However, despite enteral feeding for 6 weeks, his weight remained static with associated profound weakness inhibiting rehabilitation. There were concerns about a tracheo-oesophageal fistula so decision was made to commence supplementary parenteral feeding. Weight remained unchanged after 1 month and nutrition was reverted to enteral alone. The challenge in this case was how to support the patient in not only remaining weight neutral but also gaining weight and nutrition. Ultimately, managing the underlying process leading to any cachexia is paramount to improve a patient’s nutritional status. This proves challenging when optimised nutrition itself is needed to treat the primary consumptive process.

This is a complex case where patient-specific circumstances have required deviation from guidelines. It highlights the need to consider invasive fungal infection early in critically ill patients, including in those without classical risk factors. It also shows the challenges in improving patient’s nutrition status in the context of pre-existing cachexia and superadded consumptive pathology. Unanswered questions remain about the diagnosis and management of this patient demonstrating the need for the development of further diagnostic criteria and guidelines in the context of critical illness.

Learning points.

• Invasive pulmonary aspergillosis is an important source of mortality and morbidity in intensive care, and early identification and initiation of therapy is essential to improve outcomes. Earlier clinical suspicion is required for prompt identification and treatment.

• There is a need for more accurate and standardised assessments of patients’ nutritional requirements, especially in the critically ill population. Our cases highlighted the difficulties in estimations of these requirements and what targets to set.

• Patients with bronchopleural fistulae are complex and require input from a number of teams to allow for successful management of the lung pathology. Early involvement of the nutrition and the physiotherapist team is needed to prevent potentially irreversible complications from deconditioning and malnutrition.