Fade to black!

A 71-year-old man presented to the emergency room with chest pain, dry cough, shortness of breath and night sweats for 10 days. He had no fever, flu-like symptoms, sputum production or haemoptysis. There was no history of recent travel or sick contacts. A general review of systems was remarkable for subjective weight loss and malaise. He was an active smoker with a 100 pack-years smoking history and a past medical history of rectal squamous cell carcinoma (SCC). His cancer was treated with chemoradiotherapy and had been in remission for 8 years before the presentation. He was also known to have COPD, Global Initiative for Chronic Obstructive Lung Disease (GOLD) category B, well-controlled, with no recent hospitalisations or exacerbations over the past year.

Upon initial examination, arterial blood pressure was 123/75 mmHg, heart rate was 81 beats per min, respiratory rate was 24 breaths per min, and oxygen saturation was 93% on room air. Arterial blood gases showed partial oxygen pressure 73 mmHg, carbon dioxide tension 45 mmHg and pH 7.43. There were decreased breath sounds on the right side, up to the upper lung zone with dull percussion note and reduced tactile vocal fremitus on chest examination. The physical examination was unremarkable otherwise. The patient had a chest radiograph, which showed a right-sided pleural effusion (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Chest radiograph depicting right-sided pleural effusion.

The most likely diagnosis is malignant pleural effusion due to significant smoking history and prior history of malignancy. CAP and pulmonary TB are differentials. The patient received intravenous antibiotics for possible CAP.

A diagnostic thoracentesis is needed to establish appropriate diagnosis. Under ultrasound guidance, a diagnostic and therapeutic thoracentesis yielded 1200mL of fluid. Shortness of breath and tachypnoea improved post-procedure.

The fluid analysis results showed straw-coloured lymphocytic, exudative effusion, a negative Gram stain, and malignant cells, consistent with malignant pleural effusion (table 1). The pleural fluid analysis was not suggestive of complicated parapneumonic effusion or empyema.

On day 2 of admission, the patient developed worsening shortness of breath and desaturation. A follow-up examination revealed tracheal deviation to the left side, with decreased vocal fremitus and reduced breath sounds on the right side.

A repeat chest radiograph revealed iatrogenic pneumothorax on the side of the pleural effusion (figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Chest radiograph depicting right-sided pneumothorax.

The patient has a >2 cm iatrogenic pneumothorax and is symptomatic. Hence a chest tube was inserted and connected to the underwater seal (figure 3).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Chest radiograph showing a right-sided chest tube and partial lung re-expansion.

The patient's shortness of breath progressed despite chest tube insertion. The chest tube was functioning appropriately, as evidenced by swinging fluid levels and bubbling. CT of the chest was performed (figure 4).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Chest CT.

The right lung shows a sizeable pneumothorax, apical subpleural blebs and subpleural nodules. Other cuts of CT revealed multiple spiculated nodules throughout both lung fields.

The patient underwent right-sided video-assisted thoracoscopy. There was an upper lobe scar with surrounding nodules and continuous air leak. The thoracic surgery team performed an upper lobe wedge resection to fix the air leak and inserted a chest tube connected to an underwater seal. The colour of the fluid draining from the chest tube noted immediately post chest tube insertion was black (figure 5).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5

Black-coloured pleural fluid draining from the chest tube.

The most common cause of the black pleural effusion is malignancy. The pleural fluid analysis was consistent with empyema secondary to Pseudomonas aeruginosa and Klebsiella pneumoniae (table 1). Multiple invasive procedures, including a thoracocentesis and chest tube insertion performed before thoracoscopy on a frail, elderly patient with underlying malignancy, could have led to an iatrogenic empyema. Pyomelanin is a brown/black extracellular pigment produced by Pseudomonas; however, it does not cause black discolouration of body fluids.

The patient also had neutrophilic leukocytosis, a raised C-reative protein (324 mg·L⁻¹), and high procalcitonin (38.2 ng·mL⁻¹). He received piperacillin/tazobactam with a working diagnosis of empyema. Over the next 8 weeks, 8 L of pleural fluid was drained. Although the pleural fluid haematocrit was not measured, there was no drop in haemoglobin with massive fluid removal. Moreover, the fluid was free-flowing without any clots. The persisting dark colour of the fluid despite a decrease in RBC content over time without a haemoglobin drop indicates the absence of haemodynamically significant bleeding. The possibility of the black colour being secondary to haemolysis cannot be excluded entirely; however, the fact that the fluid colour turned to black immediately post-thoracoscopy without an initial red colour, which is more typical of a haemothorax, makes acute bleeding less likely as a cause. The other possibility is of bilothorax, which is usually reported secondary to a pancreaticopleural fistula. The absence of ascites and right-sided effusion makes that less likely.

Reported causes of black pleural fluid associated with malignancy are either due to pigment production by malignant cells (e.g. melanoma) or haemolysis. The black pleural effusion associated with most adenocarcinoma cases has been attributed to haemolysis. The authors believe that the absence of black colour from the initial pleural fluid sample, and then the appearance of this colour post-thoracoscopy, may be suggestive of the colour being attributed to the surgical intervention. However, to the best of our knowledge, such an association is not reported in the literature.

The surgical pathology report from the wedge resection showed infiltration of the lung by poorly differentiated malignant cells forming solid sheets. The tumour cells were pleomorphic with abundant eosinophilic cytoplasm, along with frequent mitotic figures and atypical forms. Immunohistochemistry was negative for P-40 and P-63, which are the markers for SCC. This immunoreactivity excluded the possibility of recurrence of the rectal SCC. The tumour cells’ reactivity for epithelial-specific antigen (MOC-31) and BER-EP4 confirmed the epithelial origin of the tumour. Based on negative lung adenocarcinoma markers, thyroid transcription factor 1 (TTF-1) and napsin-A, a lung primary was less likely. TTF-1 immunoreactivity is highly sensitive and specific in determining pulmonary versus extrapulmonary origin of adenocarcinoma [1]. In extrapulmonary adenocarcinomas (except thyroid) is so low (1%) that the negativity for TTF-1 may be interpreted as definitive evidence that the tumour is a primary from an extrapulmonary source. The cells were immunopositive for cytokeratin (CK)-7 expression and immunonegative for CK-20 expression (figure 6). A definitive conclusion on the origin of the tumour based entirely on the immuno-staining is challenging due to the lack of additional markers of pancreatic and hepatobiliary origin.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6

Histopathology slides. a) Poorly differentiated tumour arranged in solid sheets (haematoxylin and eosin stain, magnification x100). b) Highly pleomorphic tumour cells with brisk mitotic activity (haematoxylin and eosin stain x400). c) Immunoreactive for CK-7 (immunohistochemistry x100). d) Negative for CK-20 (immunohistochemistry x100). e) Negative for TTF-1 (immunohistochemistry x200). f) Negative for napsin-A (immunohistochemistry x100).

CT of the chest, abdomen and pelvis with contrast showed a dilated common bile duct measuring 16 mm with associated dilated intrahepatic biliary radicals. CK-7 positivity and CK-20 negativity in biliary tract carcinomas are associated with intrahepatic bile duct carcinomas. Tumour markers showed normal alpha-fetoprotein (AFP) and carcinoembryonic antigen (CEA), but a raised carbohydrate antigen 19-9 (CA19-9) (table 2). The serum alkaline phosphatase, total bilirubin and aspartate aminotransferase were also elevated. The patient's age, sex, tumour marker profile, immunohistochemical profile, liver function profile, and imaging data fit the possibility of metastasis from a hepatobiliary origin, likely intrahepatic bile duct carcinoma [2–5].

A multidisciplinary team meeting between the primary team, pulmonology, cardiothoracic surgery, oncology, onco-radiology and histopathology was conducted. The general condition of the patient made him unfit for further malignancy workup or chemotherapy. He received palliative care from thereon. He later passed away due to the progression of the disease.