Differential diagnosis of chylothorax in a patient on parenteral nutrition: a case report

Abstract

Parenteral nutrition (PN) is widely used when enteral feeding is contraindicated or not possible in patients who are malnourished or at risk of malnutrition, but it is not without complications. We describe the rare complication of PN-related chylothorax that can lead to extensive investigations if not picked up early with clinical suspicion and simple biochemical tests.

Introduction

A chylothorax is a type of pleural effusion. It results from lymphatic fluid (chyle) accumulating in the pleural cavity. Its source is usually leakage from the thoracic duct or one of the main lymphatic vessels that drain into it. Chylothorax is rare, with approximately 50% due to malignancy (particularly lymphoma), 25% due to trauma (especially during surgery), and the rest due to miscellaneous causes such as tuberculosis, sarcoidosis and amyloidosis. ^1,2 The effusion is characteristically white and milky in appearance and contains triglycerides in excess of 1.2 mmol/L. ³

Case

A 71-year-old male was admitted to the intensive care unit (ICU) with a duodenal perforation following an elective laparoscopic cholecystectomy. Owing to the high biliary output from the abdominal drain, he was commenced on parenteral nutrition (PN) via central venous neck catheter in order to rest his bowel. The catheter was changed as indicated by unit infection control guidelines as he required PN for more than four weeks during his ICU stay. On the third week of admission, PN was being administered via a left-sided subclavian catheter when his respiratory function deteriorated. A chest X-ray showed large bilateral pleural effusions, more extensive on the left, with some fluid in the horizontal fissure on the right and the endotracheal tube and left central neck catheter in stable positions.

Pleural tap revealed a chylous fluid that was sent for analysis (Table 1). The triglyceride concentration of 29.1 mmol/L was consistent with chyle. There is, however, no discrete cut-off value for differentiating chylous from non-chylous effusions, but it was estimated that fluid with triglyceride concentration >1.24 mmol/L had a less than 1% chance of not being chylous. ³ The patient was receiving PN and this needed to be excluded as the source. To differentiate chylothorax from pseudo-chylothorax, pleural fluid cholesterol was measured and to differentiate PN as the source, glucose and osmolality of the PN fluid were also measured (Table 1). Owing to the high glucose concentration (109.1 mmol/L) and osmolality (446 mOsm/kg), a provisional diagnosis of possible PN leak into the pleural space was considered. A sample of the PN fluid and serum were subsequently analysed (Table 1), and the similarity of the triglycerides and glucose concentrations in the pleural fluid and PN confirmed the diagnosis.

Table 1

Biochemical composition of PF, PN administered and serum

	PF	PN	Serum
Triglycerides (mmol/L)	29.1	31.9	1.1
Cholesterol (mmol/L)	3.2	5.3	1.4
Glucose (mmol/L)	109.1	94.8	9.1
Osmolality (mOsm/kg)	446	1192	319
Potassium (mmol/L)	5.8	22.8	5.0

PF, pleural fluid; PN, parenteral nutrition

The pleural effusion was drained via a chest drain, the left central venous catheter was removed and PN was continued via a right-sided central venous catheter. The patient made a good recovery and was transferred to a general ward.

Discussion

PN is widely used when the gastrointestinal tract cannot or should not be used in patients who are malnourished or at risk of malnutrition, but it is not without complications. The complications can be infections, technical (due to catheter placement) and metabolic such as hyperglycaemia, hyperlipidaemia, hypercapnia, acid–base disturbance, liver complications and metabolic bone disease. ⁴

Differential diagnosis of a turbid or milky (or both) pleural effusion is chylothorax, pseudo-chylothorax (cholesterol pleurisy) and empyema. ⁵ The later is differentiated by centrifugation, which will show a clear supernatant and other biochemical and microbiological investigations. Pseudo-chylothorax is seen when fluid has been present for a long time in the pleural space and seen especially in fibrotic pleura. ⁵ Chylothorax and pseudo-chylothorax can be discriminated by the measurement of lipids in the fluid. A true chylothorax will usually have a high triglyceride concentration, usually >1.24 mmol/L, and chylothorax can usually be excluded if the triglyceride concentration is <0.56 mmol/L. ⁶ The literature is conflicting in the use of cholesterol measurement in differentiating the two. While some authors state that a high cholesterol concentration (>5.2 mmol/L) is seen in pseudo-chylothorax, ⁵ others have shown no differences in cholesterol concentration between both conditions. ³ In our case, the concentration of pleural fluid cholesterol was 3.2 mmol/L.

Left-sided large-bore 14G catheters have been associated with increased risk of pleural effusions. ⁷ It is normal practice to confirm the position of the central catheter by radiography prior to it being used for administration of PN. However being a hyperosmolar fluid, PN can cause osmotic injury leading to vascular leakage despite an intact placement. ⁸

Clinicians should be aware of this rare complication of PN feeding, especially when there is a sudden deterioration in respiratory function without an obvious cause. In the literature we found one case report on chylothorax due to PN leak, ⁹ where the authors recommended the measurement of glucose and potassium along with the lipid profile. In this case, the potassium measurement did not contribute to the final diagnosis (Table 1).

A pleural fluid glucose concentration <3.3 mmol/L is found in exudative pleural effusions secondary to empyema, rheumatoid disease, lupus, tuberculosis, malignancy or oesophageal rupture. The lowest glucose concentrations are found in rheumatoid effusions and empyema. ⁵ On the other hand, a high pleural glucose concentration is a rare biochemical finding. There have been case reports on PN leak, ⁹ oesophageal perforation ¹⁰ and peritoneal dialysis, ¹¹ which have contributed to elevated glucose in pleural fluid. However, a suspected chylous fluid should have triglycerides and cholesterol measured, as well as glucose and osmolality in those receiving PN in order to prevent misdiagnosis and further unnecessary investigations. Finally, it should be recognized that this analysis reflects the situation at the time of the pleural tap and does not exclude another cause for an effusion at a later date.

DECLARATIONS

Conflicts of interest: None.

Guarantor: LM.

Contributorship: RPV – contributed to the drafting of the paper, literature review, critically reviewed the case report's content and approved the final version submitted for publication; JLB, – clinical management, critically reviewed the case report's content and approved the final version submitted for publication and LM – clinical management, critically reviewed the case report's content and approved the final version submitted for publication.