Abstract
Background
Chylothorax is a rare anatomical disruption of the thoracic duct associated with a significant degree of morbidity and mortality. Diagnosis usually relies upon lipid analysis and visual inspection of the pleural fluid. However, this may be subject to incorrect interpretation. The aim of this study was to compare pleural fluid lipid analysis and visual inspection against lipoprotein electrophoresis.
Methods
Nine pleural effusion samples suspected of being chylothorax were analysed. A combination of fluid lipid analysis and visual inspection was compared with lipoprotein electrophoresis for the detection of chylothorax.
Results
There was 89% concordance between the two methods. Using lipoprotein electrophoresis as gold standard, calculated sensitivity, specificity, negative predictive value and positive predictive value for lipid analysis/visual inspection were 83%, 100%, 100% and 75%, respectively.
Conclusion
Examination of pleural effusion samples by lipoprotein electrophoresis may provide important additional information in the diagnosis of chylothorax.
Introduction
Chylothorax – the presence of chyle in the pleural space – is a clinical presentation that has been recognised for over three centuries. Chylomicron formation is the primary step in the exogenous lipid transport pathway, facilitating the transport of lipids from the thoracic duct into the systemic circulation. Damage to lymphatic vessels, most commonly due to dissection of the mediastinum in thoracic surgery, can lead to leakage of chyle into the pleural space, presenting clinically as a chylothorax. A number of other non-surgical causes of chylothorax have been described, including neoplasms and infections. 1 In these other clinical contexts, patients typically present after several days, with dyspnoea, cough and chest discomfort. Although chylothorax is a rare condition, it is associated with a significant degree of morbidity and mortality. 2
A number of investigations are indicated in the assessment of suspected chylothorax; however, these are used with varying frequency. In clinically acute settings, exploratory surgery is recommended to localise the trauma and provide immediate intervention. Chest X-ray and computed tomography can aid in determining the size and location of the chylothorax. Despite these, biochemical analysis of the thoracentesis remains the most important diagnostic step for clinicians. 3 Chyle has a characteristic white, odourless and milky appearance which on standing forms an immiscible layer at the surface of the fluid. Seminal data from Staats et al. continue to provide the most widely used criteria for assessing the presence of chyle: triglycerides (TGs) >110 mg/dL (1.24 mmol/L) suggest chylothorax, and TGs < 50 mg/dL (0.56 mmol/L) make chylothorax unlikely. 4 There remains a level of uncertainty with these measurements, and it has been suggested that lipoprotein electrophoresis could contribute to the detection of chyle in pleural fluid. 5 Despite this, lipoprotein electrophoresis is not routinely used for this purpose. The aim of the current study was to evaluate the role of pleural fluid lipoprotein electrophoresis in the diagnosis of chylothorax.
Methods
Over a two-month period, nine pleural effusion samples were assayed on the Olympus AU 5400 (High Wycombe, UK) for TGs. The TG method used was a glycerol-3-phosphate oxidase/phenol/aminophenazone (GPO-PAP) enzymatic colorimetric test, with an intra-assay coefficient of variation of 1.22% at 1 mmol/L, a linear range 0.113 mmol/L to 11.3 mmol/L and a functional sensitivity 0.04 mmol/L. Samples were sent from surgical sources in a large tertiary cardio-thoracic centre (Nottingham University Hospitals) to rule out or help diagnose chylothorax. Centrifuged samples were left overnight (4℃) and visually inspected the following day for chyle. Our protocol was modified from the data of Staats et al.: TG >1.30 mmol/L – report as positive, TG <0.50 mmol/L – report as negative and TG 0.50–1.30 mmol/L – inspect the pleural fluid for chylomicrons. Lipoprotein electrophoresis was performed on the Sebia Hydrasis (Georgia, USA) to determine how this compared with lipid analysis and visual inspection. It should be noted that neither TG measurement nor lipoprotein electrophoresis have been commercially validated for use with pleural fluid.
Results
Results are summarised in Table 1. ‘Reported as’ indicates the reported result based on fluid lipid analysis/visual inspection alone. Figure 1 shows the results of lipoprotein electrophoresis, which corresponded with the reported result for samples A, C, D, E, F, G, H, and I. Sample B was reported as negative; however, a faint band was observed after lipoprotein electrophoresis (Lane 2), denoting the presence of chyle (chylomicrons remain at the application point on electrophoresis due to their size). Sample C was reported as positive for chyle due to the TG concentration; the presence of chyle was confirmed by electrophoresis (Lane 4). However, visual inspection of sample C did not reveal a distinct layer of chylomicrons. Sample H was reported as negative; however, using Staats criteria, a TG >1.24 mmol/L suggests a positive result. The lipoprotein electrophoresis confirmed that there was no chyle present (Lane 10). The negative control was a faecal fluid sample (Lane 6).
Lipoprotein electrophoresis of pleural effusion samples. Lipoprotein electrophoresis was performed on nine pleural fluid samples being investigated for the presence of chyle. Results were compared with those obtained by our current method. ?: query.
Overall, combined fluid lipid analysis/visual inspection showed 89% concordance with lipoprotein electrophoresis, with 83% sensitivity, 100% specificity, 75% negative predictive value and 100% positive predictive value.
Discussion
Our data confirm the important role that lipoprotein electrophoresis may have in the detection of chylothorax. Visual inspection alone lacks both sensitivity and specificity for the detection of chyle. Sometimes, turbidity due to elevated cholesterol can mimic the characteristic appearance of chyle (pseudochylothorax). Conversely, a study by Maldonado et al. demonstrated that a significant number of samples from patients with chylothorax (44%) did not have the characteristic milky appearance.6 Our findings are in line with this study, which concluded that gross appearance of fluid alone is not a sensitive diagnostic marker for identifying chylothorax. 6
Our analysis also illustrates the limitations of relying exclusively on the TG concentration, using the criteria of Staats et al. We identified one case where the result was positive according to the TG concentration, but chyle was absent on lipoprotein electrophoresis. Conversely, we identified another case where the result was negative based on TG concentration, but chyle was present on electrophoresis. These data support previous work 6 which reported that 14% (n = 74) of chylothoraces were associated with a total TG concentration of less than the cut-off value of 1.24 mmol/L suggested by Staats et al. 4
Conclusion
The data presented here confirm that lipoprotein electrophoresis has a role to play in the diagnosis of chylothorax. However, it is not feasible to perform lipoprotein electrophoresis as a front-line text for the detection of chyle – it is a laborious and costly investigation which is often run in batches due to small sample numbers. Therefore, we advocate its use in selected cases, to replace visual inspection when the TG is between 0.5 and 1.3 mmol/L, or where clinical suspicion of chylothorax is high. Furthermore, its use may be extended to cases where the TG criteria of Staats et al. and visual inspection disagree, for example a TG >1.24 mmol/L but no layer of chylomicrons on visual inspection. In these situations, lipoprotein electrophoresis can further contribute to the diagnosis of chylothorax.
Footnotes
Acknowledgements
We would like to thank Nottingham University Hospitals for providing the opportunity to complete this research.
Declaration of conflicting interests
The authors declare no conflict of interest.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Ethical approval
N/A.
Guarantor
FA.
Contributorship
FA conceived the research topic; SG performed the practical aspects, did the data analysis and wrote the first draft of the manuscript. FA reviewed the manuscript and SG made any necessary changes. Both SG and FA have approved the final version for submission.