Neonatal chylothorax and ultrasound: Diagnosis, measurements, and correlations with management

Abstract

BACKGROUND

Chylothorax is a rare condition in neonates. It is associated with increased mortality, prolonged hospital stay, and significant nutritional, metabolic, and immunological complications. We hypothesized that lung ultrasound may play a critical role in managing this condition. However, standardized ultrasound parameters for the management of chylothorax have not currently been identified.

AIM

To determine preliminary ultrasound measurement parameters for chylothorax in neonates and to assess their usefulness for future prospective studies.

METHODS

We retrospectively reviewed 10 cases of congenital chylothorax in neonates admitted to the neonatal intensive care unit at Fondazione Policlinico Universitario A. Gemelli. We collected the following standardized ultrasound measurements: Lung-chest wall distance, lung-diaphragm distance, and derived area. Finally, the correlation between these measurements was analyzed according to the primary outcome of the need for drainage.

RESULTS

The median gestational age of the study infants was 34.7 weeks. Five neonates required thoracentesis, and seven needed pleural drainage placement. Ultrasound evaluations were performed at birth and subsequently every 48 hours, or more frequently in cases of clinical instability. The receiver operating characteristic curves gave these values for the proposed measurements (setting the chest tube drainage as the outcome): Lung-chest wall distance [area under the curve (AUC) = 0.917; 95% confidence interval (CI): 0.810-1]; lung-diaphragm distance (AUC = 0.938; 95%CI: 0.869-1); derived area (AUC = 0.984; 95%CI: 0.954-1); cut-off values of 0.8 cm, 1.2 cm and 0.72 cm² were identified for the lung-chest wall distance, lung-diaphragm distance, and derived area, respectively.

CONCLUSION

Lung ultrasound may be a valuable tool in the management of neonates with chylothorax. Future studies with a larger cohort are needed to confirm our findings.

Key Words: Chylothorax; Newborn; Lung ultrasound; Infant; Pleural effusion

Core Tip: Chylothorax is a rare but severe condition in neonates. Lung ultrasound can be useful in managing this condition, but ultrasound parameters have not been standardized thus far. The results of this study could be useful for proposing prospective studies intended to test “measurement protocol” for chylothorax in neonates. We retrospectively reviewed 10 cases of congenital chylothorax and measured the lung-chest wall distance, the lung-diaphragm distance, and the derived area. Receiver operating characteristic curves showed interesting values for the proposed measurements. Lung ultrasound, in the future, can be studied as a valuable tool in the management of neonates with chylothorax.

INTRODUCTION

Chylothorax is a rare condition in newborns characterized as an effusion of lymphatic fluid in the pleural cavity with congenital or traumatic etiology[1,2]. Although it is an infrequent occurrence in the neonatal population, the mortality rate is high. In less severe cases, complications include long hospital stays, lung hypoplasia, and heart failure. Moreover, the loss of antibodies, nutritional components, lymphocytes, coagulation factors, and immunoglobulins in the pleural cavity causes immunological and nutritional impairment and increases the risk of infections, malnutrition, and coagulation disorders[1,3].

The diagnosis of chylothorax is based on analysis of the pleural fluid. Management includes supportive therapies, thoracentesis, parenteral nutrition, nothing per mouth, medium chain triglycerides diet, and pharmacological and surgical treatments[1-4]. Because the incidence of chylothorax is rare[3], there is a lack of data on this condition. Clinical recommendations are based on clinicians’ experiences and opinions. Currently, there is no gold standard management strategy although some algorithms have been proposed for the management and follow-up of chylothorax[1-3]. The flow rate and volume of the effusion have been the most widely proposed variables to direct the decision-making process.

Recently, lung ultrasound has described as a useful modality for the detection and characterization of pleural effusion[2,5,6]. Lung ultrasound is a valuable tool for the management of respiratory conditions in sick newborns[5,6]. Ultrasound is simple, rapid, and reproducible with no side effects. Furthermore, it is easy to learn, and its findings are very accurate when assessed with the patient’s history and other examinations[5]. Ultrasound images can provide both qualitative and quantitative information about the condition to the clinician.

Lung ultrasound is a beneficial tool for frequent bedside assessments of chylothorax. Clinicians use the ultrasound findings to determine the appropriate treatment and its duration as well as to support the decision to start, continue, or increase a diet. Ultrasound findings also corroborate flow rate evaluations proposed by different decision-making algorithms in the literature[2,7]. Therefore, lung ultrasound can be used for both diagnosis and monitoring of chylothorax[7]. Ultrasound can be performed during the outpatient follow-up of infants with chylothorax as well[2]. Currently, no standardized ultrasound parameters have been identified to determine when to drain the effusion or how to monitor the effusion in neonates.

The aim of the present study was to explore correlations between ultrasound metrics and drainage outcomes, obtaining preliminary ultrasound measurement parameters useful for future prospective studies.

MATERIALS AND METHODS

Study design and patient recruitment

This was a retrospective, observational study conducted at Fondazione Policlinico Universitario A. Gemelli IRCCS (Rome, Italy) from April 2018 to November 2023. We reviewed the data from cases of congenital chylothorax in neonates admitted to the neonatal intensive care unit at our facility. Informed consent was provided by the caregivers of the neonates. The study protocol was approved by the Institutional Review Board of Fondazione Policlinico Gemelli IRCCS.

Data collection

We reviewed the medical records of neonates diagnosed with chylothorax. Cases with incomplete or missing documentation (missing ultrasound data in our retrospectively reviewed medical records) were excluded to ensure accuracy and consistency. We systematically collected demographic characteristics, lung ultrasound data obtained during hospitalization, treatment regimens, and clinical outcomes.

Study design and measurements

The primary goal of this study was to evaluate factors that could predict the need for chest tube drainage in neonates with chylothorax. In our clinical practice, the decision to position a drainage was guided by clinical status of the newborn, in terms of respiratory deterioration or haemodynamic instability in a context of pleural effusion progression. The management of these cases followed our internal protocol that is based on the currently available literature and expert consensus in the field, an approach in which ultrasound metrics do not play a leading or critical role in the decision making[3].

Lung ultrasound examinations were performed serially by a trained physician on each newborn according to clinical need. A 12 MHz hockey stick probe was used. Images were taken with a General Electric LOGIQ E ultrasound machine. The infants were positioned in the supine position during the ultrasound examination, allowing for consistent and reliable imaging. During the ultrasound examination, we scanned the anterior, lateral, and posterior chest walls using the anterior and posterior axillary lines as boundaries. When scanning each lung region, the probe was held perpendicular or parallel to the ribs. All examinations were performed using sterile disposable probe covers as established by internal protocols.

To ensure consistency and reproducibility in our measurements, we standardized lung ultrasound parameters for each examination to ensure reproducible results, with all measurements taken along the inferior posterior axillary line (Figure 1): (1) Lung-chest wall distance (cm) measured along the mid-axillary line with a perpendicular line to the lung and the chest wall (Figure 1, line A); (2) Lung-diaphragm distance (cm) measured along the mid-axillary line at the point of maximum projection perpendicular to the diaphragm (Figure 1, line B); and (3) Derived area (cm²) calculated by multiplying the lung-chest wall distance and the lung-diaphragm distance (Figure 1, orange rectangle).

Figure 1 Illustration of the ultrasound measurements. The lung-chest wall distance (line A) and the lung-diaphragm distance (line B) are measured in centimeters to determine the derived area (orange rectangle) as a geometric approximation.

The “derived area” parameter was conceived as a simple geometric surrogate of effusion extent measurable at the bedside. All the patients were placed in supine position for at least one hour before the ultrasound examination. In patients with no respiratory assistance or assisted with conventional mechanical ventilation, the measurements were collected during the expiratory phase. In case of high frequency oscillatory ventilation, the respiratory phase was not relevant.

Statistical analysis

We conducted a detailed analysis of the correlation between the ultrasound measurements and the primary clinical outcome (need for chest tube drainage). We constructed receiver operating characteristic (ROC) curves for each parameter to identify the cut off value for each ultrasound measurements (setting the need for chest tube drainage as outcome). Categorical data are expressed as absolute n (%), and numerical data are expressed as median and interquartile range. Data were analyzed using the IBM Statistical Package for Social Science 25.0 version (SPSS Inc., Armonk, NY, United States). A P value < 0.05 was considered statistically significant.

RESULTS

Baseline characteristics of patients

We analyzed 10 patients diagnosed with neonatal congenital chylothorax. The median gestational age was 34.7 weeks (interquartile range 32.5-37.4 weeks). A prenatal diagnosis was made in 8 patients (80%), and delivery room intubation was required for 7 patients. Additionally, 4 patients (40%) presented with associated ascites at the time of the clinical onset, and pleural effusion was bilateral in 7 patients (70%). Specifically, two neonates underwent surgery for diaphragmatic hernia; four were diagnosed with a genetic disorder (one with trisomy 21 and three with Noonan syndrome); one was affected by congenital lymphangiomatosis; one had chylothorax of unknown etiology, associated with other malformations, and is undergoing follow-up for a rare disease; and two had idiopathic chylothorax. The baseline characteristics of the patients are summarized in Table 1.

Table 1 Basic population details.

Parameter	Median (interquartile range) or n (%)
Gestational age (weeks)	34.7 (32.5-37.4)
Birth weight (g)	2650 (2470-3105)
Male	5 (50)
Gestational age < 37 weeks	7 (70)
Gestational age ≤ 32 weeks	2 (20)
Prenatal diagnosis	8 (80)
Vaginal delivery	2 (20)
Apgar 1’	5 (4-6)
Apgar 5’	7 (6-8)
Delivery room intubation	7 (70)
Ascites	4 (40)
Bilateral pleural effusion	7 (70)

Chylothorax management

Five patients required thoracentesis, and seven needed pleural drainage placement. The analysis of the pleural fluid confirmed the chylothorax diagnosis in all patients. We analyzed the fluid samples obtained after drainage and the diagnosis was based on the cell count. Chylothorax in neonates is defined as: > 1000 white blood cells/μL and > 70% lymphocytes in fasting (more than 6 hours) patients; > 1000 white blood cells/μL, >70% lymphocytes, and > 110 mg/dL triglycerides in patients receiving enteral nutrition. At the time of diagnosis, all newborns had not consumed any food. Five neonates were treated with continuous octreotide infusion. The treatment had a median duration of 29 days (range: 24-35 days), and the dosage was adjusted based on the clinical response of each patient. According to our internal protocols, the patients continued fasting until the pleural drainage volume decreased to below 2 mL/kg/day or was completely resolved.

Once clinical stability was achieved and the drainage volume no longer increased, enteral feeding was initiated. Initially, specialized formula enriched with medium-chain triglycerides was introduced and gradually increased. After the patient could tolerate this formula, breast milk was introduced into the feeding regimen. Details regarding the management and outcomes of the patients are summarized in Table 2.

Table 2 Management and outcomes.

Parameter	Median (interquartile range) or n (%)
Thoracentesis	5 (50)
Pleural drainage placement	7 (70)
Duration (days)	15 (11-21)
Octreotide	5 (50)
Octreotide, start day	18 (14-20)
Octreotide, stop day	50 (39-85)
Duration (days)	29 (24-35)
Octreotide, start dose (μg/kg/hour)	2 (1-4)
Octreotide, highest dose (μg/kg/hour)	12 (9-17)
Parenteral nutrition (days)	27 (11-42)
Start enteral nutrition (days)	18 (8-27)
Start breast milk (days)	38 (33-45)
Recurrence	1 (10)
Survival	9 (90)

Ultrasound data

Ultrasound evaluations were performed at birth and every 48 hours or more frequently in cases of clinical instability during hospitalization. If pleural effusion was present, measurements of the lung-chest wall distance, lung-diaphragm distance, and derived area were taken. To assess the predictive power of each measurement, we constructed ROC curves. The ROC curves (setting the need for chest tube drainage as the outcome) gave these values: An area under the curve (AUC) of 0.984 [95% confidence interval (CI): 0.954-1.000] for the derived area, 0.938 (95%CI: 0.869-1.000) for the lung-diaphragm distance, and 0.917 (95%CI: 0.810-1.000) for the lung-chest wall distance. We found cut-off values of 0.8 cm, 1.2 cm and 0.72 cm² for the lung-chest wall distance, lung-diaphragm distance, and derived area, respectively. The sensitivity, specificity, positive predictive values, negative predictive values, and accuracy related to these the cut-off values are reported in Table 3. The ROC curves are shown in Figure 2.

Figure 2 Receiver operating characteristic curves for lung-chest wall distance, lung-diaphragm distance, and derived area. Area under the curve (AUC) of 0.984 (95%CI: 0.954-1.000) for the derived area, 0.938 (95%CI: 0.869-1.000) for the lung-diaphragm distance, and 0.917 (95%CI: 0.810-1.000) for the lung-chest wall distance.

Table 3 Reliability of ultrasound measurements for the need of chest tube drainage, 95%CI.

	Lung-chest wall distance (cm)	Lung-diaphragm distance (cm)	Derived area (cm2)
Cut-off value	0.8	1.2	0.72
Sensitivity (%)	71.4 (35-92)	100 (59-100)	100 (59-100)
Specificity (%)	100 (90-100)	81.8 (67.7-90.7)	93.2 (81-98)
PPV (%)	100	46.7	70
NPV (%)	95.7	100	100
LR+	> 15.00	5.50	14.60
LR-	0.28	0.00	0.00
Accuracy	0.96	0.84	0.94

PPV: Positive predictive value; NPV: Negative predictive value.

DISCUSSION

Chylothorax in neonates is a rare but serious condition. Currently, there are no standardized clinical and therapeutic management protocols available. We hypothesized that lung ultrasound would be an advantageous imaging modality to assist in the diagnosis, treatment decision making, and follow-up of patients with chylothorax.

A primary concern in the treatment of chylothorax is the decision to intervene with pleural drainage and/or the placement of permanent drains. These procedures are associated with parenchymal lung injury[1] and infection risk. Patients with chylothorax experience an immunocompromised state because of an associated deficiency in immunoglobulins[1,8,9]. Furthermore, when there is a significant volume of pleural fluid, the patient typically presents with tachypnea followed by dyspnea, potentially leading to hemodynamic instability. These patients require close monitoring in an intensive care setting to observe the effusion and the symptoms. When the clinical state of the patient worsens, the patient experiences a rapid decline in their overall condition.

Chylothorax is typically characterized by a prolonged hospitalization period with phases of observation, improvement, and potential reaccumulation of pleural effusion[10-13]. The treatment of chylothorax is complex and based on a stepwise approach that depends on the severity of the clinical condition, the need for drainage, and the volume of chyle. Treatment consists of parenteral nutrition, the use of special formulas, and multiple medications, with octreotide being the most frequently used[14-17].

Thoracic ultrasound is an easy-to-perform imaging technique that has many advantages for managing patients with chylothorax. Although ultrasound imaging is operator-dependent, it is a reproducible, noninvasive, fast, and convenient (can be used as a bedside procedure). However, there are no standardized protocols for managing neonatal chylothorax using thoracic ultrasound.

Our study aimed to determine possible ultrasound measurements correlating to the need for chest drainage in neonates presenting with chylothorax. Given the limitations of the study, our data need to be intended as pilot data, requiring external validation. We conducted ultrasound every 48 hours during the acute phase or more frequently in cases of clinical deterioration. We determined that the correct positioning of the probe was on the mid-axillary line in the basal position with axial orientation. This enabled us to identify the lung-chest wall distance and the lung-diaphragm distance to obtain the derived area. The “derived area” parameter was conceived as a simple geometric surrogate of effusion extent measurable at the bedside.

The population analysis showed that these measurements could be used, as a basis for further studies, to identify patients who require pleural drainage and who could benefit from the placement of a thoracic drainage catheter. Figure 3 presents a proposed practical flow-chart (intended solely for future research).

Figure 3  Hypothetical algorithm for future prospective testing.

This study serves as a preliminary foundation, intended to explore measurements to be validated in future protocols for the appropriate use of ultrasound to manage chylothorax. There are some limitations that should be acknowledged. First, this study was retrospective and conducted at a single center, limiting the generalizability of the results. As a retrospective study, a major limitation is the unfeasibility to perform a reliability analysis of the measurements. Furthermore, even if the measurements did not impact the decision making process (not influencing the decision to drain the chylothorax), the physicians performing the lung ultrasound were not blinded to the outcome. We recognize it as a bias risk. Indeed, these results should be considered not as conclusive, but as a potential and interesting substrate for future prospective studies. Second, the decision to perform drainage was based on clinical judgment. Finally, the sample size was small, although this is understandable given the rarity of the condition[18-21]. It is advisable to conduct a prospective study enrolling a larger population (sample size of at least 30-50 patients) to statistically confirm the trend of our results. Nevertheless, the proposed results are promising as they could be used for future studies designed with the intent to standardize the quantification of pleural effusion. In the future, new measurements could be incorporated into a clinical management algorithm. It is essential to design additional prospective, multicenter trials to rigorously validate the proposed method.

CONCLUSION

Lung ultrasound is a useful tool for the diagnosis and management of chylothorax in neonatal intensive care unit settings. A standardized ultrasound protocol with specific measurements for the detection of chylothorax will be beneficial for neonatologists in the decision-making process and management of this condition.

ACKNOWLEDGEMENTS

The authors acknowledge Ministero della Salute-Ricerca Corrente 2025.