Between June 14, 2022 and August 31, 2024, a total of 19 participants diagnosed with PLE following the Fontan operation were enrolled. Among these, four patients voluntarily withdrew their consents (2 patients withdrew after the 1-month follow-up and other two after completing the 3-month follow-up) due to a lack of symptom improvement or a perceived worsening of symptoms despite taking the medication. Consequently, a total of 15 patients completed the study. Since the primary outcome focused on the changes in serum albumin levels from baseline to 6 months or to the study endpoint, a PP analysis was conducted. The baseline characteristics of the PP population are summarized in Table 1. Of the 15 participants, 8 (53.3%) were male, and 7 (46.7%) were female. The median age was 15 years, with the youngest being 12 years old. The median body weight was 36.9 kg [interquartile range (IQR), 31.7–44.1], and the median time from Fontan operation to PLE diagnosis was 2.4 years (IQR, 1.0–7.2). At baseline, the median serum albumin level was 2.2 g/dL (IQR, 1.8–3.2), and the median stool alpha-1 antitrypsin level was 215.6 mg/dL (IQR, 96.9–412.0).

The primary outcome was to assess the improvement in serum albumin levels at 6 months or at the study endpoint compared to baseline. Median serum albumin increased from 2.2 g/dL (IQR, 1.8–3.2) at baseline to 2.5 g/dL (IQR, 2.3–3.5) at 6 months (p = 0.183). Although not statistically significant, this change may indicate a trend toward improvement. At the study endpoint, following treatment discontinuation, a non-significant decrease was observed (2.4 g/dL, IQR, 2.1–3.4; p = 0.345) (Fig. 1A, B).

Changes in serum albumin during the study period. A Serum albumin levels over the study period, showing an increasing trend until 6 months, followed by a decrease after discontinuation of the medication. B Comparison of serum albumin levels at baseline, 6 months, and study endpoint, with no statistically significant differences. (p = 0.124, for baseline vs. 6 months, p = 0.382, for baseline vs. endpoint)

Changes in stool alpha-1 antitrypsin level, body weight, and stool frequency are summarized in Fig. 2. Stool alpha-1 antitrypsin levels significantly decreased from a median of 215.6 mg/dL (IQR, 96.9–412.0) at baseline to 75.5 mg/dL (IQR, 29.6–236.7) at 6 months (p = 0.016). At the study endpoint, levels increased numerically to 143.7 mg/dL (IQR, 72.0–241.0), though this change was not statistically significant (p = 0.213).

Changes in stool alpha-1 antitrypsin levels, body weight, and stool frequency during the study period. A Stool alpha-1 antitrypsin levels over the study period, showing a decreasing trend until 6 months, followed by an increase after discontinuation. B Body weight over the study period, showing no statistically significant changes. C Daily stool frequency over the study period, showing a significant reduction by the study endpoint. D Comparison of stool alpha-1 antitrypsin levels at baseline, 6 months, and study endpoint, with a statistically significant reduction between baseline and 6 months (p = 0.016). E Comparison of body weight at baseline, 6 months, and study endpoint, with no statistically significant differences. F Comparison of daily stool frequency at baseline, 6 months, and study endpoint, with a statistically significant reduction between baseline and study endpoint (p = 0.007)

Body weight remained stable throughout the study. Median values were 36.9 kg (IQR, 31.7–44.1) at baseline, 36.5 kg (IQR, 32.8–43.9) at 6 months (p = 0.394), and 36.7 kg (IQR, 32.8–45.5) at the study endpoint (p = 0.256).

Stool frequency improved significantly from 3.0 times per day (IQR, 1.5–3.5) to 1.5 times per day (IQR, 1.0–1.8) by the study endpoint (p = 0.007). Despite these improvements, the grade of ascites assessed by ultrasound and the proportion of patients experiencing diarrhea and edema did not show any statistically significant differences. (Supplementary Fig. 1).

A subgroup analysis was conducted to compare the differential responses in serum albumin and stool alpha-1 antitrypsin levels based on the presence of diarrhea at baseline. Patients were divided into two groups: those with diarrhea (n = 7) and those without diarrhea (n = 8). Baseline characteristics of the two groups are summarized in Table 2. Other than the presence of diarrhea, there were no significant differences between groups in age, sex, body weight, or the time interval from Fontan operation to PLE diagnosis. The diarrhea group had a significantly higher baseline stool frequency (median, 3.5 stools/day) compared with the non-diarrhea group (median, 1.5 stools/day; p = 0.009).

Changes in laboratory values and symptoms are shown in Fig. 3. In the diarrhea group, serum albumin increased from 1.8 g/dL (IQR, 1.7–2.1) at baseline to 2.4 g/dL (IQR, 2.3–3.3) at 6 months (p = 0.138), while the non-diarrhea group showed a slight decrease from 2.7 g/dL (IQR, 2.1–3.5) at baseline to 2.6 g/dL (IQR, 2.5–3.4) (p = 0.623). At the study endpoint, serum albumin levels in the diarrhea group were 2.2 g/dL (IQR, 2.0–3.1) (p = 0.116, for baseline vs endpoint), while the non-diarrhea group showed levels of 2.6 g/dL (IQR, 2.2–3.3) (p = 0.866, for baseline vs. endpoint).

Subgroup analysis of patients based on the presence of diarrhea at baseline. A Serum albumin levels over the study period in the two groups. B Stool alpha-1 antitrypsin levels in the groups during the study period. C Comparison of stool frequency between the groups throughout the study period. D Body weight changes in the groups over the study period. Overall, patients with diarrhea demonstrated more favorable treatment responses

In the diarrhea group, stool alpha-1 antitrypsin decreased from 220.3 mg/dL (IQR, 176.2–486.5) to 75.5 mg/dL (IQR, 27.3–230.7) at 6 months, with borderline statistical significance (p = 0.075). The non-diarrhea group also showed a reduction from 194.1 mg/dL (IQR, 56.7–344.3) to 66.2 mg/dL (IQR, 40.7–210.9) (p = 0.116). At the study endpoint, the diarrhea group exhibited a statistically significant decrease to 100.3 mg/dL (IQR, 38.8–132.1; p = 0.043), while the non-diarrhea group showed a non-significant increase to 275.0 mg/dL (IQR, 165.3–396.6; p = 0.600).

Regarding stool frequency, the diarrhea group showed a significant decrease from a median of 3.5 stools/day (IQR, 3.5–4.3) to 1.5 stools/day (IQR, 1.5–1.8) at the study endpoint (p = 0.042). In the non-diarrhea group, stool frequency decreased from 1.5 stools/day (IQR, 1.0–2.3) to 1.0 stools/day (IQR, 0.5–1.25), but the change was not statistically significant (p = 0.084). No significant differences were observed in body weight changes in either group over the study period.

Of the 19 participants enrolled in the study, four withdrew before completing the trial. The reasons for their withdrawal are summarized in Table 3. Two participants experienced mild abdominal discomfort and did not observe an increase in serum albumin levels within the first month of treatment. Based on these observations, they chose not to continue participating in the clinical trial and voluntarily withdrew consent. The remaining two participants, who had pre-existing diarrhea related to PLE, reported persistent diarrhea after three months of treatment. Both patients and their guardians determined that there were no improvements and decided to withdraw consent voluntarily. None of these withdrawals were due to the side effects attributable to CM.

In this study, adverse events were reported in 8 participants, representing 53.3% of the study cohort. The predominant adverse drug reactions were gastrointestinal symptoms, including abdominal discomfort, diarrhea, and vomiting (4/8, 50%). In addition, 2 participants experienced electrolyte imbalances, 1 experienced worsening of anemia, and 1 reported chest pain. Importantly, no serious adverse events associated with CM were observed during the study. The adverse effects observed were consistent with the established side effect profile of the medication, which had been recognized previously during its approval process. No unexpected adverse events were observed.