Of the 42 fetuses with type I choledochal cyst over the study period, 3 were excluded. Of the 3 excluded fetuses, 2 were ruled out due to operations at other hospitals, image quality was poor in 1, leaving 39 for study analysis. (Supplementary Fig. S2).
Magnetic resonance imaging was performed on 1367 healthy fetuses in our hospital during the same period. 1267 were excluded. Of the 1267 excluded fetuses, 1171 were examined for other body parts other than the abdomen, image quality was poor in 35 fetuses, the incomplete sequences were in 35, 5 were twins, leaving 100 for study analysis. (Supplementary Fig. S3).
In summary, a total of 39 fetuses with type I choledochal cysts and 100 healthy controls were included in this study.
Baseline Characteristics Of The Study Population
Demographic characteristics and imaging findings of the participants were summarized in Table 1. There was no statistically significant difference in liver length, liver width and liver OD values among the healthy fetuses and fetuses with type I choledochal cysts. Compared with healthy fetuses, patients had significantly lower of the gallbladder length, width and OD in the type I choledochal cysts group. Compared with healthy fetuses, patients had significantly lower of the gallbladder length, width and OD in the type I choledochal cysts group. The portal vein diameter, spleen length, spleen width and spleen OD was greater in the type I choledochal cysts group than in the healthy group. Although there was no difference in liver morphological changes between the two groups, we found that liver fibrosis was confirmed by postoperative pathology in 18 of the 39 cases, as shown in supplementary Fig. S4.
Table 1
Baseline Characteristics of participants.
| Healthy fetuses (n = 100) | Fetuses with Type I choledochal cysts (n = 39) | P-value |
Gender(male/female) | 66(34.00%)/34(66.00%) | 8(20.51%)/31 (79.49%) | < 0.001 |
Gestational weeks | 28.84 ± 3.94 | 26.63 ± 4.09 | 0.004 |
Portal vein diameter (mm) | 2.34 ± 0.19 | 2.47 ± 0.34 | 0.007 |
Gallbladder length (mm) | 17.18 ± 3.10 | 12.84 ± 5.00 | < 0.001 |
Gallbladder width (mm) | 5.77 ± 1.26 | 5.03 ± 2.75 | 0.033 |
Gallbladder OD (mm2) | 93.28 (75.50-123.42) | 55.08 (37.05–75.60) | < 0.001 |
Liver length (mm) | 60.15 ± 9.84 | 58.93 ± 10.08 | 0.51 |
Liver width (mm) | 42.16 ± 6.33 | 43.19 ± 9.13 | 0.45 |
Liver OD (mm2) | 2565.61 (2033.62-3106.55) | 2401.33 (1888.51-3342.47) | 0.86 |
Spleen length (mm) | 19.66 ± 4.62 | 23.07 ± 5.74 | < 0.001 |
Spleen width (mm) | 8.21 ± 2.44 | 9.38 ± 3.48 | 0.026 |
Spleen OD (mm2) | 153.93 (101.16-216.95) | 207.68(141.69-245.56) | 0.003 |
Cyst volume (mm3) | —— | 931.91(342.69-2195.20) | < 0.001 |
OD: multiplication of orthogonal diameter |
Univariate Analysis For Gallbladder Size
The results of univariate analysis for gallbladder size were shown in Table 2. The results of univariate analysis showed that gestational weeks were positively correlated with gallbladder size (β = 6.47, p = 0.0002) and the same results were observed for liver width, liver OD, and spleen length (β = 1.89, p = 0.022; β = 0.02, p = 0.042; β = 3.04, p = 0.017, respectively). We also found that gender, portal vein diameter, liver length, spleen width, spleen OD and the cyst volume were not associated with gallbladder size.
Table 2
The results of univariate analysis for Gallbladder OD values in fetuses with Type I choledochal cysts.
| Statistics | β(95%CI) | p-value |
Gender | | | |
Female | 31 (79.49%) | ref | |
Male | 8 (20.51%) | 8.69 (-27.30, 44.68) | 0.64 |
Gestational weeks | 26.63 ± 4.09 | 6.47 (3.44, 9.51) | 0.0002 |
Portal vein diameter (mm) | 2.47 ± 0.34 | 11.21 (-32.70, 55.12) | 0.62 |
Liver length (mm) | 58.93 ± 10.08 | 1.15 (-0.28, 2.58) | 0.13 |
Liver width (mm) | 43.19 ± 9.13 | 1.89 (0.35, 3.43) | 0.022 |
Liver OD (mm2) | 2611.83 ± 953.17 | 0.02 (0.00, 0.03) | 0.042 |
Spleen length (mm) | 23.07 ± 5.74 | 3.04 (0.67, 5.41) | 0.017 |
Spleen width (mm) | 9.38 ± 3.48 | 2.07 (-2.23, 6.38) | 0.35 |
Spleen OD (mm2) | 207.68 (141.69-245.56) | 0.08 (-0.03, 0.19) | 0.15 |
Cyst volume (mm3) | 931.91 (342.69-2195.20) | -0.00 (-0.01, 0.01) | 0.82 |
OD: multiplication of orthogonal diameter; ref: reference. |
The Relationship Between Gestational Weeks And Gallbladder Size In Different Models
Univariate linear regression models were used to evaluate the associations between gestational weeks and gallbladder size in fetuses with type I choledochal cysts in different models. Meanwhile, the non-adjusted and adjusted models were showed in Table 3. In the non-adjusted model, gestational weeks showed positive correlation with gallbladder OD (β = 6.47, 95%CI: 3.44 to 9.51, P = 0.0002). In the minimally adjusted model (adjusted gender), the effect size was an obvious positive correlation (β = 7.28, 95%CI: 3.99 to 10.56, P = 0.0001). After adjusting for other covariates, the effect size was also positive (β = 13.47, 95%CI: 6.83 to 20.10, P = 0.0005).
Table 3
Relationship between gestational weeks and gallbladder OD of fetuses with Type I choledochal cysts in different models.
Variable | Non-adjusted (β, 95%CI, P) | Adjust I (β, 95%CI, P) | Adjust II (β, 95%CI, P) |
Gestational weeks | 6.47 (3.44, 9.51) 0.0002 | 7.28 (3.99, 10.56) 0.0001 | 13.47 (6.83, 20.10) 0.0005 |
Non-adjusted model: All of the covariants were not adjusted. |
Adjust I model: We adjusted gender. |
Adjust II model: We adjusted gender, portal vein diameter, liver length, liver width, liver OD, spleen length, spleen width, spleen OD, cyst volume. |
CI: confidence interval; OD: multiplication of orthogonal diameter |
The analyses of non-linear relationship and threshold analysis was performed between gallbladder size with gestational weeks
We analyzed the non-linear relationship between gallbladder OD with gestational weeks (Fig. 1). We found that the relationship between gallbladder OD and gestational weeks was non-linear (after adjusting gender, portal vein diameter, liver length, liver width, liver OD, spleen length, spleen width, spleen OD, cyst volume). By using a two-piecewise linear regression model, we calculated that the two inflection points were 27 and 34. On the right of the first inflection point, the effect size was 32.31 with 95%CI (7.23, 57.38), p = 0.021. On the left of the second inflection point, we also observed a positive relationship between gallbladder OD and gestational weeks (27.23, 4.62 to 49.84, p = 0.05) (Table 4).
Table 4
The results of two-piecewise regression model in fetuses with Type I choledochal cysts.
Inflection points of gestational weeks (Per 1 change) | Effect size β (95%CI) | P value |
Inflection point 27 | | |
< 27 | 3.92 (-1.94, 9.78) | 0.20 |
> 27 | 32.31 (7.23, 57.38) | 0.021 |
Log likelihood ratio test | | 0.012 |
Inflection point 34 | | |
< 34 | 27.23 (4.62, 49.84) | 0.05 |
> 34 | -17.90 (-53.07, 17.26) | 0.35 |
Log likelihood ratio test | | 0.032 |
Effect: Gestational weeks |
Cause: Gallbladder OD |
Adjusted: gender, portal vein diameter, liver length, liver width, liver OD, spleen length, spleen width, spleen OD, cyst volume. |
CI: confidence interval; OD: multiplication of orthogonal diameter |
Comparison of gallbladder size with gestational age between healthy fetuses and type I choledochal cysts group
Similarly, we found that gallbladder OD changes with gestational weeks in normal fetuses as shown by the solid line after adjusting for gender, portal vein diameter, liver length, liver width, liver OD, spleen length, spleen width, spleen OD (Fig. 2). Compared with normal fetuses, the overall trend in gallbladder OD was lower in the type I choledochal cysts group than in the normal fetuses. The change of gallbladder OD with gestational weeks in normal fetuses was shown in Supplementary Fig. S5.