The accuracy of cerebellar diameter in detection of the expected date of delivery in comparison with femur length in third trimester

Introduction

Accurate estimation of gestational age (GA) is critical for effective obstetric planning and also aids in estimating foetal malformation or growth retardation. Failure can result in iatrogenic prematurity or postmaturity, both of which have been linked to increased perinatal morbidity and mortality^[1] . Ultrasound foetal biometry is highly reliable in the first and second trimesters of pregnancy^[2,3] but the reliability of any ultrasound method decreases as gestation progresses. Many studies ^[2] show that the reliability of any single ultrasound parameter is poor in the third trimester. The cerebellum, which is the largest part of the hind brain, is located in the posterior cranial fossa. It is located dorsal to the pons and medulla, separated by the fourth ventricle. The cerebe-llum is roughly spherical, but its median is somewhat constricted.

The foetal biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC), and femur length (FL) are the most commonly used biometric parameters for estimating gestational age ^[6] .These parameters have few limitations because conditions that change the shape of the skull affect BPD, which is a well-accepted indicator of GA^[7]. The main issue is that measuring BPD in a foetus with an abnormal skull shape, such as dolichocephaly or brachycephaly, is difficult. Femur length varies with ethnicity and skeleton congenital anomalies ^[8] . As a result, transverse cerebellar diameter (TCD) was developed as an alternative parameter of foetal brain growth and gestational age estimation ^[5] .

Because the cerebellum is located in the posterior cranial fossa and is surrounded by dense petrous ridges and the occipital bone, it can withstand extrinsic pressure deformation better than the parietal bones^[5]. Because ultrasound can easily visualise the foetal cerebellum, imaging the posterior fossa is becoming an integral part of many routine foetal sonograms ^[9]  Several authors working on transverse cerebellar diameter (TCD) have found it to be a better marker for GA estimation than other clinical and biometric parameters ^[10,11,12] . The current study attempted to evaluate foetal cerebellar development ultrasono-graphically, which aids in determining foetal gestational age.

Patients and methods

Ethical Consideration:

The ethical committee of  Obstetrics & gynecology department , Faculty of Medicine gave its approval to conduct the current study. Before enrolling in the study, all women gave verbal informed consent after the study's goals were clarified. The confidentiality of the data base was ensured.

This present study was a cross-sectional study done on 71 pregnant women with history of reliable menstrual cycles (sure of dates, regular cycles, no history of hormonal contraception at last 6 months, no history of abortion or delivery at last 3 months) and did not suffer from any complications for the mother or fetus, pregnant women sub divided in 2 groups .Group 1 is from (28 to 32 wk) & Group 2 from (32 to 36 wk). They referred for routine ultrasound examination in ultrasound department of El-Minia University maternity hospital during the period from August/2021 to March/2022. Pregnancies with congenital anomalies,

intrauterine fetal death & medical disorders like hypertension and diabetes mellitus, and antepartum hemorrhage were excluded from the current study.

- All enrolled women had been evaluated as: complete history taking, particularly for (menstrual history, obstetric history, medical illness that affects pregnancy, and vaginal bleeding). By convex probe 3.5 MHz using (VOLUSON S8), the fetus examined in the following sequences, firstly the patient lie supine with exposed abdomen, mesured SFH then the probe is applied to the center of the abdomen vertically to determine the fetal lie and presentation, then the FL is measured and the gestational age was calculated, TCD was measured in mm which is equivalent to gestational age.

- TCD measurement: The TCD was measured from the sub-occipito-bregmatic view in which the anterior horn of the lateral ventricles and cavum are visualized at front of the head together with the cerebellum at the back. The lateral ventricle view required for the BPD was obtained, then rotate the probe slightly downward, toward the fetal neck. The posterior horns of the lateral ventricles will disappear from view to be replaced by cerebellum. Then cerebellum is measured from one outer margin to the other outer margin in transverse.

Statistical Methods of Analysis:

The collected data was coded, entered, and analyzed using the IBM software statistical package for social sciences (SPSS) (version 25). For categorical variables, descriptive statistics were in the form of frequency and percentage, while for numerical variables in the form of mean and standard deviation. (mean ± SD). The proper statistical significance mea-sures were used: (Independent Sample t-test, Chi-Square (χ2) test and Pearson’s correlation analysis; {r- values: 0 to 0.3 positive or negative (slight), 0.3 to 0.7 (moderate) and 0.7 to 1 (strong). Statistical significance was described at a p-value of less than or equal to 0.05. Simple charts were used to demonstrate some findings.

Results

The current study was a cross sectional observational study that conducted on 71 pregnant women with history of regular cycles, no history of hormonal contraception at last 6 months, no history of abortion or delivery at last 3 months before pregnancy. Their age was ranged from (18) to (39) with an average age of (26.56±4.8) years old, all of them was with regular menstrual cycle (100%), regarding history of previous delivery; (20%) of them have never given birth while (80%) of them have given birth before, with a number ranging from one to six times with a median of (2 times) and an average previous delivery (1.83 ±1.2 times). Regarding history of previous abortion, among the studied women, 22% had never experienced a miscarriage, while 78% had had a previous abortion, with several previous abortions ranging from one to five times, a median of (1 time) and an average previous

abortion of (1.60 ±1.1 times). Regarding the GA as calculated by the 1^st day from the LMP, it was ranged from 28 weeks to 35 weeks with an average GA of (31.59 ±2.27) weeks.

Table (1) shows the association between FL (weeks) and GA among the studied pregnant, mean FL (weeks) was significantly (p=0.001) associated with GA, it was 29.6±1.07 in pregnant with GA 28-32 weeks, 33.6±1.1 in pregnant with GA 32-36 weeks. Also, there was a statistically significant strong linear positive correlation between FL (weeks) and GA (r=0.99, p=0.001).

Regarding the association between TCD (mm) and GA among the studied pregnant, mean TCD (mm) was significantly (p=0.001*) associated with GA, it was 29.7±1.2 in pregnant with GA 28-32 weeks, 33.7±1.2 in pregnant with GA 32-36 weeks. There was strong positive significant correlation; it was 0.93 (0.001*) in pregnant with GA 28-32 weeks, 0.95 (0.001*), in pregnant with GA -32-36 weeks.

Table (2) and Table (3) demonstrate a comparison of mean GA as estimated by TCD with that assessed by FL and by SFH among the studied pregnant women. Among pregnant women in the gestational age 28-32 weeks, there was non-statistically significant difference between GA as estimated by TCD and as estimated by FL or SFH. Also, among pregnant women in the gestational age 32-36 weeks, there was non-statistically significant difference between GA as estimated by TCD and as estimated by FL (Figure-1) or SFH. That indicates that TCD is very accurate in the early 3^rd trimester period.

A correlation between GA as estimated by TCD and as estimated by FL among the studied pregnant women demonstrates that there was a statistically significant positive linear corre-lation between TCD, and GA estimated by FL and by SFH/F, (Figure 2,3).

TCD in prediction of GA according to GA weeks groups, can correctly predict GA in early 3^rd trimester (28-36 weeks) at cut-off point >34.9 mm with a sensitivity of 94%.

Table (1): The association and correlation between femoral length (FL) for GA groups among the studied pregnant women; (N= 71):

*p-value ≤0.05 is considered statistically significant by independent sample t-test analysis

r: Pearson's Correlation Coefficient

Table (2): Comparison of mean GA as estimated by TCD with that estimated by SFH/F among the studied pregnant women; (N= 71):

Table (3): Comparison of mean GA as estimated by TCD with that estimated by FL among the studied pregnant women; (N= 71):

Figure (1): comparison of mean GA as estimated by TCD with that measured by FL among the studied pregnant women

Discussion

More accurate methods for dating pregnancies have emerged with the development of diagnostic ultrasonography, most used biometric indices for dating pregnancies are the BPD, HC, AC and FL ^[2] . Among these, FL parameter is considered superior to other parameters in predicting GA accurately as it can resist to external deformation better, however newer parameters are constantly being explored for their efficiency e.g. TCD, Humerus length, fetal kidney length, nasal bone length, three dimensional fractional limb volume, umbilical venous Doppler in growth restricted fetuses, etc ^[13] .

Fetal TCD is one such fetal parameter that has evolved as a superior one in predicting GA in both singleton and twin gestations ^[10] . The aim of the current study was to assess the reliability of TCD in compare with the femur length in predicting gestational age.

The study  included 71 pregnant women &  revealed that the mean gestational age measured using the TCD 29.63±1.08 during the 28-32 weeks and 33.69 ±1.12 during the 32-36 weeks GA, which was less but without a statistically significant difference when compared with

week difference from the GA calculated using FL, so this clarify that there was no significant difference between both measured gestational age in early third trimesters using FL and TCD.

In our study, The P value comparing TCD with BPD was found to be (0.477) in group 1 (from 28 to 32 wk) & (0.688) in group 2 (from 32 to

36 wk). The P value comparing TCD with FL was found to be (0.477) in group 1 & (0.858) in group 2 which incompatible with another study conducted by (Askr et al., 2019) in 200  patients between  (29-36 wk)  in there study The P value comparing  TCD with BPD within 3 days was found to be (<0.001) showing that there was highly significant difference between correct assessment within 3 days by BPD and by TCD.

The P value comparing TCD with FL within 3 days  was found to be (0.009) and there was a significant difference between correct assess-ment within 3 days by FL and by TCD^[14] . In the current study  we found that TCD can correctly predict GA in early 3rd trimester (28-36 weeks) at cut-off point >34.9 mm with a sensitivity of  94%, which incompatible with another study conducted by (Bassiouny et al., 2020) estimated the gestational age in third trimester  of pregnancy by fetal TCD and it is accuracy in 103 patient between (31-36 wk) TCD was accurate within 88.3% of the cases , while FL was accurate in 51.5% which mean that TCD is highest statistically significant measurement which could be used in Third trimester for assessing the gestational age compared with FL&BPD without any effect of  parity ^[15] .

In the current study, among pregnant women in the GA  (28-32 ) and (32_36) weeks there  was non-statistically significant difference between  GA as estimated by TCD and as estimated by other assessed fetal biometry (SFH, BPD, HC, FL and AC), which incompatible with  another study conducted by (Reddy et al ., 2017 ) on 100 pts between (15-40 wk) were divided into two groups based on GA (15-28 and 29-40 weeks). evaluated accuracy of prediction GA used fetal TCD and compared between TCD and other existing parameters in evaluating GA.

They showed that TCD was an accurate parameter in estimation of gestational age in second and third trimesters as its values are in close relation with that of GA by LMP. It is also better predictor of the gestational age when compared to other parameters especially in third trimester ^[16] .

In our study also approved that TCD seems to be good marker for GA calculation compared with other sonographic fetal Biometry (FL, HC, AC and BPD) at (28_36 wk) which is similar to a study conducted by (Zakaria et al., 2019 ) in 200 patients at third trimester (28-40wk) concluded that TCD is the most accurate method for assessment of gestational age in Third trimester followed by FL and the least accurate is the BPD, Also, combining accuracy of TCD (89%) and FL (81%)  in our study TCD is (94%)^[17] .